Please be advised that this is an electronic - State of Idaho

Please be advised that this is an electronic - State of Idaho

Please be advised that this is an electronic reproduction of Legislative proceedings and does not represent the official record of committee actions t...

24MB Sizes 0 Downloads 10 Views

Recommend Documents

Please be advised that this is an electronic - State of Idaho
Feb 13, 2013 - provisions of Section 59-917, Idaho Code, Fred S. Martin, State. Senator, District 15, Ada, State of Idah

Please be advised that this is an electronic - State of Idaho
Jan 8, 2014 - Chase, Chairman of the Idaho Water Resource Board (IWRB). RULES ... SPEAKER: Chairman Pearce welcomed Jack

Page 1 CAUTION: PLEASE BE ADVISED THAT THE FACTS OF
4 days ago - BUSHMASTER FIREARMS INTERNATIONAL, LLC, et al. SC 19832/19833. Judicial District of Bridgeport. Panel: Palm

THE PROJECT Please be advised that VCI - City of Pasadena
PWP CONTRACTOR. VCI Construction Inc. David Johnson, Project Manager. (909) 921-8237 [email protected] We apologize f

Please be advised that recently the below complaints and problems
<Case2> Negotiation regarding rent, deposit and gratuity (key money) when moving into apartments. Students try to co

23 June 2017 Please be advised that obsolete - Gove Operations
Jun 23, 2017 - The auction will be managed by Pickles Auctions. To view the catalogue list, visit the. Pickles Auction w

Page 1 Please be advised that the Barry County Board of
Jul 17, 2017 - Chair Geiger called the regular session of the Barry County Board of ..... The hotel will be happy to ass

1. Please be advised that the man known as Johnny Chattin, claiming
William ( Bill ) Taylor is a resident of the Qualla Boundary of North Carolina, the home of the Eastern Band of Cherokee

Note: This is an alternative text that may be considered in place of the
local radicals were the more successful. Samuel Adams orchestrated a demonstration on the 8th, the day of the victims' f

Conversations That Matter is an Oh Boy Productions program. Please
We'll be right back. - [Narrator] Conversations That Matter is a not-for-profit program made possible thanks to the char

Please be advised that this is an electronic reproduction of Legislative proceedings and does not represent the official record of committee actions that are contained in the committee minute books in the Legislative Library. This media is intended for personal, educational or research purposes only and should not be used for commercial or political purposes.

MINUTES

JOINT MEETING

SENATE STATE AFFAIRS COMMITTEE HOUSE ENVIRONMENT, ENERGY & TECHNOLOGY COMMITTEE DATE:

Monday, January 16, 2017

TIME:

1:30 P.M.

PLACE:

WW02-Lincoln Auditorium

MEMBERS PRESENT:

Chairman Siddoway, Vice Chairman Hagedorn, Senators Davis, and Winder Chairman Raybould, Vice Chairman Thompson, Representatives Hartgen, Vander Woude, Anderson, Anderst, Mendive(Mendive), Trujillo, Chaney, Nate, Cheatham, Reed(Horman), Malek, Moon, Nye(Smith), Jordan (5), Rubel

ABSENT/ EXCUSED:

Senators Hill, Lodge, Lakey, Stennett, and Buckner-Webb

NOTE:

The sign-in sheet, testimonies and other related materials will be retained with the minutes in the committee's office until the end of the session and will then be located on file with the minutes in the Legislative Services Library.

CONVENED:

Chairman Raybould called the Joint Meeting of the Senate State Affairs Committee and House Environment, Energy and Technology Committee (Committees) to order at 1:32 p.m. Chairman Raybould welcomed all those in attendance and introduced Dr. Mark Peters, Director of the Idaho National Laboratory (INL), to give his presentation (see attachment 1).

PRESENTATION: Dr. Peters stated that the Idaho National Laboratory (INL) is comprised of a Research and Education Campus in Idaho Falls and 890 square miles of Idaho land. They operate three reactors and are ready to start a fourth. INL includes paved roads, railroad lines, electrical transmission and distribution lines and a mass transit system. The infrastructure contains buildings, fire stations, reactors, nuclear and radiological facilities, spent fuel pools, and 300 metric tons of used fuel. The Idaho Falls campus is a cutting-edge facility. INL is the 6th largest private employer in Idaho. In 2016 it employed 4,211 people in Idaho with an average base salary of $92,660. INL had a significant positive economic impact on the State of Idaho and served over 13,000 visitors in fiscal 2016. Funding came largely from the U.S. Department of Energy with 48% of the total funding source, 19% from the National Nuclear Security Administration, and the remaining portion came from a combination of other agencies. Financial projections for FY-2017 are positive and INL is committed to using taxpayer's dollars wisely. Great science and technology can only be done with world class infrastructure and INL is committed to that priority. There are many labs that do basic science and some that do applied science; INL does the whole spectrum. Three pillars of simultaneous excellence shape the future of INL as a research, development, demonstration, and deployment national laboratory. Dr. Peters indicated there will be additional investment in the buildings at the Integrated Research Center (Idaho Falls campus) and also in two new buildings that are being built in conjunction with the State Board of Education for a Cybercore Integration Center and the Collaborative Computing Center. Both of these areas are growing and more space is needed for the researchers as well as research facilities.

These facilities would have a positive economic benefit to the area.While the INL has a growing workforce, there is also a segment that is aging. They will need to work closely with the educational institutions to develop the next generation of talent. New plans for training these and other types of unique talent are going to require INL to work with educators to accomplish this task. INL is focusing on bringing more interns, graduate students, and postdoctoral students to do research at their facility. Dr. Peters explained that INL is collaborating with partnerships in four areas: nuclear energy competitiveness and leadership; integrated nuclear fuel cycle solutions; a regional clean energy system; and cyber and physical security organized around the Cybercore Integration Center. INL's vision is to change the world's energy future and secure their critical infrastructure. Their mission is to discover, demonstrate, and secure innovative nuclear energy solutions, other clean energy options, and develop critical infrastructure. INL has a strategic plan to help them reach their goals. Representative Anderson questioned what contributed to the increase in salaries from $88,000 to over $92,000. Dr. Peters stated that the increase came from hiring new support people, scientists, and engineers and from raising salaries to be competitive. Representative Anderson asked why INL was building two new buildings on the campus in conjunction with the State Board of Education. Dr. Peters responded that INL felt the value to INL as well as the State universities and the State was significant. The buildings could be constructed more quickly and for less cost by partnering with the State Board of Education. Representative Anderson inquired about the controversy over bringing radioactive material into INL to test and the result that it has had on INL. Dr. Peters stated that the issue has not been resolved and the Attorney General's position is that until such time as the integrated waste treatment unit is operating, INL will not be allowed to bring in new materials to test. He indicated that this has a direct negative impact on their ability to do new research now and over the long term. Representative Trujillo asked if negotiations would go forward with the opening of the Waste Isolation Pilot Plants (WIPP). Dr. Peters indicated that it only allows treated material that is currently on site to be moved and it will not address issues related to the integrated waste treatment unit. Representative Ruebel inquired about the possibility of INL partnering with institutions of higher learning to provide co-operative programs for students in Idaho as well as other states. Dr. Peters indicated that INL would be excited to have a co-op program in conjunction with the universities. Senator Winder asked what the role of the universities is in the building program and what are the benefits. Dr. Peters stated that Boise State University, Idaho State University, and the University of Idaho all have their own computing capabilities on campus, but being able to use INL high performance computing would be a great benefit. By bringing in the next generation of computers it would allow Idaho universities to stay on the edge of technology. Collaboration space could be shared and the ability to attract great people would be an added benefit. An example is the area of cyber security, where INL would be able to understand the problems from conception to testing. INL and the universities each have unique capabilities that complement each other and are exciting for the next generation of students.

JOINT SENATE STATE AFFAIRS COMMITTEE HOUSE ENVIRONMENT, ENERGY & TECHNOLOGY COMMITTEE Monday, January 16, 2017—Minutes—Page 2

Representative Hartgen questioned the ownership, plans for leasing, and whether the money for these new buildings would come from State or private sector money. Dr. Peters responded that the buildings would be bonded by the State and leased back to INL who would continue to pay for the leases for as long as they used them. Representative Hartgen inquired about the estimated cost of the buildings over time. Dr. Peters stated that the total cost was $80 million. Representative Hartgen asked what the low point of growth was. Dr. Peters said it was approximately $800 million dollars and occurred during the federal sequestration time period. Representative Thompson questioned the number of years required to pay for the buildings and if INL would still have access after that time. Dr. Peters responded that payback time would be 15 years and after that time INL would continue to rent the buildings. Representative Thompson asked Dr. Peters if he saw any problems with taking on this project. Dr. Peters stated that he was convinced that there would be growth in both the computing and cyberspace areas and INL would be committed to follow through. Senator Hagedorn voiced some concern with the viability of this project since there are so many entities entering the cyber security world. Dr. Peters responded that INL has significant backing from the Department of Defense (DOD) and most of the resources coming from the DOD is going to this particular area. PRESENTATION: Chairman Raybould introduced Mike McGough, Chief Commercial Officer of NuScale Power and turned the time to him for his presentation. (Attachment 2) Mr. McGough began his presentation with a brief overview of the company. NuScale Power is the first of current U.S. Small Modular Reactors (SMRs) to begin design of commercial Nuclear Power Plants (NPP). A NuScale Power Module (NPM) includes the reactor vessel, steam generators, pressurizer, and containment in an integral package that eliminates reactor coolant pumps and large bore piping. Mr. McGough presented a NuScale Technology Overview and explained how to determine whether and how it works. It takes three things to make the big plant design work: conduction, convection, and gravity. These components always work at Nuscale and that is what makes this plant very different from others. NuScale has built all of the components and tested them individually and then tested them as an integrated system. Safety is an important issue to NuScale Power. Mr. McGough discussed issues concerning safety. Safety is measured in nuclear plants by the probability of an event that causes damage to the nuclear core. There are many challenges and opportunities facing NuScale in the upcoming years. Mr. McGough discussed what is needed for NuScale to be successful in Idaho and what it will mean for our State. Chairman Raybould introduced Ted Rampton, Chief Governmental Affairs Officer and Project Lead, from the Utah Associated Municipal Power Systems (UAMPS). (Attachment 3) PRESENTATION: Mr. Rampton UAMPS is an energy services inter-local agency created in 1980 under Utah statutes. It is comprised of 47 members in seven western states. It is a project-based, energy services related, non-profit organization. Several types of energy resources are used in their projects, the largest being hydro energy at 28%. The future of coal fueled generation, the future of power markets, and the future of natural gas fueled generation are all concerns of UAMPS.

JOINT SENATE STATE AFFAIRS COMMITTEE HOUSE ENVIRONMENT, ENERGY & TECHNOLOGY COMMITTEE Monday, January 16, 2017—Minutes—Page 3

Mr. Rampton discussed the Carbon Free Power Project (CFPP) and the approach UAMPS plans to take to ultimately construct and operate the project. He indicated that they would take a three pronged approach on how to de-risk UAMPS' exposure to carbon regulation. It would include investigation of nuclear small modular reactors, energy efficiency, and distributed generation (rooftop solar). Mr. Rampton stated that UAMPS was aware of past and present environmental regulatory pressures and has a plan to deal with current environmental regulations. In conclusion, Mr. Rampton stated that there is a need for CFPP as a reliable replacement for carbon dioxide emitting generation sources. NuScale's technology has promise to be a cost competitive resource with reduced exposure to future environmental regulations. There are currently four acceptable sites being considered for location of the CFPP project. The next steps in the selection process are to continue engagement with the Shoshone-Bannock Tribes and various Idaho State governmental offices. There would be significant regional economic development potential upon approval of CFPP. The future schedule would include: plant licensing between 2017-2022; site mobilization, construction, and startup during 2020-2026; and commercial operations beginning in 2026. Representative Hartgen voiced concern that modules were being immersed in underground water because of the danger to acquifers. Mr. McGough stated that the NuScale plan has a "dry cooling" technique which uses electrical driven fans as opposed to using water for cooling. It is currently being evaluated for effectiveness. Mr. Rampton commented that a cooling study is being conducted that would minimize the amount of water drawn from the eastern Snack River aquifer. Senator Hagedorn asked what kind of technology capabilities were being built into their plan for long term use. Mr. McGough said that NuScale's strategy is to use standard water reactor coolant and standard enriched uranium oxide fuel. Since the regulator is familiar with both, those two things have been removed from serious questioning because they have a lot of history with other technologies. NuScale believes it will add 10 years per variable to the licensing scope once those plants reach some stage of design maturation. Representative Raybould thanked the speakers for their participation and the information they provided. ADJOURNED:

Representative Raybould adjourned the meeting at 2:45 p.m.

___________________________ Senator Siddoway

_________________________________ Twyla Melton, Secretary

Chairman _________________________________ Sharon Pennington, Assistant Secretary

JOINT SENATE STATE AFFAIRS COMMITTEE HOUSE ENVIRONMENT, ENERGY & TECHNOLOGY COMMITTEE Monday, January 16, 2017—Minutes—Page 4

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

as Franklin County Prosecutor; Donnette Fisher, 455 Fed.Appx. 541 This case was not selected for publication in the Federal Reporter. Not for Publication in West's Federal Reporter. See Fed. Rule of Appellate Procedure 32.1 generally governing citation of judicial decisions issued on or after Jan. 1, 2007. See also Sixth Circuit Rule 28. (Find CTA6 Rule 28) United States Court of Appeals, Sixth Circuit. 84 VIDEO/NEWSSTAND, INC., dba 84 Video/ Newsstand; Vine Street News, Inc., dba Adult Mart; NU Philly Video/News, Inc.; Mile, Inc., dba Lion's Den; American Pride, Inc., dba Lion's Den; Midwest Pride II, Inc., dba Lion's Den; Entertainment U.S.A. of Cleveland, Inc., dba Christie's Cabaret; Gold Restaurant, Inc., dba Gold Horse; Donna and Bato, LLC, dba Expressions; Calpal, LLC, dba Dreamgirls; NL Corp Inc., dba Diamonds Cabaret; Buckeye Association of Club Executives, Inc., Plaintiffs–Appellants, v. Thomas SARTINI, in his official capacity as Ashtabula County Prosecutor; Ross Cirincione, in his official capacity as Law Director of the City of Bedford Heights; David A. Lambros, in his official capacity as Law Director of the City of Brookpark; Robert Triozzi, in his official capacity as Law Director of the City of Cleveland; William Mason, in his official capacity as Cuyahoga County Prosecutor; Tony Geiger, in his official capacity as Law Director for the City of Lima; Jeurgen Waldick, in his official capacity as Allen County Prosecutor; Mike Minniear, in his official capacity as Law Director for the City of Milford; Robin Piper, in his official capacity as Butler County Prosecutor; Matthew E. Crall, in his official capacity as Law Director of the City of Bucyrus; Stanley E. Flegm, in his official capacity as Crawford County Prosecutor; David Kiger, in his official capacity as Law Director of the City of Jeffersonville; David B. Bender, in his official capacity as Fayette County Prosecutor; Richard C. Pfeiffer, Jr., in his official capacity as Columbus City Attorney; Ron O'Brien, in his official capacity

in his official capacity as Law Director for the City of Franklin; Rachel A. Hutzel, in his official capacity as Warren County Prosecutor; Daniel G. Padden, in his official capacity as Guernsey County Prosecutor; David A. Hackenberg, in his official capacity as Law Director of the City of Findlay; Mark C. Miller, in his official capacity as Hancock County Prosecutor; Joseph T. Deters, in his official capacity as Hamilton County Prosecutor; Joseph R. Klammer, in his official capacity as Law Director of the City of Eastlake; Charles E. Coulson, in his official capacity as Lake County Prosecutor; Richard S. Bindley, in his official capacity as Law Director of the City of Heath; Douglas Sassen, in his official capacity as Law Director for the City of Newark; Kenneth W. Oswalt, in his official capacity as Licking County Prosecutor; John T. Madigan, in his official capacity as Law Director of the City of Toledo; Paul S. Goldberg, in his official capacity as Law Director for the City of Oregon; Julia R. Bates, in her official capacity as Lucas County Prosecutor; Iris Torres Guglucello, in her official capacity as Law Director of the City of Youngstown; Paul J. Gains, in his official capacity as Mahoning County Prosecutor; Kenneth Fisher, in his official capacity as Law Director of the City of Brunswick; Dean Holman, in his official capacity as Medina County Prosecutor; Patrick Bonfield, in his official capacity as Law Director of the City of Dayton; Lori E. Kirkwood, in her official capacity as Law Director for the City of West Carrollton; Mathew Heck, in his official capacity as Montgomery County Prosecutor; Charles Howland, in his official capacity as Morrow County Prosecutor; Dave Remy, in his official capacity as Law Director for the City of Mansfield; James Mayer, in his official capacity as Richland County Prosecutor; Toni Eddy, in his official capacity as Law Director of the City of Chillicothe; Michael M. Ater, in his official capacity as Ross County Prosecutor; Andrew L. Zumbar, in his official capacity as Law Director of the City of Alliance; Joseph Martuccio, in his official capacity as Law Director of the City of Canton; John Ferrero, in his official capacity as Stark County Prosecutor; Max Rothal, in his official capacity as

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

1

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

Law Director for the City of Akron; Penelope Taylor, in her official capacity as Law Director for the City of Tallmadge; Sherri Bevan Walsh, in her official capacity as Summit County Prosecutor; Joseph T. Dull, in his official capacity as Law Director for the City of Niles; Dennis Watkins, in his official capacity as Trumbull County Prosecutor; Mike Johnson, in his official capacity as Law Director for the City of New Philadelphia; Amanda K. Spies, in her official capacity as Tuscarawas County Prosecutor; Russ Leffler, in his official capacity as Huron County Prosecutor; Derek Diveine, in his official capacity as Seneca County Prosecutor; Terry S. Shilling, in his official capacity as Law Director for the City of Elyria; Dennis Will, in his official capacity as Lorain County Prosecutor; Martin Frantz, in his official capacity as Wayne County Prosecutor; Scott

West Headnotes (5) [1]

Public Amusement and Entertainment Sexually Oriented Entertainment Ohio statute regulating the operation of sexually oriented businesses, which required that sexually oriented businesses close for six hours each day, did not violate First Amendment; legislature properly relied on the secondary-effects evidence before it, and such evidence was sufficient to support the hours–of–operation restriction, the restriction was unrelated to the suppression of speech, and posed only an incidental burden on First Amendment freedoms that was no greater than is essential to further the government interest in substantially reducing secondary effects. U.S.C.A. Const.Amend. 1; ; Ohio R.C. § 2907.40.

Hillis, in his official capacity as Law Director for the City of Zanesville; D. Michael Haddox, in his official capacity as Muskingum County Prosecutor; Stephen A. Schumaker, in his official capacity as Clark County Prosecutor; Neal M. Jamison; Peter M. Kostoff, Law Director, Defendants–Appellees, The State of Ohio, Defendant–Intervenor. No. 09–3920. | Sept. 7, 2011.

7 Cases that cite this headnote [2]

Constitutional Law Sexually oriented businesses Public Amusement and Entertainment Sexually Oriented Entertainment

Synopsis Background: Ohio sexually oriented businesses brought action against state and law directors and county prosecutors for cities, villages, and counties throughout Ohio, challenging certain regulations of their businesses as violative of the First Amendment. The United States District Court for the Northern District of Ohio granted summary judgment to defendants, and businesses appealed.

Definition of “adult bookstore” in Ohio statute regulating the operation of sexually oriented businesses was not unconstitutionally overbroad on its face in violation of First Amendment; definition included not just businesses that had a “significant or substantial portion of [their] stock in trade or inventory in” adult materials, but also those that “maintain[ ] a substantial section of [their] sales or display space for” such articles. U.S.C.A. Const.Amend. 1; Ohio R.C. § 2907.40(A)(1) .

[Holding:] The Court of Appeals, Helene N. White, Circuit Judge, held that statute did not violate First Amendment.

Affirmed.

Constitutional Law Hours of operation

Cases that cite this headnote [3]

Constitutional Law Sexually oriented businesses Public Amusement and Entertainment

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

2

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

Sexually Oriented Entertainment Definition of “adult cabaret” in Ohio statute regulating the operation of sexually oriented businesses was not unconstitutionally overbroad on its face in violation of First Amendment; under statutory definition, it was not enough that a venue regularly feature entertainment including nudity or sexual activity in the sense that they presented such material recurrently, rather, it had to be presented “consistent[ly]” and such entertainment had to constitute a substantial proportion of the venue's overall offerings. U.S.C.A. Const.Amend. 1; Ohio R.C. § 2907.40(A)(1) . Cases that cite this headnote [4]

Constitutional Law Sexually oriented businesses

Cases that cite this headnote

*544

Before: KETHLEDGE and WHITE, Circuit

Judges; and BECKWITH * , Senior District Judge. Opinion HELENE N. WHITE, Circuit Judge. **1 Plaintiffs 84 Video/Newsstand, Inc. et al. (Plaintiffs) appeal the district court's grant of summary judgment to Defendants Thomas Sartini, et al., (Defendants), law directors and county prosecutors for cities, villages, and counties throughout Ohio, and Defendant–Intervenor State of Ohio, in this action challenging certain regulations of sexually oriented businesses in Ohio as violative of the First Amendment. We AFFIRM.

Public Amusement and Entertainment Dancing and other performances No-touch provision of Ohio statute regulating the operation of sexually oriented businesses, which prohibited entertainers who were nude or semi-nude from touching each other during the course of a performance, was not unconstitutionally overbroad on its face in violation of First Amendment. U.S.C.A. Const.Amend. 1; Ohio R.C. § 2907.40(C)(2) . Cases that cite this headnote [5]

Public Amusement and Entertainment Sexually Oriented Entertainment Public Amusement and Entertainment Motion pictures, videos and games Hours-of-operation restriction in Ohio statute regulating the operation of sexually oriented businesses applied to adult bookstores and adult video stores; statute defined sexually oriented businesses to mean an adult bookstore, adult video store, adult cabaret, adult motion picture theater, sexual device shop, or sexual encounter center, but did not include a business solely by reason of its showing, selling, or renting materials that may depict sex. Ohio R.C. § 2907.40(A)(15).

I On May 16, 2007, the Ohio General Assembly adopted Substitute Senate Bill 16 (“S.B. 16”), codified at Ohio Rev.Code Ann. § 2907.40 (West 2010) (“the Law” or “§ 2907.40”), to regulate the operation of sexually oriented businesses. Section 2907.40 imposes two substantive restrictions on sexually oriented businesses. First, § 2907.40(B) limits hours of operation: No sexually oriented business shall be or remain open for business between 12:00 midnight and 6:00 a.m. on any day, except that a sexually oriented business that holds a liquor permit ... may remain open until the hour specified in that permit if it does not conduct, offer, or allow sexually oriented entertainment activity in which the performers appear nude. Ohio Rev.Code Ann. § 2907.40(B). 1 Second, § 2907.40(C) adopts a so-called “no-touch” provision, limiting physical contact with and between nude or semi-nude performers:

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

3

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

(1) No patron who is not a member of the employee's immediate family shall knowingly touch any employee while that employee is nude or seminude or touch the clothing of any employee while that employee is nude or seminude. (2) No employee who regularly appears nude or seminude on the premises of a sexually oriented business, while on the premises of that sexually oriented business and while nude or seminude, shall knowingly touch a patron ... or another employee ... or the clothing of a patron ... or another employee ... or allow a patron ... or another employee ... to touch the employee or the clothing of the employee. Id. § 2907.40(C) (ellipses in subsection (C)(2) refer to exceptions for members of employee's immediate family). Violation of § 2907.40(B) is a first-degree misdemeanor. Id. § 2907.40(D). Violation of § 2907.40(C) is a first-degree misdemeanor if the violation is achieved by touching a “specified anatomical area” 2 or clothing covering such area, and is a fourth-degree *545 misdemeanor if achieved by touching any other part of the body. Id. § 2907.40(E). Subsection (A) of the Law defines relevant terms, including “sexually oriented business”: an adult bookstore, adult video store, adult cabaret, adult motion picture theater, sexual device shop, or sexual encounter center, but does not include a business solely by reason of its showing, selling, or renting materials that may depict sex. Id. § 2907.40(A)(15). Each individual type of sexually oriented business is also defined. In particular, “adult bookstore” or “adult video store” means a commercial establishment that has as a significant or substantial portion of its stock in trade or inventory in, derives a significant or substantial portion of its revenues from, devotes a significant or substantial portion of its interior business or advertising to, or maintains a substantial section of its sales or display space

for the sale or rental, for any form of consideration, of books, magazines, periodicals, or other printed matter, or photographs, films, motion pictures, video cassettes, compact discs, slides, or other visual representations, that are characterized by their emphasis upon the exhibition or description of specified sexual activities or specified anatomical areas. 3 **2 Id. § 2907.40(A)(1). Section 2907.40(A)(2) originally provided its own definition of “adult cabaret.” It was later amended to replace the original definition with a reference to the definition contained in § 2907.39. That section provides: “Adult cabaret” means a nightclub, bar, juice bar, restaurant, bottle club, or similar commercial establishment, whether or not alcoholic beverages are served, that regularly features any of the following: (a) Persons who appear in a state of nudity or seminudity; (b) Live performances that are characterized by the exposure of specified anatomical areas or specified sexual activities; (c) Films, motion pictures, video cassettes, slides, or other photographic reproductions that are distinguished or characterized by their emphasis upon the exhibition or description of specified sexual activities or specified anatomical areas. Id. § 2907.39(A)(3). The stated purpose of S.B. 16 was to address the adverse secondary effects of sexually oriented businesses. The bill included legislative findings that: [s]exually oriented businesses, as a category of commercial uses, are associated with a wide variety of adverse secondary effects including, but not limited to lewdness, public indecency, prostitution, potential spread of disease, illicit drug use and drug trafficking, personal and

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

4

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

property crimes, negative impacts on surrounding properties, blight, litter, and sexual assault and exploitation.

Procedure 5.1, because the case involves a question of Ohio constitutional law. The State of Ohio intervened as a defendant on October 26, 2007.

S.B. 16, 127th Gen. Assem., Reg. Sess. (Ohio 2007), § 3768.03(B)(1). 4 Prior to passage of S.B. 16, the House Judiciary Committee of the Ohio General Assembly heard testimony for and against the bill *546 and received considerable documentary evidence regarding the secondary effects of sexually oriented businesses. The Senate State and Local Government and Veterans Affairs Committee also considered the legislation. The Ohio General Assembly relied on a variety of sources, including: presentations providing anecdotal accounts of the adverse secondary effects of sexually oriented businesses; summaries and full texts of studies and reports showing that adult businesses cause secondary effects; a critique of the study of one researcher who concluded that adult businesses do not cause adverse secondary effects; legal opinions from the Sixth Circuit and elsewhere upholding regulations addressing secondary effects of adult businesses; and other testimony in favor of the bill. The Legislature also heard testimony from opponents. On October 17, 2007, the day § 2907.40 went into effect, the twelve Plaintiffs filed suit seeking a temporary restraining order (TRO), preliminary injunction, permanent injunction, and declaratory judgment. The district court described the parties to the suit: Plaintiffs consist of three groups: (1) businesses throughout Ohio that sell adult books, magazines, videos, and DVDs (“bookstore Plaintiffs”); (2) businesses throughout Ohio that present nude or seminude adult performances to patrons (“cabaret Plaintiffs”); and (3) the Buckeye Association of Club Executives (“BACE”), a not-for-profit trade group that promotes and protects the rights of member adult bookstores and cabarets throughout Ohio. **3 Defendants consist of two groups: (1) law directors for cities and villages throughout Ohio in which Plaintiffs and BACE members are located; and (2) county prosecutors for the counties throughout Ohio in which Plaintiffs and BACE members are located. Defendants have the authority to prosecute violations of § 2907.40. Plaintiffs also served the Ohio Attorney General, pursuant to Federal Rule of Civil

The district court denied the motion for a TRO on October 18. The court then held a preliminary injunction hearing. At the hearing, Plaintiffs called five witnesses, including two experts: Dr. Daniel Linz, an expert on secondary-effects studies, who testified that sexually oriented businesses do not cause appreciable secondary effects and critiqued existing studies showing secondary effects; and Dr. Judith Hanna, an expert on exotic dance, who testified regarding the expressive aspects of exotic dance. Plaintiffs also offered testimony from individuals with experience operating sexually oriented businesses and submitted the declaration of Dr. Lance Freeman, in which he summarized the findings of a study concluding that the presence of an adult bookstore or adult cabaret in proximity to residential property did not depress property values. Defendants called three witnesses who testified that sexually oriented businesses cause secondary effects, including crime: Julie Taylor Schmatz, a former exotic dancer; Louis Gentile, a private investigator; and Dr. Richard McCleary, an expert on secondary-effects studies. The court denied Plaintiffs' motion for a preliminary injunction on August 8, 2008. 5 *547 While the district court proceedings were ongoing, the Ohio General Assembly amended the definition of “adult cabaret” contained in § 2907.40(A)(2), replacing the original definition with a provision incorporating by reference the definition of “adult cabaret” found in § 2907.39 of the Revised Code. Sub. S.B. No. 183, 2008 Ohio Laws 101. Plaintiffs moved for a TRO or preliminary injunction to bar implementation of this amendment. The court denied that motion in an order dated January 5, 2009. Defendants then moved for summary judgment on all Plaintiffs' claims, which the court granted on June 22, 2009. Plaintiffs timely appealed.

II

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

5

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

This court reviews a district court's grant of summary judgment de novo. Binay v. Bettendorf, 601 F.3d 640, 646 (6th Cir.2010). Summary judgment is appropriate “if the movant shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law.” Fed.R.Civ.P. 56(a); see also Binay, 601 F.3d at 646. This court must draw all reasonable inferences and view all evidence in favor of the non-moving party. Binay, 601 F.3d at 646; Wuliger v. Manufacturers Life Ins. Co., 567 F.3d 787, 792 (6th Cir.2009). **4 Plaintiffs raise a number of issues on appeal, arguing that: S.B. 16 was based on insufficient evidence that sexually oriented businesses cause certain secondary effects and is not narrowly tailored; the law's definitions of “adult bookstore or video store” and “adult cabaret” are unconstitutionally overbroad; the no-touching restriction independently violates the First Amendment; and adult bookstores and adult video stores are actually excluded from regulation by the plain language of the law. We address these arguments in order.

III Plaintiffs first assert that Ohio Rev.Code Ann. § 2907.40 violates the First Amendment because the evidence relied on by the Ohio General Assembly in passing the statute was insufficient to survive intermediate scrutiny. They further argue that the law is not narrowly tailored because it suppresses a substantial amount of speech.

A. Applicable Law It is beyond doubt that “[n]ude dancing is a form of expressive conduct protected by the First Amendment.” Sensations, Inc. v. City of Grand Rapids, 526 F.3d 291, 298 (6th Cir.2008); Deja Vu of Nashville, Inc. v. Metro. Gov't of Nashville and Davidson Cnty., 274 F.3d 377, 391 (6th Cir.2001); see also City of Erie v. Pap's A.M., 529 U.S. 277, 289, 120 S.Ct. 1382, 146 L.Ed.2d 265 (2000) (plurality opinion) (“Nude dancing ... is expressive conduct, although we think that it falls only within the outer ambit of the First Amendment's protection.”). Other forms of erotic entertainment are similarly protected by the First Amendment. City of Renton v. Playtime Theatres, Inc., 475 U.S. 41, 46–47, 106 S.Ct. 925, 89 L.Ed.2d 29

(1986) (affirming that adult theaters are protected by the First Amendment); Richland Bookmart, Inc. v. Knox Cnty. (Richland Bookmart II), 555 F.3d 512, 520 (6th Cir.2009) (“[S]exually explicit but non-obscene speech, such as adult publications and adult videos” are within a “protected category of speech”). We treat regulations targeting the “secondary effects” of sexually oriented businesses as content-neutral, and assess them under intermediate scrutiny. 6 Richland *548 Bookmart II, 555 F.3d at 521 & n. 2; 729, Inc. v. Kenton Cnty. Fiscal Court, 515 F.3d 485, 490–91 (6th Cir.2008). Where a law specifically targets the secondary effects of adult businesses, this court applies the test set out in United States v. O'Brien, 391 U.S. 367, 88 S.Ct. 1673, 20 L.Ed.2d 672 (1968), as interpreted by City of Renton v. Playtime Theatres, Inc., 475 U.S. 41, 106 S.Ct. 925, 89 L.Ed.2d 29 (1986), and City of Los Angeles v. Alameda Books, Inc., 535 U.S. 425, 122 S.Ct. 1728, 152 L.Ed.2d 670 (2002). Richland Bookmart II, 555 F.3d at 523–24 (“[W]e find it prudent to conduct our analysis in terms set forth in Renton and Alameda Books—or, equivalently, to apply the O'Brien test, incorporating evidentiary standards articulated in Renton and its progeny.”). Under the O'Brien test, courts must determine whether the legislature enacted a challenged law “(1) within its constitutional power, (2) to further a substantial governmental interest that is (3) unrelated to the suppression of speech, and whether (4) the provisions pose only an ‘incidental burden on First Amendment freedoms that is no greater than is essential to further the government interest.’ ” Sensations, 526 F.3d at 298. On the second prong of the O'Brien test, this Circuit further applies the plurality opinion in Alameda Books, which announced a three-step burden-shifting analysis applicable to secondary-effects cases. Richland Bookmart II, 555 F.3d at 525. Under Alameda Books, **5 [first,] a municipality may rely on any evidence that is ‘reasonably believed to be relevant’ for demonstrating a connection between speech and a substantial, independent government interest. This is not to say that a municipality can get away with shoddy data or reasoning. The municipality's evidence must fairly support the

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

6

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

municipality's rationale for its ordinance. [Second, if] plaintiffs fail to cast direct doubt on this rationale, either by demonstrating that the municipality's evidence does not support its rationale or by furnishing evidence that disputes the municipality's factual findings, the municipality meets the standard set forth in Renton. [Third, if] plaintiffs succeed in casting doubt on a municipality's rationale in either manner, the burden shifts back to the municipality to supplement the record with evidence renewing support for a theory that justifies its ordinance. 535 U.S. at 438–39, 122 S.Ct. 1728 (plurality opinion) (citations omitted); see also Sensations, 526 F.3d at 297 n. 5. Further, the legislature's evidentiary burden is slight: [W]e have consistently held that a city must have latitude to experiment, at least at the outset, and that very little evidence is required. As a general matter, courts should not be in the business of second-guessing factbound empirical assessments of city planners.... [The government] is entitled to rely on [its local] knowledge; and if its inferences appear reasonable, we should not say there is no basis for its conclusion. 535 U.S. at 451–52, 122 S.Ct. 1728 (Kennedy, J., concurring) (citations omitted). Indeed, state and local governments “need not conduct their own studies demonstrating that adverse secondary effects result from the operation of sexually oriented businesses or that the measures chosen *549 will ameliorate these effects.” Richland Bookmart II, 555 F.3d at 524; see also Alameda Books, 535 U.S. at 438, 122 S.Ct. 1728 (plurality opinion); id. at 451, 122 S.Ct. 1728 (Kennedy, J., concurring); Renton, 475 U.S. at 51–52, 106 S.Ct. 925. However, “a city may not attack secondary effects indirectly by attacking speech.” Alameda Books, 535 U.S. at 450, 122 S.Ct. 1728 (Kennedy, J., concurring). The regulation must be aimed

at secondary effects, and the government “must advance some basis to show that its regulation has the purpose and effect of suppressing secondary effects, while leaving the quantity and accessibility of speech substantially intact.” Id. at 449, 122 S.Ct. 1728.

B. Section 2907.40 Survives Scrutiny Under the O'Brien Test 1. Passage of § 2907.40 Was Within the Power of the Ohio General Assembly [1] The first question under the O'Brien test is whether enacting § 2907.40 was within the state's constitutional power. The parties do not dispute that the Ohio General Assembly had such power. This court has held that “regulating sexually oriented businesses to reduce negative secondary effects lies within the scope of a [government's] authority under the O'Brien test.” Sensations, 526 F.3d at 298. Therefore, the first prong of the O'Brien test is satisfied.

2. Section 2907.40 Furthers a Substantial Government Interest **6 The second step of the O'Brien test asks whether the legislature enacted the law “to further a substantial government interest.” See Sensations, 526 F.3d at 298. Under the Alameda Books burden-shifting framework, Defendants must first show that the Ohio General Assembly relied on “evidence that is ‘reasonably believed to be relevant’ for demonstrating a connection between speech and a substantial, independent government interest.” 535 U.S. at 438, 122 S.Ct. 1728 (plurality opinion). “It is now recognized that governments have a substantial interest in controlling adverse secondary effects of sexually oriented establishments, which include violent, sexual, and property crimes as well as blight and negative effects on property values.” Richland Bookmart II, 555 F.3d at 524. Thus, our inquiry hinges on the evidence the Ohio General Assembly relied on when it passed § 2907.40. This court has held that a wide variety of sources may form a sufficient evidentiary basis at this stage, including land-use studies, prior judicial opinions, surveys of relevant professionals (such as realestate appraisers), anecdotal testimony, police reports, and other direct and circumstantial evidence. See, e.g., Richland Bookmart II, 555 F.3d at 525; 729, Inc., 515 F.3d at 491–92; J.L. Spoons, Inc. v. Dragani, 538 F.3d 379, 381

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

7

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

(6th Cir.2008); see also City of Erie v. Pap's A.M., 529 U.S. 277, 296, 120 S.Ct. 1382, 146 L.Ed.2d 265 (2000) (noting that the City of Erie was permitted to rely on prior judicial decisions and that “it was reasonable for Erie to conclude that ... nude dancing was likely to produce the same secondary effects” found in the prior decisions). In the instant case, the Ohio General Assembly gathered a range of evidence demonstrating that sexually oriented businesses cause harmful secondary effects, including a combination of anecdotal evidence from live testimony and other submissions, press reports, land-use studies, expert reports, and prior judicial opinions. The evidence consists both of studies and cases from other jurisdictions, and of studies, cases, press reports, and anecdotal evidence from Ohio. This evidence is sufficient to demonstrate a reasonable belief on behalf of the General Assembly that sexually oriented businesses cause negative secondary effects, including certain *550 types of crime, decreased property values, and health risks, and that the proposed statute would address such effects. The burden thus shifts to Plaintiffs to show either that the evidence does not fairly support the General Assembly's rationale for the law—to combat secondary effects—or that the factual findings relied on by the General Assembly were incorrect. Alameda Books, 535 U.S. at 438–39, 122 S.Ct. 1728. The Sixth Circuit has previously upheld regulations similar or identical to those enacted in § 2907.40. See Richland Bookmart II, 555 F.3d at 519 (upholding, without discussion, an hours-of-operation restriction); Entm't Prods., Inc. v. Shelby Cnty., 588 F.3d 372, 393–94 (6th Cir.2009) (upholding, against overbreadth challenge, a no-touching and six-foot buffer-zone requirement between entertainers and customers and between entertainers and other entertainers); Sensations, 526 F.3d at 299 (upholding no-touching rule between performers and audience members and hours-ofoperation restriction); 729, Inc., 515 F.3d at 490–93 (upholding requirement that entertainers must “maintain a minimum distance of five ... feet from ... customers, for a minimum of one ... hour after the entertainer appears semi-nude on the establishment's premises”); Deja Vu of Cincinnati, L.L.C. v. Union Twp. Bd. of Trs., 411 F.3d 777, 789–91 (6th Cir.2005) (en banc) (upholding an hours-ofoperation limitation); Deja Vu of Nashville, Inc. v. Metro. Gov't of Nashville & Davidson Cnty., 274 F.3d 377, 396 (6th Cir.2001) (upholding regulation prohibiting physical contact between customers and entertainers); Richland

Bookmart, Inc. v. Nichols (Richland Bookmart I), 137 F.3d 435, 440–41 (6th Cir.1998) (upholding hours-of-operation restriction); DLS, Inc. v. City of Chattanooga, 107 F.3d 403, 408–13 (6th Cir.1997) (upholding a regulation prohibiting “entertainers from approaching within six feet of customers, employees, or other entertainers during a performance”). **7 Nonetheless, Plaintiffs make two main attacks on the evidence underlying § 2907.40. First, they argue that testimony by their own expert, Dr. Daniel Linz, casts sufficient doubt on the secondary-effects studies relied on by the General Assembly to create a genuine issue of material fact requiring a trial. Second, Plaintiffs argue that the General Assembly's evidence of secondary effects does not support the hours-of-operation restriction in § 2907.40(B).

i. Plaintiffs' Expert Fails to Cast Doubt on the Evidence Relied on by the Ohio General Assembly Plaintiffs argue that studies and analyses conducted by Dr. Linz substantially undermine the evidence relied on by the General Assembly At the preliminary injunction hearing, Linz testified regarding studies he conducted in cities around the country and in Toledo, Cleveland, Columbus, and Dayton, Ohio. He concluded in all these studies that sexually oriented businesses do not increase adverse secondary effects, namely crime, in surrounding areas. He also testified, as to studies conducted by other researchers, that “the methods are either so flawed or the studies so poorly conducted, that they do not, in fact, demonstrate an adverse secondary effect.” Plaintiffs additionally submitted a declaration from Dr. Lance Freeman, accompanied by a study of property values in Ohio he co-authored, concluding that “the presence of an adult bookstore or adult cabaret in proximity to residential property did not lead to a decrease in property values.” Defendants offered testimony by Dr. McCleary. McCleary discussed secondary-effects studies he conducted, and explained that “in every instance, [he] ha[s] *551 been able to corroborate the theory” that sexually oriented businesses cause secondary effects. Based on his own and other researchers' studies, he testified that “it is a scientific fact that sexually-oriented businesses have crime-related secondary effect [sic], that they pose public safety hazards to their immediate environments.” McCleary also directly criticized the validity of the

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

8

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

findings of Linz's study of four Ohio cities, arguing that the underlying data did in fact demonstrate secondary effects. Linz and McCleary primarily disagreed about study methodology, including: the appropriate criteria for judging the validity of secondary-effects studies; the best way to measure the incidence of crime; the proper geographic area surrounding a sexually oriented business within which to measure crime; and the period of time a study must analyze in order to be valid. The evidence proffered by Plaintiffs—testimony and reports of qualified experts—is indeed the type of evidence that may appropriately be used to cast doubt on the Legislature's evidence. See Richland Bookmart II, 555 F.3d at 526 (commenting that plaintiffs' evidence “is of dubious substantive import” because, “[u]nlike most plaintiffs challenging similar regulations, Plaintiffs do not introduce their own expert findings or studies” (citation omitted)); J.L. Spoons, 538 F.3d at 381–82 (describing expert testimony offered by plaintiffs at preliminary injunction hearing). The question, then, is whether the testimony of Drs. Linz and Freeman is sufficient to create a genuine issue of material fact as to the accuracy of the evidence relied on by the Legislature. **8 This court has repeatedly held that governments are not “required to demonstrate empirically that [their] proposed regulations will or are likely to successfully ameliorate adverse secondary effects.” Richland Bookmart II, 555 F.3d at 524. “[E]vidence suggesting that a different conclusion [by the legislature] is also reasonable does not prove that the [government's] findings were impermissible or its rationale unsustainable.” Id. at 527. Under Alameda Books, the touchstone is whether the legislature “reasonably believed [the evidence it relied on] to be reasonable” and whether the evidence “fairly support[s] the [legislature's] rationale” for the law. 535 U.S. at 438, 122 S.Ct. 1728. A mere difference of opinion about the conclusions to be drawn from a body of evidence cannot invalidate the legislature's decision. This court has twice decided secondary-effects cases where Dr. Linz's testimony or reports were discussed on appeal. Sensations, 526 F.3d at 295 (upholding Grand Rapids, Michigan, ordinance regulating sexually oriented businesses); J.L. Spoons, 538 F.3d at 382–83 (upholding Ohio Liquor Commission Rule restricting nude dancing and sexual contact at establishments holding Liquor Control Commission permits). But neither of these cases

gave more than cursory mention to Linz's testimony, and neither specifically discussed whether or to what extent Linz's evidence cast doubt on the evidence relied on by the legislature. Secondary effects cases in other circuits where Linz's testimony was introduced have split regarding the weight to be afforded that evidence. A majority of decisions have held that testimony and studies by Linz were insufficient to invalidate the legislative body's evidence. See, e.g., Imaginary Images, Inc. v. Evans, 612 F.3d 736, 747–48 (4th Cir.2010); Doctor John's v. Wahlen, 542 F.3d 787, 791–93 (10th Cir.2008); Fantasyland Video, Inc. v. Cnty. of San Diego, 505 F.3d 996, 1002 (9th Cir.2007); G.M. Enters., Inc. v. Town of St. Joseph, 350 F.3d 631, 635–36, 640 (7th Cir.2003); see also Galardi v. City of *552 Forest Park, No. 1:09–CV–965–RWS, 2011 WL 111586, at *6 (N.D.Ga. Jan.13, 2011). Plaintiffs rely mainly on two recent Seventh and Tenth Circuit cases that credited Linz's testimony with casting doubt on the evidence supporting the government regulations. Annex Books, Inc. v. City of Indianapolis, 581 F.3d 460, 461–65 (7th Cir.2009) (concluding that a showing that the city's evidence was not germane to the restrictions it enacted in combination with the presentation of a study by Linz raised a genuine issue of material fact and required remand for an evidentiary hearing); Abilene Retail No. 30, Inc. v. Bd. of Comm'rs of Dickinson Cnty., 492 F.3d 1164, 1187 (10th Cir.2007) (Ebel., J., concurring; concurrence joined by full panel as alternative ground of decision) (crediting Dr. Linz's studies and testimony with undermining the county's evidence, both by presenting five studies conducted by Linz that found no adverse secondary effects from sexually oriented businesses, and by offering testimony and a peer-reviewed article “challenging the validity of the County's studies”). **9 In this case, Linz's evidence is of two types: studies Linz conducted finding that sexually oriented businesses do not cause adverse secondary effects, and criticism of the validity of studies by other researchers that do find such effects. Linz's study of four Ohio cities is directly relevant to the central issue in this case and, if accurate, does tend to cast doubt on the Ohio General Assembly's evidence. McCleary's testimony, in turn, casts doubt on Linz's study. We conclude that the Linz evidence in the record before us is insufficient to create a genuine issue of material fact requiring remand. This is for two reasons. First, Linz's testimony and studies fail to cast doubt on the entire

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

9

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

body of evidence relied on by the General Assembly, including those secondary-effects studies not discussed by Linz and the significant quantity of other types of evidence relied on by the Legislature with which Linz does not engage, including prior court decisions, news reports, and anecdotal testimony by law enforcement officials and others. The legislature's evidentiary burden to justify a regulation targeted at secondary effects is slight. Here, Plaintiffs' testimony and exhibits do not show the body of that evidence as a whole to be so questionable as to undermine support for the restrictions in § 2907.40. Second, to the extent that the Linz evidence does “dispute [ ] the [government's] factual findings” at step two of the Alameda Books burden shifting test, 535 U.S. at 438–39, 122 S.Ct. 1728, the testimony and reports by Dr. McCleary introduced at the preliminary injunction hearing were sufficient “to supplement the record with evidence renewing support for a theory that justifies” the Ohio law in satisfaction of Alameda Books step three. Id. at 439, 122 S.Ct. 1728. The Linz and McCleary testimony amounts to a battle of experts who primarily disagree about study methodology. The General Assembly had before it studies by McCleary as well as a paper by McCleary critiquing Linz's methodology. The General Assembly was entitled to credit McCleary's findings as providing reasonable support for the restrictions it enacted. Even assuming that Linz's evidence, by itself, casts doubt on the Legislature's evidence, McCleary's testimony renews support for Ohio's theory of secondary effects. Thus, no genuine issue of material fact remains as to whether the Legislature properly relied on the secondary-effects evidence before it.

ii. Whether the Ohio General Assembly's Evidence was Sufficient to Support the Hours–of–Operation Restriction in § 2907.40(B) Plaintiffs also argue that the General Assembly's evidence of secondary effects *553 does not support the hoursof-operation restriction in § 2907.40(B). This court has upheld restrictions on the hours of operations of sexually oriented businesses in several cases. See Richland Bookmart II, 555 F.3d at 519 (upholding, but not discussing, an hours-of-operation restriction); Sensations, 526 F.3d at 294, 299 (upholding prohibition against sexually oriented businesses operating between 2:00 AM and 7:00 AM); Deja Vu of Cincinnati, L.L.C. v. Union Twp. Bd. of Trs., 411 F.3d 777, 789–91 (6th Cir.2005) (en banc) (upholding regulation requiring sexually oriented businesses to close at midnight); Richland Bookmart, Inc.

v. Nichols (Richland Bookmart I), 137 F.3d 435, 438, 440– 41 (6th Cir.1998) (upholding requirement that sexually oriented businesses close between midnight and 8:00 AM Monday through Saturday and all day on Sunday). Those cases gave significant deference to legislative bodies, finding that “ ‘[r]educing crime, open sex and solicitation of sex and preserving the aesthetic and commercial character of the neighborhoods surrounding adult establishments is a substantial government interest.... It is not unreasonable to believe that such regulation of hours [of operation] ... would tend to deter prostitution’ and other negative secondary effects.” Deja Vu of Cincinnati, 411 F.3d at 790 (quoting Richland Bookmart I, 137 F.3d at 440) (alterations in original). **10 In this case, the General Assembly considered evidence specifically relevant to the conclusion that closing sexually oriented businesses during the early morning hours could ameliorate negative secondary effects, including prior court cases upholding such restrictions and testimony at the legislative committee hearing by a police officer with experience investigating crime at sexually oriented businesses and by a former manager of adult-entertainment establishments. 7 To rebut these legislative findings, Dr. Linz testified in district court that one of his own studies had found no support for the theory that sexually oriented businesses attract more crime during the late-night or early-morning hours than at other times. Dr. McCleary, in turn, testified that, although he knows of no studies correlating the severity of secondary effects of adult businesses to their hours of operation, there has been a significant amount of research supporting the more general proposition that “[c]rime risks for any business that's at risk goes up” at night. Most of the evidence before the General Assembly that supports an hours-of-operation restriction discussed the late-night secondary effects of live-entertainment establishments, not adult bookstores and adult video stores. Nonetheless, proponents of S.B. 16 presented the General Assembly with several decisions by courts of appeals upholding such restrictions as applied to adult bookstores and video stores, including Center for Fair Public Policy v. Maricopa County, 336 F.3d 1153 (9th Cir.2003), from the Ninth Circuit and Richland Bookmart I from this circuit. Center for Fair Public Policy, in particular, provides support for the restriction, as it canvassed evidence, including testimony and studies from other jurisdictions, that indicated a link between

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

10

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

secondary effects and the hours of operation of non-live*554 entertainment adult businesses. Additionally, this circuit's opinion in Sensations, Inc. upheld a municipal ordinance limiting the hours of operation of all sexually oriented businesses, including book stores and video stores. 526 F.3d at 294, 298–99. The Ohio General Assembly relied on sufficient evidence to support passage of the law.

3. § 2097.40 is Unrelated to the Suppression of Speech The third prong of the O'Brien test requires this court to determine whether the challenged law is in fact “unrelated to the suppression of speech.” Sensations, 526 F.3d at 298. The parties do not argue that it is not. Targeting the secondary effects of sexually oriented businesses is permissible under O'Brien. See, e.g., Barnes v. Glen Theatre, Inc., 501 U.S. 560, 585–86, 111 S.Ct. 2456, 115 L.Ed.2d 504 (1991) (Souter, J., concurring) (“[O]n its face, the governmental interest in combating prostitution and other criminal activity is not at all inherently related to expression.”). Because the stated and unchallenged purpose of § 2907.40 is to address such secondary effects, the law survives O'Brien's third prong.

4. § 2907.40 Poses Only an Incidental Burden on First Amendment Freedoms that is No Greater than is Essential to Further the Government Interest **11 The fourth prong of the O'Brien test asks whether the restrictions “pose only an ‘incidental burden on First Amendment freedoms that is no greater than is essential to further the government interest.’ ” Sensations, 526 F.3d at 298. Justice Kennedy's concurrence in Alameda Books sharpens this inquiry, requiring that a government “must advance some basis to show that its regulation has the purpose and effect of suppressing secondary effects, while leaving the quantity and accessibility of speech substantially intact.” 535 U.S. at 449, 122 S.Ct. 1728 (Kennedy, J., concurring). In other words, “the necessary rationale for applying intermediate scrutiny is the promise that [regulations] may reduce the costs of secondary effects without substantially reducing speech.” Id. at 450, 122 S.Ct. 1728. Justice Kennedy's concurrence thus requires a proportionality analysis: a government may not seek to reduce secondary effects by reducing speech on a one-to-one basis. As he put it, “[i]t is true that cutting adult speech in half would probably reduce secondary effects proportionately. But again, a promised proportional reduction does not suffice. Content-based

taxes could achieve that, yet these are impermissible.” Id. at 451, 122 S.Ct. 1728. This court has held that Kennedy's discussion on this point is binding. 729, Inc., 515 F.3d at 491 (“Although the Alameda Books plurality did not discuss [this] requirement, Justice Kennedy expressly said that consideration of this issue was required for his concurrence in the judgment. Justice Kennedy's opinion binds us on this point.”). Plaintiffs assert that the hours-of-operation restriction fails under this proportionality analysis because it directly reduces a substantial amount of speech. We disagree. Plaintiffs' argument proceeds in two steps: First, they posit that there is insufficient evidence that closing sexually oriented businesses between midnight and 6:00 AM would significantly curtail adverse secondary effects. Second, they argue that they have demonstrated that the hoursof-operation restriction will cause a “massive reduction in speech.” Therefore, they argue, § 2907.40 seeks to reduce secondary effects by reducing the amount of speech— measured in hours of operation of the adult businesses— in direct proportion to any secondary effects ameliorated. *555 On the first point, the General Assembly did consider some evidence, including prior court cases, reports and studies from other jurisdictions and anecdotal testimony, that secondary effects of adult businesses are greater during the late night hours. As discussed above, this evidence provides sufficient basis, even if not overwhelming, to conclude that sexually oriented businesses cause secondary effects late at night that are different in severity or scope from those caused at other times of day. This conclusion is significantly bolstered by this Circuit's prior cases upholding hours-of-operation restrictions. See Richland Bookmart II, 555 F.3d at 519; Sensations, 526 F.3d at 294, 299; Deja Vu of Cincinnati, 411 F.3d at 789–91; Richland Bookmart I, 137 F.3d at 438, 440–41. **12 On the second point, Plaintiffs offered testimony at the preliminary injunction hearing about the quantity of speech that the hours-of-operation provision would suppress, arguing that it amounts to a “massive reduction.” They measured this in terms of economic effect of the law by explaining that prior to passage of § 2907.40, adult bookstores in Ohio did a significant amount of business, measured in millions of dollars per year, during the hours they will now be required to remain closed. Plaintiffs also offered evidence that “juice bars”—

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

11

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

establishments that provides nude dancing but do not sell alcoholic beverages—generate the majority of their revenues between 11:00 P.M. to 4:00 A.M., and so may become unprofitable if subject to the law.

526 F.3d at 294, 299 (upholding prohibition on sexually oriented businesses operating between 2:00 AM and 7:00 AM). Under these precedents, § 2907.40 does leave the quantity of speech substantially intact.

Evidence of a regulation's economic impact is not directly relevant to the First Amendment inquiry. See Deja Vu of Nashville, 274 F.3d at 397 (“[T]he relevant inquiry is not whether the Ordinance will cause any economic impact on the sexually oriented businesses. Although ... compliance with the Ordinance will cut into the plaintiffs' profits, the plaintiffs have failed to introduce any evidence showing that they will not have a reasonable opportunity to operate their establishments.”); DLS, Inc., 107 F.3d at 413 (“[T]he inquiry for First Amendment purposes is not concerned with economic impact. In our view, the First Amendment requires only that [the city] refrain from effectively denying respondents a reasonable opportunity to open and operate an adult theater within the city.” (quoting Renton, 475 U.S. at 54, 106 S.Ct. 925)). Plaintiffs assert, however, that the evidence of lost sales during the early morning hours is indicative of the quantity of speech suppressed, and is not offered to prove loss of profits, per se. We assess this claim under Justice Kennedy's proportionality analysis laid out

**13 Plaintiffs argue, however, that once Justice Kennedy's proportionality analysis is correctly applied, the hours-of-operation restriction cannot be sustained because the reduction in secondary effects is small while the diminution in the availability of speech is large. With regard to the adult bookstores, this argument fails. To the extent that the secondary effects of sexually oriented businesses, including adult bookstores, are more serious late at night, the closure of those businesses between midnight and 6:00 AM addresses those effects while leaving ample time—18 hours per day-when that speech remains available. Plaintiffs have not established that the hours-of-operation restriction will block a significant amount of access to speech. Individuals seeking to take advantage of these stores may do so during their remaining open hours, when the asserted secondary effects are less severe. Further, at oral argument Plaintiffs' counsel conceded that none of the adult bookstores in Ohio closed down as a result of the hours of operation restriction, indicating that the effect of the law is not as dire as feared. Plaintiffs fail to distinguish the instant case from prior Sixth Circuit cases upholding hours-ofoperation restrictions, except insofar as they argue that proportionality analysis requires a different outcome on the record presented here. The record does not support such a finding.

in Alameda Books. 8 729, Inc., 515 F.3d at 491. Under the 729, Inc. approach, this *556 court must ask whether the restriction leaves the “quantity and accessibility of protected speech substantially intact,” id. at 492 (quoting Alameda Books, 535 U.S. at 449–50, 122 S.Ct. 1728), and must ensure that it “is reasonably likely to cause a substantial reduction in secondary effects while reducing speech very little.” Id. at 493, 122 S.Ct. 1728 (quoting Alameda Books, 535 U.S. at 451, 122 S.Ct. 1728 (Kennedy, J., concurring)). The hours-of-operation restriction requires that sexually oriented businesses close for six hours each day (42 hours per week), leaving 18 hours per day (126 hours per week) when the businesses may remain open. This is less restrictive than the hours-of-operation restriction upheld in Richland Bookmart I, which required closure between midnight and 8:00 AM Monday through Saturday and all day on Sunday, amounting to 72 hours per week (leaving 96 hours per week when the businesses could remain open). 137 F.3d at 438, 440–41. Similarly, in Deja Vu of Cincinnati, this court upheld a regulation allowing such businesses to remain open for just “twelve hours a day, six days a week.” 411 F.3d at 791; see also Sensations,

Proportionality analysis also fails to invalidate the law with respect to adult cabarets. Plaintiffs argue that “[t]he statute destroys the juice bars' business model by forcing them to close during the very hours that the vast majority of their patrons attend the constitutionally protected performances presented at them.” It is true that the premise of the law must not be that the affected businesses will close. See Alameda Books, 535 U.S. at 450–51, 122 S.Ct. 1728 (Kennedy, J., concurring). On the other hand, the First Amendment is not concerned with economic effects. See City of Renton v. Playtime Theatres, Inc., 475 U.S. 41, 54, 106 S.Ct. 925, 89 L.Ed.2d 29 (1986); Young v. American Mini Theatres, Inc., 427 U.S. 50, 78, 96 S.Ct. 2440, 49 L.Ed.2d 310 (1976) (Powell, J., concurring). Juice bars are able to offer nude entertainment for 18 hours per day or, alternatively, to offer non-nude entertainment for 24 hours per day. It is not clear that if following passage of

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

12

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

§ 2907.40 juice bars can no longer successfully market their business model, the *557 overall quantity of erotic speech will have diminished. See DLS, Inc., 107 F.3d at 413 (“[W]e consider the economic effects of the ordinance in the aggregate, not at the individual level; if the ordinance were intended to destroy the market for adult cabarets, it might run afoul of the First Amendment, but not if it merely has adverse effects on the individual theater.”); see also Deja Vu of Nashville, 274 F.3d at 397 (“Although we do not doubt that compliance with the Ordinance will cut into the plaintiffs' profits, the plaintiffs have failed to introduce any evidence showing that they will not have a reasonable opportunity to operate their establishments.”). If juice bars close, erotic dancing will be able to shift more heavily toward alcohol-serving establishments providing nude and semi-nude entertainment until midnight and scantily clad entertainment after midnight. Thus, under the proportionality analysis, the restriction “is reasonably likely to cause a substantial reduction in secondary effects while reducing speech very little.” 729, Inc., 515 F.3d at 493. This court has upheld hours-of-operation restrictions on juice bars in cases decided after Alameda Books, and Plaintiffs do not successfully distinguish the present case. See Deja Vu of Cincinnati, 411 F.3d at 789–91.

IV A **14 Plaintiffs argue that the definitions of “adult bookstore [and] adult video store” and “adult cabaret” contained in § 2907.40(A)(1) and (A)(2) are unconstitutionally overbroad on their face. A law is unconstitutionally overbroad, and thus must be invalidated, when it “prohibits a substantial amount of protected speech both in an absolute sense and relative to the statute's plainly legitimate sweep.” Entm't Prods., Inc. v. Shelby Cnty., 588 F.3d 372, 379 (6th Cir.2009) (quoting Connection Distrib. Co. v. Holder, 557 F.3d 321, 336 (6th Cir.2009) and United States v. Williams, 553 U.S. 285, 293, 128 S.Ct. 1830, 170 L.Ed.2d 650 (2008)). “Overbreadth doctrine exists to allay the concern that the threat of enforcement of an overbroad law may deter or chill constitutionally protected speech.” J.L. Spoons, Inc. v. Dragani, 538 F.3d 379, 383 (6th Cir.2008). Facial invalidation of a law under the overbreadth doctrine should be sparing and careful however, Richland Bookmart, Inc. v. Knox Cnty.

(Richland Bookmart II), 555 F.3d 512, 522 (6th Cir.2009), because “there are substantial social costs created by the overbreadth doctrine when it blocks application of a law to constitutionally unprotected speech, or especially to constitutionally unprotected conduct.” Virginia v. Hicks, 539 U.S. 113, 120, 123 S.Ct. 2191, 156 L.Ed.2d 148 (2003). For this reason, “the Supreme Court has ‘vigorously enforced the requirement that a statute's overbreadth be substantial.’ ” Entm't Prods., 588 F.3d at 379 (quoting Williams, 553 U.S. at 292, 128 S.Ct. 1830). Thus, for a law to be judged facially overbroad, Plaintiffs must “demonstrate from the text of [the statute] and from actual fact that a substantial number of instances exist in which the law cannot be applied constitutionally.” Id. (quoting N.Y. State Club Ass'n v. City of New York, 487 U.S. 1, 14, 108 S.Ct. 2225, 101 L.Ed.2d 1 (1988)). Although this standard sets a high bar, this Circuit “has not shied away from invalidating a regulatory scheme in its entirety when the threat of impermissible applications and the consequent chilling effect unambiguously warranted this remedy.” Id. at 380. Such invalidation of a law on overbreadth grounds is appropriate when the language of the law is not “readily susceptible to a limiting construction.” Odle v. Decatur Cnty., 421 F.3d 386, 396 (6th Cir.2005). If a law's language lends *558 itself to an interpretation that avoids unconstitutional applications, it may be upheld. However, this court will “not rewrite statutes to create constitutionality,” Triplett Grille, Inc. v. City of Akron, 40 F.3d 129, 136 (6th Cir.1994), and so will not “accept a construction where to do so would amount to rewriting state or local law—an enterprise the federal courts are not empowered to undertake.” Odle, 421 F.3d at 397 (citing Virginia v. American Booksellers Ass'n, Inc., 484 U.S. 383, 397, 108 S.Ct. 636, 98 L.Ed.2d 782 (1988)).

B. The Definition of “Adult Bookstore” Is Not Unconstitutionally Overbroad Section 2907.40(A)(1) defines “ ‘[a]dult bookstore’ or ‘adult video store’ ” to mean a commercial establishment that has as a significant or substantial portion of its stock in trade or inventory in, derives a significant or substantial portion of its revenues from, devotes a significant or

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

13

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

substantial portion of its interior business or advertising to, or maintains a substantial section of its sales or display space for the sale or rental, for any form of consideration, of books, magazines, periodicals, or other printed matter, or photographs, films, motion pictures, video cassettes, compact discs, slides, or other visual representations, that are characterized by their emphasis upon the exhibition or description of specified sexual activities or specified anatomical areas. **15 Plaintiffs argue that this definition sweeps up businesses that deal only partially in sexually oriented materials but do not produce secondary effects, including neighborhood video stores that rent X-rated films, businesses like Borders that a have section devoted to the sale of romance novels, and drug stores, grocery stores and other retail establishments that sell adult-oriented materials such as pornographic magazines. This is so, they argue, because all of these businesses “have ‘sections of [their] sales or display space’ of adult materials that can easily be characterized as ‘substantial.’ ” [2] Plaintiffs urge that Executive Arts Studio, Inc. v. City of Grand Rapids, 391 F.3d 783 (6th Cir.2004), dictates the result in the instant case. There, this court struck down as overbroad the following definition of adult bookstore: An establishment having as a substantial or significant portion of its stock in trade, books, magazines, and other periodicals [and media] which are distinguished or characterized by their emphasis on matter depicting, describing or relating to “specified sexual activities” or “specified anatomical areas,” as defined herein, or an establishment with a segment or section devoted to the sale or display of such material. Id. at 787–88 (emphasis added). It did so, however, on the basis that “establishment[s] with a segment or section devoted to the sale or display of [adult] materials”

included “multiple establishments which would never be defined as adult bookstores in everyday English, such as a Walden's or Borders.” Id. at 796 (emphasis added). Because there was no evidence that such businesses produced secondary effects, the definition was overbroad. The definition in § 2907.40 is distinguishable from that at issue in Executive Arts because it does not contain the “a segment or section” language. Rather, it limits all four sub-parts of the definition with the modifier “substantial” and three sub-parts with the alternative modifier “significant.” The question here, then, is whether a significant number of establishments that do not produce secondary effects (like major bookstore chains or neighborhood video stores) have a “significant or substantial portion of [their] stock in trade or inventory in, derive[ ] a significant or substantial portion of [their] revenues *559 from, devote[ ] a significant or substantial portion of [their] interior business or advertising to, or maintain[ ] a substantial section of [their] sales or display space for” materials characterized by nudity or sexual activities. Ohio Rev.Code Ann. § 2907.40(A)(1) (emphasis added). Courts have upheld “significant or substantial” language against overbreadth challenges in First Amendment cases. See World Wide Video of Washington, Inc. v. City of Spokane, 368 F.3d 1186, 1198–99 (9th Cir.2004) (“Cases directly addressing the phrase ‘significant or substantial’ in this context have upheld its validity. Moreover, this phrase is readily susceptible to a narrowing construction.” (citations omitted)); Pleasureland Museum, Inc. v. Beutter, 288 F.3d 988, 996–97 (7th Cir.2002) (upholding definition of “Adult Bookstore, Adult Novelty Store, and Adult Video Store as commercial establishments that, inter alia, ‘derive [ ] a significant or substantial portion or [their] revenues' from Media ‘characterized by the depiction or description of’ nudity or sexual activities.” (some internal quotation marks omitted) (alterations in original)). Courts, including this one, have also held that laws that include “significant” or “substantial” language are not unconstitutionally vague. 9 See 511 Detroit St., Inc. v. Kelley, 807 F.2d 1293, 1295–96 (6th Cir.1986); VIP of Berlin, LLC v. Town of Berlin, 593 F.3d 179, 187–88 (2d Cir.2010); Z.J. Gifts D–4, L.L.C. v. City of Littleton, 311 F.3d 1220, 1229–30 (10th Cir.2002), overruled on other grounds, 541 U.S. 774, 124 S.Ct. 2219, 159 L.Ed.2d 84 (2004).

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

14

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

**16 None of these cases, however, dealt with a definition of adult bookstore or adult video store quite the same as that adopted by the State of Ohio. 10 Section 2907.40 includes not just businesses that have a “significant or substantial portion of [their] stock in trade or inventory in” adult materials, but also those that “maintain[ ] a substantial section of [their] sales or display space for” such articles. Although the term “significant or substantial” is readily susceptible to a limiting construction, prior cases do not squarely address whether the display-space clause of § 2907.40(A)(1) is sufficiently narrow in its reach. Plaintiffs argue that major bookstores with sections of their floor space devoted to romance novels, drugstores with sections “devoted to magazines like Playboy and Penthouse” or generalinterest video stores with “a separate section of adult X rated tapes” all fall within this definition. Although the record does not provide evidence regarding whether a significant number of such establishments could fairly be considered to devote a significant portion of their floor space to sexually oriented materials, a commonsense reading of the terms “significant” and “substantial” should exclude these businesses from regulation. 11 *560 Further, to be overturned, a “statute's overbreadth [must] be substantial.' ” Entm't Prods., 588 F.3d at 379 (quoting Williams, 553 U.S. at 292, 128 S.Ct. 1830). Plaintiffs have not “demonstrate[d] from the text of [the statute] and from actual fact that a substantial number of instances exist in which the law cannot be applied constitutionally.” Id. (quoting N.Y. State Club Ass'n v. City of New York, 487 U.S. 1, 14, 108 S.Ct. 2225, 101 L.Ed.2d 1 (1988)). Therefore, the definition of “adult bookstore or adult video store” is not facially overbroad. Should a business like Borders or a general-interest video store (businesses that do not cause secondary effects) become subject to regulation under the law, it may challenge its regulation on an as-applied basis. See New York v. Ferber, 458 U.S. 747, 773–74, 102 S.Ct. 3348, 73 L.Ed.2d 1113 (1982) (“Under these circumstances, [the law] is not substantially overbroad and ... whatever overbreadth may exist should be cured through case-bycase analysis of the fact situations to which [the law's] sanctions, assertedly, may not be applied.” (citation and internal quotation marks omitted) (some alterations in original)).

C. The Definition of “Adult Cabaret” Is Not Unconstitutionally Overbroad [3] Section 2907.40(A)(2) states that “[a]dult cabaret has the same meaning as in section 2907.39 of the Revised Code.” Section 2907.39(A)(3) provides: “Adult cabaret” means a nightclub, bar, juice bar, restaurant, bottle club, or similar commercial establishment, whether or not alcoholic beverages are served, that regularly features any of the following: (a) Persons who appear in a state of nudity or seminudity; (b) Live performances that are characterized by the exposure of specified anatomical areas or specified sexual activities; **17 (c) Films, motion pictures, video cassettes, slides, or other photographic reproductions that are distinguished or characterized by their emphasis upon the exhibition or description of specified sexual activities or specified anatomical areas. Ohio Rev.Code Ann. § 2907.39(A)(3) (emphasis added). “Regularly features” is defined to mean “a consistent or substantial course of conduct, such that the films or performances exhibited constitute a substantial portion of the films or performances offered as a part of the ongoing business of the adult entertainment establishment.” Id. § 2907.39(A)(11). Plaintiffs argue that this definition is overbroad because it applies to “liquor permit premises and non-liquor establishments, dinner theaters and ‘similar establishments' that might present serious plays and shows that involve mere depictions of sexual activity[, as well as] nightclubs of all kinds, including comedy clubs, that might include *561 performances with nudity or semi-nudity in them.” This court has upheld similar definitions of “adult cabaret” against overbreadth challenges. Entm't Prods., 588 F.3d at 381–83 (upholding definition of “adult cabaret” limited to “establishment[s] that feature as a principal use of [their] business, [employees who are nude or semi-nude]” (emphasis added)); Sensations, Inc. v. City of Grand Rapids, 526 F.3d 291, 300 (6th Cir.2008) (in context of ordinance nearly identical to Ohio's law, upholding “a regulation banning total nudity in sexually

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

15

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

oriented businesses” because it is “far narrower than a similar regulation [of nudity] applicable to the general public” outside the context of sexually oriented businesses that had been struck down in another case); Deja Vu of Nashville, Inc. v. Metro. Gov't of Nashville & Davidson Cnty., 466 F.3d 391, 397–98 (6th Cir.2006) (upholding ordinance defining sexually oriented business as one that “regularly depict[s] material which is distinguished or characterized by an emphasis on matter depicting [nudity or sexual activity]”). Consistent with these cases, the “regularly features” clause of the “adult cabaret” definition in § 2907.40 limits the statute so that it does not reach a substantial number of constitutionally-protected performances. Plaintiffs have failed to raise a genuine issue of material fact as to whether businesses that do not cause secondary effects, such as dinner theaters or comedy clubs, ever “regularly feature” regulated entertainment in the sense that they present it as a “consistent or substantial course of conduct, such that the films or performances exhibited constitute a substantial portion of the films or performances offered as a part of the ongoing business of the adult entertainment establishment.” Ohio Rev.Code Ann. § 2907.39(A)(11). Under this definition, it is not enough that a venue regularly feature entertainment including nudity or sexual activity in the sense that they present such material recurrently. Rather, it must be presented “consistent[ly]” and such entertainment must constitute a substantial proportion of the venue's overall offerings. This definition is sufficiently limited that it is unlikely to reach a large number of establishments that do not cause secondary effects, and thus that may not be constitutionally regulated by the law. The definition of adult cabaret is not facially overbroad.

V **18 [4] Plaintiffs also challenge § 2907.40(C)(2) to the extent that it prohibits entertainers who are nude or seminude from touching each other during the course of a performance. Although Plaintiffs' brief is unclear as to the First Amendment theory under which they seek to challenge the no-touch provision, their counsel clarified at oral argument that we should analyze their argument as an overbreadth challenge. The statute states:

No employee who regularly appears nude or seminude on the premises of a sexually oriented business, while on the premises of that sexually oriented business and while nude or seminude, shall knowingly touch a patron ... or another employee ... or the clothing of a patron ... or another employee ... or allow a patron ... or another employee ... to touch the employee or the clothing of the employee. Ohio Rev.Code Ann. § 2907.40(C)(2). The law thus prohibits any performer who is nude or seminude from touching or being touched by another performer, whether the second performer is clothed or not. This court has twice upheld similar prohibitions on erotic performers touching each other. In DLS, Inc. v. City of Chattanooga, 107 F.3d 403 (6th Cir.1997), this court upheld an ordinance providing that: *562 No entertainer, employee or customer shall be permitted to have any physical contact with any other [sic] on the premises during any performance and all performances shall only occur upon a stage at least eighteen inches (18 [″] ) above the immediate floor level and removed at least six feet (6″) from the nearest entertainer, employee and/or customer. Id. at 406 (first alteration in original). The court construed the plaintiffs' argument to be that the bufferzone requirement violated the O'Brien test. It thus did not discuss whether the ordinance violated the overbreadth doctrine. The court also did not discuss the constitutionality of the ordinance as applied to banning contact between entertainers, but rather focused on the ban on contact between entertainers and customers. More recently, in Entertainment Productions, Inc. v. Shelby County, 588 F.3d 372 (6th Cir.2009), we addressed an overbreadth challenge to a similar statute. The statute at issue in Entertainment Productions provided: “ ‘No entertainer, employee, or customer shall be permitted

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

16

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

to have any physical contact with any other on the premises during any performance,’ and to that effect, ‘all performances shall only occur ... removed at least six feet (6″) from the nearest entertainer, employee, or customer.’ ” Id. at 393 (alteration in original). The plaintiffs' overbreadth challenge asserted that the statute unconstitutionally interfered with performers' ability to convey their erotic messages through dance. We rejected this challenge, concluding that the plaintiffs failed to demonstrate a “substantial number of unconstitutional applications” of the restriction. Id. at 394. **19 We find no meaningful basis on which to distinguish the instant case from Entertainment Productions. Although Plaintiffs here offered evidence that some physical contact between performers communicates a message and constitutes expression within the meaning of the First Amendment, the plaintiffs in Entertainment Productions made the same argument and presented proofs as well. 12 Plaintiffs here made no greater showing of a “substantial number of unconstitutional applications” of the restriction than did the plaintiffs in Entertainment Productions, and thus their overbreadth challenge must fail.

VI [5] Plaintiffs argue that the plain language of § 2907.40 excludes adult bookstores and video stores from the hours-of-operation restriction. The hours-of-operation restriction applies to all “sexually oriented businesses.” Ohio Rev.Code Ann. § 2907.40(B). The definition of “sexually oriented business,” found at § 2907.40(A)(15), includes adult bookstores and adult video stores. But that definition also includes language that, Plaintiffs argue, excludes bookstores and video stores from the statute's reach: “ ‘Sexually oriented business' means an adult bookstore, adult video store, adult cabaret, adult motion picture theater, sexual device shop, or sexual encounter center, but does not include a business solely by reason of its showing, selling, or renting materials that may depict sex.” Id. § 2907.40(A)(15) (emphasis added). The definition of adult bookstore and adult video store, located at *563 § 2907.40(A)(1), identifies adult bookstores and video stores by the contents of the materials they sell or rent: materials “characterized by their emphasis upon [sexual activities] or [nudity].”

Plaintiffs thus point to a conflict: These businesses are defined in terms of their selling or renting materials that depict sex in § 2907.40(A)(1), but are apparently excluded from the definition of sexually oriented business on the same grounds in § 2907.40(A)(15). “In all cases of statutory construction, the starting point is the language employed by [the legislature].” Pittsburgh & Conneaut Dock Co. v. Dir., Office of Workers' Comp. Programs, 473 F.3d 253, 266 (6th Cir.2007). “Reliance on the literal language of the statute is not justified, however, if it leads to an interpretation which is inconsistent with the legislative intent or to an absurd result.” Appleton v. First Nat'l Bank of Ohio, 62 F.3d 791, 801 (6th Cir.1995). Importantly, “[c]ourts generally construe statutes in a way to avoid making provisions meaningless.” Sakarapanee v. Dep't of Homeland Sec., U.S. Citizenship & Immigration Servs., 616 F.3d 595, 600 (6th Cir.2010); see also Duncan v. Walker, 533 U.S. 167, 174, 121 S.Ct. 2120, 150 L.Ed.2d 251 (2001) (“It is [this Court's] duty to give effect, if possible, to every clause and word of a statute.” (internal quotation marks omitted)). Adopting Plaintiffs' reading of the statute would render § 2907.40(A)(1) meaningless by excepting from regulation all those businesses it purports to encompass. Because the plain language of the statute bears another meaning, however, this outcome can be avoided. Section 2907.40(A)(1) defines adult bookstores and adult video stores as commercial establishments whose inventory, revenues, interior business or advertising, or display space consists in “significant or substantial ” part of materials “characterized by their emphasis upon the exhibition or description of specified sexual activities or specified anatomical areas.” Thus, it is not just any business that shows, sells, or rents materials depicting sex that is subject to regulation, but rather only those substantially or significantly devoted to providing materials whose “essential character or quality” is to depict sex or nudity. See § 2907.40(A)(4) (defining “characterized by”). Establishments whose business consists of selling less than a significant or substantial amount of regulated material do not fall within the law. By this reading, § 2907.40(A)(15) reinforces the “significant or substantial” and “characterized by” language of § 2907.40(A)(1) by further clarifying that merely dealing in small amounts of adult material does not bring a business within the definition. Indeed, it would be strange for the Ohio General Assembly to enact a regulation on the hours of

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

17

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

operation of adult bookstores and video stores, and in the same regulation exempt all such businesses. **20 We therefore affirm the district court's rejection of this argument.

For the foregoing reasons, we AFFIRM the district court's grant of summary judgment.

All Citations 455 Fed.Appx. 541, 2011 WL 3904097

VII

Footnotes

* 1 2 3 4 5

6

7

8

9

Judge Sandra S. Beckwith, Senior United States District Judge for the Southern District of Ohio, sitting by designation. Businesses that hold liquor permits in Ohio are required to close at 2:30 A.M. Therefore, a sexually oriented business with a liquor license may remain open until 2:30 only if all nude performances cease at midnight. Section 2907.40(A)(16) defines “specified anatomical areas” as “human genitals, pubic region, and buttocks and the human female breast below a point immediately above the top of the areola.” “Characterized by” is defined as “describing the essential character or quality of an item.” Ohio Rev.Code Ann. § 2907.40(A)(4). These findings were included in the version of S.B. 16 initially passed by the Ohio General Assembly. The final legislation enacted by the Assembly that included the current language of § 2907.40 was a substitute bill. The above legislative findings do not appear in the version of Substitute S.B. 16 enacted into law. Plaintiffs appealed the denial of a preliminary injunction to this court. While the appeal was pending, the district court granted summary judgment for Defendants and entered final judgment on all claims, and Plaintiffs filed the instant appeal. This court subsequently dismissed the preliminary injunction appeal on the ground of mootness and because “[t]he denial of injunctive relief has merged into the final judgment.” 84 Video/Newsstand, Inc. v. Sartini, No. 08–4559 (6th Cir. filed Nov. 4, 2009). As this court has noted, “to some extent, the classification of restrictions on sexually explicit establishments as contentneutral is a legal fiction—but one that has been generally followed.” Richland Bookmart II, 555 F.3d at 521 n. 2. “Although five members of the Court abandoned the premise that such restrictions are content-neutral ... in City of Los Angeles v. Alameda Books, [535 U.S. 425, 122 S.Ct. 1728, 152 L.Ed.2d 670 (2002),] the Court continued to apply intermediate scrutiny to laws targeting ‘secondary effects.’ ” Id. (quoting 729, Inc., 515 F.3d at 490–91). Captain Chuck Adams of the Troy Police Department testified that most illegal activity at a strip club he had investigated occurred after midnight. That activity included open alcoholic containers in a public place, drug possession, DUIs, assault, and prostitution. David Sherman, a former regional manager of a strip-club chain, described how the incidence of illegal activity at strip clubs, including “drug dealing, solicitation, and illegal dances,” increased late at night as employees and customers became more intoxicated and disinhibited. The district court relied on Center for Fair Public Policy v. Maricopa County, 336 F.3d 1153 (9th Cir.2003), which upheld an hours-of-operation restriction against a challenge similar to that at issue here. There, the Ninth Circuit decided that the hours-of-operation restriction was constitutional because it left open “ample alternative channels for communication.” Id. at 1170 (citation omitted). The Ninth Circuit concluded that Justice Kennedy's proportionality analysis, which was adopted in the context of a case involving zoning restrictions on sexually oriented businesses, was not applicable in a challenge to hours-of-operation restrictions. The court reasoned that “the application of Justice Kennedy's proportionality analysis to this particular type of secondary effects law would invalidate all such laws, and we are satisfied that he never intended such a result. His proportionality requirement was simply not designed with this particular type of restriction in mind.” Id. at 1163. The Ninth Circuit's reasoning is foreclosed in this Circuit by 729, Inc v. Kenton County Fiscal Court, which held that Justice Kennedy's proportionality analysis is binding law in secondary effects cases. 515 F.3d at 491. Thus, the district court's reliance on Center for Fair Public Policy was misplaced. Although doctrinally distinct, the Supreme Court has “traditionally viewed vagueness and overbreadth as logically related and similar doctrines.” Kolender v. Lawson, 461 U.S. 352, 358 n. 8, 103 S.Ct. 1855, 75 L.Ed.2d 903 (1983); see also Entm't Prods., 588 F.3d at 379 (“The void-for-vagueness doctrine and the overbreadth doctrine vindicate overlapping

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

18

84 Video/Newsstand, Inc. v. Sartini, 455 Fed.Appx. 541 (2011)

10 11

12

values in First Amendment jurisprudence.... When a law implicates First Amendment freedoms, vagueness poses the same risk as overbreadth, as vague laws may chill citizens from exercising their protected rights.”). State cases cited by Defendants that find “substantial or significant” language not to be overbroad similarly deal with statutory definitions somewhat narrower than Ohio's. The definition at issue here is further limited by the proviso that the printed matter or visual representations featured by these businesses must be “characterized by their emphasis upon the exhibition or description of specified sexual activities or specified anatomical areas.” Ohio Rev.Code Ann. § 2907.40(A)(1) (emphasis added). The statute defines “characterized by” as “describing the essential character or quality of an item.” Id. § 2907.40(A)(4). Thus, the definition can be construed to exclude businesses that make available printed or visual media where nudity or sexual activities are not a central element. The Seventh Circuit has upheld a similar restriction against an overbreadth challenge on this reasoning. See Pleasureland Museum, Inc. v. Beutter, 288 F.3d 988, 996–97 (7th Cir.2002). The “characterized by” clause in § 2907.40(A)(1) defines the contents of the printed or visual media sold, not the nature of the businesses themselves. It thus narrows the reach of § 2907.40 but does not necessarily save it from overbreadth, as it is possible that “specified sexual activities” or “specified anatomical areas” could constitute the “essential character” of X-rated videos in a neighborhood video store, romance novels, or pornographic magazines. Dr. Judith Hanna, an expert in dance and the communicative aspects of dance, testified about the messages that exotic dance performances communicate. Joseph Hall, who works with adult cabarets, also described communicative touching between performers, including in the course of skits entered into a national competition. Dr. Hanna testified in Entertainment Productions as well.

End of Document

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

19

Nite Moves Entertainment, Inc. v. City of Boise, 153 F.Supp.2d 1198 (2001) 29 Media L. Rep. 1549

necessary to serve the substantial government interest. U.S.C.A. Const.Amend. 1. KeyCite Yellow Flag - Negative Treatment Distinguished by Bush v. City of San Diego, S.D.Cal., June 11, 2010

Cases that cite this headnote

 

153 F.Supp.2d 1198 United States District Court, D. Idaho.

[3]

NITE MOVES ENTERTAINMENT, INC., an Idaho corporation, Erotic City, a sole proprietorship, and Spontaneous Productions, Inc, an Idaho corporation, Plaintiffs, v. CITY OF BOISE, an Idaho municipal corporation, Defendant.

Preemption of local ordinance can either be express or implied. Cases that cite this headnote [4]

No. CV–00–449–S–BLW. | Jan. 17, 2001.

Cases that cite this headnote [5]

Constitutional Law Resolution of non-constitutional questions before constitutional questions Federal court should refrain from deciding a constitutional issue when a non-constitutional ground for decision is available.

[2]

Municipal Corporations Concurrent and Conflicting Exercise of Power by State and Municipality A local ordinance that does not, on its face, conflict with state law is not preempted if the local ordinance and state law are wholly unrelated.

West Headnotes (9) [1]

Municipal Corporations Concurrent and Conflicting Exercise of Power by State and Municipality State law will displace a local ordinance when a local ordinance is facially incompatible with or directly contradicts state law in area of substantive policy.

Suit was filed challenging city's public nudity ordinance. The District Court, Winmill, Chief Judge, held that ordinance was not narrowly tailored to achieve the city's objective, and therefore violated First Amendment. Judgment for plaintiffs.

Municipal Corporations Concurrent and Conflicting Exercise of Power by State and Municipality

Cases that cite this headnote [6]

Municipal Corporations Concurrent and Conflicting Exercise of Power by State and Municipality

Cases that cite this headnote

Obscenity Preemption

Constitutional Law Symbolic speech

City's public nudity ordinance was not, expressly or by implication, preempted by Idaho statute conferring upon state the exclusive right to legislate in the area of indecency and obscenity. I.C. § 18–4113.

Restriction on symbolic speech is valid under First Amendment if it (1) is within the constitutional power of the government to enact; (2) serves a substantial government interest; (3) is not related to the suppression of free expression; and (4) is not any greater than

1 Cases that cite this headnote [7]

Municipal Corporations

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

1

Nite Moves Entertainment, Inc. v. City of Boise, 153 F.Supp.2d 1198 (2001) 29 Media L. Rep. 1549

Concurrent and Conflicting Exercise of Power by State and Municipality Local enactments that merely extend the state law by way of additional restrictions or limitations are not invalid. Cases that cite this headnote [8]

Municipal Corporations Concurrent and Conflicting Exercise of Power by State and Municipality Where it can be inferred from a state statute that the state has intended to fully occupy or preempt a particular area, to the exclusion of local governmental entities, a local ordinance in that area will be held to be in conflict with the state law, even if the state law does not so specifically state.

Susan L. Mimura, Kevin S. Borger, Boise City Prosecuting Attorney's Office, Boise, ID, for defendant.

MEMORANDUM DECISION AND ORDER WINMILL, Chief Judge. Introduction The Court finds that Boise's nudity ordinance violates the First Amendment. By banning anything more revealing than short shorts and modest halter tops, the ordinance sweeps too broadly in curtailing expressive nude dancing, protected by the First Amendment under Supreme Court case law. The Court therefore finds the ordinance void and unenforceable. The Court's analysis is set forth below.

Cases that cite this headnote Analysis [9]

Constitutional Law Nudity in general Constitutional Law Nude dancing in general Obscenity Indecent exposure City's public nudity ordinance was not narrowly tailored to achieve the city's objective, and therefore violated First Amendment; by banning anything more revealing than short shorts and modest halter tops, the ordinance swept too broadly in curtailing expressive nude dancing, protected by the First Amendment. U.S.C.A. Const.Amend. 1. 1 Cases that cite this headnote

This action was initiated by the plaintiffs' filing of a complaint seeking declaratory judgment and injunctive relief. Plaintiffs claim (1) that Boise City Ordinance 6008 as amended by the Boise City Council on August 22, 2000, unconstitutionally infringes upon plaintiffs' right to freedom of expression guaranteed by the First Amendment to the United States Constitution, and (2) that the Boise City Council did not satisfy the requirements of Idaho Code § 50–902 in amending the ordinance, in violation of plaintiffs' procedural due process rights. Other issues presented by both parties relevant to the determination *1200 of these matters, including defendant's Motion in Limine to Exclude Expert Testimony (Docket No. 37), are also discussed herein. A trial in this matter was held on October 3, 2000, after which the matter was taken under advisement. The Court concludes that Boise City Ordinance 6008 unconstitutionally infringes upon the plaintiffs' First Amendment rights to freedom of expression. Accordingly, the relief requested by the plaintiffs will be granted.

Attorneys and Law Firms *1199 Debora K. Kristensen, Givens Pursley, Boise, ID, for plaintiffs. Robert B. White, Givens Pursley & Huntley, Boise, ID, for Jamie Collins.

I. Factual Background While the precise date is unclear, the concept of a “nudity ordinance” in Boise arose from “an event held in downtown Boise that became nude dancing” and “concerns and complaints about partial nudity appearing in other places, specifically in the parks.” Testimony

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

2

Nite Moves Entertainment, Inc. v. City of Boise, 153 F.Supp.2d 1198 (2001) 29 Media L. Rep. 1549

of Kevin Borger, Transcript of Boise City Pre–Council Meeting, August 8, 2000, pg. 1. Around the time the City first began formulating a nudity ordinance, the City Attorney's office became aware that an ordinance restricting nudity was being challenged and was currently pending before the United States Supreme Court. It appears that work on the Boise ordinance was put on hold until March, 2000 when the decision in City of Erie v. Pap's A.M., 529 U.S. 277, 120 S.Ct. 1382, 146 L.Ed.2d 265 (2000), came down. Id. at 1. In light of the Supreme Court's decision upholding the ordinance in City of Erie, Boise began to prepare an ordinance of its own. In an effort to ensure the Boise ordinance would pass constitutional muster, Mr. Borger attended a seminar in Washington, D.C. and gathered information and studies on the secondary effects of nudity and erotic dancing. Id. at 2. These efforts ultimately crystalized into a draft version of a Boise City ordinance prohibiting public nudity. On August 15, 2000, the Boise City Council met for its regularly scheduled meeting. On the agenda for the meeting was the third and final reading and public hearing on the proposed ordinance 0–44–00, creating a new Title 6, Chapter 22 of the Boise City Code, the title of which indicated that its intent was to outlaw public nudity. After hearing testimony from 77 members of the public, some of whom favored and some of whom opposed the ordinance, the Boise City Council approved the proposed ordinance (“the original ordinance”) in the early morning hours of August 16, 2000 by a vote of 4–1. On August 17, 2000, plaintiffs filed a Complaint for Declaratory Judgment and Injunctive Relief against defendant, seeking among other things a declaratory judgment that the original ordinance was unconstitutional under the First Amendment to the United States Constitution and under Article I, Section 9 of the Idaho Constitution. On the same day, plaintiffs filed a Motion for Emergency Temporary Restraining Order seeking an order preventing defendant from enforcing the Original ordinance. On August 18, 2000, after hearing argument from counsel for plaintiffs and defendant, this Court issued a Temporary Restraining Order prohibiting enforcement of the original ordinance. In so doing, this court noted concerns with subsection (b) of Section 6–22–02. 1

*1201 On August 22, 2000, the Boise City Council met in executive session to consider the status of the ordinance. At the conclusion of the executive session, with five of the six members of the council present, the City Council voted 4–1 to repeal subsection (b) of Section 6–22–02 of the original ordinance, thereby creating a new and amended ordinance (the “amended ordinance”). According to Borger, the repeal of subsection (b) was the result of a motion to reconsider made pursuant to Boise City Code § 1–02–16. The amended ordinance contained the same ban on public nudity that was contained in the original ordinance. Specifically, both ordinances made it unlawful for any person to appear in any public place in a state of nudity. The amended ordinance defined nudity as follows: a. “Nudity” means the showing, in a public place, of the human male or female genitals, pubic area, anus or buttocks or cleft of the buttocks with less than a fully opaque covering, the showing of the female breast with less than fully opaque covering of any part of the breast below the top of the areola, or the showing of covered male genitals in a discernibly turgid state. This ban on public nudity went substantially further than the ordinance that the Supreme Court found acceptable in the City of Erie decision. Specifically, the ordinance at issue in City of Erie would only require that erotic dancers “wear, at a minimum, ‘pasties' and a ‘G-string.’ ” City of Erie v. Pap's A.M. 529 U.S. 277, 120 S.Ct. 1382, 1388, 146 L.Ed.2d 265 (2000), while Boise's ordinance would have required the wearing of what may be referred to as “short shorts” and a modest bra or halter top. 2 Indeed, many of the swim suits and gowns sold in Boise department stores could not be worn in public without violating the Boise ordinance's ban on public nudity. The amended ordinance was published in the Idaho Statesman on August 26, 2000, and thereby became effective on that date. Subsequently, plaintiffs filed an amended complaint seeking declaratory judgment and injunctive relief from enforcement of the amended ordinance. On September 5, 2000, upon stipulation of the parties, the Court vacated the Temporary Restraining Order and issued, in its place, a Preliminary Injunction,

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

3

Nite Moves Entertainment, Inc. v. City of Boise, 153 F.Supp.2d 1198 (2001) 29 Media L. Rep. 1549

the duration of which was extended until a full hearing could be held on the matter. The trial in this matter was held on October 3, 2000. Testimony lasted approximately four hours, during which the Court heard from four witnesses, all testifying on behalf of the plaintiffs. Dr. Daniel Linz, a professor at the University of California–Santa Barbara whose studies involve the application of social science methodology to problems in law and public policy, testified first. The premise of Dr. Linz's testimony was that the land use studies and reports of crime regarding adult businesses upon which the Boise City Council relied are fundamentally flawed and do not comply with the standards for *1202 adjudicative expert testimony laid out in Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 113 S.Ct. 2786, 125 L.Ed.2d 469 (1993). Jamie Collins, a nude entertainer at Erotic City, testified next. Ms. Collins' testimony addressed the effect partial clothing requirements would have on the erotic message she conveys while dancing. Ms. Collins' testimony was followed by that of Scott Stewart, artistic director for Spontaneous Productions, Incorporated. Mr. Stewart testified regarding the impact the Boise ordinance would have on retaining the rights to perform and properly performing certain productions frequently produced at the theater. The final witness was Colleen Patricia Heaton, a partner of Nite Moves Entertainment. Ms. Heaton testified generally about the adult entertainment industry and the manner in which Nite Moves would carry out its business.

II. Analysis A. The City violated Idaho Code § 50–902 in adopting the amended ordinance and repealing subsection (b). Plaintiffs contend that in passing the Amended Ordinance at the August 22, 2000 meeting, the Boise City Council violated Idaho Code § 50–902 which provides that “[o]rdinances shall be read on three (3) different days, two (2) reading of which may be by title only and one (1) reading of which shall be in full, unless one half (½) plus one (1) of the members of the full council shall dispense with the rules.” Plaintiffs argue that Idaho law requires that in repealing and amending an ordinance, the City Council must follow the same procedure required for enactment of an ordinance. See Beem v. Davis, 31 Idaho 730, 175 P. 959 (1918).

The City concedes that it did not comply with Idaho Code § 50–902 in repealing subsection (b) of the original ordinance. However, the City argues that the city was not adopting a new ordinance, but was simply reconsidering its passage of the original ordinance. As such, the City argues that its procedures were governed by Boise City Code § 1–02–16, which sets forth the reconsideration process. That ordinance reads as follows: When a question has been once decided, it shall be in order for any member who voted in the affirmative to move for a reconsideration thereof, but no motion for reconsideration of a vote shall be made after adjournment of the next regularly scheduled meeting following the final reading of any ordinance or resolution to be considered or any vote to be reconsidered. No motion for such reconsideration shall be made more than once. The record reflects that the Council fully complied with § 1–02–16 in its attempt to repeal subsection (b). It is undisputed that the original ordinance was properly adopted on August 15. At the next regularly scheduled meeting (only a week later), Council member Baker, who voted in favor of the original ordinance, moved for reconsideration. No motion for reconsideration had been made prior to that time. The fact that the council met in executive session, or that an executive session was held to discuss “pending litigation,” is not material in this case. Section 1–02–16 does not require additional public notice or comment, and the session was called during the course of a regularly scheduled meeting. For these reasons, the City Council was clearly acting within the parameters of Boise City Code § 1–02–16, and thus properly reconsidering the original ordinance to address this Court's concerns with its constitutionality. The plaintiffs argue, however, that the Idaho Supreme Court's decision in Beem v. Davis, 31 Idaho 730, 175 P. 959 (1918), *1203 requires that a city council, in repealing or amending an ordinance, must follow the same procedure required for its original enactment. In Beem, the plaintiff filed a petition for a writ of mandamus to compel the village officers to enforce provisions of an

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

4

Nite Moves Entertainment, Inc. v. City of Boise, 153 F.Supp.2d 1198 (2001) 29 Media L. Rep. 1549

ordinance fixing fire limits. The evidence at trial showed that when a particular building was to be constructed, the village trustees attempted to suspend the ordinance so as to permit the building's construction. The trustees had simply agreed to repeal the ordinance indefinitely, while a more substantial procedure was required to pass the ordinance in the first instance. Under these circumstances, the court ruled that “this proceeding could not operate as a repeal of the ordinance” because “an ordinance can be repealed only by pursuance of the same method necessary for its enactment.” Id. at 962. Unfortunately, the Idaho Supreme Court did not explain the basis for its decision. It is unclear whether its conclusion was based upon a general principle of equity, or whether other ordinances or bylaws of the village imposed such a requirement. No case law is cited for the proposition, and plaintiffs have not provided any additional case law supporting this holding. The Court is likewise unaware of any Idaho case that has addressed this issue or stated its support for this proposition. The Court in Beem seems to have adopted what Courts in some jurisdictions have referred to as the doctrine of “legislative equivalency.” Matter of Rockland Properties Corp. v. Town of Brookhaven, 612 N.Y.S.2d 673, 674, 205 A.D.2d 518, 519 (N.Y.App.Div.1994), Chicago, I. & L. Ry. Co. v. Town of Salem, 166 Ind. 71, 76, 76 N.E. 631, 633 (Ind.1906), Bradley v. City of New Castle, 730 N.E.2d 771 (Ind.Ct.App.2000). Typically, the doctrine of legislative equivalency requires that an employee position created by a legislative act must be abolished by a correlative legislative act. Civil Service Employees Association, Inc. v. County of Nassau, 696 N.Y.S.2d 174, 264 A.D.2d 798 (N.Y.App.Div.1999). In some instances, however, the doctrine of legislative equivalency requires that the act which amends or repeals a law must be of equal dignity with the act which establishes it; and an ordinance can be amended or repealed only by another ordinance enacted with like formality as the original ordinances. In other words, existing legislation may be amended or repealed only by the same procedures as were used to enact it. See, Matter of Gallagher v. Regan, 42 N.Y.2d 230, 234, 397 N.Y.S.2d 714, 366 N.E.2d 804, 808 (N.Y.1977); see also Rockland Properties, at 519, 612 N.Y.S.2d 673; Bradley, at 787. The principles underlying the doctrine of legislative equivalency are also embodied within I.C. § 50–901 and

902, which sets forth the procedures that a city must follow in passing, amending, or repealing an ordinance. It seems clear from the language of those statutes, that a city may only amend or repeal an ordinance that has been duly passed, taken effect, and become part of the city code, by adopting yet another ordinance that expressly amends or repeals it. The city council, in adopting such an amendatory ordinance, must clearly follow the procedures that the Idaho legislature has prescribed for the adoption of ordinances. This, the City failed to do. This would call into question the validity of the City's reconsideration ordinance. However, this issue need not be addressed further here because a determination of the issue would not affect the ban on public nudity with which the plaintiff is concerned. Invalidating the amended ordinance would leave the original ordinance, since it was appropriately adopted. As noted above, the amended ordinance is identical to the original ordinance, except *1204 for the deletion of subsection (b). Because the Court finds herein that subsection (a) is an unconstitutional infringement on the freedom of expression, neither provision can properly be enforced.

B. It is unnecessary for the Court to reach the further issue of whether the City also violated the plaintiff's Due Process Rights in adopting the amended ordinance. [1] Plaintiffs also argue that the City Council, in enacting the amended ordinance, violated their Due Process Rights. However, it is well-settled that a federal court should refrain from deciding a constitutional issue when a non-constitutional ground for decision is available. Wolston v. Reader's Digest Ass'n, Inc., 443 U.S. 157, 161 n. 2, 99 S.Ct. 2701, 61 L.Ed.2d 450 (1979); Ashwander v. TVA, 297 U.S. 288, 341, 56 S.Ct. 466, 80 L.Ed. 688 (1936) (Brandeis, J., concurring); Gutierrez v. Immigration and Naturalization Service, 745 F.2d 548 (9th Cir.1984). In other words, when a non-constitutional ground will dispose of an issue, it is the Court's obligation to use it. The Court has already determined that even if the amended ordinance was not properly adopted, the Court would be required to consider an identical provision in the original ordinance. Given this holding, it is unnecessary to address the constitutional due process issue raised by the plaintiffs.

C. The Ordinance unconstitutionally infringes upon plaintiffs' First Amendment rights.

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

5

Nite Moves Entertainment, Inc. v. City of Boise, 153 F.Supp.2d 1198 (2001) 29 Media L. Rep. 1549

Plaintiffs claim that the amended ordinance unconstitutionally infringes upon plaintiffs' freedom of expression guaranteed by the First Amendment to the United States Constitution and by Article I, Section 9 of the Idaho Constitution. The Supreme Court has found that “being ‘in a state of nudity’ is not an inherently expressive condition.” City of Erie v. Pap's A.M., 120 S.Ct. 1382, 1391 (2000). However, the Supreme Court has consistently found that nude dancing of the type at issue here is expressive conduct falling within the outer ambit of the First Amendment's protection. Barnes v. Glen Theatre, Inc., 501 U.S. 560, 565–66, 111 S.Ct. 2456, 115 L.Ed.2d 504 (1991); Schad v. Mount Ephraim, 452 U.S. 61, 66, 101 S.Ct. 2176, 68 L.Ed.2d 671 (1981). To ascertain what level of scrutiny applies to the nudity ordinance here, a determination must be made as to whether the regulation “is related to the suppression of expression.” Texas v. Johnson, 491 U.S. 397, 403, 109 S.Ct. 2533, 105 L.Ed.2d 342 (1989); see also City of Erie, at 1391. As the Court found in the Johnson, if the governmental purpose in enacting the regulation is unrelated to the suppression of expression, then the regulation need only satisfy the “less stringent” standard for evaluating restrictions on symbolic speech established in United States v. O'Brien, 391 U.S. 367, 88 S.Ct. 1673, 20 L.Ed.2d 672 (1968) 3 . If, however, the government interest is related to the content of the expression, then the regulation falls outside the scope of the O'Brien test and must be justified under a more demanding standard. Id. at 403, 109 S.Ct. 2533. In City of Erie, the Supreme Court concluded that “government restriction on public nudity ... should be evaluated under the framework set forth in O'Brien.” *1205 120 S.Ct. at 1391. The Boise City ordinance, like the ordinances in Barnes and City of Erie, is, on its face, a general prohibition on public nudity. By its terms, the ordinance regulates conduct alone. As the Supreme Court found in Barnes and City of Erie, this sort of ordinance “does not target nudity that contains an erotic message; rather, it bans all public nudity regardless of whether that nudity is accompanied by expressive activity.” Id. at 1391; Barnes, 501 U.S. at 568, 111 S.Ct. 2456. In making such a finding, the Court must look primarily at the face of the ordinance. Despite allegations that the council was targeting what it deemed inappropriate expression, the Court cannot strike down an otherwise constitutional enactment “on the basis of an alleged illicit motive.” City

of Erie, 120 S.Ct. at 1391; see also United States v. O'Brien, 391 U.S. 367, 382–83, 88 S.Ct. 1673, 20 L.Ed.2d 672 (1968). Thus, the O'Brien test is properly applied in the instant action. [2] The O'Brien test requires a restriction (1) be within the constitutional power of the government to enact; (2) serve a substantial government interest; (3) not be related to the suppression of free expression; and (4) not be any greater than necessary to serve the substantial government interest. As the test is applied, the Boise ordinance is deemed valid if the ordinance satisfies these four factors for evaluating restrictions on symbolic speech. The first factor of the O'Brien test is whether the government regulation is within the constitutional power of the government to enact. A municipality in Idaho possesses and exercises those powers granted to it either by the state constitution or the legislature. Section 2 of Article 12 of the Idaho Constitution grants municipalities police power to make and enforce regulations. It is generally recognized that the authority granted under Section 2 of Article 12 is restricted in three ways: (1) the ordinance or regulation must be confined to the limited powers of the governmental body enacting the same, (2) it must not be in conflict with other general laws of the state, and (3) it must not be an unreasonable or arbitrary enactment. Benewah County Cattlemen's Ass'n, Inc. v. Board of County Com'rs of Benewah County, 105 Idaho 209, 212, 668 P.2d 85, 88 (1983). In the ordinance itself, the City Council articulates several purposes underlying its adoption. The preamble states that the “City of Boise is concerned with the health and safety of its citizens” and that the City desires to prevent criminal behavior such as “prostitution, sexual assault, the spread of sexually transmitted diseases and other deleterious effects.” They rely upon studies done by other cities on the effects of sexually oriented businesses which find a causal connection between such businesses and criminal behavior such as “sex offenses, property crimes and homicide, rape and robbery.” Boise City Code § 6008. Clearly, ordinances promoting public health and safety, such as the one here, are typically within the City's police power to enact. Thus, the ordinance is not arbitrary and, absent direct conflict with a state law, would appear to be within the City's power to enact.

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

6

Nite Moves Entertainment, Inc. v. City of Boise, 153 F.Supp.2d 1198 (2001) 29 Media L. Rep. 1549

However, plaintiffs claim that Boise was without power to enact the ordinance because the field of obscenity is statutorily preempted by the State of Idaho. The Idaho State Legislature has affirmatively indicated that it reserves, to the exclusion of municipal governments, the right to regulate indecency and obscenity. Idaho Code § 18–4113 provides: In order to make the application and enforcement of this act uniform throughout the state, it is the intent of the legislature to preempt, to the exclusion of city and county government, the regulation *1206 of the sale, loan, distribution, dissemination, presentation, or exhibition of material or live conduct which is obscene. To that end, it is hereby declared that every city or county ordinance adopted before the effective date of this act which deals with the sale, loan, distribution, dissemination, presentation, or exhibition of material of live conduct which is obscene shall stand abrogated and unenforceable on or after such effective date; and that no city or county government shall have the power to adopt any ordinance relating to the regulation of the sale, loan, distribution, dissemination, presentation, or exhibition of material or live conduct which is obscene on or after such effective date.

have challenged the ordinance on both grounds. Idaho case law discussing express preemption is sparse, and appears to be non-existent with respect to Idaho Code § 18–4113. Typically, state preemption analysis involves a two-step test: it must first be determined whether the local ordinance, in fact, conflicts with state law and, if so, it must then be determined whether the legislature intended to permit the conflict. State law will displace a local ordinance when a local ordinance is facially incompatible with or directly contradicts state law in area of substantive policy. Id. at 519. However, a local ordinance that does not, on its face, conflict with state law is not preempted if the local ordinance and state law are wholly unrelated. Musser, 67 Idaho at 220, 176 P.2d 199. [6] Idaho Code § 18–4113 clearly and unequivocally states that the State of Idaho will have the exclusive right to legislate in the area of indecency and obscenity. The statute expressly states its intent “to preempt, to the exclusion” of local government, the regulation of obscene conduct. The question posed, then, is whether the Boise ordinance constitutes the regulation of obscene conduct, so as to fall within the area of authority expressly reserved to the state. It is undisputed that Idaho's obscenity statute may sweep within its regulatory framework some forms of nudity. The statute makes it unlawful to sell, distribute, publish, print or exhibit obscene material. Obscene live conduct is defined in the statute to mean any physical body activity which appeals to the prurient interest and is patently offensive because it consists of sexual acts, masturbation, excretory functions, or lewd exhibition of the genitals or genital area. I.C. § 18–4101(1)(2)(a) and (b).

[7] By comparison, the ordinance at issue here prohibits an individual who appears in public “in a state of nudity.” As plaintiffs note, the Boise nudity ordinance and Idaho While it is obvious that the “obscene live conduct,” Code § 18–4113 have many similar purposes and may at prohibited by the statute, will often involve a display of times prohibit identical conduct. “Safety, morality and nudity that violates the ordinance, it is also clear that community protection are dominant themes in [the Senate the ordinance goes much further and prohibits all public Statement of Purpose of Idaho Code § 18–4113] as well nudity, whether deemed obscene under the statute or as the preamble to the Ordinance.” Plaintiffs' Trial Brief, not. Local enactments which merely extend the state law pg. 25. From this alone, however, one cannot reach the by way of additional restrictions or limitations are not conclusion—urged by the plaintiffs—that the state has invalid. See Benewah County Cattlemen's Ass'n, Inc. v. preempted the field. Board of County Com'rs of Benewah County, 105 Idaho 209, 214, 668 P.2d 85 (1983); Voyles *1207 v. City of [3] [4] [5] Preemption can either be express or implied. Nampa, 97 Idaho 597, 601, 548 P.2d 1217 (1976); Taggart State v. Musser, 67 Idaho 214, 176 P.2d 199 (1946); Caesar v. Latah County, 78 Idaho 99, 102, 298 P.2d 979 (1956). v. State, 101 Idaho 158, 610 P.2d 517 (1980). Plaintiffs

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

7

Nite Moves Entertainment, Inc. v. City of Boise, 153 F.Supp.2d 1198 (2001) 29 Media L. Rep. 1549

For this reason, the Court finds that the City's ordinance is not expressly preempted by I.C. § 18–4113.

the ordinance, and the first factor of the O'Brien test is therefore satisfied.

[8] Plaintiffs argue that even if the ordinance is not expressly preempted by statute, the legislature has preempted such bans on nudity by implication. In Caesar v. State, 101 Idaho 158, 161, 610 P.2d 517 (1980), the Court described the doctrine of implied preemption as follows:

[9] The second factor of the O'Brien test is whether the regulation furthers an important or substantial government interest. In City of Erie, the court explained this factor as it related to nudity ordinances:

Where it can be inferred from a state statute that the state has intended to fully occupy or preempt a particular area, to the exclusion of [local governmental entities], a [local] ordinance in that area will be held to be in conflict with the state law, even if the state law does not so specifically state. In Envirosafe Serv. of Idaho v. County of Owyhee, 112 Idaho 687, 735 P.2d 998 (1987), the Court reiterated the Caesar standard and further explained that “[t]he doctrine of implied preemption typically applies in instances where, despite the lack of specific language preempting regulation by local governmental entities, the state has acted in the area in such a pervasive manner that it must be assumed that it intended to occupy the entire field of regulation.” Id. at 689, 735 P.2d 998. The Court has reviewed the Idaho legislature's efforts at regulating obscenity and similar conduct, and concludes that the lack of any specific statutory prohibitions on public nudity belies any suggestion that the state's comprehensive regulation of obscenity indicates an intention to preempt the field and preclude local regulation of public nudity. Admittedly, the legislature, in its adoption of obscenity laws, may have intended to address the same societal problems that the City of Boise was attempting to address in adopting its ban on public nudity. However, the doctrine of implied preemption does not go so far as to preclude local governments from targeting, through different means, the social ills that the legislature seeks to remedy. Overlap in purpose is not tantamount to conflict in method. For these reasons, the Court finds that the State has not preempted, expressly or by implication, the City's ability to regulate public nudity. The City has the power to enact

The asserted interests of regulating conduct through a public nudity ban and of combating the harmful secondary effects associated with nude dancing are undeniably important. And in terms of demonstrating that such secondary effect pose a threat, the city need not “conduct new studies or produce evidence independent of that already generated by other cities” to demonstrate the problem of secondary effects, “so long as whatever evidence the city relies upon is reasonably believed to be relevant to the problem that the city addresses.” Renton v. Playtime Theatres, Inc., 475 U.S. 41, 106 S.Ct. 925, 89 L.Ed.2d 29. Because the nude dancing at Kandyland is of the same character as the adult entertainment at issue in Renton, Young v. American Mini Theatres, Inc., and California v. LaRue, it was reasonable for Erie to conclude that such nude dancing was likely to produce the same secondary effects.... Even in cases addressing regulations that strike closer to the core of First Amendment values, we have accepted a state or local government's *1208 reasonable belief that the experience of other jurisdictions is relevant to the problem it is addressing. City of Erie, 120 S.Ct. at 1395. In this case, the Boise City Council has explicitly relied on City of Erie in justifying its efforts to restrict negative secondary effects. 4 Moreover, the Boise City Council expressly considered and relied on the same studies that the Erie City Council used to justify their nudity ordinance. Because reliance on those studies was found reasonable by the Supreme Court, this Court cannot now reject those studies as flawed despite the testimony of plaintiff's expert to that effect. That expert testimony was substantially equivalent, if not identical, to arguments he presented in his Amicus Brief to the Supreme Court in City of Erie. Plaintiffs have not alleged that new information has come to light since that decision. Thus, an examination as to the adequacy and reasonableness of those studies is precluded. However, the Supreme Court's determination “that a legislative body may rely on foreign studies to establish

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

8

Nite Moves Entertainment, Inc. v. City of Boise, 153 F.Supp.2d 1198 (2001) 29 Media L. Rep. 1549

its interest in a regulation does not relieve that entity from the obligation of demonstrating that the study must be ‘reasonably believed to be relevant to the problem the city addresses.’ ” Alameda Books, Inc., 222 F.3d 719, 726 (2000) (quoting Renton 475 U.S. at 51–52, 106 S.Ct. 925 (1986)). Therefore, while it was not necessary for the Boise City Council to undertake its own studies, the City could only rely on studies conducted by other cities if it reasonably believed that those studies were relevant to the problems that the City Council was attempting to address by adopting the nudity ordinance. In considering the reasonableness of the Council's conclusions in this regard, the Court is mindful of the Supreme Court's admonitions about the limited role of the judiciary in scrutinizing legislative judgment. In City of Erie, the Supreme Court gave substantial deference to the members of the city council who “were familiar with downtown commercial Erie” and likely had “first-hand knowledge of what took place around nude dancing establishments in Erie.” 120 S.Ct. at 1395; see also American Mini Theatres, 427 U.S. at 71, 96 S.Ct. 2440, 49 L.Ed.2d 310 (1976)(In upholding the validity of a zoning regulation, the court found “it is not our function to appraise the wisdom of [the City Council's] decision ... Moreover, the city must be allowed a reasonable opportunity to experiment with solutions to admittedly serious problems.”); Jones v. Intercable, Inc. v. City of Chula Vista, 80 F.3d 320, 326 (9th Cir.1996) (courts “accord substantial deference to the predictive judgments” of legislative bodies when analyzing contentneutral regulations that burden speech) (quoting Turner Broadcasting System, Inc. v. FCC, 512 U.S. 622, 665, 114 S.Ct. 2445, 129 L.Ed.2d 497 (1994)). Thus, the plaintiffs in the instant action bear a burden to demonstrate that the council was clearly unreasonable in its determination that restriction of nudity in adult establishments would mitigate criminal behavior in the City of Boise. In this respect, an expert could be relevant to this action and the court will therefore admit Dr. Linz's testimony. But for the most part, Dr. Linz's testimony focused on findings that the City of Erie court found could be reasonably relied upon. Plaintiffs have not produced evidence *1209 suggesting that the studies reasonably relied upon in Erie, Pennsylvania were inapplicable to the City of Boise. Additionally, plaintiffs have not shown that the council's reasoning was patently false or even

unreasonable. As a result, the court finds that the ordinance does further an important government interest. The third factor in O'Brien is whether the government interest is unrelated to the suppression of free expression. Given the similarities between the Erie and Boise ordinances, it appears that the finding by the Supreme Court in City of Erie that this factor is satisfied is controlling. The parties do not dispute this conclusion. The fourth factor—that the restriction is no greater than is essential to the furtherance of the government interest —requires a closer examination. In City of Erie, the court found that this requirement was satisfied, because “[t]he requirement that dancers wear pasties and G-strings is a minimal restriction in furtherance of the asserted government interests, and the restriction leaves ample capacity to convey the dancer's erotic message.” at 1397. Other than to observe that least restrictive means analysis is not necessary because the ordinance was a contentneutral restriction, the Court provided little enlightenment as to how this factor of the O'Brien test is to be applied in the concrete case. This is unfortunate, because it is clear that the Boise ordinance imposes a far more substantial restriction than does the ordinance at issue in City of Erie—the latter imposing only a de minimis restriction on nude erotic dancing and the former imposing a conservative dress code. However, other decisions of the Supreme Court make clear that the fourth part of the O'Brien test involves questioning whether the statute is “narrowly tailored” to correct the social ill at which it is targeted. See, Ward v. Rock Against Racism, 491 U.S. 781, 798–99, 109 S.Ct. 2746, 105 L.Ed.2d 661 (1989); Board of Trustees of the State University of New York v. Fox, 492 U.S. 469, 478, 109 S.Ct. 3028, 106 L.Ed.2d 388 (1989). A statute is narrowly tailored if it “targets and eliminates no more than the exact source of the ‘evil’ it seeks to remedy.” Russell C. Frisby, et al., v. Sandra C. Schultz, et al., 487 U.S. 474, 485, 108 S.Ct. 2495, 101 L.Ed.2d 420 (1988); City Council of Los Angeles v. Taxpayers for Vincent, 466 U.S. 789, 808–10, 104 S.Ct. 2118, 80 L.Ed.2d 772 (1984). Thus, “the essence of narrow tailoring” is “focusing on the evils the [government] seeks to eliminate ... [without] significantly restricting a substantial quantity of speech that does not create the same evils.” Turner Broadcasting System, Inc. v. FCC (Turner II), 520 U.S. 180, 216, 117 S.Ct. 1174, 137 L.Ed.2d 369 (1997) (citing Ward v. Rock Against Racism,

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

9

Nite Moves Entertainment, Inc. v. City of Boise, 153 F.Supp.2d 1198 (2001) 29 Media L. Rep. 1549

491 U.S. 781, 799, n. 7, 109 S.Ct. 2746, 105 L.Ed.2d 661 (1989)).

vice”); Café 207, Inc. v. St. Johns County, 856 F.Supp. 641, 646 (M.D.Fla.1994).

The narrow tailoring requirement is met if the rule is “not substantially broader than necessary to achieve the government's interest.” Ward, 491 U.S. at 800, 109 S.Ct. 2746. Indeed, a complete ban can be narrowly tailored, but only if each activity within the proscription's scope is an appropriately targeted evil. Frisby at 485, 108 S.Ct. 2495. But, “a regulation is not ‘narrowly tailored’—even under the more lenient tailoring standards applied in Ward and Renton—where, ... ‘a substantial portion of the burden on speech does not serve to advance [the State's contentneutral] goals.’ ” Simon & Schuster, Inc., v. Members of the New York State Crime Victims Board, et al., 502 U.S. 105, 112 S.Ct. 501, 116 L.Ed.2d 476 (1991); see also, Ward at 799, 109 S.Ct. 2746.

The problem here is that the city's ordinance does not just require the wearing of pasties and a G-string, nor does it require the wearing of a bikini or other partial cover-up. Rather, the ordinance would ban the wearing of anything more revealing than short shorts and a modest bikini top. In doing so, it would completely interfere with the erotic message of nudity that the Supreme Court has indicated

The plaintiffs contend that the Supreme Court's approval of an ordinance which permits partial nudity—requiring only pasties and a G-string—somehow suggests *1210 that this is the only dress requirement that can be constitutionally imposed. However, given the broad discretion accorded to legislative bodies under the “narrow tailoring” standard, it is clear that the Supreme Court's decision does not impose such a bright line rule. Although pasties and a G-string restriction may be such a minimal intrusion upon the dancer's ability to convey an erotic message that it is not necessary to even consider whether the ordinance is narrowly tailored to achieve the city's objective, this does not mean that an ordinance which requires a more modest dress style would not satisfy the narrow tailoring requirement. Courts have consistently found this analysis proper. In Colacurcio v. City of Kent, the Ninth Circuit rejected a challenge to a ten-foot setback provision in erotic dancing clubs, although less restrictive regulations (i.e., four or six foot setbacks) could have achieved the same result. 163 F.3d 545, 554, 557 (9th Cir.1998). Also, the Court recognizes that ordinances requiring more covering than just pasties and a G-string have been upheld under the narrow tailoring test in some jurisdictions. Bright Lights, Inc. v. City of Newport, 830 F.Supp. 378, 384 (E.D.Ky.1993)(finding bikini requirements for erotic dancers was sufficiently narrow in advancing goals of reforming city's reputation as “home to smorgasbord of

is entitled to limited First Amendment protection. 5 It would seem clear that a city may go farther than the City of Erie and require more than just the wearing of pasties and a G-string. 6 However, by not only prohibiting nudity, but also imposing a conservative dress code which precludes any expressive nude conduct, the City has “burden[ed] substantially more speech than is necessary to further the government's legitimate interests,” and the restriction is “substantially broader than necessary to achieve the government's interest.” Ward, 109 S.Ct. at 2746, 2758. The Court won't tell the City how to draft their ordinance, except to say that the City of Erie's ordinance clearly passes constitutional muster, and the City of Boise's ordinance clearly does not. Accordingly, IT IS NOW THEREFORE ORDERED, that Boise City Ordinance 6008, creating a new Title 6 Chapter 22 of the Boise City Code title Public Indecency, as amended by the Boise City Council on August 22, 2000 is unconstitutional, void, and unenforceable and of no force and effect under the First Amendment to the United States Constitution, and that the Defendant, its agents, servants, employees or anyone else acting in concert with the *1211 Defendant or on its behalf shall be, and the same are hereby permanently enjoined from enforcing said Ordinance. IT IS FURTHER ORDERED, that defendant's Motion to Strike Expert Testimony (Docket No.37) shall be, and is hereby, DENIED.

All Citations 153 F.Supp.2d 1198, 29 Media L. Rep. 1549

Footnotes

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

10

Nite Moves Entertainment, Inc. v. City of Boise, 153 F.Supp.2d 1198 (2001) 29 Media L. Rep. 1549

1

2

3

4 5

6

Section 6–22–02(b) made it illegal for a person to “manage, produce, sponsor, present, or exhibit, a person in the state of nudity.” In this respect the Original ordinance substantially departed from the language of the ordinance upheld in City of Erie v. Pap's A.M., 120 S.Ct. 1382, 1391 (2000), by imposing not only a ban on public nudity, but also a prohibition on displays, presentations, and exhibitions that included any form of nudity. By subsequently amending the original ordinance to delete this provision, the City appears to have conceded that it presented overbreadth problems not at issue in City of Erie. Boise's ordinance is substantially more restrictive than nudity ordinances upheld in other jurisdictions. See, e.g., Bright Lights, Inc. v. City of Newport, 830 F.Supp. 378, 380 (E.D.Ky.1993) (upholding ordinance requiring G-string and a bra not “expos[ing] to view the portion of the breast below a horizontal line across the top of the areola” but excluding “cleavage ... exhibited by a dress, blouse, shirt, leotard, bathing suit, or other wearing apparel provided the areola is not exposed in whole or in part”); Café 207, Inc. v. St. Johns County, 856 F.Supp. 641, 643 (M.D.Fla.1994) (upholding ordinance providing “in net effect, a female is ‘nude’ whenever more than two-thirds of the buttocks or more than three-fourths of the breasts are exposed.”). The parties have not cited any case upholding an ordinance as restrictive as the Boise ordinance. In O'Brien, the Supreme Court found that a statute prohibiting knowing destruction of selective service certificates did not abridge free speech on its face. It was determined that “freedom of expression” does not include all modes of communication of ideas by conduct. Therefore, the Court concluded that punishment for burning a draft card penalized only the non-communicative element of the act even though the act contained communicative elements. Id. The preamble states “WHEREAS, the City Council of the City of Boise is aware of studies done by other cities into the effects of sexually oriented businesses and relies upon those studies as a basis to support the passage of this ordinance as permitted by City of Erie v. Pap's A.M. tdba “Kandyland”, 529 U.S. 277, 120 S.Ct. 1382 (2000); ...” See City of Erie, 120 S.Ct. at 1388–89, finding that “nude dancing ... is expressive conduct that is entitled to some quantum of protection under the First Amendment.” See also, Barnes v. Glen Theatre, Inc., 501 U.S. 560, 566, 111 S.Ct. 2456, 115 L.Ed.2d 504 (1991), finding that “nude dancing of the kind sought to be performed here is expressive conduct within the outer perimeters of the First Amendment, though we view it as only marginally so.” As noted above, courts in other jurisdictions have upheld ordinances requiring more clothing than required by the City of Erie statute, while the Court is unaware of any case in which a Court has upheld an ordinance more restrictive than the one presented here.

End of Document

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

11

Cleveland State University

[email protected] Law Faculty Articles and Essays

Faculty Scholarship

2012

The Association of Adult Businesses with Secondary Effects: Legal Doctrine, Social Theory, and Empirical Evidence Alan C. Weinstein Cleveland State University, [email protected]

Richard D. McCleary

How does access to this work benefit you? Let us know! Follow this and additional works at: http://engagedscholarship.csuohio.edu/fac_articles Part of the First Amendment Commons, Land Use Law Commons, and the Urban Studies Commons Original Citation Alan C. Weinstein, The Association of Adult Businesses with Secondary Effects: Legal Doctrine, Social Theory, and Empirical Evidence, 29 Cardozo Arts & Entertainment Law Journal 565 (2012)

This Article is brought to you for free and open access by the Faculty Scholarship at [email protected] It has been accepted for inclusion in Law Faculty Articles and Essays by an authorized administrator of [email protected] For more information, please contact [email protected]

+(,121/,1( Citation: 29 Cardozo Arts & Ent. L.J. 565 2011 Content downloaded/printed from HeinOnline (http://heinonline.org) Wed Oct 24 11:49:59 2012 -- Your use of this HeinOnline PDF indicates your acceptance of HeinOnline's Terms and Conditions of the license agreement available at http://heinonline.org/HOL/License -- The search text of this PDF is generated from uncorrected OCR text. -- To obtain permission to use this article beyond the scope of your HeinOnline license, please use: https://www.copyright.com/ccc/basicSearch.do? &operation=go&searchType=0 &lastSearch=simple&all=on&titleOrStdNo=0736-7694

THE ASSOCIATION OF ADULT BUSINESSES WITH SECONDARY EFFECTS: LEGAL DOCTRINE, SOCIAL THEORY, AND EMPIRICAL EVIDENCE* ALAN

C. WEINSTEIN* & RICHARD

MCCLEARY**

INTRODUCTION...............................................565 I. LAND-USE REGULATION OF ADULT ENTERTAINMENT BUSINESSES AND THE EVOLUTION OF THE SECONDARY EFFECTS DOCTRINE .............................. 567 II. CHALLENGES BASED ON DAUBERT CLAIMS . ............... 575 III. CHALLENGES BASED ONJURISDICTION-SPECIFIC STUDIES....578 IV. CHALLENGES BASED ON THE ON-SITE/OFF-SITE DISTINCTION OR THE URBAN/RURAL DISTINCTION............ 581

A. The On-Site/Off-Site Distinction ...... ....... 581 B. The Urban/Rural Distinction ............... 584

V.

THEORETICAL AND CASE STUDY SUPPORT FOR THE ASSOCIATION OF ADULT BUSINESSES WITH SECONDARY EFFECTS................................................585

A. The Prevailing Criminological Theory of Secondary Effects. ...................... ..... 586 B. The Role of Adult Business Types in Criminological Theory ................. .......... 588 C. The Role of Adult Business Location in Criminological Theory .................... 589 D. Case Studies Confirming the Prevailing Criminological Theory of Secondary Effects............591 1. Off-Site Business Case Study ....... ......... 591 2. Rural Case Study: Montrose, Illinois ............... 593 CONCLUSION

..........................................

........595

INTRODUCTION

In the decade since the U.S. Supreme Court's decision in City of Los Angeles v. Alameda Books, Inc.,' the adult entertainment * Permission is hereby granted for noncommei rcial reproduction of this Note in whole or in part for education or research pur poses, including the making of multiple copies for classroom use, subject only to the condition that the name of the author, a complete citation, and this copyright notice and grant of permission be included in all copies. * Associate Professor and Director, Law & Public Policy Program, Cleveland-Marshall College of Law & Maxine Goodman Levin College of Urban Affairs, Cleveland State University.

565

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 565 2011

566

CARDOZO ARTS & ENTERTAINMENT

[Vol. 29:565

industry has attacked the legal rationale local governments rely upon as the justification for their regulation of adult businesses: that such businesses are associated with so-called negative secondary effects. These attacks have taken a variety of forms, including: trying to subject the studies of secondary effects relied upon by local governments to the Daubert standard for admission of scientific evidence in federal litigation;' producing studies that purport to show no association between adult businesses and negative secondary effects in a given jurisdiction;' and claims that distinct business models' and/or specific local conditions" are not associated with the secondary effects demonstrated in the studies relied on by many local governments. In this Article, we demonstrate that, contrary to the industry's claims, methodologically appropriate studies confirm

**Professor, School of Social Ecology, University of California, Irvine. I City of L.A. v. Alameda Books, Inc., 535 U.S. 425 (2002).

2 Commercially available adult entertainment in the form of "adult cabarets" (or "strip clubs") featuring live dancers and retail stores selling pornographic magazines and electronic media-some of which also feature so-called "viewing booths"-is a multi-billion dollar industry. See, e.g., Ronald Weitzer, Sex Woik: Paradigmsand Policies, in SEX FOR SALE: PROSTITUTION, PORNOGRAPHY, AND THE SEX INDUSTRY 1 (Ronald Weitzer ed., 2d ed. 2010) ("In 2006 alone, Americans spent $13.3 billion on X-rated magazines, videos and DVDs, live sex shows, strip clubs, adult cable shows, computer pornography, and commercial telephone sex. Rentals and sales of X-rated filmsjumped from $75 million in 1985 to $957 million in 2006. In just one decade, the number of X-rated films released annually more than doubled, fron 5700 in 1995 to 13,588 in 2005. There are around 3500 strip clubs in America and the number has grown over the past two decades."). See genertly Christopher J. Andrew, The Secondary E/fects Docthine: The Historical Deve/opment G(arrent Application, and Potential Mischaractetization of an Eiusive Judicial Piecedent, 54 RUTGERS L. REV. 1175 (2002); John Fee, The Pornographic Secondary Effts Doctrine, 60 ALA. L. REV. 291, 292 (2009) (" [A] regulation will be treated as contentneutral and subject to intermediate scrutiny, despite its content-discriminatory form, if the primary purpose of the regulation is to control the secondary effects rather than the primary effects of speech"). 4 Daubert v. Merrell Dow Pharm., 509 U.S. 579, 590 (1993) ("[I]n order to qualify as 'scientific knowledge' [under Federal Rule of Evidence 702,] an inference or assertion must be derived by the scientific method[, but does not need to be 'known' to a certainty].").

5 See,

e.g., Bryant Paul et al., Goveinment Regulation of "Adull" Businesses Through Zoning and Anti-Nudity Ordinances:Debunking the Legal Myth of Negative Seconday EFecls, 6 CoMM. L. & POL'Y 355, 356 (2001). GSee, e.g., Daniel Linz et al., Peep Show Establishments, Police Activity, Public Place, and lTime: A Study ofSecondary Efects in San Diego, Cali/,rnia,43J. SEX RES. 182 (2006). 7 See, e.g., Encore Videos, Inc. v. City of San Antonio, 330 F.3d 288, 294 (5th Cir. 2003) (declaring adult business ordinance unconstitutional because none of the secondary effects studies cited in the legislative record had studied "take-home" adult media stores where no adult entertainment is presented or viewed on the premises), opinion clarifiec 352 F.3d 938 (5th Cir. 2003). Contra Doctor John's v. Wahlen, 542 F.3d 787, 793 (10th Cir. 2008) (rejecting claim that the "on-site/off-site" distinction is relevant in initially judging whether a local government reasonably relied on the studies in enacting its regulations); Richland Bookmart, Inc. v. Knox Cnty., 555 F.3d 512, 526 (6th Cir. 2009) (holding that a local government may rely on a study of secondary effects that did not address the particular category of adult business challenging the ordinance). S See, e.g., Abilene Retail #30, Inc. v. Bd. of Comm'rs, 492 F.3d 1164, 1175 (10th Cir. 2007), ced. denied, 552 U.S. 1296 (2008) (ruling that a local government in a rural area could not have reasonably relied on studies of secondaiy effects that did not examine businesses in an entirely rural area).

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 566 2011

2011] ADULT BUSINESSES WITH SECONDARY EFFECTS

567

criminological theory's prediction that adult businesses are associated with heightened incidences of crime regardless of jurisdiction, business model, or location and thus, such studies should have legal and policy effects supporting regulation of adult businesses. I. LAND-USE REGULATION OF ADULT ENTERTAINMENT BUSINESSES9 AND THE EVOLUTION OF THE SECONDARY EFFECTS DOCTRINE

In the late 1960s, Boston's city planners proposed to concentrate the city's adult businesses in a single, small "adult entertainment district" located in the city's downtown area near Chinatown. Popularly known as the "Combat Zone," the district was formally established in 1974.' This proposal had two theoretical advantages. First, it would keep vice activity out of the city's other districts. Second, it would allow the police to focus resources on a small area, thereby reducing the risk of crimes associated with vice. These theoretical advantages, however, were not realized in practice. Soon after the district had been established, crime and disorderly conduct escalated and the failure of Boston's "Combat Zone" experiment was obvious." At about this same time, city officials in Detroit began to notice "the emergence of clusters of 'adult' movie theatres and bookstores together with topless bars and 'go go' establishments in certain areas of the City."" Detroit then added adult entertainment businesses to an existing ordinance that already prohibited the concentration of a number of other businesses that were associated with negative effects on surrounding propertiesincluding bars, transient hotels, and poolrooms-by setting minimum distances between adult businesses and certain other uses." Several existing adult businesses that were being forced to relocate challenged the constitutionality of the Detroit ordinance, but the ordinance was upheld by the district court noting that the 9 This

article focuses exclusively on land-use regulation of adult entertainment businesses. We do not discuss, other than tangentially, two other forms of regulation that have been used extensively: restrictions on nudity in adult performances and licensing of adult businesses, owners and employees. The Supreme Court has decided two cases involving restrictions on nudity - Barnes v. Glen Theatre, Inc., 501 U.S. 560 (1991) and City of Erie v. Pap's A.M., 529 U.S. 277 (2000) - and two cases involving licensing of adult entertainment: FW/PBS, Inc. v. City of Dallas, 493 U.S. 215 (1990) and City of Littleton v. Z.J. Gifts D-4, L.L.C., 541 U.S. 774 (2004). For a discussion of these forms of regulation, See generally, BRIAN W. BLAESSER & ALAN C. WEINSTEIN, FEDERAL LAND USE LAW &

LITIGATION ch. 6 (2011 ed. 2011). 1o Nicole Stelle Garnett, RelocatinglDisorder,91 VA. L. REV. 1075, 1106-07 (2005). 11 Id. at 1107; see also Norman Marcus, Zoning Obscenity: Or, the Moral Politics o Pou, 27 BUFF. L. REV. 1, 3-4 (1978); see generally WESLEY G. SKOGAN, DISORDER AND DECLINE: CRIME AND THE SPIRAL OF DECAY IN AMERICAN NEIGHBORHOODS (1990).

12 Am. Mini Theatres, Inc. v. Gribbs, 518 F.2d 1014, 1015 (6th Cir. 1975), Young v. Am. Mini Theatres, Inc., 427 U.S. 50 (1976). 13Id. at 1016, 1018.

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 567 2011

ev'd sub nom.

568

CARDOZO ARTS & ENTERTAINMENT

[Vol. 29:565

city had relied on the opinions of social scientists and real estate experts who agreed that prohibiting adult businesses from concentrating in any one area would mitigate negative effects on surrounding properties associated with a concentration of such businesses: primarily increased crime and downward pressure on property values.14 After the Sixth Circuit reversed the district court's decision because it saw the ordinance as a prior restraint,'" a fractured Supreme Court upheld the ordinance.' Justice Stevens' plurality opinion in Young v. American Mini Theatres, Inc." viewed the pornographic speech at issue as not worthy of the most robust protection under the First Amendment, citing the Court's treatment of commercial speech as precedent for varying the protection afforded under the First Amendment." Justice Powell, who provided the fifth vote to uphold the ordinance, wrote separately to argue that Detroit was justified in enacting the ordinance because it was aimed at mitigating adverse secondary effects associated with the regulated businesses." Although Young accepted that regulations could be based on the enacting legislative body's concern with addressing the secondary effects associated with adult businesses, it said nothing about the quantity or quality of the evidence that was needed to demonstrate that such a concern was legitimate. These questions were addressed ten years later in City of Renton v. Playtime Theatres, Inc.20 In the early 1980s, the Seattle suburb of Renton, Washington, enacted a zoning ordinance that in many respects resembled the ordinance challenged in Young. Since Renton had no adult businesses at the time the ordinance was enacted, it could not base its regulations on a study of secondary effects in Renton itself and so looked to a Washington Supreme Court opinion reviewing studies from nearby Seattle.2 ' The following year, two theaters located in a district where adult businesses were prohibited began to show "X-rated" films and immediately sought

Id. at 1018. Id. at 1019-20. 16 Young v. Am. Mini Theatres, Inc., 427 U.S. 50, 72-74 (1976).

14 15

17 Id.

18 Id. at 68-69. 19 Id. at 80-82. Justice Stewart, joined by Justices Brennan, Marshall and Blackmun dissented, arguing that the ordinance was unconstitutional because its distinction between theaters was based on the content of the films they exhibited and thus was not a valid content neutral regulation of the time, place or manner of expression. Id. at 83-88. Justice Blackmun also dissented separately, objecting to the majority's refusal both to consider respondent's vagueness claims and to overturn the ordinance on those grounds. Id. at 88. 2o City of Renton v. Playtime Theatres, Inc., 475 U.S. 41 (1986). 21 Id. at 50-51 (discussing Northend Cinema, Inc. v. City of Seattle, 585 P.2d 1153 (Wash. 1978) (en banc)).

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 568 2011

2011] ADULT BUSINESSES WITH SECONDARY EFFECTS

569

a declaratory judgment that the ordinance was unconstitutional." The trial court upheld the ordinance, but the Ninth Circuit reversed." The Supreme Court reversed the Ninth Circuit, ruling that the ordinance complied with the Young standard in that its sole purpose was the mitigation of secondary effects. Further, the Court explicitly stated that a city did not have to conduct its own study of secondary effects or produce evidence of secondary effects in addition to those already available from other cities before enacting an ordinance regulating adult businesses, so long as the city reasonably believed that whatever studies or evidence it relied on were relevant to the problem the city was addressing.24 Renton thus legitimized the practice of basing the governmental purpose for enacting a local adult business ordinance on secondary effects studies from other communities. Renton also set a reliability threshold, albeit a low one, for the government's secondary effects evidence: the evidence need only be "reasonably believed to be relevant to the problem that the city addresses."1' The Supreme Court revisited this issue sixteen years later in its decision in City of Los Angeles v. Alameda Books, Inc." In 1977, Los Angeles had conducted a comprehensive secondary effects study that found, among other things, that concentrations of adult businesses were associated with high ambient crime rates. Based on this finding, Los Angeles enacted an ordinance requiring adult businesses to be separated by a minimum distance. In 1983, concerned that the ordinance contained a loophole that would allow multiple adult businesses to operate in a single structure, the city amended the ordinance to prohibit the operation of more than one adult business in the same building or structure. Instead of requiring minimum distances between adult businesses, the amended ordinance required minimum distances between distinct adult entertainment activities. Adult businesses that combined onsite coin-operated video viewing booths with sales of videos for offsite use were prohibited and existing multiple-activity businesses were forced to segregate their on-site and off-site activities." In 1995, two multiple-activity businesses challenged the

Id. at 43. 2. 1(1. 22

Id. at 51-52 (Justice Rehnquist wrote: "The First Amendment does not require a city, before enacting such an ordinance, to conduct new studies or produce ecidence independent of that already generated by other cities, so long as whatever evidence the city relies upon is reasonably believed to be relevant to the problem that the city addresses."). 25 Id. 26 City of L.A. v. Alameda Books, Inc., 535 U.S. 425 (2002). 2 Id. at 429-33. Justice Souter characterized this model as "commercially natural, if not universal." Id. at 465 (SouterJ., dissenting). 24

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 569 2011

570

CARDOZO ARTS & ENTERTAINMENT

[Vol. 29:565

amended ordinance. Since the 1977 study had said nothing about the secondary effects of combining multiple activities under one roof, they argued that Los Angeles had no evidence that multipleactivity businesses were associated with secondary effects. The district court agreed and the Ninth Circuit affirmed on the ground that because the 1977 Los Angeles study had not investigated the effects of multiple-activities under one roof, the evidence for the amended ordinance did not meet Renton's threshold of being reasonably relevant.8 But the U.S. Supreme Court took a different view. As often happens in First Amendment cases, the Supreme Court's decision in Alameda Books did not produce a clear majority holding. Justice O'Connor authored a plurality opinion, joined by Chief Justice Rehnquist and Justices Thomas and Scalia, with Justice Scalia also filing a concurring opinion. Justice Kennedy filed an opinion concurring in the judgment, but departing from the rationale announced by Justice O'Connor. Justice Souter authored a dissenting opinion, joined by Justices Stevens and Ginsburg and, in part, by Justice Breyer." While acknowledging the limitations of the 1977 study, Justice O'Connor argued that Los Angeles could infer from its study that concentrations of adult activities would also be associated with secondary effects and, thus, that Los Angeles had complied with the evidentiary requirement of Renton. Justice O'Connor's opinion criticized the Ninth Circuit for imposing too high a bar for cities that seek merely to address the secondary effects of adult businesses."' The Ninth Circuit found that the 1977 study did not provide reasonable support for the 1983 amendment because the study focused on the secondary effects associated with a concentration of establishments rather than a concentration of operations within a single establishment. While acknowledging that the city's 1977 study did not assess whether multiple adult businesses operating under one roof were associated with an increase in secondary effects, Justice O'Connor argued that the city could infer that a concentration of operations, no less than a concentration of establishments, would be associated with an increase in negative secondary effects. She also criticized the Ninth Circuit for implicitly requiring that the city must not merely provide reasonable support for a theory that justifies its ordinance, but also prove that its theory is the only plausible one." 28 Id. at 420-30 (majority opinion).

29 Id. at 428. M Id. at 436-37. 1 Id. at 4,87-38.

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 570 2011

2011] ADULT BUSINESSES WITH SECONDARY EFFECTS

571

Justice O'Connor then addressed what evidentiary standard a city would need to meet. After noting that in Renton the Court

"held that a municipality may rely on any evidence that is 'reasonably believed to be relevant' for demonstrating a connection between speech and a substantial, independent government interest," Justice O'Connor wrote: This is not to say that a municipality can get away with shoddy data or reasoning. The municipality's evidence must fairly

support the municipality's rationale for its ordinance.

If

plaintiffs fail to cast direct doubt on this rationale, either by demonstrating that the municipality's evidence does not support its rationale or by furnishing evidence that disputes the municipality's factual findings, the municipality meets the standard set forth in Renton. If plaintiffs succeed in casting doubt on a municipality's rationale in either manner, the burden shifts back to the municipality to supplement the record with evidence renewing support for a theory that justifies its ordinance." Applying this test to the case at hand, Justice O'Connor concluded that, given the early stage of the litigation, the city had complied with the evidentiary requirement of Renton." Both Justice Scalia and Justice Kennedy wrote concurring

opinions. Justice Scalia did nothing more than reiterate his longstanding claim that businesses engaged in "pandering sex" are not protected under the First Amendment and that communities may not merely regulate them with impunity, but may suppress them entirely. 4 Justice Kennedy wrote a lengthy concurring opinion to express concern that "the plurality's application of Renton might constitute a subtle expansion" of what is permitted under that case." Justice Kennedy contended that Alameda Books raised two evidentiary questions for the Court. "First, what proposition does a city need to advance in order to sustain a secondary-effects ordinance? Second, how much evidence is required to support the proposition?""' He argued that the plurality answered only the second question, and while he believed that answer was correct, in 32 Id. at 438-39. ,, Id. at 439, 442.

J., concurring) (citing his opinions in City of Erie v. Pap's A.M., 529 U.S. 277, 310 (2000) (Scalia, J., concurring), and FW/PBS, Inc. v. City of Dallas, 493 U.S. 215, 256-61 (1990) (Scalia, J., dissenting in part and concurring in part)). The holding in FW/PBS was subsequently modified by City of Littleton v. Z.J. Gifts D-4, L.L.C., 541 U.S. 774 (2004). ", Alameda Books, 535 U.S. at 445 (Kennedy,J., concurring). 36 Id. at 449. 34 Id. at 443-44 (Scalia,

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 571 2011

572

CARDOZO ARTS & ENTERTAINMENT

[Vol. 29:565

his view, more attention needed to be paid to the first." The critical inquiry that Justice Kennedy believed the plurality skipped was "how speech will fare under the city's ordinance."" In his view, shared by Justice Souter's dissenting opinion, a "city may not assert that it will reduce secondary effects by reducing speech in the same proportion."" In short, "[tihe rationale of the ordinance must be that it will suppress secondaryeffects-and not by suppressing speech.""' Applying this first step to the ordinance in this case, Justice Kennedy argued that it would have one of two effects when applied to an establishment operating two adult businesses under one roof: one of the businesses must either move or close. Since the latter of these effects cannot lawfully be the rationale for the ordinance-i.e., the city cannot lawfully seek to reduce the amount of secondary effects merely be reducing the number of adult businesses-the city's rationale must be that affected businesses will relocate rather than close and that the resulting dispersion of businesses will reduce secondary effects but not substantially diminish the number of businesses. Having identified the city's "proposition," Justice Kennedy next asked whether the city had presented sufficient evidence to support that proposition. In line with the plurality, Justice Kennedy argued for significant deference to local government fact-finding in making this inquiry. Citing Renton and Young, he contended that cities "must have latitude to experiment, at least at the outset, and that very little evidence is required."" He also cautioned that "[a]s a general matter, courts should not be in the business of second-guessing the fact-bound empirical assessments of city planners," noting: "The Los Angeles City Council knows the streets of Los Angeles better than we do. It is entitled to rely on that knowledge; and if its inference appears reasonable, we should not say there is no basis for its conclusion."" Here, Justice Kennedy found that, for purposes of surviving a motion for summary judgment, the city's proposition is supported by both its 1977 study and "common experience" and that the 1983 ordinance was reasonably likely to reduce secondary effects substantially while reducing the number of adult entertainment businesses very little." In his dissenting opinion, Justice Souter, joined by Justices 37 Id. Id. at 450. Id. at 449. 4o Id. at 449-50. 41 Id. at 451. 42 Id. at 451-52 (citations omitted). 43 Id. at 450-51. ,8 39

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 572 2011

2011] ADULT BUSINESSES WITH SECONDARY EFFECTS

573

Stevens and Ginsburg and in part by Justice Breyer, argued that imposing stricter evidentiary standards on governments would guard against potential abuses. Justice Souter was concerned about what he viewed as the significant risk that courts will approve ordinances that are effectively regulating speech based on government's distaste for the viewpoint being expressed. He stated: Adult speech refers not merely to sexually explicit content, but to speech reflecting a favorable view of being explicit about sex and a favorable view of the practices it depicts; a restriction on adult content is thus also a restriction turning on a particular viewpoint, of which the government may disapprove." For Justice Souter, the risk of viewpoint discrimination may be addressed by imposing on government a requirement that it

demonstrate empirically that the effects exist, that they are caused by the expressive activity subject to the zoning, and that the zoning can be expected either to ameliorate them or to enhance the capacity of the government to combat them (say, by concentrating them in one area), without suppressing the expressive activity itself." Justice Souter claimed that his call for empirical evidence did not impose a Herculean task on government; rather, the harms allegedly caused by adult establishments "can be shown by police reports, crime statistics, and studies of market value, all of which are within a municipality's capacity or available from the distilled

experiences of comparable communities."" He also noted that the need for "independent proof' can vary with the proposition that needs to be established and thus "zoning can be supported by common experience when there is no reason to question it." 4 7 In the final section of his dissent, which Justice Breyer did not join, Justice Souter applied this standard to the case at hand and argued that the city offered neither a rationale nor evidence to support the proposition that an adult bookstore combined with video booths would produce the claimed secondary effects.48 Although Alameda Books reaffirmed Renton in crucial respects, thereby supporting government regulation of adult businesses,

Id. at 457 (Souter,J., dissenting). 15 Id. 46 Id. at 458-59. 17 Id. at 459. 48 Id. at 461-64. 44

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 573 2011

CARDOZO ARTS & ENTERTAINMENT

574

[Vol. 29:565

Justice O'Connor's plurality opinion described exactly how an adult business could challenge such regulations: This is not to say that a municipality can get away with shoddy

data or reasoning.

The municipality's evidence must fairly

support the Imunicipality's the Inunicipality's rationale for its

ordinance. If plaintiffs fail to cast direct doubt on this rationale, either by demonstrating that the Inunicipality's evidence does not support its rationale or by furnishing evidence that disputes the Imunicipality's factual findings, the municipality meets the standard set forth in Renton. If plaintiffs succeed in casting doubt on a municipality's rationale in either

manner, the burden shifts back to the municipality to supplement the record with evidence renewing support for a theory that justifies its ordinance."

The adult entertainment industry, along with some scholars, began to "cast doubt" on the legal rationale local governments rely upon as the justification for their regulation of soon

adult businesses in both of the ways suggested by Justice O'Connor's opinion. As noted previously, these efforts have taken a variety of forms: trying to subject the studies of secondary effects relied upon by local governments to the Daubert standard for scientific evidence in federal litigation; producing jurisdictionspecific studies that purport to show no association between adult

businesses and negative secondary effects;51 and claims that distinct business models"2 and/or specific local conditions" are not associated with the secondary effects demonstrated in the studies relied on by many local governments. With a few exceptions," most of these challenges have failed. The decisions to date indicate that so long as a regulation has a plausible rationale and is supported by at least some evidence, the courts 19 Id. at 438-39 (majority opinion).

50 Datibert v. Merrell Dow Pharm., 509 U.S. 579 (1993). See, e.g., Paul, et al., supra note 5. 1 See, e.g., Linz et al., sufna note 6. 52 See, e.g., Encore Videos, Inc. v. City of San Antonio, 330 F.3d 288, 294 (5th Cir. 2003) (declaring adult business ordinance unconstitutional because none of the secondary effects studies cited in the legislative record had studied "take-home" adult media stores where no adult entertainment is presented or viewed on the premises), opinion claifIied, 352 F.3d 938 (5th Cir. 2003). Contra Doctor John's v. Wahlen, 542 F.3d 787, 793 (10th Cir. 2008) (rejecting claim that the "on-site/off-site" distinction is relevant in initially judging whether a local government reasonably relied on the studies in enacting its regulations); Richland Bookmart, Inc. v. Knox Cnty., 555 F.3d 512, 526 (6th Cir. 2009) (holding that a local government may rely on a study of secondary effects that did not address the particular category of adult business challenging the ordinance). 53See, e.g., Abilene Retail #30, Inc. v. Bd. of Comm'rs, 492 F.3d 1164, 1175 (10th Cir. 2007), cerL denied, 552 U.S. 1296 (2008) (ruling that local government in rural area could not have reasonably relied on studies of secondary effects, none of which examined businesses in an entirely rural area). 54 See, e.g., Encore Videos, 330 F.3d 288; Abilene Retail, 492 F.3d 1164.

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 574 2011

2011] ADULT BUSINESSES WITH SECONDARY EFFECTS

575

continue to show substantial deference to legislatures.' In the following sections, we describe each of the major types of challenges to the rationales for adult entertainment regulations, discuss how those have been treated in the courts, and, where appropriate, critique each in terms of both methodological validity and criminological theory. II. CHALLENGES BASED ON DAUBERT CLAIMS

In the wake of Alameda Books, Daniel Linz, a Professor in the Department of Communication at the University of California, Santa Barbara, who has frequently been retained as an expert witness for adult businesses, ' co-authored an article ' arguing that when the studies relied upon by most jurisdictions are subjected to "appropriate" scrutiny, they are revealed to be seriously flawed methodologically and should not be accepted by courts as supporting the government's claim that it is regulating adult businesses because of their demonstrated association with negative secondary effects. The methodological rules endorsed in the article are derived from the four criteria for admissibility of scientific expert witness testimony stated by Justice Blackmun's opinion in Daubertv. MerrellDow Pharmaceuticals." Following publication of the article, plaintiffs challenging adult business regulations attempted to cast doubt on the government's factual basis for regulating adult businesses by

55See, e.g., G.M. Enters., Inc. v. Town of St. Joseph, 350 F.3d 631 (7th Cir. 2003); Giovani Carandola, Ltd. v. Fox, 396 F. Supp. 2d 630, 651 (M.D.N.C. 2005) (deferring to North Carolina General Assembly's belief that "sexually oriented businesses are associated with higher incidents of crime"). But see R.V.S., L.L.C. v. City of Rockford, 361 F.3d 402 (7th Cir. 2004) (finding it unreasonable for city officials to believe that secondary effects were associated with a business where dancers performed wearing fully opaque clothing over the pubic area, buttocks, and breasts when the city had no evidence of secondary effects associated with such businesses and plaintiffs two experts testified no studies demonstrated adverse secondary effects from such businesses; nor did the experts believe such effects could be found). 5'JULES B. GERARD & ScoTT D. BERGIHOLD, LOCAL REGUILATION OF ADULT BUSINESSES 295 (2011 ed. 2010). 57 Paul et al., supra note 5. Professor Linz's co-authors, Bryant Paul and Bradley Shafer were, at the time the article was published, respectively, a Ph.D. candidate in the UC, Santa Barbara Department of Communication and an attorney in private practice in Lansing, Michigan. Bryant Paul is currently an Assistant Professor in the Department of Telecommunications at Indiana University. Bradley Shafer is still in private practice in Lansing. 58 Daubert v. Merrill Dow Pharm., 509 U.S. 579 (1993). In brief, the four criteria are: (1) Has the scientific theory or technique used by the witness been tested-or can it be tested-for reliability?; (2) Has the theory or technique used been subjected to peer review and publication?; (3) In the case of a particular scientific technique, the court ordinarily should consider the known or potential rate of error; and (4) To what degree has the theory or technique been accepted in the scientific community? Id. at 593-94. Subsequently, in Kumho Tire Co. v. Carmichael,526 U.S. 137 (1999), the Court adopted similar criteria for the admissibility of all evidence. See also Gen. Elec. Co. v. Joiner, 522 U.S. 136 (1997) (addressing issue of the standard of appellate review for admissibility decisions made by courts under Daubert).

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 575 2011

576

CARDOZO ARTS & ENTERTAINMENT

[Vol. 29:565

These attempts have largely been introducing the article.5 discussion of the article in Tenth Circuit's The unsuccessful. Doctor John's v. Whalen," and the subsequent discussion by the Minnesota federal district court in PAO Xiong v. City of Moorhead,"' illustrate the approach to the article by courts that have rejected it as a basis for casting doubt on a city's rationale or evidence. In 2008, the Tenth Circuit stated: The article's main premise is also problematic because it argues that secondary effects studies relied on by municipalities should meet the requirements of Daubert v. MerrellDow Pharm., Inc.... However, the Supreme Court has "flatly rejected [the] idea" of requiring cities to rely on empirical analysis.

. .

. In fact, among

the specific empirical studies that the Supreme Court rejected in City of Erie, were Dr. Linz's studies cited by an amicus curiae, and relied on by the dissent." One year later, the Minnesota federal district court elaborated: Requiring adherence to scientific standards of analysis would be inconsistent with the deference that municipal authorities are given to analyze and address community issues when acting in their legislative function. In fact, adopting such an analytical standard would require municipalities to ignore the valid, but not necessarily scientific, concerns expressed by average citizens in their communities. Further, other courts have concluded that the analysis advocated by the Linz article is insufficient to meet the burden to cast doubt on a municipality's reasoning, even when supplemented by additional evidence, and the Court finds the reasoning employed in those cases persuasive. See Doctor John's v. Wahlen (concluding that the Linz article failed to meet burden to cast doubt because municipality relied on studies not considered by article and because empirical evidence is not required in enactment of ordinances).

In addition, the Linz article's approach largely ignores the factfinding function in which municipalities engage when enacting ordinances. The Linz article acknowledges that existing studies a See, e.g., Doctor John's v. Wahlen, 542 F.3d 787 (10th Cir. 2008); G.M. Enters., Inc. v. Town of St.Joseph, 350 F.3d 631 (7th Cir. 2003). GO DoctorJohnv', 542 F.3d at 787. 61 PAO Xiong v. City of Moorhead, 641 F. Supp. 2d 822 (D. Minn. 2009). 2 I)octorJohn', 542 F.3d at 792 (citations omitted). The court also noted: At first glance, the article does appear to cast doubt on secondary effects studies generally in stating that its authors reviewed 107 relevant studies. However, the article only analyzes the 10 most frequently cited studies by municipalities, and the City of Roy only relies on 4 of those 10 studies. Consequently, it is difficult to see how the article casts doubt on the other 14 studies relied on by the City, let alone the other 7 reports and the many cases cited by the ordinance. Id. at 791-92.

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 576 2011

2011] ADULT BUSINESSES WITH SECONDARY EFFECTS

577

conflict as to whether negative secondary effects arise from adult businesses. Conflicting evidence does not require a municipality to find that negative secondary effects are unlikely to occur. Where a municipality is presented with conflicting evidence, municipal authorities may engage in fact-finding and ultimately may determine that a study finding such a link is more relevant or credible than a study that does not. A municipality may also decide to disregard some studies. The relevant question for courts reviewing these ordinances becomes whether the municipalities reasonably believed that secondary effects were likely to occur." There have, however, been cases in which the Linz article, when submitted with other evidence, has been sufficient to cast doubt on the government's evidence. 4 For example, in Abilene Retail, in addition to the Linz article, the plaintiff submitted five studies indicating that sexually oriented businesses were not the cause of negative secondary effects and called Dr. Linz as an expert witness to critique every study relied on by the government."" This differs from the plaintiff in DoctorJohn's who just submitted the Linz article without specifically addressing each study relied on by the City of Roy."" In the cases where the Linz article was offered along with other types of evidence, it appears that the evidence that addressed the specific studies relied on by the governments involved played a greater role in the courts' decisions that the various plaintiffs had succeeded in casting direct doubt on the government's rationale or evidence. As noted above by the Tenth Circuit and Minnesota federal district court, studies such as the one offered by Dr. Linz would only be valid to cast direct doubt if the Supreme Court required governments to produce empirical data obtained by the methodological standards required by Daubert.6 The Court has refused to impose such a requirement. Rather, the plaintiff must show that the government was not reasonable in its reliance on the evidence in order to meet the direct doubt burden required under the Alameda test. This requires more than just evidence that suggests the government could have reached a different reasonable conclusion. Courts will G, PAO Xiong, 641 F. Supp. 2d at 828 (citation omitted). 64 See Abilene Retail # 30, Inc. v. Bd. of Conin'rs, 492 F.3d 1164 (10th Cir. 2007); 22nd Ave. Station, Inc. v. City of Minneapolis, 429 F. Stipp. 2d 1144 (D. Minn. 2006); Giovani Carandola, Ltd. v. Fox, 396 F. Supp. 2d 630 (M.D.N.C.2005), affd in pait, vacated in part, ev'd in part, 470 F.3d 1074 (4th Cir. 2006). 65 Abilene Retail, 492 F.3d at 1170. G'Doctor John's, Inc. v. City of Roy, No. 1:03-cy-00081, 2007 WL 1302757, at *8-9 (D. Utah May 2, 2007), affd 542 F.3d 787 (10th Cir. 2008). G7 See supra notes 62-63 and accompanying text. See alo G.M. Enters., Inc., v. Town of St. Joseph, 350 F.3d 631, 640 (7th Cir. 2003).

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 577 2011

578

CARDOZO ARTS & ENTERTAINMENT

[Vol. 29:565

defer to local governments' legislative judgments, so long as the judgments meet the Court's standard under Renton and as long as the local government reasonably believed the evidence was relevant to the issue of reducing negative secondary effects. The fact that another, contradictory conclusion may also be reasonable does not cast doubt on a municipality's conclusion. So long as the Renton standard is met, "Alameda Books does not require a court to re-weigh the evidence considered by a legislative body, nor does it empower a court to substitute its judgment in regards to whether a regulation will best serve a community.""

III. CHALLENGES BASED ONJURISDICTION-SPECIFIC STUDIES In a number of challenges to adult business regulations, the plaintiff has produced an expert report, often authored by Dr. Linz and his colleagues, demonstrating that the adult business-or businesses-in the jurisdiction are not associated with negative secondary effects."' On the whole, these challenges have not fared well. As stated recently by a Michigan federal district court in

ABCDE Operating,L.L.C. v. City ofDetroit71 [S]everal other courts have rejected Dr. Linz's studies, finding thern insufficient to rel)lut evidence of secondary effects. See, e.g., Imaginary Images, Inc. v. Evans, 612 F.3d 736, 748 (4th Cir. 2010) ("So while the Linz study and others may well be of interest to legislatures or those formulating policy, it does not provide the kind of 'clear and convincing' evidence needed to rebut the government's showing and invalidate the

s Doctor John's, 2007 Al 1302757, at *10. Similarly, evidence that non-sexually oriented businesses also produce negative secondary effects does not diminish the reasonableness of a legislative decision to regulate only adult businesses. For example, in Peek-A-Boo Lounge of Bradenton,mIc., v. NManatee Cnty., No. 8:05-CV-1707-T-27TBM, 2009 Al 4349319 (M.D. Fla. 2009), the plaintiff offered evidence that sexually oriented businesses have no greater correlation to secondary effects than other types of businesses. Peek-A-Boo, 2009 WL 4349319, at *6. However, the District Court held that this evidence did "little to cast doubt on secondary effects associated with sexually oriented businesses." Id. The court further went on to state that the government "may regulate secondary effects in sexually oriented businesses . . . notwithstanding the existence of secondary effects in other types of businesses." Id. (citing City of Renton v. Playtime Theatres, Inc., 475 U.S. 41, 52-53 (1986)). Even when a plaintiff can cast direct doubt on one piece of evidence cited by the government, she meets her burden only when they cast doubt on all evidence cited by the government. Id. at *5 (citing Daytona Grand, Inc. v. City of Daytona Beach, 490 F.3d at 884 (11th Cir. 2007)). For example, in Doctor John's, the plaintiff presented evidence that directly questioned the legitimacy of four of the fourteen studies cited by the City of Roy in enacting its ordinance. Regardless of whether this evidence was sufficient to cast direct doubt on the studies (the court determined that it was not), the court held that the plaintiff failed to meet its burden because she offered no evidence to undermine the other ten studies cited by the City. DoctorJohn', 2007 WL 1302757 at *9. GDr. Linz and his colleagues have produced such reports in litigation involving the cities of Greensboro, North Carolina, San Diego, California, and Toledo, Ohio among others. T0ABCDE Operating, L.L.C. v. City of Detroit, No. 10-13435, 2011 MT 3607072 (E.D. Mich. Aug. 16, 2011).

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 578 2011

2011] ADULT BUSINESSES WITH SECONDARY EFFECTS

579

regulation."); 84 Video/Newsstand, Inc. v. Sartini, 2007 U.S. Dist. LEXIS 80079, at *22 ("the court finds that Plaintiffs' evidence from Dr. Daniel Linz is not sufficient at this stage to cast direct doubt on Defendants' evidence of a substantial government interest, especially in light of the fact that several courts have rejected Dr. Linz's findings under similar circumstances."); Little Mack Ent. II v. Twp. of Marengo, 625 F. Supp. 2d 570, 580 (W.D. Mich. 2008) ("Contrary to Little Mack's arguments, the affidavit of Daniel Linz, Ph.D., which concludes that the ordinances were based on shoddy data and flawed reasoning, does not undermine the legislative basis for adopting the ordinances."); Pao Xiong v. City of Moorhead, 641 F. Supp. 2d 822, 828-829 (D.Minn. 2009) ("concerns advocated by the Linz article is [sic] insufficient to meet the burden to cast doubt on a municipality's reasoning, even when supplemented by additional evidence"); Doctor John's v. G. Blake Wahlen, 542 F.3d 787 (10th Cir. 2008);J.L. Spoons, Inv. v. Dragani, 538 F.3d 379 (6th Cir. 2008).n In addition to the legal concerns that have led many courts to find that such studies fail to cast doubt on the association of adult businesses with secondary effects, there are sound methodological

reasons why these studies should be rejected. While a comprehensive critique of these various jurisdiction-specific studies is beyond the scope of this article, not to mention the reader's attention span, we note two of the methodological concerns with these studies. First, many of these studies base their measurement of crimes committed in the vicinity of adult businesses on data that is questionable because they use an inappropriate metric for the reporting of crimes. Criminologists use crime incidents (or "crimes known to the police") to measure crime risk. These are traditionally measured through Uniform Crime Reports ("UCRs")." Given this well established convention, it is surprising that many of the jurisdiction-specific reports employ a different measure for crime, Calls for Service ("CFSs") to police, which are generated through calls to the "911" emergency number or to police departments directly. A justification for the use of CFSs instead of UCRs to measure crime was provided in the study of Id. at *4. See also Entm't Prods., Inc. v. Shelby Cnty., No. 08-2047, 2011 WL 3903002 (W.D. Tenn. Sept. 6, 2011), decided after ABCDE Operating and reaching same result. 72 "The Uniform Crime Reporting ... Program was conceived in 1929 by the International Association of Chiefs of Police to meet a need for reliable, uniform crime statistics for the nation. In 1930, the FBI was tasked with collecting, publishing, and archiving those statistics." Law enforcement agencies throughout the United States collect data on crimes reported to police and then provide that data to the FBI. Unifrnm Crite Repods, FED. BUREAU OF INVESTIGATION, http://www.fbi.gov/about-us/cjis/ucr/iuci (last visited Oct. 30, 2011). 71

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 579 2011

580

CARDOZO ARTS & ENTERTAINMENT

[Vol. 29:565

secondary effects in Toledo, Ohio, prepared by Dr. Linz and a colleague: We employ calls for service in this study for four reasons: 1)

The use of these indicators of crine is compatible with criminology research; 2) Studies of secondary effects relied on by the City of Toledo have also employed this measure. It is possible, therefore, to directly compare the findings of the

present study to these studies; 3) CFS are known to be consistent with victirization data; 4) The Justice Department

endorses their use as indicators of criminal activity.' We disagree strongly with these rationales. The third and fourth rationales are unsubstantiated and, in our view, simply incorrect: CFSs are not consistent with victimization data and the U.S. Department of Justice has never endorsed the use of CFSs as a measure of "criminal activity." As regards the first two, in fact, criminologists rarely, if ever, use CFSs to measure crime risk74 and only a few of the secondary effect studies relied on by the City of Toledo used CFSs for any purpose whatsoever. Second, when the conclusions derived from the data presented in many of these studies is examined critically in light of that data, the conclusions are found to be insupportable or even contradictory to the data. Readers who lack statistical backgrounds may wonder how two teams of experts can analyze the same data with the same methods, yet, arrive at radically different conclusions. Simply put, the stark differences between the original analyses in these reports and a subsequent reanalysis are due to differences in the underlying statistical assumptions and differences in the interpretation of analytic results. In our view, the statistical analyses in these studies often are based on highly questionable assumptions and this error is then compounded through a misinterpretation of the study's results. 7 DANIEL LINZ & MIKE YAO, EVALUATING POTENTIAL SECONDARY EFFECTS OF ADULT CABARETS AND VIDEO/BOOKSTORES IN TOLEDO, OHIO: A STUDY OF CALLS FOR SERVICE TO THE POLICE 16 (Feb. 15, 2004) (unpublished study) (on file with authors).

A review undertaken by graduate students under the authors' direction of writings published between 2000 and 2004 in foui national criminology journals, Cnminolog, Justice Quarteriy, the journal of Quantitative Griminolog, and the Journal of (riminalfjutie comprised 705 bibliographic items, primarily articles. Most of the articles were either non-empirical (theoretical essays, reviews, etc.) or else, analyzed phenomena other than crime (police behavior, sentencing decisions, etc.). Of the 254 articles that analyzed a crime statistic, 134 (52.8%) analyzed UCRs; 119 (46.8%) analyzed victim or offender surveys. Only five articles (1.9%) analyzed CFSs. These data reflect the consensus View among criminologists that CFSs are not the best-or even a good-mueasure of crime. See RICHARD MCCLEARY &JAMES W. MEEKER, A METHODOLOGICAL CRITIQUE OF THE LINZ-YAO REPORT: REPORT TO THE CITY OF TOLEDO, OH 17 (May 15, 2004) (unpublished report) (on file with the authors). 75 Two-thirds of the studies relied on by the City of Toledo used UCRs. See id. at 16. 74

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 580 2011

2011] ADULT BUSINESSES WITH SECONDARY EFFECTS

581

Every statistical analysis is predicated on a set of assumptions that, taken together, constitute a "model." If one or more of the predicate assumptions is unwarranted, the model will yield analytic results that are biased in some way. The consequences of this bias can be benign. Results predicated on "wrong" assumptions can still be approximately "right." But the consequences of bias are not always benign. In many instances, the accrued bias violated assumptions can have disastrous consequences. Further, model assumptions notwithstanding, the results of every statistical analysis must be interpreted. Except for results derived from randomized controlled trials (experiments), analytic results cannot be expressed as a single number. Results derived from quasi-experimental designs, such as examining data before or after the opening (or closing) of an adult business or examining data from an area with an adult business compared to an area without an adult business, invariably consist of several numbers which must be integrated. This opens the door to subjectivity. Focusing exclusively on only one of several numerical results can lead to a misinterpretation of the larger set of results. In short, we claim that methodologically appropriate studies of secondary effects that are interpreted in an appropriate manner will always demonstrate an association between adult entertainment businesses and negative secondary effects.

IV.

CHALLENGES BASED ON THE ON-SITE/OFF-SITE DISTINCTION OR THE URBAN/RURAL DISTINCTION

A. The On-Site/Off-Site Distinction While there are a number of different adult entertainment business types," almost all such businesses can readily be classified as either a business where adult entertainment (whether in hardcopy format, electronic media or live performances) is viewed on the premises or where adult entertainment material is sold for viewing off the premises. Encore Videos, Inc. v. City of San Antonio," exemplifies challenges to adult entertainment regulations brought by operators of off-site businesses claiming that the studies relied & CONNIE COOPER, EVERYTHING You ALWAYS WANTED TO KNow ABOUT REGULATING SEX BUSINESSES 27-37 (2000), which lists the following types: 76 See, e.g., ERIC DAMIAN KELLY

Mixed-Retail Outlets, comprising mainstream retail stores with back rooms of sexually explicit material and retail percentage stores; Adults-Only or Sexually Oriented Retail Outlets, comprising adult media outlets and sex shops; Sexually Oriented Entertainment, comprising movie theatres, video-viewing booths, and live entertainment; and Touching and Encounter Businesses, comprising lingerie modeling studios, nude encounter studios, nude photography studios, massage parlors not operated by medical professionals or certified massage therapists and body-painting studios. Encore Videos, Inc. v. City of San Antonio, 330 F.3d 288 (5" Cir. 2003), opinion clarIied, 352 F.3d 938 (5' Cir. 2003), ce. dented, 540 U.S. 982 (2003).

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 581 2011

582

CARDOZO ARTS & ENTERTAINMENT

[Vol. 29:565

on by the enacting jurisdiction either dealt only with on-site businesses or failed to distinguish between the secondary effects associated with on-site businesses versus off-site businesses. In Encore Videos, the San Antonio ordinance classified off-site book and video stores as adult entertainment businesses if their inventories included twenty percent adult material. Citing Alameda Books, an off-site book store challenged the ordinance's rationale and underlying evidence, arguing that San Antonio had relied on studies that either excluded off-site adult businesses or, otherwise, had not distinguished between the effects of on-site and off-site adult businesses. The Fifth Circuit agreed; moreover, in the court's view, the city's rationale for ignoring the differences between on-site and off-site businesses was weak: Off-site businesses differ from on-site ones, because it is only

reasonable to assume that the former are less likely to create harmful secondary effects. If consumers of pornography cannot view the materials at the sexually oriented establishment, they are less likely to linger in the area and engage in public alcohol consumption and other undesirable activities." The Encore Videos approach was subsequently adopted by the Seventh Circuit in Annex Books v. City of Indianapolis."' In that case, the City of Indianapolis relied on studies of secondary effects involving on-site adult businesses to support an ordinance regulating adult bookstores. The court held that if Indianapolis could not produce sufficient evidence for off-site adult businesses, then "its ordinance must meet the same fate as San Antonio's.""

Id. at 295. The court's view on this issue was, however, significantly colored by the fact that the ordinance targeted businesses with as little as twenty percent adult material, and thus could potentially ensnare "mainstream" businesses With adult sections. The court stated: "Given the expansive reach of the ordinance in the instant case, we must require at least some substantial evidence of the secondary effects of establishments that sell adult products solely for off-site consumption. Otherwise, even ordinary bookstores and video stores with adult sections could be subjected to regulation that restricts their First Amendment rights without evidence that they cause 'secondary effects.'" Id. (citation omitted); see also, Giggles World Corp. v. Town of Wappinger, 341 F. Supp. 2d 427 (S.D.N.Y. 2004) (questioning, on motion for summary judgment, secondary effects basis for ordinance regulating businesses with only twenty percent adult material). But see PAO Xiong v. City of Moorhead, 641 F. Supp. 2d 822 (D. Minn. 2009) (approving, on motion for summary judgment, the city's use of studies based on "on-site" adult businesses, coupled with citizen testimonials, to justify an ordinance restricting "off-site" adult businesses which sold more than twenty peicent adult goods). 79 Annex Books, Inc. v. City of Indianapolis, 581 F.3d 460 (7th Cir. 2009). 80 Id. at 467 (citing Encore Videos, 330 F.3d 288); see alo, New Albany DVD, L.L.C. v. City of New Albany, 581 F.3d 556, 560 (7th Cir. 2009) (discounting city's proof as to the undesirable secondary effects of plaintiffs off-site business because the studies the city relied upon did not "fairly support[]" a causal connection between the adult business and the secondary effects the ordinance sought to address). 78

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 582 2011

2011] ADULT BUSINESSES WITH SECONDARY EFFECTS

583

Other courts, however, have explicitly rejected this approach. As early as 1994, the approach had been rejected by the Eighth Circuit in ILQ Investments v. City of Rochester," but that ruling, of course, pre-dated Alameda Books. In a ruling made after Alameda Books, the Tenth Circuit rejected the Fifth Circuit's approach in Encore Videos that the "on-site/off-site" distinction is relevant to the "casting doubt" first step of the Alameda Books analysis. In Dr. John's v. Wahlen," the Tenth Circuit held that the distinction between an on-site and off-site adult business was not relevant when determining whether a local government produced evidence to meet its initial burden. The court did acknowledge that this distinction may become relevant once the burden shifted to the plaintiffs, but the relevance cannot be assumed. Thus, a city may enact regulations on an "off-site" business based on secondary effects that do not focus solely on such businesses. The burden would then shift to the plaintiff to cast doubt on the relevancy of such studies to "off-site businesses."8 3 The Ninth Circuit also upheld a restriction on "off-site" businesses after the Encore Videos ruling, but its decision in World Wide Video of Washington, Inc. v. City of Spokane" focused more on additional evidence upon which the city relied rather than on an outright rejection of the on-site/off-site distinction. In this case, the plaintiffs expert demonstrated that the studies the city relied on did not deal exclusively with "take-out" (i.e., off-site) businesses and provided his own studies showing that such businesses did not cause negative secondary effects in Spokane. The Tenth Circuit ruled that the plaintiff had not met its burden of "casting doubt" on the city's rationale or supporting evidence, despite the plaintiffs expert's findings, because, in addition to the studies from other cities, Spokane had relied on citizen testimony linking "off-site" businesses with pornographic litter and public lewdness, and this evidence, "standing alone, was sufficient to meet the 'very little' evidence standard of Alameda Books."" In contrast, a 2009 Sixth Circuit decision, Richland Bookmart, Inc. v. Knox County, Tenn., 6 implicitly rejected the Encore Videos approach, although it did not explicitly rule that the distinction

81ILQInvs., Inc. v. City of Rochester, 25 F.3d 82 DoctorJohn's v. Wahlen, 542 F.3d 787 (10

1413 (8th Cir. 1994). Cir. 2008). 83 Id. at 793; see also Enlightened Reading, Inc. v. Jackson County, No. 08-0209-CV-W-FJG, 2009 WL 792492 (W.D. Mo. Mar. 24, 2009). 84 World Wide Video of Wash., Inc. v. City of Spokane, 368 F.3d 1186 (9 Cir. 2004), as amended on denial of ieh'g and iehg en banc (July 12, 2004). 85 Id. at 1195; see alo H & A Land Corp. v. City of Kennedale, 480 F.3d 336, 339-41 (5th Cir. 2007) (upholding the city's regulation of "off-site" businesses, and distinguishing Encore Videos, where the city had relied, in part, on evidence from surveys of real estate appraisers that focused strictly on "off-site "businesses). 86 Richland Bookmart, Inc. v. Knox Cnty., 555 F.3d 512 (6th Cir. 2009).

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 583 2011

584

CARDOZO ARTS & ENTERTAINMENT

[Vol. 29:565

between on-site and off-site businesses was irrelevant. In this case, the court reviewed an ordinance aimed at off-site businesses with at least thirty-five percent adult material, which led it to focus on a claimed distinction between businesses that carried a large percentage of adult materials and those businesses that barely met the thirty-five percent threshold for regulation. Ruling that the County had met its burden under Alameda Books, the court flatly rejected the plaintiffs' argument that the studies relied upon by the County were irrelevant on the ground that off-site "combination" stores, defined as stores primarily offering "mainstream" merchandise which barely met the thirty-five percent percentage threshold, are substantially different from stores that have a greater percentage of adult items." The court reasoned: Requiring local governments to produce evidence of secondary effects for all categories created by every articulable distinction is a Inisapprehension of the Supreme Court's holding that governments may rely on any evidence "reasonably believed to be relevant." While the 35% threshold may be arbitrarily chosen, and it very well may be that this threshold sweeps in some relatively benign establishments, it is not for us to decide that some higher, equally arbitrary percentage would lessen the burden on expression without compromising the efficacy of the

Ordinance in controlling secondary effects. The case stands for the proposition that local governments can reasonably rely on evidence of secondary effects associated with a variety of adult businesses and are not required to obtain evidence that any given category of adult business-defined by a

plaintiffs "articulable distinction" of its business category from whatever categories were included in associated with negative secondary effects.

the

city's

studies-is

B. The Urban/RuralDistinction Another type of recent challenge, conceptually similar to the

"on-site/off-site" distinction, focuses on the claim that the studies relied on by a local government are not germane to local conditions, most particularly, the rural nature of a jurisdiction. This claim was accepted by the Tenth Circuit in Abilene Retail #30, Inc. v. Board of Commissioners of Dickinson County, Kansas." The 87

See id. at 526. 88 Id. (citations omitted). 89 Abilene Retail #30, Inc. v. Bd. of Comm'rs, 492 F.3d 1164 (10th Cir. 2007), cert. denied, 552 U.S. 1296 (2008).

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 584 2011

2011] ADULT BUSINESSES WITH SECONDARY EFFECTS

585

court, finding that none of the studies of secondary effects relied upon by the Board examined businesses in an entirely rural area, concluded that the Board could not have "reasonably believed" that a single business in a rural area was associated with secondary effects and ruled that the district court had improperly granted an award of summary judgment to the County. The court stated: All of the studies relied upon by the Board examine the secondary effects of sexually oriented businesses located in ulrban environments; none examine businesses situated in an entirely rural area. To hold that legislators may reasonably rely on those studies to regulate a single adult bookstore, located on a highway pullout far from any business or residential area within the County would be to abdicate our "independent judgment" entirely. Such a holding would require complete deference to a local government's reliance on prepackaged secondary effects studies from other jurisdictions to regulate any single sexually oriented business, of any type, located in any setting."0 The panel was split, however, on precisely where in the Alameda Books analysis the Board was required to look for evidence of secondary effects other than the "pre-packaged" studies it had relied on. The majority of the panel argued that the Board's reliance on such studies failed to meet even the initial burden in Alameda Books." In a concurring opinion, one member of the

panel argued that the Board could have reasonably believed that such studies supported the notion that even a single business in a rural location could be associated with negative secondary effects, but that since the plaintiff adult business had presented evidence to refute that belief, the burden had shifted to the Board to find further evidence linking the single rural business with negative secondary effects."'

V. THEORETICAL AND CASE STUDY SUPPORT FOR THE ASSOCIATION OF ADULT BUSINESSES WITH SECONDARY EFFECTS

Taken at face value, many of the claims made by plaintiffs

But see Independence News, Inc. v. City of Charlotte, 568 F.3d 148 (4th Cir. 2009) (ruling that city need not show that an individual adult business actually produces negative secondary effects in order to enforce ordinance because the absence of any evidence of adverse secondary effects associated with a given business today is no guarantee that such effects will not be present tomorrow), cert. denied, 130 S.Ct. 507 (2009). 91 Id. at 1175-76. '2 Id. at 1181-85 (Ebel,J., concurring). 90 Id. at 1175.

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 585 2011

586

CARDOZO ARTS & ENTERTAINMENT

[Vol. 29:565

and their experts in challenges to adult entertainment ordinances seem attractive, or even compelling, from a common sense perspective. The problem is that these claims either ignore theoretically relevant characteristics of adult businesses or are methodologically flawed. In particular, such claims ignore the routine activity theory of crime associated with adult businesses or use inappropriate data sources and methods to demonstrate that adult businesses are not associated with secondary effects or both. A. The PrevailingCriminologicalTheory of Secondary Effects" The prevailing criminological theory of secondary effects is derived as a special case from the routine activity theory of crime." Applied to secondary effects, the theory can be written as:

N of Targets x Expected Value Ambient Crime Risk =

x N of Predators

PolicePresence

In simple terms, the routine activities at an adult business site attract predators, generating a "hot spot of predatory crime."' The relative attractiveness of a site is determined by the number of targets at the site, their expected value, and the level of police presence at the site. Sites with a relatively large number of high-value targets and a relatively low police presence attract a relatively large number of predators. The hotspot theory assumes a pool of rational predators who move freely from site to site, choosing sites with high-value targets and low police presence. Because these predators lack legitimate means of livelihood and devote substantial time to illegitimate activities, they are "professional thieves" by Edwin Sutherland's classic definition.' Otherwise, they are a heterogeneous group. 93

This section is adapted from Richard McCleaiy, Rural Ilotspots: The Case of Adult Businesses, 19 CRIM.JUST. POL'YREV. 153 (2008). 94 See generally Lawrence E. Cohen & Marcus Felson, Social Change and Cime Rate Tiends: A Routine Activity Approach, 44 AM. SOC. REV. 588 (1979); Marcus Felson & Lawrence E. Cohen, Human Ecology and Cuime: A Routine Activities Appoach, 8 HUMAN ECOLOGY 389 (1980). 95 Lawrence W. Sherman et. al, Hot Spots of Piedatoy Cime: Routine Activities and the Ciminology oPlace,27 CRIMINOLOGY 27 (1989); see also Paul J. Brantingham & Patricia L. Brantingham, Mtes on the Geometry (YCrine, in ENVIRONMENTAL CRIMINOLOGY 27-54 (Paul J. Brantingham & Patricia L. Brantingham ects., 1981); Patricia L. Brantingham & PaulJ. Brantingham, Ndes, Paths and Edges: Considerations on the Comlexity of(Crime and the Physical Environment, 13J. ENVTL. PSYCHOL. 3 (1993). 96 See EDWIN H. SUTHERLAND, THE PROFESSIONAL THIEF I (reprin t 1989).

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 586 2011

2011] ADULT BUSINESSES WITH SECONDARY EFFECTS

587

Some are vice purveyors who dabble in crime; others are criminals who use the promise of vice to lure and lull their victims. Despite their heterogeneity, these predators share a rational decisionmaking calculus that draws them to adult business sites. The crime-vice connection has been a popular plot device for at least 250 years. John Gay's The Beggar's Opera," written in 1728, for example, centers on the relationship between MacHeath, a predatory criminal, and the vice ring composed of Peachum, Lucy, and Jenny. This popular view is reinforced by the empirical literature on criminal lifestyles and thought processes. The earliest and best-known study, Clifford Shaw's The Jack-Roller, written in 1930, describes "Stanley," a delinquent who lives with a prostitute and preys on her clients." Criminological thinking concerning the connection between crime and vice has changed little in the eight decades since Shaw's Jack-Roller. To document the rational choices of predatory criminals, Richard Wright and Scott Decker interviewed eighty-six active armed robbers." Asked to describe a perfect victim, all mentioned victims involved in vice, either as sellers or buyers. Three of the armed robbers worked as prostitutes: "From their perspective, the ideal robbery target was a married man in search of an illicit sexual adventure; he would be disinclined to make a police report for fear of exposing his own deviance."0 0 Moreover, the rational calculus described by these prostitute-robbers echoes the descriptions of other predators." Like tourist attractions'12 and sporting events," adult business sites attract targets from wide catchment areas. Compared to the targets attracted to these other hotspots, however, adult business patrons are disproportionately male, open to vice overtures, and

97 JOHN

GAY, THE BEGGAR'S OPERA (Biyan Lotighrey & T.O. Treadwell eds., Penguin

Books 1986) (1728). 8

! CLIFFORD R. SHAW, THE JACK-ROLLER: A DELINQUENT BOY'S OWN STORY (1966); see alo JON SNODGRASS, THEJACK-ROLLER AT SEVENTY (Lexington Books 1982) (1930). 99 RICHARD T. WRIGHT & SCOTT H. DECKER, ARMED ROBBERS IN ACTION: STICKUPS AND STREET CULTURE (1997). 100 Id. at 69. 101See, e.g., TREVOR BENNETT & RICHARD WRIGHT, BURGLARS ON BURGLARY: PREVENTION AND THE OFFENDER (1984); MARK S. FLEISHER, BEGGARS & THIEVES: LIVES OF URBAN STREET CRIMINALS (1995); JACK KATZ, SEDUCTIONS OF CRIME: MORAL AND SENSUAL ATTRACTIONS IN DOING EVIL (1988); NEAL SHONER, GREAT PRETENDERS: PURSUITS AND CAREERS OF PERSISTENT THIEVES (1996); Floyd Feeney, Robbers as Decisionv-Makers, in THE REASONING CRIMINAL: RATIONAL CHOICE PERSPECTIVES ON OFFENDING 53-71 (Derek B.

Cornish & Ronald V. Clarke eds., 1986); Jack Katz, The MVotivation of the Pessitent Robber, in CRIME AND JUSTICE: A REVIEW OF RESEARCH (Michael Tonry ed., 1991). 102 See, e.g., Terry A. Danner, Violent Times: A Case Study of/the Ybor City Historic Disthict, 14 Crim. Just. Pol'y Rev. 3 (2003); Fr6duric Dimanche & Allena Lepetic, New O)rlieans Tourism and (rime: A Case Study, 38 J. TRAVlL RES. 19 (1999). 104 See, e.g., Jonathan J. Corcoran et al., Predicting the Geo-Temporal Variations of(Crime and Disorder, 9 INT'L J. FORECASTING 623 (2003); Jeffrey Westcott, Super Bowl XXXIX.- The Sucessful Respon.se of the FBI and Its Partners,FBI L. ENFORCEMENT BULL.,Jan. 2006, at 1.

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 587 2011

588

CARDOZO ARTS & ENTERTAINMENT

[Vol. 29:565

carry cash. When victimized, moreover, adult business patrons are reluctant to involve the police. They are "perfect" victims, from the rational predator's perspective. Given a choice among equally lucrative sites, of course, rational predators prefer sites with low levels of police presence. In the original statement of the routine activity theory, Lawrence Cohen and Marcus Felson included police in the pool of "capable guardians," along with the targets themselves and any potential witnesses."" The stigma associated with adult business sites limits the effectiveness of non-police guardians, however. With minor exceptions, the level of guardianship at adult business sites is proportional to the level of physical (e.g., motor or foot patrols) or virtual (e.g., security cameras) police presence. Of course, the relative effectiveness of a fixed level of police presence can be affected by broadly defined environmental factors. Because conventional police patrolling is less effective in darkness, ambient crime risk rises after dark.1 115 Architectural structures that obscure vision can have a similar effect but outdoor lighting can be used to mitigate the risk.""' Because rural areas have fewer police resources,"' rational predators may view rural adult business sites as more attractive.108 B. The Role of Adult Business Types in CriminologicalTheoy Secondary effects are realized in terms of "victimless" vice crimes (prostitution, drugs, etc.), predatory personal crimes (assault, robbery, etc.), predatory property crimes (theft, auto theft, etc.), "opportunistic" nuisance crimes (vandalism, trespass, etc.), and disruption of order (public drunkenness, disturbing the peace, etc.). Nevertheless, within these broad etiological categories, criminological theory allows for quantitative and qualitative differences among distinct adult business models. These differences accrue through either of two mechanisms. First, the distinguishing characteristic of an adult business model can create idiosyncratic opportunity structures for a particular type of crime. Second, the distinguishing characteristic can reduce the effectiveness of common policing strategies. Compared to the complementary model, for example, adult 10 See Cohen & Felson, supranote 94; Felson & Cohen, supra note 94. o5 Timothy Coupe & Laurence Blake, Daylight and Darkness Taigeting Srategies and the Risks

o/fBeing Seen at Residential Buiglaries, 44 CRIMINOLOGY 431, 437 (2006). 1o6 DAVID P. FARRINGTON & BRANDON C. WELSH, EFFECTS OF IMPROVED STREET LIGHTING ON CRIME: A SYSTEMATIC REVIEW 2-3 (2002). 0 See, e.g., QUINT C. THURMAN & EDMUND F. MCGARRELL, COMMUNITY POLICING IN A RuRAL SETTING (1997); cf. RALPH A. WEISHEIT ET AL., CRIME AND POLICING IN RuRAL AND SMALL-TOWN AMERICA 154 (3d ed. 2006) (claiming rural prosecutors must work with limited resour ces). 108 McCleary, supra note 93, at 157.

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 588 2011

2011] ADULT BUSINESSES WITH SECONDARY EFFECTS

589

businesses that serve alcohol present opportunities for noninstrumental personal crimes (assault, disorderly conduct, etc.). Proactive policing strategies are also less effective for adult businesses that serve alcohol. The private viewing booths that distinguish on-site from offsite adult businesses lead to differences through both mechanisms. First, the booths generate idiosyncratic opportunities for "victimless" vice crimes. Second, the booths pose a special problem for policing. Other than victimless vice crimes, however, there are no differences between the on-site and off-site adult business models. To the extent that both models attract highvalue targets from wide catchment areas, both are expected to attract predators to their neighborhoods, thereby generating ambient victimization risk. As will be seen in the case study discussion below, the data corroborate this theoretical expectation. C. The Role of Adult Business Location in CriminologicalTheoy The

Tenth Circuit's Abilene Retail "

decision

effectively

accepted the proposition that studies of secondary effects associated with adult businesses in urban areas could not be relied upon to support the regulation of adult businesses in a rural area. Because most criminological research has been conducted in nonrural areas, criminological theories do not necessarily generalize to rural crime. Because relatively little crime occurs in rural areas, of course, few criminologists are interested in urban-rural questions. Thus, the potential cost of the Tenth Circuit's decision was staggering. At minimum, local governments would be forced to undertake studies of adult businesses located in more rural areas, rather than relying on existing studies from less rural jurisdictions. In the absence of such studies, adult businesses could have an incentive to relocate to rural areas since local governments could not demonstrate that they had a factual basis for enacting regulatory ordinances. In either case, the ability of local governments to mitigate public safety hazards associated with adult businesses would be compromised. Of course, if criminological theories can be generalized to rural areas, then the Abilene Retail decision may be called into question. Although the generalization may be difficult for some criminological theories, the relevant theory of "hotspots," discussed previously in the context of distinguishing between onsite and off-site adult businesses as regards their association with

109 Abilene Retail #30, Inc. v. Bd. of Comm'rs, 492 F.3d 1164 (10th Cir. 2007), cer. denied,

552 U.S. 1296 (2008).

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 589 2011

590

CARDOZO ARTS & ENTERTAINMENT

[Vol. 29:565

secondary effects,"o applies to any accessible area, rural or urban, and can explain why an adult business that opens in a sparsely populated rural community could be associated with an increase in ambient crime risk, in effect making a hotspot of the community. Writing a decade after the advent of Uniform Crime Reports in 1930,"' George Vold confirmed that a city's crime rate was

proportional to its population."' The observed relationship had an obvious explanation: "[B] ehavior in the country in all probability comes under much greater informal control of the opinions and disapprovals of the neighbors than is the case in the relative anonymity of the city . . . .""

The negative correlation

confirmed not only grand sociological theory,114 but also the related criminological theory of social disorganization. As proposed by Shaw and McKay1 5 in 1942, the theory of social disorganization predicts that neighborhoods with low residential stability will have high rates of delinquency and vice versa. To the extent that a small town has the characteristics of a stable neighborhood, social disorganization theory would predict the low crime rates observed by Vold."" When a small town is disrupted by an influx of newcomers, however, the same theory predicts an abrupt increase in the town's crime rate. This increase in crime can occur in at least two ways. First, the newcomers may victimize the town's residents. Indeed, fear of victimization by newcomers is implicated in the rapid spread of gated communities.' Second, the influx of newcomers may disrupt the town's routine activities in a way that attracts predatory criminals, creating a local "hot spot of predatory crime. "118 The outline we previously presented of criminological theory regarding the creation of a local "hot spot of predatory crime" applies equally in the rural setting, with one significant difference. With respect to the quantity and quality (or value) of the targets at a site, urban and rural adult business sites are equally attractive to 110 See sjnna notes 89-104 and accompanying text.

III For information

about Uniform Crime Reports, see supra note 72.

112 George B. Vold, Cite in City and Count) Areas, ANNALS AM. ACAD. POL. SOC. SC., Sept.

1941, at 38. 11s 1(1. 114 See, e.g., FERDINAND TONNIES, COMMUNITY AND SOCIETY (Dover 2002) (reprint of 1887 title "Gemeinschaft und Gesellschaft"); EMILE DURKHEIM, THE DIVISION OF LABOR IN SOCIETY (W.D. Halls trans., 1984) (1893). 115 CLIFFORD R. SHAW & HENRY D. McKAY, JUVENILE DELINQUENCY AND URBAN AREAS

(1942). its See McCleary, supra note 93. 117 SARAH D. BLANDY ELT AL., GATED COMMUNITIES: A SYSTEMATIC REVIEW OF THE RESEARCH EVIDENCE (2003), available at http://york.academia.edu/RowlandAtkinson/Papers/272572/Gated Communities A Sy stematic Review of the Research Evidence. 18 Sherman et. al., sina note 95.

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 590 2011

2011] ADULT BUSINESSES WITH SECONDARY EFFECTS

591

the rational offender; however, police presence is generally lower at rural sites. Some part of this urban-rural disparity in policing is due to obvious factors. Rural police agencies protect larger areas with fewer personnel, for example, and drive longer distances in response to calls. Though less obvious, "fuzzier" jurisdictional lines and more complex demands for service make policing more difficult and less effective in rural areas." Because police presence is relatively lower at rural sites, controlling for the quantity and quality of targets, rural sites are more attractive to the rational offender. D. Case Studies Confirming the PrevailingCriminologicalTheory of Secondary Effects The authors have previously published two case studies that confirm the prevailing criminological theory of secondary effects as applied, respectively, to an adult off-site business and an adult business located in a rural area."" The case study involving an adult off-site business was conducted in Sioux City, Iowa. The case study involving an adult business located in a rural area was conducted in Montrose, Illinois. Each case study is briefly described below. 1. Off-Site Business Case Study Sioux City, Iowa is located on the Missouri River, which forms the border between Iowa and Nebraska. In 2010 it had a population of 82,684."' Adult businesses are nothing new to Sioux City, Iowa. Two adult businesses had operated without incident in the city's older downtown area for decades. Although both businesses sold sexually explicit DVDs for off-site use, most of their revenue came from coin-operated viewing booths. In terms of "look and feel," the two businesses were indistinguishable from adult businesses in larger cities. In March 2004, a third adult business, Dr. John's, opened in Sioux City. Unlike the two existing businesses, Dr. John's had no viewing booths. It was located in a newer area of the city and lacked the garish appearance often associated with adult businesses generally and, in particular, with Sioux City's two existing adult businesses. During subsequent litigation, the trial judge commented on this fact:

IN See supra note

107. See Richard McCleary & Alan C. Weinstein, Do "OSite"Adult Busineses Have .Secnday Effects? Legal Doctine, Social Theot, and Empihical Evidece, 31 LAw POL. 217 (2009); McCleary, supranote 93. 121 Sioux City, Iowa, U.S. CENSUS BUREAU, http://quickfacts.census.gov/qfd/states/19/1973335.html (last revised Oct. 18, 2011, 3:26 PM). 120

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 591 2011

592

CARDOZO ARTS & ENTERTAINMENT

[Vol. 29:565

[T]he first impression of the store is a far cry from the first image that most people would likely have of an "adult book store" or "sex shop." There is nothing seedy about the neighborhood, store building, or store front. In fact, from a quick drive-by, one would likely assume that the business was a rather upscale retail store for wonen's clothing and accessories. There are no "adult" signs or banners proclaiming "peep shows," "live entertainment booths," "XXX movies," "live models," "adult massage," or any of the other tasteless corme-ons all too familiar from adult entertainment stores that exist in virtually every American city of any size and which one may find scattered along interstates and highways even in rural America."' The trial judge's drive-by impression may overstate the point. Few passers-by would mistake Dr. John's for anything other than what it was. Regardless of its look and feel, Dr. John's was located in a prohibited zone. When Sioux City attempted to enforce its zoning code, Dr. John's sued, arguing that off-site adult businesses lacked the typical crime-related secondary effects associated with adult businesses. To counter this argument, Sioux City produced police reports of incidents occurring within 500 feet of Dr. John's during the four years between January 1, 2002, and December 31, 2005. This time period comprised 793 days before and 668 days after Dr. John's opened. For purposes of quasi-experimental control, reports of incidents occurring with 500 feet of a nearby motel were also retrieved. The data showed that in the area within 500 feet of Dr. John's, the annual crime rate rose from 7.8 to 22.4 incidents per year, an increase of approximately 190 percent. Crime in the 500 foot area surrounding the control area-the motel-rose as well but the increase was more modest: an increase from 20.3 to 25.1 incidents per year amounts to a twenty-five percent increase. Based on a crude comparison of these rates, Dr. John's appears to pose an ambient victimization risk. Of course, this assumes that other plausible alternative hypotheses can be ruled out. As more fully discussed in our "Off-Site" article,' statistical analysis demonstrated that the various plausible alternative hypotheses-for example, the "null hypothesis" that the before and after effect was simply due to chance-were unlikely to have produced the

122 Doctor John's, Inc. v. City of Sioux City, 389 F. Supp. 2d 1096, 1103 (N.D. Iowa 2005)

(quoting from court's ruling on plaintiffs motion for preliminary injunction). 123 McCleary & Weinstein, supra note 120.

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 592 2011

2011] ADULT BUSINESSES WITH SECONDARY EFFECTS

593

observed effect. Due to the small number of incidents that occurred in the areas surrounding Dr. John's and the control site during 20022005, the statistical significance of the secondary effects shown in the data could be arguable. Critics might characterize the evidence as interesting but not compelling. That is the nature of case study evidence. By the Renton criteria, on the other hand, the data constitute a sufficient factual predicate for the regulation of off-site adult businesses. In light of the strong underlying theory described earlier, which predicts such results, few criminologists would find the case study results surprising or controversial. 2. Rural Case Study: Montrose, Illinois An unincorporated village of fewer than 250, Montrose, Illinois is located on 1-70 midway between St. Louis, Missouri, and Indianapolis, Indiana.124 Interstate 70 separates Montrose's residential dwellings from its businesses: a convenience store-gas station, a motel, and for a short period, a tavern. Other than gas and lodging, cross-country travelers had no reason to exit I-70 at Montrose prior to February 2003. In that month, the Lion's Den, an adult business, opened on a service road within 750 feet of the 1-70 off-ramp. A large, elevated sign let 1-70 travelers know that Xrated videos, books, and novelties could be purchased "24/7." The store was successful by all accounts. The residents of Montrose did not welcome the new business. Unlike the village's other businesses, the Lion's Den was located on the residential side of 1-70. Complaining that the store disrupted their idyllic life-style, villagers picketed the site on several occasions. Traffic was a chronic complaint. The narrow gravel access road connecting the site to 1-70 could not support the weight of big-rig trucks; it soon fell into disrepair. The Lion's Den offered to build a new, larger access road from 1-70 to its site. But fearing an even larger volume of traffic, the villagers declined the offer. Like many Illinois villages, Montrose had no adult business ordinances. However, the Lion's Den was located within 1,000 feet of a public park in violation of an Illinois statute.2 When the State moved to enforce its statute, the Lion's Den sued, arguing that "off-site" adult businesses could not generate the public safety hazards associated with adult cabarets, video arcades, and other on-site adult entertainment businesses. The trial in Illinois v. Lion's Montrose (Illinois, USA), CITY http:// ww .citypopulation.de/php/usa-census-illinois.php?cityid=1750283 Apr. 3, 2011). 124 See

125 65 ILL. COMP. STAT. ANN. 5/11-5-1.5 (West 2009).

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 593 2011

POPULATION,

(last updated

594

CARDOZO ARTS & ENTERTAINMENT

[Vol. 29:565

Den lasted four days. The court upheld the statute and, in July 2005, the Montrose Lion's Den closed its doors.12 At the trial, the State presented evidence of the Lion's Den's adverse impact on the surrounding area: sexually explicit litter and decreased use of the nearby park. While neither party presented local crime data, our examination of the available data bearing on the crime-related secondary effects of the adult business in Montrose shows that during the 1,642-day period beginning January 1, 2002, the Effingham County Sheriff's Office recorded eighty-three crime incidents in the Village. The most common incidents involved the theft or destruction of property. Incidents of disorder and indecency, traffic-related incidents, and alcohol-drug offenses were nearly as common. Incidents involving danger or harm to persons (robbery, assault, etc.) were rare. When this 1,642 day period is separated into an 881-day segment in which the Lion's Den was open and a 761-day segment in which it was closed, crime rates are 22.39 and 13.92 total incidents per year for the "Open" and "Closed" segments respectively. From these raw rates, it appears that crime in Montrose rose when the Lion's Den opened and fell when the Lion's Den closed. As with the previous case-study, this assumes that plausible alternative hypotheses for the difference can be ruled out, which they were for this case study as well. 12 7 In short, following the opening of an adult business on an interstate highway off-ramp into a sparsely populated rural village, total crime in the village rose by approximately sixty percent. Two years later, when the business closed, total crime in the village dropped by approximately sixty percent. In light of the strong before and after quasi-experimental design, the only plausible explanation for this effect is that, like adult businesses in urban and suburban settings, adult businesses in sparsely populated rural areas generate ambient crime-related secondary effects. This finding was not unexpected. Although criminological theories are based largely on data collected in urban and suburban areas, the routine activity theory of hotspots12" generalizes to rural settings. Put simply, adult businesses attract patrons from wide catchment areas. Because these patrons are disproportionately male, open to vice overtures, and reluctant to report victimizations, their presence attracts offenders. The spatio-temporal conjunction of targets and offenders generates

126

Illinois ex rel. Deters v. Lion's Den, Inc., No. 2004-CH-26 (June 10, 2005), available at http://www.adultbusinesslawN.com/downloads/27_ModifiedPIOrdei.pdf, a//'d 936 N.E.2d 1240 (Ill. App. Ct 2007). 127 McCleary, supra note 93, at 158-60. 128 See Sherman, supra note 95.

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 594 2011

2011] ADULT BUSINESSES WITH SECONDARY EFFECTS

595

ambient victimization risk-a hotspot of predatory crime. This theoretical mechanism operates identically in rural, suburban, and urban areas. Moreover, because rural areas ordinarily have lower levels of visible police presence, rural hotspots may be riskier than their suburban and urban counterparts.'' In contrast, "common sense" secondary effect theories rely on the premise that an "average" patron spends less time at an off-site adult business and, hence, faces a relatively smaller victimization risk at the business. This implies that off-site adult businesses have relatively smaller secondary effects. If the secondary effects of offsite adult businesses are no larger than the effects of, say, a convenience store, there may be no legitimate basis for regulating off-site adult businesses qua adult businesses. Plaintiffs have made this theoretical argument, of course, and in the absence of either theory or evidence to the contrary, courts have accepted it." The argument and underlying theory assume that the "average" patron drives up to the store, runs in, makes a purchase, runs out, and drives off. Although this "average" behavior may be consistent with common sense, it is inconsistent with the data: an ethnographic study of an off-site adult business reports that patrons spend significant periods of time in the immediate vicinity of the site. ' Some wait outside until the business is empty. Others "case" the business on multiple occasions before deciding to enter. Some patrons park their cars a block or more away and walk to the store. These "average" behaviors attract criminal predators to the site, creating the ambient public safety hazard predicted by the criminological theory of secondary effects. CONCLUSION

The ongoing efforts by the adult entertainment industry to discredit the secondary effects rationale local governments rely upon as the justification for their regulation of adult businesses have enjoyed only limited success. The fundamental reason for this is that, contrary to the industry's claims, methodologically appropriate secondary effects studies confirm criminological theory's prediction that adult businesses are associated with heightened incidences of crime regardless ofjurisdiction, business model or location. Further, the studies that have been produced 129 For a discussion of some of the regulatory issues that are posed by adult businesses

locating in rural communities, see Matthew L. McGinnis, Note, Sex, but Nt the City: AdultEntertainment Zoning, the List Amendment, and Residential and Rural Municipalities, 46 B.C. L. REV. 625 (2005). Hn See, e.g., Encore Videos, Inc. v. City of San Antonio, 330 F.3d 288, 295 (5" Cir. 2003), opinion ctlaijied,352 F.3d 938 (5' Cir. 2003), cert. denied, 540 U.S. 982 (2003). 1st Kristen Hefley, Stigma Management of Male and Female Customers to a Non-Grhan Adult Novelty Store, 28 DEVIANT BEHAV. 79 (2007).

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 595 2011

596

CARDOZO ARTS & ENTERTAINMENT

[Vol. 29:565

on behalf of the adult industry are frequently flawed either in their methods or in the analyses of their findings.

HeinOnline -- 29 Cardozo Arts & Ent. L.J. 596 2011

SENATE BILL NO. 1144 SECONDARY EFFECTS MATERIALS

I. SELECTED STUDIES A. Connie B. Cooper, FAICP, and Eric Damian Kelly, Ph.D., FAICP, Survey of Texas Appraisers: Secondary Effects of Sexually-Oriented Businesses on Market Values, and Richard McCleary, Ph.D., Crime-Related Secondary Effects: Secondary Effects of “OffSite” Sexually-Oriented Businesses, prepared for Texas City Attorneys Association (2008) B. Richard McCleary, Rural Hot Spots; The Case of Adult Bookstores, Criminal Justice Policy Review, 2008, 19(2), pp. 153-163 C. Report of Richard McCleary, Ph.D., Illinois ex rel. Deters v. The Lion’s Den, Inc., No. 04-CH-26 (Ill. 4th Jud. Ct. 2005) D. A Report on the Secondary Impacts of Adult Use Businesses in the City of Denver (1998) E. National Law Center for Children and Families, NLC Summaries of “SOB Land Use” Studies: Crime Impact Studies by Municipal and State Governments on Harmful Secondary Effects of Sexually-Oriented Businesses (1996) F. Report of Peter R. Hecht, Ph.D. to the American Center for Law and Justice on the Secondary Impacts of Sex Oriented Businesses (1996) II. CASE STUDY: J.L. SPOONS, INC. V. OHIO DEPARTMENT OF PUBLIC SAFETY A. J.L. Spoons, Inc. v. Ohio Department of Public Safety, 31 F. Supp. 2d 933 (N.D. Ohio 2014) B. Report of Daniel Linz, Ph.D., Rule 52 (2010) C. Richard McLeary, Ph.D., & Wendy Regoeczi, Ph.D., Secondary Effects of SexuallyOriented Businesses: Report to the Ohio Attorney General (2010) III. SELECTED CASES AND GENERAL LEGAL ANALYSIS A. 84 Video/Newstand, Inc. v. Sartini, 455 Fed. Appx. 541 (6th Cir. 2011) B. Nite Moves Entertainment, Inc. v. City of Boise, 153 F. Supp. 2d 1198 (D. Idaho 2001) C. Alan C. Weinstein & Richard D. McCleary, The Association of Adult Businesses with Secondary Effects: Legal Doctrine, Social Theory, and Empirical Evidence, 29 Cardozo Arts & Ent. L.J. 565 (2012)

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/242123188

Rural HotspotsThe Case of Adult Businesses Article in Criminal Justice Policy Review · January 2008 DOI: 10.1177/0887403408315111

CITATIONS

READS

7

21

1 author: Richard Mccleary University of California, Irvine 51 PUBLICATIONS 1,979 CITATIONS SEE PROFILE

All content following this page was uploaded by Richard Mccleary on 28 October 2015. The user has requested enhancement of the downloaded file. All in-text references underlined in blue are added to the original document and are linked to publications on ResearchGate, letting you access and read them immediately.

Rural Hot Spots: The Case of Adult Businesses

Richard McCleary

Keywords: secondary effects, hotspots, ambient crime risk, adult businesses, rural crime N of words: 3,796; N of tables: 1

Criminal Justice Policy Review, 2008, 19(2), pp. 153-163. Send correspondence to [email protected] Paul Brantingham, Marcus Felson, and Alan Weinstein read early drafts. I also benefited from conversations with the late Dennis W. (“Denn”) Roncek.

Abstract A recent decision of the U.S. Court of Appeals for the Tenth Circuit questions whether the routine activity theory of hotspots applies to adult entertainment businesses located in sparsely populated rural areas. Although few criminologists are interested in urban-rural differences, the Tenth Circuit decision makes this topic acutely relevant to policy-makers and courts. To address the threshold question, I demonstrate that the hotspot theory is general to urban, suburban, and rural locations. I then present the results of a case study that corroborates the theory. When an adult entertainment business opens on an interstate highway off-ramp to a small rural village, total crime in the village rises by sixty percent. Two years later, when the business closes, total crime drops by sixty percent. Alternative explanations related to uncontrolled threats to internal validity are considered and ruled out. After reporting the results of the case study, I discuss the consequences of the theory and results for policy-makers.

Introduction Expressive activities that occur inside adult entertainment businesses, including cabarets that feature live nude or semi-nude dancing, x-rated video arcades, and bookstores, enjoy First Amendment protection. Courts have ruled that governments may regulate these businesses, nevertheless, so long as the regulations are aimed at mitigating the businesses’ potential adverse “secondary effects” (Andrew, 2002). To defend an ordinance, a government must produce evidence to show that the businesses are associated with secondary effects such as ambient noise, litter, and in particular, crime. The government’s evidence need not satisfy arbitrary standards of methodological rigor. On the contrary, the 1986 U.S. Supreme Court decision in City of Renton v. Playtime Theatres holds that governments may rely on any evidence “reasonably believed to be relevant to the problem that the city addresses.” Taking advantage of this evidentiary standard, few governments conduct local secondary effects studies; most rely on the large body of studies conducted in other places and times. The U.S. Supreme Court reviewed the evidentiary standard sixteen years later. Though reaffirming the modest “reasonably believed to be relevant” Renton standard, in City of Los Angeles v. Alameda Books, the Court allowed adult businesses to challenge the relevance of secondary effects evidence. If a business could demonstrate that the government’s evidence was irrelevant to the problem that its ordinance purported to address, the ordinance could be struck down. Relevance challenges fall into two categories. The first centers on the fact that secondary effects studies have typically ignored salient differences among distinct adult business models. In Encore Videos v. City of San Antonio, an adult bookstore argued that its products were sold for “off-site” use only and, thus, that it could not have the same secondary effects as cabarets, video arcades, and other “on-site” adult businesses. Accepting part of this argument, the Fifth Circuit struck down a San Antonio ordinance whose evidentiary predicate failed to include secondary effects studies of “off-site” adult bookstores. An ambiguous passage in the Encore Videos decision left the impression that the Fifth Circuit had endorsed an interpretation of criminological theory favoring the plaintiffs. Citing the ambiguous passage, “off-site” adult businesses argued subsequently that criminological theory precluded secondary effects for their business model. Four years later, however, in H and A Land Corp. v. City of Kennedale, the Fifth Circuit upheld an ordinance whose evidentiary predicate included studies of “off-site” adult bookstores. The three-judge panel, including one member who had participated in the Encore Videos decision, took the unusual step of retracting the passage that seemed to endorse an interpretation of criminological theory (McCleary and Weinstein, 2007). The second category of Constitutional challenges centers on the fact that secondary effect studies have ignored idiosyncratic local conditions. In 2004, an adult bookstore in rural Kansas used criminological theory to argue that the sparsely-populated rural environment precluded the

R U R A L H O TSPO TS - P AG E 2

possibility of secondary effects. And since the local government had not studied this issue prior to enactment, the ordinance should be struck down. Rejecting this argument, the trial court granted the defendant’s summary judgment motion. On appeal, however, in Abilene Retail #30 v. Dickinson County, the Tenth Circuit agreed with the plaintiff’s interpretation of criminological theory: All of the studies relied upon by the Board examine the secondary effects of sexually oriented businesses located in urban environments; none examine businesses situated in an entirely rural area. To hold that legislators may reasonably rely on those studies to regulate a single adult bookstore, located on a highway pullout far from any business or residential area within the County would be to abdicate our “independent judgment” entirely. Such a holding would require complete deference to a local government’s reliance on prepackaged secondary effects studies from other jurisdictions to regulate any single sexually oriented business of any type, located in any setting (p. 1175). Because the adult bookstore was located in an isolated rural area, and because the County had no evidence to suggest that rural adult businesses would have secondary effects, the Tenth Circuit reversed the summary judgment and remanded the case for trial. Although the question of urban-rural generality is only one of many weighed in the Tenth Circuit’s decision, it is the central question of this essay. Because most criminological research has been conducted in non-rural areas, criminological theories do not necessarily generalize to rural crime. Because relatively little crime occurs in rural areas, of course, few criminologists are interested in urban-rural questions. Following the Tenth Circuit’s Abilene Retail decision, on the other hand, urban-rural differences are acutely relevant to policy-makers and courts. The potential cost of the decision is staggering. In the best case, local governments will be forced to rewrite ordinances to cover businesses located in more rural areas. In the worst case, litigious adult businesses will have an incentive to relocate to rural areas, forcing trial courts to judge the relative ruralness of areas, case-by-case. In any case, extrapolating the Tenth Circuit’s argument to other variables not explicitly addressed by criminological theory threatens the ability of local governments to mitigate public safety hazards associated with adult businesses. This essay addresses the threshold question of whether criminological theories can be generalized to rural areas. Although the generalization may be difficult for some criminological theories, the relevant theory of “hotspots” (Sherman, Gartin, and Buerger, 1989) applies to any accessible area, rural or urban. After describing the relevant criminological theory, I report the results of a corroborating quasi-experimental case study. When an adult business is opened on an interstate highway off-ramp in a sparsely populated rural community, ambient crime risk rises precipitously, in effect making a hotspot of the community.

R U R A L H O TSPO TS - P AG E 3

The Criminological Theory of Secondary Effects Writing shortly after the advent of Uniform Crime Reports, Vold (1941) confirmed that a city’s crime rate was inversely proportional to its population. The observed relationship had an obvious explanation: “[B]ehavior in the country in all probability comes under much greater informal control of the opinions and disapprovals of the neighbors than is the case in the relative anonymity of the city (p. 38).” The negative correlation confirmed not only grand sociological theory (e.g., Tönnies, 1963[1887]; Durkheim, 1964[1893]) but, also, the related criminological theory of social disorganization. As proposed by Shaw and McKay (1942), the theory of social disorganization predicts that neighborhoods with low residential stability will have high rates of delinquency and vice versa. To the extent that a small town has the characteristics of a stable neighborhood, social disorganization theory would predict the low crime rates observed by Vold (1941). When a small town is disrupted by an influx of newcomers, moreover, the same theory predicts an abrupt increase in the town’s crime rate. This can occur in at least two ways. First, the newcomers may victimize the town’s residents. Indeed , fear of victimization by newcomers is implicated in the rapid spread of gated communities (Blandy, Lister, Atkinson, and Flint, 2003). Second, the influx of newcomers may disrupt the town’s routine activities in a way that attracts predatory criminals, creating a local “hot spot of predatory crime” (Sherman, Gartin, and Buerger, 1989). The discovery of hotspots by Sherman, Gartin, and Buerger (1989) was anticipated by the work of Brantingham and Brantingham (1981); adult business hotspots have many of the properties associated with crime “attractors” and “generators” (see also Brantingham and Brantingham, 1993). A simpler routine activity theory (Cohen and Felson, 1979; Felson and Cohen, 1980; Clarke, 1983; Felson, 1998) is sufficient for present purposes, however. In this context, the routine activity theory of crime equates ambient crime risk, generally defined as the number of crimes within 500-1000 feet of a site, with the product of four risk factors. This can be written as: Targets x Expected Value Ambient Crime Risk

=

× Offenders Police Presence

An increase (or decrease) in the number of targets at the site or in their expected value, defined in the usual way, yields an increase (or decrease) in ambient crime risk. An increase (or decrease) in police presence, on the other hand, yields a decrease (or increase) in ambient crime risk. Targets Adult business sites are crime hotspots because they attract potential victims, or targets,

R U R A L H O TSPO TS - P AG E 4

from wide catchment areas. Adult business sites are no different in that respect than tourist attractions (Dimanche and Lepetic, 1999; Danner, 2003) and sporting events (Corcoran, Wilson and Ware, 2003; Westcott, 2006). Compared to the targets found at these better known hotspots, however, the targets found at adult businesses are exceptionally attractive to offenders. This reflects the presumed characteristics of adult business patrons. They are disproportionately male, open to vice overtures, and carry cash. Most important of all, when victimized, they are reluctant to involve the police. From the offender’s perspective, they are “perfect” victims. Offenders The crime-vice connection has been a popular plot device for at least 250 years. John Gay’s Beggar’s Opera (1728), for example, describes the relationship between MacHeath, a predatory criminal, and the vice ring composed of Peachum, Lucy, and Jenny. This popular view is reinforced by the empirical literature on criminal lifestyles and thought processes. The earliest and best-known study (Shaw, 1930; Snodgrass, 1982) describes the life of “Stanley,” a delinquent who lives with a prostitute and preys on her clients. This simple application of the routine activity theory assumes a pool of rational offenders who move freely from site to site, choosing to work the most attractive site available. These offenders lack legitimate means of livelihood and devote substantial time to illegitimate activities; they are “professional thieves” by Sutherland’s (1937) definition. Otherwise, they are a heterogeneous group. Some are vice purveyors who dabble in crime. Others are predatory criminals who promise vice to lure and lull their victims. Despite their heterogeneity, the offenders share a rational decision-making calculus that draws them to adult business sites. Expected Value Criminological thinking has changed little in the 75 years since Shaw’s (1930) JackRoller. To document the rational choices of predatory criminals, Wright and Decker (1997) interviewed 86 active armed robbers. Asked to describe a perfect victim, all mentioned victims involved in vice, either as sellers or buyers. Three of the armed robbers worked as prostitutes: From their perspective, the ideal robbery target was a married man in search of an illicit sexual adventure; he would be disinclined to make a police report for fear of exposing his own deviance (p. 69). The rational calculus described by these prostitute-robbers echoes the descriptions of other predators (see Bennett and Wright, 1984; Feeney, 1986; Fleisher, 1995; Katz, 1988, 1991; Shover, 1996).

R U R A L H O TSPO TS - P AG E 5

Police Presence With respect to the quantity and quality (or value) of the targets at a site, urban and rural adult business sites are equally attractive to the rational offender. Police presence is generally lower at rural sites, however. Some part of the urban-rural disparity is due to obvious factors. Rural police agencies protect larger areas with fewer personnel, for example, and drive longer distances in response to calls. Though less obvious, fuzzier jurisdictional lines and more complex demands for service make policing more difficult and less effective in rural areas (Thurman and McGarrell, 1997; Weisheit, Falcone, and Wells, 1999). Because police presence is relatively lower at rural sites, controlling for the quantity and quality of targets, rural sites are more attractive to the rational offender. Montrose, Illinois: A Case Study An unincorporated village of 250 residents, Montrose, Illinois is located on I-70 midway between St. Louis and Indianapolis. I-70 separates Montrose’s residential dwellings from its businesses: a convenience store-gas station, a motel, and for a short period, a tavern. Other than gas and lodging, cross-country travelers had no reason to exit I-70 at Montrose prior to February, 2003. In that month, the Lion’s Den opened on a service road within 750 feet of the I-70 offramp. A large, elevated sign let I-70 travelers know that x-rated videos, books, and novelties could be purchased “24/7.” The store was successful by all accounts. The residents of Montrose did not welcome the new business. Unlike the village’s other businesses, the Lion’s Den was located on the residential side of I-70. Complaining that the store disrupted their idyllic life-style, villagers picketed the site on several occasions. Traffic was a chronic complaint. The narrow gravel access road connecting the site to I-70 could not support the weight of big-rig trucks; it soon fell into disrepair. The Lion’s Den offered to build a new, larger access road from I-70 to its site. But fearing an even larger volume of traffic, the villagers declined the offer. Like all Illinois villages, Montrose had no adult business ordinances. The Lion’s Den was located within 1,000 feet of a public park, however, in violation of an Illinois statute. When the State moved to enforce its statute, the Lion’s Den sued, arguing that “off-site” adult businesses could not generate the public safety hazards associated with adult cabarets, video arcades and other on-site adult entertainment businesses. The trial in State v. The Lion’s Den et al. lasted four days. The court upheld the statute and, in July, 2005, the Montrose Lion’s Den closed its doors. At the trial, the State presented evidence of the Lion’s Den’s adverse impact on the surrounding area: sexually explicit litter and decreased use of the nearby park. Neither party presented local crime data, however. Table 1 reports data bearing on the crime-related secondary effects of the adult business in Montrose. During the 1,642-day period beginning January 1, 2002, the Effingham County Sheriff’s Office recorded 83 crime incidents in the Village. The most common incidents involved the theft or destruction of property. Incidents of disorder and

R U R A L H O TSPO TS - P AG E 6

indecency, traffic-related incidents, and alcohol-drug offenses were nearly as common. Incidents involving danger or harm to persons (robbery, assault, etc.) were rare.

Table 1 - Montrose, IL: Crime-Related Secondary Effects Open

Closed Log Effect

Property Crimes Personal Crimes All Other Crimes

23 3 28

9.54 1.24 11.61

15 5 9

7.20 2.40 4.32

Total Crimes

54

22.39

29

13.92

Constant Open

ë -3.267 0.475

t -17.60 2.06

e0.475 . 1.61

The columns labeled “Open” and “Closed” in Table 1 break the incidents down into an 881-day segment in which the Lion’s Den was open and a 761-day segment in which it was closed. Crime rates are 22.39 and 13.92 total incidents per year for the “Open” and “Closed” segments respectively. From these raw rates, it appears that crime in Montrose rose when the Lion’s Den opened and fell when the Lion’s Den closed. This assumes that plausible alternative hypotheses for the difference can be ruled out, of course. Null Hypothesis The most obvious alternative explanation is that the difference is due to chance. To rule this out, the daily total crime count series was regressed on a binary variable representing “Open” and “Closed” days (Cameron and Trivedi, 1998) . The log-parameter values reported in Table 1 were estimated with Stata 9.2 (Stata Corporation, 2007). Because the effect estimate ë=0.475 occurs with probability p(t>2.09) < 0.035, by the conventional 95 percent confidence criterion, the chance explanation, or null hypothesis, is rejected. Although parameter estimation requires working in the natural log metric, log-parameters are not easily interpreted. The exponentiated effect estimate is approximately equal to the ratio of the segments, however. In this instance, the value (e0.475=) 1.61 is interpreted as a 61 percent difference. The rate of total crime in Montrose was 61 percent higher during the 29 months that the Lion’s Den was open, that is, compared to the period prior to February 2003, before the Lion’s Den opened, and the period after July 2005, when it closed. This is a large, statistically significant crime-related secondary effect. Internal Validity Another set of alternative explanations involve uncontrolled threats to internal validity. The switching regime (closed-open-closed) property of the quasi-experimental design controls

R U R A L H O TSPO TS - P AG E 7

many of the most common threats to internal validity. Nevertheless, authorities on quasiexperimental design (Campbell and Stanley, 1966; Cook and Campbell, 1979; Shadish, Cook and Campbell, 2003) cite maturation, history, and instrumentation as the most plausible threats to the internal validity of time series designs. The threat of maturation refers to the possibility that the effect reported in Table 1 may be due, not to the opening of the Lion’s Den, but to a natural trend in the Village’s crime rate. Because the daily time total crime time series satisfies the simple Poisson homogeneity assumption (Feller, 1968), however, the maturation hypothesis is rejected. The threat of history refers to the possibility that the effect may be due to some event in the Village that coincided with the opening of the Lion’s Den. A search of local news media found only one significant event during the 1,662-day time series. Shortly after the Lion’s Den opened, the Village’s only liquor-serving tavern closed permanently. If the tavern’s closing had any effect on crime in Montrose, however, the expected effect would be to reduce the crime rate during the 881 days that the Lion Den was open. Accordingly, history is rejected as an alternative hypothesis. Instrumentation refers to the possibility that the effect may be due, not to the opening of the Lion’s Den, but to a coincidental change in the way that crimes are recorded in the Village. If the Effingham County Sheriff stepped up the frequency of patrols in the Village when the Lion’s Den opened, for example, the effect reported in Table 1 might be a spurious artifact of heightened surveillance. Criminologists acknowledge that heightened surveillance can exaggerate “victimless” crime rates; proactive enforcement against prostitution and drugs invariably leads to higher vice crime rates. Proactive enforcement against “serious” crime does not not produce higher rates of homicide, assault, and robbery, however. On the contrary, criminologists generally agree that heightened surveillance reduces the rate of “serious” crime. The detailed incident reports do not support an instrumentation hypothesis. During the 881 days that the Lion’s Den was open, crime in the Village grew more “serious.” Although five “Personal Crimes” were reported during the 761 days that the Lion’s Den was closed, versus three when it was open, none of the five incidents involved a weapon or resulted in an injury. When the Lion’s Den was open, in contrast, two of the three “Personal Crimes” reported in the Village were armed robberies, one committed by a gang of four men wearing ski masks and armed with shotguns. Both armed robberies were committed at site of the Lion’s Den, moreover, and were the only robberies recorded in the Village’s modern history. The timing of the crime incidents is related to their seriousness. During the 761 days that the Lion’s Den was closed, Montrose’s modal crime incidents were “drive-off” thefts from the Village’s gasoline station and vandalism at the Village’s motel. Most of these incidents occurred during the day and required no immediate response from the Sheriff’s Office; and because the businesses were separated from residences by I-70, the modal incidents attracted little attention. During the 881 days that the Lion’s Den was open, on the other hand, a majority of incidents

R U R A L H O TSPO TS - P AG E 8

occurred at night and demanded immediate response; as more incidents began to occur on the residential side of I-70, crime became more noticeable to Village residents. Discussion Following the opening of an adult business on an interstate highway off-ramp into a sparsely populated rural village, total crime in the village rose by approximately 60 percent. Two years later, when the business closed, total crime in the village dropped by approximately 60 percent. In light of the strong quasi-experimental design, artifactual explanations for this effect, including maturation, history, and instrumentation are implausible. The only plausible explanation for the effect reported in Table 1 is that, like adult businesses in urban and suburban settings, adult businesses in sparsely populated rural areas generate ambient crime-related secondary effects. This finding was not unexpected. Although criminological theories are based largely on data collected in urban and suburban areas, the routine activity theory of hotspots (Sherman, Gartin, and Buerger, 1989) generalizes to rural settings. Put simply, adult businesses attract patrons from wide catchment areas. Because these patrons are disproportionately male, open to vice overtures, and reluctant to report victimizations, their presence attracts offenders. The spatio-temporal conjunction of targets and offenders generates ambient victimization risk – a hotspot of predatory crime. This theoretical mechanism operates identically in rural, suburban, and urban areas. Because rural areas ordinarily have lower levels of visible police presence, moreover, rural hotspots may be riskier than their suburban and urban counterparts. The Tenth Circuit may not have found the Montrose results useful. Every case study is unique in some respect, after all; and although the U.S. Census Bureau considers both Effingham County, Illinois and Dickinson County, Kansas to be “rural,” the Tenth Circuit may have focused on idiosyncratic, legally relevant factors. Nevertheless, the case study results demonstrate that, whether urban, suburban, or rural, hotspots are hotspots. In urban, suburban, and rural areas, adult businesses attract patrons who are disproportionately male, open to vice overtures, and reluctant to report victimizations to the police. This attracts offenders to the site with predictable consequences for ambient crime risk. In theory, of course, due to the relative scarcity of police in rural areas, offenders may find rural hotspots more attractive. Otherwise, the routine activity theory of hotspots generalizes to any site that is attractive to potential victims, or targets, and accessible to offenders. Solving the problem of rural hotspots by allocating more police resources to rural areas is politically unfeasible. Governments allocate public safety resources across regions on utilitarian grounds. Per capita allocations have the greatest impact on per capita crime rates. This poses an obstacle to rural problem-oriented policing (Weisheit, Falcone, and Wells, 1999), of course, but it is a rational policy for a government. Because the targets attracted to the rural hotspot live outside the jurisdiction, and because victimizations are under-reported, ignoring the hotspot is a more realistic strategy.

R U R A L H O TSPO TS - P AG E 9

The future is unclear. The relocation of adult businesses to rural areas parallels the postwar “flight” of inner-cities families. From the perspective of adult business proprietors, the urban environment has become hostile. Zoning codes force adult businesses into “ghettos” where their operations are strictly regulated and where competition with other adult businesses is fierce. Rural areas have few regulations, on the other hand, and little competition; access to interstate highway traffic is a bonus. As urban environments become more hostile, more adult businesses will relocate to rural areas, forcing state and county governments into policy decisions. The case study reported here can, hopefully, inform that debate. References Andrew, C. J. (2002) The secondary effects doctrine: the historical development, current application, and potential mischaracterization of an elusive judicial precedent. Rutgers Law Review, 54, 1175. Bennett, T. and R. Wright. (1984) Burglars on Burglary: Prevention and the Offender. London: Gower. Blandy, S., D. Lister, R. Atkinson and J. Flint. (2003) Gated Communities: A Systematic Review of the Research Evidence. CNR Paper #12. Bristol and Glasgow: Economic and Social Research Council. Brantingham, P.L., & Brantingham, P.J. (1981). Notes on the geometry of crime. In P.L. Brantingham, & P.J. Brantingham (Eds.), Environmental criminology (pp. 27-54). Thousand Oaks, CA: Sage. Brantingham, P.L., & Brantingham, P.J. (1993). Nodes, paths and edges: considerations on environmental criminology. Journal of Environmental Psychology, 13, 3-28. Cameron, A.C., & Trivedi, P.K. (1998). Regression analysis of count data, econometric society: Monograph 30. Cambridge: Cambridge University Press. Campbell, D.T. & Stanley, J.C. (1966). Experimental and quasi-experimental designs for research. Chicago: Rand-McNally. Clarke, R.V. (1983). Situational crime prevention: its theoretical basis and practical scope. In M. Tonry, & N. Morris (Eds.), Crime and justice: Annual review of research (pp. 1-63). Chicago: University of Chicago Press. Cohen, L.E., & Felson, M. (1979). Social change and crime rate trends: a routine activity approach. American Sociological Review, 44, 588-608.

R U R A L H O TSPO TS - P AG E 10

Cook, T.D., & Campbell, D.T. (1979). Quasi-experimentation: Design and analysis issues for field settings. Skokie, IL: Houghton-Mifflin. Corcoran, J.J., Wilson, I.D., & Ware, J.A. (2003). Predicting the geo-temporal variations of crime and disorder. International Journal of Forecasting, 19, 623-634. Danner, T.A. (2003). Violent times: A case study of the Ybor City Historic District. Criminal Justice Policy Review, 14, 3-29. Dimanche, F., & Lepetic, A. (1999). New Orleans tourism and crime: A case study. Journal of Travel Research, 38, 19-23. Durkheim, E. (1964[1893]). The division of labor in society. Glencoe, IL: The Free Press. Feeney, F. (1986). Robbers as decision-makers. In D. Cornish, & R.V. Clarke (Eds.), The reasoning criminal: Rational choice perspectives on offending (pp. 53-71). New York: Springer-Verlag. Feller, W . ( 1968). An introduction to probability theory and its applications: Volume I (3rd ed.). New York: Wiley. Felson, M. (1998). Crime and everyday life (2nd ed.). Thousand Oaks, CA: Pine Forge Press. Felson, M., & Cohen, L.E. (1980). Human ecology and crime: A routine activities approach. Human Ecology, 8, 389-406. Fleisher, M. S. (1995). Beggars and thieves: Lives of urban street criminals. Madison: University of Wisconsin Press. Gay, J. (2006[1728]). The beggar’s opera. New York: Hard Press Editions. Katz, J. (1988). Seductions of crime: Moral and sensual attractions in doing evil. New York: Basic Books. Katz, J. (1991). The motivation of the persistent robber. In M. Tonry (Ed.), Crime and justice: A review of research (pp. 277-306). Chicago: University of Chicago Press. McCleary, R. & Weinstein, A.C. (2007). Do “off-site” adult businesses have secondary effects? Legal doctrine, social theory, and empirical evidence. Paper presented at the annual meeting of the American Society for Criminology, Atlanta, GA. Shadish,W.R., Cook, T.D., & Campbell, D.T. (2002). Quasi-experimental designs for generalized causal inference. Skokie, IL: Houghton-Mifflin.

R U R A L H O TSPO TS - P AG E 11

Shaw, C.R. (1966[1930]). The jack-roller: A delinquent boy's own story. Chicago: University of Chicago Press. Shaw, C.R., & McKay, H.D. (1942). Juvenile delinquency and urban areas. Chicago: University of Chicago Press. Sherman, L.W., Gartin, P.R., & Buerger, M.E. (1989). Hot spots of predatory crime: Routine activities and the criminology of place. Criminology, 27, 27–56. Shover, N. (1996). Great pretenders: Pursuits and careers of persistent thieves. Prospect Heights, IL: Westview Press. Snodgrass, J. (1982). The jack-roller at seventy. Lexington, MA: Lexington Books. Stata Corporation (2007). Stata Version 9.2. College Station, TX: Stata Corporation. Sutherland, E.H. (1937). The professional thief. Chicago: University of Chicago Press. Thurman, Q., & McGarrell, E.G. (1997). Community policing in a rural setting. Cincinnati, OH: Anderson. Tönnies , F. (1963[1887]). Community and society (Gemeinschaft und gesellschaft). New York: Harper and Row. Vold, G.B. (1941). Crime in city and country areas. Annals of the American Academy of Political and Social Science, 217, 38-45. Weisheit, R.A., Falcone, D.N., & Wells, L.E. (1999). Crime and policing in rural and smalltown America (2nd ed.). Prospect Heights, IL: Waveland Press. Westcott, J. (2006). Super Bowl XXXIX. FBI Law Enforcement Bulletin, 75, 1. Wright, R.T., & Decker, S.H. (1997). Armed robbers in action: Stickups and street culture. Boston: Northeastern University Press. Cases and Decisions City of Renton v. Playtime Theatres, Inc., 475 U.S. 41 (1986) City of Los Angeles v. Alameda Books, Inc., 535 U.S. 425 (2002) Encore Videos, Inc. v. City of San Antonio, 330 F.3d 288 (5th Cir. 2003)

R U R A L H O TSPO TS - P AG E 12

H and A Land Corp. v. City of Kennedale, TX, 480 F.3d 336 (5th Cir. 2007) Abilene Retail #30, Inc. v. Board of Commissions of Dickinson County, Kansas, 492 F.3d 1164, 1175 (10th Cir. 2007) People of the State of Illinois ex rel. Edward C. Deters, State’s Attorney of Effingham County v. The Lion’s Den, Inc., et al. 4th Ill. Judicial Circuit Court, No. 04-CH-26 (June 10, 2005)

View publication stats

IN THE CIRCUIT COURT FOR THE FOURTH JUDICIAL CIRCUIT OF ILLINOIS EFFINGHAM COUNTY, EFFINGHAM, ILLINOIS

PEOPLE OF THE STATE OF ILLINOIS ex rel. EDWARD C. DETERS, State’s Attorney of Effingham County Plaintiff v. THE LION’S DEN, INC., a foreign business corporation incorporated in the State of Ohio; EFFINGHAM RETAIL 27, INC., a foreign business corporation incorporated in the State of Ohio, MICHELLE MOORE and REAL ESTATE located at 401 Frontage Road, in the village of Montrose, Effingham County, Illinois, Defendants.

) ) ) ) ) ) ) ) ) ) ) ) ) ) ) ) )

Case No. 04-CH-26

REPORT OF RICHARD McCLEARY, Ph.D. I am a Professor at the University of California, Irvine with appointments in the Departments of Environmental Health Science, Criminology, and Planning. My training and experience qualify me as an expert in the areas of criminology and statistics. Throughout my career, I have applied these areas of expertise to the problem of measuring site-specific public safety hazards, especially the public safety hazards associated with sexually-oriented businesses that are ordinarily called “crime-related secondary effects.” My curriculum vitae and a list of cases in which I have been deposed or testified as an expert is appended to this report. The plaintiff in this case has retained me to express an opinion on the central questions in this case. These questions are: Question 1: Do sexually-oriented businesses, as a general class, pose significant

ambient public safety hazards? Question 2: Given an affirmative answer to the first question, how valid is evidence on which this opinion is based? To form an opinion on these two questions, I reviewed the materials sent to me by the Effingham County State’s Attorney and the empirical studies of crime-related secondary effects cited in this report. I also consulted the relevant authorities on statistics and criminology and other relevant literature, cited in this report. My opinions also rely on my training and background in statistics and criminology, of course, and especially, on my research in the area of crime-related secondary effects. Based on my review of materials, studies, and authorities, at a trial in this case, I would express the following opinions. Opinion 1: As a class, sexually-oriented businesses pose significant ambient public safety hazards. These hazards involve not only “victimless” crimes (prostitution, e.g.) but, also, the “serious” crimes (robbery, e.g.) and “opportunistic” crimes (vandalism, e.g.) that are associated with vice. The ambient public safety hazard (or crime victimization risk) can be ameliorated by regulation. Opinion 2: The criminogenic nature of sexually-oriented businesses is a scientific fact. This opinion is based on two considerations. First, strong, empirically-validated criminological theory predicts that crime victimization risks will be higher around sexually-oriented business sites as a consequence of the normal commercial activities at the site. Second, this theoretically expected secondary effect has been observed in a diverse range of locations, circumstances, and times. Although the magnitude and nature of the observed crime-related secondary effect varies from REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 2

case to case, every adequately designed study has observed and reported a large, significant effect. In other legal proceedings, Experts retained by the sexually-oriented business industry have argued that the consistent empirical finding of the crime-related secondary effects literature is a methodological artifact of the weak quasi-experimental designs used in this literature.1 I disagree. These same experts have conducted analyses that purport to show that sexuallyoriented businesses have no crime-related secondary effects or, in a few instances, salutary impacts on ambient crime. Again, I disagree with the industry experts. Before addressing the empirical evidence, however, I will outline the relevant criminological theory. 1. Criminological theory of secondary effects. Crime-related secondary effects studies consistently find that sexually-oriented businesses generate significant ambient public safety hazards. The cumulative finding of this literature is more convincing because it is predicted by a strong, empirically-validated criminological theory. Modern criminological theory holds that the victimization risk at a site is determined by three factors: (1) The number of potential victims (or targets) at the site; (2) the “hardness” of the site’s targets; and (3) the number of potential offenders at the site.2 Holding target-hardness

1

See, e.g., Paul, B., D. Linz and B.J. Shafer, “Government regulation of ‘adult’ businesses through zoning and anti-nudity ordinances: de-bunking the legal myth of negative secondary effects.” Communication Law and Policy, 2001, 6:355-391. 2

The source of this so-called “Routine Activities Theory” is L.E. Cohen and M. Felson "Social change and crime rate trends: A routine activity approach." American Sociological Review, 1979, 44:588-608. See also, M. Felson’s Crime and Everyday Life, Second Edition (Thousand Oaks, CA: Pine Forge Press, 1998). Ignoring one mathematical technicality, this three-factor theory can be written formally as: CRIME RISK = [ TARGETS X OFFENDERS ] / HARDNESS Crime risk rises if the number of targets or offenders rises; crime risk falls as target hardness REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 3

constant, sites with greater numbers of offenders or targets have higher ambient risks. Holding the numbers of offenders and targets constant, on the other hand, sites with softer targets have higher ambient risks. Target-hardening strategies range from devices designed to increase security around the target (lighting, dead-bolt locks, etc.) to proactive guardianship (uniformed guards, police patrols, etc.). Reducing the density of offenders (by controlling site access, e.g.) or targets (by dispersing the targets across sites, e.g.) can also reduce ambient risk in principle. But as a practical matter, crime reduction programs operate primarily through target-hardening.3 1.1 Victims or targets. Applied to sexually-oriented businesses, this three-factor criminological theory predicts that adult bookstores, video arcades, peep-shows, nude dancing clubs and the like will generate large, significant crime risks. The risk phenomenon is driven by the victim factor. Sexuallyoriented businesses can draw relative large numbers of potential victims to a common site. The density of potential victims attracts predatory criminals to the site. Business practices designed to attract more customers to the site (sales, advertising, etc.) aggravate the risk. To the extent that many types of businesses try to attract more customers, this factor is not exclusive to sexually-oriented businesses. Customers attracted to sexually-oriented businesses have characteristics (i.e., the desire for anonymity) that make them attractive targets, however. 1.2 Target hardness The qualities of sexually-oriented business patrons make them soft targets for predatory

rises. 3

The classic statement on target-hardening is Oscar Newman’s Defensible Space: Crime Prevention Through Urban Design. (New York: MacMillan, 1973). REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 4

criminals. Many travel long distances to the site, for instance, and are often strangers or outsiders to the area.4 To avoid stigmatization, some patrons use aliases and pay in cash; and worse, when victimized, tend not to complain to or seek assistance from the police. In addition to characteristics that make sexually-oriented business patrons soft targets, the three-factor theory also points to physical properties of the site. Virtually all sites have structural features – alleyways, adjacent buildings, fences, etc. – that hide or obscure actors and activities. These structures, which constitute one important class of site-specific risk factors, aggravate ambient risk through either of two mechanisms. First, to the extent that a structural feature can be used to obscure criminal activities, the feature will make the site more attractive to predatory criminals. As more potential criminals are drawn to the site, the site’s victimization risk rises. Second, to the extent that a structural feature makes routine police patrolling of the site more difficult, the effectiveness of police crime reduction procedures diminishes. For this reason, zoning regulations often include features (security lighting, removing visual obstructions, etc.) that facilitate police patrolling and related crime reduction procedures. 1.2.1 Alcohol makes soft targets softer In theory, proximity to alcohol aggravates the ambient crime phenomenon through two mechanisms. First, proximity to a bar or tavern makes a sexually-oriented business site more attractive relative to competing sites; opening a tavern near a sexually-oriented business site will increase the concentration of targets at the site. Second, to the extent that alcohol lowers 4

In 1990, as part of an investigation, Garden Grove police officers ran registration checks on motor vehicles parked at sexually-oriented businesses. Virtually all of the vehicles were registered to addresses outside Garden Grove. The 1986 Austin, TX study arrived at the same finding. REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 5

personal inhibitions and clouds judgment, proximity to a tavern will make sexually-oriented business patrons more vulnerable to predatory criminals; i.e., it will make soft targets softer. 1.3 Expected crimes and criminals Criminological theory predicts that the public safety hazard posed by sexually-oriented businesses will be realized in three broad categories of crime: Predatory, “victimless,” and opportunistic crimes. Predatory crimes, like robbery and assault, are perpetrated in the victims’ presence and, for that reason, each is potentially fatal. “Victimless” crimes, including drugs, lewd behavior, and prostitution, are perpetrated with the victims’ active participation. Aside from the fact that these crimes are illegal per se, victimless crimes are often linked to predatory crimes through the perpetrator. This link can be a rationale for police vice control activities. Finally, crimes of opportunity, including some acts of vandalism, thefts, and burglaries, are perpetrated outside the victims’ presence, often by predatory criminals drawn to the site by opportunities for other crimes. The criminal predators who are attracted to the sexually-oriented business site by the quantity and quality of victims can be thought of as “professional” criminals. Most lack legitimate means of livelihood and devote substantial time to illegitimate activities. With few exceptions, predators and victims are not acquainted and this has theoretical implications for the types of crimes expected. 1.4 Regulatory amelioration Since predatory criminals are attracted to sexually-oriented business sites by the concentration of soft targets, ambient crime risk can be ameliorated by hardening the sites’ targets. To the extent that regulations, including those that govern public nude conduct, focus

REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 6

public safety attention and police resources on a site, the regulations will ameliorate the public safety hazard. Inspections and routine visible police presence in a neighborhood have the effect of reducing victimization risk through a complex set of pathways.5 Effective regulations vary across jurisdictions, of course, but to the extent that (apparently) idiosyncratic regulations are designed to facilitate enforcement, to maximize police officer safety, or to otherwise harden a target, they are expected to amplify the regulations’ impact on ambient crime. In that sense, these idiosyncracies represent sound public policy. 1.5 Non-sexually-oriented businesses Criminological theory predicts that non-sexually-oriented businesses, like gas stations, convenience markets, and bread stores, will have trivially small crime-related secondary effects. This is because non-sexually-oriented businesses do not draw large numbers of soft-target customers from wide catchment areas; customers drawn to the sites of non-sexually-oriented businesses spend only minutes on-site and, otherwise, lack the characteristics that make sexuallyoriented business customers soft targets. Due to the relatively low density of soft targets in areas around non-sexually-oriented businesses, rational predatory criminals are not attracted to gas stations, convenience markets, or bread stores. 1.6 On-premise vs. off-premise businesses In other legal proceedings, experts for the sexually-oriented business industry have argued that the criminological theory does not apply to businesses that rent or sell products only for off-

5

The best known statement of this effect is “Broken windows: The police and neighborhood safety.” by J.Q. Wilson and G.L. Kelling, Atlantic Monthly, 1982, 249:29-38. Wilson and Kelling argue persuasively that police visibility in a neighborhood has a greater impact on crime and victimization risk than police activities that target crime. Modern police methods are based on this theory. REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 7

premise consumption. This argument holds that, even though other sexually-oriented businesses might pose public safety hazards, the sub-class of sexually-oriented businesses that rent or sell only for off-premise consumption do not and cannot pose the same risk. In fact, however, the theoretical mechanism that generates a public safety hazard in and around an on-premise sexually-oriented business, operates as well for off-premise sexually-oriented businesses. Whether its products are consumed on- or off-premises, the success of a sexually-oriented business depends on its ability to attract large numbers of a certain type of customer from a wide catchment area. The implicit interests of these customers make them attractive targets to prostitutes and predatory criminals. Off-premise sexually-oriented businesses often must compete with book and video rental stores are not sexually-oriented (e.g., book and video rental stores that have small sexuallyoriented sections). To do this, off-premise sexually-oriented businesses might offer products (e.g., legal stimulants, sexual toys, paraphernalia) that their non-sexually-oriented competitors do not. The differences that distinguish the off-premise sexually-oriented business attract customers whose implicit interests make them soft targets for predatory criminals. In theory then, because both on- and off-premise sexually-oriented businesses must attract customers, both pose public safety hazards. The empirical evidence confirms this theory. 2. Empirical evidence: government-sponsored studies. Crime-related secondary effect studies began to appear in the mid-1970s. The literature accumulated for the next thirty years until, at present, there are at least three-dozen studies. For all practical purposes, this literature can be divided into two halves. The first half, consisting of studies conducted or sponsored by governments, consistently reports that sexually-oriented

REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 8

businesses have large, significant crime-related secondary effects. The second half, consisting of studies conducted or sponsored by sexually-oriented businesses usually reports no significant crime-related secondary effects. The contradiction between these two halves of the literature is easily reconciled. Before reconciling the contradiction, however, I will describe the results of three type government-sponsored secondary effect studies that were conducted over a period of thirty years. 2.1 Phoenix, AZ (1979) In 1979, the City of Phoenix, AZ conducted a study of crime-related secondary effects. Although the actual work was conducted by City employees, Arizona State University faculty served as advisors and consultants. I was a Professor of Criminal Justice at Arizona State University at that time and met regularly with the City employees who conducted this research.

Table 2.1 - Secondary Effects in Phoenix, AZ

Property Crime Rate Personal Crime Rate Sexual Crime Rate

Adult Business Areas

Control Areas

Secondary Effect

122.86 5.81 9.40

87.90 5.11 1.62

139.8 % 113.7 % 580.2 %

Source: ADULT BUSINESS STUDY, City of Phoenix Planning Department, May 25, 1979; Table V

To estimate the crime-related secondary effects of adult businesses, the researchers compared crime rates in areas with adult businesses to crime rates in “matched” control areas (i.e., areas that were similar but that had no adult businesses). The comparisons are summarized in Table 1. The property and personal crime rates reported in Table 1 were estimated from REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 9

Uniform Crime Report (UCR) data. The percentages reported in the right-hand column, in red, are the secondary effect estimates derived from the crime rates. Compared to crime rates in the control areas, the UCR property crime rate was 39.8 percent higher; the UCR personal crime rate was 13.7 percent higher; and the UCR sex crime rate was 480.2 percent higher in the adult business areas. By any reasonable standard, these are large, significant secondary effects. 2.2 Garden Grove, CA (1991) The salient weakness in the Phoenix study is that its “static group comparison” design6 assumes that the test and control neighborhoods are equivalent on relevant crime risk factors. If this assumption is unwarranted, observed secondary effects cannot be attributed to the sexuallyoriented businesses. The surest, simplest way to control this threat to validity is to use a beforeafter design. In the early 1990s, James W. Meeker and I were able to use a before-after design to study the secondary effects of sexually-oriented businesses in Garden Grove, CA.7 In Garden Grove study, Dr. Meeker and I observed neighborhood crime incidents before and after a sexually-oriented business opened in the neighborhood

We found that when a

sexually-oriented business opened its doors, ambient crime rose. This before-after contrast captures the essence of a crime-related secondary effect. If, as strong criminological theory predicts, sexually-oriented businesses generate public safety hazards, we expect to see crime rise when a sexually-oriented business moves into the neighborhood – otherwise the theory is wrong.

6

The most widely-cited authority on quasi-experimental designs is D.T. Campbell and J.C. Stanley, Experimental and Quasi-experimental Designs for Research. Rand-McNally, 1966. Campbell and Stanley call the design used in Phoenix a “static group comparison” design. 7

Final Report to the City of Garden Grove: The Relationship between Crime and Adult Business Operations on Garden Grove Boulevard. October 23, 1991. Richard McCleary, Ph.D. and James W. Meeker, J.D., Ph.D. REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 10

The validity of this inference requires that other plausible explanations for the beforeafter difference be ruled out, of course. Perhaps the difference is a coincidence, for example; or perhaps crime rose throughout the city, not just in the neighborhood where the sexually-oriented business opened. To control these common “threats to internal validity,” Dr. Meeker and I replicated the analyses for all other sexually-oriented businesses in Garden Grove. If a beforeafter rise in ambient crime were due to some unrelated factor, the effect would also be observed at other sexually-oriented businesses in Garden Grove. If the same effect were not observed at these control sites, on the other hand, the rise in ambient crime could be attributed, with a great degree of confidence, to the newly opened business.

Table 2.2a - Secondary Effects in Garden Grove, CA: Business Openings Total “Serious” Crime, One Year Before/After Test Sites Before

After

March, 1982 March, 1986 August, 1988

71 31 32

106 68 50

Total

134

224

Control Sites Before

After

1.49 2.19 1.56

76 80 41

78 92 40

1.03 1.15 0.98

1.67

197

210

1.06

Source: Final Report to the City of Garden Grove, pp. 26-28

Secondary effects for three business openings are reported in Table 2.2a. When a new sexually-oriented business opened, total “serious” crimes in a 500-foot radius around the site rose, on average, 67 percent. To control for the confounding effects of city-wide crime trends, changes in police activity, and other common threats to internal validity, these before/after differences were compared to the analogous differences for the addresses of existing sexuallyoriented businesses. Total “serious” crimes in a 500-foot radius around these “control” sites REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 11

rose, on average, only 6 percent. The secondary effect observed when a new sexually-oriented business opens is, thus, substantively large and statistically significant. Critics could still argue that effects of this magnitude might be observed for the opening of any new business. As a practical matter, however, if every Garden Grove business had generated a crime-related secondary effect as large as the effects found for sexually-oriented businesses, Garden Grove would have had the highest crime rate in the region; but Garden Grove did not have a particularly high crime rate. As a theoretical matter, moreover, the public safety hazard posed by sexually-oriented businesses is a consequence of the normal commercial activity of those businesses; the normal commercial activities of other types of business, including gas stations, convenience markets, bread stores, and Montessori preschools, are qualitatively different.

Table 2.2b - Secondary Effects in Garden Grove, CA: Alcohol License Total “Serious” Crime, One Year Before/After Test Sites

Violent Crimes Property Crimes

Before

After

1 38

10 48

Control Sites

10.0 1.26

Before

After

1 16

2 21

2.00 1.31

Source: Final Report to the City of Garden Grove, pp. 26-28

In addition to the findings about sexually-oriented businesses, the Garden Grove study produced a collateral finding that is of some importance. When a tavern opened less than 500 feet from a sexually-oriented business, violent crime in the vicinity of the sexually-oriented business rose significantly. While violent crime in the vicinity of a “control” also rose, the effect

REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 12

was substantially smaller. Before-after differences in property crime around the test and control sites were not significantly different. The crime-related secondary effect of sexually-oriented businesses is aggravated by proximity to a liquor license then but the effect is limited to serious violent crime. Like the major finding in Garden Grove, this collateral finding is predicted by empirically-validated criminological theory. 2.3 Centralia, WA (2003) In terms of internal validity, before-after designs (Garden Grove, e.g.) are stronger than “static group comparison” designs (Phoenix, e.g.). When the methodologically strongest design is not possible, of course. In many cases, for example, well-defined control sites, like those in Garden Grove, are not available. Small cities present an illustrative case. When their sexuallyoriented business codes are challenged, small cities often lack the resources required to hire experts; when they are able to hire experts, moreover, local conditions often preclude the use of the optimal design. Since most small cities have at most only one sexually-oriented business, the availability of optimal control sites is always problematic. Nevertheless, in such cases, strong before-after designs may still be feasible. In first week of December, 2001, a sexually-oriented business opened in an vacant residential structure in Centralia, WA. With population ca. 14,000, Centralia is located on Interstate 5 between Portland and Seattle. Total serious crimes before and after the sexually oriented business’ opening are reported in Table 2.3. In the impact area, serious crime rose by nearly 90 percent after the opening. In the rest of Centralia, during the same period, serious crime dropped by nearly four percent. The statistical significance of these before-after contrasts can be tested by forming the odds ratio, as reported in Table 2.3, and comparing its value

REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 13

(1.9559) to the value of its standard error (.8076).8 By chance alone, an odds ratio larger than 1.9559 would occur less than eight times in one thousand trials or samples.

Table 2.3 - Total Crime Before and After a Sexually-Oriented Business Opens

Impacted Area Other Centralia Odds Ratio

After

Before

Odds

17 3243

9 3358

1.8889 0.9658 1.9559

Although it is highly unlikely that the secondary effect reported in Table 2.3 occurred by chance, lacking a suitable control site, the effect could conceivably be due to some uncontrolled threat to internal validity. Is it possible, e.g., that crime would have risen in the impact area if another type of business – say, a bread store – had moved into the vacant residential structure? This is highly unlikely and inconsistent with the facts and statistics. Regarding the facts, the bulk of the “crime wave” reported in Table 2.3 occurred during late-night hours when bread stores are closed; the bulk of the increase was realized in crimes that are theoretically inconsistent with a bread store; and so forth and so on. The statistics make an even stronger argument, however. The before-after effect reported in Table 2.3 is both substantively and statistically large. If every new business that opened in Centralia generated a secondary effect of this magnitude, Centralia would be the “crime capitol” of the Pacific northwest. But since Centralia is a pleasant, crime-free city, it is highly unlikely 8

Assuming that the secondary effect estimates in Table 2.2 are due to chance, and that the crime data are Poisson-distributed, the standard error of the odds ratio is .8076. Derivations of this standard error are found in most graduate level statistics texts. See, e.g., p. 345 of Steve Selvin’s Statistical Analysis of Epidemiological Data (New York: Oxford University Press, 1991). REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 14

that the large, significant secondary effect is a simple consequence of a new business moving into a vacant residence. 2.4 Summary The three government-sponsored secondary effect studies reviewed here are typical of this half of the literature in that each finds a substantively large, statistically significant crimerelated secondary effects. The studies also typify the range of methodological rigor found in the secondary effects studies conducted or sponsored by governments. Other commonly cited government-sponsored studies which report large, significant crime-related secondary effects include Los Angeles CA (1977), Whittier, CA (1978), Minneapolis, MN (1980), Indianapolis, IN (1984), Austin, TX (1986), Seattle, WA (1989), Times Square, NY (1994), and Newport News, VA (1996). Individually, each of these studies might be faulted on narrow methodological grounds; that is the nature of non-experimental research. Since no single methodological critique applies to all of the studies, however, taken together, the literature supports the strong inference that sexually-oriented businesses pose serious ambient public safety hazards. 3. Empirical evidence: industry-sponsored studies. Beginning in 2001, secondary effect studies conducted or sponsored sexually-oriented business began to appear. Unlike the government-sponsored research studies conducted in the preceding 25 years, the industry-sponsored studies were conducted specifically for use in legal proceedings. The most prolific author of industry-sponsored studies is Dr. Daniel Linz.9 Without exception, Dr. Linz’ secondary effect studies find that sexually-oriented businesses pose no statistically significant ambient crime risks. Although Dr. Linz’ null findings might appear to

9

Dr. Linz is Professor of Communication, University of California, Santa Barbara. REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 15

contradict the consensus finding of government-sponsored studies, in fact, Dr. Linz’ research is entirely consistent with the consensus view. I will review three of his studies. 3.1 Greensboro, NC (2003) One of the most interesting industry-sponsored studies was conducted by Dr. Linz in Greensboro, NC.10 Dr. Linz used a relatively weak “static group comparison design” similar to the one used in the 1979 Phoenix study. With few exceptions, government-sponsored studies tend to use relatively weak quasi-experimental designs. Weaker designs tend to bias a study in favor of the null finding – no secondary effect, i.e. – Dr. Linz reported in Greensboro: The presence of adult cabarets and adult video/bookstores in “neighborhoods” was unrelated to sex crimes in the area. We found that several of an adult video/bookstore were located in high person and property crime incident “neighborhoods.” We examined the “neighborhoods” and local areas surrounding the adult video/bookstores (1000 foot radius) further and we found that the adult video/bookstores were not the primary source of crime incidents in these locations ... (T)here is no support for the City of Greensboro’s theory that adult businesses produce adverse secondary effects. The results of our study show that adult businesses are not associated with crime events.11 Due to the technical nature of the statistical analyses, the City of Greensboro retained me to “translate” Dr. Linz’ numerical results into plain words.12 Even for a statistician, Dr. Linz’ report was a difficult read. The numerical results supporting his conclusion were scattered over 18 pages of computer output in an appendix. When the actual numbers are examined, it was clear

10

Evaluating Potential Secondary Effects of Adult Cabarets and Video/Bookstores in Greensboro: A Study of Calls for Service to the Police by Daniel Linz, Ph.D. and Mike Yao, November 30th, 2003. Submitted by the plaintiffs in Giovani Carandola Ltd., et al v. City of Greensboro, U.S. District Court for the Middle District of North Carolina, Greensboro Division (1:03 CV 722) 11

p. 3 (counting the title sheet as p. 1) of the Linz-Yao Greensboro Study.

12

R. McCleary. A Methodical Critique of the Linz-Yao Report: Report to the Greensboro City Attorney. December 15, 2003. REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 16

that Dr. Linz had overstated the basis of his strongly-worded conclusion. Put simply, Dr. Linz’ numbers contradicted his words.

Figure 3.1 - Results of Dr. Linz’ Greensboro Study

Source: Tables 14-19, Evaluating Potential Secondary Effects of Adult Cabarets and Video/ Bookstores in Greensboro: A Study of Calls for Service to the Police. Daniel Linz and Mike Yao, November 30 th, 2003.

The results of Dr. Linz’ analyses are plotted in Figure 3.1. Dr. Linz analyzed six types of police data (relating to violent crimes, property crimes, drug crimes, sexual crimes, disorder crimes, and all other crimes). Controlling the effects of demographic and economic variables presumed to cause crime, Dr. Linz estimated crime risks four model neighborhood types. To facilitate interpretation, the green bars in Figure 3.1, representing control neighborhoods, are fixed at 100 percent. Dr. Linz found that neighborhoods with taverns, represented as blue bars in Figure 3.1, had more crime than the controls. That result was expected. What Dr. Linz did not expect, however, was that neighborhoods with adult cabarets would have more crime than the neighborhoods with taverns; and that neighborhoods with adult bookstores or video arcades REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 17

would have more crime than any other type of neighborhood. The pattern of results in Figure 3.1 are credible for three reasons. First, the results are exactly what strong criminological theory predicts. Second, the study was sponsored by the sexually-oriented business industry. Third, Dr. Linz used a study design that biased the analyses in favor of a null finding. The fact that a large, significant secondary effect survived the weak design testifies to size and nature of the secondary effects in Greensboro. 3.2 Toledo, OH (2004) The 2003 Greensboro study was not an isolated case. One year later, Dr. Linz replicated the Greensboro study in Toledo.13 In Toledo, Dr. Linz reported: ... that after controlling for variables used by criminologists and found to be related to criminal activity, the presence of adult cabarets, video/bookstores or other adult businesses in the neighborhood was unrelated to crime incidents when the control variables were considered, no matter what type of crime event we studied.14 Professor James W. Meeker and I were retained by the City of Toledo to re-analyze Dr. Linz’ data. Although we were able to replicate the Dr. Linz’ analysis, arriving at the same results, we noticed several suspicious idiosyncracies in Dr. Linz’ statistical model. With three simple “tweaks,” Dr. Linz’ statistical model produced the large, significant secondary effect estimates plotted in Figure 3.2.15

13

Evaluating Potential Secondary Effects of Adult Cabarets and Video/Bookstores in Toledo, Ohio: A Study of Calls for Service to the Police. Daniel Linz and Mike Yao, February 15, 2004. Submitted in Deja Vu-Toledo, Inc., et al., v. City of Toledo, U.S. District Court for the Northern District of Ohio, Western Division (Case No. 3:03CV7245). 14

p. 2, the Linz-Yao Toledo Report

15

Dr. Linz’ model broke crime down into five categories (personal, property, sex, drugs, and other); broke down sexually-oriented businesses into three categories (cabarets, bookstores, and other); and defined the secondary effect in terms crime totals (rather than per capita crime REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 18

To facilitate interpretation of Figure 3.2, the secondary effect estimates for control neighborhoods, which have no sexually-oriented businesses or taverns, are fixed at 100 percent and are represented as green bars. Compared to control neighborhoods, per capita UCR Part I (“serious”) and Part II (“victimless”) crime rates in neighborhoods with taverns, represented by blue bars in Figure 3.2, were 30 and 23 percent higher respectively. In neighborhoods with sexually-oriented businesses, however, the per capita UCR Part I (“serious”) and Part II (“victimless”) crime rates were 268 and 241 percent higher respectively.

Figure 3.2 - Results of the Toledo Re-analysis

The secondary effect estimates plotted in Figure 3.2 are unambiguous. Even compared to

rates). Our model broke crime down into two categories (UCR Part I or “serious” and Part II or “victimless” crimes); made no distinctions among the types of sexually-oriented businesses; and defined the secondary effect in terms of per capita crime rates (rather than crime totals). Other than these three differences, the two models are identical. Both use multiple regressions on the same explanatory variables and are estimated with the same software (SPSS). REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 19

neighborhoods with taverns, the crime-related secondary effects of Toledo’s sexually-oriented businesses are large and statistically significant. Non-technical readers may wonder how three simple model “tweaks” could transform a small, insignificant secondary effect into a large, significant secondary effect? Minor changes in the structure of a statistical model cannot reverse a secondary effect estimate (from small and insignificant to large and significant) unless the model is mis-specified; Dr. Linz’ model was clearly mis-specified so as to create a bias in favor of a null finding. 3.3 San Diego, CA (2002) Using the same weak quasi-experimental design, Dr. Linz reported a similar effect in San Diego.16 Comparing police calls-for-service in peep-show and control neighborhoods, Dr. Linz reported that there were no statistically significant differences. Professor James W. Meeker and I were retained by the City of San Diego to re-analyze Dr. Linz’ data.17 We discovered that Dr. Linz’ summary of the San Diego findings omitted an important point. What Dr. Linz actually found was that peep-show neighborhoods had 15.7 percent more calls-for-service than the control neighborhoods. Although a 15.7 percent increase in the crime rate is large by any standard, Dr. Linz argued that because the effect was statistically insignificant, the “real” secondary effect was zero: ... statistically nonsignificant result and must be interpreted, as meaning that there is no significant difference between these two averages – an indication that the

16

A Secondary Effects Study Relating to Hours of Operation of Peep Show Establishments in San Diego, California. September 1, 2002. Daniel Linz and Bryant Paul. Submitted in Mercury Books v. City of San Diego. U.S. District Court, Southern District of California (00CV2461). 17

R. McCleary and J.W. Meeker, A Methodical Critique of the Linz-Paul Report: A Report to the San Diego City Attorney’s Office. March 12, 2003. REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 20

level of criminal activity for [peep-show areas] is equal to the level of criminal activity for [control areas].18 The substantively large 15.7 percent increase is not “real,” in other words. Finding no statistically significant secondary effect, Dr. Linz then concludes that no secondary effect exists. This is a fallacy. Not finding a significant effect might imply that no secondary effect exists, as Dr. Linz claims, but it might also imply that the search for an effect was weak or defective.19 It was.

Figure 3.3 - False-negative Rates for the San Diego Report

18

p.15, A Secondary Effects Study Relating to Hours of Operation of Peep Show Establishments in San Diego, California. September 1, 2002. Daniel Linz and Bryant Paul. 19

Not finding something cannot prove that the thing does not exist. Newton made this point with his aphorism “Negativa non Probanda.” “Finding nothing proves nothing.” REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 21

Under the circumstances, Dr. Linz should have asked whether his failure to find a significant secondary effect was an artifact of the weak design used in the San Diego study. Although Dr. Linz did not address this question, his report included enough statistical detail to permit others to calculate relevant probabilities. The probabilities plotted in Figure 3.3 were calculated by Professor Meeker and me from statistics reported by Dr. Linz. As shown, the 15.7 percent secondary effect estimate found by Dr. Linz has a false-negative probability of .508. What this means, simply, is that Dr. Linz’ null finding is more likely (51 percent) to be incorrect than it is to be correct (49 percent). 3.3 Summary The three industry-sponsored secondary effect studies reviewed here are typical of this half of the literature in that each uses a relatively weak quasi-experimental design and each reports no statistically significant crime-related secondary effect. In some instances, however, as in the case of the (2003) Greensboro study and the (2004) Toledo study, re-analyses of the data have found large, statistically significant crime-related secondary effects that the authors ignored and/or obfuscated. In other instances, as in the case of the (2002) San Diego study, re-analyses have found that the statistical power of the reported null finding falls far below the conventional level; and in those cases, what the authors had mis-characterized as a null finding – no secondary effect – is an inconclusive finding. I will explain this point further in my conclusion. 4. Conclusion Table 4 lists sixteen of the most widely-cited crime-related secondary effects studies. The three columns of Table 4 correspond to three findings: an adverse effect finding, an inconclusive finding, and a null finding. Thirteen of the sixteen studies – including the two industry-

REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 22

sponsored studies reviewed in this report – had adverse effect findings. Three of the studies, all industry-sponsored, had inconclusive findings. None of the sixteen had a null finding (i.e., a finding of no adverse secondary effect).20

Table 4 - Secondary Effects Studies by Type of Finding Adverse Effect

Inconclusive

Null Effect

Confidence > 95% Power > 80%

Confidence < 95% Power < 80%

Confidence < 95% Power > 80%

Amarillo, 1977 Los Angeles, 1977 St. Paul, 1977 Whittier, 1978 Phoenix, 1979 Minneapolis, 1980 Indianapolis, 1984 Austin, 1986 Garden Grove, 1991 Times Square, 1994 Newport News, 1996 Greensboro, 2003 Toledo, 2004

Charlotte, 2001 Ft. Wayne, 2001 San Diego, 2003

Before stating the obvious inference to be drawn from Table 4, I will explain, as best 20

This set of sixteen studies is not arbitrary. It was drawn from a list of studies submitted to the court by the plaintiff in this case. In addition to thirteen crime-related secondary effect studies, the plaintiff submitted a secondary effect study of real estate values (Oklahoma City, 1986) and two studies consisting of public hearings (Houston, 1983; Seattle, 1989). Because my is limited to crime-related secondary effects, I will express no opinion in this report on the 1986 Oklahoma City study. Because my area of expertise lies in the numerical analysis of crime data, moreover, I will express no opinion in this report on the anecdotal evidence reported in the 1983 Houston and 1989 Seattle studies. My reluctance to express opinions on anecdotal evidence should not imply that scientific methods are the only means of demonstrating a secondary effect. Crime-related secondary effects can be demonstrated by methods that, while empirical, do not rely on formal scientific design structures (before-after contrasts, cross-sectional control comparisons, etc.). On the contrary, anecdotal or other “nonscientific” data can be used to demonstrate secondary effects and legislatures can rely on these nonscientific data. See, e.g., Center for Fair Public Policy v. Maricopa County, 336 F.3d 1153, 1168 (9th Cir. 2003). REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 23

possible, the criteria used to decide whether a finding is adverse, an inconclusive, or null. The criteria are so widely accepted in the scientific community that no research can be accepted as scientific unless it adheres to the conventions of statistical “confidence” and “power.”21 The idea behind the confidence and power criteria is that any statistical hypothesis test can be wrong. Recognizing this point, scientists who conduct hypothesis tests calculate error rate. In any hypothesis test, there are two types of error rates. The “false-positive” error rate comes into play when an adverse effect finding emerges from the research; the “false-negative” error rate comes into play when the research produces a null finding.

Figure 4 - False-Positives and False-Negatives

Do sexually-oriented business areas have more crime than control areas?

Yes

False- positive (5%)

No

Confidence (95%)

Power (80%)

False-negative (20%)

Figure 4 illustrates the relationship between the two complementary error rates.22 The

21

Since general acceptance in the scientific community is one of four criteria noted in Daubert v Merrell Dow Pharmaceuticals 509 US 579 (1993), this would also affect admissibility in U.S. District Court trials. 22

In statistical hypothesis testing, a false positive is called a “Type I” or “alpha-type” error. A false negative is called a “Type II” or “beta-type” error. The terms “false positive” and REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 24

research question “Do sexually-oriented business areas have more crime than control areas?” has two possible answers, “Yes” or “No.” Since data vary from sample to sample, any study can arrive at an incorrect answer. Incorrect “Yes” answers are false-positives. Incorrect “No” answers are false-negatives. Conventional false-positive and false-negative rates are 5 and 20 percent.23 Complements of the false-positive and false-negative rates, “confidence” and “power,” are 95 and 80 percent respectively. These conventional levels imply that “Yes” decisions are correct 95 percent of the time, “No” decisions are correct 80 percent of the time respectively. Returning now to Table 4, adverse effect findings are required to have 95 percent confidence (i.e., false positive rates less than 5 percent). The thirteen studies listed in the first column of Table 4 satisfy this criterion. Lacking 95 percent confidence, one can conclude that the finding is a null effect if – and only if – the finding has 80 percent power (i.e., false negative rates less than 20 percent).

Findings that have neither 95 percent confidence nor 80 percent

power are inconclusive and must be done over. Three industry-sponsored studies listed in Table 4 fall into this category.

“false negative,” which come from the field of public health screening, are widely used in popular discourse. I use the terms “false positive” and “false negative” for descriptive simplicity. 23

The most comprehensive authority on this issue is Chapter 22 of The Advanced Theory of Statistics, Vol. 2, 4th Ed. by M. Kendall and A. Stuart (Charles Griffin, 1979). This authority requires a strong background in mathematics, however. J. Cohen’s Statistical Power Analysis for the Behavioral Sciences, 2nd Ed. (L.E. Erlebaum Associates, 1988) and M. Lipsey’s Design Sensitivity: Statistical Power for Experimental Research. (Sage Publications, 1990). Both Cohen (pp. 3-4) and Lipsey (pp. 38-40) set the conventional false-positive and false-negative rates at .05 and .2, respectively. These rates can be set lower, of course. The convention also sets the ratio of false-positives to false-negatives at 4:1, implying that false-positives are “four times worse than” false-negatives. The 4:1 convention dates back at least to 1928 (J. Neyman and E. Pearson, “On the use and interpretation of certain test criteria for purposes of statistical inference.” Biometrika, 1928, 20A:175-240). It reflects a view that science should be conservative. In this instance, for example, the 4:1 convention works in favor of the sexually-oriented businesses. When actual decision error costs are known, the actual ratio is used. REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 25

The issue of low statistical power, depicted in Figure 3.3 for the San Diego study, has become one of the most serious methodological issues in the secondary effects literature. The “dirty little secret” of social science research is that anyone with a modest research background can design a study so weak as to guarantee statistically insignificant results.

Richard McCleary, Ph.D.

REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 26

Appendix In the last five years, I have been deposed or testified as an expert witness in six cases: Alaska Inter-Tribal Council v. State. Alaska Superior Court, Dillingham Branch. Artistic Entertainment v. City of Warner Robins. U.S. District Court, Middle District of Georgia (Case No. 97-00195-CV-4-HL-5); U.S. Court of Appeals, Eleventh Circuit (Case No. 02-10216). Scamp’s v. California Alcoholic Beverage Commission and City of Westminster, CA. Alcoholic Beverage Control Board Administrative Hearing. Washington Retailtainment, Inc. v. City of Centralia, WA. U.S. District Court, Western District of Washington at Tacoma (Case No. C03-5137FDB). Giovani Carandola Ltd. et al. v. Ann Scott Fulton et al. U.S. District Court for the Middle District of North Carolina, Greensboro Division (Case No. 1:01 CV 115) Fantasyland Video, Inc. v. County of San Diego, U.S. District Court for the Southern District of California (Case Nos. 02-CV-1909 LAB (RBB) and 02-CV-2023 LAB (RBB). My curriculum vitae is appended.

REPORT O F R IC HA R D M C C LEA R Y , P H .D. - P AG E 27

J.L. Spoons, Inc. v. Ohio Dept. of Public Safety, 31 F.Supp.3d 933 (2014)

31 F.Supp.3d 933 United States District Court, N.D. Ohio, Eastern Division.

West Headnotes (7) [1]

J.L. SPOONS, INC., et al., Plaintiffs, v. OHIO DEPARTMENT OF PUBLIC SAFETY, et al., Defendants.

Courts, including Sixth Circuit, apply Alameda Books burden-shifting test to determine whether government has substantial interest in enacting regulation targeting secondary effects associated with adult establishments: (1) municipality may rely on any evidence that is reasonably believed to be relevant for demonstrating connection between speech and substantial, independent government interest, though it cannot get away with shoddy data or reasoning and its evidence must fairly support its rationale for regulation, (2) if plaintiffs fail to cast direct doubt on this rationale, either by demonstrating that municipality's evidence does not support its rationale or by furnishing evidence that disputes municipality's factual findings, municipality meets Renton standard, and (3) if plaintiffs succeed in casting doubt on municipality's rationale in either manner, burden shifts back to municipality to supplement the record with evidence renewing support for theory that justifies its ordinance. U.S.C.A. Const.Amend. 1.

Case No. 1:04 CV 314. | Signed July 9, 2014. Synopsis Background: Group of strip club owners brought action challenging Ohio liquor control rule precluding liquor permit holder from knowingly or willfully allowing nudity or sexual activity on its premises. The United States District Court for the Northern District of Ohio, Ann Aldrich, J., 2007 WL 14581, declared parts of rule unconstitutionally overbroad. State appealed. The Court of Appeals, 538 F.3d 379, reversed. On remand, the United States District Court for the Northern District of Ohio, Dan Aaron Polster, J., 2010 WL 3370184, rejected plaintiffs' renewed facial challenge, and held that first decision effectively decided their as-applied challenge and foreclosed further litigation of that claim on remand. Plaintiffs appealed. The Court of Appeals, Helene N. White, Circuit Judge, 509 Fed.Appx. 464, affirmed in part, reversed in part, and remanded.

Holdings: The District Court, Dan Aaron Polster, J., held that:

Cases that cite this headnote [2]

[2] rule allowed for reasonable alternative avenues of communication;

So ordered.

Constitutional Law Nude dancing in general Although it falls within the outer ambit, nude dancing is expressive conduct that is protected by the First Amendment. U.S.C.A. Const.Amend. 1.

[1] rule was content-neutral place regulation;

[3] rule served a substantial governmental interest and was not unconstitutional as applied to plaintiffs.

Constitutional Law Freedom of speech, expression, and press

Cases that cite this headnote [3]

Constitutional Law Secondary effects Government can enact content-neutral time, place and manner regulations that are aimed at ameliorating the deleterious secondary

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

1

J.L. Spoons, Inc. v. Ohio Dept. of Public Safety, 31 F.Supp.3d 933 (2014)

[4]

effects of sexually oriented establishments. U.S.C.A. Const.Amend. 1.

of liquor. U.S.C.A. Const.Amend. 1; OAC 4301:1–1–52.

1 Cases that cite this headnote

Cases that cite this headnote

Constitutional Law Secondary effects Sixth Circuit assesses constitutionality of regulations that purport to ameliorate the deleterious secondary effects of sexually oriented establishments under Renton intermediate-scrutiny standard. U.S.C.A. Const.Amend. 1. Cases that cite this headnote

[5]

Constitutional Law Nude or semi-nude dancing Intoxicating Liquors Licensing and regulation Ohio liquor control rule precluding liquor permit holder from knowingly or willfully allowing nudity or sexual activity on its premises, the application of which was challenged as violative of First Amendment, was content-neutral place regulation aimed at ameliorating adverse secondary effects associated with nude dancing in environment that sold liquor. U.S.C.A. Const.Amend. 1; OAC 4301:1–1–52. 1 Cases that cite this headnote

[6]

Constitutional Law Nude or semi-nude dancing Intoxicating Liquors Licensing and regulation Ohio liquor control rule precluding liquor permit holder from knowingly or willfully allowing nudity or sexual activity on its premises, the application of which was challenged as violative of First Amendment, allowed for reasonable alternative avenues of communication; “speech” at issue, the expressive messages conveyed by exotic dance where female breast was fully exposed, was not in any way diminished by the absence

[7]

Constitutional Law Nude or semi-nude dancing Intoxicating Liquors Licensing and regulation Ohio liquor control rule precluding liquor permit holder from knowingly or willfully allowing nudity or sexual activity on its premises, the application of which was challenged as violative of First Amendment, served a substantial governmental interest; state presented evidence that requiring dancers at establishments that served alcohol at a minimum to wear pasties would combat adverse secondary effects associated with nude dancing at those establishments, and plaintiffs failed to meet their burden of casting doubt on that rationale. U.S.C.A. Const.Amend. 1; OAC 4301:1–1–52. 1 Cases that cite this headnote

Attorneys and Law Firms *934 J. Michael Murray, Raymond V. Vasvari, Jr., Steven D. Shafron, Lorraine R. Baumgardner, Berkman, Gordon, Murray & Devan, Cleveland, OH, Luke C. Lirot, Clearwater, FL, for Plaintiffs. Elise W. Porter, Damian W. Sikora, Paul R. Kulwinski, Anthony D. Siciliano, Charles E. Febus, Office of the Attorney General, Columbus, OH, for Defendants.

ORDER DAN AARON POLSTER, District Judge. This case involves a constitutional challenge to Ohio Administrative Code 4301:1–1–52, known as “Rule 52.” Rule 52 prohibits the display of nudity and sexual behavior at establishments operating with a liquor license. Plaintiffs in this case are three Ohio strip clubs and a stripclub association. On February 17, 2004, Plaintiffs filed

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

2

J.L. Spoons, Inc. v. Ohio Dept. of Public Safety, 31 F.Supp.3d 933 (2014)

an action in this Court seeking to enjoin Defendants, the Ohio Liquor Control Commission, the Ohio Department of Public Safety and individual members of each agency (together the “State”), from enforcing sections (A)(2), (B)(2) and (B)(3) *935 of Rule 52. 1 Plaintiffs initially asserted both facial and as applied challenges to Rule 52. On January 3, 2007, the Court permanently enjoined Defendants from enforcing Rule 52 on the ground that it was facially overbroad. On appeal, a divided panel of the Sixth Circuit held that Rule 52 is not facially overbroad and reversed. J.L. Spoons, Inc. v. Dragani, 538 F.3d 379, 382 (6th Cir.2008) (J.L. Spoons I ). Following the Sixth Circuit's ruling, Plaintiffs filed a new motion for a temporary restraining order and preliminary injunction, asking the Court to rule on their as applied challenge (Doc. # 61). On August 26, 2010, the Court 2 denied Plaintiffs' motion on the ground that their as applied claim was foreclosed by the Sixth Circuit's decision in J.L. Spoons I and lifted the stay of the enforcement of Rule 52. On appeal, the Sixth Circuit held that its decision in J.L. Spoons I had not foreclosed Plaintiffs' as applied claim and remanded the case for further proceedings. J.L. Spoons, Inc. v. Ohio Dept. of Public Safety, 509 Fed.Appx. 464, 472 (6th Cir.2012) (J.L. Spoons II ). Accordingly, the issue now before the Court is whether Rule 52 is unconstitutional as applied to Plaintiffs.

I. From the start of this litigation, Defendants have defended Rule 52 primarily on the ground that it was enacted to combat the undesirable secondary effects that result when there is nude dancing at establishments that serve alcohol. In J.L. Spoons I, a divided panel of the Sixth Circuit found that “Rule 52 is a constitutional, contentneutral regulation of the undesirable secondary effects, including prostitution, drug trafficking, and assault, associated with nude dancing in an environment serving alcohol. It is not overbroad.” 538 F.3d at 382. Following the panel's decision, Plaintiffs filed a new motion for preliminary injunction, asking the Court to rule on their as applied challenged. The Court declined to do so, concluding that the J.L. Spoons I decision *936 foreclosed further consideration of Plaintiffs' as applied challenge. J.L. Spoons, Inc. v. Collins–Taylor, 2010 WL 3370184, *2 (N.D.Ohio, Aug. 26, 2010). Plaintiffs filed a timely appeal.

[1] On December 27, 2012, the Sixth Circuit reversed the Court's dismissal of Plaintiffs' as applied challenge. The Sixth Circuit ruled that in J.L. Spoons I the prior panel had “simply accepted the established proposition that regulation targeting the secondary effects of strip clubs are presumed constitutional” and, therefore, it had not resolved Plaintiffs' challenge to the secondary effects evidence. J.L. Spoons II, 509 Fed.Appx. at 472. The panel explained that “[a]lthough laws targeted against secondary effects are presumed constitutional, it is a separate question whether, in a specific situation, there are secondary effects that need to be addressed.” Id. at 471. The panel then discussed two Supreme Court cases in which the Court analyzed laws targeting adverse secondary effects associated with adult establishments; City of Renton v. Playtime Theaters, Inc., 475 U.S. 41, 106 S.Ct. 925, 89 L.Ed.2d 29 (1986) and City of Erie v. Pap's A.M., 529 U.S. 277, 120 S.Ct. 1382, 146 L.Ed.2d 265 (2000). Id. In Renton, “the [Supreme] Court held that a government may rely on any evidence that is ‘reasonably believed to be relevant’ ” in evaluating adverse secondary effects associated with adult establishments, including the experiences of “other cities.” Id. (citing Renton, 475 U.S. at 51–52, 106 S.Ct. 925). In Pap's A.M., the Court's plurality opinion “reiterated that a government may reasonably rely on the experience of other jurisdictions relevant to the secondary-effects problem it is addressing.” J.L. Spoons II, 509 Fed.Appx. at 471 (citing Pap's A.M., 529 U.S. at 297, 120 S.Ct. 1382) (O'Connor, J., plurality op.). The panel determined that “[t]hese cases establish that there is a general presumption that a government may regulate secondary effects associated with strip clubs by relying on a body of prior experience.” Id. The panel then recognized that the Supreme Court has also laid out a procedure for plaintiffs who seek to rebut this presumption. Id. (quoting City of Los Angeles v. Alameda Books, Inc., 535 U.S. 425, 453, 122 S.Ct. 1728, 152 L.Ed.2d 670 (2002)) (internal quotations omitted). The “procedure” that the panel referred to is the burden-shifting test, set forth by the Supreme Court's plurality opinion in Alameda Books, 535 U.S. at 453, 122 S.Ct. 1728. Id. at 472. Courts, including the Sixth Circuit, apply the test to determine whether a government has a substantial interest in enacting a regulation targeting secondary effects. Id. The burdenshifting test includes three steps: [First,] a municipality may rely on any evidence that is ‘reasonably believed to be relevant’ for demonstrating a connection

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

3

J.L. Spoons, Inc. v. Ohio Dept. of Public Safety, 31 F.Supp.3d 933 (2014)

between speech and a substantial, independent government interest. This is not to say that a municipality can get away with shoddy data or reasoning. The municipality's evidence must fairly support the municipality's rationale for its [regulation]. [Second, i]f [P]laintiffs fail to cast direct doubt on this rationale, either by demonstrating that the municipality's evidence does not support its rationale or by furnishing evidence that disputes the municipality's factual findings, the municipality meets the standard set forth in Renton. [Third, i]f [P]laintiffs succeed in casting doubt on a municipality's rationale in either manner, the burden shifts back to the municipality to supplement the record with evidence renewing support for a theory that justifies its ordinance. Id. (citing Alameda Books, 535 U.S. at 438–39, 122 S.Ct. 1728) (O'Connor, J., plurality op.) (internal citations omitted). Because *937 Plaintiffs' evidentiary challenge to the adverse secondary effects of strip clubs had yet to be resolved under the Alameda Books standard, the panel remanded the case so that the Court could rule on Plaintiffs' as applied claim. Id. at 465, 122 S.Ct. 1728. Following the Sixth Circuit's decision remanding the case, the Court took testimony on three occasions. The first hearing, which took place on December 17, 2013 (hereinafter referred to as the “December 17 Evidentiary Hearing”), featured testimony from Dr. Daniel Linz, Dr. Richard McCleary and Agent Andrew Bouza. Prior to the hearing, Plaintiffs submitted an expert report from Dr. Linz, and the State submitted an expert report from Drs. McCleary and Wendy Rogeczi. Dr. Linz's report analyzes data from a study he conducted in which he attempted to gauge the impact that Rule 52 has had on adverse secondary effects in the cities of Dayton, Cleveland and Toledo. Dr. Linz testified that Rule 52 has had little to no effect in reducing adverse secondary effects. During his testimony, Dr. McCleary questioned the methodology that Dr. Linz used in his study, and he testified that his analysis of Dr. Linz's data shows that Rule 52 has reduced secondary effects crime.

During the December 17 Evidentiary Hearing, the State also presented evidence of crime occurring within adult cabarets. 3 The State presented this evidence through the testimony of Agent Bouza, Enforcement Agent for the Ohio Department of Public Safety's Investigative Unit (“Investigative Unit”). Agent Bouza testified about criminal activity that he observed while working undercover in adult cabarets located in Toledo. Based upon Mr. Bouza's testimony, the Court asked the State's counsel if the State was advancing the argument that the combination of alcohol and nude dancing may lead to criminal activity within the premises of adult cabarets. Counsel confirmed that, in addition to the State's secondary effects justification for enacting Rule 52, the State was making this argument as an additional rationale to support Rule 52. The Court responded that the State should have put the Plaintiffs and the Court on notice. Counsel for the State indicated that the State had made this argument in the earlier stages of the litigation, 4 but that it had not done so following the Sixth Circuit's decision remanding the case back to this Court. Accordingly, the Court gave Plaintiffs the opportunity to develop evidence to counter the State's argument and reconvened the hearing on April 14, 2014, the second evidentiary hearing in this case (hereinafter referred to as the “April 14 Evidentiary Hearing”). At the April 14 Evidentiary Hearing, Plaintiffs presented the testimony of Greg Flaig and Jim StJohn. Messrs. Flaig and StJohn have experience working, on the management side, with adult entertainment businesses. Mr. Flaig is the Executive Director of the Owners Coalition (the “Coalition”) in Ohio and the Secretary of the Buckeye Association of Club Executives *938 (“BACE”). Mr. StJohn is the President of ACE, the national chapter of BACE, and the Senior Vice–President of Acquisitions for Spearmint Rhino, an adult cabaret operation with twentytwo locations throughout the United States. Both Messrs. Flaig and StJohn testified that they agree with most of Rule 52's prohibitions, and were only objecting to the portion of Rule 52 that prohibits the showing of the nipple and areola areas of the female breast. As a result, the Court asked Plaintiffs' counsel if Plaintiffs were only objecting to Rule 52's requirement that dancers wear, at a minimum, pasties. April 14, 2014 Evidentiary Hearing Transcript, Doc. # 121 (“April Hrg. Tr.”) at 119. Counsel for Plaintiffs responded that while

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

4

J.L. Spoons, Inc. v. Ohio Dept. of Public Safety, 31 F.Supp.3d 933 (2014)

this was Plaintiffs' main objection to Rule 52, “and the thing that is most damaging to the plaintiffs and to their patrons,” they were not abandoning their objection to Rule 52's other prohibitions. Id. at 123. At the end of the hearing, the Court directed counsel to advise the Court as to whether either side intended to introduce additional witnesses to address Rule 52's requirement that dancers wear pasties. Id. at 132. On April 28, 2014, counsel notified the Court that Plaintiffs intended to call additional witnesses, including Dr. Judith Hanna, a cultural anthropologist and sociologist who researches and writes about art, dance, and society. The Court therefore scheduled a third, and final, evidentiary hearing for June 19, 2014 (hereinafter referred to as the “June 19 Evidentiary Hearing”). Prior to the hearing, Plaintiffs submitted an expert report from Dr. Hanna. At the June 19 Evidentiary Hearing, Plaintiffs called, in addition to Dr. Hanna, Judith Molter and Sue Russell, two entertainers who currently work at adult cabarets in Ohio. All three witnesses testified about the important expressive elements contained in dancing where the breast is fully exposed.

Renton v. Playtime Theatres, 475 U.S. 41, 106 S.Ct. 925, 89 L.Ed.2d 29 (1986).” Entm't Prods., Inc. v. Shelby Cnty., Tenn., 721 F.3d 729, 734 (6th Cir.2013) (citation omitted). [5] Rule 52 is a content-neutral regulation aimed at ameliorating the adverse secondary effects associated with nude dancing in an environment that sells liquor. See Renton, 475 U.S. at 47, 106 S.Ct. 925 (finding that the Renton ordinance is content-neutral because it “is *939 aimed not at the content of the films shown at ‘adult motion picture theaters,’ but rather at the secondary effects of such theaters on the surrounding community” (emphasis in original)). Furthermore, Rule 52 does not prohibit nude dancing in Ohio. Rather, Rule 52 limits where in Ohio nude dancing can take place, that is, it cannot take place at establishments operating with a liquor license. Accordingly, the Court analyzes it as a content-neutral “place” restriction under Renton. 5

[6] The Court first addresses Renton's second requirement, whether Rule 52 allows for reasonable alternative avenues of communication. The Sixth Circuit has recognized that “notwithstanding the reasonableness of the state's rationale, the statute must leave ‘the quantity and accessibility of speech substantially intact.’ ” Entm't Prods., 721 F.3d at 735 (quoting Alameda Books, 535 II. U.S. at 449, 122 S.Ct. 1728) (Kennedy, J., concurring [2] [3] [4] Although it falls within the “outer ambit of in judgment). The Court finds that this requirement the First Amendment's protection,” it is well-settled that is satisfied. The “speech” in this case, the expressive nude dancing is expressive conduct that is protected by the messages that are conveyed by exotic dance where the First Amendment. J.L. Spoons I, 538 F.3d at 389 (quoting female breast is fully exposed, is not in any way diminished Pap's A.M., 529 U.S. at 289, 120 S.Ct. 1382). The Supreme by the absence of liquor. Plaintiffs' expert, Dr. Hanna, Court has upheld laws restricting protected speech where testified that the messages that are expressed by nudity the governmental interest in enacting the law is unrelated in exotic dance are unrelated to the consumption or to the content of the speech. For instance, in Renton, the availability of alcohol on the premises: Supreme Court recognized that the government can enact The Court: And the—the freedom and expressiveness content-neutral time, place and manner regulations that that that dancer is creating, in that mood, in an are aimed at ameliorating the deleterious secondary effects ambience, fantasy, eroticism, all the things you testified of sexually oriented establishments. Renton, 475 U.S. at to, alcohol has nothing to do with that, right? 47, 106 S.Ct. 925. The Supreme Court held that contentneutral time, place and manner regulations of protected Dr. Hanna: Well in the sense that it makes people—I speech will survive constitutional scrutiny “so long as mean a lot of people just want to drink to relax. So that they are designed to serve a substantial governmental creates the ambience. interest and do not unreasonably limit alternative avenues of communication.” Id. (citations omitted). “[The Sixth The Court: Well, so are you saying that—well, what is Circuit] assess[es] the constitutionality [sic] of regulations the expression? Is it the drinking or the nudity? that purport to ameliorate the deleterious secondary Dr. Hanna: Well it's the nudity is the expression. effects of sexually oriented establishments under the intermediate-scrutiny standard announced in City of

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

5

J.L. Spoons, Inc. v. Ohio Dept. of Public Safety, 31 F.Supp.3d 933 (2014)

The Court: All right. *940 Dr. Hanna: And the alcohol, whether somebody's drinking a coke or a beer, is irrelevant. The Court: All right. That's my point. It's irrelevant to the dancer, and it's irrelevant to the patron, correct. Dr. Hanna: But a lot of patrons want to have a drink. I mean most people who go out at night do. The Court: But that's independent. You can go out and have a drink because you want to socialize with your friends or just have a drink. Dr. Hanna: Right. The Court: There's no freedom of—there's no freedom of expression or eroticism or fantasy or whatever with having a beer or a vodka, right? Dr. Hanna: Right. The Court: The expression, the fantasy, the eroticism, is created by the nudity and the movement? Dr. Hanna: Yes. June 19, 2014 Evidentiary Hearing Transcript, (“June Hrg. Tr.”) at 32–33. Therefore, as the Sixth Circuit explained in J.L. Spoons I, “Rule 52 has a minimal impact on the marketplace of ideas because persons desiring to perform mainstream works of art involving nudity and sexual activity may do so in an establishment that is not licensed to sell liquor.” 538 F.3d at 384.

A. Evidence of Secondary Effects [7] Turning to the requirement that Rule 52 serve a substantial governmental interest, the Court, as discussed above, must apply the Alameda Books burden-shifting analysis. When a district court is faced with analyzing a regulation targeting adverse secondary effects, the regulation at issue is usually challenged just prior to, or immediately after, the effective date of the regulation. 6 The court therefore does not have an opportunity to analyze evidence of the regulation's actual impact on adverse secondary effects. Here, because of the unique procedural history of the case, Rule 52 has been in effect since September 8, 2010. 7 As a result, this case presents

the Court with a rare opportunity to evaluate what impact, if any, regulations such as Rule 52 have on adverse secondary effects. Plaintiffs argue that “[t]he evidence on which [the State] rel[ies] is not sufficient *941 to support the rationale for Rule 52's statewide ban on nudity and touching on premises licensed to serve alcohol.” (Doc # : 62–1 at 4.) Additionally, Plaintiffs contend that the expert testimony of Daniel Linz, Ph.D. regarding his study of adverse secondary effects casts direct doubt on the State's rationale for Rule 52. Id. at 14. The Court disagrees with both contentions. Under the first step of the Alameda Books burden-shifting analysis, the State must have a reasonable evidentiary basis for concluding that Rule 52 would ameliorate adverse secondary effects. Courts have recognized that “combating the harmful secondary effects associated with nude dancing [is] undeniably important,” which include violent, sexual, and property crimes. Pap's A.M., 529 U.S. at 296, 120 S.Ct. 1382; see also Richland Bookmart, 555 F.3d at 524. The Sixth Circuit has shed light on a state's burden for establishing that it has a reasonable basis to conclude that the regulation will combat adverse secondary effects. In Entm't Prods., the Sixth Circuit upheld a similar ordinance that requires “adult oriented establishments” to obtain a license and regulates the manner in which entertainment may be provided. 8 721 F.3d at 732–33. The panel held that the government's burden, “[a]lthough not extraordinarily high, this evidentiary burden requires that the state show that the evidence upon which it relied was reasonably believed to be relevant to the problem that the entity sought to address.” Id. at 734 (citing 729, Inc. v. Kenton Cnty. Fiscal Ct., 515 F.3d 485, 491 (6th Cir.2008)) (quoting Renton, 475 U.S. at 51–52, 106 S.Ct. 925) (internal quotations marks omitted). “[T]here is no hard-and-fast rule that the government have any empirical data directly supporting a link between a given regulation and the secondary effect it is purported to ameliorate.” Id. (emphasis in original). “[T]he state may rely upon any evidence that is reasonably believed to be relevant for demonstrating a connection between speech and a substantial, independent government interest.” Id. at 739 (quoting Alameda Books, 535 U.S. at 438, 122 S.Ct. 1728 (plurality opinion) (emphasis added) (quoting Renton, 475 U.S. at 51–52, 106 S.Ct. 925) (internal quotation marks

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

6

J.L. Spoons, Inc. v. Ohio Dept. of Public Safety, 31 F.Supp.3d 933 (2014)

omitted)). Evidence includes “land-use studies, prior judicial opinions, surveys of relevant professionals (such as real-estate appraisers), anecdotal testimony, police reports, and other direct and circumstantial evidence.” Id. (quoting 84 Video/Newsstand, 455 Fed.Appx. at 549). Here, the State has shown that the evidence it relied upon provided it with a reasonable basis for believing that Rule 52 would combat adverse secondary effects that result when there is nude dancing at establishments that serve alcohol. The State presented the Court with several types of evidence that it relied upon, both from a 2003 public hearing that the Ohio Liquor Control Commission (the “Commission”) conducted before enacting Rule 52 and at the 2004 preliminary injunction hearing before the Court. The record shows that when promulgating Rule 52, the State relied on cases and a study regarding nudity in liquor-serving establishments and testimonials from people with experiences in these businesses. 9 For instance, *942 at the 2003 public hearing, the Commission heard testimony from Bruce A. Taylor, an attorney with experience in prosecuting vice crimes, including obscenity, prostitution and liquor violations. Transcript of 2003 Public Hearing, Exhibit 1 to Doc. # 18 at 11. Mr. Taylor testified that criminal activity, including drug trafficking and prostitution, occurs more frequently in and around liquor-serving establishments that have nude dancing than those that do not provide nude dancing. Id. at 18. Specifically, “[t]he amount of prostitution, the amount of drug traffic, the amount of fights and brawls that occur in and around bars or juice bars that have nude dancing is an escalated statistical number, and it's not sort of an uncommon or unreasonable result to imagine, because sexual performances tend to excite sexual thoughts and feelings in men who are the predominant customers of these kinds of places.” Id. Plaintiffs argue that the State has not relied on evidence that shows that covering the nipple and areola areas of the female breast, as required by Rule 52, will reduce adverse secondary effects. As noted above, at the April 14 Evidentiary Hearing, Plaintiffs' witnesses, Messrs. Flaig and StJohn, testified that their main objection to Rule 52 is the portion of the Rule that requires female dancers to wear, at a minimum, pasties. Specifically, Mr. StJohn testified that he does not think there is any evidence that establishes a connection between dancers at adult cabarets

wearing pasties, as opposed to being fully topless, and criminal activity: Court: So the only objection you have to Rule 52 is the showing of the female breast with less than fully opaque covering over the nipple and areola? Mr. StJohn: That's a main concern of mine. All the rest, sexual activity— Court: The rest of Rule 52 you don't have a problem with. Mr. StJohn: Sexual activity and all those other rules should be on the books, no question about that. I'm not going to condone that type of activity. If it's in there, I can live with that. Someone has to prove to me why we're doing the things we're doing. If there's a reason to that, Judge, I'll understand. Somebody has to— Court: Bottomless, you know, covering the lower portions, you don't have a problem with that? Mr. StJohn: I don't. Actually I like that. Court: So what you're saying is that you don't see any, any effect on any criminal conduct whether or not the establishment permits the entertainers to be fully topless or wear pasties. *943 Mr. StJohn: None whatsoever. I just don't. I mean, it just baffles me but, you know, stranger things have happened. April Hrg. Tr. at 117. Following Mr. StJohn's testimony, Plaintiffs' counsel stated that, while his clients were not abandoning their objections to Rule 52's other prohibitions, this requirement “was especially indefensible as a matter of evidence and constitutional law.” Id. at 127. At the June 19 Evidentiary Hearing, Plaintiffs presented additional testimony concerning Rule 52's requirement that dancers at adult cabarets wear, at a minimum, pasties. Mss. Molter and Russell testified that covering the areola and nipple areas of the female breast takes away from the “beauty of breasts.” June Hrg. Tr. at 42, 47. Ms. Molter also testified that wearing pasties distracts from the “sensual fantasy” she tries to create during her performances. Id. at 42. In addition, Dr. Hanna testified that she has counted “18 kinds of messages” that are conveyed by the nude breast; such as, honesty, independence, lack of pretense and eroticism. Id. at 26.

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

7

J.L. Spoons, Inc. v. Ohio Dept. of Public Safety, 31 F.Supp.3d 933 (2014)

The Court credits the testimonies of Mr. StJohn, Mr. Flaig, Dr. Hanna, Ms. Molter and Ms. Russell; however, it finds that Rule 52's requirement that dancers at establishments that serve alcohol wear, at a minimum, pasties, passes constitutional scrutiny. First, in Pap's A.M., the Supreme Court found that the “requirement that dancers wear pasties and G-strings is a minimal restriction in furtherance of the asserted government interests, and the restriction leaves ample capacity to convey the dancer's erotic message.” Pap's A.M., 529 U.S. at 301, 120 S.Ct. 1382 (internal citations omitted). Second, a plurality of the Court in Pap's A.M. rejected the dissent's view that the City of Erie must come forward with evidence showing that pasties and G-strings reduce crime: As to the second point—whether the regulation furthers the government interest—it is evident that, since crime and other public health and safety problems are caused by the presence of nude dancing establishments like Kandyland, a ban on such nude dancing would further Erie's interest in preventing such secondary effects. To be sure, requiring dancers to wear pasties and G-strings may not greatly reduce these secondary effects, but O'Brien requires only that the regulation further the interest in combating such effects. Even though the dissent questions the wisdom of Erie's chosen remedy, ‘the city must be allowed a reasonable opportunity to experiment with solutions to admittedly serious problems,’.... It also may be true that a pasties and G-string requirement would not be as effective as, for example, a requirement that the dancers be fully clothed, but the city must balance its efforts to address the problem with the requirement that the restriction be no greater than necessary to further the city's interest. 529 U.S. at 300–01, 120 S.Ct. 1382 (internal citations omitted). The evidence establishes that the State has met its burden under the first step of the Alameda Books burden-shifting analysis. The State passed Rule 52 to advance a government interest that the Supreme Court has recognized to be important. The State relied upon a variety of evidence that it reasonably believed demonstrated that establishments that serve alcohol and permit nude dancing are susceptible to adverse secondary effects. As the Supreme Court recognized in Pap's A.M., the government must be allowed a reasonable opportunity to experiment with solutions to address undesirable

secondary effects *944 associated with nude dancing in an environment that sells alcohol. Here, the State has decided to address this problem by requiring that dancers wear, at a minimum, pasties and G-strings. Thus, the first factor of the Alameda Books analysis is satisfied. Turning to the second step of the Alameda Books analysis, the Court finds that Plaintiffs have not met their burden. The evidence put forth by Plaintiffs, the testimony of Dr. Linz and the study he performed, does not cast direct doubt on the secondary-effects rationale advanced by the State. Dr. Linz's study involved analyzing criminal activity incidents occurring within a 500–foot radius of adult cabarets in the cities of Toledo, Cleveland and Dayton. 2013 Linz Expert Report (“2013 Linz Report”). In order to be included in the study, the adult cabaret had to be open for a period of 32 months, immediately before and after Rule 52 went into effect, and it had to be within a geographical area where police records existed. All of the adult cabarets satisfying these requirement were included in the study. As for the controls, they were comprised of non-adult establishments that served alcohol by the drink. 10 Id. at 3. For every adult cabaret, Dr. Linz chose one non-adult establishment that was closest to the adult cabaret. In total, Dr. Linz collected data on 13 adult cabarets and 13 control non-adult establishments. Id. at 2–3. Dr. Linz separated the crime data into three categories: (1) “Person Crimes;” (2) “Property Crimes;” and (3) “Societal Crimes.” 11 Id. at 2. For each category, Dr. Linz calculated the “weighted average” and the “unweighted average.” Id. at 4–7. Both numbers represent the gross average number of crime incidents that occurred within a 500–foot radius of the adult cabaret or control. December 17, 2013 Evidentiary Hearing Transcript, Doc. # 116 at 9 (“Dec. Hrg. Tr.”). However, for the “weighted average,” Dr. Linz also assigned each crime incident a numerical value depending on its proximity to the adult cabaret or control. 12 Id. at 74. Based on his analysis of the data, Dr. Linz arrived at two conclusions in his report. First, he concluded that “often crime events are more frequent at non-adult alcohol serving control locations compared to adult alcohol serving locations.” 2013 Linz Report at 7. Second, he concluded that “the implementation of Rule 52 has had little to no effect reducing the amount of crime

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

8

J.L. Spoons, Inc. v. Ohio Dept. of Public Safety, 31 F.Supp.3d 933 (2014)

in the vicinity surrounding the adult cabarets.” Dec. Hrg. Tr. at 78. As an initial matter, in Entm't Prods., the Sixth Circuit found that Dr. Linz's work “is a minority viewpoint within the secondary-effects literature,” and “federal court is simply not the appropriate forum for Dr. Linz and his colleagues to wage methodological combat with other studies on the secondary effects of adult speech.” 721 F.3d at 738. Even taking Dr. Linz's study in the most favorable light, it fails to cast direct doubt on the State's contention *945 that adverse secondary effects result when there is nudity at establishments that serve alcohol. First, as Dr. Linz admitted during his testimony, much of his data for Dayton and Toledo is inconclusive. When Dr. Linz collected data from the police department in Dayton, there was only one adult cabaret that satisfied the requirements of his study. Dec. Hrg. Tr. at 60. Therefore, Dr. Linz testified that, because his data on Dayton is based on only one adult cabaret and control site, it is “the least reliable indicator.” Id. As for Toledo, Dr. Linz admitted that his data on Person Crimes is also inconclusive. Dr. Linz's data on Person Crimes in Toledo shows that there was a dramatic increase in the number of Person Crimes that occurred in Toledo after Rule 52 went into effect, for both adult cabarets and non-adult establishments. Id. at 99. Dr. Linz admitted that the only conclusion that can be drawn from this data is that, after Rule 52 went into effect, there was an overall increase in the number of Person Crimes that were being committed in Toledo: The Court: It looks like there was a huge increase in the intensity of person crimes [in Toledo] 13 in both surrounding cabarets and the control group. All right. You are talking about factors of three to four times. Is that right?

Dr. Linz: That is correct, your Honor. The Court: So I am not sure what you can draw from table 1, other than Toledo became a whole lot more dangerous in the three years since 2010 than before no matter where you were downtown. Dr. Linz: I think that is correct, your Honor. ******

The Court: So I guess I am a little hard to understand your general conclusion for Toledo and Cleveland that Rule 52 has no effect. Isn't it more nuance[d], that it seems to have effect on some type[s] of crime but not to others? Dr. Linz: Well, I would say when we combine the weighted average and unweighted averages and we look across all crime categories, that in effect sometime[s] things go up; sometimes things go down. And that overall, it would be my best conclusion that the effect for Rule 52 was not significant. The Court: But isn't it a fact that someone looking at this data could draw a different conclusion? Say, for example, you could say, all right, looking at Toledo— I don't know what you draw quite frankly for person crime intensity.... There is nothing in any of your other charts like table 1, correct, that is kind of a dramatic increase or decrease in crime. Dr. Linz: That is correct, your Honor. The Court: So, you know, as a scholar and analyst, professor, when something seems aberrational, it requires further analysis, correct. *946 Dr. Linz: I would be interested in doing further analysis, that is correct. Dec. Hrg. Tr. at 99, 103–104. Second, one can draw multiple conclusions from Dr. Linz's data, including conclusions that directly contradict Dr. Linz's. For instance, Dr. Linz concludes that “for the most part crime rates either remained constant or increased in the areas surrounding the adult businesses from the pre Rule 52 to post Rule 52 period.” 2013 Linz Report at 7. However, if one were to ignore the data that Dr. Linz admitted is inconclusive, 14 Person Crimes in Cleveland is the only category where there was not a decrease in crime incidents after Rule 52 went into effect. After Rule 52 went into effect, there was a decrease in crime incidents for Property Crimes and Societal Crimes in Toledo and Cleveland. 15 As to Societal Crimes in Cleveland, Dr. Linz testified that a decrease in crime incidents after Rule 52 went into effect could support the conclusion that Rule 52 had a positive impact on secondary effects crimes in this category:

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

9

J.L. Spoons, Inc. v. Ohio Dept. of Public Safety, 31 F.Supp.3d 933 (2014)

The Court: I am saying the control group [for Societal Crimes in Cleveland] went from 10 before Rule 52 to 12 after[,] unweighted 28 to 30. 16 All right. So I will call it a modest increase, but for the cabarets, you have virtually identical, no change weighted and a modest decrease unweighted. So it is either no impact or slight reduction in crime, depending on whether you use weighted or unweighted versus the control group where crime went up. So could not one draw the conclusion that, at least for societal crimes in Cleveland, Rule 52 had a positive effect? Dr. Linz: I would draw the conclusion that there is little to suggest an effect either way. The Court: All right. Well, are you saying that if I look at the data and draw the conclusion I did, that I have drawn the wrong conclusion? Dr. Linz: No, your Honor, I would not. Dec. Hrg. Tr at. 106–107. Therefore, similarly, the decrease in crime incidents around adult cabarets in the other categories also supports the conclusion that Rule 52 has had a positive impact on secondary effects crimes in these categories. To refute Dr. Linz's report and conclusions, the State presented the testimony of Dr. Richard McCleary and a report he and Dr. Wendy Regoeczi authored. Drs. McCleary and Regoeczi disagree with the methodology that Dr. Linz used in his study. Their main criticism of Dr. Linz's study concerns his methodology for choosing control sites. They posit that Dr. Linz's analysis and conclusions are flawed because he did not carefully choose control sites. In their report, Drs. McCleary and *947 Regoeczi opine that Dr. Linz's conclusion “assumes that 13 [adult cabaret] and Control sites are comparable, of course, and ... they are not even roughly comparable. This raises serious questions about the validity of Dr. Linz'[s] ... conclusion. If Control sites were not carefully selected to be similar to [adult cabaret] sites, [adult cabaret]-Control differences are expected.” McCleary & Regoeczi 2013 Expert Report (“2013 McCleary Report”) at 3 (emphasis in original). In addition, Dr. McCleary testified that his analysis of the crime data that Dr. Linz gathered shows that Rule 52 reduced crime by seven percent. Dec. Hrg. Tr. at 199. Dr. McCleary cautioned that due to the size of the study,

the seven percent calculation has an error rate of eleven percent, which is above the five percent error rate that scholars strive to achieve in a well-conducted study. Id. at 200. Nonetheless, Dr. McCleary noted that “if you went to the legislature and told them that you had some program that would reduce crime in the State of Ohio by 7 percent, they would write a check.” Id. at 207. Based on their analysis of Dr. Linz's data, Drs. McCleary and Regoeczi concluded that: Although the evidence taken as a whole suggests that Rule 52 had [a] positive effect—or at worst, a benign effect—Dr. Linz arrives at the contrary conclusion. Ignoring reductions in nine Cleveland and Dayton sites, for example, Dr. Linz focuses on increases in four Toledo sites to argue that the effects of Rule 52 were negative. One can always find small subsamples that are larger and smaller than average, of course. Based on the entire sample, however, and on the average of all thirteen sites, Dr. Linz'[s] contrary opinion is flawed. 17 2013 McCleary Report at 13. Similar to Entm't Prods., “[a]t best, the [Plaintiffs] have demonstrated that the [State] faced a choice between two reasonable alternative viewpoints when assessing the need for the challenged regulations.” 721 F.3d at 738–39. As discussed above, the State has established that it relied on multiple evidentiary sources to support its rationale that enacting Rule 52 will combat adverse secondary effects. In response to that evidence, Plaintiffs attempt to “cast direct doubt” on the State's rationale by relying on Dr. Linz's report. However, despite his conclusions in his report, during his testimony, Dr. Linz admitted that much of his data is inconclusive. Dr. Linz also admitted that multiple conclusions can be drawn from his data, including a conclusion that supports the State's rationale for enacting Rule 52. “The [Plaintiffs] effectively ask [the Court] to second-guess the deliberative judgments of [the State].” Id. at 739. Like the court in Entm't Prods., the Court refuses to do so.

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

10

J.L. Spoons, Inc. v. Ohio Dept. of Public Safety, 31 F.Supp.3d 933 (2014)

B. Direct Evidence of Criminal Activity The State offered compelling evidence in support of Rule 52 through the testimony of Andrew Bouza under a related but distinct theory. Agent Bouza testified that, while working undercover in adult cabarets located in Toledo, he observed criminal *948 activity occurring within adult cabarets that were not complying with Rule 52's requirement that dancers cover the areola and nipple areas of their breasts. The State contends that the combination of alcohol with nude dancing increases the probability of sexual activity, prostitution and unlawful drug transactions on the premises themselves. Mr. Bouza has worked as an Enforcement Agent for the Investigative Unit since July 2010. Dec. Hrg. Tr. at 150. The Investigative Unit is responsible for enforcing Title 43 of the Ohio Revised Code, the portion of the Code that deals with issues related to liquor permits. Id. Mr. Bouza generally handles cases dealing with “[u]nderage sales, underage drinking, ... drug sales, prostitution, gambling, food stamps, [and] food stamp fraud” and has conducted undercover investigations in five adult cabarets located in or near Toledo. 18 Id. at 150, 182. During the course of his investigation, Mr. Bouza witnessed sexual activity between dancers and other criminal activity in establishments that served alcohol and permitted fully topless dancing:

Q: Generally, for the Court, the locations with this nude dancing, can you describe generally the setup or how it works from what you've observed? A: Out on the main dance floor, there is a main dance stage. Typically, when you go in, most of the girls will go up either by themselves or with another girl up on stage. Their first song they will typically dance in like a G string bottom with a bra or bikini top style covering. Their second song, they typically take off the top to expose their naked breasts and continue on with their second song. Q: And what types of activity outside of just the dancing have you observed in these locations? A: I have seen dancers lay other dancers on the stage, suck on their naked breasts, put their mouths on their clothed genitals, go over to the side of the stage,

pull up female patron shirts, suck on their naked breasts.... Dec. Hrg. Tr. at 156–57. Mr. Bouza testified that during a “private dance,” (a one-on-one dance between a patron and dancer typically in a private room or blocked-off area away from the general stage) “there were incidents where a female dancer would place her mouth over my clothed genitals, [and] either hum or blow hot air through [my] jeans.” Id. at 168. Mr. Bouza even “paid two female dancers to perform oral sex on each other for additional money” and has also witnessed patrons engage in sexual acts with the dancers: Q: Did you observe patrons engage in sexual activity with these dancers? A: Yes, I did. I was in a private dance where right across from me within three feet there was a male patron, his pants were down by his ankles. A female, her bottoms were off and having sex with him right next to me. I let the dancer know, I said, “do you see what's going on?” She says “yeah, we got girls that do that.” On my way out, they have a *949 manager that keeps track of the number of songs that girls give dances to so they can get their money. Id. at 169. In addition, Mr. Bouza testified, in responding to questions from the State's attorney, that he has witnessed solicitation of prostitution and narcotics trafficking: Q. Have you ever had occasion to discuss hiring a dancer for sex during your investigations? A. Yes. Q. And where did these solicitations occur? A. For the most part, in the private dance room. Q. Okay. And that would still be on the permit premise? A. Correct. Q. Have you ever been lured by drugs at these locations? A. Yes. Q. And where do these drug transactions occur?

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

11

J.L. Spoons, Inc. v. Ohio Dept. of Public Safety, 31 F.Supp.3d 933 (2014)

A. Usually, they are set up out on the general dance floor, but the transactions take place in the private dance room. Q. And that's still in the permit location? A. Correct. Dec. Hrg. Tr. at 162. When asked whether he has observed dancers appearing to be intoxicated or impaired, Mr. Bouza responded that he has, and when asked how he could tell that they were intoxicated, Mr. Bouza stated that “they use chairs, tables, and walls to support themselves as they are walking. I had a dancer tell me that she was so intoxicated one night that she fell down and broke her tooth in the women's restroom.” Id. at 158. Mr. Bouza also testified about his experiences in establishments that do not offer nude dancing. Id. at 181. Mr. Bouza testified that he has never witnessed prostitution or dancers engaging in unlawful sexual activity at these establishments. Id. In establishments where Mr. Bouza has witnessed dancers engaging in unlawful sexual activity, cabarets where women were dancing fully nude, Mr. Bouza testified that the dancers consumed alcohol with the patrons and that every 10 to 15 minutes he was approached by a dancer or wait staff to buy the dancers a drink. Id. at 157–58. It is not surprising that when alcohol is consumed in an environment where there is nude or topless dancing, inhibitions of both dancers and patrons are lowered. This can lead to patrons and dancers engaging in unlawful sexual activity. Mr. Bouza's testimony, therefore, provides direct evidence that the product of nude or topless dancing and alcohol creates an environment conducive to unlawful sexual activity, prostitution and drug trafficking, criminal behavior that the State clearly has an interest in curtailing. The report and testimony of Plaintiff's expert, Dr. Hanna, actually provides support for this conclusion. In her report, Dr. Hanna discusses the messages that are communicated by nudity in exotic dance, including, “messages of eroticism, temptation and allurement, pretense of sexual availability and longing.” 2014 Hanna Expert Report at 8. In addition, at the June 19 Evidentiary Hearing, Dr. Hanna testified that the “whole notion” of adult entertainment businesses is to create the fantasy of

intimacy that exists in the bedroom. June Hrg. Tr. at 27. These messages, which the nude dancers are trying to convey, when coupled with the lowering of inhibitions brought on by patrons and dancers consuming alcohol, increases the likelihood that the unlawful sexual contact and conduct that Agent Bouza discussed during his testimony will occur. *950 In response to Agent Bouza's testimony, at the April 14 Evidentiary Hearing, Plaintiffs presented the testimony of Greg Flaig. Mr. Flaig provided extensive testimony about his involvement with BACE and the Coalition. BACE is a legal organization that “makes the owners in the state aware of the larger laws and how they're going through the legislature.” April Hrg. Tr. at 22. The Owners Coalition is an organization made up of fortyseven adult cabarets that “operate under certain systems and programs ... [to] mak[e] sure that the club is working under the law as it should.” Id. at 8–9. Members of the Coalition are subject to the Coalition's Due Diligence Program, a program that the Coalition set up to ensure that its members are operating in compliance with the law. Id. at 11. For instance, to address the Coalition's concerns with human trafficking, the Program requires members to “sign off” on the Coalition's Human Trafficking Manual. Id. at 16. In addition, members are required to post “the Human Trafficking sign” in every dressing room. Id. As part of the Due Diligence Program, club owners and managers have to perform an audit, “a very thorough 10–page document” that covers “all aspects” of what law enforcement and human resource programs are concerned with. Id. at 13. The audit is comprised of a list of questions that address the requirements of the Due Diligence Program. For instance, one of the questions that is part of the audit is “Are the ‘Human Trafficking’ Signs Posted?” April 14 Evidentiary Hearing, Exhibit 8, Due Diligence Program Manual. If a member of the Coalition is not in compliance with the Coalition's Due Diligence Program, it runs the risk of having its membership in the Coalition revoked. Mr. Flaig provided testimony about the adult entertainment businesses that were the subject of Agent Bouza's investigation and testimony. Mr. Flaig testified that, after being cited by the Commission, two of the adult cabarets that Agent Bouza investigated, Hush and Fantasy Land, joined the Coalition. April Hrg. Tr. at 24– 25. Since joining the Coalition, the Commission has not

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

12

J.L. Spoons, Inc. v. Ohio Dept. of Public Safety, 31 F.Supp.3d 933 (2014)

issued any citations to Hush and it has issued only one citation to Fantasy Land. Id. at 25. Mr. Flaig also testified about the process by which dancers are able to perform at adult cabarets. In Ohio, dancers are not employees of the adult cabarets where they perform. Rather, a dancer who wants to perform at a particular cabaret, will, prior to her performance, enter into a lease agreement with the cabaret's manager, the Entertainment Tenant Lease (“Lease Agreement”). Id. at 38–39. The Lease Agreement allows the dancer to lease space within the cabaret (the area in which the dancer is permitted to perform is limited to the space that she leases). Id. Mr. Flaig explained that “the lease itself is set up so that they have total autonomy and control over their own entertainment function.” Id. at 39. For instance, nothing in the lease prohibits a dancer, when she's not performing on stage, from receiving and consuming an alcoholic beverage purchased by a customer. Id. at 43. The Court finds that while Mr. Flaig's testimony is informative as to the efforts that adult entertainment businesses are making to comply with Rule 52, it is not sufficient to counter Mr. Bouza's testimony. First, Mr. Flaig testified that approximately sixty percent of the adult cabarets in Ohio are members of the Coalition. Id. at 35. Therefore, assuming that membership in the Coalition helps to prevent criminal activity that occurs within adult cabarets, nearly half of the adult cabarets in Ohio are not members of the Coalition. For the adult cabarets that are not part of the Coalition, Rule 52 is necessary to deter criminal activity that may be caused by the *951 combination of nude dancing and the consumption of alcohol. Second, while Mr. Flaig's testimony establishes that the Coalition's Due Diligence Program is aimed at deterring criminal activity within adult cabarets, there is no evidence that the Program, in isolation, is adequate to deter this criminal activity. Mr. Flaig testified that sometime around 2010, after Rule 52 went into effect, he began tracking the number of adult cabarets that received liquor citations for unlawful sexual conduct. Id. at 65. According to Mr. Flaig, in the four-year period that Rule 52 has been in effect, the Commission has only issued twenty-four citations for unlawful sexual conduct. Id. at 80. Mr. Flaig testified that this is a relatively low number considering that there are “over 3,000 young ladies that entertain per night” at more than 100 clubs. Id. This suggests to the

Court that as a result of various efforts, including Rule 52 and the Coalition's Due Diligence Program, criminal activity within adult cabarets has decreased. The Court posed this suggestion to Mr. Flaig: Court: Mr. Flaig, just picking up on what you said at the very end, you said “For the entire state we have almost full compliance, we've come so far in four years. Enforcement people gave me the answers.” What I take from that is that Rule 52 and the combination of State enforcement and then your effective due diligence and compliance has—serves to keep prostitution, sex acts and drug trafficking out of the adult entertainment business. Is that a fair way to put it.” Id. at 84–85. Mr. Flaig responded that he did not agree. Id. In explaining why he did not agree with the Court's conclusion, Mr. Flaig stated that “when it really boils down to Rule 52, Rule 52 is a superfluous law ... because if it didn't exist tomorrow, the very enforcement agencies, such as vice and all the others, would be able to arrest people and put them away when they break [the law].” Id. at 86. The Court comes to the opposite conclusion. As noted above, Agent Bouza's testimony established that the product of nude or topless dancing and alcohol creates an environment conducive to criminal activity. There is no evidence that in the absence of Rule 52 the Due Diligence Program would adequately deter this criminal activity. In addition, Mr. Flaig's response misses the point. Rather than “arrest people and put them away when they break [the law],” the State has decided to address the problem by eliminating at least one of the causes of the criminal activity that occurs within adult cabarets.

III. For the forgoing reasons, the Court finds that Rule 52 serves a substantial governmental interest. The evidence establishes that the State, when it promulgated Rule 52, had a reasonable evidentiary basis for concluding that nude dancing in adult cabarets leads to undesirable secondary effects. Dr. Linz's report and testimony concerning the impact of Rule 52 does not “cast direct doubt” on the State's secondary effects rationale. In addition, the State has presented evidence that the combination of nude dancing with the consumption of alcohol by dancers and patrons increases the likelihood of criminal activity within the premises of adult cabarets.

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

13

J.L. Spoons, Inc. v. Ohio Dept. of Public Safety, 31 F.Supp.3d 933 (2014)

Accordingly, the Court upholds Rule 52 and dismisses Plaintiffs' as applied challenge.

All Citations

IT IS SO ORDERED.

31 F.Supp.3d 933

Footnotes

1

2 3 4

5

Ohio Administrative Code 4301:1–1–52 in relevant part: (A) Definitions as used in this rule: (2) “Nudity” means the showing of the human male or female genital, pubic area or buttocks with less than a fully opaque covering; the showing of the female breast with less than a fully opaque covering of any part of the nipple and/ or areola; the exposure of any device, costume, or covering which gives the appearance of or simulates the genitals, pubic hair, natal cleft, perineum anal region or pubic hair region; or the exposure of any device worn as a cover over the nipples and/or areola of the female breast, which device simulates and gives the realistic appearance of the nipples and/or areola. (B) Prohibited activities; no permit holder, his agent, or employee shall knowingly or willfully allow in and upon his licensed permit premises any persons to: (2) Appear in a state of nudity; (3) Engage in sexual activity as said term is defined in Chapter 2907 of the Revised Code. OAC Ann. 4301:1–1–52. Under Chapter 2907 of the Ohio Revised Code: (A) “Sexual conduct” means vaginal intercourse between a male and female; anal intercourse, fellatio, and cunnilingus between persons regardless of sex; and, without privilege to do so, the insertion, however slight, of any part of the body or any instrument, apparatus, or other object into the vaginal or anal opening of another. Penetration, however slight, is sufficient to complete vaginal or anal intercourse. (B) “Sexual contact” means any touching of an erogenous zone of another, including without limitation the thigh, genitals, buttock, pubic region, or, if the person is a female, a breast, for the purpose of sexually arousing or gratifying either person. (C) “Sexual activity” means sexual conduct or sexual contact, or both. ORC Ann. 2907.01. This case was initially assigned to Judge Ann Aldrich. Following Judge Aldrich's death, this case was reassigned to my docket on May 6, 2010. Adult cabarets refers to establishments that offer adult entertainment and serve alcohol. In response to Plaintiffs' second motion for a preliminary injunction and temporary restraining order, the State filed a brief in which it argued that Plaintiffs were unlikely to succeed on the merits because the State had a substantial interest in enacting Rule 52. (Doc. # 71). In support, the State pointed to the testimony of Scott Pohlman. Mr. Pohlman testified at the 2004 preliminary injunction hearing. At the time of his testimony, he was the Assistant Deputy Director of the Ohio Department of Public Safety. Id. at 7. Mr. Pohlman testified that, while investigating adult cabarets, he witnessed explicit sexual acts take place between dancers and patrons. Id. at 23. The Court recognizes that in J.L. Spoons I, the Sixth Circuit noted that had this Court, in the first round of litigation, struck down Rule 52 on the ground that it violated the First Amendment guarantee of freedom of expression, then the Supreme Court's decision in Pap's A.M. would be “directly on-point.” J.L. Spoons I, 538 F.3d at 383. In Pap's A.M., the Supreme Court upheld a regulation making it “a summary offense to knowingly or intentionally appear in public in a ‘state of nudity.’ ” Id. (quoting Pap's A.M., 529 U.S. at 283, 120 S.Ct. 1382). The Sixth Circuit's opinion in J.L. Spoons I suggests that because the Court in Pap's A.M. evaluated the regulation under the framework set forth in United States v. O'Brien, 391 U.S. 367, 377, 88 S.Ct. 1673, 20 L.Ed.2d 672 (1968), then Rule 52 should, in turn, be analyzed under O'Brien. See J.L. Spoons I, 538 F.3d at 383 (“Pap's A.M. would be directly on-point and would decide the issue were it not for the fact that the district court struck down Rule 52 on the grounds that it was overbroad, not that it violated the First Amendment guarantee of freedom of expression under O'Brien.”). The Court believes that because Pap's A.M. and O'Brien involved restrictions on conduct (public nudity and burning of draft cards, respectively), rather than restrictions on the time, place or manner of protected speech, Rule 52 should be analyzed under Renton. Sixth Circuit cases post-J.L. Spoons I suggests that this is the appropriate approach. See Richland Bookmart, Inc. v. Knox Cnty., Tenn., 555 F.3d 512, 523 (6th Cir.2009) (“[O]ur first step is to determine whether the [ordinance] purports to be a regulation of conduct that

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

14

J.L. Spoons, Inc. v. Ohio Dept. of Public Safety, 31 F.Supp.3d 933 (2014)

6

7

8

9

10 11 12 13

incidentally burdens expression (as in O'Brien ), a time, place, or manner regulation targeting secondary effects (as in Renton ), or a regulation comprising both (as in Pap's A.M.)”). Furthermore, while the Court applies Renton, it adheres to the Sixth Circuit's position that “the O'Brien and Renton inquiries embody much the same standards.” Id. at 521 (internal quotation and citation omitted). See e.g., 84 Video/Newsstand, Inc. v. Sartini, 2011 Fed.App. 0655N, 455 Fed.Appx. 541, 546 (6th Cir.2011) (“On October 17, 2007, the day § 2907.40 went into effect, the twelve Plaintiffs filed suit seeking a temporary restraining order (TRO), preliminary injunction, permanent injunction, and declaratory judgment.”); Entm't Prods., Inc. v. Shelby Cnty., 2009 Fed.App. 0406P, 588 F.3d 372, 377 (6th Cir.2009) (“On January 25, 2008—prior to the expiration of the 120–day grace period for obtaining licenses—Plaintiffs filed suit in the United States District Court for the Western District of Tennessee against Shelby County and the City of Memphis, seeking injunctive relief and a declaratory judgment.”); Deja Vu of Cincinnati, L.L.C. v. Union Twp. Bd. of Trs., 2005 Fed.App. 0270P, 411 F.3d 777, 781 (6th Cir.2005) (One month after the Board of Trustees of Union Township enacted Resolution No. 99–15, Deja Vu filed its complaint “alleging that various provisions of Resolution No. 99–15 violate the First and Fourteenth Amendments to the United States Constitution.”). As part of the Court's August 26, 2010 order dismissing Plaintiffs' as applied challenge, the Court stayed the enforcement of Rule 52 until September 8, 2010 to permit Plaintiffs to seek relief, including a stay, from the Sixth Circuit. On August 27, 2010, Plaintiffs filed a Notice of Appeal. On the same day Plaintiffs also filed a Motion for an Injunction Pending Appeal under Rule 8(a) of the Federal Rules of Appellate Procedure, which requires plaintiffs to seek injunctive relief pending appeal in the first instance in the district court. The Court denied Plaintiffs' motion (Doc. # 87), and Plaintiffs did not seek a stay from the Sixth Circuit. The limitations include: (1) prohibiting nudity; (2) prohibiting certain sexual activities, touching of certain anatomical areas, and all physical contact during performances; (3) prohibiting the sale or consumption of alcohol on the premises; and (4) requiring that all performances take place on a stage at least 18 inches above floor level and that all performers stay at least six feet away from customers and other performers. Id. At the 2004 preliminary injunction hearing, the State introduced a study from Adams County Sheriff's Department in Colorado, which found that there was a direct correlation between the increases in crime, the increases in alcohol related offenses, and the increased transiency of the patronage for businesses of this nature. 2004 Preliminary Injunction Hearing, Exhibit U at 3. The study concluded “that nude entertainment establishments were an attractant to a class of patronage which was very much undesirable to the good citizens of Adams County and represented a very real danger to the safety of the nearby residential citizenry.” Id. The State further supported its position with a Memorandum of Law by the National Law Center for Children and Families that summarizes several significant court cases involving the regulation of adult entertainment in liquor-serving establishments. 2004 Preliminary Injunction Hearing, Exhibit X. Additionally, at the 2003 public hearing, the Commission heard testimony from Ed Duvall, a member of the Ohio Department of Public Safety's Investigative Unit. Mr. Duvall testified that the combination of nudity and the consumption of alcohol increases secondary effects crime around adult cabarets. Transcript of 2003 Public Hearing, Exhibit 1 to Doc. # 18 at 74–76. Controls had to be no more than 2.5 miles away from the adult business, but also not closer than 1000 feet from any adult business or control. Id. at 3. (1) “Person Crimes” includes murders, assaults, and similar crimes; (2) “Property Crimes” includes burglaries, criminal mischief, shoplifting, auto thefts, and other property crimes; and (3) “Societal Crimes” includes alcohol or drug intoxication, possession of narcotics, prostitution, etc. Id. For example, a robbery that occurred 100 feet away from an adult cabaret would have been given more weight than a robbery that occurred 300 feet away. In his report, Dr. Linz included the below table for Person Crimes in Toledo, which is referred to as “Table 1”: Point Source

Weighted Average

Unweighted Average

Adult Businesses (01/15/08–09/07/10)

3.85

12

Adult Businesses (09/08/10–05/01/13)

14.82

37.5

4.42

10.5

12.89

30.25

Non–Adult Bars and Nightclubs (01/15/08–09/07/10) Non–Adult Bars and Nightclubs (09/08/10–05/01/13)

14

2013 Linz Report at 4. As noted above, Dr. Linz admitted during his testimony that the data for Dayton and the Person Crimes data for Toledo is inconclusive.

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

15

J.L. Spoons, Inc. v. Ohio Dept. of Public Safety, 31 F.Supp.3d 933 (2014)

15 16

The decrease in crime incidents around adult cabarets in these categories was based on either the unweighted average or weighted average, and in many instances on both. In his report, Dr. Linz included the below table for Societal Crimes in Cleveland: Point Source

Weighted Average

Unweighted Average

Adult Businesses (01/28/08–09/07/10)

5.15

17.13

Adult Businesses (09/08/10–04/18/13)

5.03

14.25

10.04

28.25

12.02

30.12

Non–Adult Bars and Nightclubs (01/28/08–09/07/10) Non–Adult Bars and Nightclubs (09/08/10–04/18/13)

17

18

2013 Linz Report at 6. Similarly, when the Court asked Dr. McCleary whether Dr. Linz's data supports his overall conclusion that Rule 52 has had no impact on secondary effects crime, Dr. McCleary responded that it did not. Dr. McCleary went on to explain that “[the data] is very strained. [Dr. Linz] was able to arrive at that conclusion because he sliced and diced everything up. You can take any large sample or before-after sample and cut it up into small enough segments so that you basically have a hodge podge of results. His findings are very, very confusing.” Dec. Hrg. Tr. at 202. Four of the adult cabarets are located in Toledo, and one is located in Bucyrus. At the time of Mr. Bouza's investigation, the four cabarets located in Toledo had a liquor permit and the cabaret located in Bucyrus served alcohol, even though it did not have a liquor permit. Four of the investigations that Mr. Bouza conducted are closed cases; the Commission imposed significant fines on three of the cabarets, and it revoked the liquor permit of the fourth cabaret. At the time of his testimony, the fifth investigation that Mr. Bouza was involved with was an open case, and Mr. Bouza did not provide the name of the adult cabaret. Id. at 182–86.

End of Document

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

© 2017 Thomson Reuters. No claim to original U.S. Government Works.

16

S ECONDARY E FFECTS OF S EXUALLY-O RIENTED B USINESSES:

R EPORT TO THE O HIO A TTORNEY G ENERAL

Richard McCleary, Ph.D.

Wendy Regoeczi, Ph.D.

March 12th, 2010

Introduction The Attorney General has retained us to review the facts and materials and to formulate opinions on the secondary effects issues raised in J.L. Spoons.1 We reviewed reports by Drs. Daniel Linz2 and Lance Freeman3 for that purpose as well as other documents and references. Our assessment of these materials, along with our educational backgrounds and training, have lead us to develop three general opinions:

Opinion 1: The criminological theory of ambient crime risk, known as the “routine activity theory of hotspots,” predicts that sexually-oriented businesses (SOBs) will generate large, significant crime-related secondary effects. These effects are the result of three factors. (1) SOBs draw patrons from wide catchment areas. (2) Because they are disproportionately male, open to vice overtures, reluctant to report victimizations to the police, etc., SOB patrons are “soft” targets. (3) The high density of “soft” targets at the site attracts predatory criminals, including vice purveyors who dabble in crime and criminals who pose as vice purveyor in order to lure or lull potential victims. The conjunction of these three factors generates an ambient public safety hazard in SOB neighborhoods. Opinion 2: Although the “hotspot” model applies to all SOB subclasses, qualitative differences across subclasses may exist with respect to the nature of the hazard. In this particular instance, the secondary effects of SOBs that serve alcohol and offer on-site entertainment will differ qualitatively from the effects of other SOBs. Opinion 3: In the last thirty years, the predicted theoretical relationships have been confirmed by empirical studies employing a wide range of quasi-experimental designs. Given that secondary effects are predicted by a strong theory, and given that the predicted relationships are corroborated consistently by a diverse empirical literature, it is a scientific fact that SOBs pose ambient crime risks. “Secondary effects” include litter, noise, traffic, real estate values, crime, and the general quality

1

J.L. Spoons, Inc. v. Nancy J. Dragani, et al. U.S. District Court, Northern District of Ohio, Cleveland, Case No. 04-00314. Hereafter, “J.L. Spoons.” 2

Report of Daniel Linz, Ph.D., January 15th, 2010.

3

Examining the Relationship between Adult Oriented Businesses and Surrounding Property Values in Ohio. Lance Freeman and Derrick Hamilton, no date. It is our understanding that Dr. Hamilton will not be called as a witness in J.L. Spoons. Accordingly, hereafter, we refer to this report as “Report of Lance Freeman, Ph.D.” R ICHARD M CC LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 2

of life. Although this report focuses on the secondary effects of SOBs on crime and real estate, virtually all secondary effects are correlated. Our opinions about crime and real estate secondary effects also extend to the broader, general category of secondary effects. This report begins with an introduction to the criminological theory of secondary effects. Over the last forty years, the consensus finding has emerged that SOBs generate large, significant ambient secondary effects. This consensus finding is credible because it is predicted by a strong theory. The theory serves as both an explanation of why SOBs are expected to have secondary effects but, also, how to lessen these effects. After developing the underlying theory, in §2 below, we review some of the early studies that corroborate the theory. Scientific theory leads us to expect to find secondary effects in SOB neighborhoods and that is exactly what we find. Since 1975, studies conducted across all types of geographical regions (rural, urban, suburban) and examining virtually all subclasses of SOBs have produced results corroborating the theory. Given that the strong theoretical expectation has been tested and corroborated, the consensus finding that SOBs pose large, significant secondary effects is a scientific fact. Nevertheless, in the last decade, experts retained by SOB plaintiffs have argued that the early studies are methodologically flawed and that, when more rigorous methods are used, the empirical evidence demonstrates that SOBs have no secondary effects at all. This argument is incorrect, of course. In §3 below, we review some of the studies conducted since 2001. Since the plaintiff in J.L. Spoons serves alcohol and offers live adult entertainment, we concentrate on that SOB subclass. Contrary to the claims made by plaintiffs’ experts, the consensus finding of earlier studies demonstrating large, significant secondary effects is further reinforced by those more recent studies. In §4-5, we respond to the reports by Drs. Linz and Freeman. Dr. Linz’s report concerns the secondary effects of SOBs on ambient crime and is organized around three general opinions which we paraphrase here as: !

No criminological theory predicts a correlation between SOBs and crime. The routine activity theory of hotspots, in particular, does not apply to SOBs.

!

Secondary effects studies conducted in other jurisdictions generally find an SOB-crime correlation. These studies are unreliable, however.

!

Secondary effects studies conducted by Dr. Linz and colleagues are reliable and lead to the conclusion that SOBs have no secondary effects. The State of Ohio ignored these studies.

We disagree with all three of these general opinions. Dr. Linz uses the results of several analyses to support his opinions. In each case, however, we interpret the analytic results differently and R ICHARD M CC LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 3

disagree with Dr. Linz’s conclusions. In addition to relying on the broader secondary effects literature, the State relied on evidence collected in public hearings. We review this evidence in §4.6 below. Dr. Linz does not refute this evidence. Dr. Freeman’s report concerns the secondary effects of SOBs on real estate prices. To estimate the magnitude of this secondary effect, Dr. Freeman statistically adjusts the sales prices of homes in Cleveland, Columbus, Dayton and Toledo for some of the variables that determine the value of a home – number of bedrooms, baths, etc. – and then computes correlations between proximity to an SOB and adjusted prices. If proximity to an SOB adversely affects the value of a home, the correlations should be large and negative. Finding mostly small correlations, however – except in Columbus where he finds a large, positive correlation – Dr. Freeman concludes that there is no evidence to support the consensus opinion that SOBs have adverse secondary effects on real estate values. Although Dr. Freeman’s “hedonic method” is widely used in the planning literature, it is relatively novel in secondary effects research. In our opinion, salient differences between SOBs and other NIMBY (“not in my back yard”) sites raise doubts about the utility of hedonic methods in secondary effects research. The large, positive correlation found in Columbus is symptomatic of the problems of the method. If one were to take Dr. Freeman’s results seriously, then opening an SOB next door to a Columbus residence would raise the property’s value by 127 percent. We know of no legislators who would find this inference reasonable. A more likely interpretation of Dr. Freeman’s results is that they are an artifact of an inappropriate method. A more widely used method for investigating the secondary effects of SOBs on property values relies on surveys of real estate appraisers. In §5.2, we report the result of a survey of real estate appraisers that is particularly relevant to J.L. Spoons. In the opinion of a large sample of real estate appraisers, opening an adult cabaret (the SOB subclass at issue in J.L. Spoons) within 500 feet of a residential or commercial property adversely affects the property’s value. Opening a non-adult cabaret (liquor and live non-adult entertainment) within 500 feet will have a smaller adverse affect in the appraisers’ opinions. One might discount survey data on the grounds that they are subjective. They support a reasonable conclusion, however, and are optimally relevant to J.L. Spoons. Dr. Freeman’s statistical results are unreasonable, in contrast, and not necessarily relevant. Although hedonic pricing methods have been used to model the secondary effects of other land uses (NIMBY sites, parks, etc.) and pollution point-sources, they are a novelty in this field. Theoretical differences between SOBs and these other land uses question the suitability of the methods. The perverse implications of Dr. Freeman’s statistical results reinforce this view. After discussing the opinions of Drs. Linz and Freeman, we address several important methodological issues. In §6, we discuss the use of 911 calls to measure ambient crime risk. As a general rule, Dr. Linz and other plaintiffs’ experts prefer this measure. Criminologists prefer the R ICHARD M CC LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 4

more traditional measure on statistical and substantive grounds. In §7, we discuss statistical hypothesis tests, particularly the conventions of statistical power. Dr. Linz and other plaintiffs’ experts have conducted secondary effects studies that, in their opinion, demonstrate that SOBs have no secondary effects. When the widely accepted statistical power conventions are taken into consideration, however, these null findings are shown to be inconclusive. Finally, in §8, we list our references and authorities.

R ICHARD M CC LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 5

1.

The Criminological Theory of Secondary Effects

It is a scientific fact that SOBs, as a class, pose large, statistically significant ambient public safety hazards. The public safety hazard consists not only of “victimless” crimes (prostitution, drugs, etc.) but, also, the “serious” crimes (assault, robbery, etc.) and “opportunistic” crimes (vandalism, trespass etc.) associated with vice. Table 1 - Secondary Effect Studies Relied on by Legislatures Los Angeles, CA Whittier, CA St. Paul, MN Phoenix, AZ Minneapolis, MN Indianapolis, IN Austin, TX Garden Grove, CA Manhattan, NY

1977 1978 1978 1979 1980 1984 1986 1991 1994

Times Square, NY Newport News, VA Dallas, TX San Diego, CA Greensboro, NC Centralia, WA Daytona Beach, FL Montrose, IL Sioux City, IA

1994 1996 1997 2002 2003 2003 2004 2005 2006

The SOB-crime relationship qualifies as a “scientific fact” on several grounds. First, the crime-related effects of SOBs are predicted by a strong scientific theory. Second, these expected theoretical relationships have been corroborated empirically. On the second point, Table 1 lists eighteen empirical studies whose findings corroborate the claim that SOBs pose large, significant ambient public safety hazards. The extensive range of time-frames, locations, and circumstances represented by these studies suggests that the uniform finding is general and robust. 1.1

The Routine Activity Theory of Crime Hotspots

The consensus finding of this literature becomes scientific fact when it is interpreted in the context of a scientific theory. In this instance, the SOB-crime relationship is predicted by a mainstream theory of modern scientific criminology. The so-called routine activity theory4 answers the what-when-where questions of victimization risk. As applied to “hotspots of predatory crime,” such as SOB sites, the theory holds that ambient crime risk, generally defined as the number of crimes within 500-1000 feet of a site, is the product of four risk factors. This can be written as:

4

This theory is due to Cohen and Felson (1979; Felson and Cohen, 1980; Felson, 1998). The routine activity theory is one of the most validated theories in modern social science. In 2005 alone, according to the Social Science Citation Index, the 1979 Cohen-Felson article was cited 621 times. The “hotspot” application of the theory is due to Sherman, Gartin, and Buerger (1989) and to Brantingham and Brantingham (1981; 1993). R ICHARD M CC LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 6

N of Targets x Average Value Ambient Crime Risk

=

× Offenders Police Presence

An increase (or decrease) in the number of targets at the site or in their average value yields an increase (or decrease) in ambient crime risk. An increase (or decrease) in police presence, on the other hand, yields a decrease (or increase) in ambient crime risk. 1.1.1 Targets SOB sites are crime hotspots because they attract potential victims, or targets, from wide catchment areas. In this respect, SOB sites are comparable to tourist attractions (Dimanche and Lepetic, 1999; Danner, 2003) and sporting events (Corcoran, Wilson and Ware, 2003; Westcott, 2006). However, compared to the targets found at these better known hotspots, the targets found at SOBs are exceptionally attractive to offenders. This reflects the presumed characteristics of SOB patrons. The patrons do not ordinarily live in the neighborhood but travel long distances to the site.5 They are disproportionately male, receptive to vice overtures, and carry cash. Most important of all, when victimized, they are reluctant to involve the police. From the offender’s perspective, they are “perfect” victims. 1.1.2 Offenders The crime-vice connection has been a popular plot device for at least 250 years. John Gay’s Beggar’s Opera (1728), for example, describes the relationship between MacHeath, a predatory criminal, and the vice ring composed of Peachum, Lucy, and Jenny. This popular view of the relationship between vice and predatory crime is reinforced by the empirical literature on criminal lifestyles and thought processes. In an ethnographic study of London’s Underworld, Victorian reformer Henry Mayhew (1851, 75-88) describes a type of prostitute who conspires with predatory criminals to rob patrons. Seventy-five years later, sociologist Clifford R. Shaw (1930; Snodgrass, 1982) documents the life of “Stanley,” a delinquent who lives with a prostitute and preys on her clients. This routine activity theory of hotspots assumes there exists a pool of rational offenders who travel freely from site to site, opting to work the most attractive site available. These

5

In 1990, as part of an investigation, Garden Grove police officers ran registration checks on motor vehicles parked at SOBs. Virtually all of the vehicles were registered to addresses outside Garden Grove. The 1986 Austin, TX study arrived at the same finding. More recently, the Effingham County Sheriff’s Department ran registration checks on motor vehicles parked at an SOB in the Village of Montrose. Except for employees’ vehicles, all were from outside the county.

R ICHARD M CC LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 7

offenders lack legitimate means of livelihood and devote substantial time to illegitimate activities; they fit Sutherland’s (1937) definition of “professional thieves.” Outside of that, they are a heterogeneous group. Some are vice purveyors who dabble in crime. Others are predatory criminals who lure and lull their victims with offers of vice. Despite their heterogeneity, these offenders share in common a rational decision-making calculus that leads them to seek out adult business sites. 1.1.3 Target Value Criminological thinking has changed little in the 75 years since Shaw’s (1930) Jack-Roller. To document the rational choices of predatory criminals, Wright and Decker (1997) interviewed 86 active armed robbers. Asked to describe a perfect victim, all mentioned victims involved in vice, either as sellers or buyers. Three of the armed robbers worked as prostitutes: From their perspective, the ideal robbery target was a married man in search of an illicit sexual adventure; he would be disinclined to make a police report for fear of exposing his own deviance (p. 69). The rational calculus described by these prostitute-robbers echoes the descriptions of other predators (see Bennett and Wright, 1984; Feeney, 1986; Fleisher, 1995; Katz, 1988, 1991; Shover, 1996). Individuals involved in drugs as a form of vice are also rationally selected as robbery targets. As one robber noted in Jacobs’ study (2000:29), “you can never report a drugrelated robbery.” 1.1.4 Police Presence Holding constant the quantity and value of the targets at a site, rational offenders choose sites with the lowest level of visible police presence. In strictly physical terms, increasing (or decreasing) the number of police physically on or near a site reduces (or increases) ambient risk. However, police presence can also be virtual through remote camera surveillance and similar processes. The effectiveness of both physical and virtual police presence can be affected – for better or worse – by broadly defined environmental factors. For example, due to the reduced effectiveness of conventional patrolling after dark, crime risk rises at night, peaking around the time that taverns close. Darkness has a lesser effect on other policing strategies, which raises the general principle of optimizing the effectiveness of police presence. One theoretical reason why SOB subclasses might have qualitatively different ambient risks is that they have different optimal policing strategies. 1.2

Theoretical Role of Subclasses

In lawsuits, SOB plaintiffs have argued that their narrowly-defined SOB subclass is exempt from criminological theory. But in fact, the relevant criminological theory applies to all R ICHARD M CC LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 8

subclasses. To the extent that two SOB subclasses attract similar patrons from similarly wide catchment areas, theory predicts similar ambient crime risks. Put simply, similar causes (the presence of many high-value targets and low levels of police presence) have similar effects (i.e., high ambient crime risk). Existing data support these predicted relationships. Although the theory applies equally to all SOB subclasses, at the same time, it allows for qualitative differences among the subclasses. Subclass-specific risks arise in some instances. This occurs when the defining property of the subclass implies (or creates) idiosyncratic opportunities (or risks) for particular types of crime. Compared to the complementary subclass, for example, SOBs that serve alcohol present idiosyncratic opportunities for non-instrumental crimes, especially simple assault, disorderly conduct, etc. SOBs that provide on-premise entertainment present idiosyncratic opportunities for vice crime, customer-employee assault, etc. Criminologists call this etiological crime category “opportunistic.” There are many obvious examples and SOB regulations often treat subclasses differently because of their varying ambient opportunity structures. Qualitative differences also arise when the defining property of the subclass compromises the effectiveness of common policing strategies. Policing SOBs that offer on-site entertainment (adult cabarets, peep shows, etc.) may require that police officers inspect the interior premises, for example. Because this places officers at risk of injury, policing on-site SOBs requires specially trained and equipped officers, prior intelligence, specialized backup manpower, and other resources. Furthermore, routine drive-by patrols to “show the flag” are less effective because such premises offer cover for potential offenders who can wait inside without arousing suspicion. The optimal policing strategies for two subclasses are sometimes incompatible or even mutually exclusive. To illustrate, an optimal policing strategy for SOBs that do not offer on-site entertainment, such as adult video and book stores, often involves neighborhood patrols by uniformed officers in marked cars. Visibility is a key element of this strategy. For peep shows and adult cabarets, on the other hand, the optimal policing strategy often involves boots-on-theground deployments of plainclothes officers and unmarked cars. Avoiding visibility is a key element of this strategy. Obviously, neighborhood patrols by plainclothes officers driving unmarked cars would defeat a major purpose of drive-by patrols; likewise, sending uniformed officers into an adult cabaret would be an inefficient method of control and might pose a physical danger to the officers, patrons, and employees. As a general rule, distinct SOB subclasses may require distinct policing strategies to mitigate ambient crime risks. To some extent, differences among the optimal policing strategies for SOB subclasses amount to differences in cost. In many (but certainly not all) instances, the least expensive policing strategy involves drive-by patrols by uniformed officers in marked cars. Beyond the deterrent value of visible drive-by patrols, patrol officers can keep watch for known offenders and suspicious activity. When potential problems are spotted, the patrol officers can forward the information to a specialized unit or, if necessary, handle it on the spot, requesting backup

R ICHARD M CC LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 9

resources only as needed.6 It is important to realize, nevertheless, that the implementation of a policing strategy is determined in large part by local exigencies. 1.3

Theoretical Role of Alcohol

Proximity to alcohol is a key component of the criminological theory of secondary effects. Alcohol aggravates an SOB’s already-high ambient crime risk by lowering the inhibitions and clouding the judgments of the SOB’s patrons. In effect, alcohol serves to further soften the already soft targets found at the SOB site. The available data corroborate this theoretical expectation in all respects. Predatory criminals prefer inebriated victims,7 and SOBs that serve alcohol or that are located near liquor-serving businesses pose accordingly larger and qualitatively different ambient public safety hazards.8 Governments rely on this consistent finding generated from crime-related secondary effect studies as a rationale for limiting nudity in liquor-serving businesses. Though not explicitly represented as a risk factor in the criminological theory, laboratory research demonstrates a relevant adverse interaction between consumption of alcohol and adult entertainment. Experimental subjects who drank alcohol and viewed pornography were more aggressive than subjects who drank alcohol or viewed pornography (Norris et al., 2002; Davis et al., 2006). Relying on this research finding, several states have limited the availability of alcohol in SOBs. 1.4

Theoretical Role of Mitigation Strategies

The routine activity theory points to strategies for mitigating the crime-related secondary effects of SOBs. In principle, the effects of a mitigation strategy can be direct or indirect . Direct effects are typically realized through direct manipulation of the risk factors to reduce ambient risk. Indirect effects are realized by making the risk factors more efficient. In practice, of course, some of the strategies are expensive or otherwise impractical. Our review of these

6

See, e.g., National Research Council. Fairness and Effectiveness in Policing: The Evidence. National Academies Press, 2004. 7

See, e.g., Wright and Decker (1997, p. 87): “[E]ach of (the armed robbers) expressed a preference for intoxicated victims, who were viewed as good targets because they were in no condition to fight back.” (p. 70); “Several [armed robbers] said that they usually chose victims who appeared to be intoxicated because, as one put it, ‘Drunks never know what hit them.’” 8

A 1991 study of Garden Grove, California by McCleary and Meeker found a large, significant increase on ambient crime risk when an alcohol-serving establishment opened within 500 feet (ca. one city block) of an SOB. Secondary effect studies in Greensboro (2003) and Daytona Beach (2004) found that alcohol-serving SOBs had larger secondary effects than retail alcohol outlets. These studies are reviewed in §2-3 but see, especially, footnote 21 below. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 10

strategies begins with one of the most expensive, least practical mitigation strategies. 1.4.1 Increased Police Presence The simplest, surest way to mitigate ambient crime risk is to assign more police to SOB neighborhoods. Although the relationship between police presence and ambient crime risk is complicated and complex, criminologists generally accept the aphorism: “more police, less crime.”9 Unfortunately, this simplest, surest mitigation strategy is expensive and impractical. From the government’s perspective, increasing the number of police patrols in a neighborhood is prohibitively expensive. From the perspective of the SOB and its patrons, police presence can be highly intrusive, bordering on “harassment.” In principle, fixed levels of police presence can be made more effective by fine-tuning existing policing strategies. Police patrols can be made more visible, e.g., by using uniformed officers in marked vehicles instead of plain-clothes officers in unmarked vehicles. Most police departments have already optimized their strategies, however. Police effectiveness can also be enhanced by incorporating rational enforcement policies into SOB codes. Several examples are described in subsequent sections. Figure 1.4.2 - Robbery Risk by Distance from an SOB

9

See, e.g., S.D. Levitt. Using electoral cycles in police hiring to estimate the effect of police on crime. American Economic Review, 1997, 87:270-290. “Increases in police are shown to substantially reduce violent crime but have a smaller impact on property crime. The null hypothesis that the marginal social benefit of reduced crime equals the costs of hiring additional police cannot be rejected.” (p. 270). Some “victimless” vice crimes are an exception to the rule, of course. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 11

1.4.2 Increased Distance from Sensitive Uses Reducing the density of targets in an SOB neighborhood is a more economical, practical mitigation strategy. As a rule, the most problematic secondary effects are associated with dense concentrations of SOBs (e.g., Boston’s “combat zone” model). Accordingly, many governments require minimum distances between SOB sites (e.g., the Detroit model). In addition to reducing per-site target density, in turn decreasing aggregate risk, this model minimizes many obstacles to routine policing. Figure 1.4.2 demonstrates the rationale for a related mitigation strategy.10 The vertical axis of this “risk-distance function” is calibrated in units of Part I personal crime (homicide, aggravated assault, robbery, and rape) risk, relative to the neighborhood risk, for 28 Greensboro SOBs for 1996-2005. The horizontal axis is calibrated in distance from an SOB. The unit of distance is a city block which, in the Greensboro neighborhoods from which these data are taken, is approximately 400 feet. Suppose that a person exits a building five city blocks (i.e., 2,000 feet) from an SOB. As this person walks toward the SOB, his or her victimization risk rises. For the first few blocks, the risk increments are modest; thereafter, the risk increments grow large. At two blocks from the SOB, the person’s risk is double what it was at the start of the five-block walk. At one-half block, the risk is six times higher. If the person walks away from the SOB site, his or her victimization risk falls until, at a distance of three blocks from the site, the risk decrements are imperceptible. Governments can take advantage of the risk-distance relationship plotted in Figure 1.4.2 by setting minimum distances between SOBs and other sensitive land uses. SOB patrons have no choice but to “run the gauntlet.” The victims of some ambient crime incidents are not SOB patrons, however, but rather, are neighborhood residents and passers-by. By setting minimum distances between SOBs and the land uses frequented by these people, the government mitigates the SOB’s ambient crime risk secondary effect.11 1.4.3 Limited Hours of Operation Another economical and practical strategy for mitigating the ambient crime risk of SOBs is

10

Risk-distance functions are revisited in §3-4 below.

11

We are often asked to specify a distance sufficient to fully mitigate an SOB’s ambient crime risk. The correct answer to this question – “As far as possible” – is not helpful. Although the risk-distance function plotted in Figure 1.4.2 seems to answer this question, remember that it is the average of 28 SOB sites. By definition, some sites are “better,” some “worse.” Planners must assume a worst case scenario while balancing this assumption with practical (and legal) considerations. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 12

to limit the hours of operation. Routine activities theory reduces to the aphorism, “more targets, more crime.” And in the overnight hours when businesses close and people go home, the crime rate drops. While the crime rate drops, however, the per-target risk rises. When a business stays open around-the-clock, its victimization risk rises steadily after sundown, peaking in the early morning. Darkness softens a target, increasing its attractiveness to predatory criminals. Several mechanisms operate here but the most salient is that routine policing is more difficult and less effective in darkness. When bars and taverns close, police resources are stretched thinner, making already soft targets even softer. Governments typically mitigate this risk by closing high-risk public places (playgrounds, beaches, parks, etc.) from dusk to dawn; by imposing curfews on high-risk persons (teenagers, parolees, etc.); and by limiting the operation of high-risk businesses (bars, SOBs, etc.) during times of acute risk. Not surprisingly, this theoretical prediction is confirmed by the empirical evidence. 1.4.4 “Hardening” SOB Sites12 In principle, ordinances can mitigate ambient crime risk by requiring SOBs to “harden” their properties. Mandating outdoor lighting, parking lot surveillance cameras, and anti“cruising” structures illustrate strategies for hardening the site’s exterior. There are very few options for exterior hardening, unfortunately; and although the effectiveness of exterior hardening strategies depends to some extent on local circumstances and conditions, there is little evidence that any of the typical options can mitigate ambient crime risk. Regulating the interior configurations of SOBs, in contrast, has a stronger rationale in criminological theory. Moreover, interior hardening strategies are often less costly, more practical, and in theory, more effective. Three widely used strategies illustrate the general principle: !

Ordinances that eliminate interior blind spots

!

Ordinances that prohibit closed viewing booths

!

Ordinances that restrict entertainers to raised stage areas

Each of these strategies reduces the risk of on-premise victimization of patrons and employees.13 In some respects, the risk reduction mechanism is obvious. Removing blind spots and opening up closed booths obviously reduces the opportunity for lewd behavior, e.g. Though less obvious, to

12

The classic statement on “hardening” is Oscar Newman’s Defensible Space: Crime Prevention Through Urban Design. (New York: MacMillan, 1973). 13

The strategies also facilitate routine enforcement while minimizing the risk of injury to police officers. Those topics are discussed separately in the next section. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 13

the extent that patron-on-patron, patron-on-employee, and employee-on-patron confrontations are precipitated by lewd behavior, these strategies also reduce the risk of assault. The highest risk of patron-on-patron, patron-on-employee, and employee-on-patron crime occurs inside SOBs that feature live entertainment; and of course, alcohol aggravates the risk. The risk can be mitigated by separating patrons and entertainers. Ideally, separation is achieved by mandated structures, such as raised stages. By creating a tangible “wall” between employees and patrons, raised stages reduce unintentional (or intentional) “touching,” thereby decreasing the risk of patron-on-employee and employee-on-patron crime. 1.4.5 Police Officer Safety While assaults on police officers are rare, they are among the most serious crimes that occur inside SOBs. In theory, moreover, they are preventable. Since the risk of assault extends from the time officers enter the SOB until they leave, mitigation strategies are aimed at minimizing the number of times officers must enter SOBs and the length of time they must spend inside upon entry. Strategies that focus on the latter factor are more practical. Police officers enter SOBs either in response to a reported crime incident or to inspect the premises as part of routine enforcement. By reducing the risk of on-premise crime incidents, the interior target-hardening strategies described in the preceding section reduce the number of times that officers must enter SOBs to respond to reported incidents. Otherwise, there are few options for reducing the number of times that officers must enter SOBs. Not withstanding the risk to officers, routine inspection can be an effective mitigation strategy. By focusing attention on SOB sites, routine inspection reduces ambient risk through a complex set of pathways collectively referred to as “broken windows.”14 Regardless of how officers come to be inside an SOB, any strategy that minimizes the amount of time spent inside reduces the risk of injury. Ordinances aimed at improving interior visibility illustrate these strategies. In many instances, officers can accomplish their purpose with a quick visual inspection. If the interior of the SOB is well lit and obstacle-free, the inspection can be completed by a single officer in a minute or two. If the interior is dark and/or labyrinthian, the same inspection may require two (or more) officers for a longer period of time. In SOBs that feature live entertainment, a raised stage reduces the risk of injury to police officers through the same mechanism. If an ordinance mandates, say, a six-foot distance between patrons and entertainers, absent a raised stage, enforcing (and/or detecting willful violations of)

14

The best known statement of this effect is “Broken windows: The police and neighborhood safety.” by J.Q. Wilson and G.L. Kelling, Atlantic Monthly, 1982, 249:29-38. Wilson and Kelling argue persuasively that police visibility in a neighborhood can have a greater impact on victimization risk than police activities that target crime per se. Modern police methods are based on this theory. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 14

the ordinance may require that several plainclothes officers spend an hour or more inside. With a raised stage, on the other hand, a comparable level of enforcement and detection of violations can be accomplished with shorter, more superficial inspections. Raised stages also facilitate selfenforcement. Lacking a raised stage, ensuring that patrons and entertainers comply with a distance rule demands constant attention and keen judgement by the SOB. A raised stage facilitates self-enforcement by the SOB, thereby reducing the risk of patron-patron and employeepatron confrontations. 1.4.6 Tailoring Regulations to Local Needs The ideal SOB ordinance combines low compliance costs for the SOB with low enforcement costs for the government. To some extent, compliance and enforcement costs depend on local circumstances and conditions and these often dictate differences in codes and/or enforcement strategies. A code or strategy that is optimal for one set of circumstances may be less than optimal for another. If a local variation is aimed at rationalizing regulation and optimizing mitigation, it should be encouraged. By definition, local conditions are too numerous to list. Nevertheless, the principle is straightforward. Legislatures adapt and modify codes to address local idiosyncracies. In most instances, modifications are designed to facilitate compliance and minimize enforcement costs. Toward that end, legislatures often consult local enforcement officers and, to the extent possible and appropriate, incorporate the views of experts into the regulations. 1.5

Concluding Remarks on Criminological Theory

The legal debate over the crime-related secondary effects of SOBs ignores the crucial role of criminological theory. Without exception, criminological theory predicts that SOBs will generate ambient public safety hazards. Witnesses for SOBs produce study after study to show that SOBs have no crime-related secondary effects or, sometimes, that SOBs have beneficial public safety impacts on their neighborhoods. We discuss the details of these studies at a later point. For present purposes, the criminological theory that we have described is internally consistent, compelling, and widely accepted. The historical record indicates the wide acceptance of this theory. Recapitulating the opinions of experts in Young v. American Mini-Theaters (1976), e.g., Justice Stevens outlines the crux of the theory: In the opinion of urban planners and real estate experts who supported the ordinances, the location of several such businesses in the same neighborhood tends to attract an undesirable quantity and quality of transients, adversely affects property values, causes an increase in crime, especially prostitution, and

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 15

encourages residents and businesses to move elsewhere.15 Three decades later, in a case that has much in common with J.L. Spoons, the Illinois Supreme Court describes the theory in more explicit detail:

The City also adduced testimony from Dr. Wesley Skogan, a professor of political science at Northwestern University and a member of the University’s Institute for Policy Research. According to Professor Skogan, research shows that establishments serving alcohol attract a significant amount of additional crime. Such establishments create the opportunity for crime by bringing the potential victim and the criminal together. Victims become more vulnerable because of alcohol's debilitating effects, thus creating an attractive situation for potential offenders.16 As it turns out, the criminological theory is well corroborated by data. For reasons that will become obvious shortly, it will be useful to separate the secondary effects studies into “earlier” and “later” categories conducted, respectively, prior to and after 2001. One major difference between earlier and later studies is that the former were government-sponsored for the most part while the latter were often sponsored by SOB plaintiffs. For reasons that are not yet apparent, secondary effects studies sponsored by governments and plaintiffs use different methodologies and arguments. When idiosyncratic differences are taken into account, however, the findings of the earlier and later studies are wholly consistent.

15

Young v. American Mini-Theaters (1976)

16

City of Chicago v. Pooh Bah Enterprises, Inc., 865 N.E.2d 133 (Ill.2006) R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 16

2

Early Studies Corroborate the Theory

Scientific theory leads us to expect secondary effects in SOB neighborhoods and, in fact, that is exactly what the early studies found. Table 1 lists eighteen studies conducted over a 30year period in rural, urban, and suburban settings; the studies span all regions of the U.S. and every conceivable SOB subclass. Despite this diversity, the eighteen studies have one thing in common: Each reports what we call the “consensus finding of the literature” – a substantively large, statistically significant crime-related secondary effect. Given the theoretical expectation and ample corroboration, this consensus finding is a scientific fact. The eighteen studies listed in Table 1 are also methodologically diverse. Some of the studies use a before/after difference to estimate a secondary effect. Others use SOB-control differences for that purpose.17 Some of these SOB-control studies select control zones by “matching.” Others use statistical models (e.g. regression) to adjust irrelevant differences between the SOB and control zones. Methodological attacks on the literature typically focus on idiosyncratic design features of each study. Despite their methodological idiosyncracies, the studies all report remarkably similar findings. This consensus renders any methodological challenge implausible. Ideally, one could read each of the eighteen studies listed in Table 1 and draw inferences from their similarities and differences. Given the broad consensus finding, however, there is little to learn from the minor details of specific studies. Our review will focus on SOB subclasses and, to a lesser extent, on methodological idiosyncracies. We revisit the methodological issues in subsequent sections. 2.1

SOB-Control Contrasts: Phoenix, 1979

In many respects, true experiments are the strongest designs.18 But since true experiments are not possible, crime-related secondary effect studies rely on quasi-experimental designs. Except for random assignment, quasi-experimental and true experimental designs use similar structures to control threats to validity. The strongest quasi-experimental design compares ambient crime risk at a site before and after the opening of an SOB. Before-after contrasts are

17

Our authority on quasi-experimental design is Experimental and Quasi-Experimental Designs for Research by D.T. Campbell and J.C. Stanley (Rand-McNally, 1966). Campbell and Stanley call before/after designs “pretest-posttest” designs; they call SOB-control designs “static group comparison” designs. In general, before/after comparisons are prone to fewer threats to internal validity and, hence, are “stronger” than SOB-control designs. 18

An experimental design controls common threats to validity by random assignment. To estimate the crime-related secondary effects of SOBs experimentally, e.g., we would compile a list of the business sites in a jurisdiction and open SOBs in a random sample of sites. Random assignment (and hence, experimenting) is not possible, of course. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 17

not always possible, unfortunately. A somewhat weaker quasi-experimental design compares ambient crime risk at an SOB site to ambient crime risk at a control site. Though weaker in principle, SOB-control contrasts are often more practical. For this comparison to be valid, the two sites must be comparable on all other relevant causal variables. Barring outright dishonesty, the differences will be small and approximately random, thereby favoring neither side. In 1979, the City of Phoenix conducted a study of crime-related secondary effects. To estimate the crime-related secondary effects of SOBs, the researchers compared crime rates in areas with SOBs to crime rates in “matched” control zones (i.e., similar areas containing no SOBs). The comparisons are summarized in our Table 2.1. The property and personal crime rates reported in Table 2.1 were estimated from Uniform Crime Report (UCR) data. The percentages reported in the right-hand column (in red) are the secondary effect estimates derived from the crime rates. Compared to crime rates in the control zones, the UCR property crime rate was 39.8 percent higher; the UCR personal crime rate was 13.7 percent higher; and the UCR sex crime rate was 480.2 percent higher in the adult business areas. By any reasonable standard, these are large, significant crime-related secondary effects. Table 2.1 - Secondary Effects in Phoenix, AZ

Property Crime Rate Personal Crime Rate Sexual Crime Rate

SOB Areas

Control Areas

Secondary Effect

122.86 5.81 9.40

87.90 5.11 1.62

139.8 % 113.7 % 580.2 %

Source: ADULT BUSINESS STUDY, City of Phoenix Planning Department, May 25, 1979; Table V

Over the last 30 years since this study was conducted, legislatures around the U.S. have accepted and relied upon its findings. Witnesses retained by SOBs and SOB plaintiffs, on the other hand, have argued that the 1979 Phoenix study is “fatally flawed” and that its findings are wholly implausible. It is our opinion that this position is incorrect. Although the design of this study leaves much to be desired – especially by today’s standards – many of the study’s methodological shortcomings function to produce smaller effect sizes. If anything, a stronger design would have produced a larger effect estimate.

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 18

2.2

Before-After Contrasts: Garden Grove, 1991

Prior to 1990, virtually all crime-related secondary effect studies compared crime rates in police districts with SOBs to crime rates in districts without SOBs.19 By contemporary standards, this is a weak study design. Existing police districts covered several square miles, e.g., and sometimes contained several SOBs. Researchers handled these problems as best they could by matching and, infrequently, by statistical adjustment. To some extent, the wide use of weak “static group comparison” designs was dictated by economics. Prior to 1990, relatively few police departments had sophisticated information management systems. Citing these methodological flaws, experts retained by SOB plaintiffs characterized these studies as illustrations of “shoddy research” whose findings are not to be trusted. Ironically, the methodological flaws in these early studies favor a null finding.20 If anything, stronger designs would most likely have produced larger, more significant effect estimates. This issue aside, the “static group comparison” design assumes that SOB and control neighborhoods are equivalent on relevant crime risk factors. If this assumption is unwarranted, observed secondary effects cannot be attributed to the SOBs. The surest, simplest way to control this threat to validity is to use a before-after design. In the late 1980s, Richard McCleary and James W. Meeker conducted a secondary effects study in Garden Grove, CA designed to correct many of the methodological “flaws” identified by SOB plaintiffs.21 The Garden Grove design used location-coded crime data so that crime rates could be estimated in 500-foot circles centered on an SOB, e.g.; the design used a full decade of crime data, so that relatively stronger before/after contrasts could be used; and finally, the design took advantage of several nearly ideal control businesses for controls. Due in large part to these design innovations, two decades later, the 1991 Garden Grove study is considered to be the most rigorous, valid study of secondary effects in the literature.22

19

Studies in Los Angeles (1977), Amarillo (1977), Whittier (1978), St. Paul (1978), Phoenix (1979), Indianapolis (1984), and Austin (1986) used this design. 20

“Null finding” means “finding that SOBs have no secondary effects.”

21

Final Report to the City of Garden Grove: The Relationship between Crime and Adult Business Operations on Garden Grove Boulevard. October 23, 1991. Richard McCleary, Ph.D. and James W. Meeker, J.D., Ph.D. 22

E.g., in City of Chicago v. Pooh Bah Enterprises (224 Ill.2d 390, 865 N.E.2d 133): “According to Professor [Wesley G.] Skogan ... establishments serving alcohol attract a significant amount of additional crime ... by bringing the potential victim and the criminal together. Victims become more vulnerable because of alcohol’s debilitating effects, thus creating an attractive situation for potential offenders ... Professor Skogan discussed a number of studies and articles showing the relationship between alcohol consumption and crime ... The Garden R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 19

Examining ambient crime before and after an SOB opened, McCleary and Meeker found that crime risk rose whenever an SOB opened for business; when an SOB closed, crime risk fell. The validity of a before/after design requires that other plausible explanations for the rise and fall of crime be ruled out. The change may be a coincidence, e.g.; the observed patterns may reflect more general trends of crime rising or falling throughout the city. To control these common “threats to internal validity,” McCleary and Meeker replicated each before/after analysis for other SOBs in Garden Grove. If a rise or fall in ambient crime were a coincidence and reflective of a more general trend, the effect would be observed at other Garden Grove SOBs. Failing to observe the effect at these control sites, on the other hand, the effect could be attributed confidently to the newly opened SOB. Table 2.2 - Secondary Effects in Garden Grove, CA: Business Openings Total “Serious” Crime, One Year Before/After Test Sites Before

After

March, 1982 March, 1986 August, 1988

71 31 32

106 68 50

Total

134

224

Control Sites Before

After

1.49 2.19 1.56

76 80 41

78 92 40

1.03 1.15 0.98

1.67

197

210

1.06

Source: Final Report to the City of Garden Grove, pp. 26-28

Secondary effects for three business openings are reported in Table 2.2. When a new SOB opened, the total number of “serious” crimes in a 500-foot radius around the site increased, on average, 67 percent. To control for the confounding effects of city-wide crime trends, changes in police activity, and other common threats to internal validity, these before-after differences were compared to the analogous differences for the addresses of existing SOBs. Total “serious” crimes in a 500-foot radius around these “control” sites rose, on average, only six percent. The secondary effect observed when new SOBs open is, thus, substantively large and statistically significant. Social scientists and their government clients learned two things from the 1991 Garden Grove study. First and foremost, the availability of relatively stronger before-after quasiexperimental designs produces the same ambient public safety hazards. The Garden Grove findings corroborate the findings in the Los Angeles (1977), Phoenix (1979), and Indianapolis

Grove study is particularly noteworthy because it showed that the combination of retail alcohol sales and adult uses created a higher incidence of crime than resulted from retail alcohol sales or adult uses operating in isolation.” R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 20

(1984) studies. Second, however, the 1991 Garden Grove study demonstrated how expensive a crime-related secondary effect study can be. We will have more to say about this at a later point. 2.3

What We Learned from the Early Studies

By contemporary standards, the earliest secondary effects studies were relatively unsophisticated. In particular, the early studies compared ambient crime risk in existing police precincts rather than in the smaller impact areas suggested by criminological theory. The use of weak quasi-experimental designs in these early studies was dictated by fiscal reality.23 Yet despite their design weaknesses, these studies produced a consistent picture that came into sharper focus as stronger, more sophisticated studies added to the consensus finding of large, statistically significant crime-related secondary effects generated by SOBs.

23

In their 1991 Garden Grove study, McCleary and Meeker spent more than $100,000 (adjusted for inflation) for a stronger, more sophisticated quasi-experimental design. The study’s cost was a minor scandal in some city departments. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 21

3.

Adult Cabarets

The relatively weak designs used in the early literature opened the door to charges by SOB plaintiffs that the strong consensus finding of the literature is an artifact. Had the studies used stronger designs, according to the plaintiffs’ argument, all would have arrived at the opposite conclusion.24 But in fact, the very consistency of the early literature rules out an artifactual explanation. First, virtually all design weaknesses bias the study in favor of the null finding. Second, more recent studies that use stronger, more sophisticated designs yield the same finding as the weaker, less sophisticated early studies. In addition to their greater methodological sophistication, secondary effect studies conducted after 2001 differ from the earlier studies in two respects. First, many of the later studies were commissioned by SOB plaintiffs. Second, reacting to legal arguments by SOB plaintiffs, many of the later studies investigated the secondary effects of specific SOB subclasses. Following this recent development, our review of the modern literature focuses on adult cabarets, the oldest and most interesting SOB subclass. And since the plaintiffs in J.L. Spoons is an adult cabaret, it is the relevant SOB subclass. In principle, estimating the secondary effect of an adult cabaret is straightforward. If we agree that live nude entertainment is the essential difference between adult cabarets and other businesses that sell alcohol by the drink (or “taverns”), the secondary effect can be estimated by comparing the ambient crime rates for adult cabarets and taverns. Although the differences between adult cabarets and taverns are often more complicated than conceded by this simplest, straightforward design, several studies have used taverns as controls for adult cabarets. All find that adult cabarets have higher ambient crime rates than taverns. 3.1

25

NC.

Greensboro, 2003

In 2003, Dr. Daniel Linz conducted a crime-related secondary effect study in Greensboro, Analyzing 911 calls-for-service Dr. Linz concluded that: The presence of adult cabarets and adult video/bookstores in “neighborhoods” was unrelated to sex crimes in the area. We found that several of an (sic) adult video/bookstore were located in high person and property crime incident “neighborhoods.” We examined the “neighborhoods” and local areas surrounding

24

The best-known statement of this view is B. Paul, D. Linz, and B.J. Shafer. Government regulation of ‘adult’ businesses through zoning and anti-nudity ordinances: de-bunking the legal myth of negative secondary effects. Communication Law and Policy, 2001, 6:355-391. 25

Evaluating Potential Secondary Effects of Adult Cabarets and Video/Bookstores in Greensboro: A Study of Calls for Service to the Police by Daniel Linz, Ph.D. and Mike Yao, November 30th, 2003. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 22

the adult video/bookstores (1000 foot radius) further and we found that the adult video/bookstores were not the primary source of crime incidents in these locations ... (T)here is no support for the City of Greensboro’s theory that adult businesses produce adverse secondary effects. The results of our study show that adult businesses are not associated with crime events.26 Due to the technical nature of Dr. Linz’s statistical analyses, the City of Greensboro retained Richard McCleary to “translate” Dr. Linz’s numerical results into plain words.27 Dr. Linz’s report was a difficult read, even for statisticians. The numbers on which his conclusion was based were scattered across 18 pages of computer output in an appendix. Few report readers consult appendices under any circumstances. In this particular instance, a critical reading of the report’s appendices required technical skills (that most of the report’s readers lack) and great tolerance for numerical detail. When the actual numbers were finally examined, it became clear that Dr. Linz had exaggerated the basis of his strongly-worded conclusion. Put simply, Dr. Linz’s numbers contradicted his words. Figure 3.1 - Results of the 2003 Greensboro Study

Source: Tables 14-19, Evaluating Potential Secondary Effects of Adult Cabarets and Video/ Bookstores in Greensboro: A Study of Calls for Service to the Police. D. Linz and M. Yao, November 30th, 2003.

The results of Dr. Linz’s analyses are plotted in Figure 3.1. The green bars show the

26

P. 3 (counting the title sheet as p. 1) of the Linz-Yao Greensboro Study.

27

R. McCleary. A Methodological Critique of the Linz-Yao Report: Report to the Greensboro City Attorney. December 15, 2003. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 23

ambient crime levels28 for Greensboro’s “control” neighborhoods that have no taverns and no SOBs. The blue and red bars report the ambient crime levels for neighborhoods with taverns and neighborhoods with adult cabarets, respectively. To facilitate interpretation, we have fixed the ambient crime levels in control neighborhoods at 100 percent. This permits an easy interpretation of the ambient effects in tavern neighborhoods (blue bars) and adult cabaret neighborhoods (red bars) as multiples of the control neighborhood effects (green bars). Since the social, demographic, and economic variables that are presumed to “cause” crime vary across neighborhoods, unadjusted crime levels may be deceiving. To control for these confounding effects, Dr. Linz adjusted his raw numbers with a statistical model whose technical details will not be discussed here. As the adjusted effects plotted in Figure 3.1 show, Dr. Linz found that ambient crime in tavern neighborhoods (blue bars) range from 148 percent (violent crimes) to 229 percent (sexual crimes) of the ambient crime in control neighborhoods. Since tavern neighborhoods are the criminological “gold standard” of ambient crime, that result was expected.29 What Dr. Linz did not expect, however, was that adult cabaret neighborhoods (red bars) would have more crime than the tavern neighborhoods (blue bars). Crime-related secondary effects in Greensboro’s adult cabaret neighborhoods ranged from 175 percent (for property crime) to 307 percent (for sexual crime) of the ambient crime levels in control neighborhoods. These effect estimates are large in every sense and, of course, they are not surprising. To us, the only surprise was that the estimates in Figure 3.1 were reported in a study commissioned by a consortium of SOB plaintiffs. 3.2

Daytona Beach, 2004

In 2004, Dr. Linz collaborated with Dr. Randy D. Fisher on a Daytona Beach secondary effect study.30 With minor exceptions, the design of the Daytona Beach study was identical to the Greensboro design.31 Analyzing 911 calls once again, Drs. Linz and Fisher concluded that

28

We use the term crime “levels” because, strictly speaking, crime “rates” are difficult to tease out of 911 calls. 29

Most of the research on the relationship between taverns and ambient crime risk is due to our late colleague, Dennis W. (“Denn”) Roncek. Roncek and Pravatiner (1989) and Roncek and Maier (1991). 30

Evaluating Potential Secondary Effects of Adult Cabarets in Daytona Beach, Florida: A Study of Calls for Service to the Police in Reference to Ordinance 02-496 by Daniel Linz, Ph.D., Randy D. Fisher, Ph.D. and Mike Yao, April 7th , 2004. Dr. Fisher is Associate Professor of Psychology at the University of Central Florida. 31

Since the Daytona Beach SOBs were adult cabarets, Linz, Fisher, and Yao excluded bookstores and video arcades from the study. Instead of defining “neighborhoods” as Census R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 24

adult cabarets, had no significant crime-related secondary effects: We are able to account for crime events in Daytona Beach with a moderately high level of accuracy using variables found by other researchers to be related to crime…The social disorganization variables and especially the presence of an (sic) alcohol beverage retail sale establishments in the blocks (that did not feature adult entertainment) accounts largely for this explanatory power. The presence of an adult cabaret in the census block explained only (sic) a trivial amount of variability in crime incidents when these other variables were considered ... From these analyses we are able to reliably conclude that once we control for variables known to be related to crime there is not a meaningful relationship between the presence of an adult cabaret in the neighborhood and crime events.32 This conclusion was worded more cautiously than the conclusion in Greensboro. Indeed, the authors went so far in the Daytona Beach report as to admit that, as in Greensboro, the Daytona Beach results demonstrate statistically significant crime-related secondary effects: There are analyses reported below where there are small but statistically significant relationships due to the exceptionally large N (sample size) employed in the analyses (at times over 1,100 census blocks)...[But] we favor “strength” over a technical “significance.”33 This is an esoteric statistical issue, of course. In our opinion, Drs. Linz and Fisher misunderstand the assumptions of their model as well as the statistical problem of an “exceptionally large N” that, in their opinion, obviates the statistical model. Put simply, they are incorrect. Not withstanding the large statistical size of their effect estimates, the effect estimates reported by Drs. Linz and Fisher in Daytona Beach are substantively large. Figure 3.2 plots the results of the Daytona Beach analyses using the same conventions used in Figure 3.1 (for Greensboro). Once again the ambient crime levels in control neighborhoods (green) are fixed at 100 percent so that the levels in tavern neighborhoods (blue) and adult cabaret neighborhoods (red) can be interpreted as multiples of the controls. With two exceptions, adult cabaret neighborhoods have higher ambient crime levels than tavern neighborhoods. Given the wellknown relationship between taverns and ambient crime, the Daytona Beach analyses corroborate the consensus finding of the literature. Like SOBs generally, adult cabarets pose large, statistically significant ambient public safety hazards.

Block Groups, in Daytona Beach, Linz, Fisher, and Yao used Census Tracts. The Greensboro and Daytona Beach designs are otherwise identical. 32

P. 36 (counting the title sheet as p. 1) of the Linz-Fisher-Yao Daytona Beach study.

33

P. 23 (counting the title sheet as p. 1) of the Linz-Fisher-Yao Daytona Beach study. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 25

Figure 3.2 - Results of the 2004 Daytona Beach Study

Source: Tables 7-19, Evaluating Potential Secondary Effects of Adult Cabarets in Daytona Beach, Florida: A Study of Calls for Service to the Police in Reference to Ordinance 02-496. D. Linz, R.D. Fisher, and M. Yao, April 7th, 2004.

Figure 3.2 speaks for itself. Tavern neighborhoods (blue) have 90 percent more total crime than control neighborhoods (green). Adult cabaret neighborhoods (red) have 270 percent more total crime than control neighborhoods (green). In substantive terms then, taverns have large secondary effects and adult cabarets have even larger secondary effects. The fact that these effect estimates are also statistically large adds little to our understanding of Figure 3.2. The estimates are statistically large, of course – i.e., statistically significant – and that poses a dilemma for Drs. Linz and Fisher. If the estimates were statistically small, Drs. Linz and Fisher could argue that they were due to chance (regardless of their substantive size). Since they cannot make this claim, Drs. Linz and Fisher argue that statistical significance is an artifact of an “exceptionally large N.” This is a specious argument, however, on two grounds. First, samples of 1,100 are not large enough to obviate the statistical model used by Drs. Linz and Fisher. But second, if samples of 1,100 were large enough to obviate the statistical model, as claimed, all of effect estimates would be statistically significant. In fact, of the 84 parameter estimates reported by Drs. Linz and Fisher, 42 are statistically significant and 42 are not. We will return to this issue in §3.4 below. 3.3

Palm Beach County, 2004

Comparing 911 calls to the addresses of nine adult and seven non-adult cabarets in Palm Beach County, FL, Dr. Terry A. Danner found that the adult cabaret addresses had fewer crimerelated 911 calls (2.5 per month vs. 2.9 for SOB addresses) but more order-related 911 calls (3.1 per month vs. 2.0 for SOB addresses). Based on these comparisons, Dr. Danner concluded that the contrast “does not provide compelling evidence that the addition of various levels of nude dancing to the ‘nightclub type environment’ produces a pattern of crime and public disorder that R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 26

appears to be uniquely attributable to the adult cabaret category of business.”34 Dr. Danner’s idea of comparing adult and non-adult cabarets makes good sense. The legal difference between adult and non-adult cabarets is, after all, the quantity and/or quality of clothing worn by employees. Changing the quantity and/or quality of clothing changes the non-adult cabaret into an adult cabaret and vice versa. The logical implication of this argument is that ambient crime rate differences between adult and non-adult cabarets must be due to nudity. Aside from the idea of comparing adult and non-adult cabarets, however, Dr. Danner’s study is problematic in two respects. First, Dr. Danner used 911 calls to measure ambient crime risk. Although 911 calls are correlated with ambient crime risk, however, the correlation is weak at best. Second, Dr. Danner considered only the subset of 911 calls to the immediate addresses of the adult and non-adult cabarets. Calls to nearby addresses were excluded. If, as predicted by criminological theory, ambient crime risk “seeps out” across the adult cabaret neighborhood, excluding these calls will of course bias the secondary effect estimate in an unknown way. Drs. McCleary, Valerie Jenness, and James W. Meeker were retained by Palm Beach County to evaluate and, if necessary, replicate Dr. Danner’s study. Given the problematic use of address-specific 911 calls, they questioned Dr. Danner’s conclusion.35 Their replication used the same adult and non-adult cabaret sites. Instead of using 911 calls, however, they used crime incident reports; and instead of restricting the analyses to the specific addresses, they included all crime incidents that occurred within 1,100 feet of the adult and non-adult cabarets. The results of the replication are plotted in Figure 3.3. In terms of total crime, SOBs (in red, nine adult cabarets) and controls (in blue, seven non-adult cabarets) are both risky places. Moving toward an “average” site, whether SOB or control, victimization risk rises. Moving away, risk diminishes. With that said, compared to control sites, SOB sites are much riskier on average. How much riskier? At 500 feet (approximately one long city block), ambient risk at the SOB is four times greater. At 1,000 feet, the risk is substantially lower for all sites. But even at that distance, SOB sites are 3.5 times riskier than control sites.

34

P. 8, The Crime-related Secondary Effects of Adult Cabarets in Palm Beach County by Terry A. Danner, Ph.D. Report submitted in Palm Beach County v. Casablanca East, CA-0203813 AF, Circuit Court, Fifteenth Judicial Circuit, Palm Beach County, 2005. Dr. Danner is professor of criminal justice at St. Leo’s University. 35

Crime-Related Secondary Effects of Sexually-Oriented Businesses: Report to the County Attorney, Palm Beach County, Florida. Valerie Jenness, Ph.D., Richard McCleary, Ph.D., and James W. Meeker, J.D., Ph.D. August 15, 2007. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 27

Figure 3.3 - Palm Beach County Risk-Distance Functions, Total Crime

Although risk-distance plots have been widely used to document the ambient crime risks at “nuisance” sites, including SOBs,36 few have included statistical significance tests of the plots. Given the quantity and quality of data that were available in Palm Beach County, Drs. Jenness, McCleary and Meeker were able calculate confidence intervals for the risk-distance functions plotted in Figure 3.3. At the conventional 95 percent confidence level, both the SOB (red) and control (blue) functions are statistically significant as is their difference. Rejecting both null hypotheses then, the functions plotted in Figure 3.3 have the obvious interpretation. 3.4

Concluding Remarks

The three studies of adult cabarets reviewed here employ a range of designs. Two use 911 calls, one uses crime incident reports. Two compare SOB and control neighborhoods, one compares adult and non-adult cabarets. Yet all three studies support the conclusion that adult cabarets have large, statistically significant secondary effects. Nevertheless, there is a remarkable difference in how the studies interpret their findings. Whereas the Greensboro and Daytona Beach studies find large, statistically significant secondary effects, the authors give their findings the opposite interpretation. The secondary effect study summarized in Figure 3.2 was commissioned by the plaintiffs in Daytona Grand v. City of Daytona Beach.37 Drs. Fisher and Linz used a two-prong argument to challenge the City’s secondary effects evidence. First, they argued that the studies relied on by the City were

36

See, e.g., An Analysis of the Relationship between Adult Entertainment Establishments, Crime, and Housing Values. M. McPherson and G. Silloway. Minnesota Crime Prevention Center, Inc. October, 1980. 37

Daytona Grand, Inc. v. City of Daytona Beach, 410 F. Supp. 2d 1173 (M.D. Fla.

2006). R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 28

methodologically flawed. Second, they claimed local data showed that neighborhoods with adult businesses had the same number of 911 calls as other neighborhoods. To refute these arguments, the City cross-examined the experts. The trial court was unimpressed, however, and struck down those parts of the Daytona Beach ordinance that regulated nudity. The trial court’s decision in Daytona Grand provoked a mild panic among Florida governments. Two years later, however, the U.S. Eleventh Circuit reversed the trial court.38 The Eleventh Circuit decision reaffirmed the Renton standard in the most crucial respect: If the government’s interpretation of its secondary effects evidence is “reasonable,” there is no need to show that its interpretation is the only reasonable interpretation. The fact that plaintiffs can draw alternative conclusions from the evidence does not bar the government from “reaching other reasonable and different conclusions.” The Eleventh Circuit addressed three other relevant issues. First, the panel explicitly rejected the methodological arguments of Paul, Linz and Shafer.39 Second, the panel rejected the use of 911 calls to demonstrate the absence (but not the presence) of a secondary effect. Third, the panel noted, as we have, that several of the secondary effect estimates reported by Drs. Linz and Fisher were statistically significant. The experts are no doubt correct that factors other than the presence of adult theaters affect crime rates in Daytona Beach: crime is plainly caused by many factors. But that does little to undermine the City’s conclusion that adult theaters also affect crime rates, especially when the experts’ own analysis shows a statistically significant correlation between adult theaters and increased crime in half of the areas in the study.40 This observation by the Eleventh Circuit panel is consistent with Figure 3.2 above.

38

Daytona Grand, Inc. v. City of Daytona Beach, 490 F.3d 860 (11th Cir. 2007).

39

Paul, Linz, and Shafer (2001) argue that the government’s secondary effects evidence must satisfy Daubert admissibility criteria. 40

Daytona Grand, Inc. v. City of Daytona Beach (2007) at 47-48 R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 29

4.

The Opinions of Daniel Linz, Ph.D.

The factual predicate of the State of Ohio’s Rule 52 includes secondary effects studies conducted in other jurisdictions and times. In Dr. Linz’s opinion, these studies are not reliable evidence: THERE IS NO RELIABLE EVIDENCE FROM STUDIES CONDUCTED IN OTHER COMMUNITIES THAT ADULT BUSINESSES CREATE SUFFICIENT ADVERSE SECONDARY EFFECTS TO JUSTIFY THE TYPE OF RESTRICTIONS INTRODUCED BY THE STATE OF OHIO.41 The word “sufficient” and the phrase “the type of restrictions” seem to distinguish this opinion from the opinions expressed by Dr. Linz in other suits. In addressing this opinion, we assume that the word “sufficient” does not imply a logical property of the evidence; and that “the type of restrictions” is roughly synonymous with “regulations like Rule 52 that govern alcohol and adult entertainment.” With that understanding, we disagree with Dr. Linz’s opinion. The voluminous, diverse collection of secondary effects studies relied on by the State supports the consensus view that SOBs have large, significant secondary effects; and that these effects are aggravated by proximity to alcohol. In addition to its reliance on studies conducted in other jurisdictions and times, Dr. Linz faults the State of Ohio for ignoring secondary effects studies commissioned by SOB plaintiffs. Specifically: ABSENT FROM THE STATE OF OHIO'S JUSTIFICATION FOR RULE 52 ARE A LARGE NUMBER OF SCIENTIFICALLY SOUND STUDIES THAT SHOW NO ADVERSE EFFECTS OF ADULT BUSINESSES.42 Assuming that Dr. Linz refers to secondary effects studies commissioned by SOB plaintiffs and conducted by Dr. Linz and his colleagues, we disagree with this opinion as well. None of these studies is more “scientifically sound” (or reliable) than the government-sponsored studies relied on by the State. The results or findings reported in these studies are entirely consistent with the results or findings reported in government-sponsored studies. 4.1

Dr. Linz’s Methodological Authority

Both of Dr. Linz’s opinions rest on the authority of a 2001 article written by Drs. Paul and Linz, and Mr. Shafer. The methodological rules endorsed in the Paul-Linz-Shafer article are derived from the four criteria suggested by Justice Blackmun’s opinion in Daubert v. Merrell

41

Report of Daniel Linz, Ph.D., p. 1.

42

Report of Daniel Linz, Ph.D., p. 1. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 30

Dow. Dr. Linz paraphrases these criteria as: Studies of secondary effects should be examined in order to determine whether they answer the following questions which reflect important principles guaranteeing reliable information. Colloquially, these questions are: (1) “Compared to what?” (2) “Is this just a one-time fluke?” (3) “Is crime measured according to a reliable source?” and “did the government go looking for more crime to justify its legislation?” and (4) “Did the investigators talk only to people who would give them answers they wanted to hear?” These criteria are neither difficult nor cumbersome to apply, nor are they novel in nature – they are, however, essential features of any project attempting to gain reliable knowledge about secondary effects.”43 We disagree with Dr. Linz’s opinion both in general and in specific. In general, legislatures would indeed find it “cumbersome” to apply any set of strictly academic methodological rules – much less this set of rules – to the fact-finding process. Forcing a legislature to conduct its fact-finding in compliance with academic rules would bring representative democracy to a halt. The courts seem to have arrived at this same conclusion.44 Judged by purely academic standards, the specific methodological rules endorsed by PaulLinz-Shafer are not a necessary-sufficient methodological canon. The rules are not derived from the primary authorities on quasi-experimental design or statistics but, rather, from out-of-print sophomore-level psychology texts.45 Although the article is well known to SOB plaintiffs, it has had virtually no impact on the social science literature. Excluding citations by Dr. Linz and his colleagues, as of May, 15, 2007, the Paul-Linz-Shafer article had been cited only twice in peerreviewed journals. We know of no natural or social scientists who use or endorse the Paul-LinzShafer methodological canon. Although Dr. Linz claims that the U.S. Tenth Circuit Court of Appeals “specifically cited [the Paul-Linz-Shafer] article ... as a basis for overturning the evidence relied upon” by the plaintiff in Abilene Retail #30, Inc. v. Board of Commissioners of Dickinson County, Kansas, we do not find a citation to the Paul-Linz-Shafer article in the Court’s published decision.

43

Report of Daniel Linz, Ph.D., p. 3.

44

See, e.g., G.M. Enterprises. Inc. v. Town of St. Joseph, Wisconsin.

45

In discussing sampling, e.g., Paul-Linz-Shafer cite the authority of The Practice of Social Research 8th Ed. (E.R. Babbie, Wadsworth, 1998) and Approaches to Social Research, 2nd Ed. (R.A. Singleton et al., Oxford, 1993). Both texts were written explicitly for sophomorelevel social psychology courses. Both were out-of-print by 2001. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 31

4.2

The Paul-Linz-Shafer Validity Tests in Practice

Dr. Linz’s methodological critiques of government-sponsored secondary effects studies consist of identifying some weakness in a study’s design and then characterizing the weakness as a “fatal flaw.” This style of argument reflects a basic misunderstanding of quasi-experimental design.46 Since all secondary effect studies use quasi-experimental designs – including those conducted by Dr. Linz and his colleagues – all have potential shortcomings. Most shortcomings have trivially small impacts on the study’s findings or, else, bias the findings in favor of a null finding – in favor of the plaintiff’s argument. But a more serious problem with Dr. Linz’s methodological critiques of government-sponsored secondary effects studies is that the Linz-PaulShafer validity tests are arbitrary in practice. 4.2.1 The Paul-Linz-Shafer Rules are Arbitrary in Practice To be of any use, a methodological rule must have objective (numerical) benchmarks that characterize the relative validity of a study design. Lacking objective benchmarks, a rule cannot be used a priori to design a study with optimal validity or a posteriori to evaluate the validity of a study’s findings or results. The four Paul-Linz-Shafer rules lack objective benchmarks. The Paul-Linz-Shafer “compared-to-what” test illustrates this point. The gist of this test is that SOB and control areas must be “statistically adjusted” (or “matched”) to control for crime risks unrelated to the hypothesized secondary effects of SOBs. Assuming that the design uses control areas – and not all quasi-experimental designs use control areas – no methodological authority would disagree with this rule. As defined by Paul-Linz-Shafer, however, the rule is wholly arbitrary. In their critique of a 1979 Phoenix study, e.g., Paul-Linz-Shafer declare the SOB-control matching to be “unacceptable.”47 They reveal no objective numerical threshold value that separates “acceptable” SOB-control matching and “unacceptable” matching. The Paul-Linz-Shafer “one-time fluke” test suffers from the same lack of objectivity. The sense of the “one-time-fluke” test is that the secondary effect (e.g., ambient crime risk) should be estimated over a long enough period of time to ensure that a finding satisfies conventional levels of statistical confidence. Again, while no methodological authority would disagree with this rule, Paul-Linz-Shafer do not specify the length of time required to pass the “one-time-fluke” test. In the 1979 Phoenix study, Paul-Linz-Shafer decide that the duration of the study was not “sufficient” to pass the test.48 But they do not tell us how long a duration would be required to

46

“Quasi-experimental” refers to any design lacking random assignment. Our authorities on quasi-experimental design are Campbell and Stanley (1966), Cook and Campbell (1979), and Shadish, Cook, and Campbell (2002). 47

Paul, Linz, and Shafer (2001), p. 380

48

Paul, Linz, and Shafer (2001), p. 380 R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 32

pass the test. One year? Five years? In this particular instance, fortunately, the Central Limit Theory provides an objective benchmark for the “one-time fluke” test. Consideration of this objective benchmark highlights another weakness of the Paul-Linz-Shafer test: studies that fail the “one-time fluke” test invariably yield findings that favor the plaintiff. 4.2.2 Violations of the Paul-Linz-Shafer Tests Favor the Null Finding Since every (quasi-experimental) secondary effects study has several methodological shortcomings, a proper methodological critique must show, first, that the consequences of the shortcoming are not trivially small. If the consequences of a methodological shortcoming are trivially small or benign, the shortcoming can be ignored without cost to either side. Second, assuming that the consequences of a methodological shortcoming are non-trivial, a critique must show that the consequences create a bias against the plaintiff’s interests. If the consequences create a bias in favor the defendant’s interests, on the other hand, the plaintiff’s experts can ignore the methodological shortcoming. As it turns out, the methodological shortcomings associated with the four Paul-LinzShafer validity tests almost always bias a study’s results in favor of the null hypothesis. That is, they almost always work to make the secondary effects of SOBs appear smaller than they really are. In the case of the “one-shot fluke” test, e.g., the Central Limit Theorem implies that the statistical significance of a secondary effect is inversely proportional to the duration of the study. The shorter the study duration, in other words, the less significant the finding. When a study fails the “one-time fluke” test, the study is likely to conclude (erroneously) that there is no difference between SOB and control areas – i.e., no significant secondary effects. Likewise, studies that fail the Paul-Linz-Shafer “compared-to-what” test are biased in favor of the plaintiff’s interests and, hence, are likely to find no differences between SOB and control areas. A mathematical proof of the postulate that “matching” errors inflate the standard error of the SOB-control difference is straightforward.49 As the proportion of total variance due to matching error rises – which is to say, as the quality of the match weakens – secondary effect estimates approach zero while the standard error of the estimate inflates. As a result, the study is likely to find (erroneously) secondary effects that are substantively and statistically small. 4.2.3 The “Looking-for-More-Crime” Test The Paul-Linz-Shafer “looking-for-more-crime” test is not a validity test in the normal sense but, rather, a claim that an SOB-control difference in ambient risk can be a spurious artifact of proactive policing. In a 1977 Los Angeles study, Paul-Linz-Shafer note that, while SOB areas

49

See, e.g., Table 1 and supporting text in Cochrane, W.G. Matching in analytical studies. American Journal of Public Health, 1953, 43, 684-691. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 33

had higher crime rates than matched control areas, the difference could have been an artifact of heightened police presence in the SOB areas: Although the findings...suggested high levels of criminal activity [in SOB areas], any implication that this is connected to the presence of adult businesses is invalidated by the fact that the researchers admitted to “stepped up” surveillance within these areas. Put simply, the police most likely found greater amounts of crime in the adult establishment areas because they were trying harder to find it.50 Most criminologists would disagree. Although a “looking-for-more-crime” hypothesis might be plausible for some crimes, it is wholly implausible for most crimes. Vice crimes (prostitution, drugs, etc.) often arise through the activity of specialized undercover police who are indeed “looking-for-more-crime.” Crimes like robbery, vandalism, and auto theft, on the other hand, are ordinarily initiated – and hence, known to the police – through victim complaints. Higher levels of police presence in an area generally reduces the ambient risk of these crimes. The 1977 Los Angeles study reports higher levels of homicide, robbery, and assault in SOB areas. 4.2.4 The “Known Error Rate” Rule Though not one of their four explicit validity tests, Paul-Linz-Shafer dismiss the findings of “nearly all” government-sponsored studies on the grounds that the studies failed to report “an error rate—a test of the reliability of findings in science.”51 The sense of the “known error rate” rule is that studies should demonstrate the reliability of their findings by reporting error rates (or probabilities). Assuming that error rates are defined and appropriate – and this is not always the case – no methodological authority would disagree. No methodological authority would endorse the manner in which Paul-Linz-Shafer apply the rule, however. Our first objection is that Paul-Linz-Shafer apply the “known error rate” only to studies that report significant secondary effects. Studies that report null findings – i.e., that fail to find significant secondary effects – are given an implicit pass on the test. Since this issue is relatively technical and esoteric, we postpone its discussion until §6 below. Our second objection is that, as Paul-Linz-Shafer apply the rule, secondary effects evidence not collected with formal quasi-experimental methods is excluded. No matter how useful, informative, or valid the evidence might be, under the Paul-Linz-Shafer rule, secondary effects evidence collected with other methods is non-admissible. Among the excluded categories of secondary effects evidence is first-hand observations.

50

Paul, Linz, and Shafer (2001), p. 379

51

Paul, Linz, and Shafer (2001), p. 355. “Known error rate” is one of the four criteria suggested by Justice Blackmun in Daubert. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 34

Especially in cases where formal quasi-experimental studies are not feasible or economical, secondary effects can be documented with first-hand observations or archival records. Because it is difficult (if not impossible) to calculate an error rate for these data, the Paul-Linz-Shafer rule excludes them. Most scientific fields admit these data, however. The medical case studies and ethnographies published in prestigious scientific journals (New England Journal of Medicine, The Lancet, etc.), e.g., are collected without benefit of probability sampling, controls, or other formal quasi-experimental design structures. Although it is difficult (if not impossible) to calculate an error rate for observational data, this is not to say that there are no methodological conventions for assessing their reliability. On the contrary, scientists, clinicians, legislators, and judges use the same obvious criteria to assess the relative credibility (i.e., reliability) of first-hand observations. Other things equal, e.g., firsthand observations are more reliable than second-hand observations; expert observations are more reliable than non-expert observations; corroborated observations are more reliable than uncorroborated observations; and so forth. 4.3

Criminological Theory: Routine Activity Theory of Hotspots

In §1 above, we outlined the routine activity theory of hotspots. Ordinarily attributed to Sherman, Gartin, and Buerger (1980; Cohen and Felson, 1979; 1980), this widely used – and widely accepted – criminological theory explains why predatory offenders are attracted to SOB sites. Dr. Linz argues that the routine activity theory of hotspots does not apply to SOBs: While routine activities theory can be very useful in explaining how crime can be distributed spatially or geographically in a given area, McCleary’s application of the theory to sexually oriented businesses is unwarranted for several reasons: 1) The hotspot routine activities theory as applied to adult businesses assumes that the sexually-oriented businesses are, indeed, a [sic] crime hotspots without any empirical evidence that this true; 2) the theory focuses only on concentric circles in the area immediately surrounding the adult business and fails to take into account the myriad characteristics of the larger neighborhood in which the sexuallyoriented business is situated that may account for crime; and, 3) the theory, as stated, cannot be falsified.52 We disagree with each of these points. In fact, 1) the earliest research on hotspots includes SOBs among the most serious hotspots. 2) Virtually all spatial criminologists, including those cited by Dr. Linz, use concentric circular zones to model ambient crime risk. And 3) the routine activity theory of hotspots is eminently testable (or in Dr. Linz’s words, falsifiable).

52

Report of Daniel Linz, Ph.D., p. 6 R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 35

4.3.1 Evidence that SOBs are Hotspots In Dr. Linz’s opinion the prevailing routine activity theory of hotspots is predicated on the “not empirically sound” assumption that SOBs are crime hotspots. Dr. Linz’s opinion stands on three distinct (but related) claims: !

Legal - Dr. Linz argues that recent U.S. Circuit Court of Appeals decisions have held that there is no reason to assume that SOBs are hotspots.

!

Empirical - Dr. Linz argues that the Minneapolis study by Sherman et al. (1989) did not identify SOBs as hotspots.

!

Methodological - Dr. Linz argues that primary methodological authorities require that exploratory methods must identify SOBs as hotspots prior to conducting hypothesis tests.

We disagree not only with these three specific claims but, also, with the relevance of the claims to any question about the prevailing routine activity theory of hotspots. Legal - Under the rubric “The Courts do not assume adult businesses are hotspots of crime,” Dr. Linz argues that two recent Seventh Circuit Court of Appeals decisions support his opinion. We have read both decisions and disagree with Dr. Linz’s interpretation. Although we claim no legal expertise, in our reading, the Seventh Circuit panel decided both cases on legal grounds, not methodological or criminological grounds. Empirical - It is not difficult to find published research that identifies SOBs as crime hotspots. Dr. Linz’s contrary opinion is incorrect. In fairness, Dr. Linz’s opinion relies on the opinion of Dr. Jeffrey M. Cancino whose opinion rests on two published articles: [R]esearch conducted by Sherman et al. (1989) revealed that three and a half percent of the intersections and some addresses were overwhelming responsible for most crime. These addresses included department stores, 24-hour convenient [sic] stores, liquor stores, public housing projects, apartment buildings, and homeless shelters in Minneapolis. Weisburd et al. (2004) found that between four and five percent of street segments in the city accounted for 50 percent of crime incidents for each year over 14 years. As Cancino (2009) has pointed out, these studies did not identify adult businesses as hotspots. Nevertheless, McCleary assumes confirmation that adult businesses are hot-spots.53 Since Dr. Linz does not claim that he personally read these articles, we assume that he relied on Dr. Cancino’s reading. If Dr. Cancino read Sherman et al. (1989), however, he could not have read it thoroughly. As Dr. Linz claims, the list of Minneapolis hotspots identified by Sherman et

53

Report of Daniel Linz, Ph.D., p. 6 R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 36

al. included “department stores, 24-hour convenient [sic] stores, liquor stores, public housing projects, apartment buildings, and homeless shelters.” But the list also included SOBs. Three of the 25 hottest spots in Minneapolis were SOBs.54 Although Dr. Linz’s mistaken impression of the Minneapolis research conducted by Sherman et al. might be attributed to a misreading (though perhaps by Dr. Cancino), his mistaken impression of the research conducted by Weisburd et al. (2004) has no obvious explanation. Weisburd et al. had no interest in identifying hotspots and did not report a list of hotspots. Dr. Linz’s claim (though perhaps due to Dr. Cancino) that Weisburd et al. “did not identify adult businesses as hotspots,” he is not literally incorrect. But it is misleading. Weisburd et al. neither looked for nor found any hotspots. Methodological - Informed by his vicarious misreadings of Sherman et al. (1989) and Weisburd et al. (2004), Dr. Linz argues that, as a methodological convention, the routine activity theory of hotspots cannot be applied to SOBs until after an exploratory spatial analysis has identified a crime hotspot centered on an SOB. Dr. Linz’s authority for this opinion is a single sentence on the first page of Anselin et al. (2000): Special attention is given to some practical and accessible methods of exploratory data analysis that arguably should be the starting place of any empirical analyses of the relationship of place to crime.55 Dr. Linz interprets this sentence to mean that a confirmatory (i.e., hypothesis testing) analysis must necessarily be preceded by an exploratory (i.e., descriptive) analysis. Anselin et al. (2000) do not make this argument in this sentence, however, or anywhere else. Ironically, Dr. George E. Tita is a co-author of the Anselin et al. (2000) chapter. An affidavit by Dr. Tita is attached to this report. 4.3.2 Concentric Circles Dr. Linz’s second argument against the routine activity theory of hotspots concerns the use of concentric circular zones centered on the SOB. Some part of Dr. Linz’s argument stems from his conviction that secondary effects estimates must be adjusted for the effects of social disorganization variables; we address this part of Dr. Linz’s argument in §4.4 below. Another part of Dr. Linz’s argument is based on geometrical properties. Specifically, The epidemiology literature concerning sources of pollution and physical health is instructive in illustrating the weakness of the small area analysis (e.g., concentric zones) ... According to Lawson and Waller (1996: 473), when examining risk (in

54

See Sherman et al. (1989), Table 5, p. 43.

55

P. 213, Anselin et al. (2000). R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 37

this case crime) to a single source of pollution (in this case an adult business), and the assumption is that distance is a surrogate for risk, then a circular region (e.g., concentric zones) centered on the source provides the least sampling bias. However, if the source of pollution is not central to the area (Diggle, 1990) or population structure and other factors dictate the area (Lawson, 1995), then a circular zone has no advantage over more traditional square, rectangular, or polygon shapes (e.g., census tracts, block-groups, blocks). Use of polygons and tabular data are better able to account for many of the truncated factors that small area studies (e.g., concentric zones) tend to exclude (Lawson and Waller, 1996: 473).56 Dr. Linz argues that, under some circumstances, concentric circular zones offer no statistical advantage over polygonal zones. We agree. But under most circumstances – and especially for testing hotspot null hypotheses – concentric circular zones yield unbiased estimates of ambient crime risk that “other polygonal regions do not.”57 If circumstances dictate the use of polygonal regions, “one should make some effort to provide ‘similar’ sampling detail in all directions.”58 But if the regions are census tracts, this is practically impossible. Figure 4.3.2 - Lawson and Waller (1996, pp. 473-4) on Circular Regions

“[A] circular region centred on the source yields the least sampling bias”

“Square, rectangular, or other polygonal regions do not provide such unbiasedness”

If Dr. Linz has read the authority that he cites for his idiosyncratic opinions on circular sampling regions – and this is not clear – he has misunderstood the authorities, particularly “Lawson and Waller (1996: 473).” Figure 4.3.2 uses hypothetical map regions to illustrate the

56

Report of Daniel Linz, Ph.D., p. 8

57

Lawson and Waller (1996), p. 473

58

Lawson and Waller (1996), p. 474 R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 38

only point made by Lawson and Waller: Circular sampling regions yield unbiased estimates of crime risk; regular and irregular polygonal regions do not. In light of the well-known statistical properties of circular regions, criminologists have always used concentric circles to analyze spatial distributions of crime. This tradition extends back to the early “delinquency area” studies of Shaw and McKay (1942[1929]) – which Dr. Linz cites as the authority for his research – as well as the replications by Hayner (1933) and Schmid (1960). Ironically, studies that Dr. Linz has submitted in J.L. Spoons, particularly Paul and Linz (2002) and Linz et al. (2004), also use circular regions. 4.3.3 The Routine Activity Theory of Hotspots Is Testable In our description of the routine activity theory of hotspots, we noted that, due to potential embarrassment, SOB patrons are reluctant to report victimizations to the police. This makes the patrons attractive targets to predatory criminals. In Dr. Linz’s opinion, This makes [the] theory impossible to falsify. That is, if one were to attempt to empirically test the theory and find that there was substantial crime risk associated with the adult business, we would conclude that we found evidence for the theory. However, even if one were to attempt to empirically test the theory and found no evidence that the adult business was associated with crime, one could explain this finding by asserting that the only reason an effect was not found was because of the propensity of the victims to not report their victimization to the police. Therefore, the theory has no way of being proven false--any contrary evidence can always be “explained away” by the “reluctance to report” corollary to the theory.59 We disagree. Criminologists have known for at least fifty years that some witnesses and some victims are reluctant to report some crimes to the police. In the 2004 Charlotte study reviewed at §4.5.2 below, e.g., Dr. Linz and colleagues note the possibility of under-reporting: Perhaps victims of crime in areas surrounding adult clubs are not motivated to report crime incidents to the police. If this were the case, there may not be stable crime reporting across study and control sites. It could be that, compared to the control sites, more of the crime that occurs in the adult dance club zone goes unreported. It seems plausible that many of the victims of crime in these areas might not want to draw attention to themselves.60 The threat of under-reporting is not uniquely criminological. The threat of under-reporting arises whenever social scientists study sensitive issues. People are reluctant to discuss their purchases of “embarrassing” merchandise (Goodwin, 1992), e.g.; their sexual histories, especially STDs (Fenton et al., 2001); their tax returns, especially “cheating” (Grasmick, Bursik, and Cochrane,

59

Report of Daniel Linz, Ph.D., p. 8

60

Report of Daniel Linz, Ph.D., p. 8 R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 39

1991), and their smoking habits (Ebert and Fahy, 2007). In each of these fields, social scientists recognize the threat of under-reporting and take steps to control it. Criminologist control the threat either by design or by incorporating the errors into the statistical models. Since most studies find secondary effects, Dr. Linz’s complaint is moot. Under-reporting might be relevant when a study reports a null findings, of course, but we know of no study where a null finding was ruled out on under-reporting grounds. Since Dr. Linz cites no such study, we assume that his complaint is hypothetical. 4.4

Social Disorganization Theory Cannot Explain Secondary Effects

In Dr. Linz’s opinion, the criminological theory of social disorganization is more relevant to secondary effects phenomena than the prevailing theory described in §1 above.61 But in fact, social disorganization theory is wholly irrelevant to secondary effects phenomena. An authority cited by Dr. Linz describes the two related questions addressed by the theory: The theory of social disorganization speaks not only to the ability of a community to achieve common values (e.g., to defend itself against predatory victimization), but also to community processes that produce offenders.62 This is the consensus view. Criminologists use the theory of social disorganization to explain why well organized communities produce relatively few delinquents: The institutions that define a well organized community exert a higher degree of control on youth while focusing resources on predatory criminals who live outside the community. Since SOBs are located in commercial and/or industrial zones, social disorganization has nothing to say about secondary effects. 4.4.1 Social Disorganization Theory Social disorganization theory was developed in the 1920s by sociologists of the “Chicago School” particularly Clifford R. Shaw and Henry D. McKay. Shaw and McKay (1942 [1929]) posited that residential instability, ethnic homogeneity, and persistent poverty would produce socially disorganized communities which would, in turn, produce higher rates of delinquency. Subsequent elaborations of the theory focused on intervening structures (Bursik and Grasmick, 1993; Kornhauser, 1978; Simcha-Fagan and Schwartz, 1986; Taylor, Gottfredson, and Brower, 1984; Warner, 2007) and on the ability of community residents to organize and to use their

61

For his opinions on social disorganization theory, Dr. Linz cites Testing Theories of Crime and Deviance (Charis E. Kubrin, Thomas D. Stucky, and Marvin D. Krohn, Oxford University Press). We were unable to locate a book with this title. We assume that the correct citation is Researching Theories of Crime and Deviance by the same authors and publisher. The website for Oxford University Press describes this book as an undergraduate text. 62

Sampson and Groves (1989), pp. 785-6. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 40

“collective efficacy” to control crime and delinquency (Sampson and Groves, 1989; Sampson, Raudenbush, and Earls, 1997). Figure 4.4.1 shows a causal model of the social disorganization theory from Sampson and Groves (1989).63 According to Sampson and Groves (1989) five neighborhood characteristics (Low Economic Status, Ethnic Heterogeneity, Residential Mobility, Family Disruption, and Urbanization) cause three social disorganization factors (Sparse Local Friendship Networks, Unsupervised Teen Peer Groups, and Low Participation in Organizations) which, in turn, cause crime and delinquency. In addition to their indirect causal effects operating through the social disorganization factors, the five neighborhood characteristics have direct causal effects on crime and delinquency. Figure 4.4.1 - Social Disorganization Theory According to Sampson and Groves (1989)

“The theory of social disorganization speaks not only to the ability of a community to achieve common values (e.g., to defend itself against predatory victimization), but also to community processes that produce offenders.” Sampson and Groves (1989, pp. 785-6)”

The core mechanism of social disorganization theory is the breakdown of neighborhood social institutions. Socially disorganized neighborhoods have disproportionately more short-term (transient) residents. Because poorer residents are less able to move, poverty becomes endemic; families break down. Informal institutions, including friendship networks and social organizations (churches, PTAs, etc.), wither. The weakened neighborhood’s institutions no longer exert sufficient control on youths; rates of delinquency and crime rise.

We seldom see this core mechanism in SOB neighborhoods. Located in commercial and

63

Figure 4.4.1 is based on Figure 1 in Sampson and Groves (1989, p. 783), an authority cited by Dr. Linz. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 41

industrial zones, SOB neighborhoods have few if any residences, churches, schools, or families.64 Neither Shaw and McKay (1942 [1929]) nor Sampson and Groves (1989) would recognize the area around a typical SOB as a “neighborhood.” Dr. Linz’s description of social disorganization theory ignores this point: Social disorganization theory posits that the level of crime in a particular neighborhood is a function of the “social disorganization” in that neighborhood. A community is considered “disorganized” if there is a lack of social solidarity, social cohesion, and integration among the residents of that community (Kubrin, Stucky, & Krohn, 2009). The lack of these characteristics in a community leads to crime because informal social control is unable to be established without them, and it is this informal social control that helps to deter crime in a community (Kubrin et al, 2009). Perhaps most important to social disorganization’s application to the study of negative secondary effects is the insistence that the characteristics of the community in which the crime is occurring are important.65 Although Dr. Linz’s description of social disorganization theory correctly focuses on the actions of neighborhood’s families, residents, and informal institutions, he does not explain how the theory might work in neighborhood’s that have few residents, families, and institutions. Dr. Linz’s weak understanding of social disorganization theory is further evident in his treatment of alcohol-serving establishments as a social disorganization variable: I should mention, one of the elements that social disorganization theory points to as a possible mediator of crime, or I should say as a possible predictor of crime in a particular area, is the presence or absence of an alcohol-serving establishment.66 No primary authority on criminological theory, including those cited by Dr. Linz, would include “presence or absence of an alcohol-serving establishment” as a social disorganization variable for two reasons. First, the hypothetical residential neighborhoods at the theory’s core tend not have bars or taverns. Second, although the spatial correlation between alcohol-serving businesses and crime is widely known and well accepted (see, e.g., Roncek and Pravatiner, 1989; Roncek and Maier, 1991), the correlation is predicted by the routine activities theory of hotspots, not social

64

In Cleveland, Dayton, and Toledo, e.g., there were only 306 residential property sales within one-quarter mile of an SOB. 65

Report of Daniel Linz, Ph.D., p. 11.

66

Testimony of Daniel Linz, Ph.D., March 12, 2004, p. 111 R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 42

disorganization theory.67

67

Five of the top ten hotspots in Minneapolis were bars or taverns (Sherman et al., 1989, Table 5, p. 43). R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 43

Figure 4.4.2 - Social Disorganization Effects for Six Crimes

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 44

4.4.2 Evidence from Dr. Linz’s Studies To test a scientific theory, we (1) model the theory; (2) collect reliable, appropriate data; (3) estimate model parameters from the data; and (4) use the parameter estimates to test crucial hypotheses. Assuming that the model accurately represents the theory, and that the data are reliable and appropriate, the theory will predict specific results for the hypothesis tests. If the observed and predicted results are similar, the data corroborate the theory. If the observed and predicted outcomes are discrepant, on the other hand, the data reject the theory. Most rejected theories can be modified to account for the discrepancy; but in rare instances, the discrepancy may demand that the theory be categorically rejected. Figure 4.4.2 plots the standardized parameter estimates for the effects of nine social disorganization variables on six categories of crime in Greensboro.68 A few of the 54 observed effects corroborate Dr. Linz’s social disorganization theory to some extent. Neighborhoods with higher median ages are expected to have lower levels of crime, e.g. As predicted, Dr. Linz find a negative correlation for Property, Drug, Disorder, and All Other crimes (though not for Personal and Sex crimes). Most of the 54 observed effects do not corroborate the theory, however, and several actually contradict the theory. !

15-24 year-olds. Neighborhoods with younger populations should have higher crime rates. Dr. Linz finds negative effects on Personal, Property, Sex, Disorder, and All Other crimes, however, but curiously, no effect on Drug crimes.69

!

Family disruption. Neighborhoods with high numbers of female-headed, non-family, and unmarried households should have high crime. Dr. Linz finds no effect in eight cases, however, negative effects in four cases, and the expected positive correlation in only six of eighteen cases.

!

Residential instability. Neighborhoods with high rates of vacant, renteroccupied, and vacant owner-occupied housing should have high crime rates. Dr. Linz finds that residential instability has no effect on crime in fourteen of eighteen cases.

68

See §3.1 above. Figure 4.4.2 is based on Tables 14-19 in Evaluating Potential Secondary Effects of Adult Cabarets and Video/Bookstores in Greensboro: A Study of Calls for Service to the Police. Daniel Linz and Mike Yao, November 30th, 2003. 69

Dr. Linz’s results for the 15-24 year-old population lead to an absurd conclusion: More juveniles, less juvenile delinquency. We suspect that this result is an artifact of his misspecified regression model. Our suspicion is reinforced by a San Antonio study reviewed at §4.5.5 below. Using the same regression model in San Antonio, Enriquez, Cancino, and Varano (2007) also find less juvenile delinquency in neighborhoods with more juveniles. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 45

Inconsistent effects – positive for one crime, negative for another – pose a more serious challenge to the theory. To illustrate, female-headed households has a positive effect on two crimes (Drug and All Other), a null effect on three other crimes (Personal, Sex, and Disorder), and a positive effect on Property crime. Female-headed households appears to be uncorrelated with crime. Except for median age, none of Dr. Linz’s effects corroborates social disorganization theory. Most of the effects reject the theory. Used appropriately, social disorganization is a powerful criminological theory. Used inappropriately, on the other hand, it supports incorrect inferences. Judging from the perverse pattern of effects in Figure 4.4.2, critical readers should ask why Dr. Linz would want to rely on social disorganization theory to estimate the secondary effects of SOBs. The answer, put simply, is that by incorporating a large number of extraneous explanatory variables in his model, Dr. Linz biases his results in favor the null finding. We will return to this topic at §4.5.1 below when we discuss multiple regression. 4.5

Studies That Find No Secondary Effects

In addition to his opinion that the State relied on secondary effects studies that were “not reliable,” Dr. Linz faults the State for ignoring a set of “scientifically sound” studies, conducted by Dr. Linz and his colleagues, that find no secondary effects: ABSENT FROM THE STATE OF OHIO’S JUSTIFICATION FOR RULE 52 ARE A LARGE NUMBER OF SCIENTIFICALLY SOUND STUDIES THAT SHOW NO ADVERSE EFFECTS OF ADULT BUSINESSES.70 We disagree. Although we do not know whether the State ignored these studies as Dr. Linz claims, a fair review of those studies would not have shaken the State’s reliance on the broader secondary effects literature and its own local evidence. On the contrary, studies that find no secondary effects, especially those conducted by Dr. Linz and his colleagues, are consistent with the consensus finding of the broader literature and the State’s local evidence. Dr. Linz’s claim that his studies are “scientifically sound,” compared to other studies, rests on four design elements that distinguish them, in Dr. Linz’s opinion, from other studies: !

Dr. Linz’s studies use exploratory spatial analyses: “Rather than assuming that [SOBs] are crime hotspots without any empirical evidence ... we have tested this assumption.”71

!

Dr. Linz’s studies analyze crime rates in polygonal regions (e.g., census

70

Report of Daniel Linz, Ph.D., p. 1.

71

Report of Daniel Linz, Ph.D., p. 11. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 46

tracts): “Further, we focus not only on concentric circles ... but we also take into account the myriad characteristics of the larger neighborhood in which the [SOB] is situated that may account for crime.”72 !

Dr. Linz’s studies compare crime rates in SOB areas to analogous crime rates in control areas: “[I]t is necessary ... to engage in a comparative analysis in order to ascertain if crime is higher or lower at [SOBs] compared to other venues such as taverns and bars.”73

!

Dr. Linz’s studies use multiple regression to analyze SOB-control differences: “[I]t is necessary to employ as [sic] specific analytic technique, multivariate regression when studying secondary effects.”74

According to Dr. Linz, each of these four design elements enhances the validity of a secondary effects study; and each is found in his studies but are not found in the studies relied on by the State. We disagree with both claims. None of the four elements is a necessary-sufficient condition for validity. The four elements are found in many of the studies relied on by the State, moreover, and are not found in many of Dr. Linz’s studies. The first two design elements stem from Dr. Linz’s rejection of the criminological theory of hotspots and from a misreading of two primary authorities. !

Dr. Linz’s insistence that spatial null hypotheses75 cannot be tested until an exploratory analysis identifies SOBs as hotspots misinterprets Anselin et al. (2004), a primary authority on spatial criminology.76 Exploratory and confirmatory (i.e., hypothesis testing) methods are “apples and oranges,” appropriately used for very different purposes.

!

Dr. Linz’s insistence that existing polygonal map regions (e.g., census tracts) have better statistical properties than purposefully constructed circular regions is a vicarious misreading of Lawson and Waller (1996: 473), a primary authority on spatial sampling.

72

Report of Daniel Linz, Ph.D., p. 11.

73

Report of Daniel Linz, Ph.D., p. 12.

74

Report of Daniel Linz, Ph.D., p. 12.

75

The spatial null hypothesis holds that “The SOB site is not a crime hotspot:” To reject the spatial null hypothesis is to conclude that “The SOB site is a crime hotspot.” 76

See the affidavit from Dr. George Tita, a co-author of Anselin et al. (2004) which Dr. Linz cites as the authority for his opinion. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 47

Neither of these design elements makes a study more “scientifically sound.” On the contrary, according to the primary methodological authorities cited by Dr. Linz, either design element can threaten the validity of a study. Dr. Linz infers his second two distinguishing design elements – which he characterizes as “methodological dictums” – from the Seventh Circuit Court of Appeals decision in Annex Books v. Indianapolis. Regarding SOB-control contrasts, e.g., the Court said: Nor can we tell whether 41 arrests at one business over the course of 365 days is a large or a small number. How does it compare with arrests for drunkenness or public urination in or near taverns, which in Indianapolis can be open on Sunday and well after midnight? If there is more misconduct at a bar than at an adult emporium, how would that justify greater legal restrictions on the bookstore — much of whose stock in trade is constitutionally protected in a way that beer and liquor are not.77 No methodologist would quarrel with this dictum. Without an appropriate control, secondary effects estimates are uninterpretable. But the control must be appropriate. The costs in validity of an inappropriate control – a “crack house,” on one hand, or a church on the other – are dear. We return to this point in §4.5.2 below when we discuss Dr. Linz’s Charlotte study. Although we might agree with Dr. Linz’s opinions on SOB-control contrasts – at least in principle – Dr. Linz’s opinion on the methodological necessity of a “specific analytic technique, multivariate regression,” presumably derived from the Seventh Circuit’s Annex Books decision, is surely incorrect. The Court’s only reference to “multiple regression” is: One may doubt that Linz’s work is the last word; a multivariate regression would provide a better foundation than either a time series or a geographic cross-section. See Daniel L. Rubinfeld, Reference Guide on Multiple Regression, Reference Manual on Scientific Evidence (2nd ed.) (Federal Judicial Center 2000).78 We do not read this passage as a categorical dictum in favor of multiple regression. Since Dr. Linz’s Indianapolis study used “time series” and “geographic cross-section” designs, the Court’s “important methodological dictum” might have this more plausible interpretation: The design of Dr. Linz’s Indianapolis study compares unfavorably to alternative designs, including a multiple regression. The methodological authority cited by the Court reinforces this interpretation. According to Rubinfeld (2000),

77

Annex Books, Inc., et al. v. City of Indianapolis, Indiana

78

Annex Books, Inc., et al. v. City of Indianapolis, Indiana R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 48

Multiple regression analysis can be a source of valuable scientific testimony in litigation. However, when inappropriately used, regression analysis can confuse important issues while having little, if any, probative value.79 Would the Court issue a categorical dictum in one sentence and cite a contradictory authority in the next? That is implausible. Nevertheless, if Dr. Linz’s interpretation were correct – i.e., if the Court did endorse “a specific analytic technique, multivariate regression when studying secondary effects” – then the validity of Dr. Linz’s multiple regressions must be judged against Rubinfeld (2000), the multiple regression authority cited by the Court. By that standard, Dr. Linz’s use of multiple regression is inappropriate, leading to inferences that are wholly invalid. 4.5.1 Rubinfeld (2000) on Multiple Regression Quantitatively-oriented social scientists of our generation are familiar with Rubinfeld’s Econometric Models and Economic Forecasts, 2nd Ed. (1981, written with Robert S. Pindyck) and with his methodological reviews, including Rubinfeld (2000). In sum, Daniel L. Rubinfeld is an acknowledged authority on multiple regression. Contrary to Dr. Linz’s opinion, Rubinfeld (2000) does not recommend multiple regression in all situations but, rather, only in situations where the following assumptions are warranted: !

Strong Causal Theory. Multiple regression assumes a strong causal theory connecting a limited number of explanatory variables (causes) to a single response variable (effect).

!

No “Feedback.” Multiple regression assumes unidirectional causality. An explanatory variable is assumed to cause the response variable, in other words, but not vice versa.

!

No Omitted Explanatory Variables. Multiple regression assumes that no relevant, significant explanatory variable has been omitted from the model.

!

No Extraneous Explanatory Variables. Multiple regression assumes that no irrelevant, insignificant explanatory variable has been included in the model.

!

No “Fishing.” Multiple regression assumes that the theoretically specified model is unique.

After discussing these assumptions, we will compare Dr. Linz’s use of multiple regression

79

Rubinfeld (2000), p. 183 R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 49

analyses against the standards laid out in Pindyck and Rubinfeld. (1) Strong Causal Theory. Prior to any other design consideration, multiple regression requires a causal theory that connects a response (or dependent) variable to a limited number of explanatory (or independent) variables: Causality cannot be inferred by data analysis alone; rather, one must infer that a causal relationship exists on the basis of an underlying causal theory that explains the relationship between the two variables.80 As a practical matter, the theory must be strong enough to also specify the proper unit of analysis for cause and effect (e.g., individuals), the “best” response variable (i.e., the measured effect), and a limited number of explanatory variables (i.e., the measured causes). The prevailing theories in most fields are not strong enough to warrant multiple regression modeling. Rubinfeld (2000) cites the field of pay discrimination law as an exception. In a typical lawsuit, the ith employee’s “expected” salary is given by the multiple regression model: Salary i = á + â [Market Variables] i + ã [Race, Sex, etc.] i + g i Strong economic and legal theories hold that the ith employee’s salary is determined by a small number of legitimate labor market variables (employee’s education, experience, tenure, etc.) – but not by illegitimate variables (employee’s race, sex, etc.). Employees with identical scores on the labor market variables may not have identical salaries in the short run. The model’s random error term, g i, accounts for these small, unexplained salary differences. In the long run, or in expectation, the random errors vanish, resulting in Expected Salary = á* + â* + ã* where á*, â*, and ã* are least-squares estimates of the effects of the legitimate and illegitimate variables on employee salaries. Small values of ã* support the defendant’s case; large values support the plaintiff’s case.81 Multiple regression is well suited to pay discrimination suits because strong economic and legal theories specify the response variable (employee’s salary in dollars), the appropriate unit of analysis (employee), a limited number of legitimate explanatory variables (education, experience, tenure, etc.), and a limited number of illegitimate explanatory variables (race, sex, etc.). Because the underlying theory is so strong, defendant, plaintiff, and court can agree on the correct interpretation of the regression model and results. Where the prevailing theory is not so strong,

80

Rubinfeld (2000), p. 184

81

In this context, “small” means “not different than zero” or “not statistically significant at the conventional 95 percent confidence level.” R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 50

on the other hand, multiple regression may be inappropriate: Failure to develop the proper theory, failure to use variables that are appropriate for the multiple regression study, and failure to use alternative procedures when they are more appropriate can substantially bias the multiple regression results.82 Rubinfeld (2000) might have been describing secondary effects research. In J.L Spoons, e.g., a typical secondary effects suit, the Defendant and Plaintiff cannot agree on a response variable (911 calls, total crimes, sex crimes, etc.), the unit of analysis (500-foot circle, census tract, faceblocks, etc.), and most important of all, the explanatory variables. This reflects the fact that the Defendant and the Plaintiff have different secondary effects theories. Figure 4.5.1 - “Feedback” Loops in the Four-City Ohio Study

(2) No “Feedback.” The fact that SOB areas have more crime or lower property values than non-SOB areas has two interpretations. Either SOBs cause high crime and low property values; or else, high crime and low property values cause SOBs to move into an area. Courts have emphasized this interpretive dilemma: [T]he studies [on which the City relied] are simple cross-sectional analyses that leave causation up in the air. (In other words, they may show no more than that adult businesses prefer high-crime districts where rents are lower.)83 In some substantive fields, the interpretive dilemma can be resolved on theoretical grounds. In a salary discrimination suit, e.g., the defendant could not argue that the plaintiff’s salary caused his or her race or sex. In secondary effects research, the underlying theory is not strong enough to

82

Rubinfeld (1986), p. 1072 as cited in Rubinfeld (2000).

83

Annex Books, Inc., et al. v. City of Indianapolis, Indiana R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 51

resolve the dilemma. Figure 4.5.1 uses three explanatory variables from Dr. Linz’s Four-City Ohio study, reviewed at §4.5.4 below, to illustrate the interpretive dilemma.84 To estimate the effects of SOBs on crime in Cleveland, Columbus, Dayton, and Toledo, Dr. Linz regresses the number of crimes in a neighborhood on the neighborhood’s median home price, vacancy rate, and owner-occupancy rate. To interpret the regression, Dr. Linz must assume that a change in any or all of the explanatory variables will cause a change in crime – but not vice versa. In the multiple regression framework, the expert often assumes that changes in explanatory variables affect the dependent variable, but changes in the dependent variable do not affect the explanatory variables—that is, there is no feedback. In making this assumption, the expert draws the conclusion that a correlation between an explanatory variable and the dependent variable is due to the effect of the former on the latter and not vice versa. Were the assumption not valid, spurious correlation might cause the expert and the trier of fact to reach the wrong conclusion.85 But in fact, falling home prices, falling owner-occupancy rates, and rising vacancies are among the likely causal effects of rising crime rates. “Feedback” from the response variable (crime) to the explanatory variables violates the unidirectional causality assumption. Dr. Linz’s regression results are biased as a consequence and inferences based on the results are invalid. (3) No Omitted Explanatory Variables. In addition to specifying the “best” response variable and unit of analysis, the strong causal theory must specify the explanatory variables to be used in the multiple regression model. If the theory is inadequate, relevant explanatory variables may be omitted from the model with serious consequences. Failure to include a major explanatory variable that is correlated with the variable of interest in a regression model may cause an included variable to be credited with an effect that actually is caused by the excluded variable. In general, omitted variables that are correlated with the dependent variable reduce the probative value of the regression analysis. This may lead to inferences made from regression analyses that do not assist the trier of fact.86 The consequences of omitting a relevant explanatory variable depend on the relative importance of the variable in the causal system and magnitude and sign of its correlation with the variables

84

Our Figure 4.5.1 is based on Rubinfeld’s (2000, p. 196) Figure 1.

85

Rubinfeld (2000, p. 195)

86

Rubinfeld (1986), p. 1072. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 52

included in the model. Although the magnitude and nature of the bias cannot be predicted, the inferential consequences of an omitted variable are always serious in secondary effects research. To discover the variables that might have been omitted from Dr. Linz’s regression models, return to Figure 4.4.1. Sampson and Groves (1989) included measures of eight variables in their social disorganization model: socio-economic status, local friendship networks, participation in local organizations, supervision of local youth, residential stability, ethnic heterogeneity, family disruption, and urbanization. Dr. Linz’s regression models include measures of socio-economic status, residential stability, ethnic heterogeneity, and family disruption; but omit measures of local friendship networks, participation in local organizations, supervision of local youth, and urbanization. The inferential consequences of omitting these four variables is unknown. But it is unlikely that the consequences are benign. (4) No Extraneous Explanatory Variables. In specifying the exact set of explanatory variables to be included in the multiple regression model, the strong causal theory excludes all other variables. Incorporating extraneous, irrelevant variables in a multiple regression model biases significance tests. The major effect of the inclusion of irrelevant variables is to reduce the statistical significance of the regression results ... If the variable that is included, however, is inappropriate for theoretical reasons but does have an effect on the dependent variable, then the problem can be more serious. As will measures of goodness-offit, t-values may rise, and the trier of fact may reach inappropriate conclusions from the statistical analysis. The courts should be wary of the practice of including a large number of variables solely to overfit the data.87 If tests of statistical significance are not an issue, incorporating extraneous variables in the multiple regression model can have benign consequences. When the overfit model is used to test the significance of a secondary effect estimate, however, the consequences will be severe and will always favor the null hypothesis. Dr. Linz’s Four Ohio City study, reviewed at §4.5.4 below, illustrates the extent of the problem. Table 4.5.1 reports the parameter estimates from the four models available to us. Nearly 60 percent (31/52 = .596) of parameter estimates in the four models are not statistically significant at the 95 percent confidence level. With the statistically insignificant variables in the model, Dr. Linz adds a dummy variable, coded for the present or absence of SOBs in a Census Block Group, to the model. Since the incremental proportion of variance explained by the SOB dummy variable is not statistically significant, Dr. Linz concludes that SOBs in the four Ohio cities have no secondary effects. The statistically insignificant effect estimates in his model pose a threat to the statistical conclusion validity of this inference.

87

Rubinfeld (1986), p. 1077. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 53

Table 4.5.1 - Parameter Estimates in the Four Ohio Cities Study

Area Population Median age Nonwhite Female-headed households Married households Less than 9th grade College degree Median household income Median home price Percent below the poverty level Vacant housing units Owner-occupied housing units

Table 4a

Table 6b

Table 8c

Table 10d

–––– 2.485 –––– 2.985 –––– -3.463 –––– -2.191 -3.096 –––– 3.409 3.520 2.326

–––– –––– -2.631 –––– -3.613 –––– –––– –––– –––– –––– 2.690 6.256 ––––

–––– –––– -2.631 –––– -3.613 –––– –––– –––– –––– –––– 2.690 6.256 ––––

–––– –––– -2.663 –––– –––– –––– –––– 2.271 –––– -2.014 4.162 3.656 ––––

8:5

4:9

4:9

5:8

Significant : Non-significant a

Columbus sex crimes; b Dayton forcible rapes; c Dayton prostitution; d Dayton sex crimes

(5) No “Fishing.” Also known as “data torturing”88 and “data dredging,”89 “fishing” refers to the all too common practice of searching through a menu of multiple regression models to find one that supports a particular inference. The large number of possible explanatory or control variables available tempts statisticians to run a large number of regressions. It can be dangerous, however, if the expert uses this approach to search for the best coefficient that he or she can find.90 Allowed to choose from even a modest menu of potential multiple regression models, naive analysts can produce a null (or non-null) finding at will. In experimental research, design structures, including random assignment and placebo

88

Mills (1993, p. 1196): “If you torture your data long enough, they will tell you whatever you want to hear.” 89

Federal Judicial Center (2000, p. 358).

90

Rubinfeld (1986), p. 1073 R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 54

blinding, minimize the opportunity for “fishing.” In quasi-experimental research, where these design structures are not available, rigidly enforced conventions minimize the potential for “fishing.” In multiple regression modeling, where neither experimental design structures nor quasi-experimental conventions are available, the opportunity for “fishing” can only be minimized by strict adherence to a strong causal theory. Departures from the theoretical specification must be explained and if the explanation is unconvincing, critics must conclude that the results are an artifact of “fishing.” The unusually wide range of designs found in Dr. Linz’s studies suggest that “fishing” is a potent, plausible threat to the statistical conclusion validity of his results. 4.5.2 Adult Cabarets in Charlotte To estimate the secondary effects of SOBs in Charlotte, Dr. Linz and his colleagues compared ambient UCR Part I crime at the sites of twenty adult cabarets and three control businesses: a McDonald’s fast food restaurant, a Kentucky Fried Chicken take-out store, and an Exxon mini-mart gas station. Comparing ambient crime at the SOB and control sites, Dr. Linz and his colleagues found that [The] presence of an adult nightclub does not increase the number of crime incidents reported in localized areas surrounding the club (defined by circular areas with 500 and 1,000 feet radii) as compared to the number of crime incidents reported in comparable localized areas that do not contain such an adult business.91 Before commenting on the most controversial aspect of the Charlotte study’s design, note that it did not include a exploratory spatial analysis; that it measured crime in circular (vs. polygonal) regions centered on the SOB and control sites; and that it did not control for extraneous variables with multiple regression. In short, it had none of the design elements that, in Dr. Linz’s opinion, make a study “scientifically sound.” But the most memorable design element of the Charlotte study must be Dr. Linz’s choice of control sites.92 In their report to Dr. Linz, the North Carolina co-authors explicitly note the potential threat to validity posed by the choice of these particular control businesses. [I]t may be more appropriate to compare adult club sites with non-adult club sites so that one can determine whether the type of club activity affects the level of crime. This comparison may be implicit (if not explicit) in the minds of citizens

91

Report of Daniel Linz, Ph.D., p. 13

92

The Charlotte study was a collaboration between Drs. Linz and Paul in California and Drs. Land, Ezell, and Williams in North Carolina. After designing the study, Dr. Linz retained Drs. Land, Ezell, and Williams to collect and analyze the data and write a report. Drs. Linz and Paul added “introduction” and “methods” sections to the report and submitted it for publication. Descriptions of data collection and analysis written Drs. Linz and Paul are inaccurate. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 55

and justices when considering whether an adult club should be allowed to locate in a particular area. Methodologically, using basic service type businesses such as fast food restaurants as control sites may confound the comparisons being made in the research, even if they are located in areas equivalent to those in which adult dance clubs are located.93 The Seventh Circuit Annex Books panel would surely agree with Dr. Linz’s co-authors. Control is a fundamental principle of design. Implicit in this principle, however, is the idea that analog controls must be theoretically appropriate. Given the obvious inappropriateness of Dr. Linz’s controls, the Charlotte study says very little – if anything – about the secondary effects of adult cabarets. Had Dr. Linz chosen more appropriate control businesses, the Charlotte design would still have had another, more serious shortcoming. Calling attention to the shortcoming, Dr. Linz’s coauthors attribute it to a combination of not enough control sites and too little crime data: [D]ue to the limited number of clubs/controls and the relatively large number of parameters in comparison (especially the club-specific fixed-effects that were included to account for unobserved heterogeneity at the club/neighborhood level), we ... were not able to reliably and robustly estimate these models ... If we had access to either more years of crime data or more clubs/controls, these models would have been more feasible and more appropriate. Dr. Linz’s co-authors are describing the statistical power of the study design. Statistical power is a property of secondary effects null hypothesis tests. The SOB-control difference in Charlotte was too small to reject the null hypothesis. This does not imply that the null hypothesis is true, however. While the null hypothesis might be true, it is also possible that the study design lacked the statistical power required to detect a secondary effect. If the adult cabarets had had relatively large secondary effects, as Dr. Linz’s co-authors note, the effect would have passed unnoticed in the background “noise” of the study design. Statistical power is an endemic shortcoming of the secondary effects studies conducted by Dr. Linz and his colleagues. We discuss this relatively esoteric topic in detail in §8 below. For present purposes, we note that when a study is designed without an adequate level of statistical power, the failure to find a secondary effect says nothing about the hypothetical secondary effects of SOBs. The study is inconclusive. Most of Dr. Linz’s “scientifically sound” studies belong in this category.

93

Linz et al. (2004), p. 87. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 56

4.5.3 Adult Cabarets in Fort Wayne The 2001 Fort Wayne study by Drs. Linz and Paul compares a subset of crime incidents recorded within 1,000 feet of eight adult cabarets to the analogous incidents recorded in eight “matched” 1,000-foot control circles. Since Drs. Linz and Paul do not report their secondary effect estimates and standard errors, critics cannot calculate the statistical power of their null finding.94 Since the number of SOB and control sites in the Fort Wayne study is smaller than the number of sites in the Charlotte study, however, and since the number of crimes is smaller, we can assume that the Fort Wayne null finding is inconclusive. Given the weak statistical power of their design, the null finding is expected. One contribution to the low statistical power of the Fort Wayne design warrants special notice. Drs. Linz and Paul used Part I UCR crimes “cleared by arrest” to measure ambient crime risk. Since most Part I crimes are not “cleared by arrest,” their design excludes most of the crime in Fort Wayne. Critics can only wonder what their findings might have been if Drs. Linz and Paul had used a conventional ambient crime risk measure (e.g., all Part I crimes).95 Table 4.5.3 - “Fishing” as a Threat to Statistical Conclusion Validity 500-foot circles

911 calls

Census block groups

Fort Wayne

Cleared crimes Serious crimes

1,000-foot circles

Charlotte Greensboro

The use of this unconventional measure of ambient crime risk raises the specter of “fishing,” a methodological threat that Cook and Campbell discuss in the context of statistical conclusion validity. To illustrate this methodological threat, consider the designs used in Dr. Linz’s Fort Wayne, Charlotte, and Greensboro studies. Although the three studies were all designed and implemented by the same actors, the studies use three different measures of ambient crime risk (cleared crimes, serious crimes, and 911 calls) and three different areal units of analysis

94

Paul, Linz, and Shafer (2001, 355) fault the government-sponsored studies relied on by the State for not reporting these statistics: “[T]he most frequently cited studies ... fail to meet the basic assumptions necessary to calculate an error rate — a test of the reliability of findings.” 95

An area that has fewer “crimes cleared by arrest” may have fewer crimes; alternatively, the area’s crimes may be more difficult to “clear.” Criminologists use “cleared” crimes to study investigative processes but not to measure ambient crime risk. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 57

(1000-foot circles, 500-foot circles, and Census block groups). Crossing the design variations in these three studies yields nine potential quasi-experimental designs. Finally, we are surprised to learn that the Fort Wayne study “received a top award from the U.S. Department of Justice” and that it was “vetted by peer review for its methodological soundness.” An internet search reveals that the Fort Wayne study did indeed win a student paper award. But all (three) of the student papers submitted to the 5th Annual International Crime Mapping Research Conference won “Top Student Paper” awards. None of the student papers submitted to the 5th Annual International Crime Mapping Research Conference was peer reviewed. 4.5.4 Sex Crimes in Four Ohio Cities The Four Ohio Cities study was completed in 2005.96 An abridged version (omitting the statistical details reported in our Table 4.5.1 above) appeared in 2007.97 To estimate the effects of proximity to an SOB on sex crimes in Cleveland, Columbus, Dayton, and Toledo, Dr. Linz and his colleagues regress the number sexually-related 911 calls in a Census Block Group on demographic characteristics suggested by the routine activities and social disorganization theories. With these variables in the model, a dummy variable, coded for the presence of an SOB in the Census Block Group, is added to the model. The difference in R2 statistics between the two hierarchical models is used to test the significance of the secondary effect estimate. Since the R2 differences are small, Dr. Linz and his colleagues conclude that SOBs in the four Ohio cities have no secondary effects. In our opinion, however, this conclusion does not follow from the statistical results reported by Dr. Linz and his colleagues. First, the underlying regression model violates all five of Rubinfeld’s (2000) assumptions. Second, more important, the model’s response (dependent) variable is sex-related 911 calls. In a recent decision, a panel of the Eleventh Circuit Court ruled that statistical analyses of 911 calls cannot be used to show the absence of secondary effects Such crimes are often “victimless,” in the sense that all of those involved are willing participants, and, therefore, they rarely result in calls to 911 ... [A]n encounter between a prostitute and a “john” rarely leads to a 911 call.98

96

Linz, D. and M. Yao. Evaluating Potential Secondary Effects of Alcohol Serving Adult Cabarets: A Four City Study of Police Activity. September 28, 2005. 97

Linz, D., M. Yao, and S. Byrne. Testing supreme court assumptions in California v. la Rue: Is there justification for prohibiting sexually explicit messages in establishments that sell liquor? Communication Law Review, 2007, 7:23-53. 98

Flanigan’s Enterprises v. Fulton County, Eleventh Circuit Court of Appeals, No. 0817035, February 16th , 2010. The decision is quotation from the earlier Daytona Grand decision R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 58

Although criminologists use 911 calls for specific purposes, their use as a measure of ambient crime risk is limited. In light of Dr. Linz’s contrary opinions on 911 calls, we discuss the details of the disagreement at §6 below. 4.5.5 Crime in San Antonio A published study by Enriquez, Cancino, and Varano (2007) replicates the design used by Dr. Linz and his colleagues in secondary effects studies of SOBs in Greensboro (reviewed at §3.1 above), Daytona Beach (reviewed at §3.2 above) and the four Ohio cities (reviewed at §4.5.4 above). The results of the San Antonio replication raise the same questions. First, none of the regression models reported by Enriquez, Cancino, and Varano (2007) fit the data.99 Second, more important, the results are inconsistent with the theory of social disorganization. Nearly half of the variables that should affect crime, according to social disorganization theory, are statistically null; they have no effects on crime. Of the 32 statistically significant effects, 17 contradict the theory. Most notably, the estimated effects of the 15-29 year-old male population on crime are negative for five of seven crimes; neighborhoods with high populations of 15-29 year-old males, in other words, have low crime rates. Proportions of renter-occupied housing, divorced couples, Latinos, and Blacks also have negative effects on crime, contradicting the theory. Median household income is the only social disorganization variable whose effects are consistent with the theory. Oblivious to the theoretical perversity of their results, Enriquez, Cancino, and Varano (2007) conclude that the consensus view of the secondary effects literature is unwarranted: Instead, the results point to weak institutions, namely alcohol outlets and community characteristics associated with social disorganization theory as causes and correlates of crime.100 The reported results do not support this conclusion, however, but rather, strongly suggest that the theory of social disorganization is inappropriate for secondary effects phenomena. 4.6

Dr. Linz Does not Refute Evidence Presented in the State’s Hearing

In addition to relying on the broader secondary effects literature, the State relied on

which also ruled out the use of 911 calls to prove the absence of secondary effects. 99

The reported R2 statistics (interpreted as the proportions of variance explained by the models) range from 3.1 to 10 percent. R2 statistics as low as these preclude publication in a peerreviewed journal. The journal that published this article is edited by law students. It is not a peerreviewed journal. 100

Enriquez, Cancino, and Varano (2007), p. 34. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 59

evidence submitted in a public hearing. Dr. Linz does not refute this evidence. Testimony of David Raber. Mr. Raber encouraged the Liquor Control Commission to consider the broader circumstances in which violations occur so as not to overly penalize permit holders for the actions of rogue employees. Testimony of Bruce Taylor. Mr. Taylor described his experiences as a Cleveland attorney and how they prosecuted nudity at juice bars using indecent exposure statutes. When confronted with the argument that paying customers would not be offended by the nudity, they began prosecuting these cases under obscenity statutes instead. He discussed the Rule 52 that has been subject to recent amendments and how it has been challenged on the basis that it violates the First Amendment to the U.S. Constitution. The Cuyahoga County Court of Appeals argued that even greater protection of dancing exists within the Ohio Constitution. A number of studies have been conducted examining the harmful secondary effects of places of adult entertainment, resulting in a common body of knowledge. The Supreme Court held that cities need not conduct their own studies but instead can draw on the experiences of other places. Bars or juice bars with nude dancing experience increased levels of prostitution, drug trafficking, fights and brawls, even compared to other bars. Federal court in Cleveland ruled the 1998 version of the Rule 52 to be overly broad. In 2000, the U.S. Supreme Court, upholding an Erie ordinance, ruled that requiring dancers to wear pasties and G-strings does not violate their First Amendment rights. Mr. Taylor argued that the new proposed Rule 52 is constitutionally safer because (1) it is more specific and (2) the rule will not be subject to an overbreadth challenge. He also argued that places that sell alcohol give up some of their rights by obtaining a permit to do so, which gives the Commission the power through liquor regulation to deny permits to establishments that have certain types of entertainment. Mr. Taylor presented an affidavit from an undercover investigator in Monroe, Ohio demonstrating that topless lap dancers in the Bristol Show Club had physical contact with customers during lap dances in back rooms. The club took a cut of this extra money as well. Mr. Taylor submitted affidavits for search warrants issued by police in El Paso, Texas. Police determined there were strip clubs in the city hiring underage girls as dancers. Mr. Taylor submitted testimony given to the Ohio legislature before the Ohio Senate’s Judiciary Committee by (1) David Miller from Citizens for Community Values of Cincinnati; (2) Carol McKenzie of Memphis, Tennessee who works for an organization to help young girls get out of the strip club business; (3) Dave Sherman, manager of Deja Vu in the midwest region who testified about how girls are recruited for the business and what they are really told to do. He also submitted copies of testimony given in 2000 before the Michigan House Committee from Dave Sherman, Caroline McKenzie, and two former strippers. Mr. Taylor provided a copy of a study conducted in Adams County, Colorado, in 1991, discussing the high levels of crime associated with their strip club businesses and the effectiveness R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 60

of their ordinance in reducing these crimes. He also provided a copy of “Strip Club Testimony” by Kelly Holsopple, which chronicles her experience as a stripper. Mr. Taylor also discussed the 1993 federal district case of Bright Lights vs. the City of Newport, Kentucky, where a federal district judge upheld an ordinance requiring dancers to wear more than pasties, instead requiring that they wear clothing covering most of the lower half of women’s breasts. The judge justified the ordinance on the basis of the secondary effects caused by adult entertainment businesses. Mr. Taylor’s testimony ended with a description of how the exploitation of young girls produces a slippery slope from dancing at a strip club to prostitution. Finally, he argues that the proposed Rule 52 is constitutional; a First Amendment challenge to the proposed Rule 52 is not likely to be accepted. During questioning, Mr. Taylor argued that the Supreme Court of Ohio has stated that the Ohio Constitution does not grant greater protection to speech or expression than the First Amendment and most appeal courts in the State have sided with the Ohio Supreme Court (Cleveland being one exception). He further argued that the language in the proposed Rule 52 is not vague. Mr. Taylor was questioned regarding whether any studies have produced contrary findings to the particularly harmful effects found with the combination of nudity/partial nudity and alcohol. He responded that he was unaware of any studies that show the contrary. Testimony of Ed Duvall. Duvall, of the Ohio Department of Public Safety, presented two concerns. The first was a request to expand the definition of Section A-2 in the proposed Rule 52 to include not just the nipple but also the areola area. The second was a request that section B-6 be expanded to include language specifically referring to electronic benefit transfer (EBT) cards. During questioning, Mr. Duvall was asked whether it has been experience over the course of his many years in law enforcement that the combination of nudity/partial nudity and alcohol increases antisocial behavior. He stated that this was indeed his experience and that complaints were common in bars with impaired customers and nudity or lap dancing. He also reported receiving complaints from both local law enforcement and concerned citizens regarding prostitution, drug selling, and violent crime. He further confirmed that harmful activity outside of the establishment increases as well. He described how “street individuals” know that customers, out-of-town individuals, and tourists go to these establishments - that overcharge for everything with money to spend. Cars in the parking lots are subject to breaking and entering. Domestic situations also arise, e.g., where a boyfriend of a dancer will show up with a weapon.

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 61

4.7

Concluding Remarks

Dr. Linz’s critiques of the secondary effects literature relied on by the State are based on the idiosyncratic methodological rules endorsed by Paul, Linz, and Shafer (2001), not on primary methodological authorities. Some (but not all) of the secondary effects studies conducted by Dr. Linz and his colleagues find no secondary effects. Dr. Linz interprets these null findings to mean that SOBs have no secondary effects. In every case, however, Dr. Linz’s null findings fail to achieve the conventional level of statistical power required for that interpretation. Lacking the convention level of power, a null finding is an inconclusive result. We expand on this technical point in §7 below.

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 62

5.

Opinions of Lance Freeman, Ph.D.

Dr. Freeman uses “hedonic” pricing methods to estimate the secondary effects of SOBs on residential real estate values in Cleveland, Columbus, Dayton, and Toledo during 2001-2003. In simple terms, hedonic pricing methods assume that the value of a home can be inferred from its sales price; and that the sales price of a home is determined by its features (age, construction, living space, bathrooms, garage, etc.) and its by neighborhood (SES, schools, parks, etc.). Using a statistical model to control for the effects of these variables, homes near SOBs should sell for more or less the same price as any other home. If not, the price difference is interpreted as a secondary effect of SOBs. Hedonic methods do not guarantee a unique statistical model but, rather, allows the analyst to pick and choose model structures and assumptions from a menu. As a consequence, independently constructed models will always have small (or sometimes even large) differences. Parameter estimation introduces another menu of potential differences. When a model finally emerges from this process, it must be interpreted conditionally. Different model structures and assumptions and different estimation algorithms may support different interpretations. Although Dr. Freeman discusses some of the ambiguities in his statistical results, he does not consider whether other model specifications, assumptions, and estimation algorithms might have produced different results. Informed by the same hedonic methods, McCarthy, Renski and Linz (2001) chose very different structures and assumptions for their model and used a different algorithm to estimate its parameters. Dr. Freeman characterizes McCarthy, Renski, and Linz (2001) as “[p]robably the most sophisticated examination of the secondary effects thesis”101 but does not explain why he chose a different set of model structures with different assumptions and estimated the model parameters with a different algorithm. Ultimately, Dr. Freeman dismisses the ambiguities in his statistical results and finds “little evidence to conclude that the secondary effects thesis is a robust theory that explains how adult oriented businesses impact surrounding neighborhoods.”102 However, our experience with hedonic methods, especially in secondary effects research, suggests an alternative interpretation: Hedonic methods are poorly suited to modeling the secondary effects of SOBs. The two most problematic aspects of the statistical results support this interpretation. First, Dr. Freeman’s results have perverse implications.103 . Second, more important, his results cannot be reconciled with the results of other presumably valid methods.

101

Report of Lance Freeman, p. 23.

102

Report of Lance Freeman, p. 23.

103

We use “perverse” as economists use the term. See, e.g., Capozza and Van Order (1978). The term “unreasonable” might have a similar meaning for legislators.

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 63

Theoretical considerations also question the suitability of hedonic methods. Although hedonic methods have been used successfully to estimate the secondary effects of public amenities (parks, transportation hubs, etc.), public disamenities (landfills, power lines, etc.) and pollution point-sources, the secondary effects of these other land uses are qualitatively different. Whereas these other land uses might affect residential property, e.g., most SOBs are located in commercial and industrial zones. The secondary effects of SOBs cover relatively small areas and are realized through independent (though correlated) pathways: crime, blight, deterioration, etc. 5.1

Dr. Freeman’s Perverse Statistical Results104

Figure 5.1a illustrates one perverse implication of Dr. Freemans’s statistical model.105 The horizontal axis gives the distance in miles (measured in quarter-mile increments) to the nearest SOB. The vertical axis gives the relative change in the sales price of an “average” home after the differences in features and neighborhoods have been controlled. Relative prices in Dayton and Cleveland change little with distance to the nearest SOB. Relative prices in Columbus and Toledo, on the other hand, continue to change with distance to the nearest SOB. Figure 5.1a - Secondary Effects on Home Prices (Table A.6, Panel 1)

Ignoring other considerations for the moment, the price-distance plots in Figure 5.1a seem

104

We use “perverse” in the sense that economists use the term. See, e.g., Capozza and Van Order (1978). Many economists cite Merton’s (1936) “law of unintended consequences” as the source. 105

Report of Lance Freeman, Table A.6, Panel 1, p. 35. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 64

to support Dr. Freeman’s summary conclusion. If SOBs had the secondary effects that claimed, we might expect to see the same positive price-distance relationships in Cleveland, Columbus, and Dayton that we see in Toledo. Instead, we see null price-distance relationships in Cleveland and Dayton and a negative price-distance relationship in Columbus. The Achilles’ heel in this argument, of course, is that we expect to see no relationship whatsoever at the scale (quarter-mile increments) and distances (miles away from an SOB) that Dr. Freeman’s statistical model assumes. Finding these relationships hints at the model’s shortcomings. Figure 5.1b - Secondary Effects on Home Prices (Table A.6, Panel 4) One SOB opens within a quartermile of a home...

Several SOBs open within a quarter-mile of a home...

Dr. Freeman also reports the results of statistical models that were not estimated with the assumption of an unbounded price-distance relationship. The blue bars in Figure 5.1b plot the estimated effect on the sales price of an “average” home located within a quarter-mile of an SOB, controlling for differences in features and neighborhoods.106 One interpretation of these estimates is that, if an SOB opens within a quarter-mile of a home, the sales price of the home will rise by two percent in Cleveland and by seven percent in Toledo; but the sales price will fall by seven percent in Columbus. The red bars in Figure 5.1b plot the analogous effects for several SOBs. If more than one – i.e., several – SOBs open within a quarter-mile, the sales price of an “average” home will rise by 40 percent in Cleveland, 127 percent in Columbus, and seven percent in Toledo. These effect estimates have perverse implications for the economic behavior of buyers, sellers, and real estate agents. Those who are not well versed in economic theory might discount these estimates on the grounds that they are unreasonable. From either perspective, the statistical results raise serious questions about the suitability of hedonic methods for modeling secondary effects.

106

Report of Lance Freeman, Table A.6, Panel 4, p. 35. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 65

5.2

Surveys of Real Estate Appraisers

A related problem is that Dr. Freeman’s statistical results are not easily reconciled with the results of other presumably valid methods. Secondary effects studies in Los Angeles (1977), Indianapolis (1984), Austin (1986), and Garden Grove (1991) used the opinions of real estate professionals to document the effects of SOBs on nearby property. Allowing for substantive and methodological variations, these surveys support the conventional wisdom that SOBs adversely affect the value of nearby property. In their methodological critique of the secondary effects literature, Paul, Linz, and Shafer (2001) argue that, compared to the opinions of real estate professionals, analyses of actual home sales prices are preferred: Survey evidence is not comparable to, nor can it replace, the evidence supplied by objective comparisons of ... property values ... within areas containing adult entertainment businesses, with property values ... within areas containing no such businesses. Such a comparative analysis is the preferable social scientific means by which to establish a relationship between the presence of adult entertainment businesses and ... decreases in property values.107 We disagree. Paul, Linz, and Shafer (2001) offer no methodological authority for their opinion. “Survey evidence” is widely used across the social sciences because survey data are reasonably valid, relatively inexpensive to collect, and optimally flexible. This explains why surveys of real estate professionals are the preferred method for demonstrating secondary effects. We do not argue that an hedonic analysis (such as Dr. Freeman’s) and a survey of real estate professionals (such as those relied on by the State) must yield identical results. The results of hedonic and survey methods should be reconcilable, however, and that is not the case here.108 A recent survey by Cooper and Kelly (2008) has typical results and, furthermore, seems particularly relevant to J.L. Spoons. Cooper and Kelly asked a sample of 195 Texas appraisers certified as Members of the Appraisal Institute (MAIs) or Senior Residential Appraisers (SRAs) two questions: !

If located within 500 feet, how would [a Gentleman’s Club/Strip Club] potentially affect the market value of a single-family home?

!

If located within 500 feet, how would [a Gentleman’s Club/Strip Club]

107

Paul, Linz, and Shafer (2001), pp. 374-5

108

Brookshire et al. (1982) demonstrate that a properly designed hedonic study and an analogous sample survey will yield findings that are at least consistent. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 66

potentially affect the market value of a community shopping center? These questions were nested in a list of other land uses that included other SOB subclasses, businesses that serve alcohol along with non-adult entertainment, businesses that serve alcohol without live entertainment, businesses that sell (but do not serve) alcohol, typical NIMBY sites, and public amenities. Figure 5.2 reports several relevant contrasts. Figure 5.2 - Opinions of 196 Texas MAI or SRA Appraisers (Cooper and Kelly, 2008)

If located within 500 feet, how would the listed land use potentially affect the market value of a Single-Family Home?

If located within 500 feet, how would the listed land use potentially affect the market value of a Community Shopping Center?

Practically all of the appraisers (96.2 percent) believed that opening an SOB within 500 feet of a home would adversely affect the home’s value. Roughly the same proportion (92.4 percent) believed that opening a non-adult cabaret (i.e., a lounge that serves alcohol and offers non-adult live entertainment) within 500 feet of a home would have the same adverse effect. For residential property then, the appraisers saw only minor differences between adult and non-adult cabarets. For commercial property, in contrast, appraisers saw major differences. Whereas an overwhelming majority of the appraisers (79.6) believed that opening an SOB within 500 feet a shopping center would adversely affect the shopping center’s value, only a minority (41.9 percent) – albeit a large one – believed that opening a non-adult cabaret within 500 feet of a shopping center would have the same adverse effect. None of the appraisers believed that an SOB operating within 500 feet would affect the value of a single-family home or a community shopping center positively. Aside from their reasonableness, the results depicted in Figure 5.2 illustrate the salient advantages of survey methods. Due to practical constraints, Dr. Freeman was forced to estimate the secondary effects of a generic SOB (vs. the SOB subclass of interest) on residential property (vs. commercial and/or industrial property) within a quarter-mile (vs. 500 feet) of the SOB site. Survey methods do not have the same practical constraints. Surveys can focus on the specific R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 67

SOB subclass (e.g., “Gentleman’s Club/Strip Club”), real estate market (e.g., commercial), and effect zone (e.g., 500 feet) of interest in J.L. Spoons. 5.3

Hedonic Methods Are Unsuited to Secondary Effects Research

Although Rosen (1974) developed the hedonic theory to describe market phenomena, the application to real estate markets, along with a set of methodological rules, is generally attributed to Freeman (1974, 1979). Subsequently, hedonic methods have been used to estimate the effects of air and water pollution, proximity to NIMBY sites, parks, public transportation and a range of other amenities or disamenities on property values. Other than McCarthy, Renski, and Linz (2001), however, and Dr. Freeman’s work, hedonic methods have not been widely used to model the secondary effects of SOBs. Based on this admittedly small sample, every hedonic analysis finds no secondary effects. It is no surprise that SOB plaintiffs prefer hedonic methods. In contrast, every survey of real estate professionals finds secondary effects. Given the results of surveys in Los Angeles (1977), Indianapolis (1984), Austin (1986), and Garden Grove (1991), it seems unlikely that any survey would yield plaintiff-friendly results. This result does not explain why government defendants prefer surveys, however. In our opinion, the preference is based on practical considerations of flexibility, cost, and understandability. Understandability is a major consideration. Few readers have the backgrounds required to critically evaluate the statistical results of an hedonic analysis. In contrast, survey results are easily understood by general audiences. Practical considerations of cost are also important. The cost of a survey of real estate professionals will often be less than half the cost of an hedonic analysis. The most important practical advantage of a survey, however, is the relative flexibility of the method. Whereas a survey can be designed to focus on the variables that are relevant to, say, J.L. Spoons, the design of an hedonic model is constrained by circumstances. Accordingly, the hedonic model can be more relevant than circumstances allow. 5.3.1 SOBs vs. the Relevant SOB Subclass The survey by Cooper and Kelly (2008) was designed to compare the secondary effects of an SOB subclass (“Gentleman’s Club/Strip Club”) to the secondary effects of a control business (“Lounge with Live Entertainment”). The design feature optimizes the relevance of the survey results to J.L. Spoons. Generalizing Dr. Freeman’s results to the relevant SOB subclass would require unwarranted assumptions. Dr. Freeman does not explain why he chose to estimate the secondary effect aggregated across SOB subclasses. On purely statistical grounds, however, it is likely that estimating a disaggregated effect would have demanded a larger sample of home sales than was available. 5.3.2 Residential Properties v. Commercial-Industrial Properties The survey by Cooper and Kelly (2008) was designed to estimate the secondary effects of SOBs on the values of both residential (“Single-Family Home”) and commercial (“Community R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 68

Shopping Center”) properties. Since most SOBs are located in commercial-industrial zones, the secondary effects on commercial and industrial property may be more relevant than the effects on residential property. Dr. Freeman does not explain why commercial and industrial properties were excluded from the design. Considerations of statistical power (i.e., sample size) certainly played a role in this decision. The inherent difficulty of comparing commercial-industrial properties must have played a larger role, however. Hedonic methods make sense only when the price difference between two properties can be attributed to a small set of objectively measurable variables. Residential properties meet this requirement in many instances.109 If two homes have similar features (age, construction, living space, etc.) and are located similar neighborhoods (SES, schools, parks, etc.), any difference in their sales price is expected to fall in the range of “sampling error.” Commercial and industrial properties are not obviously comparable, however, at least not on a few objectively measurable variables. Because used car lots, coffee shops, and medical offices have few common features, hedonic models of price differences make little sense. Hedonic models of rent differences among commercial-industrial properties constitute a small proportion of the total literature, usually focusing on specialized market niches. If a model can be limited to small homogeneous subclass – say, offices – rent differences can be attributed to a few objectively measurable features (size, parking, security, etc,) and neighborhood variables (accessibility, theme, density, etc.). Without this focus, however, hedonic pricing methods are not well suited to the problem. 5.3.3 Omitted Variables Bias One of the most crucial assumptions of any regression model concerns the set of variables included in the model. If a variable has been omitted from the model, and if the omitted variable is correlated with an included variable, ordinary least-squares parameter estimates are biased and inconsistent. 110 Lacking these essential estimator properties, Dr. Freeman’s statistical results are uninterpretable. The most obvious omitted variables in Dr. Freeman’s models are neighborhood characteristics, particularly: !

School Quality - Other things equal, homes in “good” school districts will

109

Dr. Freeman’s models include seven distinct residential housing markets: apartments, condominiums, duplexes, mobile homes, town homes, triplex homes and single family homes. Homes in each of these markets have distinctive features, neighborhoods, sellers and buyers. Combining the distinct markets seems to violate the conventional assumption, due to Rosen (1974), of a homogeneous market. Comparing homes in distinct markets as if they were bought and sold in same market is misleading. 110

For proof, see any graduate econometrics text, e.g., Judge et al. (1985, p. 857). R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 69

command higher sales prices. !

Crime Rate - Other things equal, homes in “safe” neighborhoods will command higher prices.

!

Access to Public Amenities - Other things equal, homes within walking distance of parks and playgrounds will command higher prices.

And so on and so forth. Dr. Freeman relies on fixed-effects estimators to account for these (and all other) omitted neighborhood characteristics.111 In principle, fixed-effects estimators can account for omitted neighborhood characteristics if the characteristics vary discretely across neighborhoods. Dr. Freeman’s neighborhoods are not neighborhoods in the ordinary sense, however. Lacking data on “organic neighborhoods,” Dr. Freeman uses Census Tracts as a proxy. Unfortunately, the neighborhood characteristics that affect home prices vary within and “spill out” across Census Tracts. Figure 5.2.3 - Four Hypothetical Census Tracts and a Park

Hypothetical Census Tracts A, B, C, and D in Figure 5.2.3 illustrate the inadequacy of fixed-effects estimators. Tracts A and B have a (shared) Park while Tracts C and D do not. Fixed-effects estimators assume that each of the homes in Tracts A and B benefit equally from proximity to the Park but that none of the homes in Tracts C and D benefit at all. In fact, some of

111

Report of Lance Freeman, p. 14: “To control for the effect of neighborhood characteristics we used a census tract fixed effects approach.” R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 70

the homes in Tracts C and D are closer to the Park – and thus, derive a greater benefit from the Park – than some of the homes in Tracts A and B. In this hypothetical (albeit realistic) example, the fixed-effects estimator does not solve the statistical problems raised by the omitted variable, but rather, aggravates the problem. In a curious footnote, Dr. Freeman alludes to replicating his models without fixed-effects estimators: Although the results are not included in this report we also estimated models that did not include neighborhood fixed effects. One may argue that inclusion of neighborhood controls with distance from adult business measures could lead to an overidentified model which would bias our results towards not finding a significant relationship between distance from an adult business and property value. However, the results from the models that exclude the neighborhood controls do not substantively change the general conclusions of this report.112 The fact that fixed-effect and ordinary least-squares estimators yield indistinguishable results would suggest to many economists that the basic model is improperly specified. We concur in that opinion. In this particular instance and in general, the straightforward solution to the omitted variables problem – measuring and incorporating all of the relevant variables in the model – is impractical. 5.3.4 Spatial Dependence Spatial dependence poses another obstacle to the hedonic method. The sales price of a home is ordinarily affected by the condition of nearby properties and by the sales price of similar homes in the neighborhood. This implies spatial dependence or spatial autocorrelation among home sales. Dr. Freeman notes this problem: [W]e argue that real property may indeed be affected by surrounding neighborhood attributes including the value of adjacent properties, which is a violation of [model error terms] being independently distributed. So to correct for this autocorrelation ... as well as any violation of the assumption that [model error terms are] identically distributed, we estimate our models using the Huber/White/Sandwich corrected standard errors.113 Huber-White-Sandwich estimators adjust standard errors of estimation for the effects of spatial dependence, particularly for violations of the model’s homoskedasticity assumption. HuberWhite-Sandwich estimators do not correct for estimator bias, however. Freedman’s (2006, p. 1) statement on this point is typical.

112

Report of Lance Freeman, footnote #1, p. 16.

113

Report of Lance Freeman, Ph.D., p. 16. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 71

The “Huber Sandwich Estimator” can be used to estimate the variance of the [maximum likelihood estimator] when the underlying model is incorrect. If the model is nearly correct, so are the usual standard errors, and robustification is unlikely to help much. On the other hand, if the model is seriously in error, the sandwich may help on the variance side, but the parameters being estimated by the [maximum likelihood estimator] are likely to be meaningless – except perhaps as descriptive statistics. If Dr. Freeman’s Huber-White-Sandwich estimators are biased – and that is a certainty – knowing that the standard errors of the biased estimators are consistent is no consolation. The hedonic pricing literature has used three basic modeling approaches to control the estimator bias due to spatial dependence: spatial lag models (Armstrong and Rodriguez, 2006; Kim, Phipps and Anselin, 2003); spatial error models (Boxall, Chan and McMillan, 2005); and spatial Durbin models (Brasington and Hite, 2005; Li and Saphores, 2009). Each approach has unique strengths and weaknesses. McCarthy, Renski, and Linz (2001) use a spatial lag models estimated in a system of “seemingly unrelated regressions.” 5.4

Concluding Remarks

Although hedonic methods have been widely used to estimate the effects of public amenities, public disamenities, and pollution sources on property values, they are a novel method in this field. Ignoring the question of whether Dr. Freeman’s secondary effect estimates are credible, they are irrelevant to J.L. Spoons: !

Dr. Freeman’s estimates apply to the broad class of SOBs. There are no estimates for the SOB subclass of interest in J.L. Spoons.

!

Dr. Freeman’s estimates apply to residential properties only. There are no estimates for the commercial or industrial properties that are likely to be involved in J.L. Spoons.

!

Dr. Freeman’s estimates assume that secondary effects can be detected at a distance of one quarter-mile. The secondary effects in J.L. Spoons affect a smaller area and are overlooked by Dr. Freeman’s hedonic model.

Ignoring the question of relevance, theoretical considerations suggest that hedonic pricing methods are unsuited to the secondary effects of SOBs. The perverse implication of Dr. Freeman’s statistical results reinforce this view. Forced to choose between Dr. Freeman’s results (Figure 5.1a-b above) and the results of a traditional survey (Figure 5.2 above), reasonable people would reject Dr. Freeman’s results and would accept the survey results.

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 72

6.

911 Calls as a Measure of Ambient Crime Risk

Secondary effects studies have used three measures of ambient crime risk: crime incident reports (e.g., UCRs), 911 calls, and to a lesser extent, victimization surveys. Given the “rare event” property of crime, victimization surveys are an impractical means of demonstrating the presence (or absence) of secondary effects. Crime incident reports are a more reliable measure, compared to 911 calls, and preferred by criminologists. For the same reason, plaintiffs’ experts prefer to use 911 calls. Dr. Linz’s Greensboro and Daytona Beach studies, reviewed at §3.1-2 above, and his Four Ohio City study, reviewed at §4.5.4 above, are typical. All large police agencies record 911 calls for planning and budgeting purposes.114 In a pinch, 911 databases can generate “quick and dirty” snapshots of crime problems. In the long run, however, police agencies use crime incident reports to measure ambient crime risk, not 911 calls. Criminologists share this view. The reasons why plaintiffs’ experts prefer 911 calls would be obvious to most criminologists. Asked to defend their preference, plaintiffs’ experts argue – incorrectly – that criminologists routinely use 911 calls to measure ambient crime risk. Many criminologists have employed citizens’ telephone calls-for-service (CFSs) to police dispatch centers to measure crime at the address (Sherman, Garten, & Burger, 1989), neighborhood (Bursik, Grasmick & Chamlin, 1990; Warner & Pierce, 1993), and city (Bursik & Grasmick, 1993) levels. According to its proponents, the CFSs measure offers a more valid description of aggregate levels of crime than either police records collated in the FBI’s Uniform Crime Reports (UCR) or victimization data collected in the National Crime Survey.115 This is a half-truth at best. Shortly after the advent of computerized 911 systems, criminologists did indeed experiment with 911 calls, sometimes even using them as surrogate measures of crime risk. The research cited by these experts was conducted during this early period. However, the results of this early research led to the consensus view that 911 calls are not the best – or even a good – measure of ambient crime risk. Today, few criminologists study 911 calls for any reason; but no criminologists study 911 calls to learn about ambient crime risk. The published literature review summarized in Table 6 supports this point. During a recent five-year period, four general criminology journals published 705 items. Most of the items were either non-empirical (essays, reviews, etc.) or else, analyzed phenomena other than crime (police behavior, sentencing decisions, etc.). Of the 254 articles that analyzed a crime statistic, 134 (52.8 percent) analyzed UCRs; 119 (46.8 percent) analyzed victim or offender surveys. Only five items (1.9 percent) analyzed 911 calls. Of these five, only one

114

These legitimate uses of 911 calls are discussed in most undergraduate policing texts. See, e.g., Roberg, Crank and Kuykendall (1999). 115

Linz, Paul, and Yao (2006), p. 191. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 73

used 911 calls as a crime risk measure. Table 6 - Crime Statistics in Criminology Journals, 2000-2004

Criminology Justice Quarterly J of Quantitative Criminology J of Criminal Justice

Total Items

Crime Stats

UCRs

Survey

911 Calls

193 152 95 265

52 48 47 107

37 23 30 44

16 23 17 63

0 2 0 3

(705)

(254)

(134)

(119)

(5)

The 1997 Fulton County study and the 2001 replication are unique among governmentsponsored secondary effects studies. Otherwise, the use of 911 calls to measure ambient crime risk is as rare as hen’s teeth. More important, of course, recent case law supports the views of criminologists and governments.116 In short, analyses of 911 calls are not sufficient to meet the standards required under Alameda Books to cast doubt on the secondary effects evidence relied on by the government to support an ordinance. 6.1

Why Plaintiffs’ Experts Prefer 911 Calls

Plaintiffs’ experts prefer to use 911 calls because, first, 911 calls are relatively “noisy.” This statistical property obscures the difference between high- and low-risk public safety hazards. Second, compared to crime incident reports, 911 calls provide a biased picture of the vice crime incidents that weigh so heavily in the secondary effects of SOBs. Third, compared to crime incident reports again, 911 calls are easily manipulated. We cover these three points in order. 6.1.1 911 Calls Are a “Noisy” Measure of Crime Risk “Sound” is a useful analog to “ambient crime risk.” Whereas the sound of a party is easily detected by next-door neighbors, neighbors who live two blocks away must listen carefully to separate the voices, music, and other party sounds from the sounds of wind, traffic, and other natural background “noise.” Four blocks away, party sounds are overwhelmed by background 116

The Eleventh Circuit outlined the limitations of 911 calls in Daytona Grand and pointed out (fn. 33) that three other Circuits had rejected attempts by plaintiffs to use 911 calls to cast direct doubt on an ordinance: Gammoh v. City of La Habra, 395 F.3d 1114, 1126-27 (9th Cir. 2005), G.M. Enter., Inc., 350 F.3d 631, 639 (7th Cir. 2003), and SOB, Inc., 317 F.3d 856, 863 & n.2 (8th Cir. 2003).

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 74

“noise.” Neighbors who live four blocks away would need exotic sound detection equipment to separate the meaningful sounds of a party from meaningless background “noise.” The analogous problem arises when we try to attribute crimes to an SOB point-source. Crime incidents occur sporadically in any large neighborhood with no apparent causal source. In purely statistical terms, these sporadic incidents are background “noise.” When a crime occurs within, say, 500 feet of the SOB, we can be reasonably confident that it is a secondary effect of the SOB, not a sporadic incident. But as one moves away from the SOB, the relative frequency of sporadic crimes rises and confident attribution becomes more difficult. In either case, of course, the statistical task is more difficult problem when the criterion signal – party sounds or crime incidents – is not easily distinguished from background noise. To illustrate, suppose that the party music were muted and atonal or that the guests spoke in foreign languages. The task of separating the sounds of such a party from typical background noise might be relatively difficult. The analogous principle applies to ambient crime. Suppose that burglary 911 calls are used to measure the burglary rate. Since most of these 911 calls are triggered by malfunctioning alarms, it is difficult to distinguish real burglaries from false alarms.117 Since most addresses are not protected by burglar alarms, moreover, it is difficult to distinguish high- and low-risk addresses. Figure 6.1.1 - Correlation of Crime Incidents and 911 Calls

As a measure of ambient crime risk, the salient shortcoming of 911 calls is that they are too “noisy.” Figure 6.1.1 depicts this fact. Since 911 calls outnumber crime incidents by a great factor in every jurisdiction, the area accorded to 911 calls (in red) is larger than the area accorded to crime incidents (in blue). The correlation between the two measures is proportional to their overlapping area. The larger the area of overlap, relative to the combined total area, the stronger

117

In their San Diego study, Drs. Linz, Paul, and Yao (2006) counted 147,127 burglary 911 calls. Of these, 74.8 percent were alarm-initiated, 99.1 percent of which turned out to be false.

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 75

the correlation. In this case, the correlation is relatively small or weak. The non-overlapping areas in Figure 6.2.1 fall into two categories. The first consists of 911 calls that have nothing to do with crime. Examples include duplicated or unfounded calls; calls that have no apparent basis; and calls that are precipitated by false alarms. 118 The second consists of crimes that circumvent the 911 system and, thus, leave no record. Examples include crimes that the police discover through routine or proactive patrolling and crimes that the police discover through specialized unit activity, especially “victimless” vice crimes, particularly drugs and prostitution. This raises the issue of statistical biases in 911 calls. 6.1.2 911 Calls Are a Biased Measure of Some Crime Risks Whereas “noise” affects all sub-samples equally, tending to obscure differences between subsamples, “bias” affects subsamples differently. There are two known biases in 911 calls. The first affects the reporting of the vice crimes that weigh heavily in secondary effects. To illustrate the magnitude of this bias, Drs. Linz and Paul analyzed 21,132,503 New York City 911 calls. Of these, 41were initiated by prostitution incidents.119 These numbers speak for themselves. A second bias affects the address-specific (“hotspot”) analyses used in the 1997 Fulton County Police Department reports. Address-specific comparisons assume that the address recorded on a 911 record is the address where the precipitating crime occurred. The address on a 911 record instructs responding patrol units where they go to “see the man,” however, and this is often not the address of the precipitating incident. If X calls 911 to report a disturbance at Y’s house, e.g., the responding patrol unit will be asked to “see the man” at X’s address. Of course, significantly many 911 calls are attributed to places (“Third and Main”), not addresses (“321 East Main”). Business proprietors who are aware of this convention are able to manipulate the 911 system. 6.1.3 911 Calls Are Easily Manipulated One reason why SOB plaintiffs might prefer 911 calls is that, because relatively few “victimless” crimes (drugs, prostitution, etc.) come in through 911 channels, 911 calls understate the incidence of these crimes by a large factor. Another reason is that 911 calls can be used to mask an address-specific public safety hazard. This last problem merits special comment. If a business is familiar with the coding conventions, 911 records can be manipulated to make the

118

More than 80 percent of the 607,903 911 calls analyzed by Linz, Paul, and Yao (2006) were cancelled, duplicated, unfounded, disposed of without report, or had some other non-crime disposition. 119

Linz. D. and B. Paul. Measuring the Secondary Effects of 60/40 Businesses in New York City: An Analysis of Calls for Service to the Police. April 14th, 2005.

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 76

business look more or less in need of police service. To build a case for more police service, the proprietor can complain to the police about problems that might otherwise be handled informally. Or alternatively, to mask a public safety hazard, the proprietor can handle problems informally, thereby creating fewer 911 records and making the business seem safer than it actually is. The testimony of a private investigator retained by Sandy Springs, a city in Fulton County, documents the manipulation 911 records by the management of local SOBs. After witnessing the beating of an unruly, inebriated patron, the undercover investigator asked an employee why the management had not called the police: And she said: Well, the police are never called. The police don’t come around here ... It’s obvious that the clubs are attempting to hide crime statistics. If they don't call the police or call for service, then there’s no report made and there’s no evidence that there’s been a problem at the clubs ... We did look at the crime statistics for Fulton County and in that area, and there are very few comp1aints. I witnessed two fights that night, the 25th, that I was in there. While we were in at other times, other fights were observed. No police calls were shown on those particular nights. The club didn’t call. Other patrons didn’t call. The person who got beat up didn’t call. Again, it just seems like they’re handling this inside the club for there not to be statistics out there.120 With obvious exceptions, manipulations of this sort are legal, strictly speaking. At the extreme, however, manipulating the 911 record-keeping system crosses the line. In a recent Manatee County case, for example, an SOB bribed at least two deputies to illegally circumvent and/or to falsify 911 records. Another Manatee deputy, Daniel E. Martin, 35, told sheriff's investigators that one of the Cleopatra's door girls had his cell phone and would call him personally to quell customer disturbances ... Former Manatee deputy Joshua R. Fleischer, 25, who resigned this month, told a detective that whenever he was dispatched to Cleopatra's for a disturbance he listed the address as the “3900” block of U.S. 41 – deliberately misidentifying the actual address in the 3800 block. Fleischer, according to the detective, did not want his reports associated with the club.121 The investigation into this scandal has spread to surrounding counties. The relevant point, for our purposes, is that business proprietors who are familiar the 911 geo-coding conventions can (and in Manatee County, at least, do) attempt to manipulate the system.

120

Mr. Guy Watkins, Business Consulting and Investigations, Inc., pp. 91-2, December 20th, 2005 transcript of the Sandy Springs City Council Meeting . 121

StripClub News, September 22nd , 2006, “Investigation tied to strip club leads to resignations and charges.” R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 77

6.2

Concluding Remarks

As a measure of ambient crime risk, 911 calls are less reliable than crime incident reports; are biased against the crimes most closely associated with SOBs; and are easily manipulated to mask address-specific public safety hazards. None of these statistical shortcomings is “fatal.” On the contrary, given adequate time, resources, and data, each of these threats to validity can be controlled – albeit at great expense. This raises the question of why anyone might want to use 911 calls to measure ambient crime risk. One answer is that each of the threats favors the null; secondary effects are always “smaller” when 911 are used as the risk measure. Experts retained by SOB plaintiffs offer other answers. First, compared to crime incident reports, 911 calls are a “better” measure of ambient crime risk. Second, 911 calls are a widely used measure of risk, not only in the secondary effects literature relied on by legislatures but, also, in basic criminological research. Both arguments are incorrect, of course. In purely statistical terms, 911 calls are not an acceptable measure of ambient crime risk. Nor are 911 calls widely used in secondary effects studies or criminological research. The published literature review summarized in Table 6 depicts the use of 911 calls in criminological research accurately. Few criminologists study 911 calls for any reason; but no criminologists study 911 calls to learn about ambient crime risk. An analogous statement can be made for the secondary effects literature. Relatively few government-sponsored studies have used 911 calls any purpose whatsoever. Other than the 1997 Fulton County study, studies based on 911 calls are as rare as hen’s teeth. Studies commissioned by SOB plaintiffs are another matter. Since 2001, 911 calls have become the preferred measure of ambient risk in plaintiff-sponsored studies.

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 78

7.

Negativa Non Sunt Probanda122

To cast doubt on a government’s secondary effects evidence, plaintiffs’ experts conduct a study that finds no secondary effects – a null finding, in other words. Dr. Linz’s study of Four Ohio Cities, reviewed at §4.5.4 above, is typical. Although plaintiffs’ experts interpret their null findings to mean that SOBs have no secondary effects, in fact, their null findings have a simpler interpretation: the design of their studies lacked the level of statistical power required to detect a secondary effect. A mundane analogy illustrates the interpretational dilemma by a scientific null finding. If you cannot find your car keys, you might be tempted to conclude that your car keys do not exist. Although that is entirely possible, it is also possible (and more likely) that you did not look hard enough for your car keys or that you were looking in the wrong place. By analogy again, if a “quick and dirty” study finds no statistically significant secondary effects – a null finding – it is entirely possible that no secondary effects exist. But it is also possible (and more likely) that the study was “too quick” or “too dirty.” Analyzing San Diego 911 calls, Drs. Linz, Paul, and Yao (2006) found that SOB areas had 15.7 percent more calls than control areas. This is a substantively large effect. Indeed, the budgetary implications of a 15.7 percent increase in 911 calls boggles the mind. The effect was statistically small, however – i.e., not statistically significant. Ignoring the substantive size of the effect, Drs. Linz, Paul, and Yao (2006) argued that the statistically insignificant effect proved that the real effect was zero – or in other words, that no real effects existed. Their null finding has other interpretations, of course. Reanalyzing the San Diego data, McCleary and Meeker (2006) discovered that Drs. Linz, Paul, and Yao (2006) had failed to find the significant secondary effect because they had not looked “hard enough” for it. A reanalysis using a more powerful statistical design found the significant secondary effect. Figure 7a - Jury Trials and Hypothesis Tests But in Reality, the Defendant is ...

122

Guilty

Not Guilty

The Jury Convicts

95% Confidence

5% False Positives

The Jury Hangs

?

?

The Jury Acquits

20% False Negatives

80% Power

This translates roughly to “Finding nothing proves nothing.” R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 79

Figure 7a summarizes the principles of statistical hypothesis testing by analogy to a jury trial. Suppose that an SOB stands accused of posing an ambient crime risk. After hearing the evidence, the jury can convict, acquit, or hang. If the jury convicts, there is a small (but non-zero) probability that the jury convicted an innocent SOB; i.e., a false-positive (or “Type I” or “á-type”) error. If the jury acquits, on the other hand, there is a small (but non-zero) probability that the jury acquitted a guilty SOB; i.e., a false negative (or “Type II” or “â-type”) error. Finally, if the jury hangs, there was no decision and, hence, no possibility of error. In real-world courtrooms, the probabilities of false-positive and false-negative verdicts is unknown. Courts enforce strict procedural rules to minimize these probabilities but we can only guess at their values. In statistical hypothesis testing, on the other hand, the values are set by rigid conventions, to five percent for false-positives and twenty percent for false negatives.123 Adopting these same values, to convict, the jury must be 95 percent certain of the SOB’s guilt. To acquit, the jury must be 80 percent certain of the SOB’s innocence. To ground the 95 and 80 percent certainty levels, we could try each case in front of a large number of independent juries. To convict, 95 percent of the juries would have to return the same guilty verdict; in the case of an acquittal, 80 percent would return the same not guilty verdict. Correct decisions are painted blue in Figure 7a. Five percent of all convictions are falsepositives and 20 percent of all acquittals are false-negatives. Incorrect decisions are painted red in Figure 7a. When the levels of certainty are too low to support conviction or acquittal, of course, the jury hangs. Non-decisions, painted yellow in Figure 7a, depend on factors such as the strength of evidence, credibility of witnesses, and so forth. So as not waste a jury’s time, the prosecutor doesn’t bring obviously weak cases to trial. Likewise, faced with strong evidence of guilt, the defense counsel seeks a plea bargain in order to avoid trial. The analogy to statistical hypothesis testing is nearly perfect. The researcher considers two complementary hypotheses. The SOB either has secondary effects; or alternatively, the SOB does not have secondary effects. Based on the magnitude of the expected and estimated effects, the researcher then accepts one of the two hypotheses. ! If the false-positive rate for the estimated is smaller than five percent, the hypothetical secondary effect is accepted with 95 percent confidence. The SOB

123

The most comprehensive authority on this issue is Chapter 22 of Kendall and Stuart (1979). This authority requires a strong background in mathematics, however. Cohen (1988) and Lipsey (1990) are more accessible. Both Cohen (1988, pp. 3-4) and Lipsey (1990, pp. 38-40) set the conventional false-positive and false-negative rates at á=.05 and â=.2, respectively. These rates can be set lower, of course. The convention also sets the ratio of false-positives to falsenegatives at 4:1, implying that false-positives are “four times worse than” false-negatives. The 4:1 convention dates back at least to Neyman and Pearson (1928). It reflects a view that science should be conservative. In this instance, e.g., the 4:1 convention works in favor of the SOB. When actual decision error costs are known, the actual ratio is used. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 80

has a large, significant secondary effect. If the false-positive rate is larger than five percent, the researcher does not automatically accept the alternative hypothesis but, rather, conducts a second test. ! If the false-negative rate for the expected effect is smaller than twenty percent, the alternative hypothesis is accepted with 80 percent power. The SOB does not have a secondary effect. But lacking both 95 percent confidence and 80 percent power, neither hypothesis is accepted; the results are inconclusive. Since inconclusive results invariably arise from weak research designs, and since the relative strength of a design is known a priori, inconclusive results should be rare. But in fact, many of the secondary effects studies sponsored by SOB plaintiffs have inconclusive results. An example illustrate the plaintiffs’ rationale. Figure 7b - False-negative Rates for the San Diego Finding

McCleary and Meeker (2006) calculated the false-negative error rates plotted in Figure 7b from statistics reported by Drs. Linz, Paul, and Yao (2006). As shown, the reported 15.7 percent secondary effect estimate has a false-negative rate of .508. What this means, simply, is that the reported null finding is more likely (51 percent) to be incorrect than it is to be correct (49 percent). The effect would have to exceed 22.7 percent (304.5 calls) before it could be detected with the conventional level of 80 percent power. The mathematics of statistical hypothesis testing is so demanding that few social scientists

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 81

understand the concepts or their importance to research.124 The conventional 80 percent power level was proposed and adopted in the 1920s when statistical hypothesis testing was in its infancy. The convention has survived for eighty years because it serves two useful, crucial functions. ! Anyone with a modest background in research methods can design a study in a way that favors – or even guarantees – a null finding. The convention minimizes abuses by malicious investigators. ! Haphazardly designed “quick and dirty” studies favor the null finding. The convention minimizes the impact of spurious findings generated by naive (but benign) investigators. Lay audiences, who must rely on common sense, cannot always distinguish between weak and strong designs or between benign and malicious investigators. Scientific conventions guard against both abuses. In this particular instance, the 80 percent power convention allows the lay audience to trust the validity of a null finding. Recognizing the conventions, secondary effect studies can assigned to one of three categories: studies that report secondary effects with 95 percent confidence; studies that report null findings with 80 percent power; and studies that are inconclusive. All of studies listed in Table 1 above either report large, significant secondary effects or else are inconclusive. No studies report null findings with the conventional 80 percent power. This reinforces a statement that we made in the introduction to this report: It is a scientific fact that SOBs pose large, significant ambient crime risks.

124

E.g.,“I attributed this disregard of power to the inaccessibility of a meager and mathematically difficult literature...” (p. 155, Cohen, 1992).

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 82

8.

References, Authorities, and Documents

In formulating the opinions described in §1-7 above, we relied upon research reports written by ourselves and others, and on standard authorities that would be available in any research library. 8.1

References and Authorities

Akers, R.L. Criminological Theories: Introduction and Evaluation. Los Angeles: Roxbury, 2000. Anderson, S.T. and S.E. West. Open space, residential property values, and spatial context. Regional Science and Urban Economics, 2006, 36:773-789. Armstrong, R.J. and D.A. Rodriguez. An evaluation of the accessibility benefits of commuter rail in eastern Massachusetts using spatial hedonic price functions. Transportation, 2006, 33:21-43. Bennett, T., and R. Wright. Burglars on Burglary: Prevention and the Offender. Gower, 1984. Blalock, H.M. Measurement and Conceptualization in the Social Sciences. Sage, 1982. Bourgeois, P. “Man shot outside video store in Kennedale.” Fort Worth Star-Telegram, March 1st, 2007. Boxall, P.C., W.H. Chan and M.L. McMillan. The impact of oil and natural gas facilities on rural residential property values: A spatial hedonic analysis. Resource and Energy Economics, 2005, 27:248-269. Brasington, D.M. and D. Hite. Demand for environmental quality: A spatial hedonic analysis. Regional Science and Urban Economics, 2005, 35:57-82. Brookshire, D.S., M. A. Thayer, W.D. Schulze and R.C. D’Arge. Valuing public goods: A comparison of survey and hedonic approaches. American Economic Review, 1982, 72:165177. Bursik, R.J. Jr. and H.G. Grasmick. Neighborhoods and Crime: The Dimensions of Effective Community Control. Macmillan, 1993. Cameron, A.C. and P.K. Trivedi. Regression Analysis of Count Data, Econometric Society Monograph 30. Cambridge U Press, 1998. Campbell, Donald T. and Julian C. Stanley, Experimental and Quasi-Experimental Designs for Research. Rand-McNally, 1966. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 83

Capozza, D.R. and R. Van Order. A generalized model of spatial competition. American Economic Review, 1978, 68:896-908. Clarke R. V. and M Felson. Routine Activity and Rational Choice: Advances in Criminological Theory New Brunswick: Transaction Publishers, 1993. Cohen, J. Statistical Power Analysis for the Behavioral Sciences, 2nd Ed. (L.E. Erlebaum Associates, 1988). Cohen, L.E. and M. Felson. Social change and crime rate trends: A routine activity approach. American Sociological Review, 1979, 44:588-608. Cohen, L.E., J.E. Kluegel and K.C. Land. Social inequality and predatory criminal victimization: An exposition and test of a formal theory. American Sociological Review, 1981, 46:505-524. Cook, Thomas D. and Donald T. Campbell, Quasi-Experimentation: Design and Analysis Issues for Field Settings (Houghton-Mifflin, 1979). Cooper, C.B. and E.D. Kelly, Ph.D. Survey of Texas Appraisers: Secondary Effects of Sexually-Oriented Businesses on Market Values. Texas City Attorneys Association, June, 2008. Cooper, C.B. and E.D. Kelly. Everything You Always Wanted to Know About Regulating Sex Businesses. 2001, American Planning Association (PAS/495/496). Danner, T.A. Violent times: A case study of the Ybor City Historical District. Criminal Justice Policy Review, 2003, 14:3-29. Davis, K.C., J. Norris, W.H. George, J. Martell, and R,J. Heiman. Men’s likelihood of sexual aggression: The influence of alcohol, sexual arousal, and violent pornography. Aggressive Behavior, 2006, 32:581-589. Davis, K.C., J. Norris, W.H. George, J. Martell, and R.J. Heiman. Rape myth congruent beliefs in women resulting from exposure to violent pornography: effects of alcohol and sexual arousal. Journal of Interpersonal Violence, 2006, 21:1208-1223. Diggle, P,J. Statistical Analysis of Spatial Point Patterns, 2nd Ed. Arnold, 2002. Ebert, L.M. and K. Fahy. Why do women continue to smoke in pregnancy? Women and Birth, 2007, 20:161-168. Enriquez, R., J.M. Cancino and S.P. Varano. A legal and empirical perspective on crime and adult establishments: A secondary effects study in San Antonio, Texas. American University Journal of Gender, Social Policy and the Law, 2006, 15:1-42. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 84

Erdfelder, E., F. Faul and A. Buchner. G*POWER: A general power analysis program. Behavior Research Methods, Instruments and Computers, 1996, 28, 1-11. Feeney, F. Robbers as Decision-Makers. Pp. 53-71 in Cornish D. and R. Clarke (eds.), The Reasoning Criminal: Rational Choice Perspectives on Offending. (Springer-Verlag, 1986). Feller, William. An Introduction to Probability Theory and its Applications, Volume I, 3rd Ed. (Wiley, 1968 [1st Ed., 1950]). Felson, M. Crime and Everyday Life, 2nd Ed. Pine Forge Press, 1998. Felson, M. Crime and Nature. Sage, 2006. Fenton, K.A., A.M. Johnson, S. McManus, and B. Erens. Measuring sexual behaviour: methodological challenges in survey research. Sexually Transmitted Infections, 2001, 77: 84-92 Fisher, R.A. Statistical Methods for Research Workers. Edinborough, Oliver and Boyd, 1925. Fleisher, M. Beggars and Thieves: Lives of Urban Street Criminals. (U Wisconsin Press, 1995). Frank, K. G-Strings and Sympathy: Strip Club Regulars and Male Desire. Duke University Press, 2002. Freedman, D. On the so-called “Huber Sandwich estimator” and “robust standard errors.” University of California-Berkeley, Department of Statistics, 2006. Freeman, A.M. On estimating air pollution control benefits from land value studies. Journal of Environmental Economics and Management, 1974, 1:74-83. Freeman, A.M. Hedonic prices, property values and measuring environmental benefits: A survey of the issues. Scandinavian Journal of Economics, 1979, 81:154-173. Garnett, N.S. Relocating disorder. Virginia Law Review, 2005, 91:1076-1134. Goldstein, H. Problem-Oriented Policing. Wiley, 1990. Goodwin, C. A conceptualization of Motives to seek privacy for nondeviant consumption. Journal of Consumer Psychology, 1992, 1:261-284. Grasmick, H.G., R.J. Bursik, Jr. and J.K. Cochran. “Render unto Caesar what is Caesar’s”: Religiosity and taxpayers’ inclinations to cheat. Sociological Quarterly, 1991, 32:251-266. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 85

Greenberg, D.F. Mathematical Criminology (Rutgers U Press, 1979). Haight, F. Handbook of the Poisson Distribution (Wiley, 1967). Hintze, H. Power Analysis and Sample Size, Version 6 (PASS). Kayesville, UT, NCSS, Inc. 2000. Hoenig, J.M. and D.M. Heisey. The abuse of power: the pervasive fallacy of power calculations for data analysis. The American Statistician, 2001, 55:1-6. Jacobs, B. A. 2000. Robbing Drug Dealers: Violence Beyond the Law. (New York: Aldine de Gruyter). Judge, G.G., W.E. Griffiths, R.C. Hill, H. Lütkepohl, and T-C. Lee. The Theory and Practice of Econometrics. Wiley, 1985. Katz, J. Seductions of Crime: Moral and Sensual Attractions in Doing Evil (Basic Books, 1988). Katz, J. The Motivation of the Persistent Robber. In Tonry, M. (ed.), Crime and Justice: A Review of Research (U Chicago Press , 1991). Kendall, M. and A. Stuart, Chapter 22 of The Advanced Theory of Statistics, 4th Ed. (Charles Griffin and Co., 1979 [1st Ed., 1946]). Kennedy, L.W. and D.R. Forde. Routine activities and crime: An analysis of victimization in Canada. Criminology, 1990, 28:137–152. Kim, C.W., T.T. Phipps and L. Anselin. Measuring the benefits of air quality improvement: A spatial hedonic approach. Journal of Environmental Economics and Management, 2003, 45:24-39. Klinger, D. and G.S. Bridges. Measurement errors in calls-for-service as an indicator or crime. Criminology, 1997, 35:529-41. Kornhauser, Ruth. 1978. Social Sources of Delinquency. Chicago: University of Chicago Press. Laverne, R.J. and K. Winson-Geideman. The influence of trees and landscaping on rental rates at office buildings. Journal of Arboriculture, 2003, 29(5), 281-290. Levitt, S.D. Using electoral cycles in police hiring to estimate the effect of police on crime. American Economic Review, 1997, 87:270-290.

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 86

Li, W. and J-D. Saphores. Estimating the benefits of urban forests in the Los Angeles (CA) multi-family housing market: A spatial hedonic approach. Paper presented at the 50th Anniversary Association of Collegiate Schools of Planning Conference, Crystal City, VA, October 2009. Linz, D., B. Paul, K.C. Land, M.E. Ezell and J.R. Williams. An examination of the assumption that adult businesses are associated with crime in surrounding areas: A secondary effects study in Charlotte, North Carolina. Law and Society Review, 2004, 38:69-104. Lipsey, M. Design Sensitivity: Statistical Power for Experimental Research. (Sage Publications, 1990). Longo, A. and A. Alberini. What are the effects of contamination risks on commercial and industrial properties? Evidence from Baltimore, Maryland. Journal of Environmental Planning and Management, 2006, 49:713-737. Mayhew, H. London’s Underworld. Edited by P. Quennel. Bracken Books, 1983. McCarthy, G.W., H. Renski and D. Linz. 2001. Measuring Secondary Effects Using Spatio-Temporal Estimation of Real Estate Appreciation. Unpublished manuscript. McCleary, R. Rural hotspots: the case of adult businesses. Criminal Justice Policy Review, 2008, 19:153-163. McCleary, R. and J.W. Meeker. Do peep shows “cause” crime? Journal of Sex Research, 2006, 43:194-196. McCleary, R. and A.C. Weinstein. Do “off-site” adult businesses have secondary effects? Legal doctrine, social theory, and empirical evidence. Law and Policy, 2009, 31:217-235. Miethe, T.D. and R.F. Meier, Crime and its Social Context: Toward and Integrated Theory of Offenders, Victims, and Situations Albany, NY: State University of New York Press, 1994. Merton, R.K. The unanticipated consequences of purposive social action. American Sociological Review, 1936, 1:894-904. Mills, J.L. Data torturing. New England Journal of Medicine, 1993, 329:1196-1199. Mustaine, E.E. and R. Tewksbury. Predicting risks of larceny theft victimization: a routine activity analysis using refined lifestyle measures. Criminology, 1997, 36:829–858. National Research Council. Fairness and Effectiveness in Policing: The Evidence. National Academies Press, 2004. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 87

Newman, O. Defensible Space: Crime Prevention Through Urban Design. MacMillan, 1973. Neyman, J. and E.S. Pearson. On the use and interpretation of certain test criteria for purposes of statistical inference. Biometrika, 1928, 20A:175-240. Norris, J., K.C. Davis, W.H. George, J. Martell, and J.R. Heiman. Alcohol's direct and indirect effects on men's self-reported sexual aggression likelihood. Journal of Studies on Alcohol, 2002, 63, 688-695. Norris, J. and K. L. Kerr. Alcohol and violent pornography: Responses to permissive and nonpermissive cues. Journal fo Studies on Alcohol, 1993, 54:118-127. Rantala, R.R. and T.J. Edwards. Effects of NIBRS on Crime Statistics. U.S. Bureau of Justice Statistics, Office of Justice Programs, February 23, 2001. Roberg, R.R., J. Crank, and J. Kuykendall, Police and Society Wadsworth, 1999. Roncek D.W. and M.A. Pravatiner “Additional Evidence that Taverns Enhance Nearby Crime,” Social Science Research 1989, 73:185-188. Roncek, D. and P. A. Maier. Bars, blocks, and crimes revisited: Linking the theory of routine activities to the empiricism of “hot spots.” Criminology, 1991, 29: 725-753. Rosen, Sherwin. 1974. Hedonic Prices and Implicit Markets: Product Differentiation in Pure Competition. Journal of Political Economics 82:34-35. Rubin, D. Matched Sampling for Causal Effects. Cambridge University Press, 2006. Sampson, R. J. and W.B. Groves. Community structure and crime: testing social disorganization theory. American Journal of Sociology, 1989, 94:774-802. Sampson, R.J., S.W. Raudenbush and F. Earls. Neighborhoods and violent crime: A multilevel study of collective efficacy. Science, 1997, 277:918-924. Sanchez, L.E. Sex, violence citizenship, and community: an ethnography and legal geography of commercial sex in one American city. Ph.D. Dissertation, Criminology, Law and Society, University of California, Irvine, 1998. Saphores, J-D. and W. Li. Estimating the benefits of urban forests: An application of fixed effects models to the Los Angeles (CA) single family housing market. Paper presented at the 56th Annual North American Meetings of the Regional Science Association International, San Francisco, November 2009.

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 88

Schmid, C,F, Urban crime areas: Part I. American Sociological Review, 1960, 25:527542. Scott, M.S. Assaults in and Around Bars. U.S. Department of Justice, Office of Community Oriented Policing. 2002. Shadish, William R., Thomas D. Cook, and Donald T. Campbell, Quasi-Experimental Designs for Generalized Causal Inference (Houghton-Mifflin, 2002). Shaw, C.R. The Jack-Roller: A Delinquent Boy's Own Story. University of Chicago Press, 1966 [1930]. Sherman, L.W., P.R. Gartin and M.E. Buerger. Hot spots of predatory crime: routine activities and the criminology of place. 1989, Criminology, 27:27–56. Shover, N. Great Pretenders: Pursuits and Careers of Persistent Thieves. (Westview, 1996). Simcha-Fagan, O. and J.E. Schwartz. Neighborhood and delinquency: An assessment of contextual effects. Criminology, 1986, 24:667-704. Skogan, W.G. Disorder and Decline: Crime and the Spiral of Decay in American Neighborhoods. University of California Press, 1992. Stiger, M. and R. McCleary “Confirmatory spatial analysis by regression of a Poisson variable” Journal of Quantitative Anthropology 1989, 2:13-38. StripClub News. Investigation tied to strip club leads to resignations and charges. September 22nd, 2006. Taylor, R.B, S.D. Gottfredson and S. Brower. Block crime and fear: Defensible space, local social ties, and territorial functioning. Journal of Research in Crime and Delinquency, 1984, 21:303-331. Thurman Q. and E.G. McGarrell. Community policing in a rural setting. Anderson Publishing, 1997. Warner, B.D. Directly intervene or call the authorities? A study of forms of neighborhood social control. Criminology, 2007, 45:99-121. Weisheit, R.A., D.N. Falcone, and L.E. Wells. Crime and Policing in Rural and Smalltown America. Waveland Press, 1999. Wilson, J.Q. and G.L. Kelling. Broken windows: The police and neighbor-hood safety. R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 89

Atlantic Monthly, 1982, 249:29-38. Wright, R.T. and S.H. Decker. Armed Robbers in Action: Stickups and Street Culture. Northeastern University Press, 1997. 8.2

Secondary Effects Studies

Los Angeles, CA, 1977. Study of the Effects of the Concentration of Adult Entertainment Establishments in the City of Los Angeles. Department of City Planning, June, 1977. Amarillo, TX, 1977. A Report on Zoning and Other Methods of Regulating Adult Entertainment in Amarillo. City of Amarillo Planning Department, September 12th, 1977. Whittier, CA, 1978. Staff Report, Amendment to Zoning Regulations, Adult Businesses in C-2 Zone with Conditional Use Permit, Case No. 353.015. January 9th, 1978. St. Paul, MN, 1978. Effects on Surrounding Area of Adult Entertainment Businesses in St. Paul. Department of Planning and Economic Development and Community Crime Prevention Project, June, 1978. Phoenix, AZ, 1979. Adult Business Study. City of Phoenix Planning Department, May 25, 1979. Minneapolis, MN, 1980. An Analysis of the Relationship between Adult Entertainment Establishments, Crime, and Housing Values. Minnesota Crime Prevention Center, Inc. M. McPherson and G. Silloway, October, 1980. Indianapolis, IN, 1984. Adult Entertainment Businesses in Indianapolis, An Analysis. Department of Metropolitan Development, Division of Planning. March, 1984. Austin, TX, 1986. Report on Adult Oriented Businesses in Austin. Office of Land Development Services, May 19th, 1986. El Paso, TX, 1986. Effects of Adult Entertainment Businesses on Residential Neighborhoods. Department of Planning Research and Development Oklahoma City, OK, 1986. Adult Entertainment Businesses in Oklahoma City: A Survey of Real Estate Appraisers. Community Development Department, Planning Division , Oklahoma City. Garden Grove, CA, 1991. Final Report to the City of Garden Grove: The Relationship between Crime and Adult Business Operations on Garden Grove Boulevard. October 23, 1991. Richard McCleary, Ph.D. and James W. Meeker, J.D., Ph.D.

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 90

New York Times Square, NY, 1994. Report on the Secondary Effects of the Concentration of Adult Use Establishments in the Times Square Area. Insight Associates. April, 1994. Newport News, VA, 1996. Adult Use Study. Department of Planning and Development. March, 1996. Dallas, TX, 1997. An Analysis of the Effects of SOBs on the Surrounding Neighborhoods in Dallas, Texas. Peter Malin, MAI. April, 1997. Ft. Wayne, IN, 2001. Measurement of Negative Secondary Effects Surrounding Exotic Dance Nightclubs in Fort Wayne, Indiana. Daniel Linz and Bryant Paul. February 13th, 2001. San Diego, CA, 2002. A Secondary Effects Study Relating to Hours of Operation of Peep Show Establishments in San Diego, California. Daniel Linz and Bryant Paul. September 1, 2002. San Diego, CA, 2003. A Methodical Critique of the Linz-Paul Report: A Report to the San Diego City Attorney’s Office. R. McCleary and J.W. Meeker. March 12, 2003. See also, McCleary, R. and J.W. Meeker (2006). Do peep shows “cause” crime? Journal of Sex Research, 43, 194-196. Greensboro, NC, 2003. Evaluating Potential Secondary Effects of Adult Cabarets and Video/Bookstores in Greensboro: A Study of Calls for Service to the Police. Daniel Linz and Mike Yao, November 30th, 2003. Greensboro, NC, 2003. A Methodical Critique of the Linz-Yao Report: Report to the Greensboro City Attorney. R. McCleary. December 15th, 2003. Centralia, WA, 2004. Crime Risk in the Vicinity of a Sexually Oriented Business: A Report to the Centralia City Attorney’s Office. Richard McCleary, Ph.D. February 28, 2004. Toledo, OH, 2004. Evaluating Potential Secondary Effects of Adult Cabarets and Video/Bookstores in Toledo, Ohio: A Study of Calls for Service to the Police. Daniel Linz and Mike Yao. February 15th, 2004. Toledo, OH, 2004. A Methodological Critique of the Linz-Yao Report: Report to the City of Toledo, OH. R. McCleary and J.W. Meeker. May 15th, 2004. Daytona Beach, FL, 2004. Evaluating Potential Secondary Effects of Adult Cabarets in Daytona Beach Florida: A Study of Calls for Service to the Police in Reference to Ordinance 02-496. D. Linz, R.D. Fisher, and M. Yao. April 7th, 2004. New York City, NY, 2005. Measuring the Secondary Effects of 60/40 Businesses in New R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 91

York City: An Analysis of Calls for Service to the Police. Linz, D. and B. Paul. April 14th, 2005. Daytona Beach. FL, 2006. A Methodological Critique of Evaluating Potential Secondary Effects of Adult Cabarets in Daytona Beach, Florida: A Study of Calls for Service to the Police in Reference to Ordinance 02-496. R. McCleary, May 1st, 2006. Palm Beach County. FL, 2007. Crime-Related Secondary Effects of Sexually-Oriented Businesses: Report to the County Attorney. Jenness, V., R. McCleary and J.W. Meeker. Los Angeles, CA, 2007. Crime-Related Secondary Effects of Sexually-Oriented Businesses: Report to the City Attorney. R. McCleary. 8.3

Other Relevant Documents

Abilene Retail #30, Inc. v. Board of Commissioners of Dickinson County Annex Books, Inc. v. City of Indianapolis City of Chicago v. Pooh Bah Enterprises, Inc. City of Los Angeles v. Alameda Books, Inc. City of Renton v. Playtime Theatres, Inc. Daubert v. Merrell Dow Pharmaceuticals Daytona Grand, Inc. v. City of Daytona Beach J.L. Spoons, Inc. v. Nancy J. Dragani, et al. Young v. American Mini Theaters, Inc. Danner, T. A. The Crime-related Secondary Effects of Adult Cabarets in Palm Beach County. Report submitted in Palm Beach County v. Casablanca East, CA-02-03813 AF, Circuit Court, Fifteenth Judicial Circuit, Palm Beach County, 2005. Fisher, R. D., D. G. Linz and C. Seaman. Supplemental Crime Analysis: Club Eden v. Town of Davie, Florida. March 3, 2008. Freeman, L. Curriculum vitae. Freeman, L. and D. Hamilton. Examining the Relationship Between Adult Oriented Businesses and Surrounding Property Values in Ohio. No date.

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 92

Linz, D. Rule 52 Report, January 15th, 2010. Linz, D. Curriculum vitae. Linz, D. Direct Testimony in J.L. Spoons, Inc. v. Nancy J. Dragani, et al. U.S. District Court, Northern District of Ohio, Cleveland, Case No. 04-00314.March 12, 2004. Linz, D. Cross Examination in J.L. Spoons, Inc. v. Nancy J. Dragani, et al. U.S. District Court, Northern District of Ohio, Cleveland, Case No. 04-00314. March 12, 2004. Linz, D. Measuring the Secondary Effects of Adult Cabarets in Greensboro, NC: A Study of Police Reports. No date. Linz, D. and M. Yao. A Study of Secondary Crime Effects of Adult Cabarets in Seattle Washington. October 12, 2006. Linz, D. and M. Yao. Measuring the Secondary Effects of Adult Businesses in the Rancho Cordova Area: A Study of Crime Incidents. February 20, 2005. Linz, D., M. Yao, and S. Byrne. Testing Supreme Court assumptions in California v. la Rue: Is there justification for prohibiting sexually explicit messages in establishments that sell liquor? Communication Law Review, 2007, 7:23-53. Liquor Control Commission of the State of Ohio, Transcript of the Public Hearing re: Rule 4301:1-1-52. Paul, B. and D. Linz. Testing assumptions made by the Supreme Court concerning the negative secondary effects of adult businesses: A quasi-experimental approach. International Communication Association: Communication Law and Policy. 2002. Paul, B., D. Linz, and M. Yao. Evaluating the potential secondary effects of adult video/bookstores in Indianapolis, IN. Freedom of Expression Division of the National Communication Association. 2007. Report of the Attorney General’s Working Group on the Regulation of Sexually Oriented Businesses. Attorney General’s Working Group, St. Paul, MN, 1989. Scott, M. S. And K. Dedel. Assaults in and Around Bars. 2nd edition. Office of Community Oriented Policing Services, U.S. Department of Justice. Seaman, C. and D. Linz. Erotic dancing, liquor, and crime: An empirical critique of Virginia statute changes restricting liquor service and adult entertainment. Annual Convention of the National Communication Association, November 12-15, 2009.

R ICHARD M C C LEARY . P H.D . AND W ENDY R EGOECZI, P H.D . - P AGE 93