Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion Eric M. Uslaner December 2012 A working paper from the Center for American Progress
1 Introduction and summary 4 The dream and the reality of inequality 10 Summary of findings 13 Inequality and pessimism: detailed results 22 Conclusion 24 Appendix 33 About the author and acknowledgements 34 Endnotes
Introduction and summary Rising inequality over the past decades led Americans to lose faith in some key aspects of the “American Dream” of a future that will inevitably be better than the past. We are less likely to believe that we have much in common with people who are not like ourselves and also that the people in power listen to or care what ordinary Americans think. Inequality makes people less likely to believe that what affects me affects you—and that ordinary people have the power to control the future or their political leaders. I use data from the American National Election Studies surveys between the years 1966 and 2008, the General Social Survey (1972-2010), and the Pew Values Surveys (1987-2009) to examine the effects of rising inequality on indicators of optimism and pessimism, social cohesion, confidence in institutions, and personal expectations for the future. Key findings include: FIGURE 1
Americans' trust in government is in decline
Americans are now less optimistic about economic growth
Americans are more pessimistic about the condition of the average person
Percentage of respondants who believe there is no real limit to growth
Percentage of respondants who believe the lot of the average person is getting worse
Share of respondants who have trust in government 80% 70%
Source: American National Election Studies
Source: Pew Values Survey
Source: General Social Survey
1 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
• Americans have become more pessimistic and less trusting as inequality has increased in recent decades—especially as the income share of the top 20 percent rose while the share of the bottom 80 percent declined.
Americans have become less trusting of other people Percentage of respondants who answered most people can be trusted
• Between 1973 and 1994 (the first and last times these questions were asked in a major national survey) the share of Americans who believed that the “lot of the average person is getting worse” rose from 56 percent to 69 percent —and all of this rise is attributable to increasing inequality.
• Between 1968 and 2006 the share of Americans who believed that “most people can be trusted” fell from 56 percent to 34 percent. Almost all of this drop is attributable to increasing inequality.
• In 1968, 63 percent of Americans believed that the government in Washington could be trusted to do the right thing, but only 14 percent agreed in 2008. Approximately a third of this drop can be attributed to increasing inequality. • Between 1975 and 2010, the share of Americans who believe that “officials don’t care what people like me think” rose from 56 percent to 71 percent. This decline clearly tracks increasing inequality. • Rising inequality leads people to become less optimistic about the future of our country and more skeptical of both their fellow citizens and their leaders. But it does not make them more pessimistic about their own prospects for success. Such overarching pessimism—notwithstanding the finding that Americans remain individually optimistic about their individual futures—leads to less social cohesion and to greater polarization. This makes finding common ground on policy issues more difficult.
70% 60% 50%
30% 20% 10% 0%
Source: General Social Survey FIGURE 5
Americans have become more distrusting of public officials Percentage of respondants who believe public officials don't care about people like them 80% 70% 60% 50% 40% 30% 20% 10% 0%
Source: General Social Survey
2 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
I find modest direct links between pessimism and the economic outcomes for which I could obtain data—owning homes, investing in stocks or starting a business. But there are probably greater effects for economic decisions that I could not test directly. And the effects of inequality on social and political cohesion may lead to negative economic outcomes. Polarization has already made it difficult for the leaders of the two political parties in the United States to reach accord on budgetary issues, which have clear FIGURE 6 consequences for the economy.
Rising income inequality has caused increasing pessimism
Policymakers should be concerned with the negative consequences of rising inequality and pessimism for the American dream. We are a society in which people have a fundamental belief that things are going to get better and that everyone will have a chance to succeed if she works hard enough. This dream—or promise, as most Americans interpret it—seems further away for many people.
Increase in pessimism caused by Gini Coefficient going from lowest value to highest value 35% 30% 25% 20% 15% 10% 5% 0%
Lot of the average It’s not fair to person would bring a child into be worse the world
Trust in other people
Source: Author's calculations
3 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Trust in government
Officials care about average person
The dream and the reality of inequality Pessimism is not part of the American mindset. The progressive theorist Herbert Croly summarized the American Dream as the belief that “the future will have something better in store for them individually and collectively than has the past or the present.” The historian Henry Steele Commager argued, “Nothing in all history had succeeded like America, and every American knew it.” This creed guided immigrants to come to a land where streets were paved with gold, to join a “people of plenty,” in historian David Potter’s words.1 Not only were Americans optimistic about the future, but they also believed that they had the power to make the American Dream come true. The Economist expressed this ideal well in 1987: “Optimism, not necessity, has always been the mother of invention in America. To every problem–whether racial bigotry or putting a man on the moon—there has always been a solution, if only ingenuity and money were committed to it.” People are optimistic in part because they believe that they can make things better.2 But as inequality arose over the past 50 years, Americans became less optimistic. They are less likely to believe that tomorrow will be better than today and that they can make it happen. As they become pessimistic, they are less likely to believe that what happens to me affects you, especially if you are not just like me. The poor don’t see a common destiny with the rich, whites don’t see a linked fate with minorities, and we are less likely to trust people of different backgrounds. The growing income gap makes us less confident that we all have common interests, especially when we perceive that some—those at the top—have far more control over our economic and political life. Rising inequality leads to pessimism and frayed social and political ties. Equality is central to the American Dream, as noted by Lord Bryce, a British visitor to the United States. wrote in the late 19th century: “There is no rank in America, no external and recognized stamp, marking one man as entitled to any social privileges, or to deference and respect from others.” Yet, as economists Christian
4 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Bjornskov, Benno Torgler, and their co-authors argued in 2006, high levels of inequality are linked with frustration and perceptions of unfairness in many societies.3 Equality is thus the social glue underlying the American Dream—and growing inequities lead to both pessimism and a weakened sense of community. As the level of economic inequality has grown over time, Americans have become more pessimistic about their futures, less connected to people of different backgrounds, and more convinced that political leaders only respond to the wealthy— and these effects are more pronounced for those who see themselves at the bottom of the economic ladder. Yet, increasing disparities in wealth do not make Americans less optimistic about their own economic fate. Even as growing inequality leads people to become more pessimistic about the country’s future, to become more wary of people who are different from themselves, and to have less confidence in government and the responsiveness of public officials, they still remain optimistic about their personal economic opportunities. The Occupy Wall Street protests pitted the vast bulk of the American population against the “top 1 percent.” Is that really the fault line in American life? When either scholars or popular commentators discuss the growing economic divide in American society, they focus on a single measure, such as the Gini index or what share of national wealth the very richest have. The Gini index is the most commonly used measure: It ranges between zero (perfect equality) and one (perfect inequality) and has risen from 0.397 in 1966, already relatively unequal, to 0.468 in 2009 (the highest of any Western country).4 Yet the Gini index doesn’t tell us anything about where the fault lines of inequality are. So I also use measures of the income shares of each 20 percent of American society and the top 5 percent—to see where differences in income matter most My findings come from three analyses. I use the major academic and media surveys that have been conducted over time and are publicly available: • The American National Election Studies, incorporating data from 1966 to 2008 • The General Social Survey, with data from 1972 to 2010 • The Pew Values Surveys, with data from 1987 to 2009 I examine 28 measures of overall optimism and pessimism; social cohesion; confidence in government; and expectations for one’s personal well-being that have
5 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
been asked in at least eight surveys within these three sources of data over time. Based upon statistical models, I examine how much more likely people are to hold onto the American Dream and other measures of social and political cohesion as income inequality rose over the past five decades. (see the Methodology box on page 8) I then turn to some economic consequences of pessimism. My second set of analyses is based upon aggregate data on housing purchases and business starts and failures from Census Bureau data Does pessimism make people less likely to buy houses? Are the rates of business starts and failures linked to less favorable outlooks for the future? Finally (third), I examine whether pessimism makes investors less likely to their plans for investments and their retirement goals with data from a 1998 Gallup/UBS survey of people with savings and investments worth at least $10,000 and whether the larger public is more likely to buy homes or stock in a 2009 survey of the Pew Economic Mobility Project Americans became more pessimistic as the income shares of the bottom four groups fell and those of the top 20 percent and top 5 percent rose. The social chasm isn’t just between those at the very top and everyone else—but between 80 percent of Americans and the other 20 percent. Moreover, growing inequality led to increased pessimism for lower- and /working and middle-income Americans classes, but not for those at the top of the economic scale. As the rich got richer, they did not become more optimistic. But the general pessimism affecting most other Americans did not extend to them. If inequality generally leads to greater pessimism, is there a spillover from pessimism to economic activity? We might expect to find that when people fear for the future, they would be less likely to invest in the stock market and to spend more on big-ticket items such as new homes. Yet, I find only modest evidence that this is the case either from survey or aggregate data. Some of these mixed results may reflect the available data, which are sparse. Yet, most investments (especially in stocks) are made by people at the top of the economic ladder. Their income shares have not fallen—indeed, they have risen.—over time. And they have not become as pessimistic as other Americans. So the link between pessimism and economic activity may be more nuanced than we might expect.
6 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
The average person’s lot is worse, but people see their own future positively People are approximately 30 percent more likely to believe that the lot of the average person is getting worse as the income share of the bottom 20 percent of income earners falls from its maximum value of 4 percent when the question was first asked in the General Social Survey in 1972 to 3.6 percent in 1994, the last time the question was asked. While the shift in income share may not seem very strong, even this modest shift has a very strong effect on people’s pessimism. It barely matters which measure of inequality I use: Inequality leads to a 25-to-30 percent increase in the belief that the average person is worse off whether it is measured by the bottom, second, third, or fourth quintile in the income distribution—or by the Gini index. There also is a 25 percent increase in pessimism as the rich get richer (the income shares of the top 20 percent and top 5 percent rise). As the rich get richer and the poor (or even the upper middle class) get poorer, people become more pessimistic. But measures of personal optimism, such as whether hard work is the key to getting ahead, admiring hard work or satisfaction with one’s personal economic situation, are either unrelated or weakly tied to levels of inequality.
Social cohesion is lower There are similar findings for the most commonly used measure of social cohesion, generalized trust, or the belief that “most people can be trusted.”As the income share of the bottom 20 percent of income earners fell from 4.2 percent in 1968 to three percent in 2010, the likelihood that someone would agree that “most people can be trusted” fell by 20 percent—regardless of which measure of inequality I use—or whether I examine data from the American National Election Studies surveys or the General Social Survey. Americans have become less likely to “trust the government in Washington to do the right thing” by about 20-to-25 percent as inequality has increased from 1968 to 2010. Yet measures of personal optimism such as whether hard work is the key to getting ahead, admiring hard work or satisfaction with one’s personal economic situation are either unrelated or weakly tied to levels of inequality.
7 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Survey questions used in this analysis
I estimate statistical models to derive estimates of the effects of the measures of inequality (income shares of each quintile, the proportion of national income received by the top 5 percent of Americans, and the Gini index) on the 28 measures of overall optimism/pessimism, social cohesion, confidence in government, and expectations for one’s personal well-being that have asked in at least eight surveys over time. The measures I chose are all standard indicators of optimism, social cohesion, confidence in institutions, and personal expectations for the future in the academic (and popular) literature. (see the next box for list of survey questions). The “effects” I estimate (see the Appendix for a technical discussion) are the changes in probability of an outcome as inequality (or any other variable) increases from its minimum to its maximum value: How much less likely is someone to be an optimist, for example, as inequality increases over time? I estimate the impact of each measure
of inequality separately (since they are all highly correlated) in statistical models that control for key demographics (education, income, race, and class) and national economic conditions such as unemployment, inflation, and growth in gross domestic product. I estimate the effect of economic inequality on pessimism, social cohesion and government controlling for other factors that are likely to impact these variables. Measures of economic fault lines:
The U.S. Census Bureau has data on the income shares by quintile from 1967 to 2010—measuring how much of the national income went to the bottom fifth, the second fifth, etc.—as well as to the top 5 percent. The bottom four quintiles all lost ground over this 44- ear period. The income share of the upper middle class fell as the top 20 percent (and top 5 percent) prospered. These measures let us get a better idea of where the fault lines are.
8 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Survey questions used in this analysis Measuring optimism versus pessimism • In spite of what some people say, the lot (situation/condition) of the average man is getting worse—agree or disagree (GSS)* • It is hardly fair to bring a child into the world with the way things look for the future—agree or disagree (GSS) • Success in life is pretty much determined by forces outside our control—agree or disagree (Pew)* • I don’t believe that there are any real limits to growth in this country— agree or disagree (Pew) • As Americans we can always find a way to solve our problems and get what we want—agree or disagree (Pew)
• The strength of this country today is mostly based on the success of American business— agree or disagree (Pew) • Generally speaking, elected officials in Washington lose touch with the people pretty quickly—agree or disagree (Pew) • People like me don’t have any say about what the government does— agree or disagree (Pew) • Voting gives people like me some say about how government runs things—agree or disagree (Pew) Personal Optimism and Pessimism • Some people feel they can run their lives pretty much the way they want to; others feel the problems of life are sometimes too big for them. Which one are you most like? (ANES)
Social cohesion • Generally speaking do you believe that most people can be trusted, or can’t you be too careful in dealing with people? (GSS, ANES)* • We have gone too far in pushing equal rights in this country—agree or disagree (Pew) • I don’t have much in common with people of other races— agree or disagree (Pew)
• When you do make plans ahead, do you usually get to carry out things the way you expected, or do things usually come up to make you change your plans? (ANES) • Some people say that people get ahead by their own hard work; others say that lucky breaks or help from other people are more important. Which do you think is most important? (GSS) • I admire people who get rich by working hard— agree or disagree (Pew)
• Today it’s really true that the rich just get richer while the poor get poorer—agree or disagree (Pew)
• Hard work offers little guarantee of success— agree or disagree (Pew)
• Many people today think they can get ahead without working hard and making sacrifices— agree or disagree (Pew)
• How much of the time do you trust the government in Washington to do the right thing? (ANES) • Public officials don’t care much what people like me think—agree or disagree (ANES, GSS, Pew)
• I’m pretty well satisfied with the way things are going for me financially—agree or disagree (Pew) • I often don’t have enough money to make ends meet—agree or disagree (Pew)
• Over the years, how much attention do you feel the government pays to what the people think when it decides what to do—a good deal, some, or not much? (ANES)
*Key to survey organizations: ANES: American National Election Studies (www.electionstudies.org) GSS: General Social Survey (http://www3.norc.org/gss+website/) Pew: Pew Research Center for The People & The Press (http://www.people-press.org/)
• The government is really run for the benefit of all the people—agree or disagree (Pew)
9 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Summary of findings When we put all of the findings together, we see remarkably similar results across the domains. When the income shares of the first, second, third, and fourth quintiles are highest (in the late 1960s or early 1970s in most of the time series), Americans are more optimistic overall, have the greatest sense of social cohesion, and the highest levels of confidence in their leaders and the representativeness of the political system more generally—mostly by about 15-to 20 percent margins When overall inequality is highest and when the people at the top of the economic ladder fare the best, we become more pessimistic—overall, about our social ties, and about our political system (by about 10 percent). Yet inequality seems to have inconsistent, insignificant, and mostly trivial effects on how we view our own prospects. Inequality bothers Americans, but we remain a people committed to individual effort—and we continue to believe that each of us can fare well regardless of whether the system is stacked against us.
Are fraying ties universal? Are the trends I have described universal? Or is the loss of faith in the American Dream and fraying social and political ties concentrated on those who have lost out as inequality has increased? Two of the three major surveys I used in my analysis, the American National Election Studies surveys and the General Social Survey, but not the Pew Values surveys, ask people which class they belong to. These are not objective measures of economic status, but they give us a good idea of where people believe that they fit into American society. It would seem reasonable that people who see themselves as less well off would be more pessimistic as their objective situations got worse as inequality increases. Yet the support for this argument is limited. Overall, inequality seems to have made most people more pessimistic and less trusting.
10 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
I consider (when there are sufficient cases for analysis) the same models by people’s perceived social class that I estimated for all respondents. Here we don’t see big differences by perceived social class. The General Social Survey asked people if they belonged to the upper, middle, working, or lower class. The American National Election Studies surveys did not include “lower class” as an option. There is far more agreement across (subjective) classes than disagreement about whether the “lot of the average person” is getting worse. The “top” three classes in the General Social Survey were all about 30 percent more likely to agree with this pessimistic world view. Lower-class respondents were less likely to be more pessimistic (by a 10 percent to 15 percent margin across inequality measures)—largely because they started out as more convinced that the economic system was stacked against them. Trust in other people declined by similar levels (about 20 percent) for each class for the American National Election Studies data. Trust fell at somewhat lower levels than in the General Social Survey data for most people, but barely at all for upper-class respondents. The upper and middle classes have become less trusting in government by about 30-to-35 percent in the American National Election Studies surveys from 1968 until 2008, with the biggest decline as inequality grows registered among the upper and middle classes (30-to-35 percent), compared to 20 percent for the working class. The upper and lower classes, but not the middle and working classes, have become less likely to agree (by 20-to-30 percent) that hard work is the path to success as inequality has grown from 1973 to 2010. These results are based upon analyses of survey data for individual respondents that also include measures of inequality for each year of each of the surveys. An alternative method of analysis is to derive estimates of pessimism and cohesion for each class for each year of the surveys—and then to see whether these aggregate measures go up or down as inequality rises. This analysis (see Methodology Box 2 on page 9) provides somewhat stronger support for the argument that people who identify with the working and lower classes have become less trusting of others, less convinced that hard work leads to success, and that public officials don’t care what the average person thinks. But there is no support that the links between inequality and whether the lot of the average person is getting worse or trust in government vary by perceived class.
11 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
There is limited evidence that people who see themselves as working or lower class have become more pessimistic as inequality increased compared to middle and upper-class Americans. Rising inequality is a tax on optimism across the board. Those who identify with the working or lower class become more pessimistic, but so do members of the middle and even upper classes.
Pessimism and economic activity Finally, I turn to the effects of pessimism on indicators of economic activity. I find only modest (simple) correlations between inequality and aggregate measures of houses sold and business starts and failures. I then examine surveys of investors in 1989 and of the full public in 2009 and estimate the impact of optimism in statistical models similar to those described above. Investors are only slightly more likely to make retirement plans or to start businesses when they are optimistic about the future—perhaps because investors are a self-selected group to begin with. The 2009 survey shows greater impacts for optimism on investing in stocks and especially in plans to buy a house. Since more “average” people buy homes than invest in the stock market, the greater impact of optimism on house purchases makes sense. Americans who are optimistic about the future are more likely to say they own their homes, but there is, perhaps surprisingly, no impact of any measure of optimism on home ownership. The available measures of economic activity are sparse so we should not conclude that pessimism does not affect how we behave. Yet, the negative impacts of inequality on optimism, social cohesion, and institutional trust—and the strong link between inequality and social cohesion5—mean that both inequality and pessimism matter more for social than for economic life.
12 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Inequality and pessimism: detailed results I turn now to a more detailed explication of the results summarized above. This is not a technical discussion—but I do provide more specifics about the estimates I briefly described above. The class divisions enhanced by growing inequality may not sap Americans’ faith in their own individual efforts, but they do lead to a greater pessimism for the future, especially for the long-term and to feelings of helplessness. Optimism and control, I argue, are the key to generalized trust—trust in strangers, who are likely to be different from yourself: “Pessimists withdraw into their own communities. They see others as malevolent...Just as optimists believe that they have the power to change the world, pessimists see a dark future beyond their control.”6 And thus trusting those whose interests are opposed to yours seems like a bad bet. Growing inequality leads to less optimism and weaker feelings of control over our lives—and, hence, to less trust. Inequality makes us less likely to believe that what affects you also affects me. And the belief that inequality stems from unfairness leads us to be both more pessimistic and less connected to those who have benefited from the stacked system.7 As inequality has risen, Americans have become more pessimistic about the future. The sharpest effects are for the beliefs that the lot of the average person has gotten worse and that “it is not fair to bring a child into the world” anymore, according o the General Social Survey. Both of these measures falls by 30 percent as inequality has risen from 1972 to 1994. These results hold no matter which measure of inequality I use: the rising Gini index, the falling shares of income for the bottom four quintiles and the increasing portions of national income for the top 20 percent and top 5 percent. The effects are somewhat smaller for three other measures: Rising inequality has led to drops of 10 percent in the belief that that there are no limits to growth in the United States and to five percent in the arguments that Americans can solve
13 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
all of their problems if they set their minds to it and to disagreement that success in life is outside our control. These are core elements of the American sense of optimism—and, while Americans still adhere to most of them, they are not as optimistic as they once were. Some of the drop in the latter three measures may reflect the time frames of the surveys: The “lot of the average person” and “not fair to bring a child” questions are from the General Social Survey and were asked from 1972 to 1994; the other three questions were asked by Pew from 1987 to 2009. There was a slightly larger rate of increase in inequality in the former period compared to the latter.8 Class doesn’t seem to matter much in shaping the impact of inequality on optimism. Rising income gaps have similar effects on whether it is fair to bring a child into the world. Between 25-and 30 percent of upper, middle, and working class people are more likely to say that it is unfair to bring a child into the world in 1994 compared to 1973, while the drop is only about 20 percent for people who identify with the lower class The only measure of optimism for which I could estimate class differences over time is the lot of the average person (discussed above): I found only modest correlations for this measure with inequality for the middle class and working classes, and weaker effects for the upper and lower classes. The smaller impacts reflect the greater optimism of the upper class (only 55 percent are pessimistic) and the strong pessimism of lower class respondents, where between 73 and 86 percent agreed that the “lot of the average person is getting worse.” Inequality makes people less optimistic about the future, regardless of class. Optimism is at the root of social cohesion. It leads us to look at people who are different from ourselves as part of our “moral community” rather than as threats to our culture and wellbeing. Rising inequality leads to social fragmentation, both through increased pessimism and by fraying the bonds that reach across class lines. Growing income gaps, I argued above, leads to declines in generalized trust of more than 20 percent. As inequality has risen by each measure, we are about 10 percent less likely in 2009 than in 1987 to say that we have a lot in common with people of other races and are more likely to agree that we have gone too far in granting equal rights—and that the rich have gotten richer as the poor have gotten poorer. Since these questions come from Pew surveys, I cannot estimate perceptions by class.
14 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Distrusting institutions Americans have always distrusted authority, so distrust in government is not surprising. As inequality has increased, so have beliefs that the views of ordinary citizens don’t matter, rising from a third of the public in the 1960s to over 70 percent in 2008. Trust in government has also fallen, from three-quarters of Americans in 1958 to 22 percent in 2010, although the decline in confidence in Washington is less uniform from the American National Election.9 More than 50 years ago the political scientist E.E. Schattschneider bemoaned the lack of influence of ordinary Americans: “The flaw in the pluralist heaven is that the heavenly chorus sings with a strong upper-class accent. Probably about 90 percent of Americans cannot get into the pressure system...The system is skewed, loaded, and unbalanced in favor of a fraction of a minority.”10 And there is evidence that Schattschneider was correct: The wealthy participate much more in political life than do those with fewer financial resources and communications skills.11 Policy makers respond to the opinions of upper-income constituents, not to those with fewer resources.12 The views of upper-income Americans are far more conservative than those in the middle and lower classes , so there is little support for programs that would reduce inequality. As gaps in income increase, public opinion moves in a more conservative direction, toward even less support for redistribution—even though there is at best mixed evidence that greater inequality leads to lower participation by the poor and greater activity by the rich, with the most sophisticated analyses showing no support for this thesis.13 Political leaders respond to the upper classes not simply because these citizens participate more or are more likely to communicate their views. They also give money to candidates. Presidential and congressional election campaigns may cost up to or even more than—$3 billion in 2012. While a quarter of (self-identified) members of the upper class say that they have donated to a campaign, only 11 percent of the middle class and 6 percent of the working class say that they have given money to candidates or parties, according to data are from the American National Election Studies. It is no surprise that many Americans might agree with the tongue-in-cheek skepticism of Rep. Barney Frank (D, MA) who said that politicians “are the only people I know who are expected to take large amounts of money from perfect strangers and
15 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
have it have no effect on our behavior.”14 While the share of working class citizens who have contributed to campaigns has fluctuated around six percent over time, the percentage of upper class survey respondents saying that they have made donations increased from 20 percent to 31 percent between 1956 and 2008. I presented results above that rising inequality leads to less trust in government. The growth in the income gap between 1968 and 2008 led to declines in confidence by 30 percent to -40 percent. Inequality had insignificant effects (for the survey data) on trust in government for the upper class, but led to drops in confidence for the other three classes. The ANES, the GSS, and Pew all asked if people agreed that “public officials don’t care what the average person thinks.” For the American National Election Studies data, as inequality rose Americans became 35 percent to 40 percent more likely to believe that officials don’t care. The results were somewhat smaller (20 percent to 25 percent) for the General Social Survey data, and even lower for the Pew Values Surveys data (about 5 percent to 10 percent). The differences arise from the longer time series of the American National Election Studies surveys, from 1968 to 2008, compared to the General Social Survey (1973-1994) and Pew Values Surveys (1987-2008). Inequality matters for all classes for the ANES series, especially for the middle class, where the change in income gap led to a decline in perceptions of responsiveness of up to 40 percent. The shorter General Social Survey time series leads to smaller changes, about 20 percent to 25 percent for the middle, working, and lower classes but only to insignificant drops for the upper class. Other measures of institutional confidence show more mixed results in the Pew Values Survey from 1987 to 2009. When asked if members of Congress pay attention to their constituents, Americans are about 10 percent to 15 percent less likely to see responsiveness by most measures. The exception is for the income share of the upper middle class (the fourth quintile), where the estimated effect is estimated at 70 percent.15 Americans are between five and 10 percent less likely to say that government benefits everyone and that voting gives you a say in politics as inequality has risen. But the rising income gap has had minimal effects on whether Americans believe they have no say in politics more generally, whether officials lose touch with people, and whether the strength of the United States lies in business. The small effects for these last three measures reflect the high confidence Americans have of their own efficacy that has not waned even as they believe that the “system” has become less responsive.
16 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Over 70 percent of Americans still believed (in 2009) that voting gives them a say, down from more than 80 percent in 1987, but still very high. In 1968, 41 percent of Americans held that they had no say in politics, rising only to 45 percent in 1996. And while 81 percent of respondents said that officials lost touch in 2009, 76 percent held this view at the start of the time series in 1987. There is only a one percent change (from 80 percent to 79 percent, clearly not significant) in the share of Americans who say that the strength of the country lies in business. So where we find small effects for inequality, we find small changes overall. When Americans have become more pessimistic about the responsiveness of their institutions, the effects of inequality are stronger
Yet Americans still expect a better future for themselves Americans have a distinctive view of equality. They accept inequality of results as long as there is equality of opportunity. Political scientist Jennifer Hochschild conducted in-depth interviews on fairness and found that “most respondents, rich and poor, endorse differentiation in the economic domain...[and] insist that productivity should dominate in decisions on wages.” In a 1975-77 survey, political scientists Herbert McCloskey and John Zaller found that more than two-thirds of Americans believed that “laziness is almost like a sin”: Americans hold that everyone should be able to advance on his or her own merits. Political scientists Benjamin I. Page and Lawrence R. Jacobs report that 58 percent of Americans in their national survey three decades later believe that large differences in pay are necessary.16 Americans still believe in the value of individual effort. They support individualism and believe that one can still achieve the American Dream despite growing economic inequality. These results stand out even as Americans increasingly say that they are finding it difficult to make ends meet and that hard work is no longer a guarantee of success. Economic mobility seems out of reach for many Americans—largely because it really has become more difficult to move up the economic ladder.17 Americans’ attitudes toward inequality do not follow a simple pattern: While most Americans believe that inequalities are too large and should be reduced, they are now and have long been skeptical of government’s ability to make the appropriate corrections.18
17 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Jeff Faux of the Economic Policy Institute argues: People are worried about their jobs and income, and majorities think that the next generation will be worse off than this one. Yet polls show that they have faith that they, personally, and their kids will be OK, which reinforces the belief that government is irrelevant to the future.19 Lower-income people are substantially less likely to believe that they control their own future, but a majority of those at the bottom are still relatively optimistic—and no less so today than 25 years ago.20 While such views may seem— indeed, are—unrealistic, they have persisted over long periods of time even as Americans have lost faith in the country’s economic future, have become less connected to their fellow citizens, and see their political leaders as not paying heed to what people want. Inequality largely doesn’t matter (is not statistically significant) for believing that hard work is admirable, that many people won’t work hard, that you can run your life the way you wish, that you can’t make ends meet, that you can carry out life plans, and that you are satisfied with your financial situation.21 The only measure of personal optimism that are shaped by inequality are the perception that hard work leads to success. Yet inequality matters for whether hard work leads to success for the upper and lower classes. For most Americans—the middle and working classes—and for the full sample, inequality does not lead to a more jaundiced view of the benefits of hard work. But for the working and lower classes, there is a strong negative correlation between the perceived benefits of hard work and rising inequality. Yet, inequality doesn’t affect just the working and lower classes. As the income shares of the middle—and even the upper-middle—classes declined, optimism for the nation’s future, social cohesion, and confidence in our institutions have all fallen. As epidemiologists Richard Wilkinson and Kate Pickett argue, a plunging tide sinks most (if not) all boats.22 Yet, this is not a collective sense of despair, a belief that we are all in this together, as in the Great Depression. Quite the contrary. We are moving toward a more zero-sum society, which is reflected in the growing polarization in American political and social life. As we remain optimistic about our own personal life, we deflect accountability away from ourselves and toward others—people who are different from ourselves and our political system. The social and political polariza-
18 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
tion makes it difficult to find common ground, either between political parties or with our fellow Americans, on how—or even whether—to tackle inequality. Unless we do, our social fabric may become even more fragile.
Pessimism and investment When things don’t look so bright, do people limit their economic activity? Clearly, if incomes fall, even relatively, people might become more wary of making large investments. Optimists should be more likely to take the risks involved in investment. Pessimists should seek to preserve their resources—and are likely to believe that the value of their investments may be as likely (or more so) to go down as to increase. Economists Manju Puri and David T. Robinson report survey results indicating that more optimistic people put greater effort into their work and invest and save more.23 Economists Christopher D. Carroll, Jeffrey C. Fuhrer, and David W. Wilcox, Daron Acemoglu and Andrew Scott, and Stephanie Dees and Pedro S. Brinca all show that people are more likely to consume more when they have optimistic views of the economy.24 People are also more likely to make investments if they are sanguine about the economic future.25 Does optimism lead to investments in housing or stocks I investigate these is links using available aggregate data, sparse though it is, on business formed per capita, business failures per capita, houses sold per capita, and consumer sentiment on the economy. I also examine two surveys, a Gallup/UBS poll of people with savings and investments worth at least $10,000 in 1998, and of the larger public in the Pew Economic Mobility Study of 2009.26 The aggregate measures provide little support for the argument that increasing pessimism leads people to refrain from investing. People are more likely to buy new houses when consumer sentiment is strong, but new businesses are less likely to be formed when the public is upbeat—and existing businesses seem strangely more likely to fail. Consumer sentiment is modestly related to the beliefs that you can get ahead by hard work and that the lot of the average person getting worse. New businesses are more likely to be started when trust is low and when people believe that officials don’t care and when government is run by a few big interests—but businesses are also more likely to fail under the same conditions. Houses sold per capita are higher when trust is low, which is counterintuitive. A more reasonable conclusion is that optimism and pessimism may not drive busi-
19 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
ness starts or failures or housing sold—or even consumer sentiment. Instead, each of these economic indicators reflects real economic conditions. Business failures and starts are almost perfectly correlated. Business starts are more likely when gross domestic product (in real dollars) is highest, but so are failures. Houses are also more likely to be sold when income is strongest. And consumer sentiment is strongest when the change in gross domestic product from the previous year is highest. The aggregate data do not suggest a strong link between pessimism and investment decisions; the survey data tell a more nuanced story. The Gallup/UBS survey of investors asked about whether respondents would start businesses when they retire, whether they have goals for retirement and if those goals are written, and whether respondents intend to start a business. There are four measures of optimism and they have mixed effects on investment plans. Investors who are optimistic about their own income are more likely to have retirement plans and to start new businesses in the near future. Seeing a bright immediate future for yourself, doesn’t lead to longer term plans for starting a business or having written goals. No other indicator of optimism matters for any of the four investment measures—not overall optimism for the economy or expectations about the stock market or one’s own portfolio. Since this is a survey of investors, we should not regard these results as conclusive. The best predictor of investment plans is not optimism, education, or even income. It is age: Older inventors are more likely to say that they will start a business either soon or when they retire and to have written goals—but are less likely to have a retirement goal overall. The Pew Economic Mobility Survey of 2009 asked a cross-section of Americans if they own stock or own a home. And here we see strong evidence of optimism. People who are optimistic about their own economic situation are substantially more likely to own both stocks and their own home. Believing that it has been easier for you to achieve the American Dream than it was for your parents also leads to both more investments and home ownership. Yet Americans seem to compare the present with the past more than with the future. If you believe that your children will have a better life than you do, you will be more likely to own a home, but not necessarily to own stocks. Believing that your children will earn more than you do doesn’t matter for either invest-
20 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
ment decision. In one important sense, Americans champion the idea of economic opportunity if you put out enough effort: You don’t need to believe that you must be born into a wealthy family to invest in either stocks or real estate. Yet ambition matters, at least for owning a home. For real estate investment, however, both the belief that you must be ambitious and that your children will have a better life matter.27 Optimism seems to matter more for home ownership, where four of the measures are significant, than for holding stocks, where only the ease of the American Dream matters. Owning a home seems to be the fulfillment of the American Dream. Some core elements of optimism play a key role in home ownership: belief that the dream has been easier for you than for your parents, that life will be better for the next generation, that ambition matters for economic mobility but that inherited status doesn’t. The Pew survey doesn’t have a question on income, but education is at least in part a surrogate for economic status—and it matters a lot more for owning stock than for home ownership. One’s personal economic situation matters twice as much for stocks as it does for homes—so home ownership is far more aspirational than are market investments. Aggregate trends in home ownership don’t seem strongly based on optimism. Investors do not concern themselves with aspirations since people with market stakes have already achieved at least part of the American Dream. But home ownership has long been linked with the Dream and is a key part of the hopes of Middle America.28 Historically, the level of home ownership in the United States has increased dramatically over time—from under half of the population at the turn of the 20th century to two-thirds of Americans by 2000. Yet only a quarter of people who moved from one residence to another in 2009 were new homeowners, compared to 35 percent a decade earlier, revealing that the dream of owning your own home has become less attainable. While the American Dream of owning your own residence is slipping away, 84 percent of Americans still aspire to buying a home.29 While the American Dream may be more elusive than in the past, it is difficult to trace either home ownership—or especially the desire to buy your own residence—to rising income inequality. Home ownership rates continued to rise even as pessimism (and inequality) grew. Pessimism does not seem to dampen the formation of new businesses. Optimists may buy more houses, but this tendency seems likely to hold in good times and bad.
21 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Conclusion Why, then, should we care about pessimism and inequality? Increasing pessimism may have negative effects on other outcomes that I have not covered. Yet inequality and pessimism have other deleterious effects. First, growing disparities in income lead to more pessimism, less faith in our institutions, and less social cohesion. Greater inequality—and the belief that some people have too much power, that they exercise it unfairly, and that ordinary folks have no say in our political life—does not demobilize Americans. Rates of participation are not uniformly down—turnout increased in the 2008 presidential election compared to other recent contests. Nor do rates of political activity track trends in either inequality or pessimism. Pessimism and inequality are troublesome because they lead to less social cohesion, especially to less trust in others who are different from ourselves, even more than they lead to negative economic consequences. In our social and political life this leads to more conflict, especially more polarization. American political and social life has become more polarized on multiple dimensions—partisan, religious, and social and this is largely attributable to rising inequality.30 Increasing polarization makes it difficult to make public policy.31 People on opposite sides of issues see each other as evil. Conservative blogger Michelle Malkin has rebuked Republican Presidential nominee Mitt Romney for a “disastrous, bend-over bipartisanship” when he refuses to say that President Obama is “not a nice man.” The polarization that stems from high inequality and low trust in people of different backgrounds or even political views makes it close to impossible to do anything about the economic and social consequences of inequality—as Center for American Progress economists Heather Boushey and Adam S. Hersh,32 and Wilkinson and Pickett, among others, have detailed.
22 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Inequality feeds pessimism. We are twice as likely to say that our leaders don’t care what we think compared to 50 years ago, and we have increasing evidence that they pay more heed to those with greater resources. As the gap between the rich and the poor increases—and the rich put additional resources to ensure that they prevail, pessimism becomes more rational. The inevitable result is even greater social tensions and to more inequality. A disjunction rises between the American can-do spirit that individual effort will lead to a prosperous life for you and your children and the reality that economic mobility is declining. Can Americans maintain their persistent optimism about their own success? If not, the development of a permanent underclass might lead to even greater social conflict.
23 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Appendix The measures of income inequality The income distribution data point to a sharp increase over time in income inequality. The most telling part of the story is that the Gini index and the income shares are almost perfectly correlated with each other, either negatively (for the first four quintiles) or positively (for the top five percent and the top quintile, which is correlated with the Gini index at r = 0.9997). Only income shares for the fourth quintile depart at all from a simple linear downward trend, but not by much. Income shares have fallen for each group except for the top quintile and the top five percent. Now most people, even those who pay considerable attention to public affairs— don’t follow economic trend data. So is it reasonable to use such indicators as a predictor of pessimism, social cohesion, and confidence in government? Political scientists Nathan J. Kelly and Peter K. Enns report that the public understands increasing inequality quite well,33 so economic data are a good proxy for how people perceive economic disparities.
Data selection and analysis I use survey data from the American National Election Studies and the General Social Surveys,the two main academic surveys in the United States, as well as the Pew Values Survey from the Pew Research Center for The People and The Press. Pew’s surveys are widely respected and used in the academic community and the chief researchers are all widely respected political scientists. The ANES surveys began in 1948 but since the inequality data startbeing collected in 1967, I can only use surveys from 1968. The surveys are conducted biennially but there was no survey in 2006. The ANES cumulative file extends to 2008. However, not all questions have been asked during the entire period The generalized trust question, for example, was asked in 1968, 1972, 1974, and 1976 and then not again until 1992;
24 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
it has been asked every two years since. The GSS began in 1972 and was conducted annually through 1994 (except in 1979 and 1981), after which it has been conducted biennially. The GSS cumulative file extends through 20120. Not every question was asked each year; the two key measures of pessimism were only asked from 1972 to 1994. They are “In spite of what some people say, the lot (situation/condition) of the average man is getting worse “ and “It is hardly fair to bring a child into the world with the way things look for the future.” In The Moral Foundations of Trust, I showed that these are good measures of long-term optimism and that they are strongly related to generalized trust (social cohesion).34 The Pew Values Survey began in 1987 and has been conducted annually to the present. But the cumulative file publicly available at this time extends only to 2009. Again, not every question is asked every year, but the continuity is greater for this survey than others. Nevertheless, the Pew surveys began after inequality had begun its steep rise in the United States so the estimated effects (see below) are generally smaller than those for the ANES and GSS. Many of the measures are dichotomies; I recoded other variables into dichotomies to derive comparable effects (see below) I analyze these survey data through probit analysis, which is appropriate for binary data. Probit coefficients are not linear, so they have no clear interpretation as regression coefficients do. The standard approach in political science and sociology is to estimate what political scientists Steven J. Rosenstone and J. Mark Hansen call “effects.”35 Probit estimates can be readily converted into the probability of a positive outcome (trusting others, being optimistic). To determine the “effect” of a variable, we recode each case to its minimum value (or another value of interest) and then estimate the probability letting each of the other variables take their original values. We then recode the predictor to its maximum value and obtain the probability, again setting the other variables at their original values. The difference in probabilities between the second and the first estimates is the “effect.” In simpler language, consider the Gini index as the predictor of interest and trust as the dependent variable. For the ANES data, the minimum Gini index in the database is 0.391 in 1971 and the maximum is 0.466 in 2004 and 2008. A simple bivariate probit predicting trust from the Gini index gives an effect of -0.087, indicating that as the Gini rose from 0.391 to 0.466, trust fell by 8.7 percent. During the period covered by the ANES data, trust fell from 56 percent (1968) to 30
25 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
percent (2008), so the inequality-induced decline of 8.7 percent is a third of the total fall in trust. These change in probabilities (here it would be written as -0.087) are directly comparable across variables and are easy to understand. I estimated the effects through a Stata program I wrote that is available upon requestI present a sample probit in Table 1 . A full listing of all the effects can be found in Table 2. TABLE 1
Probit of trust (General Social Survey) Change in Pessimism Due to Increase in . . Variable
* p < .10 ** p < .05 *** p < .01 **** p < .0001 (all tests one-tailed) + Effect estimated at 18 and 75 years old; ++ Coefficient in the wrong direction Estimated R2 = .256 -2*Log Likelihood Ratio = 28209.18 N = 24066 Percent predicted correctly: 66.7 (model) 59.5 (null)
How changes in different income inequality measures affect pessimism Poll Question
Share of Income Going to Bottom Fifth
Share of Income Going to Second Fifth
Share of Income Going to Middle Fifth
Share of Income Going to Fourth Fifth
Share of Income Going to Top Fifth
Share of Income Going to Top 5 Percent
Belief that Next Generation's Life Will Be Better
Lot of the Average Person Will Be Worse
It's Not Fair to Bring a Child into the World
There are No Real Limits to Growth
Success is Outside My Control
Americans Can Solve Our Problems
Trust in Other People
Trust in Other People
The Rich Are Not Getting Richer
26 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Share of Income Going to Bottom Fifth
Share of Income Going to Second Fifth
Share of Income Going to Middle Fifth
Share of Income Going to Fourth Fifth
Equal Rights Have No Gone Too Far
Have Something in Common with Other Races
The Government Benefts All
Voting Gives You a Say in the Government
Strength of US Business Disagree
Americans Do Have a Say in Government
Trust in Government
Officials Pay Attention to Average Person
Officials Care About Average Person (1)
Officials Care About Average Person
Officials Care About Average Person
Officials Don't Lose Touch
Hard Work Will Get A Person Ahead
Hard Work Will Get A Person Ahead
Many People Won't Work Hard
I Admire Hard Work
I Can't Make Ends Meet
Satisied with Personal Finances
Share of Income Going to Top Fifth
Share of Income Going to Top 5 Percent
Can Run Life as Wished
Can Carry Out Plans in Life
Source: Author’s calculations
The probit in Table 1 is for generalized trust from the General Social Survey. The table shows the probit coefficient, its standard error, the MLE/SE (maximum likelihood ratio/standard error, which is approximately a z ratio), and the effect. The model includes the Gini Index; education; income; race (black) class identification for the lower, working, and middle classes (the upper class is omitted to prevent perfect collinearity); age; and the inflation and unemployment rates as contextual factors to ensure that the effects of inequality are not overestimated. Inequality (as here measured by the Gini index) strongly affects trust. As the Gini index rose from 0.401 in 1972 (the first year of the GSS) to 0.466 in 2008, I estimate that the likelihood that a respondent would agree that “most people can be trusted” fell by 19 percent. Trust fell by 23 percent (from 56 percent to 33 percent over these years), so inequality alone can account for over 80 percent of the decline of faith in other people.
27 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Of course, this is not a uni-causal world, and trust is shaped by other factors. Most important is education, which elsewhere I showed to be the most important socio-demographic factor underlying trust.36 A respondent with a graduate education is 64 percent more likely to trust people than one with just a grade school education. African Americans are 21 percent less likely to have faith in others. There is a modest effect for income with people becoming more trusting as income rises (not always reflected in other research), about 14 percent, while class matters much less: Working class people are six percent less likely to trust others, middle class respondents three percent more likely, and there is no effect for working class people. Older respondents (75 years old) are 21 percent more likely to be trusting compared to 18 year olds. The coefficient on unemployment is insignificant, while that for inflation is incorrectly signed. These results confirm the argument I made in The Moral Foundations of Trust, in which I argued that long-term optimism, not short-term economic trends, shapes trust. Overall, this analysis shows the strong impact of inequality on trust—more powerful than any factor other than education and almost equal to that of race.37 I repeat the analyses for selected measures by class—to test whether inequality matters more for the working class or the lower class than it does for the middle and upper classes. I do this in two ways. First, I run the same probits for each class. The ANES and the GSS include self-identified social class questions. The Pew data set does not have a question about subjective social class. The ANES and GSS subjective class categories did not overlap strongly enough to use income as a surrogate so I excluded the Pew surveys from my analyses using social strata.38 Second, for five measures that have questions asked repeatedly over time, I derive aggregate measures of optimism, social cohesion, institutional confidence, and personal expectations for each class. I could do this only for five survey questions: the lot of the average person, generalized trust, confidence in government (“How much of the time do you trust the government in Washington to do the right thing?” and “Do officials care what people like you think?”), and whether most people get ahead in life through hard work. Some of the measures for the upper class in the surveys must be taken cautiously since the sample sizes for each year are very small. I then employ the aggregate measures for each class in a simple correlation analysis (reported in the text). However, I also verified that the results are somewhat robust by estimating regressions including unemployment and inflation rates. Since the number of cases in the time seriesare often very small, including any additional variables would be hazardous. I present the correlations in Table 3.
28 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Correlations of aggregate optimism/pessimism and social cohesion measures with inequality Variable (number of cases, Gini/ income shares)
Trust (28, 29)
Trust in Government (39, 38)
Life Better for Next Generation (19 , 20)
People get ahead by hard work (21, 23)
Lot of the average person getting worse (15)
Not fair to bring a child into the world (15)
Success determined by factors outside our control (14)
Success determined by hard work (22, 13)
No limits to growth (16)
Americans can solve problems (14)
Gone too far with equal rights (14)
Have much in common with other races (9)
Can't make ends meet (14)
Satisfied with finances (14)
Strength of U/S. is in business (13)
Government benefits all (13)
Officials don't care about average person (36)
Officials lose touch (14)
Voting gives you a say in politics (12)
Have no say in politics (14)
Numbers in parentheses are numbers of cases for Ginis and income shares, respectively. Only one entry indicates same values for each.
Pessimism and investment This is a more technical discussion of the linkage between pessimism and investment. When things don’t look so bright, do people limit their economic activity? Clearly, if incomes fall, even relatively, people might become more wary of making large investments. I investigate this link using available aggregate data, sparse though it is, on business formed per capita, business failures per capita, houses sold per capita, and consumer sentiment on the economy. I also examine two surveys, a Gallup/UBS poll of people with savings and investments worth at least $10,000 in 1998 and of the larger public in the Pew Economic Mobility Study of 2009.
29 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
The aggregate measures provide little support for the argument that increasing pessimism leads people to refrain from investing. People are more likely to buy new houses when consumer sentiment is strong (r = 0.5), but new businesses are less likely to be formed when the public is upbeat—and existing businesses seem strangely more likely to fail (r = 0.289). Consumer sentiment is modestly related to the beliefs that you can get ahead by hard work and that the lot of the average person getting worse (r = 0.35 and -0.35). New businesses are more likely to be started when trust is low and when people believe that officials don’t care and when government is run by a few big interests—but businesses are also more likely to fail under the same conditions. Houses sold per capita are higher when trust is low (r = -0.5). A more reasonable conclusion is that optimism and pessimism may not drive business starts or failures or housing sold—or even consumer sentiment. Instead, each of these economic indicators reflect real economic conditions. Business failures and starts are almost perfectly correlated (r = 0.992): Business starts are more likely when gross domestic product (in real dollars) is highest, but so are failures (r = 0.917, 0.853).39 Houses are also more likely to be sold when income is strongest (r = 0.5). And consumer sentiment is strongest when the change in GDP from the previous year is highest. The aggregate data do not suggest a strong link between pessimism and investment decisions. The survey data tell a more nuanced story. The Gallup/UBS survey of investors asked about whether respondents would start businesses when they retire, whether they have goals for retirement and if those goals are written, and whether respondents intend to start a business. There are four measures of optimism and they have mixed effects on investment plans. Investors who are optimistic about their own income are more likely (by 25 percent) to have retirement plans and (by 32 percent) to start new businesses in the near future. (see Table 3). Seeing a bright immediate future for yourself doesn’t lead to longer term plans for starting a business or having written goals. No other indicator of optimism matters for any of the four investment measures—not overall optimism for the economy or expectations about the stock market or one’s own portfolio. Since this is a survey of investors, we should not regard these results as conclusive. The best predictor of investment plans is not optimism, education, or even income. It is age: Older inventors are more likely to say that they will start a business either soon or when they retire and to have written goals—but are less likely to have a retirement goal overall.
30 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Effects of measures of optimism and pessimism on investments Gallup UBS Wealth Management Research Survey of Investors 1998 Variable
Start Business When Retire
Have Goal for Retirement
Have Written Goal for Retirement
Plan to Start Business
Optimistic about economy next 12 months
Optimistic own income next 12 months
Optimistic stock market next 12 months
Expected rate of return on portfolio next 12 months
* Not statistically significant at p < .05 (one-tailed test)
The Pew Economic Mobility Survey of 2009 asked a cross-section of Americans if they own stock or own a home. (see Table 4) And here we see strong evidence of optimism. People who are optimistic about their own economic situation are substantially more likely to own both stocks and their own home. Personal optimism matters more for owning stock than for owning a home (28 percent compared to 16 percent), but this may be because other aspects of optimism shape homeownership but not investment decisions. Believing that it has been easier for you to achieve the American Dream than it was for your parents also leads to both more investments and home ownership (by 16 percent and 12 percent. Believing that is important to be born rich to move up the economic ladder doesn’t make one more likely to own stock or to own a home. Saying that ambition leads to more mobility does not lead people to own stock, but it does boost home ownership by 18 percent. Yet Americans seem to compare the present with the past more than with the future. If you believe that your children will have a better life than you do, you will be more likely to own a home (by 21 percent), but not necessarily to own stocks. Believing that your children will earn more than you do doesn’t matter for either investment decision. For real estate investment, however, believing that you must be ambitious is associated with an 18 percent greater likelihood of owning your home.
31 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Effects of measures of optimism and pessimism on investments Pew Economic Mobility Study 2009 Own Stock
Own a Home
Personal economic situation evaluation
Easier for you to achieve American Dream than it was for your parents
How important to be born rich for mobility
How important to be ambitious for mobility
Life better for next generation **
One's own children will earn more **
* Not statistically significant at p < .05 (one-tailed test) ** Estimated in separate equations with 219 cases for owning stock and 218 cases for owning a home. All other variables in this table estimated in models with 1251 cases for owning stock and 1256 cases for owning a home. Other variables in the model are income, race (African-American), and Hispanic ethnicity.
Optimism seems to matter more for home ownership, where four of the measures are significant, than for holding stocks, where only the ease of the American Dream matters. Owning a home seems to be the fulfillment of the Dream. Some core elements of optimism play a key role in home ownership: belief that the dream has been easier for you than for your parents, that life will be better for the next generation, that ambition matters for economic mobility but that inherited status doesn’t. The Pew survey doesn’t have a question on income, but education is at least in part a surrogate for economic status—and it matters a lot more for owning stock than for home ownership. One’s personal economic situation matters twice as much for stocks as it does for homes—so home ownership is far more aspirational than are market investments.
32 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
About the author
Eric M. Uslaner is a professor of Government and Politics at the Department of Government and Politics, University of Maryland–College Park, and a Senior Research Fellow, Center for American Law and Political Science at the Southwest University of Political Science and Law, Chongqing, China. Professor Uslaner is also an Honorary Professor of Political Science, University of Aarhus (Denmark)
Acknowledgements I am grateful for the many comments from David Madland and the research support from the Center for American Progress, especially from Matt Separa. I am also grateful to the institutional support of the University of Maryland—College Park, especially from the Graduate School for a Research and Scholarship Award.
33 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
Endnotes 1 Herbert Croly, The Promise of American Life. Ed. by Arthur M. Schlesinger, Jr. (Cambridge, MA: Belknap, 1965); Henry Steele Commager, The American Mind (New Haven: Yale University Press, 1950); David Potter, People of Plenty (Chicago: University of Chicago Press, 1954). 2 AUTHOR, “Whatever Happened?”, The Economist, September 12, 1987, p. 11-12; Adam B. Seligman, The Problem of Trust (Princeton: Princeton University Press, 2000). 3 Justina Fischer and Benno Torgler, “Social Capital and Relative Income Concerns: Evidence from 26 Countries.” Research Paper Series No. 38 (Thurgau Institute of Economics and Department of Economics, University of Konstanz: October 2006), available at http://kops.ub.uni-konstanz.de/bitstream/handle/ urn:nbn:de:bsz:352-opus-74683/60_No38_08_10_TWI_ RPS_Fischer_Torgler.pdf?sequence=1.; and Christian Bjornskov and others, “On the Relation Between Income Inequality and Happiness: Do Fairness Perceptions Matter?” (Center for European Development and Economic Development Research: 2009), available at http://www.kof.ethz.ch/en/publications/p/kof-workingpapers/245/. 4 The data for the United States come from http://www. census.gov/hhes/www/income/histinc/ineqtoc.html (last accessed November 2012). The cross-national data come from the World Income Inequality Database v.20c, United Nations University World Institute for Development Economic Research, available at http:// www.wider.unu.edu/research/Database/en_GB/database/ (last accessed November 2012). 5 Eric M. Uslaner, The Moral Foundations of Trust (New York: Cambridge University Press, 2002), chs. 2, 4, 6, 8. 6 Eric M. Uslaner, The Moral Foundations of Trust (New York: Cambridge University Press, 2002). 7 Uslaner, The Moral Foundations of Trust.; Christian Bjornskov, “The Determinants of Trust,” School of Business, Aarhus University, available at http://www.ratio.se/pdf/ wp/cb_trust.pdf, March 22, 2005. 8 The Gini index rose from .401 to .456 from 1972 to 1994 and from .426 to .468 from 1987 to 2009. The income share of the top 20 percent rose from 43.9 percent to 49.1 percent across the GSS time frame and from 46.2 percent to 50.2 for the Pew time period. 9 Data are from the 1948-2008 cumulative files of the American National Election Studies: “Data Center,” available at www.electionstudies.org (last accessed November 2012). 10 E.E. Schattschneider, The Semisovereign People (New York: Holt, Rinehart, and Winston, 1960).. 11 Sidney Verba, Kay Lehman Schlozman, and Henry E. Brady, Voice and Equality: Civic Voluntarism in American Politics (Cambridge, MA: Harvard University Press, 1995). 12 Larry M. Bartels, Unequal Democracy (Princeton, NJ: Princeton University Press, 2008).; and Martin Gilens, “Inequality and Democratic Responsiveness,” Public Opinion Quarterly 69 (5) (2005): 778-96. Bartels finds that policy also reflects middle class opinions, while Gilens finds no impact of the views of these citizens.
13 Verba, Schlozman, and Brady, Voice and Equality.; Nathan J. Kelly and Peter K. Enns, “Inequality and the Dynamics of Public Opinion: The Self-Reinforcing Link Between Economic Inequality and Mass Preferences,” American Journal of Political Science 54 (4) (2010): 869. On the lack of a connection between participation rates and inequality, see Henry E. Brady, “An Analytical Perspective on Participatory Inequality and Income Inequality.” In Kathryn M. Neckerman, ed., Social Inequality (New York: Russell Sage Foundation, 2004), and Jan E. Leighley and Jonathan Nagler, Who Votes Now? Unpublished manuscript, American University, 2012. For an alternative argument, see, James K. Galbraith, Inequality and Instability. (Oxford: Oxford University Press, 2012), 160, and James C. Garand and N. Kim Nguyen, “The Effects of Income Inequality on Political Attitudes and Behavior,” presented at the 2007 Annual Meeting of the Midwest Political Science Association, Chicago, April. 14 Quoted on Frank’s website, available at http://barneyfrank.ngpvanhost.com/quotes (last accessed May 24, 2012). 15 This estimate is too high to be taken as accurate, since its first measurement in 1968 (when income shares are also available) was 34 percent, and it fell to 17 percent in 1980, the last survey in which the question was asked. 16 Jennifer Hochschild, What’s Fair: American Beliefs About Distributive Justice (Cambridge, MA: Harvard University Press, 1981).; Herbert McCloskey and John Zaller, The American Ethos (Cambridge, MA: Harvard University Press, 1984).; Benjamin I. Page and Lawrence R. Jacobs, Class War? What Americans Really Think about Economic Inequality (Chicago: University of Chicago Press). 17 Katharine Bradbury, “Trends in U.S. Family Income Mobility, 1969–2006.” Working Paper 11-10 (Federal Reserve Bank of Boston, 2011), available at http://www. bostonfed.org/economic/wp/wp2011/wp1110.pdf.. 18 Page and Shapiro, Class War?. 19 Jeff Faux, “Who will save the middle class?”, Economic Policy Institute, May 30, 2012, available at http://www. epi.org/publication/who-will-save-middle-class/. 20 Andrew Kohut and others, Partisan Polarization Surges in Bush, Obama Years: Trends in Values: 1987-2012. (Washington: The Pew Research Center For The People and The Press, 2012), available at http://www.peoplepress.org/values/. 21 There are sporadic significant coefficients across measures of inequality and class, but they follow no clear pattern. While some of the “effects” for carrying out plans may seem stronger, they are not consistently signed and they are not statistically significant. 22 Richard Wilkinson and Kate Pickett, The Spirit Level: Why Equality is Better for Everyone (London: Penguin, 2010). 23 Manju Puri and David T. Robinson, “Optimism and Economic Choice,” Journal of Financial Economics 86 (1) (2007): 71-99. 24 Daron Acemoglu and Andrew Scott, “Consumer Confidence and Rational Expectations: Are Agents’ Beliefs Consistent with the Theory?”, Economic Journal 104 (NO.) (1994): 1-19, available at http://www.jstor.org/disc
34 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
over/10.2307/2234671?uid=3739584&uid=2129&uid=2 &uid=70&uid=4&uid=3739256&sid=47698776411847.; Christopher D. Carroll, Jeffrey C. Fuhrer, and David W. Wilcox, “Does Consumer Sentiment Forecast Household Spending? If So, Why?”American Economic Review 84 (5) (1994): 1397-408, available at http://www.jstor.org/disc over/10.2307/2117779?uid=3739584&uid=2129&uid=2 &uid=70&uid=4&uid=3739256&sid=47698776411847.; Stephanie Dees and Pedro S. Brinca, “Consumer Confidence as a Predictor of Consumption Spending: Evidence for the United States and the Euro Area.” Working Paper Series No. 1349, European Central Bank, June 2011, available at http://www.ecb.europa.eu/pub/ pdf/scpwps/ecbwp1349.pdf. 25 Bruce D. Grundy and Li Hui. “Investor Sentiment, Executive Compensation, and Corporate Investment,” Journal of Banking and Finance 34 (2010): 2439-49, available at http://papers.ssrn.com/sol3/papers.cfm?abstract_ id=1442922.; Luca Rigotti, Matthew Ryan, and Rhema Vaithianathan, “Optimism and Firm Formation,” Economic Theory 46 (1) (2011), 1-38, available at http:// ideas.repec.org/a/spr/joecth/v46y2011i1p1-38.html. 26 The housing data for 1966-2009 come from “Historical Data,” available at http://www.census.gov/construction/ nrc/historical_data/ (last accessed April 2, 2012). Businesses started and failure data from 1966-1997 come from www.gpoaccess.gov/usbudget/fy03/sheets/b96. xls (last accessed November 2012) . Adjustments for population size come from population data at “Population,” available at http://www.census.gov/compendia/ statab/cats/population.html (last accessed April 2, 2012). The University of Michigan Survey of Consumer Sentiment quarterly data from 1966-2009 are available at http://www.sca.isr.umich.edu/data-archive/mine. php?kind=EXCL (last accessed April 2, 2012). I aggregated the quarterly figures to annual estimates. The Gallup/UBS data are available at the Roper Center; for a description see: http://webapps.ropercenter.uconn. edu/CFIDE/cf/action/catalog/abstract.cfm?label=&key word=USAIPOUBS1998+INVEST11&fromDate=&toDat e=&organization=Any&type=&keywordOptions=1&st art=1&id=&exclude=&excludeOptions=1&topic=Any& sortBy=DESC&archno=USAIPOUBS1998-INVEST11&ab stract=abstract&x=34&y=3 (last accessed April 2, 2012: login required). The Pew Economic Mobility Survey is available at http://www.pewstates.org/research/datavisualizations/poll-results-economic-mobility-and-theamerican-dream-85899374448 (last accessed April 5, 2012). The data were provided by Erin Currier of the Economic Mobility Project. 27 Both “life will be better for the next generation” and “one’s children will earn more” are asked to only half of the sample. This missing data leaves a sharply truncated sample when including these questions. So I estimate models with the other four optimism questions for larger samples. The coefficients for the next generation variables are from a model with far fewer cases. 28 Morgan Brennan, “Is The American Dream Of Home Ownership Dead? Not Really,” Forbes, February 11, 2011, available at http://www.forbes.com/sites/morganbrennan/2011/02/11/is-the-american-dream-of-homeownership-dead-not-really/. 29 United States Bureau of the Census, “Historical Census of Housing Tables,” available at http://www.census.
gov/hhes/www/housing/census/historic/owner.html (last accessed November 2012).; Stuart A. Gabriel and Stuart S. Rosenthal, “Homeownership Boom and Bust 2000 to 2009: Where Will the Homeownership Rate Go From Here?” (Research Institute for Housing America: 2011), available at http://www.anderson.ucla.edu/ Documents/areas/ctr/ziman/Research_RIHA_Gabriel_Rosenthal_Report7.14.11.pdf.;Woodrow Wilson International Center for Scholars, Program on America and the Global Economy, New National Poll: Americans Still Want to be Homeowners, May 22, 2012, available at http://www.wilsoncenter.org/article/new-national-pollamericans-still-want-to-be-homeowners. 30 Uslaner, The Decline of Comity in Congress, chs. 2, 4-6; Uslaner, “Is the Senate More Civil Than the House?” in Burdett Loomis, ed., Esteemed Colleagues: Civility and Deliberation in the Senate (Washington: Brookings Institution, 2000).; McCarty, Poole, and Rosenthal, Polarized Politics. 31 Uslaner, The Decline of Comity in Congress, ch. 6. 32 Heather Boushey and Adam S. Hersch, “The American Middle Class, Income Inequality, and the Strength of Our Economy: New Evidence in Economics” (Washington: Center for American Progress: 2012), available at http://www.americanprogress.org/issues/2012/05/pdf/ middleclass_growth.pdf. 33 Kelly and Enns, “Inequality and the Dynamics of Public Opinion.” 34 Uslaner, The Moral Foundations of Trust, ch. 4. A frequently used question, which I used in my book, is whether the future will be better for your children than for yourself. However, this question showed a strange time series (increasing over time, increasing as inequality rose) and individual level and aggregate relationships (more positive for the working class than for other classes and for minorities and low-income respondents) that defied logic. An inquiry about coding to the GSS director did not resolve these anomalies. 35 Steven J. Rosenstone and John Mark Hansen, Mobilization, Participation, and Democracy in America. New York, Free Press, 1993. 36 Uslaner, The Moral Foundations of Trust, 98-104. 37 Compare these results with Uslaner, The Moral Foundations of Trust, ch. 4. I estimated the effects of age between 18 and 75 years old. Surveys almost never include respondents under 18 years old. There may be a handful of respondents above 75, but there will be few such respondents and estimating effects beyond 75 years old may lead to misleading results. The probit estimation is for respondents of all age. However, the effects are only estimated between 18 and 75. 38 The GSS codes four classes (lower, working, middle, upper). The ANES codes eight classes, but most are hybrid. “Lower class” is a volunteered response, chosen by only .4 percent of respondents. 39 This may reflect the fact that both measures are per capita figures. However, even without the adjustment for population size, the correlation is substantial ( r = .82).
35 Income Inequality in the United States Fuels Pessimism and Threatens Social Cohesion
The Center for American Progress is a nonpartisan research and educational institute dedicated to promoting a strong, just, and free America that ensures opportunity for all. We believe that Americans are bound together by a common commitment to these values and we aspire to ensure that our national policies reflect these values. We work to find progressive and pragmatic solutions to significant domestic and international problems and develop policy proposals that foster a government that is “of the people, by the people, and for the people.”
1333 H STREET, NW, 10TH FLOOR, WASHINGTON, DC 20005 • TEL: 202-682-1611 • FAX: 202-682-1867 • WWW.AMERICANPROGRESS.ORG