Control is good, trust is better? Trust and its implications on agency problems An empirical study of audit fees and management remuneration in Europe
BY York Philipp Wagner Mierendorffstr. 21 10589 Berlin [email protected]
Student-ID. 368467 International Business Administration Exchange (IBAEx) Berlin School of Economics and Law Department of Business and Economics Due date: September 12th 2016 SUPERVISED BY Prof. Dr. Nadja Jehle, Berlin School of Economics and Law Prof. Dr. Madeleine Janke, Berlin School of Economics and Law
Table of Contents List of Abbreviations...........................................................................................................................II List of Figures.....................................................................................................................................III 1. Introduction......................................................................................................................................1 2. Agency Theory.................................................................................................................................3 2.1 Basic Assumptions and Conclusions of Agency Theory...........................................................3 2.2 Monitoring Expenditures...........................................................................................................5 2.3 Audit Fees..................................................................................................................................6 2.3.1 Audit Pricing Influences....................................................................................................6 2.3.2 Disclosure Regulation .....................................................................................................10 2.4 Variable Management Remuneration......................................................................................11 2.4.1 Design and Applicability of Different Remuneration Systems........................................11 2.4.2 Accounting for Different Remuneration Systems............................................................13 2.4.3 Disclosure of Remuneration.............................................................................................15 3. Trust................................................................................................................................................16 3.1 Definition.................................................................................................................................16 3.2 How Trust is Measured............................................................................................................16 3.3 Trust and Trustworthiness.......................................................................................................18 3.4 Economic Effects of Trust.......................................................................................................20 3.5 Origins of Trust........................................................................................................................22 4. Methodology...................................................................................................................................25 4.1 General Purpose and Sample Selection...................................................................................25 4.2 Dependent and Independent Variables of the Model in Detail................................................27 4.2.1 Complete Model...............................................................................................................27 4.2.2 Audit Fees........................................................................................................................28 4.2.3 Remuneration ..................................................................................................................28 4.2.4 Assets...............................................................................................................................30 4.2.5 Revenue............................................................................................................................31 4.2.6 Employees........................................................................................................................31 4.2.7 Trust.................................................................................................................................32 4.2.8 Currency Conversion.......................................................................................................32 4.3 Data Assessment......................................................................................................................33 5. Results............................................................................................................................................34 6. Discussion.......................................................................................................................................35 7. Conclusion......................................................................................................................................38 Appendix............................................................................................................................................40 Bibliography.......................................................................................................................................45 Directory of Legal Documents...........................................................................................................63 Affidavit.............................................................................................................................................64
List of Abbreviations CEAOB
Committee of European Auditor Oversight Bodies
European Group of Auditors’ Oversight Bodies
European Securities and Markets Authority
Full time equivalent
Generally accepted accounting principles
Gross domestic product
Handelsgesetzbuch (German Commercial Code)
International Auditing and Assurance Standards Board
International Accounting Standards
International Financial Reporting Standards
International Standards on Auditing
New institutional economics
Public interest entity
World Values Survey
List of Figures Figure 1: Types of remuneration systems, their distinction by incentive and accounting aspects and the instruments they use.....................................................................................................................13 Figure 2: Trust and parking tickets per diplomat across different countries......................................20 Figure 3: Elements of remuneration considered in and excluded from in the empirical model.........30 Figure 4: Descriptive statistics for the sample obtained.....................................................................33 Figure 5: Regression coefficients for regression 1 (audit fees) and regression 2 (variable remuneration).....................................................................................................................................34 Figure 6: Results of regression 1 logarithmized.................................................................................35 Figure 7: Results of regression 2 logarithmized.................................................................................35
1. Introduction An old German saying, which is originally attributed to Wladimir Iljitsch Lenin, quotes: “Vertrauen ist gut, Kontrolle ist besser” (trust is good, control is better) (Klosa et al. 1998: 766). Agency Theory assumes that its actors follow this principle. The agent is opportunistic and selfish and the principal can therefore not trust her. Unable to oversee all actions the agent takes, due to a high workload or task complexity, the principal still needs to delegate tasks and substantial decision-making power to her agent. The control that is lost in this process is recaptured by the use of monitoring and incentive mechanisms. On the other hand the social capital literature continuously emphasizes the positive impact of mutual trust within societies. Numerous large-scale sociological studies show that trust levels greatly vary throughout the world. Examining more and less trustful societies, scholars find untold positive effects of trusting behavior at microand macroeconomic level. Examples of those effects are higher income growth rates, better tax compliance or less corruption in trustful environments. Trust shows a strong impact particularly with regard to economic exchanges that require credible commitment. If commitment of agents is not credible it can to a certain degree be replaced by control. This thesis assumes that trust can widdershins also replace control and that it hence has a positive impact on agency problems. It expects trust to decrease control and to hereby significantly lower the 'control costs' that arise in situations in which agents are employed. To test this assumption, it firstly further defines what is referred to as 'control costs' here (section 2.1 and 2.2). Falling back to Agency Theory, it argues that the (monitoring) expenses caused by audit fees and variable management remuneration are examples of these 'control costs'. It finds that both, audit fees and remuneration are subject to an ever more similar regulatory environment within the European Union (EU). With this knowledge, audit fees and variable management remuneration in 119 companies from four European countries with very different trust levels (France, Germany, Sweden, the United Kingdom) are then examined (section 4.3 and 5). This study finds that the proxies for 'control costs', audit fees and the variable fraction of
remuneration, are indeed higher in low-trust environments. Running two multivariate linear regressions, which take advantage of the harmonized regulatory environment within the EU, the paper concludes that trust apparently has a diminishing effect on control and the costs it causes. In order to make the research results in section 5 more comprehensive, the quantitative part of this work is accompanied by a very extensive literature review on audit fees (section 2.3), variable remuneration (section 2.4) and trust (section 3). The literature review covers threefold: economic theory, former empirical research and accounting aspects. Section 4 detailedly explains the design of the regression whereas section 6 discusses the findings and section 7 draws a final conclusion. Even though it is limited in scope, the work I present contributes to research inasmuch as it empirically analyzes the aforementioned relations of trust and agency costs for the first time. Determinants of audit fees and remuneration on the one hand and effects of trust on economic performance on the other hand have been extensively studied. The overlap of these fields has not really been in the focus of science, however. Although it has not received much attention, this research area is of great importance: We confront agency situations every day and they do not only cause costs in the sphere of multinational enterprises. By delegating tasks and responsibilities to others, non-commercial private actors apply agency and face agency costs as well. In an ever more diverse environment we are subject to increasing but possibly unjustified distrust within organizations and society (Mayer et al. 1995: 710) and this imposes overdrawn 'control costs' on us. Increasing audit and legal fees are just one indicator for this trend. This study hence sees substantial cost-saving potential in a “trust is better” approach to social and economic interaction and recommends further research on the issue of trust formation.
2. Agency Theory 2.1 Basic Assumptions and Conclusions of Agency Theory Agency Theory is part of the new institutional economics (NIE). NIE uses concepts from both, the earlier institutional economics and the neoclassical paradigm. From the “old” institutional economics, NIE differs inasmuch as it makes use of a set of neoclassical ideas, such as the self-interested, rational and utility-maximizing individual and subsequent distribution and efficiency assumptions (Bardmann 2011: 345). From neoclassical economics NIE differs on the other hand, as – just like the preceding institutional economics – it explicitly takes into account institutions as important factors in economic interaction (Bardmann 2011: 335). Ménard and Shirley (2008: 1) define institutions as “the written and unwritten rules, norms and constraints that humans devise to reduce uncertainty and control their environment”. According to their definition, institutions appear in the form of laws, norms of behavior and beliefs as well as in organizational arrangements, such as markets, firms and other contractual agreements. Institutions are not static. They represent the views of a large majorities and thus slowly change over time (Bachmann / Inkpen 2011: 285 f.). Agency Theory, which was strongly influenced by Michael Jensen and William Meckling, against this backdrop then analyses the relationship of different actors in the organizational sphere. It assumes that at some stage of growth of an enterprise the ownership role and the management function, which are initially both taken by one single individual (the owner-manager), need to be separated due to a high work load and task complexity (Jensen / Meckling 1983: 2 f.). The basic idea is that holders of tradable securities, the owners or principals of the company, at that point delegate the exertion of control to professional managers, the agents. The delegation of control tasks goes hand in hand with a transfer of decision making power (Marten et al. 2015: 49). Jensen and Meckling (1976: 308), as other scholars of NIE, presume self-interest and utility maximization for principals and agents. Especially the agents face a high incentive to cheat on their principals (Greif 1989: 865 f.), as the model also presupposes that the principals are not able to oversee and evaluate all actions the agents take (Blum et al. 2005: 155). Therefore 3
and since the interests of principals and agents frequently differ 1, it is highly probable that agents will not always act in the best interest of their principals (Marten et al. 2015: 49). As the agents have a fairly high degree of control and can make substantial decisions, the divergence of interests inside the firm imposes considerable costs on the principals and agents. These costs are known as agency costs and they can be split up into three parts: (1) monitoring expenditures by the principal, (2) bonding expenditures by the agent and (3) residual loss (Jensen / Meckling 1976: 308). Bonding expenditures (2) are costs which occur to the agents if they enter certain agreements in order to signal that their actions are in the best interest of the principal. In practice, these bonding expenditures could be potential contract penalties or external audits causing additional effort, which agents voluntarily accept (Jensen / Meckling 1976: 325). As the primary focus of this work lies on costs that shareholders, i.e. principals, directly bear, bonding costs will not be further discussed here, though. The residual loss (3) on the other hand are any costs that directly arise from the divergence of the principals and the agents interests: If the agents indeed use the firm-owned assets for private consumption or shirk their duties, for instance, this results in lower profits or share rate losses, which are costs the principals eventually have to bear (Marten et al. 2015: 51). The residual loss is always taken by the principals (Jensen / Meckling 1976: 308). Therefore, the principals are prompted to establish certain measures, which either monitor the work of the management or align the interests of the agents with their own goals and subsequently lower the residual loss (ibid.: 323; Schulz 2009: 77 f.). Monitoring instruments are the hiring of independent auditors or supervised restricted budgets, among others (ICAEW 2005: 6 f.). Interest-aligning control mech1 Security holders most probably have the goal of profit and equity maximization while the agents, maximizing their utility, might also try to minimize their own work effort or privately consume the firm-owned assets (Miller 2008: 352 f.). Obviously, those goals are conflicting. Also, agents might have a different attitude towards entrepreneurial risk as they just manage and do not own the firm (Beccera / Gupta 1999: 184). 4
anisms could in practice be variable management remuneration or the division of the company into profit centers (Blum et al. 2005: 155; Köster 2013: rec. 3). Commonly, monitoring and interest-alignment are both applied in combination (Miller 2008: 350). All the instruments, however, also create costs, which Agency Theory refers to as monitoring expenditures (1) (Jensen / Meckling 1976: 308). At this point it is important to note again that both, the monitoring expenditures and the residual loss are summands of the total agency costs. Hence, principals will only opt for any monitoring expenditure if it is lower than or equal to the the residual loss it eliminates (Jensen / Meckling 1976: 323 f.). Other scholars argue that the managerial labour market, making managers constantly worry about their reputation (Fama 1980), and the market for corporate control, making poorly managed firms an easy target to hostile takeovers (Manne 1965), also have a confining effect on the residual loss and thus on Agency Costs. These approaches will not be further examined here, though. A main argument of this paper further developed in section 4.1 is the statement, that populations with high levels of mutual trust suffer less agency costs, precisely lower monitoring expenditures than populations with low mutual trust. Because of this, section 2.2 further defines the central variable monitoring expenditures and discusses how it could be quantified. Sections 2.3 and 2.4 then examine two types of quantifiable monitoring expenditures, explain which factors influence them and outline how they are disclosed and accounted for. 2.2 Monitoring Expenditures In section one restricted budgets, independent auditors, the establishment of profit centers and variable management remuneration were named as examples of monitoring and incentive mechanisms. These measures all create monitoring expenditures, but not all of the expenditures can be quantified. Some of the expenditures can be quantified, but figures which help to estimate them are regularly not disclosed by companies (Baetge et al. 2014: 6). The only monitoring expenditures which are both, quantifiable and widely accessible and therefore of use for our model are firstly fees independent auditors receive for financial statement audits and secondly the variable part of the compensation of managers. As this study will detailedly outline in 2.3 and 2.4, those two expenditures are accounted for in bookkeeping departments and they 5
are compulsorily disclosed in financial reports, which are prepared in accordance with the International Financial Reporting Standards (IFRS) and published by European enterprises (IAS 24.17; Directive 2006/43/EC, art. 49). Therefore they serve well as proxies for monitoring expenditures. 2.3 Audit Fees 2.3.1 Audit Pricing Influences Due to the aforementioned owner-manager conflict, owners are assumed to suspect their agents of providing them with limited or even wrong information (ICAEW 2005: 8 f.). The purpose of audits is to assure that management provides a true and fair view of the company's financial performance and position in the financial statements (ICEAW 2005: 9). In order to verify wether the financial statements indeed provide a true and fair view of the financials, auditors consider the generally accepted accounting principles (GAAP) which their client is obliged to use (PWC 2013: 7). They check if the financial statements are prepared in accordance with those mandatory GAAP. As no GAAP can detailedly stipulate every possible recognition, measurement and disclosure companies could make, auditors on top of that need to use their professional judgement to give an audit opinion (Hachmeister 2008: 62). Unlike commonly expected, auditors are not generally obligated to assess and evaluate the economic condition of the audited company, however (Marten et al 2015: 19; Naumann 2008: 98). In contrast to other professions, auditors are not subject to fiscal regulation in the pricing of their services. Audit fees are determined in a free market process (Niehus 2008: 158). Following an invitation by their potential clients the audit firms in this process commonly make proposals, which among other things contour the scope, schedule and price of the audit, and the potential clients then agree or decline (AICPA 2015: 8). Two factors make the audit market distinct, though: On the one hand statutory audit regulation provides the market with constant demand (Marten et al. 2015: 37). On the other hand strict professional rules pose market entry barriers and limit competition among suppliers (Hachmeister 2008: 59 f.). For this reason, the price of a financial statement audit intuitively is first and foremost influenced by the scope of the work that needs to be conducted. The scope of 6
the work, in turn, in the first place depends on the size of the audited entity – the bigger the client, the higher the fee (Zerni 2012: 313). The size of companies can be defined in different ways. For my empirical model, I will simply fall back on the aforementioned Directive on Annual Financial Statements, also using total assets / liabilities, total revenues and employees as determinants of the company size. Another important factor influencing the scope of work is the regulatory environment of the audit: If the auditor is bound to vast professional standards, the workload will increase (Eilifsen / Willekens 2008: 10 f.) Equally, if the legal liability of the auditor in case of misstatements is high, the auditor will possibly invest more time to get to a sound audit opinion (Hachmeister 2008: 66). The European Union offers a very good experimental setting for this study as not only the financial reporting standards, but also professional auditor regulation is ever more similar in the member states: The Directive on Statutory Audit first allowed the European Commission (EC) to adopt international auditing standards, particularly the International Standards on Auditing (ISA) by the International Auditing and Assurance Standards Board (IAASB), for application in the EU in 2006 (Directive 2006/43/EC, art. 26). The ISA are an extensive set of professional standards, which give guidance for audit planning, audit evidence, audit reports and other parts of the statutory audit (Marten et al. 2015: 116). The enforcement of those standards is coordinated by the Committee of European Auditor Oversight Bodies (CEAOB)2, an EU-wide supranational umbrella organization that will also impel the homogenous application of those ISA (EY 2014: 5). The CEAOB works closely together with the European Securities and Markets Authority (ESMA) (ibid: 5 f.), an organization coordinating the enforcement of IFRS on the European level (ESMA 2016). One of ESMAs main objectives is the further convergence in the application of the IFRS (ESMA 2016: 6 ff.). Given all those European institutions3, one can assume that the auditors across the selected European states already now approach their audit assignments in a fairly similar manner, invest similar amounts of time and effort and will thus charge similar audit fees4. The legal liability of auditors in case of misstatements on the other hand still differs 2 The CEAOB succeeds the European Group of Auditors’ Oversight Bodies (EGAOB), which previously was responsible for auditor oversight and professional standards harmonization (EC 2016). 3 Note that 'institution' in this context again follows the definition given in section 2.1. 4 Of course, general price levels still significantly differ across the EU, this issue will however be taken into consideration in the later empirical model (refer to section 4.2.7). 7
across member states (Naumann 2008: 117).5 While in some member states the liability of the auditor to the audited company and third parties 6 is only of contractual nature, auditors in other states also face tort claims, which exposes them to the risk of additionally paying high damages apart from the contractual fines (EC 2001: 7). Of the countries examined France, Germany and the United Kingdom provide tort law for auditor liability to both, the audited companies and third parties (ibid.: 7). In Sweden, the nature of claims of the audited company against the auditor is contractual only, but third parties can also bring forward tortious charges (ibid.: 18). In France, Sweden and the United Kingdom, the audit firms and the signatories of the audit report can be legally charged, while in Germany even the assistants of the auditors can be held responsible for misstatements (ibid.: 11 f., 18 f.; § 323 (1) HGB). Financially, the liability is not limited in France, Sweden and the United Kingdom (EC 2001: 11, 18 f.). In Germany the maximum charge against audit firms is one million Euros for audits of non-listed companies and four million Euros for audits of listed companies (§ 323 (2) HGB). The limitation period varies between three years in France and ten years in Sweden (EC 2001: 12, 18). Weighting up these differences in auditor liability, one can eventually say that none of the examined regimes is particularly harsh or permissive with statutory auditors and that the different liability aspects thus should overall roughly counterbalance each other in their effects on audit fees. The liability issue, however, also has an indirect effect on the cost of audits: Like other enterprises, audit firms cover their risks with (partly mandatory) insurances (ibid.: 82). As the availability and coverage of those insurances is limited, the big audit networks have additionally built up so called captives (London Economics 2006: 92, 99). Captives are some form in-house insurance: member firms pay premiums into a mutual fund owned by the network and are then to a certain extent covered against the various types of charges affecting the audit profession (ibid.: 92). The Big Four firms all make use of captives within their networks, while this form of insurance is not so widespread among the looser and less centralized mid-tier and small accounting organizations (ibid.: 92 f.). This keeps mid-tier and smaller firms 5 The 2014 Directive on Statutory Audits and previous EU legislation tackle this issue in a very vague manner only (Directive 2006/43/EC, art. 30 ff. and Directive 2014/56/EU, art. 30 ff.) 6 Third parties play an important role in this context as shareholders and the various other stakeholders also suffer from misstatements and therefore might have claims against the auditor (EC 2001: 7). 8
from taking audit assignments with a disproportionately great audit scope or with high liability risks that their insurance might not cover (ibid.: 46). Therefore, the segment of big and risky mandates is particularly affected by a strong audit supplier concentration (Le Vourc’h / Morand 2011: 131, 136). The few firms which serve these markets can charge premiums more easily. As most markets for audit services, the European market is generally strongly dominated by the Big Four audit firms (EC 2008: 12 f.): In the countries examined in this study, the Big Four market share ranges between 40 % in Germany and 79 % in Sweden (Francis et al. 2013: 334 f.). On top of that, the supply concentration in the Big Four market segment itself is also fairly high: The values here range between 42 % in the United Kingdom and 54 % in Germany, meaning that one firm alone covers this respective percentage of the Big Four market share (ibid.: 330, 334 f.). In the PIE market segment, the Big Four firms have an even more unchallenged position: Of the listed Blue Chip7 companies which are examined in this work 118 out of 119 were (at least partly, taking into account the joint audits in France) audited by a Big Four firm. Various scholars find that Big Four audit firms can commonly charge premiums on top of the average audit fee (Eilifsen / Willekens 2008: 4 f.). Reasons they name for these premiums are on the one hand higher audit quality (Clatworthy et al. 2009: 141) and industry specialization within the Big Four networks (Castarella et al. 2004: 130 ff.), but also brand name effects (Fergusson et al. 2006: 105), reputational endorsement through prominent audit assignments (EC 2010: 15) and even oligopolistic structures within the audit market (Moizer 1997: 63). On the other hand, the size of the audited companies influences audit fees contrariwise: Large enterprises with an extensive audit scope might, as previously emphasized, pay higher total fees, the marginal returns in this function are diminishing, though (Castarella et al. 2004: 135). This is because the bargaining power of large companies is higher and because those companies usually have a more extensive internal audit system lowering the control risk (ibid.: 135, 138). For the sample used in this study one can therefore again note two oppositely acting influences, which might not offset each 7 A Blue Chip is a well-known company with a high market capitalization. There are no common threshold values which distinguish Blue Chips from non-Blue-Chips, however (Breuer et al. 2012: 80). 9
other completely, but should at least straighten out over a large sample. Other authors state that specific industry expertise enables audit firms to charge higher fees independent of the firm's size. The research results show, that premiums on the reputation of being an expert-firm are up to 34 % of the average fee (ibid.: 130 ff. or Craswell et al. 1995: 318). Zerni (2012: 336) breaks down this effect to the employee level within firms and similarly finds that not only the reputation of the firms, but also the perceived expertise of specific employees at partner-level has a significant positive effect on audit fees. 2.3.2 Disclosure Regulation With the IAS-Regulation (Regulation 1606/2002/EC), which entered into force on January 1st 2005, the EU made the application of IAS / IFRS compulsory for all corporations whose securities are traded on a regulated European market (Regulation 1606/2002/EC, art. 4). The Directive on Annual Financial Statements in 2013 then stipulated, that the “financial statements of public-interest entities [PIE], medium-sized and large undertakings are [to be] audited by one or more statutory auditors or audit firms […]” from 2015 on (Directive 2013/34/EU, art. 34). A definition of company size, which determines wether entities are small, medium-sized or large by using total assets / liabilities, total revenues and employees, is provided in this directive. The directive, however, gives member states the right to choose thresholds for the size determinants within a certain margin (Directive 2013/34/EU, art. 3). Also, even though it was due until July 20th 2015, the directive has not been implemented in every member state yet (FEE 2016: 1 f.). Consequently, the thresholds for statutory audit still widely differ within the EU, with Finland and Malta having the lowest, and Switzerland having the highest thresholds (FEE 2016: 3 f.). The securities of the companies studied in this work are all traded on a regulated European market, though (ESMA 2016 b). Moreover, all the companies examined are governed by the law of a member state. By being governed by member state law and being traded on a regulated European market, all companies of the sample are not only obliged to apply IFRS in their financial statements. They also all match the most recent EU-definition of a PIE provided by the 2014 Directive on Statutory Audits (Directive 2014/56/EU, art. 1), meaning that their financial statements certainly need to be audited by a statutory auditor according to the Directive on 10
Annual Financial Statements. The statutory audits of those IFRS-pursuant financial reports obviously lead to expenditures for the audited companies, which, as many other expenses, are accounted for in the other operating expenses account (IAS 1.99) and therefore not clearly identifiable to the reader of the report. The IFRS themselves do not require the explicit disclosure of audit expenditures. The Statutory Audit Directive (Directive 2006/43/EC, art. 49), took up this issue, however, and prescribed the separate disclosure of audit-, tax advisory- and non-audit fees charged by assurance firms in the notes of the financial reports from 2008 on (Velte / Wernicke 2007: 73). In conclusion, this means that data on audit fees should be publicly available at least for all corporate entities whose securities are traded on a regulated European market. This gives a sample of approximately 6.300 companies (ESMA 2016: 21 f.). 2.4 Variable Management Remuneration 2.4.1 Design and Applicability of Different Remuneration Systems Variable remuneration is very popular as an interest-aligning instrument, especially among bigger enterprises (Heckeler / Lübbig 2016: rec. 1). In a variable management remuneration scheme the compensation of personnel with decision-making power is based on certain determinants – assessment bases – which the respective personnel has influence on. Taking a look at the data sets in the appendix (see appendix A), one will note that the compensation of senior management is never completely variable. As the figures show, the top level managers are paid abundant fixed salaries, which already cover their existential needs very well, and the variable remuneration rather serves as a 'bonus' on top of the fixed income. However, the data summary in 4.3 and the figures in appendix A show that the variable portion still often makes up more than 50 % of the total compensation. Variable remuneration comes in various forms and it is thus accounted for in very different ways. For senior management, the assessment bases determining the remuneration are typically tied to equity instruments: Among others, the development of the share price or the paid out dividends serve as the determinants of their variable salary (Baetge et al. 2014: 9). For managers on lower hierarchy levels other performance indicators, such as sales figures, capital use or profit contribution are more 11
common as assessment bases, because those middle managers have hardly any measurable influence on the performance of the equity instruments (Schulz 2009: 81). If different variable remuneration schemes are installed throughout one company, there is a potential they set conflicting incentives.8 This creates a certain degree of competition inside the entity, which, however, is acceptable as long as those conflicting incentives eventually serve the profit maximization, the increase of shareholder value or any other main objective the enterprise might have (Schulz 2009: 79). How effectively variable remuneration systems can be used significantly depends on outcome measurability and on task programmability: If the outcomes of the assessment bases are easily measurable, it is more probable a variable remuneration system is installed, since compensation can then be easily adjusted to the outcome figures (Björkmann / Furu 2000: 699). If the tasks of managers can in turn be well defined, it is rather rather uncommon to make use of extensive variable remuneration. In those cases it is easier to just monitor if the assigned tasks are executed according to the (fairly simple) job descriptions (Björkmann / Furu 2000: 699; Miller 2008: 354). The extent to which variable remuneration schemes are used also depends on the company size and age. Variable pay systems, especially share-based systems, are more popular among big companies and young technology startups. As scholars find, the equity-based fraction of total pay in those firms is higher than in medium-sized and small companies (Bebchuk / Grinstein 2005: 289 f.). A possible explanation for this is the strong competition for the most capable managers among those startups and big companies: Startups, which usually do not have vast cash reserves, are somewhat forced to make extensive use of equity-based payment. The big companies, which look for similar high performing and risk-taking individuals, then follow this trend in remuneration design and also expand their variable pay programs (Bebchuk / Grinstein 2005: 298). Concerning the payment of the remuneration and it's incentive effects, there are two prevalent ways to design a scheme: While linear remuneration systems with share 8 One remuneration assessment basis inside the company could be the decrease of inventory, which would most expectedly be in conflict with another common assessment basis, the sales figures. 12
characteristics treat the management personnel as if they were normal risk-taking equity owners and account for increases and decreases of the assessment basis, limited linear systems with option characteristics only pay out the amount that is derived from gains of the underlying base value (Schulz 2009: 81 f.; Pellens et al. 2014: 524). By excluding managers from share losses, those remuneration schemes enhance managerial risk-taking (Ménard / Shirley 2008: 11). In practice, shares and phantom shares9 are widespread means to make the actual compensation payments in linear systems (Schnabel 1998: 71 ff.). In limited linear systems, stock options10 and share appreciation rights11 are preferably used for remuneration (Schulz 2009: 81 f.). 2.4.2 Accounting for Different Remuneration Systems Incentive aspects →
S y s t e ms w i th s ha r e Systems with option characterischaracteristics tics
Accounting aspects ↓ 2) Equity-settled share- Shares based payment transactions
3) Cash-settled share- Phantom shares based payment transactions
Share appreciation rights
1) Systems based on --other performance indicators (IAS 19)
(Immediate) cash settlement
Figure 1: Types of remuneration systems, their distinction by incentive and accounting aspects and the instruments they use (Following Wulf 2009: 458)
Unlike the incentive perspective, which as aforementioned usually distinguishes between systems with share characteristics and systems with option characteristics, accounting splits up the diverse types of remuneration systems into (1) systems based on other performance indicators after IAS 19, (2) systems with equity-settled share9 Phantom shares are a fictitious transfer of stocks used as a remuneration instrument. They follow the exact price development of their underlying share. The remuneration is paid out in cash after an agreed vesting period (Breuer et al. 2012: 422). 10 Stock options guarantee their holder a fixed purchase price for a certain stock. The difference between the fixed price and the fair value of the share is the profit / loss of the holder. The holder of the option will of course not exercise the option if fixed purchase price > fair value. (Breuer et al. 2012: 544) 11 Instruments which entitle their holder to the appreciation amount of shares during a given period. In contrast to stock options, appreciation rights do not cause an actual transfer of shares (Breuer et al. 2012: 544). 13
based payment transactions and (3) systems with cash-settled share-based payment transactions (Pellens et al. 522 ff.). Systems based on other performance indicators (1) use figures from accounting (Economic Value Added, return on capital employed, free cashflow…) or from resource planning (sales, stock movement, …) unrelated to equity financing as determinants of the variable salary (Köster 2013: rec. 17). As outlined above, they are popular for middle management variable remuneration. Among top managers they are hardly ever the sole scheme, but rather used in combination with other remuneration systems (Schulz 2009: 81). These 'other performance indicator'-systems account for past events (IAS 19.19) and are therefore comparably easy to display in bookkeeping: When compensation is granted, personnel expenses credit cash or – in case the actual payment is done in a later period – liabilities or provisions (Pellens et al. 2014: 522). If systems with share-based payment transactions (2 and 3) are in use, the bookings are more complicated: In those cases the companies account for variable remuneration every business period, but wether managers can indeed claim those payments is often conditional upon their achievement of certain performance conditions and on a minimum period they need to serve the business (IFRS 2.19). The timeframe in which managers fulfill those agreed conditions is called vesting period and the remuneration is disbursed at the end of this period only (IFRS 2.15). In accounting, the total variable management remuneration of share-based plans is to be allocated over their respective vesting period and it is valued with the estimated probability that the vesting conditions are fulfilled (Coenenberg et al. 2014: 128).12 This means, that the reckoned personnel expenses per period are fairly accurate, even if one considers that they might have to be reversed due to non-fulfillment of the vesting conditions at the end of the vesting period. On the other hand the estimation of the claim probability also gives a wide valuation margin to companies (Pellens et al. 532). As “equity-settled” implies, management is paid with stocks, more precisely with discounts on stocks, or stock options in equity-settled share-based systems (2). This means that the accounts personnel expenses and equity are affected. For stock payments the accounting post is: 12 For equity-settled schemes, this valuation happens once when the remuneration is granted, for cash-settled schemes this valuation happens every period (Köster 2013: rec. 38). 14
Dr.: Personnel expenses, cash
The cash account is also affected in this posting record, as the companies usually still receive an amount smaller than the market value (but greater than zero), a previously agreed subscription price, for their stocks by the managers (Heckeler / Lübbig 2016: rec. 17; Pellens 2014: 529). For payment with options the booking is: Dr.: Personnel expenses
The personnel expense corresponds the value of the options granted in this posting record. Once the granted options can be exercised, the booking is: Dr.: Cash
The cash account rises, as management now buys stocks at a previously agreed subscription price. The second booking is not considered a personnel expense anymore, though (Pellens et al. 2014: 529 f.) In cash-settled share-based systems (3) management is in turn paid with cash or other assets and the share value just serves to determine the amount of the paid out remuneration (Köster 2013: rec. 46). The remuneration is the difference between the fair value at exercise date and a previously agreed subscription price (ibid.). The respective booking is: Dr.: Personnel expenses
Here the provisions account is credited, because the amount (and possibly also the timing) of the future payment is uncertain (Pellens et al. 2014: 540). For companies, cash-settled remuneration has the advantage that it does not change their ownership structure. On the other hand, cash-settled models regularly cause cash outflows, while equity-settled models sometimes even increase the liquidity (Köster 2013: rec. 4). 2.4.3 Disclosure of Remuneration Unlike the disclosure of audit fees, the disclosure of management remuneration is ruled in the IFRS themselves: IAS 24.17 requires the disclosure of “key management personnel compensation in total and for […] t h e […] categories […] short-term employee benefits, post-employment benefits, other long-term benefits, termination 15
benefits and share-based payment” (IAS 24.17). It further defines key management as “those persons having authority and responsibility for planning, directing and controlling the activities of the entity, directly or indirectly, including any director (whether executive or otherwise) of that entity” (IAS 24.9). 3. Trust 3.1 Definition Trust is of particular interest in social capital research (Woolclock 1998: 153). This research deals with “the assets that may be mobilized through” (Nahapiet / Ghoshal 1998: 243) a “network of relationships possessed by an individual or social unit” (ibid.). In daily use, trust is commonly defined as the “the belief that somebody […] is good, sincere, honest, etc. and will not try to harm or trick you” (Deuter et al. 2015: 1679). In the context of NIE a slightly different definition by Mayer et al. is more precise though. Their definition of trust as “the willingness of a party to be vulnerable to the actions of another party based on the expectation that the other will perform a particular action important to the trustor, irrespective of the ability to monitor and control that other party” (Mayer et al. 1995: 712) contains the important features vulnerability and absence of monitoring. These features are so crucial in Agency Theory (Howorth / Moro 2012: 164) that they should not be omitted in a definition used in this work. As Beccera and Gupta (1999: 181) remark reviewing the literature on the topic, trust can develop “between individuals, between organizations, between individuals and social institutions, between individuals and organizations, within public organizations, as a general characteristics of different societies and as a personal trait”. This study will mainly deal with trust between individuals and with trust as a general characteristics of societies. One should also keep in mind though, that there are diverse links and interdependencies between the different dimensions of trust (Child / Möllering 2003: 76). 3.2 How Trust is Measured Trust is not only defined but also perceived very differently across different populations. This poses many restrictions on the measurement of trust, especially if one wants to obtain results about the trusting behavior of larger groups (Alesina / La
Ferrara 2002: 213). Many authors therefore only make use of comparatively smallscale experiments to measure trust: Garfinkel (1967) and Barr (2003), do field research in small communities, Deutsch (1973), Berg et al. (1995) and Glaeser et al. (2000) conduct experiments in a laboratory setting and Rotter (1967) and Beccera et al. (2008) rate exclusive preselected groups by their answers to specially designed comprehensive questionnaires, for example. For large samples and cross-country data the literature regularly falls back to broader and more general sociological studies: Alesina and La Ferrara (2002) use results from the General Social Survey (GSS) by the American National Opinion Research Center (NORC) and Keefer and Knack (1997 and 2008) and La Porta et al. (1997) rely on the outcomes of the World Values Survey (WVS) by the World Values Survey Association, for instance. From those broader sociological studies the particular question “generally speaking, would you say that most people can be trusted or that you need to be very careful in dealing with people?” with the answer choices “most people can be trusted”, “need to be very careful” and “don't know” (WVS Association 2005: 3) has been repeatedly used to make assumptions about trust levels in different regions and countries (Alesina / La Ferrara 2002: 212 f.; Knack / Keefer 1997: 1256). With slight variation in the answer choices, this question appears in many studies, such as the aforementioned GSS, the European Social Survey and in the German General Social Survey (NORC 2014: 115; ESS ERIC 2014: 4; Baumann / Wasmer 2015: 32). Some scholars justifiably emphasize the limits of the informational value of such general studies if they are used in a very specific context. Bachmann and Inkpen (2011: 284) state that one can only speak of trust when actual trusting actions by individuals are taken. According to them, the statement of a trusting attitude alone does not qualify as an indicator of real trust. Glaeser et al. (2000: 840 f.) use laboratory experiments and find that experimentees who say that most people can be trusted do not necessarily act more trustingly than experimentees who state that one can not be careful enough. In other words, their answers to general survey questions on trust do not predict the trusting behavior of individuals well (Keefer / Knack 2008: 704). Reasons for that could be different perceptions wether someone is trustworthy or not, different concepts of trust in others or a different idea what group “most people” are (Glaeser et al. 2000: 815). If survey results and actual behavior indeed diverge, this of course also challenges the use of 'stated trust' as an explana17
tory variable in an empirical model. Glaeser et al. (2000: 840 f.), however, also find a strong correlation of 'stated trust' and trustworthy behavior. This means that individuals who perceive themselves as trusting at least act in a way which should eventually raise the trust of others. On top of that there is a great spread in the country-level aggregated answer results of the WVS: While in some countries, such as Norway and Sweden, more than 65 % of the population think that most people can be trusted, these values range below 20 % in Poland, Slovenia or Spain (WVS Association 2014). This wide spread also holds for many neighboring countries with similar cultures, economic conditions and social structures13, which diminishes the chance that different survey results are just caused through different perceptions or concepts of trust. Moreover, the trust results of the WVS seem to be relatively persistent over time, even though the samples of respondents change with every wave of the study (La Porta et al. 1997: 334; Bjørnskov 2006: 3). So for these reasons and as more precise data for cross-country or regional comparisons has not been gathered yet anyways, I will make use of the WVS results as well. 3.3 Trust and Trustworthiness Since the linkings between trust14 and trustworthiness are strong and manifold (Knack / Keefer 1997: 1258), trustworthiness and its features should also briefly be highlighted in this chapter. Generally speaking, trustworthy behavior can be seen as the counterpart to trust. It is defined as behavior “that you can rely on to be good, honest, sincere, etc.” (Deuter et al. 2015: 1680). Following the definition of trust adopted above, it especially means that the vulnerability of others is not exploited for one's own advantage, even if that is easily possible and will not be discovered. In the literature on trustworthiness, the most common features of persons worth being trusted – trustees – are ability, benevolence and integrity (Mayer et al. 1995: 717 ff.). In comparison to trust, trustworthiness is even harder to measure: The only way to make assumptions about how trustworthy individuals are is to analyze their behavior in situations in which they could exploit their fellow citizens without
13 The percentage-point difference between Norway and Finland is 15,7 % and the difference between France and Germany is 15,1 %, for instance (WVS Association 2014). 14 Even though there are the aforementioned restrictions to it, 'stated trust' from social survey results will also just simply be referred to as 'trust' from this point on. 18
any consequences.15 The American magazine Readers Digest has repeatedly conducted a “lost wallet” test, in which they “accidentally” drop twelve wallets with a cellphone number and the equivalent of 50 US-Dollars in different cities across the globe and then count how many of those wallets are returned (Beres 2016). This test is of course far from a scientific experiment, but its setup would, if scientific sampling criteria as a greater sample size or changing 'loss-locations' within the city (Columbia University 2016) were fulfilled, generally serve for drawing some conclusions about the trustworthiness of the population in those places (Keefer / Knack 2008: 703). Just as with the trust-statements, there is great spread within the results of the wallet test, with some cities where more than 90 %, and other cites where less than 20 % of the wallets are returned (Beres 2016). On top of that, in spite of the small sample sizes the different editions of the test show certain tendencies, with some cities constantly ranking better and some ranking worse. A more scientific approach to the issue of trustworthiness is a parking ticket experiment by Fisman and Miguel. They investigate the parking violations of United Nations (UN) mission diplomats in New York City (Fisman / Miguel 2006: 2). The setting of the UN diplomats in New York is somewhat special, since those individuals enjoy diplomatic immunity, meaning that they used to be16 shielded against any legal prosecution or other consequences if they violated parking regulations and did not pay their tickets (ibid.: 4).17 The study results show that also the number of New York parking violations varies greatly across the different nations: While some missions accumulate more than 100 parking tickets per diplomat, other missions don't commit any parking violations at all (ibid.: 19 ff.). Fisman and Miguel in their study focus on the issue of corruption – they indeed also find a strong correlation with the values of the Transparency International Corruption Perception Index (ibid.: 7) – but their study serves for making some conclusions about trustworthiness just as well as the diplomats are clearly put in a situation in which they can easily exploit 15 Anything untrustworthy respondents state about themselves would – by definition – not be worth trusting. Therefore survey data concerning this characteristic should not be reliable at all. 16 Legislation has changed since the experiment data was gathered as the freeriding of the diplomats became a real problem (in five years diplomats piled up 150.000 parking tickets and fines of 18 million US-Dollars) resulting in wide media coverage (Miguel / Fisman 2006: 4 f.). 17 Of course, this did not exempt the diplomats from being obliged to pay the fees or stick to the regulations. 19
the 'vulnerability' of the municipality of New York without having to fear any consequences. Looking at the trust answers of the WVS and the wallet experiment, Keefer and Knack (2008: 704 f.) find a correlation of 0,44 between trust and trustworthiness. Using more current data, the correlation is even slightly stronger. Following their regression, I examine the relation of the trust results from the WVS and the trustworthiness results from the parking ticket research and equally discover a moderate 18 negative correlation of 0,30, indicating that citizens from nations with lower trust levels might rather be prone to commit parking violations.
Figure 2: Trust and parking tickets per diplomat across different countries (Following Keefer / Knack 2008: 705; Data: Fisman / Miguel 2006; WVS 2014)
3.4 Economic Effects of Trust Scholars from social capital research find untold positive effects of trust on economic activity. This section will firstly address the underlying theoretical assumptions of discovered trust-effects and then give a concise summary of selected research findings. 18 Following Quatember (2014: 72), this study regards correlations of corr < |0,20| as weak, correlations of |0,20| ≤ corr < |0,60| as moderate and correlations of |0,60| ≤ corr < |1,00| as strong. 20
It is argued that trust has substantially contributed to the rise of contracting (Keefer / Knack 2008: 706). In comparison to spot markets, which are distinguished by immediate delivery (Breuer et al. 2012: 327), contracting allows for long-term economic relationships with extensive compliance periods, such as financial lending, permanent employment and fixed investments (Keefer / Knack 2008: 706). If long-term relationships are established, individuals can in turn specialize in different fields of activity (ibid.). Agency problems are not completely overcome, but appear less frequently (due to high trustworthiness levels) and receive less consideration in planning processes (due to high trust levels) (Beccera / Gupta 1999: 196). Costs of economic interaction are therefore lower: Performance monitoring and the need for complex written contracts covering every possible contingency decrease (Keefer / Knack 2008: 708). In national economies this leads to higher income growth rates (Keefer / Knack 1997: 1260 f.), higher investment rates (ibid: 1263), better tax compliance, greater judicial efficiency, higher bureaucratic quality and less corruption (La Porta et al. 1997: 335). Especially poorer countries seem to profit from increases of trust levels, since they struggle to provide a well-functioning legal enforcement or, more general, institutions that could to some extent make up for low trust between individuals (Keefer / Knack 1997: 1260). In the political sphere trust also has positive effects. On the one hand at individual level, where politicians who are trustworthy and thus trusted intuitively get better election results (Keefer / Knack 2008: 706 f.). On the other hand, mass-based democracies are also more stable as a whole if they are embedded in high trust environments (Ingelhart 1990: 23). In the private sector trust inside merchant coalitions gave rise to long distance trade at times when formal institutions facilitating exchanges were not developed yet: Eleventh century Maghribi traders heavily relied on the judgement and recommendations of their trader alliances when they interacted with agents overseas and they made extensive use of collective ostracism when needed (Greif 1989: 864 ff.). Towards the end of the 20th century trust was crucial to the emergence of alternative organizational forms, so-called networked businesses, which combined hierarchical firm- and market exchange features (Möllering / Sydow 2006: 70). Nowadays trust accounts for better performance of individuals inside organizations, for higher
degrees of interorganizational cooperation (Beccera / Gupta 1999: 181), for better information exchange inside companies and for higher speed in decision-making, eventually resulting in environments which foster innovation (ibid: 197). Research conducted in China shows that subsidies of multinational corporations, which are trusted rather than strictly controlled by their parent company have higher growth rates in sales and profits (Child / Möllering 2003: 76). Moreover, Howorth and Moro (2012: 173 f.) find that the cost of credit for small Italian companies diminishes at significant level if their managers are perceived trustworthy, i.e. capable, benevolent and integer by the lending banks. 3.5 Origins of Trust When exploring the origins of trust, it is helpful to distinguish between interaction-based trust at micro-level and institution-based trust at macro-level (Bachmann / Inkpen 2011: 282). As outlined in 3.1, trust at micro-level is trust between individuals who are in direct contact, while trust at macro-level is rather a generalized characteristics of large organizations or whole societies (ibid.). Trust of individuals at micro-level is a product of the so-called 'propensity' to trust, of the perceived trustworthiness of the counterpart(s), of risk-taking culture and of experience gained in earlier interaction with others (Mayer et al. 1995: 715 ff.). Propensity in this case describes what general expectations we have of our fellow citizens, wether we have a rather positive or a negative image of most people19 before dealing with them (ibid.: 715). Perceived trustworthiness, or reputation of the trustee, on the other hand determines how we then assess our counterpart in a very particular transaction. Perceived trustworthiness is highly influenced by prejudice, meaning that familiarity and cultural similarity positively influence how we asses others, while anything strange and unknown is very likely to be perceived as untrustworthy (Child / Möllering 2003: 71, 78). Even though it is surely not unsurmountable, cultural and social heterogeneity can in some cases be obstacles to interaction-based trust (Zucker 1986: 78). Risk-taking influences trust inasmuch, as one always needs to appraise what happens in case of misplaced trust: What are the stakes involved and how much pressure can be put on the counterpart if they behave untrustworthy 19 One could thus say that the WVS and other social studies actually measure the propensity to trust in societies. 22
are common considerations we intuitively take (Mayer et al. 1995: 725 f.). If we are risk-averted, we might not place trust in others after assessing the risk (Möllering / Sydow 2006: 68). Both, perceived trustworthiness and to a lesser degree also propensity to trust draw on past experiences of the trustor when she decides if she wants to trust or not (Greif 1989: 867 f.). This in turn also means that trust evolves over time if the involved parties repeatedly successfully interact (La Porta 1997: 333). Scholars assign professional and social networks a crucial role in the evolution of the interaction-based dimension of trust. Their members obtain “habits of cooperation, solidarity and public-spiritedness”, which nourish mutual trust (Putnam 1993: 89 f.). At macro-level institutions20 are seen as the main factor influencing trust levels (Child / Möllering 2003: 71; Keefer / Knack 2008: 711 ff.; Bachmann / Inkpen 2011: 284 f.). With their research in China, Child and Möllering (2003: 76, 78) show that institutions even have effects on the micro-level, with trust in business partners being higher if the institutions in their home countries are considered trustworthy. At macro-level institutions precisely affect trust by evoking 'institutional isomorphism', the phenomenon that (organizational) actors behave similarly when they are exposed to the same institutional environment (Bachmann / Inkpen 2011: 286). Institutional isomorphism makes the actions of the trustee foreseeable and thus reduces the risk that she will unexpectedly behave untrustworthily (ibid.: 285). Uncertainty, a major obstacle to successful economic and political interaction (North 2008: 28), is reduced. An example of an institution evoking institutional isomorphism would be professional regulations that closely guide the behavior of employees at work (Bachmann / Inkpen 2011: 288). Since the term is so widely defined, institutions can on the other hand also foster trust by acting as guaranteeing third-party agents who confirm statements of trustees, i.e. assure their trustworthiness. In practice, universities fulfill this function by testing and certifying the ability of their students, for instance (ibid.). Bachmann and Inkpen repeatedly stress the advantages of institutions in trustbuilding. According to them, institutions are a very efficient instrument for trust-creation, because they cost the actors far less time and resources than repeated interaction when making a relationship trustful. This is of great importance, especially in advanced globalized national economies, where actors get in contact with new busi20 As defined in section 2.1. 23
ness partners again and again at an ever higher pace (Bachmann / Inkpen 2011: 282, 285). Even though he does not explicitly use the term, Gambetta (1988) points out what happens if institutions are not present or too weak to fulfill their purpose: In his essay about Southern Italy he explains the power of mafia and all its negative consequences with a lack of well-functioning institutions. He pictures a state which is undermined by criminal clusters, does not provide effective law enforcement and therefore is not trusted by its citizens (Gambetta 1988: 162, 167). This leads to a high degree of uncertainty for economic actors, who due to the weak state also do not trust each other (ibid.: 162). Mafiosi take advantage of the 'market niche' that lacking institutions leave and create their own rules and markets, while mutually further destabilizing the 'competing' official fiscal institutions (ibid.: 169 ff., 173). By making trust an economic good that is available only to few, namely their protegees, mafia can then put very high premiums on the price of trust (ibid.: 172 f.). This makes mafia significantly different from legal private entrepreneurs who fill regulatory gaps and rather profit from network effects. Arguing that differences in formal institutions neither explain the great variety in trust, nor the general downward trend in trust levels across different U.S. states, Keefer and Knack (2008: 714 f.) go a bit further in explaining the origins of trust and name social norms as another important factor beyond institutions. As a main power that leads to trust and trustworthy behavior caused by social norms, they see social ostracism, since experiments show that formally legal free-riding, i.e. the breach of norms, is very likely to be punished socially, even if the punishers do not have any benefits from the punishment (ibid: 715). Accordingly, trust is higher in places where norms are of great importance. Additionally, they mention social heterogeneity as a factor beyond institutions leading to lower trust. Their argument is that social heterogeneity, no matter whether it is characterized by monetary, cultural, educational or political segregation, leads to less interaction and ergo to less trust (ibid.: 716). Precisely, actors from different social groups who do not interact can firstly not ostracize each other well, they secondly behave differently and thus perceive each other as less trustworthy, they thirdly do not necessarily approach the other social group by themselves and they also feel less altruism towards members of other social classes (ibid.: 716 f.). Empirical results, implying that linguistic homogeneity
increases trust, while income inequality decreases it, support their argument (Zak / Knack 2001: 316). 4. Methodology 4.1 General Purpose and Sample Selection As previously outlined, high trust significantly facilitates economic interaction, both on macro- and on micro-level. Especially with regard to agency problems scholars find numerous positive effects of high trust levels in populations. This study picks out a classic principal-agent-setting, the owner-manager conflict (Marten et al. 2015: 48), and examines the effects of trust on this setting. It puts special attention to two types of agency costs, more precisely two types of monitoring expenditures, which arise here, namely audit fees and variable remuneration. According agency theory both, the financial statement audit and variable remuneration, are measures introduced because of agents who are expected to have different preferences and goals than their principals. The theory concludes that those different preferences and goals result in opportunism, i.e. untrustworthy behavior of the agents. By drawing on the findings of trust research and incorporating them into agency theory I postulate that the trust the principal places in her agent significantly influences her extent of monitoring and incentivisation and thus the monitoring expenditures. I formulate the following hypotheses: 1) Audit fees are lower in populations with high levels of mutual trust. High trust diminishes audit fees. 2) The variable amount of remuneration is lower in populations with high levels of mutual trust. High trust decreases the variable friction of remuneration. To validate these hypotheses, I analyze the financial statement audit fees and variable management remuneration of 119 European blue chip companies. I focus on publicly listed European enterprises, as those particular companies are situated in an ever more similar regulatory environment. As previously mentioned, they all use the same accounting standards, they apply those standards similarly21 and they are audited by the same few audit firms. Their auditors are also bound to similar professional standards and apply those standards similarly. Concluding, many factors, which other21 Since the enforcement is coordinated on a supranational level 25
wise would need to be controlled for in an empirical model, can be treated ceteris paribus. Referring to the earlier remarks on institutional isomorphism one could now argue that the similar institutional environment in the EU leads to very similar trust levels in the affected companies and that the WVS data used does not reflect the reality at enterprise-level. This issue is not so severe, though, as institutional isomorphism is a continuos process rather than an immediate change (Tuttle / Dillard 2007: 388 f.). The IAS-Regulation entered into force in 2005 and all harmonization efforts referred to in this work took effect after that. Therefore, one can expect that trust levels should still differ across the companies from different EU-countries, even in the segment of PIEs, which is exposed to a very similar regulatory environment. The sample of this study contains companies from the stock market lead indices in France, Germany, Sweden and the United Kingdom. These countries were chosen, because they belong to the biggest national economies in the EU: Germany, the United Kingdom and France in this order rank highest in total gross domestic product (GDP) and Sweden ranks seventh (EC 2016 b: 18 f.). Sweden was added to the sample, because by total GDP it is the biggest economy in Scandinavia (EC 2016 b: 19), a region with very high trust levels 22. Across the four sample countries the trust levels diverge greatly. Sweden, with 65,2 % of the respondents answering that most people can be trusted (WVS Association 2014), can be categorized as a high-trust nation. Germany, with 33,8 %, and the United Kingdom, with 30,0 % of the respondents answering that most people can be trusted (ibid.), can be categorized as midtrust nations. France, with only 18,7 % of the respondents stating that most people can be trusted (ibid.), can be categorized as a low-trust nation. On European level there are about 6.300 companies which are listed on a regulated market and prepare IFRS-pursuant financial statements (ESMA 2016: 21). France, Germany and the United Kingdom each have more than 500 issuers which fulfill those criteria (ibid: 22). In Sweden there are more than 250 entities which are listed on a regulated market and prepare IFRS-pursuant financial statements (ibid.). The companies listed on a lead index (i.e. regulated market) all prepare IFRS-pursuant financial statements (Regulation 1606/2002/EC, Art. 4) and thus fall into the group of those aforementioned 500+ French, German and British, respectively 250+ 22 In Denmark 74,9 % of the respondents state that most people can be trusted (GESIS 2013: 77). In Norway this figure is at 73,7 % and in Finland it is at 58,0 % (WVS 2014). 26
Swedish companies. FTSE 100, the lead index in the United Kingdom comprises the stocks of 100 companies (LSE 2016), while the French lead index CAC 40 consists of stocks of 40 companies (Euronext 2016). German DAX 30 and Swedish OMX Stockholm 30 as their names imply both comprise 30 stocks (Deutsche Börse 2016; NASDAQ Stockholm 2016). For each of the countries selected 30 listings were included in the study sample. This means that all companies from the German and the Swedish23 lead index were added to the sample. For France and the United Kingdom 30 enterprises from the respective index were randomly selected. 4.2 Dependent and Independent Variables of the Model in Detail 4.2.1 Complete Model To test the aforementioned hypotheses, this study runs two fairly simple multivariate linear regressions: 1) audit fee = ß0 + ß1 assets + ß2 revenue + ß3 employees + ß4 trust + u Regression 1 explains what factors influence the costs of statutory audits. A special focus is put on the explanatory variable trust. The values used for trust are the previously discussed WVS answer results. Hypothesis 1 (refer to section 4.1) holds, if ß4 is smaller than zero. Control variables are total assets, total revenue and the number of employees As explained in section 2.3.2, these control variables together are common determinants of the company size. In section 2.3.1 this work argues that company size has a positive effect on audit fees, therefore ß1, ß2 and ß3 are expected to be greater than zero. The constant u are unobserved factors determining the audit fee. 2) var. remuneration = ß0 + ß1 assets + ß2 revenue + ß3 employees + ß4 trust + u Regression 2 explains what determines the variable proportion of top level executive management. All regressors are the same as in Regression 1. Hypothesis 2 holds, if ß4 is again smaller than zero. As this work argues that company size has a positive effect on variable remuneration (refer to section 2.4.1), ß1, ß2 and ß3 are expected to be greater than zero. The constant u are again unobserved determinants. 23 OMX Stockholm 30 lists the stock of Altlas Copco AB twice (NASDAQ Stockholm 2016). This means that Sweden actually only contributes 29 companies to the sample. Accordingly, the total sample size is 119. 27
Except for the VWS results on trust, data for the entire empirical model, i.e. assets, revenue and employee figures, were collected from the latest annual reports and registration documents of the respective companies. These publications are all available on the internet. While the data on audit fees is comparably easy to asses, remuneration disclosure is quite extensive and the issue which parts of the compensation are variable often requires some consideration and judgement. 4.2.2 Audit Fees In annual reports, audit fees are commonly listed in a chart which contains all fees paid to the assurance firms and their networks. As those fees are not exclusively charged for assurance services, but regularly also for tax consulting or certain advisory services, the disclosing companies differentiate between expensed audit fees, audit-related fees, tax fees and other fees in their reports (Ericsson AB 2016: 101; Danone S.A. 2016: 166; Beiersdorf AG 2016: 77; British American Tobacco plc 2016: 139, for instance). The name of the statutory audit firm and its responsible auditor is always provided (Directive 2006/43/EC, art. 24). Some companies break down wether the fees were charged for services provided to the parent company or for services to subsidiaries (Danone S.A. 2016: 166, for instance). All French and some other companies are audited by more than one audit firm (Deng et al. 2014: 1030). If they are audited by more than one auditor, the companies commonly disclose the different types of fees (audit, tax, …) paid at auditor-level (Danone S.A. 2016: 166; Ericsson AB 2016: 2016: 101, for instance). This study regards fees for parent and subsidiary audits paid to all auditors as audit fees. That respective figure is incorporated into the regression. Audit-related fees, tax fees, advisory fees and any other fees are omitted. 4.2.3 Remuneration Remuneration consists of more elements and its disclosure is therefore more extensive. All companies in the sample publish remuneration reports, which cover the compensation of the executive management and the board of directors. This study focuses on the payment of executive management because executives on the one hand have more influence on the development of the enterprise and therefore receive higher remuneration (Andreas 2011: 11 f.). This makes the owner-(executive)
manager conflict a lot more interesting than agency relationships the owners have with directors or other individuals inside the company. On the other hand performance-based variable remuneration is a lot less popular among directors (Hölz 2013: 203, 206). As previously mentioned, the IFRS oblige all entities to disclose their compensation of key management personnel (IAS 24.17). Across the whole sample, the compensation of key management personnel consists of some fixed and some variable elements (which sum up to total compensation). The figures for total compensation (and its fixed and variable components) show a great spread across the examined companies (appendix A, columns 'remuneration' and 'incentive'). To a certain extent, this is owed to the fact that each company pays each of their managers differently. Some companies, such as Berkley Group and Volkswagen (Berkley Group plc 2016: 70; Volkswagen AG 2016: 69 ff.) pay relatively high compensation to most of their executive board members, while others, such as Boliden and Pearson (Boliden AB 2016: 86 f.; Pearson Group plc 2016: 105) do not. However, the aggregate companyfigures for compensation also differ because the size of the executive boards can be very different. Some insurance companies and banks have around ten members (Deutsche Bank AG 2016: 7; Allianz SE 2016: 20 f.; AXA S.A. 2016: 112, for instance) on the executive board. Other companies, such as the real estate business Vonovia only have four executive directors (Vonovia SE 2016: 58). In order to eliminate the disruptive factor of different board sizes, which makes a comparison of the aggregate remunerations (and their components) per company across the sample impossible, this study uses the relative amount of variable remuneration, i.e. the variable remuneration over total remuneration as the dependent variable in the regression. The variable part of the total remuneration again comprises some sub-elements: most firms pay short term and long term incentives. Short term incentives are determined by other performance indicators and commonly settled in cash (ABB AB 2016: 49; thyssenkrupp AG 2016: 24, 26, for instance). Thus they fulfill the criteria of immediate cash settlement after IAS 19 explained in section 2.4.2. This means that for those variable parts the accounted expense and the disbursed amount (i.e. the amount the manager received) are mostly the same. The figure stated in the annual financial
report can be used in the empirical model without too much further consideration. For long term incentives the accounted expense and the disbursement can differ, however. Here, due to the regulation of IFRS 2, managers do not necessarily receive the entire expenses accounted for immediately (refer to section 2.4.2). Some companies therefore provide figures of 'compensation in respect of the year in question' or of 'granted compensation' (i.e. expensed amount accounted for) on the one hand and 'compensation paid in the year in question' or 'inflows' (i.e. amounts the managers actually receive) on the other hand (Capgemini S.A. 2016: 61; Deutsche Börse AG 2016 b: 124 ff., for instance). As discussed in 2.4.2, the amounts classified as expenses in accounting are fairly accurate: Long term incentive plans are not only evenly allocated over the entire period in which the compensation is vested. Accounting also takes into account the scenario that vesting conditions are not fulfilled and expenses possibly need to be reversed. Therefore, this study preferably uses the expense figures as long term variable compensation whenever they are provided. Even though they are employee benefits (IAS 19.4), pensions are excluded from the empirical model. They are subtracted from total compensation as it is difficult to assess to which extent they are variable and to which extent they are fixed. Exceptions are made when companies explicitly state wether paid pensions are fixed, respectively variable. Benefits in kind and ancillary benefits are regarded as fixed remuneration on the other hand. They of course vary over time, this variation usually is not (solely) tied to performance though. For executive managers, they also just constitute a very small friction of the total remuneration. Therefore their interest-aligning power is rather low (Schulz 2009: 81 f.). Total remuneration
Excluded from model: Fixed elements: fixed remunera- Variable elements: short term p e n s i o n s ( u n l e s s clearly indicated as tion, benefits in kind, ancillary incentive, long term incentive fixed / variable) benefits
Figure 3: Elements of remuneration considered in and excluded from in the empirical model 4.2.4 Assets The control variables in both regressions are assets, revenues and headcount. Total assets are a single unambiguous figure that does not leave room for any interpretation 30
once it is disclosed in the financial report (Freiberg / Lüdenbach 2013: rec. 42). The data for assets used in this work was taken from the balance sheets of the respective companies. 4.2.5 Revenue Revenue disclosure is not so uniform in turn. Banks and insurances do not always present their revenues the same way as companies from other sectors: Banks on the one hand commonly disclose their income and expenditures separately for each of their business areas (Sollanek 2005: 38). In their income statements they list interest-, fee-, commission-, and investment income / expenses, for instance (Commerzbank AG 2016: 149; Svenska Handelsbanken AB 2016: 67, for instance). Those subtotal values were all added up and their sum is the value used for revenue in the empirical model. Some insurances on the other hand separate their income statements into insurance-related and non insurance constituents (Sollanek 2004: 36; c.f. Munich Re AG 2016: 164, for instance). This constraints the quick assessment of the total revenues of the group. The item 'net premiums earned', which comes closest to revenue in this context (Sollanek 2004: 36 f.) was therefore used as the revenue figure in those cases. Otherwise 'total income' or actual revenues were regarded revenue (Allianz SE 2016: 136; AXA S.A. 2016: 200, for instance). 4.2.6 Employees The control variable employees also leaves some room for interpretation: The number of employees can be stated as headcount, i.e. as the number of all individuals working full time and part time for the enterprise. Many companies state their employee figure in full-time equivalent (FTE), however (Deutsche Bank AG 2016: 32; Royal Bank of Scotland plc 2016: 87, for instance). FTE converts all part-time employment relationships into full-time employments by dividing the total hours worked in the enterprise by the average annual hours of a full-time job (OECD 2001). This regularly 'shrinks' the stated employee figure. FTE figures were used in the model whenever they were available. Some of the examined companies conduct franchising besides their internalized activities and state employee figures for wholly owned operations and franchises (Accor Hotels S.A. 2016: 37, for instance). Employees of franchises are not included 31
in the empirical model. Equally, when companies make use of contracting and disclose the employee figures of their contractors (Antofagasta plc 2016: 21, for instance), this does not find consideration in the model. Exceptions are made when the share of contractors in the total workforce exceeds 20 %. In those cases contractor employees are counted as part of the company workforce. The assumption here is that contracting is not only deployed temporarily, but over longer periods. Vast contracting is is then presumed to be part of the business model. If companies and contractors have long-term business relations, contracting becomes an important factor affecting economic success in turn (BCG 2015: 12 f.) and should hence also have an influence on audit fees and variable remuneration. 4.2.7 Trust As it is the most current data set with a greater number of European countries, the trust values used are from the fifth wave of the WVS only. The data of wave five was obtained in the years 2005-2009 (WVS Association 2014). This survey period is not congruent with the survey periods of the other model variables. Since more current data is not available and trust values are relatively persistent over time (refer to section 3.2) this is a minor deficiency to the model, however. 4.2.8 Currency Conversion The EU-endorsed IFRS do not prescribe the use of a particular presentation currency in annual reports (IAS 21.38). Most of the examined French and German companies state their financial figures in Euros. The British and Swedish enterprises of the sample mostly disclose in Pound Sterling or Swedish Crowns, however. Some companies use other foreign currencies, such as Swiss Francs or US-Dollars (Lafarge Holcim Ltd. 2016: 132; Lunedin Petroleum AB 2016: 85 ff., for instance). In order to make them comparable, all financial figures were converted into Euros. For simplification the conversion of foreign currencies was conducted following the modified closing rate method of IAS 21.39. This means that revenues, audit fees and remuneration were converted with the 2015 average rates (IAS 21.39 f.). Assets were converted with the closing rates as at December 31st 2015 (IAS 21.39 a). In order to reach further comparability, the converted financial data were adjusted
with national price levels24. 4.3 Data Assessment As previously mentioned 119 companies were examined. Every company of the sample discloses figures for each of the variables, therefore the number of observations is 119 for all regressands and regressors. The following chart summarizes the data used in the model: Variable
145.000.000.000 360.000.000.000 3.244.545
total remunera- 14.600.000 tion
Figure 4: Descriptive statistics for the sample obtained All variables show fairly high standard deviations. This great dispersion of the data serves the empirical model. A check for correlations in order to avoid collinearity issues shows that total remuneration and incentive are strongly related at 0,943 (see appendix B). In their plain form, these variables are neither used in regression 1 nor in regression 2, therefore the strong correlation is not problematic. Assets and audit fee are also strongly related at 0,744. Perfect collinearity is not reached, this high value should be considered when assessing the regression results, though. Interestingly, all size measures (assets, revenue, employees) show weak or moderate correlations with each other only.
24 The national price level measure “makes it possible to compare the cost of the bundle of goods that make up […] GDP across countries. It tells how many dollars are needed to buy a dollar's worth of goods in the country as compared to the United States” (Worldbank 2016). 33
5. Results Running both regressions, the following results are obtained: Variable
Coefficient in regression 1 Coefficient in regression 2 (explaining audit fees) (explaining var. remuneration)
Figure 5: Regression coefficients for regression 1 (audit fees) and regression 2 (variable remuneration) Even though both regressions explain a fairly high amount of the variation in the sample – the adjusted R2 of regression 1 is 0,71 and the adjusted R 2 of regression 2 is 0,34 – and the overall F-test confirms significance for both regressions, the coefficients are still somewhat unsatisfactory in explaining the dependent variable: The statement that an increase in assets of one causes an audit fee increase of 0,0000217 is a bit difficult to evaluate. So is the statement that an increase in employees of one causes a decrease in variable remuneration of 0,0000000298. Only the coefficient ß3 of regression 1, indicating that an increase in employees of one makes the audit fee rise about 16 Euros, is tangible in a way. The cause for the marginally small coefficients is the wide range of numbers used in the model: While employees are figures between 91 and 610.076, audit fees mostly range in the millions and most asset and revenue figures are in the billions. Regression 1, however, already confirms the assumed tendencies of hypothesis 1 and the remarks concerning the control variables, a positive impact of assets, revenue and employees and a negative impact of trust on audit fees. In order to facilitate the assessment of the coefficients, all variables are logarithmized in a second set of regressions. This provides the following results:
Regression 1 (explaining audit fees): Variable
Significant at 95 %?
log assets (ß1)
log revenue (ß2)
log employees 0,253844 (ß3)
log trust (ß4)
R2: 0,6222; Adj. R2: 0,6089; Prob. > F: 0,0000
Figure 6: Results of regression 1 logarithmized
Regression 2 (explaining variable remuneration): Variable
Significant at 95 %?
log assets (ß1)
log revenue (ß2)
log employees -0,118252 (ß3)
log trust (ß4)
R : 0,2896; Adj. R : 0,2640; Prob. > F: 0,0000 2
Figure 7: Results of regression 2 logarithmized The coefficients are now easier to assess. A one percent increase in assets causes a 0,22 percent increase of audit fees (see figure 6), a one percent increase in employees causes a 0,12 decrease in variable remuneration (see figre 7), and so forth. The adjusted R2 of the logarithmized models are slightly lower, the overall F-test confirms significance of both regressions, though. 6. Discussion Regression 1 confirms confirms hypothesis 1 and all remarks concerning the expected effects of assets, revenue and employees on audit fees made in section 4.2.1. Trust indeed seems to have a diminishing effect on audit fees. On the other hand, audit fees are high when the audited company is big, i.e. has many employees, assets and revenues. The coefficients of assets, employees and trust are significant at 35
95 %-level. The regression explains a fairly high degree of about 60 % of the variation in the sample. According to the regression, of all explanatory variables trust hast the greatest effect on audit fees. The total value of assets in turn affects the amount of audit fees the least. Regression 2 also confirms the expected effects of trust. Here, trust seems to have an even bigger impact causing a decrease in variable remuneration of 0,71 percent if it is increased 1 percent. On the other hand the coefficients of assets and employees are unexpected and refute my suggestions made in section 4.2.1 as they cause decreases in variable remuneration when they are increased. A possible explanation for this could be that big companies and their remuneration practices, especially the variable remuneration schemes, are rather subject to public criticism than small companies (FAZ 2012; Haufe 2012, for instance). The big firms might react to this criticism in an anticipatory obedience and limit their variable schemes. However, the explanatory power of the regression is not too high either with only slightly more than 25 % of the variation of the sample explained. Following Björkmann and Furu (2000: 699) and Miller (2008: 354), who find a substitutional effect between monitoring and incentive payments, in an additional regression variable remuneration was incorporated into the regression function of audit fees serving as another control variable. Equally, audit fees were incorporated into the function of variable remuneration. The results confirmed their findings, indicating that higher audit fees decrease the variable portion of remuneration and vice versa, the coefficients were not significant, though. As the adjusted R 2 also slightly decreased after incorporating these additional variables, the results will not be presented in detail here. With only four regressors both models are comparably simple. Regression 2 has some severe deficiencies as it shows unexpected effects for assets and employees and additionally explains little of the sample variation. The trust effect is suspiciously high. Therefore, I conclude that the model does not explain variable remuneration too well and that it hence can not really support the underlying hypothesis. A redesign of the regression is recommended. Regression 1, however, confirms the findings of the literature and has a comparably high explanatory power. The evidence of this first regression serves well as a basis 36
for further research. Further research should, however, take account of certain issues: Firstly, this study relies on WVS country-level data. WVS data is not always perfectly accurate as Glaeser et al. (2000: 840 f.), Beugelsdijk (2006: 382 f.) and others justifiably criticize. Particularly, it is questionable wether this macro-data can really explain behavior at micro-level. It would hence be helpful to run similar regressions with firm-specific or at least city- or regional trust data to test my results. As large-scale survey- or experimental data does not exist yet at this level, it would need to be collected by researchers. Secondly, big prime standard-listed companies are used in this study since they are subject to vast EU-wide regulation and disclose very many of their figures, but small and medium-sized enterprises most probably represent national cultures and their corresponding trust levels better: They are not as internationally active, culturally diverse and institutionalized as the big firms (Drori 2008: 456 ff.). Thus, WVS data should intuitively better describe the behavioral patterns in those smaller companies. As regulation of small and medium-sized companies is not so harmonized at European level yet25, a model examining those companies would most probably have to control for a set of further existing regulatory differences, though. Also it is more difficult to obtain data from this market segment. Thirdly, with only four regressors the model is far from considering all factors influencing audit fees. Even in this small and highly regulated fraction of the market audits and their fees are dependent upon many more influences. A more detailed model could make use of the findings about auditor liability, auditor and client bargaining power and industry expertise presented in section 2.3.1 and translate them into further control variables. If they are wisely chosen, these control variables would probably increase the explanatory power of the model. Eventually, by exclusively looking at Western European countries the study limits itself to a very wealthy and well-developed part of the world. Taking the institutionalist perspective, this region provides some very effective arrangements, such as efficient legal systems, consequent law enforcement and very active free markets (Shirley 2008: 621). It would be interesting to examine the impacts of trust in other 25 As previously outlined (section 2.3.1), statutory audit thresholds still greatly differ across the EU, for instance. While all European PIEs are subject to statutory audit (Directive 2013/34/EU, art. 34), medium-sized companies which would need an auditor in some countries do not need to be audited in other countries (FEE 2016: 3 f.). 37
regions where these institutions are not well-developed yet existent. Would trust have an even greater diminishing influence on monitoring expenditures in those regions and confirm Keefer and Knacks (1997: 1260 ff.) findings for trust effects on income growth in less economically developed regions? 7. Conclusion Even though this analysis is limited in scope, it generally confirms the suggestion that trust diminishes monitoring costs. To a certain degree trust apparently makes laborious audits and ingenious remuneration schemes redundant. Given these findings, this on the one hand poses the question wether monitoring and incentivisation are really necessary to the extent at which we can currently observe them. If trust is indeed able to partly replace the examined control instruments, it is maybe worth to further think and research about how trust can be fostered in order to lower the 'overdrawn' agency costs. Doing this, one should keep in mind, that trust is not a panacea, though. Other than monitoring or incentive, it can not be instantly applied in any agency situation: The creation of trust is a long process and it depends upon many, partly unknown factors. It might be impossible to create a perfectly trustful environments, especially in bigger, and socially unequal or heterogenous populations (Bjørnskov 2006: 12). It might cause high costs, too. But research shows that once it is established, the advantages of trust are manifold and as the reviewed literature emphasizes, the positive effects would surely exceed the sphere of the owner-manager conflict. On the other hand, this study questions a basic assumption of Agency Theory, the selfish and opportunistic agent. With reference to the literature review on trustworthiness, I assume that some people and groups behave more trustworthy than others. This can also be assumed for agents and it obviously influences the behavior of principals. Agency theorists do not take account of this varying trust factor. They unconditionally assume self-interest and opportunism and thus unrestrictedly recommend control in the form of monitoring and incentivisation. This work regards the extensive use of control that we can observe in many environments as overdone and sees
some cost-saving potential26 in a more trustful approach to social and economic interaction. To express it with the words of the initially quoted German saying: For many situations it regards control as good, but as a general attitude it regards trust as better.
26 Further research should also try to name and quantify the costs that are related to the creation trust. Costs of trust-creation could then be compared to the part of monitoring costs that high distrust causes. 39
Appendix A) Data Sample Variables: company: Company name relincentive: incentive divided by remuneration auditfee: Audit fee in EUR assets: Total assets in EUR revenues: Total revenues in EUR employees: Number of employees remuneration: Total remuneration incentive: Variable part of remuneration company
relincen- auditfee tive
employ- remuneration ees
Deutsche Te- 59,92% lekom
Fresenius 76,78% Medical Care
Infineon 51,56% Technologies
relincen- auditfee tive
employ- remuneration ees
Alfa Laval AB 18,42%
ASSA ABLOY AB
Atlas Copco AB
Electrolux AB 44,39%
Fingerprint Cards AB
Hennes & Mauritz AB
Investment AB Kinnevik
Lundin Petro- 84,68% leum AB
Swedish Match AB
relincen- auditfee tive
employ- remuneration ees
Telia Compa- 2,16% ny AB
Credit Agrico- 43,33% le
Astra Zeneca 82,18%
Berkeley Group Holding
British Ameri- 65,94% can Tobacco
relincen- auditfee tive
employ- remuneration ees
Coca Cola HBC AG
Int. Cons. Air- 83,72% lines
Marks & Spencer
Royal Bank of Scotland
Smith & Nephew
Saint James Place
B) Correlation of Model Variables rel. incentive
Bibliography Printed Literature –
Alesina, Alberto / La Ferrara, Eliana, 2002. Who trusts others? Journal of Public Economics, Vol. 85, pp. 207–234.
Andreas, Jörn, 2011. Determinanten der Aufsichtsratsvergütung in deutschen Aktiengesellschaften. München: Rainer Hampp Verlag.
Baetge, Jörg / Kirtsch, Hans-Jürgen / Thiele, Stefan, 2014. Bilanzen, 13th Edition. Düsseldorf: IDW Verlag.
Bardmann, Manfred, 2011. Grundlagen der Allgemeinen Betriebswirtschaftslehre. Wiesbaden: Gabler.
Barr, Abigail, 2003. Trust and expected trustworthiness: experimental evidence from Zimbabwean villages. The Economic Journal, Vol. 113, pp. 614–630.
Bebchuk, Lucian / Grinstein, Yaniv, 2005. The growth of executive pay. Oxford Review of Economic Policy, Vol. 21, No. 2, pp. 283-303.
Beccerra, Manuel / Gupta, Anil, 1999. Trust within the organization: integrating the trust literature with Agency Theory and Transaction Cost Economics. Public Administration Quarterly, Vol. 23, No. 2, pp. 177-204.
Becerra, Manuel / Lunnan, Randi / Huemer, Lars, 2008. Trustworthiness, Risk, and the Transfer of Tacit and Explicit Knowledge Between Alliance Partners. Journal of Management Studies Vol. 45, No. 4, pp. 691-713.
Berg, Joyce / Dickhaut, John / McCabe, Kevin, 1995. Trust, Reciprocity and Social History. Games and Economic Behavior, Vol. 10, pp. 122-142.
Beugelsdijk, Sjoerd, 2006. A note on the theory and measurement of trust in explaining differences in economic growth. Cambridge Journal of Economics, Vol. 30, pp. 371–387.
Björkmann, Ingmar / Furu, Patrick, 2000. Determinants of variable pay for top managers of foreign subsidiaries in Finland. International Journal of
Human Resource Management, Vol. 11, No. 4, pp. 698-713. –
Bjørnskov, Christian, 2006. Determinants of generalized trust: A crosscountry comparison. Public Choice, Vol. 130, pp. 1-21.
Blum, Ulrich / Dudley, Leonard / Leibbrand, Frank / Weiske, Frank, 2005. Angewandte Institutionenökonomik. Theorien – Modelle – Evidenz. Wiesbaden: Gabler.
Breuer, Claudia / Breuer, Wolfgang / Schweizer, Thilo (eds.), 2012. Gabler Lexikon Corporate Finance (2. Auflage). Wiesbaden: Gabler.
Child, John / Möllering, Guido, 2003. Contextual Confidence and Active Trust Development in the Chinese Business Environment. Organization Science, Vol. 14, No. 1, pp. 69-80.
Clatworthy, Mark / Makepeace, Gerald / Peel, Michael, 2009. Selection bias and the Big Four premium: new evidence using Heckman and matching models. Accounting and Business Research, Vol. 39, No. 2, pp. 139-166.
Coenenberg, Adolf / Haller, Axel / Schultze, Wolfgang, 2014. Jahresabschluss und Jahresabschlussanalyse, 23 rd Edition. Stuttgart: Schäffer Poeschel.
Craswell, Allen / Francis, Jere / Taylor, Stephen, 1995. Auditor brand name reputations and industry specializations. Journal of Accounting and Economics, Vol. 20, Issue 3, pp. 297–322.
Deng, Mingcherng / Lu, Tong / Simunic, Dan / Ye, Minley, 2014. Do Joint Audits Improve or Impair Audit Quality? Journal of Accounting Research, Vol. 52 No. 5, pp. 1029-1060.
Deuter, Margaret / Bradbery, Jennifer / Turnbull, Joanna, 2015. Oxford Advanced Learner's Dictionary of Current English. 9th Edition. Oxford: Oxford University Press.
Deutsch, Morton, 1973. The Resolution of Conflict. New Haven: Yale University Press.
Drori, Gili, 2008. Institutionalism and Globalization Studies. In: Greenwood,
Royston / Oliver, Christine / Sahlin, Kerstin / Suddaby, Roy (eds.), 2008. The SAGE Handbook of Organizational Institutionalism. London: SAGE. Pp. 449-472. –
Eilifsen, Aasmund / Willekens, Marleen, 2008. In the name of trust. Some thoughts about trust, audit quality and audit regulation in Europe. In: Quick, Reiner / Turley, Stuart / Willekens, Marleen (eds.), 2008. Auditing Trust and Governance. Developing regulation in Europe. London: Routledge. Pp. 1-18.
Fama, Eugene, 1980. Agency Problems and the Theory of the Firm. The Journal of Political Economy, Vol. 88, No. 2, pp. 288-307.
Ferguson, Andrew / Francis, Jere / Stokes, Donald, 2006. What matters in audit pricing: industry specialization or overall market leadership? Accounting and Finance Vol. 46, Issue 1, pp. 97–106.
Francis, Jere / Michas, Paul / Seavey, Scott, 2013. Does Audit Market Concentration Harm the Quality of Audited Earnings? Evidence from Audit Markets in 42 Countries. Contemporary Accounting Research Vol. 30, No. 1, pp. 325–355.
Freiberg, Jens / Lüdenbach, Norbert, 2013. § 23 Anteilsbasierte Vergütungsformen (Share-Based Payment). In: Lüdenbach, Norbert / Hoffmann, Wolf-Dieter (eds.), 2013. Haufe IFRS-Kommentar. Das Standardwerk, 13th Edition. Freiburg / München: Haufe.
Garfinkel, Harold, 1967. Studies in Ethnomethodology. Englewood Cliffs: Pretice-Hall.
Glaeser, Edward / Laibson, David / Scheinkman, José / Soutter, Christine, 2000. Measuring trust. The Quarterly Journal of Economics, August 2000 Issue, pp. 811-846.
Greif, Avner, 1989. Reputation and Coalition in Medieval Trade: Evidence on the Maghribi Traders. Jounal of Economic History, Vol. 49, No. 4, pp. 857-882.
Hachmeister, Dirk, 2008. Regulierung von Abschlussprüfern und Prüfungsgesellschaften im Spannungsfeld von Qualitätssicherung und Wettbewerb. In: 47
Ballwieser, Wolfgang / Grewe, Wolfgang (eds.), 2008. Wirtschaftsprüfung im Wandel. Herausforderungen an Wirtschaftsprüfung, Steuerberatung, Consulting und Corporate Finance. München: Beck. Pp. 55-74. –
Heckeler, Udo / Lübbig, Maike, 2016. § 24. Anteilsbasierte Vergütung (Aktienoptionsprogramme). In: Driesch, Dirk / Riese, Joachim / Schlüter, Jörg / Senger, Thomas, 2016. Beck'sches IFRS-Handbuch. München: C.H. Beck. Pp 1087-1114.
Hölz, Christiane, 2013. Aufsichtsratsvergütung 2012 – eine Bestandsaufnahme. BOARD, Vol. 5/2013, pp. 203-206.
Ingelhart, Ronald, 1990. Cultural Shift in Advanced Industrial Society. Princeton: Princeton University Press.
Jensen, Michael / Meckling, William, 1976. Theory of the Firm: managerial behavior, Agency Costs, and ownership structure. Journal of Financial Economics, Vol. 3, No. 4, pp. 305-360.
Jensen, Michael / Meckling, William, 1983. Reflections on the corporation as a social invention. Harvard Business School Negotiation, Organizations and Markets Unit Working Paper Series Vol. 18. Boston: Harvard Business School.
Keefer, Philip / Knack, Stephen, 1997. Does social capital have an economic payoff? A cross-country investigation. The Quarterly Journal of Economics, November 1997 Issue, pp. 1251-1288.
Keefer, Philip / Knack, Stephen, 2008. Social Capital, Social Norms and the New Institutional Economics. In: Ménard, Claude / Shirley, Mary (eds.), 2008. Handbook of New Institutional Economics. Berlin: Springer. Pp. 701725.
Klosa Annette / Scholze-Stubenrecht, Werner / Wermke, Matthias, 1998. Vertrauen ist gut, Kontrolle ist besser. In: Klosa Annette / Scholze-Stubenrecht, Werner / Wermke, Matthias (eds.), 1998. DUDEN Redewendungen und sprichwörtliche Redensarten. Idiomatisches Wörterbuch der deutschen Sprache. Mannheim: Brockhaus. P. 766.
Köster, Oliver, 2013. Anteilsbasierte Vergütung nach IFRS 2. Kommentierung der Share-based Payments. Wiesbaden: Springer Gabler.
Manne, Henry, 1965. Mergers and the Market for Corporate Control. The Journal of Political Economy, Vol. 73, No. 2., pp. 110-120.
Marten, Kai-Uwe / Quick, Reiner / Ruhnke, Klaus, 2015. Wirtschaftsprüfung. Grundlagen des betriebswirtschaftlichen Prüfungswesens nach nationalen und internationalen Normen, 5 th Edition. Stuttgart: Schäffer-Poeschel.
Mayer, Roger / Davis, James / Schoorman, David, 1995. An integrative model of organizational trust. Academy of Management Review, Vol. 20, No. 3, pp. 709-734.
Ménard, Claude / Shirley, Mary, 2008. Introduction. In: Ménard, Claude / Shirley, Mary (eds.), 2008. Handbook of New Institutional Economics. Berlin: Springer. Pp. 1-20.
Moizer, Peter, 1997. Auditor Reputation: The International Empirical Evidence. International Journal of Auditing, Vol. 1, Issue 1, pp. 61-74.
Möllering, Guido / Sydow, Jörg, 2006. Organisationen vertrauen – Organisationales Vertrauen in Kunden-Lieferanten-Beziehungen. In: Bauer, Hans / Neumann, Marcus / Schüle, Anja (eds.), 2006. Konsumentenvertrauen: Konzepte und Anwendungen für ein nachhaltiges Kundenbindungsmanagement. München: Vahlen. Pp. 63-76.
Nahapiet, Janine / Ghoshal, Sumantra, 1998. Social Capital, Intellectual Capital, and the Organizational Advantage. The Academy of Management Review, Vol. 23, No. 2., pp. 242-266.
Naumann, Klaus-Peter, 2008. Abschlussprüfung in einem geänderten regulatorischen Umfeld. In: Ballwieser, Wolfgang / Grewe, Wolfgang (eds.), 2008. Wirtschaftsprüfung im Wandel. Herausforderungen an Wirtschaftsprüfung, Steuerberatung, Consulting und Corporate Finance. München: Beck. Pp. 97-120.
Niehus, Rudolf, 2008. Der Wechsel des Abschlussprüfers – Faktum oder 49
Farce? In: Ballwieser, Wolfgang / Grewe, Wolfgang (eds.), 2008. Wirtschaftsprüfung im Wandel. Herausforderungen an Wirtschaftsprüfung, Steuerberatung, Consulting und Corporate Finance. München: Beck. Pp. 141-172. –
North, Douglass, 2008. Institutions and the Performance of Economies over Time. In: Ménard, Claude / Shirley, Mary (eds.), 2008. Handbook of New Institutional Economics. Berlin: Springer. Pp. 21-30.
Pellens, Bernhard / Füllbier, Rolf / Gassen, Joachim / Sellhorn, Thorsten, 2014. Internationale Rechnungslegung. IFRS 1 bis 13, IAS 1 bis 41, IFRICInterpretationen, Standardentwürfe, 9th Edition. Stuttgart: Schäffer Poeschel.
Putnam, Robert / Leonardi, Robert / Nanetti, Raffaella, 1993. Making Democracy Work: Civic Traditions in Modern Italy. Princeton: Princeton University Press.
Quatember, Andreas, 2014. Statistik ohne Angst vor Formeln. Das Studienbuch für Wirtschafts- und Sozialwissenschaftler, 4th Edition. Hallbergmoos: Pearson.
Rotter, Julian, 1967. A new scale for the measurement of interpersonal trust. Journal of Personality, Vol. 35, Issue 4, pp. 651–665.
Schnabel, Hansjörg, 1998. Wertorientierte Vergütung von Führungskräften. Wiesbaden: Gabler.
Schulz, Arne, 2009. Aktienkursorientierte Vergütungssysteme für Führungskräfte. Eine Analyse aus rechtlicher und betriebswirtschaftlicher Sicht. Wiesbaden: Gabler.
Shirley, Mary, 2008. Institutions and Development. In: Ménard, Claude / Shirley, Mary (eds.), 2008. Handbook of New Institutional Economics. Berlin: Springer. Pp. 611-638.
Sollanek, Achim, 2004. Versicherungsbilanzen nach deutschem Handelsrecht. Betriebswirtschaftliche Handlungshilfen Jahresabschluss und Bilanzanalyse (115). Düsseldorf: edition der Hans-Böckler-Stiftung.
Sollanek, Achim, 2005. Bankbilanzen nach deutschem Handelsrecht (153). Düsseldorf: edition der Hans-Böckler-Stiftung. 50
Tuttle, Brad / Dillard, Jesse, 2007. Beyond Competition: Institutional Isomorphism in U.S. Accounting Research. Accounting Horizons, Vol. 21, No. 4, pp. 387–409.
Velte, Patrick / Wernicke, Thomas, 2006. Themenschwerpunkte der 5. Hamburger Reviosions-Tagung vom 12. und 13. Oktober 2006. In: Altes, Peter / Freidank, Carl-Christian (eds.), 2007. Rechnungslegung und Corporate Governance. Reporting, Steuerung und Überwachung der Unternehmen im Umbruch. Berlin: Erich Schmidt Verlag. Pp. 67-77.
Woolclock, Michael, 1998. Social Capital and Economic Development: Toward a Theoretical Synthesis and Policy Framework. Theory and Society, Vol. 27, No. 2, pp. 151-208.
Wulf, Inge, 2009. Bilanzierung von aktienbasierten Vergütungen. In: Brösel, Gerrit / Zwirner, Christian (eds.), 2009. IFRS-Rechnungslegung. Grundlagen, Aufgaben, Fallstudien. München: Oldenbourg. Pp. 457-472.
Zak, Paul / Knack, Stephen, 2001. Trust and growth. The Economic Journal, Vol. 111, pp. 295-321.
Zerni, Mikko, 2012. Audit Partner Specialization and Audit Fees: Some Evidence from Sweden. Contemporary Accounting Research, Vol. 29, No. 1, pp. 312–340.
Zucker, Lynne, 1986. Production of Trust: Institutional Sources of Economic Structure, 1840-1920. Research in Organizational Behavior, Vol. 8, pp. 53-111.
Online Sources –
ABB Ltd., 2016. Creating the future. The ABB Group Annual Report 2015. Available at: http://new.abb.com/docs/default-source/investor-center docs/annual-report/annual-report-2015/abb-group-annual-report-2015english.pdf. Accessed on July 1st 2016.
Accor Hotels S.A., 2016. 2015 Registration Document and Annual Report. Available at: http://www.accorhotels51
group.com/fileadmin/user_upload/Contenus_Accor/Finance/Documentation/2 015/accor_drf2015_livre_uk.pdf. Accessed on July 4th 2016. –
Allianz SE, 2016. Wir führen unsere stolze Tradition weiter. Allianz Konzern Geschäftsbericht 2015. Available at: https://www.allianz.com/v_1458046976000/media/investor_relations/de/beri chte_und_finanzdaten/geschaeftsbericht/gb2015/gb2015_gruppe.pdf. Accessed on July 29th 2016.
American Institute of CPAs (AICPA), 2015. Hiring a Quality Auditor: Your Guide to the Selection Process. Available at: http://www.aicpa.org/InterestAreas/PrivateCompaniesPracticeSection/QualityServicesDelivery/KeepingUp/DownloadableDocuments/Hiring_%20a_ %20quality_%20auditor.pdf. Accessed on August 26th 2016.
AXA S.A., 2016. Registration Document. Annual Financial Report 2015. Available at: https://cdn.axa.com/www-axa-com%2F73719a96-c3b1-456babaf-63b80c06968c_axa_reference_document_2015.pdf. Accessed on July 4th 2016.
Baumann, Horst / Wasmer, Martina, 2015. German General Social Survey 2014. English Translation of the German “ALLBUS“-Questionnaire. Available at: https://dbk.gesis.org/DBKSearch/SDESC2.asp? no=5242&tab=3&db=D&dab=0. Accessed on August 6th 2016.
Beiersdorf AG, 2016. Geschäftsbericht 2015. Available at: http://www.geschaeftsbericht2015.beiersdorf.de. Accessed on July 29th 2015.
Beres, Damon, 2016. Most Honest Cities: The Reader’s Digest “Lost Wallet” Test. Available at: http://www.rd.com/culture/most-honest-cities-lost-wallet-test/. Accessed on August 7th 2016.
Berkeley Group plc, 2016. Annual Report 2015. Available at: http://www.berkeleygroup.co.uk/media/pdf/l/2/berkeley-annual-report2015.pdf. Accessed on July 7th 2016.
Boliden AB, 2016. Competitive mines and smelters. Annual Report 2015. Available at: http://reports.boliden.com/globalassets/pdfs/boliden_annual-report-2015_eng.pdf. Accessed on July 1st 2016. 52
Boston Consulting Group (BCG), 2015. Value creation in mining 2015. Beyond basic productivity. Available at: https://www.bcgperspectives.com/Images/BCG-Beyond-Basic-ProductivityAug-2015.pdf. Accessed on August 19th 2016.
British American Tobacco plc, 2016. Delivering today. Investing in tomorrow. Annual Report 2015. Available at: http://www.bat.com/group/sites/uk__9d9kcy.nsf/vwPagesWebLive/DO9DCL3B/$FILE/medMDA87PVT.pdf?openelement. Accessed on July 7th 2016.
Capgemini S.A., 2016. 2015 Registration Document, Annual Report. Available at: https://www.capgemini.com/resource-file access/resource/pdf/2015_registration_document_1.pdf. Accessed on July 4th 2016.
Columbia University, 2016. Finders, Keepers? Not in New York. Available at: http://www.columbia.edu/cu/news/07/03/Wallet.html. Accessed on August 23rd 2016.
Commerzbank AG, 2016. Laufend besser. Geschäftsbericht 2015. Available at: https://www.commerzbank.de/media/de/aktionaere/haupt/2016_5/ geschaeftsbericht_2015.pdf. Accessed on July 29th 2016.
Danone S.A., 2016. Danone 15. Registration Document, Annual Financial Report. Available at: http://www.danone.com/fileadmin/user_upload/2015Registration-Document_DANONE.pdf. Accessed on July 4th 2016.
Deutsche Bank AG, 2016. Leistung aus Leidenschaft. Geschäftsbericht 2015. Available at: https://www.db.com/ir/de/download/Deutsche_Bank_Geschaeftsbericht_201 5.pdf. Accessed on July 29th 2016.
Deutsche Börse AG, 2016. DAX. Zugehörige Werte. Available at: http://www.boerse-frankfurt.de/index/zugehoerige-werte/DAX#Zugehörige Werte. Accessed on August 15th 2016.
Deutsche Börse AG, 2016 (b). Finanzbericht 2015. Available at: http://deutscheboerse.com/blob/2535924/cfa92cf1c8486ad893a0871e75ab5a5e/data/annu53
al-financial-report-2015_de.pdf. Accessed on July 29th 2016. –
Ericsson AB, 2016. Annual Report 2015. Available at: https://www.ericsson.com/res/investors/docs/2015/ericsson-annual-report-2015-en.pdf. Accessed on July 1st 2016.
Euronext N.V., 2016. CAC 40 Factsheet. Available at: https://www.euronext.com/en/products/indices/FR0003500008-XPAR/market-information. Accessed on August 15th 2016.
European Central Bank (ECB), 2016. Euro foreign exchange reference rates. Available at: https://www.ecb.europa.eu/stats/exchange/eurofxref/html/index.en.html. Accessed on July 30th 2016.
European Commission (EC), 2008. Comission recommendation concerning the limitation of the civil liability of statutory auditors and audit firms. Available at: http://ec.europa.eu/internal_market/auditing/docs/liability/impact_assessment_en.pdf. Accessed on July 30th 2016.
European Comission (EC), 2010. Green Paper. Audit Policy: Lessons from the Crisis. Available at: http://eur-lex.europa.eu/legal-content/EN/TXT/PDF/? uri=CELEX:52010DC0561&from=EN. Accessed on August 3rd 2016.
European Commission (EC), 2016. European Group of Auditors’ Oversight Bodies (EGAOB). Available at: http://ec.europa.eu/finance/auditing/egaob/index_en.htm. Accessed on July 22nd 2016.
European Commission (EC), 2016 (b). Statistical Annex of European Economy. Available at: http://ec.europa.eu/economy_finance/publications/eeip/pdf/ip025_statistical_ annex_ ee_spring_2016_en.pdf. Accessed on August 15th 2016.
European Securities and Markets Authority (ESMA), 2016. Who we are. Available at: https://www.esma.europa.eu/about-esma/who-we-are. Accessed on August 14th 2016.
European Securities and Markets Authority (ESMA), 2016 (b). Register 54
of Regulated Markets. Available at: http://registers.esma.europa.eu/publication/searchRegister?core=esma_registers_mifid_rma. Accessed on July 23rd 2016. –
European Securities and Markets Authority (ESMA), 2016 (c). Report on Enforcement and Regulatory Activities of Accounting Enforcers in 2015. Paris: ESMA. Available at: https://www.esma.europa.eu/sites/default/files/library/2016410_esma_report_on_enforcement_and_regulatory_activities_of_accounting _enforcers_in_2015.pdf. Accessed on August 15th 2016.
European Social Survey European Research Infrastructure Consortium (ESS ERIC), 2014. Source Questionnaire Round 7, 2014/2015. Available at: http://www.europeansocialsurvey.org/docs/round7/fieldwork/source/ESS7_so urce_main_questionnaire.pdf. Accessed on August 6th 2016.
EY, 2014. European Union audit legislation. Available at: http://www.ey.com/Publication/vwLUAssets/EY__European_Union_audit_legislation/$FILE/EY-pov-european-union-audit-legislation-april-2014.pdf. Accessed on July 22nd 2016.
Federation of European Accountants (FEE), 2016. Audit exemption thresholds in Europe. Available at http://www.fee.be/images/publications/auditing/1605_Audit_exemption_thre sholds_update.pdf. Accessed on August 1st 2016.
Fisman, Raymond / Miguel, Edward, 2006. Cultures of corruption: evidence from diplomatic parking tickets. National Bureau of Economic Research Working Paper Series Working Paper 12312. Available at: http://www.nber.org/papers/w12312. Accessed on August 7th 2016.
Frankfurter Allgemeine Zeitung (FAZ), 2012. EU will offenbar strikte Schranken für Manager-Boni. Available at: http://www.faz.net/aktuell/wirtschaft/nicht-hoeher-als-fixgehalt-eu-will-offenbar-strikte-schranken-fuer-manager-boni-11955412.html. Accessed on August 20th 2016.
GESIS - Leibniz Institute for the Social Sciences (GESIS), 2013. Euro55
pean Values Study (EVS) 2008 - Variable Report Denmark. Cologne: GESIS. Available at: http://www.gesis.org/en/services/data-analysis/survey-data/rdcinternational-survey-programs/european-values-study/4th-wave-2008/. Accessed on August 15th 2016. –
Haufe, 2012. Top-Management-Vergütung zu Unrecht in der Kritik. Available at: https://www.haufe.de/personal/hr-management/boni-top-management-verguetung-zu-unrecht-in-der-kritik_80_138610.html. Accessed on August 20th 2016.
Institute of Chartered Accountants of England & Wales (ICAEW), 2005. Agency Theory and the role of Audit. Available at: https://www.icaew.com/~/media/corporate/files/technical/audit%20and %20assurance/audit%20quality/audit%20quality%20forum/agency %20theory%20and%20the%20role%20of%20audit.ashx. Accessed on August 27th 2016.
LafarageHolcim Ltd., 2016. Building a new leader for a new world. Annual Report 2015. Available at: http://reports.lafargeholcim.com/2015/annual-report/. Accessed on July 5th 2016.
Le Vourc’h, Joelle / Morand, Pascal, 2011. Study on the effects of the implementation of the acquis on statutory audits of annual and consolidated accounts including the consequences on the audit market. Available at: http://ec.europa.eu/i nternal_market/auditing/docs/other/full_study_en.pdf. Accessed on August 3rd 2016.
London Economics, 2006. Study on the Economic Impact of Auditors’ Liability Regimes (MARKT/2005/24/F). Available at: http://ec.europa.eu/internal_ market/auditing/docs/liability/auditors-finalreport_en.pdf. Accessed on August 3rd 2016.
London Stock Exchange plc (LSE), 2016. FTSE 100 constituents. Available at: http://www.londonstockexchange.com/exchange/prices-andmarkets/stocks/indices/summary/summary-indices-constituents.html? index=UKX. Accessed on August 15th 2016.
Lundin Petroleum AB, 2016. Value driven transformation. Annual Report 56
2015. Available at: https://www.lundinpetroleum.com/Documents/ar_2015_e.pdf. Accessed on July 2nd 2016. –
Münchener Rückversicherungs-Gesellschaft AG, 2016. Konzerngeschäftsbericht 2015. Available at: https://www.munichre.com/site/corporate/get/documents_E1989201206/mr /assetpool.shared/Documents/0_Corporate%20Website/_Financial%20Reports/2016/Annual%20Report%202015/302-08842_de.pdf. Accessed on July 30th 2016.
NASDAQ Stockholm AB, 2016. OMXS 30, OMX Stockholm 30 index. Available at: http://www.nasdaqomxnordic.com/index/index_info?Instrument=SE0000337842. Accessed on August 15th 2016.
National Opinion Research Center (NORC), 2014. U.S. General Social Study Questionnaire 2014 (Cross-Section V3). Available at: http://gss.norc.org/documents/quex/BALLOT3_AREA_2014.pdf. Accessed on August 6th 2016.
Organisation for Economic Co-operation and Development (OECD), 2001. Full-time equivalent employment. Available at: http://stats.oecd.org/glossary/detail.asp?ID=1068. Accessed on August 18th 2016.
Pearson Group plc, 2016. Focused on delivery. Pearson Annual report and accounts 2015. Available at: https://www.pearson.com/ar2015.html. Accessed on July 8th 2016.
PricewaterhouseCoopers (PWC), 2013. Understanding a financial statement audit. Available at: http://www.pwc.com/gx/en/audit-services/publications/assets/pwc-understanding-financial-statement-audit.pdf. Accessed on August 23rd 2016.
Royal Bank of Scotland Group plc, 2016. Building a strong, simple, fair bank. Annual Report and Accounts 2015. Available at: http://www.investors.rbs.com/annual-report-2015.aspx. Accessed on July 8th 2016.
Svenska Handelsbanken AB, 2016. Annual Report 2015. Available at:
https://www.handelsbanken.com/shb/inet/icentsv.nsf/vlookuppics/investor_re lations_en_q-reports_hb_2015_eng_annualreport/$file/hb_2015_eng_annualreport.pdf. Accessed on July 2nd 2016. –
thyssenkrupp AG, 2016. engineering. tomorrow. together. Geschäftsbericht 2014/2015. Available at: https://www.thyssenkrupp.com/media/unternehmen/compliance/ thyssenkrupp_2014_2015_gb.pdf. Accessed on July 30th 2016.
Volkswagen AG, 2016. Menschen bewegen. Geschäftsbericht 2015. Available at: http://www.volkswagenag.com/content/vwcorp/content/de/misc/pdfdummies.bin.html/downloadfilelist/downloadfile/downloadfile_34/file/Y_20 15_d.pdf. Accessed on July 30th 2016.
Vonovia SE, 2016. Es ist Zeit Wohnen neu zu denken. Geschäftsbericht 2015. Available at: http://investoren.vonovia.de/websites/vonovia/German/4050/finanzberichte_-praesentationen.html. Accessed on July 30th 2016.
World Values Survey (WVS) Assosciation, 2014. Wave 5 (2005-2008) Official Aggregate v.20140429. Available at: http://www.worldvaluessurvey.org/WVSOnline.jsp. Accessed on July 3rd 2016.
Worldbank, 2016. Price level ratio of PPP conversion factor (GDP) to market exchange rate. Available at: http://data.worldbank.org/indicator/PA.NUS.PPPC.RF?view=map. Accessed on August 19th 2016.
Annual Reports and Registration Documents Used for Data Collection Company
Creating the future. The ABB Group Annual Report 2015. 2016
Accor Hotels S.A.
2015 Registration Document and Annual Report.
How we create value 2015. Adidas Group Geschäftsbericht.
Air Liquide S.A.
2015 Reference Document
Alfa Laval AB
Chili con care. Annual Report 2015.
Wir führen unsere stolze Tradition weiter. Allianz Konzern 2016 Geschäftsbericht 2015.
Annual report and financial statements 2015.
Annual Report 2015.
ASSA ABLOY AB
Annual Report 2015.
Associated British Foods 2015 Annual report and accounts. plc
What science can do. Annual Report 2015.
Atlas Copco AB
Annual Report 2015.
Registration Document. Annual Financial Report 2015.
Ökonomische, ökologische und gesellschaftliche Leistung. BASF Bericht 2015.
Berkeley Group plc
Annual Report 2015.
BHP Billiton plc
Resourcing global growth. Annual Report 2015.
BNP Paribas S.A.
Registration Document and Annual Financial Report 2015.
Competitive mines and smelters. Annual Report 2015.
British American Tobacco plc
Delivering today. Investing in tomorrow. Annual Report 2015.
Burburry. London, England. Annual Report 2015 / 2016.
2015 Registration Document, Annual Report.
Coca Cola Hellenic Bottling Company AG
Refreshing business. 2015 Integrated Annual Report.
Laufend besser. Geschäftsbericht 2015.
Horizonte der Digitalisierung. Geschäftsbericht 2015.
Crédit Agricole S.A.
Annual Financial Report 2015, Registration Document.
Danone 15. Registration Document, Annual Financial Report.
Deutsche Bank AG
Leistung aus Leidenschaft. Geschäftsbericht 2015.
Deutsche Lufthansa AG
Konsequent Zukunft sichern. Geschäftsbericht 2015.
Deutsche Post AG
Intelligente Logistik. Geschäftsbericht 2015.
Deutsche Telekom AG
Das Geschäftsjahr 2015. Antworten für die digitale Zukunft.
Annual Report 2015.
How 20 years have flown. Annual report and accounts 2015.
Annual Report 2015.
2015 Registration Document including annual financial report.
Annual Report 2015.
Essilor International S.A. 2015 registration document.
Annual Report 2015.
Fingerprint Cards AB
Annual Report 2015.
Fresenius Medical Care
Verstehen was zählt. Geschäftsbericht 2015.
AG & Co. KGaA Fresenius SE & Co. KGaA
Konzernabschluss und Konzern-Lagebericht 2015.
Driving Long Term Value from Solid Foundations. Annual Report 2015.
Annual Report 2015.
Wachstum in Metropolen. Geschäftsbericht 2015.
Henkel AG & Co. KGaA
Hennes & Mauritz AB
Annual Report 2015.
Infineon Technologies AG
Integration mit System. Geschäftsbericht 2015.
Progress and Performance. Annual Report and Financial Statements 2015.
InterContinental Hotels Group plc
Great Hotels Guests Love. Annual Report and Form 20-F 2016 2015.
International Consolidated Airlines Group, S.A.
Achieving together. Annual report and accounts 2015.
Investment AB Kinnevik
Annual Report 2015.
100 years of Investor. Building best-in-class companies since 1916.
Delivering strong growth and building scale. ITV plc Annual Report and Accounts for the year ended 31 December 2015.
Financial Report 2015.
Building a new leader for a new world. Annual Report 2015.
Land Securities plc
Annual Report 2015.
Nahe am Kunden. Geschäftsbericht 2015.
Registration Document 2015, Annual Report.
Lundin Petroleum AB
Value driven transformation. Annual Report 2015.
Marks & Spencer plc
Annual report and financial statements 2015.
Stärker. Geschäftsbericht 2015.
Mondi Group plc
Integrated report and financial statements 2015.
Münchener RückversicherungsGesellschaft AG
Document de Référence 2015.
Nordea Bank AB
Annual Report 2015.
Innovation and possibilities. Nokia in 2015.
Pearson Group plc
Focused on delivery. Pearson Annual report and accounts 2015.
2015 Registration document. Including the annual financial report.
ProSiebenSat.1 Media SE
Growing to the next level. Geschäftsbericht 2015.
Long-term thinking for life. Prudential plc Annual Report 2015.
Publicis Groupe S.A.
2015 Registration document, annual financial report.
Rangold Resources Ltd. Breaking new ground. Annual report 2015.
Registration document including the annual financial report 2015.
Royal Bank of Scotland Group plc
Building a strong, simple, fair bank.
Registration document 2015 including the annual financial 2016 report.
Compagnie de SaintGobain S.A.
2015 Registration Document including the 2015 annual financial report and the corporate social responsibility report.
Annual Report 2015.
Reimagine Your Business. Geschäftsbericht 2015.
Annual Report 2015.
Schneider Electric SE
Financial and Sustainable Development - Annual Report, Registration Document 2015.
Annual Report 2015.
Annual Report 2015.
We build for a better society.
Annual Report 2015.
Smith & Nephew plc
Better outcomes. Supporting healthcare professionals for over 150 years.
Smiths Group plc
Bringing technology to life. Annual Report 2015.
Société Generale S.A.
2016 Registration document, annual financial report 2015. 2016
Fiscal 2015 Registration Document including the Annual Financial Report.
Asking more. 2015 Annual Report.
Toward industry-leading profitability. Annual Report 2015. 2016
St. Jame's Place plc
Annual report & accounts 2015.
Standard Life plc
Growth, value, responsibility. Annual report and accounts 2015.
Svenska Handelsbanken AB
Annual Report 2015.
A sustainable financial system – a crucial element in building a society. Annual Report 2015.
Swedish Match AB
Making strides in smokeless. Annual Report 2015.
2015 Reference document. Including the Annual Financial 2016 Report.
Annual Report 2015.
What we did in 2015. Annual + Sustainability Report 2015.
engineering. tomorrow. together. Geschäftsbericht 2014/2015.
Travis Perkins plc
Annual report & accounts 2015.
Consistency. Financial Report 2015.
2015 Registration Document including the Integrated Report, the Annual Financial Report, the Corporate
Governance and Sustainability Report. Vinci S.A.
2015 Annual Report.
Consolidated financial statements at 31 December 2015.
2015 Annual Report
Vodafone Group plc
Unifying communications. Annual Report 2015.
Menschen bewegen. Geschäftsbericht 2015.
Annual and Sustainability Report 2015.
Es ist Zeit Wohnen neu zu denken. Geschäftsbericht 2015.
Making everyday experiences special. Annual Report and 2015 Accounts 2014/15.
Directory of Legal Documents –
Directive 2006/43/EC of the European Parliament and of the Council of 17 May 2006 on statutory audits of annual accounts and consolidated accounts, amending Council Directives 78/660/EEC and 83/349/EEC and repealing Council Directive 84/253/EEC.
Directive 2013/34/EU of the European Parliament and of the Council on the annual financial statements, consolidated financial statements and related reports of certain types of undertakings, amending Directive 2006/43/EC of the European Parliament and of the Council and repealing Council Directives 78/660/EEC and 83/349/EEC.
Directive 2014/56/EU of the European Parliament and of the Council of 16 April 2014 amending Directive 2006/43/EC on statutory audits of annual accounts and consolidated accounts.
Handelsgesetzbuch (HGB) in der im BGBl Teil III, Gliederungsnr. 4100-1, veröffentlichten bereinigten Fassung, das durch Artikel 5 des Gesetzes vom 5. Juli 2016 (BGBl. I S. 1578) geändert worden ist.
Regulation 1606/2002/EC of the European Parliament and of the Council of 19 July 2002 on the application of international accounting standards.
Affidavit Ich erkläre hiermit, dass ich die vorliegende Arbeit selbständig verfasst und nur die angegebenen Quellen und Hilfsmittel benutzt habe. Wörtlich oder dem Sinn nach aus anderen Werken entnommene Stellen sind unter Angabe der Quelle als Entlehnung kenntlich gemacht. Die Arbeit enthält kein Material, das ich bereits zu einem früheren Zeitpunkt als Prüfungsleistung eingereicht habe. Mir ist bewusst, dass die ungekennzeichnete Übernahme fremder Texte oder fremder Ideen als Plagiat gilt, selbst wenn diese zusammengefasst, umschrieben, gekürzt, oder anderweitig verändert wurden. Die Konsequenzen eines Plagiats sind mir bekannt. Die möglichen Konsequenzen umfassen, unter anderem, ein Nichtbestehen der Bachelorarbeit, den Ausschluss von weiteren Prüfungsleistungen im Studiengang, oder zivilrechtliche Konsequenzen, die mit dieser eidesstattlichen Erklärung verbunden sind. Diese Konsequenzen können auch nachträglich zur Anwendung kommen, also nachdem die Arbeit angenommen und korrigiert wurde. (I declare that I wrote this thesis independently and on my own. I clearly marked any language or ideas borrowed from other sources as not my own and documented their sources. The thesis does not contain any work that I have handed in or have had graded as a Prüfungsleistung earlier on. I am aware that any failure to do so constitutes plagiarism. Plagiarism is the presentation of another person's thoughts or words as if they were my own—even if I summarize, paraphrase, condense, cut, rearrange, or otherwise alter them. I am aware of the consequences and sanctions plagiarism entails. Among others, consequences may include nullification of the thesis, exclusion from the BA program without a degree, and legal consequences for lying under oath. These consequences also apply retrospectively, i.e. if plagiarism is discovered after the thesis has been accepted and graded.)
Name: York Philipp Wagner Title of the thesis: Control is good, trust is better? Trust and its implications on agency problems. An empirical study of audit fees and management remuneration in Europe. Date: Signature: