Accounting for varieties of capitalism: The case against a single set of global accounting standards

The British Accounting Review 42 (2010) 137–152

Contents lists available at ScienceDirect

The British Accounting Review journal homepage: www.elsevier.com/locate/bar

Accounting for varieties of capitalism: The case against a single set of global accounting standardsq Martin Walker Manchester Business School, University of Manchester, Booth Street West, Manchester M15 6PB, United Kingdom

a r t i c l e i n f o

a b s t r a c t

Article history: Received 22 October 2009 Received in revised form 29 March 2010 Accepted 6 April 2010

This paper argues that the optimal design of accounting standards may depend on the institutional characteristics of the political and economic system. There are several varieties of capitalism, and it is not obvious which of these varieties is best. Moreover, the existence of different varieties of capitalism arguably promotes economic progress. This being the case the paper urges a cautious approach to the imposition of a single set of global accounting standards for all companies. The forced adoption of single form of accounting runs the risk of severely restricting the different forms of capitalism that can develop. It also privileges one particular way of doing business over alternative forms that currently exist or, more importantly, may exist in the future. In effect the forced adoption of a single form of accounting can be viewed as a form of restrictive practice that prevents alternative and superior ways of doing business from taking shape. International accounting standards optimised for stock market based capitalism are not necessarily optimal for other forms of capitalism and, since stock market capitalism has lost credibility as a way of doing business, the world may be better served by encouraging alternative forms of capitalism to develop with accounting standards tailored to their needs. Ó 2010 Elsevier Ltd. All rights reserved.

Keywords: Conceptual framework Economics of uncertainty Evolution of institutions International accounting standards Political economy of institutions Varieties of capitalism

“As we know, there are known knowns. There are things we know we know. We also know there are known unknowns. That is to say we know there are some things we do not know. But there are also unknown unknowns, the ones we don’t know.” Donald RumsfelddFeb. 12, 2002, US Department of Defense news briefing. We live in interesting times. The UK, in particular, has suffered a major setback as a result of a worldwide recession caused by excessive credit expansion. These events have revealed enormous gaps in our understanding of how financial markets and financial governance systems work. Viewed through the philosophical prism of Rumsfeld we can safely conclude that the number of known unknowns has increased dramatically. Also the number of known knowns has decreased, in the sense that things we thought were known are now known to be unknown. Of course it is unknown whether the number of unknown unknowns has also increased. All we can say with certainty is that we now know the world is more uncertain than we previously knew. These momentous events have given rise to searching questions about the pace and fragility of stock market based globalisation as an economic and political process. Whereas just a couple of years ago the intellectual and political proponents

q This paper was developed from Martin Walker’s plenary paper delivered at the British Accounting Association annual conference in Dundee following his award of the 2008 BAA/ACCA Distinguished Academic. E-mail address: [email protected] 0890-8389/$ – see front matter Ó 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.bar.2010.04.003

138

M. Walker / The British Accounting Review 42 (2010) 137–152

of stock market based globalisation seemed to be defeating all who stood in their way, the intellectual and political opponents of the rush to stock market based globalisation now have an opportunity to be heard. Moreover even some of the proponents of stock market based globalisation as a long term aim, are beginning to ask more intelligent questions about the most appropriate pace with which this aim should be achieved. In particular the wisdom of replacing national institutions and regulatory safeguards built up over centuries with largely unrestricted free global markets in the presence of highly fragile global governance structures is beginning to run into serious opposition. One major issue requiring attention in relation to these events is the role of financial reporting in general and financial accounting standards in particular. The recent credit crunch has undermined confidence in the competence of accounting standard setters worldwide. How can it be that the financial reports of banks and other financial institutions failed to alert investors to the massive risk exposures they faced, often on the scale of billions of dollars? Are the accounting standards of the Financial Accounting Standards Board (FASB) and the International Accounting Standards Board (IASB), and other major standard setters good enough? Are they fit for purpose? World leaders have called for improved international accounting standards. On the 15th of November 2008, following a meeting of the G20 Heads of State, leaders of the World Bank, the International Monetary Fund, the United Nations, and the Financial Stability Forum, the participants published a Declaration of the Summit on Financial Markets and the World Economy. This declaration outlined an action plan for the reform of international financial institutions and markets. This action plan contained measures for strengthening transparency and accountability. The G20 plan calls for the “key global accounting standards bodies” to work intensively toward the objective of creating a single set of high-quality global accounting standards (see Appendix 1). Implicit in this statement is the view that current international accounting standards are not high quality. In the light of the recent collapse of the global financial system it is difficult to disagree with this view. However, the call by the G20 for the key accounting bodies to work toward a single high-quality global accounting standard is, in my view, highly questionable. When major financial scandals or collapses occur politicians often respond in haste. Sometimes the cures they propose turn out to have undesirable and unanticipated consequences e.g. Sarbanes–Oxley. Sometimes the proposed cure turns out to be a curse. Thus the purpose of this lecture is to explain why I believe that the call of the G20 for a single accounting standard is unwise. Why, that is, I think it would be a curse. My arguments are conditioned by a conviction that accounting policy, like all areas of social and economic policy, should be shaped as much as by acknowledging what we do not know as by what we do know. A call to frame a single set of accounting standards, based on an assumption that we know all we will ever need to know in order to frame such standards, seems to me to fail to comprehend the nature of the problem. 1. The wisdom of creating a global monopoly accounting standard setter The right to set the accounting standard for the entire world would represent a very significant monopoly power. It is somewhat ironic that the political supporters of unrestrained and competitive markets appear to think that free trade requires a monopoly over the setting of accounting standards. The purpose of this paper is to argue that there are fundamental reasons for doubting the wisdom of current attempts to establish a single set of global accounting standards. However, before discussing the fundamental arguments against a single set of global standards it is important for me to point out that my arguments do not involve or imply any judgements on my part about the motivations or competence of the IASB or the FASB. With regard to motivations I do not doubt the generally good intentions and trustworthiness of these two bodies. Also I would agree that both of these bodies have done many good things for the quality of financial reporting across the world. With regard to competence one can point to concerns about the track record of FASB and IASB. In particular, the recent financial collapse has occurred under the watch of both the FASB and the IASB and this has inevitably given rise to questions about their competence. However, even if recent events had raised no doubts about the ability of existing regulators to set standards, there are in my view more lasting and more fundamental reasons to doubt the wisdom of granting any accounting standard setting body a monopoly power to set a single set of global standards. Even if FASB and IASB were found to be entirely blameless for the current fiasco, which they may be,1 there are fundamental reasons for resisting the call for a single set of global standards. In this paper I present three mutually reinforcing arguments against having a single set of global standards. I start by pointing to the existence of significant disagreements between accounting experts on the scope and objectives of financial reporting. I then go on to discuss the economic literature concerned with how economies respond to risk and uncertainty. This discussion highlights how markets may fail to achieve optimal outcomes under conditions of risk and uncertainty. The role of firms as a response to market failure is highlighted, and the possibility of there being different institutional designs, across which the role and importance of stock markets varies, is explained. Finally, I discuss the literature on Varieties of Capitalism, and explore reasons why the variety of accounting models needs to be large enough to accommodate more than one variety of capitalism. The final section discusses the implications of my arguments for the future of international accounting standards.

1

This is a known unknown.

M. Walker / The British Accounting Review 42 (2010) 137–152

139

2. Fundamental disagreements over accounting concepts Accounting standard setters, professional accountants, and academic accountants do not agree on what constitutes good accounting. There are fundamental disagreements about the scope of financial reporting, the objectives of financial reporting, and the means of achieving these objectives. In the last four years, the IASB and FASB have made strenuous efforts to win assent for a joint conceptual framework for financial reporting. The responses the IASB and the FASB received to this consultation exercise reveal significant disagreements about the scope, objectives, and the means of achieving the objectives of financial reporting. Fig. 1 reproduces the informed response of the American Accounting Association (AAA), which identifies a number of major concerns. One of the concerns raised by the AAA was the attempt by FASB/IASB to subsume the stewardship objective under the general heading of decision-usefulness rather than acknowledging stewardship as a separate, some would argue prime, objective of financial reporting.

Fig. 1. Letter from the AAA’s financial accounting standards committee Nov 29, 2006.

140

M. Walker / The British Accounting Review 42 (2010) 137–152

Fig. 2. Comment letter on the draft conceptual framework. 2006.

Many of the written responses to the 2006 draft of the conceptual framework (IASB, 2008) challenged the IASB and the FASB to recognise stewardship as a separate financial reporting objective. Strong disagreement with the stance of FASB/IASB were voiced by Governance for Owners, the III Working Group, the Institut Der Wirtschaftprufer, the UK Audit Commission, the Danish Shareholders Association, the Financial Reporting Advisory Board, the Investment Management Association, the Institute of Chartered Accountants of Scotland, the National Audit Office, the Conseil National de la Comptabilité, the Association of British Insurers, the German Accounting Standards Board, the UK Accounting Standards Board, the European and Financial Reporting Advisory Group, and others. The response of the UK Shareholders Association was unusually blunt in its response to the FASB/IASB proposals, and even raised serious doubts about the underlying motives of the FASB and IASB (Fig. 2). In the second (IASB, 2006) draft of the joint conceptual framework the FASB/IASB partially, but one senses reluctantly,2 accepted some of these arguments. However a number of authorities have pointed to the half hearted nature of this response. Highly critical remarks on the 2008 revised conceptual framework were received from the Austrian Financial Reporting and Auditing Committee, the British Accounting Association Special Interest Group in Financial Accounting and Reporting (FARSIG), the Institute of Chartered Accountants in England and Wales, the Institute of Chartered Accountants of Scotland, G100, Shell International B.V., CIPFA, the Audit Commission, the BBA, the Conseil National de la Comptabilité, the Association of British Insurers, the Hundred Group, BNP Paribas, the UK Accounting Standards Board, the Investment Management Association, the CBI, the European Financial Reporting Advisory Group, the Federation of European Accountants and others. Whatever one’s personal views on the stewardship controversy, it is clear from a reading of the submissions to FASB and IASB that the meaning and importance of stewardship is a fundamental matter on which accounting academics, accounting standard setters, professional accountants, and the users of financial reporting disagree. Moreover, the controversy over stewardship is just one of several areas where there is disagreement over fundamental accounting issues. Other key issues include: the relevance and reliability of fair value accounting measures, the nature and usefulness of accounting conservatism, the role of disclosure versus recognition and measurement in the financial statements, the framing of performance statements and the importance of earnings numbers within these, the entity concept versus the equity concept etc. These are not minor areas of disagreement. The fair value debate is concerned with how asset values are measured in the financial statements and/or the demand for measures of value other than those used in the audited accounts to be disclosed as other information. The conservatism debate points to a strong implicit demand by capital

2 The new title of the revised exposure draft focuses exclusively on decision-usefulness and makes no concession to the stewardship point of view. Conceptual Framework for Financial Reporting: The Objective of Financial Reporting and Qualitative Characteristics and Constraints of Decision-Useful Financial Reporting Information.

M. Walker / The British Accounting Review 42 (2010) 137–152

141

providers for conservatism that FASB and IASB are reluctant to acknowledge. The recognition versus disclosure issue forces one to confront the extent to which financial statements should contain forward-looking estimates, given that an option always exists for such estimates to be disclosed outside the financial statements. For example, most of the benefits of fair value reporting could be achieved by disclosing fair values as supplementary information. The entity versus equity concept debate has profound implications for the information and control rights of shareholders. In the light of these very substantial and fundamental disagreements between accounting experts it can be argued that forcing a single global consensus may not be the wisest way to proceed. If you are not sure what to do, you do not place all your eggs in one basket. It may be better to allow markets, economies, governments and firms to experiment with alternative systems of accountability based on alternative perspectives about what constitutes good accounting. Moreover, we need to consider whether it is possible to accommodate these disagreements within a single logically consistent conceptual framework. For example, it can be argued that the opposing sides of the stewardship debate have two quite different conceptual frameworks. One side adopts a traditional stewardship perspective the other adopts a value relevance perspective. If this is the case then we need to be aware of the dangers of trying to force a compromise between these two viewpoints. If you attempt to merge two different conceptual frameworks, you do not end up with a better framework. You end up with a logical mess. So it might be better to have two different sets of standards based on two different, but logically consistent, conceptual frameworks, rather than a single set of standards based on a logically inconsistent conceptual framework. 3. Advances in the economics of uncertainty and information 3.1. The market context for accounting standard setting The conceptual frameworks of the IASB and the FASB say very little about the market, organisational, and institutional context of accounting. This is surprising given the very considerable advances in our understanding of the economic, organisational, and institutional context of accounting over the last half century or so. In this section I point to several areas where I believe that accounting standard setters need to update their knowledge of the economic context of accounting in order to better appreciate the forces that condition the optimal design of accounting systems. In the next section I point to areas where accounting standard setters need to update their knowledge of the political and institutional context of accounting. A particularly important advance in Economics is the recognition of the importance of market incompleteness in understanding economic behaviour under risk and uncertainty. Arrow (1964) and Debreu (1959) showed how, in principle, the standard general equilibrium models of economics could be extended to incorporate uncertainty and the passage of time. This is achieved by assuming that it is possible to agree contracts to exchange goods and services contingent on time of delivery and contingent on the state of the world at the time that delivery is agreed. Implicit in the Arrow and Debreu approach is the assumption that all economic agents are aware of all possible future states of the world that can possibly occur. Their model does not assume that all individuals agree on the probabilities of future possible events. However they do assume that everyone agrees on the set of events that could occur with positive probability. In the words of Rumsfeld they assume that all unknowns are known unknowns. This assumption contrasts with a different approach to the economics of uncertainty championed by Knight (1921) that allows for the possibility that there may also be unknown unknowns. Within the last 50 years or so mainstream economics and finance has worked within the Arrow and Debreu assumption. This work has shown that it is possible to make some progress in understanding economic behaviour under uncertainty within the Arrow and Debreu assumption. However some features of economic and political life cannot be understood within a framework that rules out the possibility of unknown unknowns. In particular the role and evolution of many social and economic institutions cannot by fully understood without allowing for the possibility that they evolved, at least in part, as a response to Knightian uncertainty. Thus a weakness of the Arrow and Debreu set up is that it cannot explain how the presence of uncertainty leads to the creation of certain types of institutions and contractual forms. I will say more about this in later sections. Within the Arrow and Debreu set up, a key distinction that has led to a number of useful insights about economic behaviour under uncertainty is the distinction between complete and incomplete markets. Crudely put markets are perfect and complete if the number and variety of competitively traded securities is sufficiently rich to permit every economic agent to achieve any pattern of state contingent consumption that they choose. In other words there is a finite competitive market price for any conceivable state contingent delivery pattern of goods and services.3 For example, if there are S conceivable states of the world, then for every state, s, it is possible, by constructing an appropriate portfolio of the set of traded securities, to construct a payoff pattern that pays off one unit of consumption in state s and zero consumption in all other states. Beaver and Demski (1979) were among the first accounting academics to recognise the importance of the assumption of perfect and complete markets for accounting. Their paper demonstrated that financial accounting cannot be a fundamental measurement discipline. In particular, they showed that the central feature of financial reporting cannot be the fundamental measurement of income. There will be no demand for fundamental income measurement when markets are perfect and

3

For a formal definition of complete security markets see Christensen and Feltham (2003, p. 147).

142

M. Walker / The British Accounting Review 42 (2010) 137–152

complete and that, in general, a fundamental income measure will not exist when markets are imperfect and incomplete. Thus, they concluded, a motivation for income reporting must rest on the improved resource allocations which such reporting provides and not on such criteria as “more income is better than less”. Moreover, they conclude (page 45), “choice of an income rule cannot be resolved by applying fundamental measurement arguments.” At the time their paper was written, relatively few accounting academics took seriously the possibility that accounting book values might be the main purpose of financial reporting. Beaver and Demski were writing within the then mainstream view of the world that income measurement was the central measure that was the ultimate and final product of the financial reporting process. However their argument applies equally well to any attempt to portray accounting as the measurement of some relevant value. Just as the choice of an income rule cannot be resolved by appeal to measurement arguments so the same can be said of asset valuation rules such as the choice between historic costs and fair values. All such arguments ultimately turn on the economic costs and benefits of alternative information signals. Thus we can rephrase the above conclusion as “choice of an income or asset rule cannot be resolved by applying fundamental measurement arguments.” Similarly, the choice between conservative accounting, and what IASB/FASB call “neutral” accounting systems, is also a matter of economic choice. This choice should not be determined by appeal to fundamental measurement arguments, rather it should be determined by the preferences of investors. Unfortunately, the IASB and the FASB seem to me to be somewhat confused on this matter. On the one hand, FASB and IASB acknowledge that accounting standards must be subject to a cost benefit test. On the other hand the new conceptual framework simply asserts that neutrality is essential for “faithful representation” of an economic phenomenon. However, there is persuasive international evidence that the capital markets exhibit a strong implicit demand for conservatism (i.e. the market views the benefits of conservatism as being greater than the costs). For example Bushman and Piotroski (2006) find that firms in countries with strong judicial systems, that respect the information demands of shareholders, reflect bad news faster in earnings than firms with weak judicial systems. Similarly a recent paper by Giner et al. (2009) demonstrates that there is an international demand for accounting conservatism. This paper shows that international portfolio investors favour countries with conservative accounting systems. The paper also shows that international investors tailor the form of their investments to the level of accounting conservatism. Thus there seems to be a clash between the IASB/FASB’s, somewhat theological, commitment to “neutrality”, and the commitment by the IASB/FASB to examine the costs and benefits of conservatism. Unfortunately the FASB and IASB do not have any costs benefit calculus in place to test the relative costs and benefits of conservative versus neutral accounting. As things stand it appears that IASB and FASB are only willing to bow to the demands of capital market investors where these demands are consistent with the conceptual framework. In general the ranking of accounting alternatives generated by applying the logic of the conceptual framework should be in line with the ranking of accounting alternatives based on estimates of the markets’ preferences. Where these ranking disagree (i.e. when the conceptual framework fails the market test) then one would expect to see either revisions to the conceptual framework, or firms choosing to provide other information to meet the market demand, or (assuming this is not ruled out by the granting of a monopoly power to a single standard setter) a rival set of accounting standards to emerge. 3.2. The market completion role of information A weakness of the Beaver and Demski (1979) paper is that they sought to compare a given structure of incomplete markets with the theoretical ideal of perfect and complete markets. What Beaver and Demski did not allow for is that the set of markets is itself endogenous. In general the level of completion of markets, and the degree of competition within a given market is a function of the quality of information available to economic agents. As explained below, the existence of markets is driven to a significant extent by the guarantee of reliable information to both sides of the transaction. Research on the economics of uncertainty has shown that there are only three ways in which an improved information environment can lead to welfare improvements. Firstly, improved forecasts of the future state of the world enable improved economic choices ex ante. For example, if you know for certain that nuclear fusion will be working well by 2020 you will not invest in conventional nuclear power today. Secondly, if economic actors can forecast what information will be available in the future then this increases the number of risk sharing opportunities available to society. To the extent that future contingencies can be anticipated, economic actors can co-insure against such events. This is most obvious in the case of state confirmation information. For example, suppose that everyone knows that at some future point in time, the economy will be either in state A (boom) or state B (bust). If everyone is confident that in future it will be possible to distinguish, say, between state A and state B, then it becomes possible for claims to consumption in state A to be traded against claims for consumption in state B thereby enabling improved risk sharing and facilitating trade between people with different beliefs about the relative likelihood of state A and state B. The third and final way in which improved information can improve risk sharing only arises when the set of tradable state contingent claims is incomplete ex ante (i.e. when the number of linearly independent claims is less than the number of possible states of the world). Under such conditions it may nevertheless be possible to effectively complete the market by allowing

M. Walker / The British Accounting Review 42 (2010) 137–152

143

State 1

State 2

State 3

Initial Trades Further trades

State 4 Final Settlement

Fig. 3. A world with 4 states of the world and intermediate trading of two securities.

additional rounds of trading in the less than complete set of trade-able claims. The possibility that improved intermediate information might improve risk sharing can be traced back to Arrow (1964), Kreps (1982), and Duffie and Huang (1985). To explain how this works, suppose there are four possible states of the world (see Fig. 3). Then if there is no intermediate information, four linearly independent tradable securities will be required to enable economic agents to achieve all possible state contingent patterns of consumption. However suppose there are only two traded securities. In this case then the market is incomplete and less than perfect risk sharing can be achieved. Suppose though that, at time point 0, everyone knows that intermediate information will be available and everyone knows that there will be a further opportunity to trade when this information is released. Then it can be shown that the possibility of re-trading increases the set of final state contingent consumption patterns that can be implemented. For example suppose the intermediate information system can perfectly distinguish between two events: event A (either state 1 or state 2 will occur) and event B (either state 3 or state 4 will occur). Then it can be shown that re-trading based on this intermediate information will effectively complete the market.4 Note that the risk sharing role of information depends crucially on the ability of investors to predict what the quality of information will be at future points in time. Among other things this implies that the producers of the relevant information must be able to pre-commit to produce the information and to make it available to the contracting parties. An interesting issue worthy of attention is the extent to which producers of accounting information can pre-commit to producing fair value information compared to historic cost accounting information. It seems likely to me that it may be easier to pre-commit to produce historic cost accounting information (or current cost accounting information based on government sanctioned indices) than it is to pre-commit to produce fair value information. However I would be interested to see this tested. 3.3. The nature of the firm So far we have seen that one solution to the lack of state contingent claims markets, is to improve the flow of information and to encourage trading on that information. Another important response to imperfect or incomplete markets is to replace the market by a firm. Coase (1937) established that firms exist because of the costs of using the price mechanism. Ball (1989) is a systematic attempt to work out the implications of Coase’s insight for understanding of accounting and auditing as a set of institutional practices. Ball argues that, since firms exist to overcome the limits to markets, then firms need to find a replacement for market prices as a basis of coordinating and controlling activities within the firm. He introduces the notion of quasi-prices to capture the idea of coordination by a price system that is largely unobservable, often implicit rather than explicit, and internal to the firm. He points out that if quasi-prices were the same as the prices that would have been agreed by trading in markets then the firm would be irrelevant. There would be no reason for the firm to exist. Firms exist because their internal implicit quasi-price systems are superior to market prices. This point is important in relation to accounting and accounting standard setting. Understanding that firms exist in order to replace poorly functioning or non-existent markets points to the need for accounting measures that can be used to plan and control economic activity in the absence of the market. The idea that “fair values”, which are largely based either on replacement cost, or realisable market values, should form the bedrock of accounting measurement needs to be reconsidered in this light. There is an obvious gap in the logic of an argument that says we should use fair values, largely based on market prices, to measure the performance of entities that exist because of market imperfections. If firms exist because markets don’t work then what is the logic of using market prices to measure their performance? So far as I am aware Ball (1989) was the first paper in the accounting and auditing literature to discuss the functional completion of contracts in contrast to contracts complete by enumeration. Contracts that are completed by enumeration

4

Chapter 6 of Christensen and Feltham (2003) explains the formal conditions under which multi-period markets will be dynamically complete.

144

M. Walker / The British Accounting Review 42 (2010) 137–152

attempt to enumerate the set of all possible states ex ante and specify a one to one mapping from states to agreed payoffs i.e. the contract spells out ex ante the payoffs to every party to the contract in every conceivable state of the world. In contrast Ball’s “functionally-completed contracts” set out ex ante a set of implicit and explicit rules, practices and procedures for determining payoffs ex post. Such arrangements often require some form of adjudication function that has some degree of independence from the parties to the contract, and that is committed to implement some principle of “fairness” in the ex post sharing of payoffs. Working from the transaction costs perspective of Coase (1937) and Ball (1989), Waymire (2009) points out a fundamental limitation of the FASB/IASB conceptual framework. Specifically, he points to an “exchange guidance” demand for accounting information which, he argues, is more fundamental than either the stewardship perspective or the value relevance perspective that drives FASB/IASB. With regard to the conceptual framework of the FASB and the IASB he concludes (p. 59) “These efforts, if successful, will minimize the stewardship demand for information by eradicating Conservatism in favour of Neutrality...They will also render the Matching Concept a dead letter within accounting practice, lessen Objectivity by allowing balance sheet account revaluations based on hypothetical (i.e., non-consummated) transactions, and have the paradoxical effect of undermining the usefulness of accounting income, which is the metric actually used most often for valuation purposes.”. We saw above that Beaver and Demski (1979) were unable to explain why a demand exists for measures of income. Can we come closer to providing a rationale for income measurement by recognising the role of the firm as an efficient contracting technology, along the lines suggested by Ball (1989) and Waymire (2009)? I believe that we can. However, in order to achieve this I believe we need to recognise two other features of the modern company, in addition to the general contracting role it fulfils. First, it is important to recognise that firms compete against other firms. Moreover, since the main reason for the existence and survival of firms is the quality and uniqueness of their quasi-price systems, it is bound to be the case that firms will want to maintain a good deal of secrecy about how their internal contracting technologies. Thus the nature of the firm implies a demand for commercial secrecy about how its internal affairs are managed. Black (1993) points out that one of the contradictions of stock market based capitalism is that the ability of a firm to compete often depends on its ability to keep information secret from competitors and other third parties (e.g. the tax authorities, and trade unions). However the demands of the stock market require all investors to be equally well informed, at least with respect to information relevant for valuing the company. To some extent capitalism solves this problem by providing summary indicators that communicate the value relevant information of company insiders to external investors. Black (1993) argues that a measure of sustainable earnings is the single most important summary value statistic. However Black (1993) does not explain why periodic measures of performance, such as earnings for the financial year, are demanded. An answer to this question can perhaps be provided by taking account of the conflicts of interest that arise between insider managers and outside investors. In particular the existence of an agency relationship between management and investors opens up the possibility that management will attempt to control the flow of information so as to present their performance in the best possible light. For example, they may attempt to delay the reporting of bad news and they may attempt to release information gradually (perhaps starting with a small number of “understanding” investors) so as to smooth the share price impact of their information. A (quasi) contractual pre-commitment to report on a strict periodic basis serves to limit the scope for such “selective” reporting. Agency considerations also help one to understand why investors seek to relate indicators of current earnings to estimates of capital invested in the firm i.e. the demand for measures of profitability. Investors need to be able to calculate how well the capital invested in the firm has been utilised relative to the cost of supplying that capital. Such calculations require information that allows the current level of earnings to be related to the amounts invested in the firm in the past, including previous levels of retained earnings. Thus, the demand for periodic performance measures arises because there is a stewardship problem. Viewed in this light the attempt by FASB/IASB to provide a conceptual framework for periodic financial reporting which fails to acknowledge stewardship as a fundamental feature of accounting looks illogical. If stewardship is the reason why external investors demand periodic reporting, then it makes no sense to design a periodic accounting system based on the assumption that stewardship is not a central concern. 3.4. The nature of financial systems Any attempt to provide a conceptual framework for Financial Accounting must reflect the way real world financial systems work or, looking at it another way, the factors that cause real world financial systems not to work. Allen and Gale (2000) provide a thought provoking guide to the main characteristics that distinguish one type of financial system from another. This sub-section is based on the insights provided by their work. The FASB/IASB conceptual framework of the financial system is rudimentary. It is based on five unrealistic assumptions 1. Security Markets Dominate the Financial System. The financial system is a set of markets for the securities of limited companies. 2. Corporate Finance. The main sources of finance for companies are the security markets. 3. Individual Ownership of Corporate Securities. The investors in security markets are private individuals.

M. Walker / The British Accounting Review 42 (2010) 137–152

145

4. Security Prices Supply All Relevant Information. Security market prices communicate all the information that investors and corporate managers need to know in order to make economically efficient investment choices and optimal risk sharing. 5. Corporate Governance by the Market. The trading choices of investors can be relied upon to govern the investment and operating choices of company managers. Given zero transaction costs and the possibility of instantaneous rebalancing of portfolios, any manager who failed to maximize firm value would be instantly replaced. If these assumptions were true then the assumption of FASB/IASB that information for stewardship is of second order importance would be correct. However in this world, there would be no demand for any accounting information of any kind. Market prices would convey all the information any investor would ever need to know. However, the last 30 years or so of research on corporate finance, corporate governance, and financial economics suggests that real world financial systems are not like this at all. First of all, financial systems in the real world involve a plethora of banks and other financial intermediaries. In this world information is needed not just to monitor and control the behaviour of company managers. It is also needed to monitor and control the behaviour of banks and other financial intermediaries. Second, in the real world, corporate investment is often financed from retained earnings. There is no automatic repayment of the capital invested in projects to the investors in those projects. Companies invest in a sequence of projects over time using the profits from past investments to finance current investments. In this world information about the profitability of current investments and the potential profitability of further investment in the firm is needed to allow investors to work out whether companies should be allowed to reinvest past profits or whether they should be forced to pay them out as dividends. For companies with debtholders as well as shareholders information is also needed to ensure that dividends are not paid out of capital, also most companies do not issue securities for trade. It is only a small percentage of companies that choose to have their equity actively traded. Even within the set of firms with actively traded equity, many of these borrow money from banks rather than, or in some cases as well as, issuing tradable debt. Third, most private individuals do not invest directly in corporate securities. They often invest their personal savings, including their pension funds, in banks and other financial intermediaries. Investment decision making in companies is often delegated by private investors to financial intermediaries. Fourth, security prices supply all the information needed for efficient investment and risk sharing when state contingent claim markets are perfect and complete. In reality many markets are imperfect and many of the markets that would be needed to achieve perfect risk sharing do not exist. Markets fail to exist for a variety of reasons including: the limited ability of real people to comprehend the complete set of states of the world; the limited ability of people to agree and enforce contracts on the states that are comprehensible; and asymmetries in the distribution of information leading to adverse selection problems; moral hazard problems arising when investors delegate decisions to others whose interest are less than perfectly aligned with their own. Fifth, there is a tension between optimum risk sharing through security markets and corporate governance. Efficient risk sharing typically requires all investors to hold highly diversified portfolios in which the securities of one particular firm are of limited significance. Investors with little or no stake in a company have little or no incentive to collect the information needed to ensure that corporate managers are doing a good job. Some kind of institutional “fix” is needed to allow investors to concentrate their voting power in order to force managers to behave in investors’ interests. In some economies, this is achieved by encouraging takeover contests whereby an underperforming management team can be replaced by a new one. In other economies control over management is achieved by financial intermediaries who hold a large percentage of the claims on future cash flows. For example in the UK some institutional investors hold a large percentage of the equity through which they are able to exert influence of company managers. In other countries, like Germany, it is often the company debtholders, notably the banks, who fulfil this role. Allen and Gale (2000) show that economies differ considerably in the way that the finance and governance of corporate investment is achieved. They also differ in the ways that economic risks are ultimately shared. In addition to the cross-sectional sharing of risk between individuals, they also point to the importance of having mechanisms in place to ensure efficient risk sharing over time. For example, how does society ensure that risks do not fall disproportionately on one generation of a society? Different institutional set ups affect the way that risks are/are not shared across the generations. Finally, they show that different institutional set ups can influence the probability and the scale of financial crises such as the recent credit crunch. For present purposes we need to note four key points arising from this analysis 1. There is considerable worldwide variation in the organisational and institutional forms of financial systems; 2. Some of this variation may be accidents of history, but most of it reflects the outcome of successful adaptations to the needs of a particular economy (i.e. the variations embody valuable institutional capital); 3. Financial systems are complex and highly interdependent. The components of a financial system tend to have been adapted to each other over time. Importing a component that was adapted to another financial system runs a risk that it may not be well adapted to its new institutional environment. 4. The demands for forward-looking value relevant information and for information for stewardship and control purposes may differ considerably between alternative financial systems. It is not safe to assume that an information system optimised for a stock market based economy will also be optimal for a different financial system.

146

M. Walker / The British Accounting Review 42 (2010) 137–152

Reflecting on the above one can begin to understand that the role of information in the economy goes far beyond the relatively routine role of forecasting future cash flows. The ability of an economy to share risks and ameliorate the destructive effects of moral hazard and adverse selection on the functioning of markets and firms is, in general, far more important than the relatively minor role of facilitating the secondary exchange of corporate securities. The focus of FASB and IASB on the information needs of secondary traders is way too narrow. 4. Advances in the political economy of institutions “Conformity can be costly in a world of uncertainty. In the long run it produces stagnation and decay as humans confront ever new challenges in a non-ergodic world that requires innovative institutional creation because no one can know the right path to survival. Therefore, institutional diversity that allows for a range of choices is a superior survival trait, as Hayek has reminded us”. North (2005, p. 42). A recurring theme in accounting thought is the link between accounting and the development of capitalist institutions. Hopwood, for example, has consistently stressed the important causal links between developments in accounting and developments in organisations and institutions (see e.g. Hopwood (1983, 1987, 2007)). More recently the works of Waymire and Basu (2008), and Chapman, Cooper, and Miller (2009) have highlighted the possibility that accounting may both shape and be shaped by institutional developments. In the context of thinking about the design of the accounting system for the whole world, it is important to be aware of the nature and role of institutions in the economy, and the way that institutions evolve. Furthermore, as scholars of accounting, we need to understand how accounting technologies both shape and are shaped by economic and political institutions. To borrow an insight from Hopwood (2007), we need to understand accounting within its institutional and historical context. For insights on the nature and role of institutions in the economy I highlight the relatively recent literature on varieties of capitalism. I find this literature helpful because it gels fairly well with the economic frameworks discussed above. However I should mention that there are other literatures, of a more sociological nature, that have contributed useful insights to our understanding of the nature and role of institutions in the economy and society. I do not discuss these literatures here because of my extremely limited knowledge of sociological theories. Chapman et al. (2009) provide a useful introduction and guide to sociological perspectives on accounting, organisations and institutions. Whitley (2007) shows how sociological methods can be applied to understand the co-evolution of firms and institutions. For insights on the importance of and the evolution of institutions, I have learned a lot over the years from reading the works of Hayek (notably Hayek (1944, 1948, 1979)). More recently I have benefitted considerably from reading North (2005). North himself was also hugely influenced by Hayek. 4.1. Varieties of capitalism Research on the comparison of alternative economic systems has been a major area of research in the social sciences for at least a century. In the early part of the twentieth century the great political and economic debate was concerned with comparing capitalism against socialism/communism. Hayek (1944, 1948) contain decisive contributions to these debates. In the second half of the twentieth century growing post war prosperity in the West, gave rise to a substantial body of literature concerned with the identification, analysis, and comparison of alternative forms of capitalism. The growth rate of this literature increased following the demise of the Soviet Union, partly because the former intellectual advocates of state sponsored socialism had to find a new intellectual project that was relevant to the “post socialist” order. For present purposes the main point to emerge from this vast literature is the importance of coordinating institutions other than markets and individual firm hierarchies. Such arrangements have come to be known as non-market coordinating institutions (NMCIs). Important examples of NMCIs are industrial trade bodies, cross-firm arrangements for the education and training of the workforce, and various arrangements to share R&D costs and information involving multiple firms. The contributors to this literature disagree on the relative importance of the many different types of NMCIs. However, they generally agree that an important contextual factor that distinguishes different forms of capitalism is the way that their NMCIs are structured and operated. Thus we need to augment the markets and hierarchies view of an economy of Coase, with a richer view that allows for NMCIs. This opens up the possibility of economic coordination between firms (and between firms and their stakeholders) being achieved partly through NMCIs and partly through markets. An important stream of research on alternative forms of capitalism has come to be known as the Varieties of Capitalism (VOC) approach. An influential volume edited by Hall and Soskice (2001) established the VOC approach as a landmark contribution to the literature on comparative capitalism. In particular the VOC approach served as an intellectual meeting point between researchers interested in the economics of business strategy and researchers interested in the political economy of alternative forms of capitalism. The unique contribution of the VOC approach is its recognition of the role of NMCIs as part of the core competencies of both firms and national economies. The starting point of the varieties of capitalism approach is the adoption of a relational view of the firm that moves beyond the idea that all economic activity is coordinated either within the firm, or by market transactions. Conventional models of business strategy, such as resource-based theories of the firm, focus on the development of core competencies within the firm

M. Walker / The British Accounting Review 42 (2010) 137–152

147

that improve its ability to compete in the market place. These traditional models assume that the relations between the firm and its customers and suppliers consist exclusively of spot market transaction for goods and services. Relational views of the firm allow for the development of firm competencies that rely on the development of long term relationships of various kinds between the firm and its key stakeholders, the firm and other firms, and the firm and society. These relationships are created and managed through various NMCIs. Hall and Soskice (2001, p. 6–7) identify five spheres in which the development of institutional relationships serve to enhance the core competencies of the firm. These are: 1. The internal structure of the firm Firms differ in the allocation of decision and control rights. They also differ in the types of performance measurement systems they use and in the ways in which rewards and promotion are linked to measures of performance. Hall and Soskice (2001, p. 29) point out that in liberal market economies “top management normally has unilateral control over the firm, including substantial freedom to hire and fire.” Such economies also tie top management rewards to short term measures of stock market performance. 2. Corporate and financial governance As explained above, countries and firms differ in the extent to which they rely on the provisions of risk capital through stock markets relative to long term bank finance. They also differ in the degree of separation of ownership and control, and the extent to which the market for corporate control through mergers and acquisitions is structured and regulated. 3. Industrial relations Countries differ in the roles and powers of trade unions and other employee organisations. Wage bargaining structures differ in terms of the extent to which wages are bargained for through supra firm relationships. Countries differ in the extent to which employment protection laws are enacted and enforced. 4. Vocational training and education Investment in high-quality industry and business specific skills requires significant long term commitments by employees and employers. The provision of a pool of well-trained employees for a specific industry requires cooperation between employers including agreements on how to share the costs of training provision. 5. Inter-firm relations Firms differ in the extent to which suppliers and customers are involved in long term relationships. Long terms relationships facilitate information sharing and are one possible solution to hold up problems.5 A particularly important set of long term relationships are those that govern the development of new technology and knowledge. Technical and quality standards are often agreed through inter-firm networks. Whitley (2007) provides an authoritative account of the way that different forms of capitalism support technological development. According to Hall and Soskice (2001) the way in which these five spheres operate serves to distinguish one form of capitalism from another. Central to the VOC thesis is the idea of institutional complementarities. For example, labour market arrangements that encourage workers to invest in firm specific human capital, for example, are complemented by corporate financing models which encourage investors to take a long term view. What Hall and Soskice call patient capital. Hall and Soskice (2001, p. 17) define institutional complementarity as follows: “.two institutions can be said to be complementary if the presence (or efficiency) of one increases the return (or efficiency of) the other.” Different varieties of capital are characterised by specific clusters of complementary NMCIs. The VOC approach identifies two polar types of capitalism: liberal market economies (LMEs) and coordinated market economies (CMEs). Liberal market economies are those that are close to the Coasian economy illustrated above (i.e. those in which NMCIs have a limited role). Coordinated market economies are those in which NMCIs play a major coordinating role in addition to the coordination achieved by markets and hierarchies. The efficiency of a coordinated market economy depends on the extent to which its NMCIs are complementary. The VOC approach has been highly influential and has provoked many fruitful debates. Hancké, Rhodes, and Thatcher (2005, chap. 1) provide a useful overview of these debates, and concludes that the VOC approach has (p. 38) “.significantly enhanced our understanding of the world”. However Hancké et al. also point to an important limitation of the early VOC literature, which paid limited attention to the role of the state in the coordination of economic activity. In some countries the state may actively encourage the growth of NMCIs. However, in other countries, the state may work against NMCIs if powerful state actors view them as a rival to state

5 A hold up problem arises when one partner in a supply chain partnership can hold the other partner to ransom once the other partner has committed relevant specific investment. For example, if a car parts manufacturer invests in highly specific machinery, they open themselves up to being exploited by the purchasers of their car parts once the relationship specific investment is in place. See Klein, Crawford, and Alchian (1978) for further discussion of this well known example of the hold up problem.

148

M. Walker / The British Accounting Review 42 (2010) 137–152

power or to the power of state funded bureaucracies. For example, if NMCIs are capable of achieving improved levels of coordination in ways that would eliminate the need for a bloated state civil service, it is possible that the state civil service will work against the formation of NMCIs or that they will work to create NMCIs that provide highly paid jobs for themselves. For example, the penchant of the UK government for state appointed “national tsars” and state controlled quangos may reflect a reluctance of the UK to encourage a more decentralised approach to the creation of voluntary NMCIs. 4.2. The contractual and informational arrangements of NMCIs A general issue that emerges from recognising the importance of NMCIs is that they require the enforcement of cooperative agreements between diverse parties. Often the agreements and their enforcement arrangements will be implicit rather then explicit, and often they will be maintained on the basis of trust and reputation. However, ultimately such cooperative institutions only survive if all of the key stakeholders are able to persuade themselves that the expected long run benefits from continued cooperation exceed the expected long run costs. This raises a number of research questions 1. How do NMCI members keep track of the costs and benefits of the NMCI? 2. How do NMCI members keep track of the way that the net benefits have been shared in the past between the NMCI members? 3. How do NMCI members establish mutually consistent expectations about future costs and benefits? 4. How do NMCI members establish mutually consistent expectations about how future net benefits will be shared? In general terms an NMCI can be thought of as a multi-person, multi-period, game under uncertainty (Aoki, 2001). The main issues to be modelled in the game are how do the members decide what to do, how do the members establish mutually consistent expectations about what the costs and benefits will be, and how do they establish mutually consistent expectations about how the costs and benefits will be shared?6 Sustainable and successful NMCIs are those that are able to consistently deliver net benefits and that inculcate a sense of confidence that any net benefits accruing to the NMCI will be fairly shared amongst the members. Thus in capitalist economies with powerful NMCIs a more general type of stewardship problem arises. How are NMCIs made accountable to their members? A key consideration likely to influence the survival of NMCIs in the long run, is the arrangements they operate in order to share unexpected costs and benefits. How do they accommodate the costs and benefits arising from unknown unknowns? Casper (2001, chap. 12) points to a fundamental distinction between the classical and regulatory approaches to contract law. Broadly speaking the classical approach is associated with liberal market economies and the regulatory with coordinated market economies. The distinction between the classical and regulatory approaches comes to centre stage in the context of incomplete contracts. Casper (2001, chap. 12) notes that when events occur that are unanticipated by the parties to a contract the courts in a classical regime will focus on enforcing the written terms of the contract irrespective of the relative contracting strengths of the two parties, and irrespective of the extent to which one side of the contract is disproportionately affected by the unforeseen events. On the other hand under a regulatory contract law approach the courts will import an assumption of fairness into the interpretation of the contract. To some extent the courts will try to work out what the parties would have agreed if the unforeseen events had been anticipated at the time the contract was written. Thus, although a full understanding of alternative forms of contract law requires sophisticated legal knowledge, it is possible to understand why the regulatory approach may be preferable in the context of coordinated market economies in which NMCIs play a major role. Having a regulatory contract law regime that is able to develop the legal principles of ex post fair sharing of unexpected costs and benefits is likely to be more supportive of long term cooperative agreements than the classical contract law approach. Looked at from the perspective of Ball (1989) we can see that different types of capitalism have different ways of achieving the functional completion of contracts. 4.3. Institutional evolution: the importance of variety Europe is blessed with a significant variety of ways of doing business. Germany is often cited as the leading example of a successful CME and the UK as a leading LME. In versions of VOC that highlight the role of the state in encouraging the economy to develop strengths in particular industries France is often cited as a classic case of state-led capitalism, and the success of the nuclear power industry in France as evidence of a major industrial development that was fostered to a considerable extent by state aid and intervention. It is at times of great uncertainty and change that the advantages of variety can be appreciated. The presence of variety creates a greater menu of options from which to respond to unexpected events. Ashby (1956, p. 207) summarises his Law of Requisite Variety as follows:

6 Often these arrangements and understandings will be implicit rather than explicit, and often they may not be documented other than through close observation of custom and practice.

M. Walker / The British Accounting Review 42 (2010) 137–152

149

“only variety can destroy variety” In biology, the organisms that become extinct are those that are unable to generate mutations sufficiently rapidly to respond to an evolving environment. Mutation in genes is a fundamental concept in evolutionary theory. Darwin’s theory of evolution is incomplete without an explanation of how new life forms emerge as potential candidates for the process of natural selection. Natural selection weeds out the life forms that are least well adapted to current environmental conditions. Genetic mutation is the process whereby new life forms enter into the competition by accidents to the DNA of existing life forms Smith (1993). However, North (2005) points to some of the difficulties in attempting to apply the biological theory of evolution to the evolution of human societies. Human societies are, in some ways more difficult to analyse than the physical systems of cybernetics, or the biological organisms and ecosystems of Darwinism. In particular North (2005, p. 66) stresses two aspects of the evolution of human societies which are poorly understood. First, our understanding of how new institutional possibilities emerge is rudimentary, and, second, in contrast to biological selection, institutional selection is influenced, at least in part by the beliefs of humans about their consequences. North stresses the importance of Knightian uncertainty, in understanding the role of institutions in society. He argues that human belief systems and political and economic institutions are central features of what he calls the artifactual structure of a society. According to North the main factor that determines the ability of a society to learn from and adapt to uncertainty is the richness and variety of its artifactual structure. He states: “Successful economic development will occur when the belief system that has evolved has created a “favorable” artifactual structure that can confront the novel experiences that the individual and society face and resolve positively the novel dilemmas.”(North, 2005, p. 695). and “Put simply the richer the artifactual structure the more likely are we to confront novel problems successfully. That is what is meant by adaptive efficiency; creating the necessary artifactual structure is an essential goal of policy.” (North, 2005, p. 70). North (2005) shows how the economic and political histories of different countries have been shaped to a considerable extent by their ability to adapt to unexpected events. In particular he points to the rich variety of the artifactual structures of Europe as a major reason why Western Europe became the leading driver of economic, political and technological development from the medieval times up to the 20th Century. He concludes: “In that competitive decentralized environment lots of alternatives were pursued; some worked, as in the Netherlands and England, some failed, as in the cases of Spain and Portugal, and some, as in France, fell between these two extremes. But the key to the story is the variety of the options pursued and the increased likelihood (as compared to a single unified policy) that some would turn out to produce economic growth. Even the relative failures in western Europe played an essential role in European development.” (North, 2005, p. 138). Thus, looked at through the lenses of cybernetics, biology, and in particular the institutional economics of North, we Europeans should be grateful for the variety of economic life forms we have. We should also be optimistic about the new economic life forms that interaction between these varieties is capable of producing. In my view, it is crucial for the member countries of the EU to understand the strategic importance of this variety. There is an unfortunate tendency amongst the politicians, bureaucrats and lawyers who make a living from the EU taxpayer to frown on this variety. The single European market agenda should be more focused on encouraging and allowing competition and cooperation between varieties of capitalism and much less concerned with standardising the EU economies onto a single, and ultimately less competitive, economic model. In particular it would be unwise to attempt to impose a single set of arrangements for contract law, company law and corporate governance, or financial reporting on the EU nation states. The VOC approach has shown how these aspects of an economy are interweaved with the NMCIs that are crucial to the coordination of economic activity especially in the CMEs of the EU. Understanding the importance of requisite variety to the long term development of Europe shows how it is possible to be enthusiastically pro European whilst at the same time being passionately against measures designed to homogenise the detailed workings of individual EU economies. 5. The future of international accounting standards To date, the issue on which international accounting policy makers have focused is the choice between a single set of global standard standards and separate sets of standards for all, or at least all major, national economies. However, there is no reason for the world to confine itself to this restricted choice set. It is possible for the number of sets of accounting standards to be greater than one, but less than the number of economies. For example there could easily be a separate set of standards for every major trading bloc. More interestingly, I have argued in this paper that the form of financial reporting should be tailored to the type of economic system. The literature on Varieties of Capitalism shows that there is more than one way of doing business, and I have

150

M. Walker / The British Accounting Review 42 (2010) 137–152

argued that in order to accommodate different varieties of capitalism, we may need to allow accounting standards to reflect the type of economic system. The relevance of the Varieties of Capitalism view of the world had been called into question by an apparent long term global trend toward stock market based capitalism. However, the recent collapse of stock market based capitalism has halted this trend. It is not clear at the present time whether this will be only a temporary halt. It is possible that the trend may resume once we emerge from the present malaise. However it is also possible that recent events will encourage policy makers to consider whether other forms of capitalism are possible, and whether measures should be put in place to encourage other forms of capitalism to survive and prosper alongside, or perhaps instead of, stock market based capitalism. The literature reviewed above on the economics of uncertainty revealed that accounting is only important in the context of a world where markets are imperfect and incomplete. When markets are perfect and complete market prices provide all the information required to guide economic decisions. In reality, markets are imperfect and incomplete, and it is in this context in which demands for various forms of accounting emerge. It is also in this context that different institutional and market arrangements for the sharing of risks, investment decision making, and corporate governance and accountability evolve. The recent crash serves to remind us that stock market based capitalism may not necessarily the best all round solution to these problems, and it surely makes sense to experiment with alternatives. Thus, if one accepts the view that alternative forms of capitalism should be encouraged and allowed to exist alongside stock market based capitalism then measures that inadvertently favour the development of globalised stock market capitalism over other forms of capitalism should be avoided. Mandating a single set of accounting standards designed to accommodate the needs of liberal stock market economies, risks doing harm to the alternative forms of capitalism that may be necessary for the long term development of the world. If the world is serious about allowing alternative forms of capitalism to prosper and survive, as I think it should be, then it needs to allow for the possibility that different forms of capitalism may require different forms of accounting. Imposing a single form of accounting, designed for a particular form of capitalism, runs the risk of preventing alternative forms of financial, economic, and legal governance from evolving. Such restrictions of institutional choice represent one of the worst forms of restrictive practice. In their authoritative and highly instructive analysis of worldwide financial reporting Benston, Bromwich, Litan, and Wagenhofer (2006) conclude “On balance, therefore, we believe that “constrained competition” within a small set of high-quality standards offers the most feasible and flexible setting to cope with increasingly global capital markets. This option would achieve some of the benefits of both competition and standardisation. In addition it promises the most harmonization without losing most of the benefits from competition.” They go on to suggest that “IFRS and U.S. GAAP are natural candidates that currently compete in some areas already” I agree with the general conclusion of Benston et al. that some degree of variety in accounting standards should be encouraged. However in this paper I have not sought to justify this conclusion on the basis of the need for competition between standards. Rather I have tried to argue that some variety in accounting standards is desirable because there are different types of capitalism, and because we do not know, and have no way of knowing, which form of capitalism is best. Having a single set of accounting standards, that are tailored to the specific needs of one particular form of capitalism, stock market capitalism, is not neutral with respect to the need to allow alternative forms of capitalism to compete on a level playing field. Rather than focusing on the narrow question of encouraging competition between accounting standards, I focus on the broader question of the need to encourage and foster competition between different varieties of capitalism. If recent events have taught us anything it is that it is not obvious that unregulated free market stock market capitalism is necessarily the best way to run the economy. Please note I am not saying here that I believe other ways of organising the economy are obviously better than stock market based capitalism. Rather I am saying that I do not know, and I believe nobody else knows, which form of capitalism is best. This being the case we should not seek to impose a single form of accounting that was designed for one form of capitalism on all other forms of capitalism. Requiring all other forms of capitalism to adopt the accounting of stock market based capitalism loads the dice in favour of stock market based capitalism. Thus I conclude that the world should consider establishing at least two sets of global accounting standards: Accounting standards for ‘liberal stock market economies’, and accounting standards for ‘coordinated market economies’. The first set of standards would focus on the information needs of stock market based investors where there is a high level of separation of ownership from control. The particular information requirements of investors in corporate debt markets, as well as equity markets, are also important here. Since such economies also rely heavily on the stock market for corporate control (through takeovers and mergers) information about the success or otherwise of past acquisitions is also important. Such economies exhibit a high demand for conditional conservatism, because of the high degree of separation between managers and investors. The role of shareholders as owners is paramount and it makes little sense to adopt an entity approach to financial reporting for such economies. In contrast coordinated market economies tend to have a stakeholder

M. Walker / The British Accounting Review 42 (2010) 137–152

151

approach to corporate governance matters, and so an entity based approach to financial reporting makes more sense for such regimes. Such economies exhibit a high demand for inter-temporal smoothing and so accounting standards that encourage transparent unconditional conservatism and transparent income smoothing are likely to pass the market (and NCMI test) for such regimes. In such regimes the ownership and information rights of owners are given a lower level of priority, and the importance of takeovers as a control mechanism is also much lower. I suggest that liberal stock market economies, like and the U.S. and the U.K. should work together to produce a single set of accounting standards designed for use in liberal stock market economies. The current standards of the IASB may well prove to be a useful starting point for this purpose. Countries that are less wedded to stock market capitalism should be encouraged to devise a new set of international accounting standards that have general applicability in economies that are less dominated by the stock market. Frankly I doubt if the current set of standards devised by the IASB are suitable for this purpose. IASB standards have largely been designed with the needs of stock market economies in mind. If a set of coordinated market economy standards could be developed then they could be used by both listed and unlisted firms in such regimes, thereby facilitating better comparisons of unlisted and listed firms in and across coordinated regimes. A further important advantage of this approach is that the unlisted companies of coordinated stock market economies will not need to adopt a completely foreign way of accounting in order to gain a stock market listing. They will be able to adopt the new international accounting standards for coordinated market economies rather than having to adopt accounting standards designed for a completely different form of capitalism. Until such time that a set of international accounting standards for CMEs exists, I think the CMEs would be well advised to allow their listed firms to choose between their own domestic accounting standards and IFRS. Similarly The UK and the US should consider allowing their own listed firms to choose between their own domestic standards and IFRS. Institutional rules that narrow choice and restrict the development of requisite varieties of capitalism should be resisted. What would be the role of the IASB in this new accounting world order? My view is that there would be an important role for a body like the IASB within the varieties of capitalism approach to international accounting standard setting. This role would be to provide a set of international accounting standards that, perhaps with some tailoring, would be suitable for use in the capital markets of both liberal stock market economies and coordinated market economies. The IASB standards would facilitate the activities of multinational companies wishing to raise capital across liberal stock market and coordinated market regimes. Listed companies based in liberal or coordinated market regimes might be allowed to choose between liberal or coordinated accounting standards or the accounting standards of the IASB. This approach would provide a high degree of comparability of accounts across international stock markets.7

Acknowledgement I am grateful to Mike Jones and an anonymous referee for detailed comments and advice. Any remaining errors are my sole responsibility. Appendix 1Declaration of the summit on financial markets and the world economy Strengthening transparency and accountability Immediate actions by March 31, 2009 * The key global accounting standards bodies should work to enhance guidance for valuation of securities, also taking into account the valuation of complex, illiquid products, especially during times of stress. * Accounting standard setters should significantly advance their work to address weaknesses in accounting and disclosure standards for off-balance sheet vehicles. * Regulators and accounting standard setters should enhance the required disclosure of complex financial instruments by firms to market participants. * With a view toward promoting financial stability, the governance of the international accounting standard setting body should be further enhanced, including by undertaking a review of its membership, in particular in order to ensure transparency, accountability, and an appropriate relationship between this independent body and the relevant authorities. * Private sector bodies that have already developed best practices for private pools of capital and/or hedge funds should bring forward proposals for a set of unified best practices. Finance Ministers should assess the adequacy of these proposals, drawing upon the analysis of regulators, the expanded FSF, and other relevant bodies.

7 Leuz (2009) proposes a “Global Player Segment” designed to encourage firms to commit to IFRS plus a higher level of enforcement and transparency. Such an approach could also accommodate an accounting standard for ‘liberal stock market economies’, and an accounting standard for ‘coordinated market economies’ as well as IFRS.

152

M. Walker / The British Accounting Review 42 (2010) 137–152

Medium-term actions * The key global accounting standards bodies should work intensively toward the objective of creating a single high-quality global standard. * Regulators, supervisors, and accounting standard setters, as appropriate, should work with each other and the private sector on an ongoing basis to ensure consistent application and enforcement of high-quality accounting standards. * Financial institutions should provide enhanced risk disclosures in their reporting and disclose all losses on an ongoing basis, consistent with international best practice, as appropriate. Regulators should work to ensure that a financial institution’ financial statements include a complete, accurate, and timely picture of the firm’s activities (including off-balance sheet activities) and are reported on a consistent and regular basis. Source: http://www.whitehouse.gov/news/releases/2008/11/print/20081115-1.html. References Allen, F., & Gale, D. (2000). Comparing financial systems. Cambridge, Mass.: MIT Press. Aoki, M. (2001). Towards a comparative institutional analysis. Cambridge, Mass.: MIT Press. Arrow, K. J. (1964). The role of securities in the optimum allocation of risk bearing. Review of Economic Studies, 31, 91–96. Ashby, W. R. (1956). An introduction to cybernetics. London: Chapman & Hall. Ball, R. (1989). The firm as a specialist contracting intermediary: application to accounting and auditing. University of Rochester, Unpublished Working Paper, April. Beaver, W. H., & Demski, J. S. (1979). The nature of income measurement. The Accounting Review, 54(1), 38–45. Benston, G. J., Bromwich, M., Litan, R. E., & Wagenhofer, A. (2006). Worldwide financial reporting: the development and future of accounting standards. Oxford: Oxford University Press. Black, F. (1993). Choosing accounting rules. Accounting Horizons, 7(4), 1–17. Bushman, R., & Piotroski, J. (2006). Financial reporting incentives for conservative accounting: the influence of legal and political institutions. Journal of Accounting and Economics, 42, 107–148. Casper, S. (2001). The legal framework for corporate governance. The influence of contract law on company strategies in Germany and the United States. In P. A. Hall, & D. Soskice (Eds.), Varieties of capitalism: The institutional foundations of comparative advantage. New York: Oxford University Press. pp. 387–417. Chapman, C. S., Cooper, D. J., & Miller, P. (Eds.). (2009). Accounting, organizations, and institutions: Essays in honour of Anthony Hopwood. Oxford: Oxford University Press. Christensen, P. O., & Feltham, G. A. (2003). Economics of accounting: Volume 1, Information in markets. Norwell, Massachusetts: Kluwer Academic Publishers. Coase, R. H. (1937). The nature of the firm. Economica, 4, 386–405. Debreu, G. (1959). Theory of value. New Haven and London: Yale University Press. Duffie, J. D., & Huang, C. (1985). Implementing Arrow–Debreu equilibria by continuous trading of a few long-lived securities. Econometrica, 53, 1337–1356. Giner, B., Tahoun, A., & Walker, M. (2009). Do cross-country differences in accounting conservatism explain variations in the degree of investor diversification. Working Paper, Manchester Business School. Hall, P. A., & Soskice, D. (Eds.). (2001). Varieties of capitalism: The institutional foundations of comparative advantage. New York: Oxford University Press. Hancké, R., Rhodes, M., & Thatcher, M. (2005). Introduction: beyond varieties of capitalism. In R. Hancké, M. Rhodes, & M. Thatcher (Eds.), Beyond varieties of capitalism. New York: Oxford University Press. Hayek, F. A. (1944). The road to serfdom. Chicago: University of Chicago Press. Hayek, F. A. (1948). Individualism and economic order. Chicago: University of Chicago Press. Hayek, F. A. (1979). Law, legislation, and liberty. London: Routledge and Kegan Paul. Hopwood, A. G. (1983). On trying to study accounting in the contexts in which it operates. Accounting, Organizations and Society, 8(2–3), 287–305. Hopwood, A. G. (1987). The archaeology of accounting systems. Accounting, Organizations and Society, 12(3), 207–234. Hopwood, A. G. (2007). Whither accounting research? The Accounting Review, 82(5), 1365–1374. IASB, International Accounting Standards Board. (2006). Discussion paper: preliminary views on an improved conceptual framework for financial reporting. London: IASB. IASB, International Accounting Standards Board. (2008). Exposure draft: conceptual framework for financial reporting: Chapter 1 the objective of financial reporting: Chapter 2 qualitative characteristics and constraints of decision-useful financial reporting information. London: IASB. Klein, B., Crawford, R. G., & Alchian, A. (1978). Vertical integration, appropriable rents, and the competitive contracting process. Journal of Law and Economics, 21, 297–326. Knight, F. H. (1921). Risk, uncertainty, and profit. New York: Harper and Row. Kreps, D. M. (1982). Multiperiod securities and the efficient allocation of risk: a comment on the Black–Scholes option pricing model. In J. McCall (Ed.), The economics of uncertainty and information. Chicago: University of Chicago Press. Leuz, C. (2009). Different approaches to corporate reporting regulation: how jurisdictions differ and why. Paper presented at the December 2009 information for better markets conference in London. North, D. C. (2005). Understanding the process of economic change. Princeton, New Jersey: Princeton University Press. Smith, J. M. (1993). The theory of evolution (Canto ed.). Cambridge: Cambridge University Press. Waymire, G. B. (2009). Exchange guidance is the fundamental demand for accounting. The Accounting Review, 84(1), 53–62. Waymire, G. B., & Basu, S. (2008). Accounting is an evolved economic institution. Foundations and Trends in Accounting, 2(1–2), 1–174. Whitley, R. (2007). Business systems and organisational capabilities: the institutional structuring of competitive competences. Oxford: Oxford University Press.