A political economy of accounting standard setting

Abstract

In recent years accounting researchers have identified “political” lobbying as a problem for accounting standard setting. This paper presents a simple game-theoretic analysis of the political process to identify situations where companies have incentives to lobby the political principal instead of participating in the usual due process of accounting standard setting. Analysis of the model suggests that “political” lobbying is more likely to happen in the EU than in the US. Furthermore it is suggested that if the relevant standard setters wish to achieve harmonization of accounting standards between the EU and the US, European companies have more lobbying leverage than their American counterparts because there are more European veto players than American ones.

1 Introduction

Accounting standard setters have often been criticised for giving undue influence to individual, mostly corporate, actors and being subject to regulatory capture. A recent addition to the criticisms voiced is the fear of too much political interference preventing “objective” accounting standards. This paper identifies public and/or political organizations which have veto power over accounting standards. It then presents a simple model of the political process to determine how this veto power influences the standard setting process and to identify situations where companies and managers have incentives to engage in “political” lobbying of the accounting standard setter, i.e., to try to influence accounting standards by approaching political actors instead of participating in the due process of accounting standard setting. The model is applied to the different institutional frameworks in which accounting standard setting takes place in the United States and in Europe.

Managers or corporations may wish to retain the ability to conceal unpleasant financial information or the ability to manage earnings to present constant growth or positive financial results (e.g., Burgstahler and Dichev 1997; Burgstahler and Eames 2003, 2006). In order to do so they have incentives to exert influence on financial reporting standards. Major international accounting standard setters follow a due process approach giving companies the ability to express their views and have them taken into account. However, managers pursuing one of the objectives mentioned above generally do not wish to express their preferences in full view of the investing public. Instead, they may use good personal contact to political decision-makers in order to gain leverage over the standard setter. Former FASB Chairman Dennis Beresford (2001) states that Congressional intervention in the standard setting process is taken very seriously by the Board. A point can be made that such activities have a detrimental effect on the objectivity and unbiasedness of financial information reported according to standards thus politically influenced (Zeff 1993, 2002).¹

Traditionally, research in lobbying has used empirical methodology and has taken comment letters sent to the standard setter as the basis of their analysis (e.g., Watts and Zimmerman 1978; Deakin 1989; Dechow et al. 1996 for lobbying in the United States; MacArthur 1988; Georgiou 2002 for lobbying in the United Kingdom; Larson 1997 for lobbying of the International Accounting Standards Committee).² Georgiou (2004) finds that the use of comment letters is correlated with the use of other means of lobbying. This gives some justification to the use of comment letters as a proxy for a company’s overall lobbying posture. However, more lobbying may go on “behind the scenes” and recent work has begun to use monetary contributions to politicians as a proxy for political lobbying. Ramanna (2008) finds that firms that have incentives to lobby against the elimination of pooling as an acceptable accounting method for business combinations can be linked via political contributions to Congresspersons who became involved against the Financial Accounting Standards Board (FASB) proposal on this issue. Johnston and Jones (2006) identify three significant accounting issues under consideration by the FASB from 1999 to 2000 and find that firms’ political lobbying expenditures are associated with their interest on those issues. Both studies present evidence of political influence on accounting standards under a private sector standard setting regime.

The American Institute of Certified Public Accountants (AICPA) is quite open about its efforts to influence public policy making in Washington. In its journal it prides itself of having influence on policy outcomes by understanding the policy-making process and by being able to deliver timely information (Lee and Rudd 1988; Lee 1988). Unsurprisingly, an empirical study of political action committee contributions found that members of the accounting profession gave significantly greater contributions to legislators who were members of committees having jurisdiction over accounting affairs (Thornburg and Roberts 2008). The study’s findings are consistent with the access hypothesis which predicts that special interest groups donate in order to get in a position where they can provide information to a relevant policy-maker. Despite the subject’s empirical relevance, so far little published economic theory work on lobbying in accounting standard setting exists.³ The present paper is an attempt to fill this void and presents a game-theoretic analysis of financial reporting standard setting in a political context.

The remainder of the paper is organized as follows: Sect. 2 discusses different systems of accounting regulation and the possibilities they respectively offer for participation in the regulatory process. Section 3 introduces and analyzes a model of accounting standard setting in a political context. Section 4 builds on these results to analyze corporate lobbying. Section 5 discusses the findings and concludes.

2 Systems of accounting regulation

“Political” lobbying in order to influence financial reporting standards has been observed in many jurisdictions. Zeff (2006) cites instances from the United States, Canada, the UK, Sweden and international lobbying on IASB standards. However, possibilities for exerting political influence vary according to country. This paper analyzes the institutional setup in the United States and in the European Union. Traditionally, studies in comparative international accounting have distinguished between an Anglo-Saxon or American and a Continental model of accounting regulation. Whereas the US has a long tradition of standard setting by an organism belonging to the private sector⁴ Continental European countries are characterized by a larger role given to legislation in accounting regulation.⁵ In Germany, the government sets broad principles in accounting via the commercial code while referring to the Grundsätze ordnungsmässiger Buchführung (GoB), literally the principles of orderly book-keeping, for situations not covered by the law (Flower and Ebbers 2002). GoB are determined in a market for interpretations by the decisions of judges in court, the expertise of practitioners and publications by academic accountants (Merkl-Davies 2004). This gives companies and managers the opportunity to influence accounting practice on either the legislative or the interpretative level. A further notable difference is in the participation of academic accountants in the development of accounting rules. Whereas there is traditionally high involvement by accounting academics in Germany (McLeay et al. 2000), few academics participate in the standard setting process in the US by submitting comment letters (Tandy and Wilburn 1996). In France, state involvement in accounting goes back to Colbert’s Edict of 1673. In the twentieth Century, the Vichy Government sponsored development of a general accounting code which was absorbed into later regulation (Colasse and Standish 1998, 2004).

A 2002 act by the European Union, known as the IAS regulation, radically altered this situation for listed companies (Véron 2007). The regulation which went into effect in 2005 mandates consolidated group reporting by companies with securities traded on a stock exchange to adhere to International Accounting Standards/International Financial Reporting Standards (IAS/IFRS) as issued by the International Accounting Standards Board (IASB) and endorsed by the European Union.⁶ Listed companies in the EU henceforth have to apply accounting standards which are developed by a private organisation operating in a very similar manner to the US standard setter. However, the political system in which this standard setter is embedded differs strongly between the two jurisdictions. It will be argued here that despite the similarities of the immediate process of developing accounting standards the differences on the political level influence companies’ lobbying incentives and are likely to lead to different standard setting outcomes. This is consistent with the view that the European Union’s implicit objective in mandating IFRS was to counter US hegemony in accounting standard setting (Dewing and Russell 2008).⁷

In the United States, the legislator has endowed the Securities and Exchange Commission (SEC) with the power to promulgate and enforce financial reporting standards in the Securities Acts of 1933 and 1934. The SEC in turn has delegated the rule-making power to organisations of the private sector, since 1973 to the Financial Accounting Standards Board. Since the right to regulate can be removed by the body that granted it accounting regulation can be seen as a two-level principal-agent relationship between the Congress and the SEC and between the SEC and the FASB.⁸ Both the legislator and the SEC thus have veto power over standards promulgated by the FASB (Beresford 1995). Horngren (1985) illustrates this hierarchical relationship as depicted in Fig. 1.

Fig. 1

Accounting standard setting hierarchy in the United States

The institutional setup for standard setting in the European Union is less straightforward. The IAS regulation stipulates that before IAS/IFRS become applicable in the European Union the Commission has to decide that they are conducive to the European public good and meet certain qualitative criteria. In this task the Commission follows a so-called comitology procedure (e.g., Bergstrom 2005) and is supported by three newly created bodies. The European Financial Reporting Advisory Group (EFRAG) represents private-sector interests and is financed by federations representing business, accountants and auditors, banks and similar organizations. Its work is reviewed on behalf of the European Commission by the Standards Advice Review Group (SARG), which is composed of independent accounting experts appointed by the Commission. The Accounting Regulatory Committee (ARC) represents member states’ governments and is staffed by civil servants from national ministries. After a new standard or interpretation is promulgated by the IASB, EFRAG’s Technical Expert Group (TEG) reviews it and issues an advice on its adoption to the Commission. SARG reviews this advice within three weeks in order to assess whether it is well balanced and objective. Taking into account EFRAG’s advice, the Commission prepares draft regulation to adopt the new standard or interpretation and sends it to the ARC which either agrees or recommends rejecting it. If the ARC recommends rejection, the Commission can either return the matter to EFRAG for further review or send it to the Council of Ministers for a final decision (Brackney and Witmer 2005). A recent change in the European Union’s comitology procedures has made the so-called regulatory procedure with scrutiny applicable to the endorsement process. This means that both the European Parliament and the Council can overturn an implementing measure of the Commission within three months’ time on the grounds that the Commission has exceeded its implementing powers or that the draft is not compatible with the aim or the content of the basic instrument (i.e., the IAS regulation). The process is illustrated in Fig. 2.

Fig. 2

Endorsement process for accounting standards in the EU

Assuming that the member states’ governments’ interests are represented on a civil servant level by the Accounting Regulatory Committee and that therefore the Council will not veto a standard deemed acceptable by the ARC this process gives veto power to three institutions. First, the European Commission which takes into account EFRAG’s advice but is not bound to it under applicable comitology procedures. Second, the ARC representing national governments. And third, the European Parliament whose members are elected by the population of the European Union. Arguably, these actors are constrained by the terms of the IAS regulation in their veto decisions. However, the phrasing employed in the regulation which requires a new standard to be “conducive to the European public good” to be endorsed seems to give carte blanche to the actors involved. Figure 3 illustrates this veto relationship between the IASB and public European actors. Note that there are two substantial differences to the hierarchical relationship in the US depicted in Fig. 1: First, there is no hierarchy between the individual veto players, i.e., no actor can undo another player’s veto. And second, the hierarchy suggested in Fig. 3 only extends to individual standards which can be vetoed by the European veto players. In contrast to this situation the two veto players in the US, the SEC and Congress, can threaten the standard setter’s very existence.

Fig. 3

Veto players in accounting standard setting in the European Union

The description of the respective political processes in accounting regulation in the European Union and the United States suggests that despite a similar structure of the immediate process of accounting standard setting—the IASB and the FASB follow an almost identical due process approach to standard setting—the possibilities for influencing standards are fairly different in the two regions due to differences in the political environment. In the US a new standard has to overcome the potential rejection by two veto players: the Securities and Exchange Commission and the legislator. The SEC can veto any standard or simply refuse to enforce it. However, the SEC is subject to oversight by Congress and Congress can therefore undo an SEC veto by legislatively mandating the enforcement of the standard. On the other hand, Congress also has the ultimate veto power over the standard. There is, at the same time, a countervailing force working against political mingling in the standard setting process: since a legislative act requires the consent of both Chambers of Congress and the President a legislative veto over an accounting standard will only occur if all three agree on it. In the European Union there are three actors with de facto veto power over accounting standards, the Commission, the Parliament and the ARC. However, contrary to the US, neither is superior to the other and can undo its decisions. Furthermore, all three institutions can in principle decide on a veto on their own without having to reach agreement with other institutions. These facts suggest that vetoing an accounting standard is easier in Europe than in the United States.

3 A model of accounting standard setting in a political context

This section establishes a simple model of the political process similar to Holburn and Vanden Bergh (2004) in order to analyze power relations in accounting regulation. Alternatives in the standard setting process are represented by a single continuous dimension. This can be thought of, for instance, as the level of discretion that the standard allows or the level of disclosure required. Not promulgating an accounting standard then also corresponds to a point on the policy line, and can be interpreted as leaving the incumbent regulation in place. The sequence of events is such that a private sector accounting standard setter (i.e., the FASB or the IASB) promulgates a new standard or a new interpretation subsequent to which public or political actors with veto power decide whether to accept that standard or to veto it and thereby reinstate the status quo before issuance of the new standard. All actors involved in this process are assumed to have single-peaked, linear and symmetrical bliss point preferences over accounting standards on the single-dimensional policy space. Actor i’s utility is then represented by $U_{i} = - \left| {\hat{s} - s_{i} } \right|$ where $\hat{s}$ represents the ultimate outcome of the standard setting process and s _i represents actor i’s optimal outcome. Each political or public body, e.g., the SEC or the European Commission, is assumed to have one single bliss point preference, independent of its internal composition. Furthermore, complete information is assumed throughout the analysis, i.e., all actors’ preferences are common knowledge.

3.1 Accounting standard setting in the United States

In the United States, accounting standards are promulgated by the Financial Accounting Standards Board. However, the right to develop mandatory standards has been delegated to the FASB by the Securities and Exchange Commission (SEC) and the SEC retains veto power over individual standards and can refuse to apply them. Furthermore, any accounting standard can be voided of its content by law, which then has higher authority than a SEC decision as well. Alternatively, the law could theoretically also order the SEC to take back a veto and acknowledge a given standard. The legislative process is such that passing a bill requires the consent of the Senate, the House and the President. To keep the model simple and tractable we abstract from the possibility of Congress to overcome a presidential veto. Then the game is played as follows: FASB develops and promulgates an accounting standard, the SEC decides whether to veto it, and the legislator decides whether to override any regulation with a law. Such a dynamic game is solved by the usual technique of backward induction: the institutions involved in the legislative process, i.e., the House, the Senate and the President, decide whether to intervene in the standard setting process by vetoing either the new standard or by undoing a veto by the SEC. Anticipating this, the SEC will only veto a new accounting standard if it prefers the incumbent standard to the new one and foresees that its veto will not be overruled by a legislative act. Under complete information the FASB will in turn anticipate both the SEC’s and the legislator’s reaction and choose its preferred standard among the set of standards that will not be vetoed.

For illustrative purposes, Fig. 4 shows a possible distribution of preferences over accounting standards.⁹ The subscripts denote the respective organizations potentially involved in the standard setting process. The range between the extreme preferences on either side of the three actors involved in the legislative process, i.e., the Senate, the House and the President, is called the Political Core and highlighted in Fig. 4 by a bold line. Points farther to the left on the single-dimensional policy line correspond to smaller numerical values in the following analysis. The eventual outcome in this situation depends on where on the policy line the incumbent standard lies as this is the fallback that will prevail if a potential new standard gets vetoed. If, for instance, the incumbent standard lies to the right of the President’s preferred point s _President, the FASB can fully impose its own preferences, $\hat{s} = s_{\text{FASB}} .$ The SEC will not veto this standard because −|s _FASB − s _SEC| > −|SQ − s _SEC| where SQ denotes the status quo, i.e., the incumbent standard. Similarly, the legislator will not veto the standard because −|s _FASB − s _Senate| > −|SQ − s _Senate| implying that at least the Senate will not agree to a legislative solution vetoing the new standard and reinstating the status quo. If, on the other hand, the incumbent standard falls sufficiently to the left of the President’s preferences, the legislators will veto any new standard s if −|s _Senate − s| < −|SQ − s_Senate| as in that case all three legislative institutions will prefer the incumbent standard to the new one. The SEC would not veto any standard acceptable to the legislator under the preference structure outlined above. Anticipating these veto strategies the FASB will maximize its utility by setting a standard $\hat{s} = s_{\text{senate}} - \left( {SQ - s_{\text{senate}} } \right).$ More generally, these findings can be stated as follows.

Fig. 4

A possible distribution of preferences over accounting standards in the US

Under the US accounting regulatory system, the FASB can fully impose its preferred standard if and only if either both the SEC and at least one of the three legislative institutions prefer this standard to the incumbent one or if all three legislative institutions prefer this standard to the incumbent one.

The logic is straightforward: if the SEC prefers the new standard to the incumbent one it will not veto it and if at least one of the legislative institutions prefers it, no law will be passed vetoing the standard. Similarly, if all three legislative institutions prefer the new standard to the incumbent one they will overrule a potential SEC veto by law. If, on the other hand, all three legislative institutions prefer the incumbent standard to the new one, they will veto it independently of the SEC’s preferences. Analogously, if the SEC and at least one legislative institution prefer the incumbent standard to the new one, the SEC will veto it and the legislator will not agree on overturning this veto. Anticipating a veto, the FASB will, of course, adopt a standard that is as close to its preferred one as possible without being vetoed. This analysis implies that if the FASB is not fully satisfied with the incumbent standard, having the right of initiative of setting a new standard, it will only promulgate a new standard under the following conditions which ensure that there exists some new standard which the FASB prefers to the incumbent one and which will not be vetoed.

If s _FASB ≠ SQ, then the FASB will promulgate a new standard if and only if neither of the following two conditions apply: (a) SQ lies outside of the Political Core and s _FASB is even farther outside on the same side of the Political Core than SQ; (b) SQ lies on one side of s _SEC and at least one legislative institution’s preferred point and s _FASB lies even farther on the same side.

If conditions (a) or (b) apply, the logic delineated above implies that any standard s which the FASB prefers to the status quo, i.e., which is closer to its preferred point on the single-dimensional policy line, will ultimately be vetoed. In case (a) the political institutions agree on issuing a law vetoing the standard. In case (b) the SEC vetoes the standard and the political institutions will not agree on overriding that veto. Anticipating this, the FASB will not attempt to promulgate a new standard. On the other hand, if neither of these two conditions applies, there is at least some standard s that is closer to the FASB’s optimal standard which will not be vetoed though not necessarily its optimal standard s _FASB itself.

3.2 Accounting standard setting in the European Union

The European Union follows a different approach to developing accounting standards. It uses standards promulgated by the international accounting standard setter IASB after a specific endorsement procedure designed to ensure that a new standard is conducive to European interests. The European Commission is advised in this process by EFRAG and has to obtain the assent of the ARC. The European Parliament and the Council subsequently assess whether the Commission has not exceeded its implementing powers by suggesting endorsement of a given standard and has adhered to the terms set out in the IAS regulation. Arguably, as discussed in Sect. 2, this gives veto power to three institutions acting independently from one another: the Commission, the ARC and the European Parliament. In a first step it is assumed that the IASB wants any new standard to be endorsed by the European Union even if this implies compromising on its contents. The implications of relaxing this assumption will be discussed below. To begin the discussion, Fig. 5 shows a possible distribution of preferences over accounting standards.

Fig. 5

A possible distribution of preferences over accounting standards in the EU

The subscripts denote the respective organizations potentially involved in standard setting. Points farther to the left on the policy line again correspond to smaller numerical values in the analysis. A major difference to the graphical depiction of the US system illustrated above lies in the absence of a Political Core. This is because under the European system any veto player can act on its own whereas a legislative veto in the United States requires the consent of three actors. Since in the EU no veto player is superior to any other in the sense that it can override a veto from another actor the veto players’ sequence of actions is inconsequential for solving the game by backward induction. The IASB moves first by promulgating a new standard, anticipating the veto players’ actions. After the promulgation of the standard the Commission, the ARC and the European Parliament will decide independently whether to veto the new standard, or, more correctly, to refuse its endorsement. Again, the outcome of the game depends on the fallback option, that is to say the incumbent standard SQ. In the situation depicted in Fig. 5, if s _ARC ≤ SQ ≤ s _Parliament, i.e., if the incumbent standard falls somewhere between the extreme preferences of the veto actors, no change of the incumbent standard is possible. This is because any change would leave at least one veto player worse off and would therefore be vetoed. If either SQ < s _ARC or SQ > s _Parliament the IASB can impose some new standard closer to its own preferences. For SQ ≤ s _ARC − (s _IASB − s _ARC) or SQ ≥ s _Parliament + (s _Parliament − s _IASB) the IASB can fully impose its own preferences, $\hat{s} = s_{\text{IASB}}$ since all veto players will then prefer this new standard to the status quo. For less extreme values of SQ the IASB has to compromise and will issue a new standard that leaves the ARC (for SQ < s _ARC) or the European Parliament (for SQ > s _Parliament) indifferent between the new standard and the incumbent one. These findings can be generalized as follows.

Under the EU accounting regulatory system, the IASB can fully impose its preferred standard if and only if one of the following cases applies: (a) SQ ≤ min{s _ARC, s _Commission, s _Parliament}–|s _IASB–min{s _ARC, s _Commission, s _Parliament}| or (b) SQ ≥ max{s _ARC, s _Commission, s _Parliament} + |max{s _ARC, s _Commission, s _Parliament} − s _IASB|.

Conditions (a) and (b) ensure that the IASB’s preferences relative to the incumbent standard and to the three veto players’ preferences are such that the IASB can promulgate a standard that it finds fully satisfying without having to compromise in order to prevent a veto. This necessitates that all three veto players prefer this standard to the incumbent one. Intuitively, this will be the case if either the incumbent standard and the IASB’s preferred standard are both on the same side of all veto players’ preferences but the IASB’s preferences are less extreme than the status quo or if the status quo is so extreme compared to all veto players’ preferences that the IASB’s favourite standard is preferred to the incumbent one by all even though it may “go too far” from at least some veto players’ point of view.

If at least one veto player prefers the incumbent standard to the proposed new one it will be vetoed. As in the American case, the IASB would anticipate a veto and refrain from issuing such a standard. However, the fact that the IASB cannot promulgate a standard fully compatible with its preferences does not necessarily mean that it cannot pass some new standard that it prefers to the incumbent one. A standard exists which is both preferred to the incumbent one by the IASB and by all veto players in the following cases.

If s _IASB ≠ SQ the IASB will promulgate a new standard if and only if either (a) SQ < min{s _ARC, s _Commission, s _Parliament} and s _IASB > SQ or (b) SQ > max{s _ARC, s _Commission, s _Parliament} and s _IASB < SQ. Intuitively, this means that for a new standard to be feasible the incumbent standard has to fall outside of the range between preferences of all three veto players. If this is not the case, any new standard will leave at least one veto player worse off and therefore get vetoed. A second necessary condition for the IASB to be able to obtain a standard preferred to the status quo is that its own preferences are at the same side relative to the status quo as all the veto players’ preferences are. If its own preferences relative to the veto players’ are even more extreme than the incumbent standard, it is not in a position to ensure a move towards its own preferred point.

3.3 Implications for accounting harmonization

The relevant accounting standard setters for the United States and the European Union, the FASB and the IASB respectively, follow a very similar approach to standard setting. However, they are embedded in two fairly different political environments. An analysis of the model suggests that the range of potential standards that will not be vetoed by a political actor is much larger in the United States since more individual veto players exist in the European Union. In the US, all three political actors have to prefer the status quo to a proposed standard in order to override it. In the EU, to the contrary, three actors have individual veto-power over a standard. This suggests a higher propensity of EU political actors to veto an IASB standard than of US political actors to veto a FASB standard.

Interestingly, this institutional structure also suggests an important difference in incentives for political interference in accounting standard setting. In the United States, the final outcome of the standard setting process is more likely to differ from the standard setter’s initial preference if the political actors’ preferences are similar. This is because the three legislative institutions have to agree in order to overturn a FASB standard. Conversely, in the European Union, the final outcome of the standard setting process is more likely to differ from the standard setter’s initial preference if the political actors’ preferences are dissimilar. In the EU, political actors have individual veto powers and do not have to agree in order to overturn a standard. Therefore the more dissimilar their individual preferences are the more likely it is that at least one political actor will have incentives to veto any given standard proposed by the IASB.

The FASB and the IASB are engaged in an effort towards international convergence of accounting standards (e.g., Schipper 2005). This has both been welcomed as leading to more international comparability in financial reporting numbers (Tarca 2004) and criticized as too restrictive to take into account regional differences (Stecher and Suijs 2007). Given the similarity in the structure of the two standard setting bodies it is reasonable to treat them as belonging to the same epistemic community (Haas 1992) and assume that their preferences over accounting standards are fairly similar. However, their effort to achieve convergence is subject to veto from the relevant political authorities. Any common standard the two standard setters wish to promulgate therefore needs to fulfil the requirement of not being subject to a veto by either the US or the EU political actors who have veto power over it. The analysis above suggests that if there are continental differences in preferences over an accounting standard and if the two standard setters nevertheless wish to achieve convergence in the area of that standard the final outcome is likely to be closer to European preferences than to American ones. This is the case because any common standard has to “pass” more veto threats on the part of European actors than American ones. This result stands in contrast to claims that the European Union gives too much ground to American hegemony in financial reporting regulation (Dewing and Russell 2004).

3.4 Extensions of the base model

The base model outlined above is deliberately kept parsimonious in order to present political forces affecting accounting standards in as simple a framework as possible. It is, however, straightforward to extend the model by adding some more “realistic” features without altering its qualitative conclusions. Evidence shows that politicians pay at least lip service to letting accounting standard setting remain in the private sector. This fact can be incorporated into the model by including costs of political mingling in the standard setting process. All actors’ utility function would then be $U_{i} = - \left| {\hat{s} - s_{i} } \right| - \Sigma_k(b_{k} \cdot C_{ik})$ where C _ik  ≥ 0 denote the costs to actor i if actor k gets involved in standard setting and b _k is a dummy variable with b _k  = 1 indicating involvement by actor k. C _ii  > 0 implies that it is costly to the (political) actor herself to get involved in standard setting. This could be the case because the politician subjects herself to criticism of exceeding her area of competence or of politicising a field better left to experts. In the case of the European Union’s endorsement of an IFRS such costs can also be justified by the desire to avoid developing “European IFRS” which would limit worldwide comparability. In the model the effect of such costs would be to create a symmetrical “non-intervention zone” around each political actor’s preferred point within which the costs of intervening exceed the benefits of getting a standard closer to one’s preferences. Such costs would limit political intervention and shift some power from the political actors to the standard setters.¹⁰ C _i≠k > 0 implies that a political actor’s involvement is costly to some other actor. Under complete information, such costs would not alter the power relation between political actors and the standard setter since they would only be incurred if a standard was vetoed. As vetoes are anticipated the standard setter would never pass a standard which stands to be vetoed anyway. Such costs, however, would in some instances persuade actors to veto a standard they in fact agree with. For instance, it has been asserted that the SEC dissuaded the FASB from passing its preferred standard on stock option expenses, threatening not to enforce it, because it feared legislative intervention (Levitt 2002, p. 110). The reason the SEC acted this way was because it feared that legislative intervention in the accounting standard setting process might create a harmful precedent.

The model assumes that each standard is evaluated separately by the political actors. Allowing for multiple standards of course introduces the possibility that veto players trade concessions over time, i.e., refrain from vetoing a standard in exchange for a promise to reciprocate by another actor preferring this particular standard to be endorsed. Including this possibility in the model makes vetoes less likely, similarly to the analysis of costs of political mingling. This in turn gives greater flexibility to the standard setter. The impact of the possibility to trade concessions should be greater in the European Union where more actors have individual veto power. In principal, greater heterogeneity of preferences of the political actors allows for more give-and-take between veto players, countervailing to some degree the increased likelihood of a political veto resulting from more heterogeneous preferences in the base model.

An important assumption made in the analysis above is that the IASB will only pass a standard which stands to be endorsed in the European Union. This assumption can be justified by the EU arguably being by far the IASB’s most important “client”. However, it is conceivable that the IASB is not willing to compromise in all instances. In the model this fact can be incorporated by relaxing the assumption and instead introducing costs of losing the EU as a client for the standard considered. The effect would be to create a symmetrical “compromise zone” around the IASB’s optimal point on the policy line. If a standard can be found that fulfils the requirements for not being vetoed by any actor in the EU and that falls within this range it will be passed. On the other hand, if no such standard exists, the IASB will pass a standard $\hat{s} = s_{\text{IASB}}$ and accept this standard not being endorsed.

Finally, the model assumes single, well-defined preferences of all actors. In reality, of course, these are bodies made up of individuals with diverging opinions. The position that a standard setter has in the context of the overall regulatory regime determines its internal balance of power. The threat of a potential external veto to a promulgated standard can be used strategically by individual board members in order to further their own agenda. For instance, the likely effect of the existence of numerous European veto players is to strengthen the position of those members of the IASB who oppose unilateral harmonization towards the American model. The influence of the existence of veto players on the internal balance of power depends on the amount of uncertainty concerning the veto players’ preferences. Individual board members can only strategically use the potential threat of a subsequent external veto if there is asymmetry of information about the likelihood of such a veto.

3.5 Recent experiences with political interference in the European Union

There have been several instances in which European institutions have mingled with accounting standard setting by the IASB in recent years. Two cases will be briefly presented to illustrate the logic outlined above. In 2004 IFRIC 3 Emission Rights was issued. In its recommendation to the European Commission EFRAG concluded that this Interpretation did not meet the requirements of the ‘true and fair view’ principle and that it did not meet the criteria of understandability, relevance, reliability and comparability. EFRAG therefore suggested not endorsing it. Anticipating a rejection by the EU, the IASB decided to withdraw the Interpretation and has not yet announced an estimated completion date for its subsequent Emissions Trading Schemes project.¹¹ Despite considering the IFRIC a correct Interpretation of existing standards, the Board thus withdrew it when foreseeing that it would not be endorsed, highlighting the power of the European Commission and its advisory body, EFRAG.

A second instance where a European Union institution initially opposed endorsement of a standard was IFRS 8 Operating Segments. In this case the European Parliament expressed concerns about the impact of the standard and requested an impact assessment from the Commission. Contrary to IFRIC 3, this initial opposition only delayed endorsement which was ultimately granted. It is interesting to compare these two cases, one of which ended with the endorsement of the standard as originally promulgated by the IASB while the other one led to the withdrawal of the Interpretation by the standard setter. Arguably, the different outcomes were caused by some major differences between the two cases. First, while the European Parliament expressed reservations about the IFRS 8 both EFRAG and the ARC supported it. Preventing its endorsement would therefore probably have imposed political costs on individual Members of the European Parliament (MEPs). Second, while IFRIC 3 was in principal universally applicable, it was to a degree tailored towards the European new cap-and-trade emissions trading scheme whereas IFRS 8 brought international segment reporting in line with the respective US standard SFAS 131. Rejecting IFRS 8 would therefore have impacted more on the harmonization effort between international and US standards than rejecting IFRIC 3, again leading to potential political costs for MEPs. Arguably, then, Parliament did not expect to ultimately prevent endorsement of the standard but rather wanted to make its voice heard in order to manifest its clout and gain leverage over future standards by demonstrating its veto power.

4 Corporate lobbying in accounting standard setting in the US and EU context

The analysis presented in Section 3 has implications for firms’ lobbying decisions. No specific assumptions are made about the origin of the lobbyists’ preferences but according to empirical results (e.g., Beresford 1993) she is more likely to belong to the preparer sector (e.g., corporate management or auditors) than to be a user of financial statements (e.g., banks or financial analysts). Assuming that lobbying is costly and marginally shifts the preferences of the actor towards whom it is directed, firms have to decide whom to lobby in order to have an impact on the final outcome of the standard setting process. They will not expend resources to influence an actor’s preferences if ultimately the standard passed is not affected. We will call the actor whose preferences are decisive the pivotal actor (Holburn and Vanden Bergh 2004) and resort to a graphical depiction of the situation in the European Union in order to present the argument. Assume the preferences are as presented in Fig. 6.

Fig. 6

A possible distribution of preferences over accounting standards in the EU

In this case the IASB can fully impose its preferred standard since all veto players prefer s_IASB to the incumbent standard SQ. In this case a lobbyist wishing to change the outcome of the standard setting process will direct his lobbying towards the standard setter as shifting the IASB’s preferences will directly impact the ultimate standard. If, on the other hand, the status quo lay somewhat to the right of its current position such that s_ARC − SQ < s_IASB − s_ARC the IASB would no longer be in a position to impose its will. If it promulgated a standard corresponding to its own preferences the ARC would prefer the status quo and veto the proposed new standard. In this case directing lobbying towards the IASB and marginally shifting its preferences would be a waste of money since the ultimate outcome of the standard setting process would not be altered. In that case the pivotal actor is the ARC and rational lobbyists will direct their efforts towards it. Note that if one actor’s preferences are pushed far enough by means of lobbying another actor may become pivotal. In Fig. 6, once the IASB’s preferences have shifted beyond the point characterized by 2 s _ARC − SQ the ARC will become pivotal.

Generalizing this argument we observe that the pivotal actor, towards whom lobbying will be directed in the first place, will be the accounting standard setter in exactly those cases where the standard setter can fully impose its preferred standard. It will be a political actor in all other cases. Since we have argued in Sect. 3 that the private sector standard setter enjoys more flexibility in the US than in the EU in the sense that it can impose its preferred standard more often it follows that lobbying will be political rather than directed towards the accounting standard setter more frequently in the European Union than in the United States.

It can be seen from the discussion in Sect. 3 that if the standard setter cannot fully impose its preferred standard the pivotal actor in general will be the political actor whose preferences are closest to the FASB in the US and the political player whose preferences are farthest from the IASB in the EU. The reason for this is that due to the way the legislative system works in the United States, it is sufficient to convince one political actor in order for a new standard not to be vetoed. In the European Union, on the other hand, all three political players need to be convinced since they have veto power individually.

5 Conclusion

The present paper discusses accounting standard setting and lobbying in a political context. It establishes and analyzes a simple model of the political process and derives predictions about the outcome of accounting standard setting under the threat of a political veto. The base model is applied in a US and an EU context. The relevant accounting standard setters, the FASB in the United States and the IASB for the EU show great similarities and follow an almost identical due process when developing new accounting standards. They are, however, embedded in an entirely different political context and as a result the outcome of the standard setting process will often differ even when both standard setters have identical preferences over an ideal accounting standard.

An analysis of the respective systems of financial reporting regulation in the European Union and the United States finds that a higher number of individual players are endowed with veto power over accounting standards in the EU than in the US. Anticipating potential vetoes, the relevant standard setter will promulgate only standards that all players who individually have the power of veto prefer to the status quo. Holding everything else constant the analysis gives rise to three main predictions:

• (a) Political and public sector actors in the European Union have a bigger say in financial reporting regulation than their counterparts in the United States.

• (b) Harmonization between standards applicable in the European Union (i.e., IFRS) and standards applicable in the US should be biased towards the European position.

• (c) The presence of more veto players implies greater weight of political actors in Europe. Therefore, lobbying is more likely to be addressed towards a political actor in the EU than in the US.

Ironically, the relative power advantages of the European Union stemming from the existence of numerous veto players in the endorsement process has been somewhat offset by a recent decision by the SEC which has been pushed for by the European Union. In 2007 the SEC decided to accept from foreign private issuers in their filings with the Commission financial statements prepared in accordance with IFRS (SEC 2008; Jamal et al. 2008). However, only those financial statements prepared in accordance with IFRS as issued by the IASB are deemed acceptable. This in turn punishes any deviation of IFRS as endorsed by the European Union from the complete set of IFRS since that imposes costs of reconciliation of financial statements to US GAAP on European firms listed in the US. It is, however, not clear that the implicit power shift towards the US position motivated the SEC’s decision.

The propositions presented above can be operationalized and tested empirically. Measuring political lobbying is easier in the United States than in Europe due to the availability of data on political contributions. However, public statements of politicians on accounting matters, such as the famous letter French President Jacques Chirac sent to Commission President Romano Prodi, can be taken as proxies for political positions.

An interesting venue for future research would be the comparison of publicly available comment letters which can be used as a proxy for European public positions, such as the ones issued by EFRAG with reactions by the IASB. Arguably, if the European Union is considered its most important client by the IASB, divergences between positions of the standard setter as evidenced by exposure drafts and public European positions should lead to a reaction of the IASB in the direction of the EU position.

While direct lobbying of the accounting standard setter via comment letters is a much researched subject, studies on the political aspects are rare despite their acknowledged importance. Both conceptual and empirical work on the politics of accounting standard setting represent a fruitful area for research.