Understanding of “Ideological Capture” Through Changes in the FASB Conceptual Framework

The fall of “reliability:” Towards an understanding of “ideological capture” through changes in the FASB Conceptual Framework

Abstract:

In 2010, the FASB updated its conceptual framework of thirty years to eliminate “reliability” as a fundamental accounting property. It argued that this amendment reflected its original intent and that “reliability” was misunderstood in practice. Drawing on primary archival resources and field interviews with FASB and IASB employees, I provide evidence that the change also sought to validate the rise of fair values in GAAP. By eliminating the need for accounting to be “reliable,” the change attempted to neutralize opposition to fair-value accounting. The change illustrates “ideological capture,” wherein regulators, seeking to diffuse contentious political economies, manufacture conceptual narratives for their actions.

Note: Interviewees who requested anonymity on certain subjects have been accorded this courtesy and identified by a unique subject number; in all other cases, interviewees have been identified by name. In cases of extended quotes from interviewees who were granted anonymity, audio recordings of the interviews have been archived.

1. Introduction

The Financial Accounting Standards Board’s (FASB) conceptual framework is an important force in shaping the nature of U.S. generally accepted accounting principles (GAAP). The conceptual framework articulates fundamental principles of accounting that are expected to guide and inform the FASB in its rulemaking activities.[1] In 2010, the FASB revised its conceptual framework to drop “reliability” as one of two “fundamental qualitative characteristics” of accounting (the other fundamental characteristic being “relevance”). The notion of reliability was replaced with that of “faithful representation.” Moreover, the concept “verifiability,” which had been previously considered an element of reliability, was reclassified as an “enhancing” (as opposed to “necessary”) characteristic of accounting. (See Figure 1 for a summary of these changes and for brief definitions of the concepts.) The changes were effected as part of a joint project with the International Accounting Standards Board (IASB) to harmonize the conceptual frameworks of the two accounting rulemaking bodies.[2]

In outlining its basis for these changes, the FASB, in its official record, argued that its original intent had always been for reliability to be understood as faithful representation – that an accounting estimate “represent what it purports to represent” – but that the term reliability had come to be misunderstood in practice to mean, among other things, measurement precision.[3] In this sense, the FASB argued that the shift away from reliability simply clarified the conceptual framework.

Some prominent observers of the change, including one former member of the IASB, argued that the change was more significant than a clarification of original intent.[4] They argued that the change was partly driven by the need to reconcile the FASB and IASB conceptual frameworks with the growing use of fair-value accounting. Since the early 1990s, the FASB had increasingly deployed fair values – estimates of current values of assets and liabilities – as the basis for accounting measurements, in lieu of the traditional practice of relying on historical costs from realized arm’s-length transactions. Because fair values could not always be reliably measured, the notions of reliability and verifiability in the conceptual framework posed a serious intellectual impediment to the growing use of fair-value accounting in GAAP.

In this paper, I principally examine these two explanations for the changes to the FASB and IASB conceptual frameworks. I draw on primary archival evidence from FASB and IASB board meeting minutes, internal and external staff documents, and communication by key FASB constituents. I also rely on original field interviews with thirteen board members, six senior staffers, and eight active constituents of the FASB and IASB at the time. Finally, I supplement these sources with secondary data from academic and media articles on the changes.

The evidence does not entirely rule out either explanation. A small core group of senior staff members at the FASB, in particular, had long held the view that the term reliability was misinterpreted in practice. To these FASB insiders, the revisions to the conceptual framework did in fact represent a clarification of original intent. But to several board members and constituents, the revisions reflected a need to update the framework to be consistent with changes in the nature of GAAP, especially fair-value accounting. As one individual who served as a board member during the deliberations noted, “The major thing hanging over our head then was fair-value measurement… This issue of reliability versus faithful representation was very core to the decisions in fair-value measurement… No one came out and said that, but it was pretty clear that this was one of the big things… Fair value was used in thirty-plus places in the accounting literature… I might not have known it in the beginning [when I joined the board], but in the end it was pretty clear.”[5]

The evidence consistent with this latter explanation is noteworthy for at least two reasons. First, it suggests that the FASB’s own official explanation for the changes to the conceptual framework is at best incomplete. This official explanation is part of the GAAP record that academics, practitioners, and even standard setters refer to as they attempt to interpret accounting rules. Selective reporting in this record risks compromising trust in the FASB’s internal practices and procedures.

Second, it illustrates the importance of ideological (or conceptual) narratives in regulation, even if such narratives are constructed ex post facto: The changes to the conceptual framework were effected in 2010, nearly twenty years after the FASB first initiated a major thrust towards fair-value accounting. The evidence relating the changes in the conceptual framework to fair-value accounting is consistent with the proposition of “ideological capture” in GAAP.[6] I expand below.

Fair-value accounting is one of the most important developments in U.S. GAAP over the last twenty-five years. The period from the 1930s through the 1980s saw very little fair-value use in GAAP, largely due to a regulatory moratorium motivated by experiences with fair values in the timeframe leading up to the stock-market crash of 1929.[7] But starting in the early 1990s, the proportion of GAAP standards using fair values grew substantially, a pattern that persisted at least through the financial crisis of 2008–09. (See Figure 2.)

Under certain circumstances, the use of fair values in accounting is consistent with longstanding accounting principles such as verifiability and conservatism – in particular cases where the fair values are obtained from liquid markets and where the fair values are used to write down already capitalized assets that are suspected to have “impaired” value. But the use of fair values in U.S. GAAP today extends in some cases to situations where asset and liability valuation estimates have to be generated from mathematical models using a number of subjective assumptions about future economic conditions, including growth and interest rates, a practice that is not verifiable. Moreover, for certain financial assets and liabilities, fair values are permitted for asset write-ups (in addition to write downs), a practice that is not conservative. Unverifiable and non-conservative accounting practices can facilitate opportunistic financial reporting.[8]

The rise of fair-value accounting in U.S. GAAP coincides with the financialization of the U.S. economy.[9] For instance, fair-value accounting practices are consistent with both the economic incentives and conceptual preferences of asset managers and investment bankers. Especially in periods of rising economic fundamentals (e.g., during the five years leading up to 2008), the use of fair-value accounting for certain financial assets results in recognizing unrealized gains in reported income, which in turn can drive management compensation.

The FASB’s revisions to its conceptual framework to bring it in line with the growth of fair-value accounting help diffuse claims that such growth is due to the influence of certain special interests from the financial sector. In this sense, the revisions are an illustration of ideological capture, wherein regulators fabricate conceptual justifications for their actions in an attempt to diminish claims of capture. Note that it is not necessary for the regulator to be actually captured; just the perception of capture is sufficient to precipitate the conceptual justifications. Ideological capture is costly to regulators not only because producing conceptual narratives is time consuming but also because embracing a given conceptual justification usually involves taking sides in an ideological debate. This approach can lower public perception of regulatory neutrality.[10]

With the changes to the conceptual framework now in place, objections to fair values on conceptual grounds, such as their insubstantial reliability, will be less persuasive in standard-setting deliberations. The conceptual framework changes may then foreshadow still more fair values in GAAP. This conclusion, if true, is concerning in light of the role of fair values in prior accounting scandals such as Enron and financial crises such as those of 1929 and 2008.[11]

In terms of prior literature, this study is most closely related to that of Erb and Pelger.[12] In providing a comprehensive history of the evolution of “reliability” in standard setting, Erb and Pelger discuss the process culminating in the 2010 FASB-IASB revisions to their respective conceptual frameworks. But there are significant distinctions between the two studies as well. First, whereas Erb and Pelger are focused on the sociology of accounting reliability, this paper is a study of the political implications of eliminating reliability from the FASB conceptual framework. Second, whereas Erb and Pelger rely on interviews with seven IASB protagonists, this study draws from twenty-seven interviews with both FASB and IASB players. Moreover, only three interviewees overlap across the two studies, suggesting the data herein can augment the analysis by Erb and Pelger.

The remainder of the paper is organized as follows. Section 2 provides an overview of the FASB’s conceptual framework. Section 3 outlines various plausible explanations for the changes to reliability and verifiability in that framework. Section 4 provides a detailed description of the events leading up to the relevant changes in the conceptual framework. This section draws on primary sources at the FASB and IASB, including board meeting minutes and staff documents. Section 5 introduces evidence from the original field interviews with key players to help examine the various possible explanations for the removal of reliability from the FASB’s conceptual framework. In this section, I also interpret the evidence in the context of the hypotheses in Section 3 and develop the proposition of ideological capture. Section 6 concludes.

2. Background and overview of the conceptual framework

In the Securities Exchange Act of 1934, Congress vested in the Securities and Exchange Commission (SEC) the authority to determine the standards under which listed corporations in the U.S. have to report financial information.[13] The SEC has, since the late 1930s, effectively deferred that authority to private organizations. At first, the authority to organize and determine accounting principles was held by the Committee on Accounting Procedures (CAP) of the American Institute of Certified Public Accountants (AICPA, known from 1917 to 1957 as the American Institute of Accountants). The CAP held sway through 1959 when it was replaced by another AICPA organ, the Accounting Principles Board (APB).[14]

Both the CAP and the APB represented attempts at self-regulation – the accounting profession determined accounting standards through its primary membership organization, the AICPA. Indeed, members of the CAP and the APB served part time, often holding fulltime senior positions at leading audit firms. Moreover, these audit firms directly supported the research activities of their standard setters.[15] Largely due to concerns about the lack of independence in accounting rulemaking, and in the face of some particularly political decisions by the APB, this organization was replaced by the FASB in 1973.[16] Since then, the de facto authority to determine U.S. GAAP has vested in the FASB.

When the APB was first constituted in the 1950s to replace the CAP, the AICPA recommended that it adopt a set of conceptual principles for its standard-setting activities. The CAP’s demise had been at least partly attributed to the ad hoc and political nature of its rulemaking, and the AICPA was keen to avoid this failure again.[17] The fledgling APB did in fact commission a set of research studies that it hoped would provide the basis for a conceptual framework, but the board did not formally adopt the studies partly because they departed from prevailing accounting practice in many ways.[18]

Another attempt at a conceptual framework for the APB in the 1960s, again at the behest of the AICPA, resulted in a non-binding publication in 1970 – APB Statement No. 4, Basic Concepts and Accounting Principles Underlying Financial Statements of Business Enterprises – but the APB itself did not last much longer. As noted earlier, it suffered a similar fate as its predecessor, the CAP, for some of the same reasons. A former APB member and Stanford professor Charles Horngren noted, “My experience as a member of the APB taught me many lessons. A major one was that most of us have a natural tendency and an incredible talent for processing new facts in such a way that our prior conclusions remain intact.”[19]

Not surprisingly then, when the FASB was constituted in the early 1970s, one of its first priorities was to put together a conceptual framework for its standard-setting activities. In fact, the Trueblood Study Group of the AICPA (1971–73), which was constituted in response to a crisis of confidence in the APB – a crisis that eventually led to the APB being supplanted by the FASB, had already sketched an outline for a conceptual framework for accounting. [20]

In the words of two longtime FASB insiders John Foster and Todd Johnson, a conceptual framework at the FASB was expected to “reduce political pressures in making accounting judgments” and also reduce “the influence of personal biases on standard-setting decisions.” They noted, “In an environment in which standard setting is based on the personal conceptual frameworks of individual standard setters, agreement on specific standard-setting issues will only occur when a sufficient number of those personal frameworks intersect. However, even those agreements may prove to be transitory because, as the membership of the body changes over time, the mix of individual conceptual frameworks will change as well. As a result, that standard-setting body may reach significantly different conclusions about similar—or even identical—issues than it did earlier, resulting in standards not being consistent with one another and past decisions not being indicative of future ones. Standard setting therefore becomes more or less ad hoc. Moreover, without a framework, rational debate cannot occur because positions about the appropriate accounting treatment for a given transaction can neither be defended nor refuted—the appropriate treatment is simply ‘in the eye of the beholder.’ That was the case with the AICPA’s Committee on Accounting Procedure which preceded the Accounting Principles Board (APB) and, it largely was true of the APB as well.”[21]

The first official FASB document on its conceptual framework was published in 1978, five years after the FASB began operations.[22] Titled CON 1, Objectives of Financial Reporting by Business Enterprises, the document’s most significant position was to establish financial reporting’s objective as “decision usefulness” for “investors and creditors.” This was the position of the Trueblood Study Group that had had a formative influence on the early FASB.[23] While this objective might have seemed innocuous, it did mark a significant departure from the long-held tradition that holding management accountable for its “stewardship” of firm resources was the key objective of financial reporting. Decision usefulness could include stewardship, but not to the exclusion of investors’ valuation demands. CON 1 thus laid a conceptual foundation for current-value estimates in financial statements.

As the first of the FASB conceptual framework documents, CON 1 also laid out the aims for this project: “The purpose of the series is to set forth fundamentals on which financial accounting and reporting standards will be based… The Board itself is likely to be the major user and thus the most direct beneficiary of the guidance provided by the new series. However, knowledge of the objectives and concepts the Board uses should enable all who are affected by or interested in financial accounting standards to better understand the content and limitations of information provided by financial accounting and reporting, thereby furthering their ability to use that information effectively and enhancing confidence in financial accounting and reporting.”[24]

CON 1 also made clear that the conceptual framework documents would not hold the official status of GAAP standards. While this was a “constitution” for GAAP, intended to insulate GAAP from idiosyncrasies and politics, it would be an advisory constitution.

The second, and perhaps most significant conceptual framework document until 2010, was CON 2, Qualitative Characteristics of Accounting Information, issued in 1980.[25] As the title suggests, this document established the principles to guide recognition and measurement criteria in financial statements. The document arose in the context of “fierce” debates in the 1970s among accountants about how to account for inflation, in particular, when current-value adjustments were meaningful in financial reporting. The debates were not resolved, and CON 2 effectively codified principles that reflected existing practice.[26]

Arguably the primary accomplishment of CON 2 was to recognize that accrual estimates represented balancing considerations of “relevance” and “reliability.” While these two principles, which were both recognized as “fundamental,” did not always conflict with each other, they often did, and CON 2 set out for FASB standards to reflect this balance. Reliability in CON 2 was defined to include “verifiability,” “faithful representation,” and “neutrality.” Besides these principles, CON 2 also recognized the importance of “comparability” of financial statements across firms and industries and “consistency” across time.

CON 2 defined verifiability as a quality “that may be demonstrated by securing a high degree of consensus among independent measurers using the same measurement methods.”[27] Neutrality – defined as a “freedom from bias” – stood in contrast to the longstanding accounting principle of “conservatism,” which called for a higher standard for recognizing gains versus losses.[28] CON 2 recognized “a place for a convention such as conservatism” in GAAP, but somewhat reluctantly so.[29] Several observers have noted that CON 2 had at least the subtle flavor of endorsing balance-sheet primacy over an income-statement view to determining GAAP.[30]

Beyond CON 2, two other FASB conceptual framework documents are relevant to the changes in the 2000s that effected the purge of “reliability.” (See Table 1 for a list of FASB documents relevant to the conceptual framework changes on reliability and verifiability.) The first is CON 5, issued in 1984, and which dealt with Recognition and Measurement in Financial Statements of Business Enterprises.[31] The second is CON 7, issued in 2000, and which dealt with Using Cash Flow Information and Present Value in Accounting Measurements.[32]

In CON 5, “reliability” was tightly connected with the principles for financial statement recognition. In fact, it was established as one of four conditions for such recognition (the other three being measurability, relevance, and that the item in question met the definition of a financial statement element). CON 5 noted: “Although recognition involves considerations of relevance and comparability, recognition criteria, conventions, and rules are primarily intended to increase reliability—they are means of coping with the uncertainty that surrounds business and economic activities. Uncertainty in business and economic affairs is a continuum, ranging from mere lack of absolute sureness to a degree of vagueness that precludes anything other than guesswork. Since uncertainty surrounds an entity’s incomplete cash-to-cash cycles in varying degrees, measuring progress reliably involves determining whether uncertainty about future cash flows has been reduced to an acceptable level.”[33] To reinforce the centrality of reliability, CON 5 added: “Information about some items that meet a definition [of an element of financial statements] may never become sufficiently reliable at a justifiable cost to recognize the item.”[34]

Despite its scope and its extensive coverage of recognition principles, CON 5 gave “virtually no guidance” on measurement principles.[35] This necessitated the issuance of CON 7, which when published in 2000 was the first FASB conceptual framework document in fifteen years. On the question of current-value estimates, CON 5 had stated that information based on current prices should be recognized if it is sufficiently relevant and reliable to justify the costs involved and more relevant than alternative information. CON 7 went beyond this terse language, recognizing that by the time the FASB had been regularly issuing fair-value-based GAAP standards for about a decade. CON 7 provided “a framework for using future cash flows as the basis for accounting measurements at initial recognition or fresh-start measurements.”[36]

But even CON 7, with its emphasis on legitimizing current-value estimates, recognized the centrality of reliability established in CON 2: “The accounting problem is to balance the cost of obtaining additional information against the additional reliability that information will bring to the measurement.”[37]

The final FASB conceptual framework document I consider in this paper is referred to as CON 8. This is the document that when issued in 2010 replaced both CON 1 and CON 2. Most importantly for the subject of this paper, CON 8 eliminated reliability as a “fundamental qualitative characteristic” of accounting and reclassified verifiability from being a “necessary” to being an “enhancing” property of financial reporting.

As already noted, CON 8 was issued as a joint project with the IASB, in an effort to expand convergence between IFRS and U.S. GAAP. The IASB had not developed its extant conceptual framework; it had inherited it from its predecessor body, the IASC. That framework, which was published in 1989, was largely inspired by the FASB’s. Camfferman and Zeff note that the similarities were due to the IASC’s desire to improve its standing with the International Organization of Securities Commissions, of which the SEC was an important member.[38] In fact, beyond inspiring the IASC’s conceptual framework, the FASB’s conceptual framework had prompted similar such initiatives in Australia, Canada, New Zealand, and the U.K. by 1999.[39]

Besides being a terser version of the FASB framework, the IASB framework differed in two important ways. First, it did not explicitly mention “verifiability,” although the principle was obliquely referred to in the framework’s definition of reliability. Second, it afforded a warmer reception to the longstanding accounting principle of conservatism, which, in keeping with English traditions, it referred to as “prudence.” English accounting principles had evolved closely with company law,[40] and in this interplay between law and accounting the notion of conservative accrual estimates helped limit liability for overstatement of unrealized gains.

Given that it had inherited its conceptual framework from the IASC, Camfferman and Zeff report that the IASB was from its inception keen to revisit the framework.[41] But, the newly formed board, perhaps recognizing the magnitude of this challenge, did not want to “comprehensively reconsider all components of the framework in the pyramid, but rather [focus] principally on those issues that are more likely to yield standard-setting benefits in the near term.” These were defined as “troublesome conceptual issues that reappear time and time again in different standard-setting projects and in a variety of different guises.”[42] Ostensibly, the notion of accounting reliability was one such troublesome issue.

3. Some plausible hypotheses to explain the fall of “reliability”

3.1. The rise of fair-value accounting

By the time the FASB began reconsidering the role of reliability in its conceptual framework in 2004, that framework had been in place for twenty-four years. That time period, however, had witnessed a significant shift in the nature of accounting, from a largely historical-cost basis to a mixed basis utilizing both historical values and fair-value estimates.

Allen and Ramanna report that the proportion of FASB standards with fair-value measures increased in the 1990s.[43] While the period from the FASB’s initiation of operations in 1973 through 1990 witnessed some fair-value based standards in GAAP, the period from 1990 through 2007 represented the relative heyday for fair-value accounting. (See Figure 2.)

There are several hypotheses for the growth of fair-value accounting. Zeff argues that the unleashing of fair values in GAAP can be attributed to one key individual, Richard Breeden, who in 1989 was appointed chairman of the SEC.[44] Breeden’s appointment came at the time of the savings-and-loan crisis in the United States, where a number of financial institutions were found to have obfuscated their precarious financial position by using historical-cost accounts to represent effectively insolvent long-term assets. The marking of these assets to market, it was argued, could have mitigated the crisis by alerting investors to the situation in a timelier fashion.

In 1992, Breeden appointed Walter Schuetze as the SEC’s chief accountant. Schuetze was sympathetic to the arguments on fair-value accounting, in a stark break from the tradition at the SEC.[45] The commission had been established in the shadow of the stock-market crash of 1929, where the unverifiable use of fair values had been assigned some of the blame for the market run-up in the 1920s. Since then, the SEC had taken a hardline approach to fair-value accounting, effectively banning its use in GAAP with very few exceptions.[46] But Schuetze was not attached to the SEC’s anti-fair-value position, and, moreover, the savings-and-loan crisis provided a context where marking-to-market would have been prudent.

While the appointments of Breeden and Schuetze and the savings-and-loan crisis were the plausible sparks that lit the fair-value fire, the subsequent growth of fair-value accounting can be attributed to the increasing financialization of the economy in the 1990s.[47] Greenwood and Scharfstein report that in the early 1970s, when the FASB got started, the financial-services sector represented about 4 percent of U.S. gross domestic product (GDP); by 1990, that sector represented about 6 percent of GDP; and by 2006 it accounted for 8.3 percent of GDP.[48] The growth of derivatives, options-based compensation, and the increased use of hedging by even non-financial corporations created a demand for more fair-value based estimates in accruals. Moreover, as the financial markets underlying derivatives became more developed, the prices in these markets started to look more reliable, further justifying the use of fair values in GAAP.[49]

Allen and Ramanna associate the rising proportion of FASB standards using fair-value measures to the increasing proportion of FASB board members from the asset management and investment banking sectors.[50] It is difficult to say conclusively why these sectors enjoyed greater representation on the FASB, but one plausible explanation is the increased prestige associated with these industries that accompanied that financialization of the economy in the 1990s. Sociologists of elites have argued that the stature of those employed in the financial sector grew through the 1990s and beyond, as that sector became more central to economic activity.[51]

In the asset management and investment banking sectors, fair-value accounting has long been employed for internal-control purposes. In both sectors, firms prepare daily balance sheets at current values to estimate their outstanding assets and liabilities. As these sectors came to enjoy increased influence at the FASB, they likely advocated for the expanded use of fair values throughout GAAP.[52] Such advocacy can be attributed to their own conceptual familiarity with fair values – their ideological predisposition to a current-value basis for accounting.

But it can also be attributed to their economic incentives. Both investment banking and asset management have benefited from the use of fair-value accounting. In the case of the former, one noteworthy example is the use of fair values in accounting for goodwill acquired in corporate acquisitions. The elimination of the historic-cost practice of goodwill amortization in favor of a fair-value approach has dampened accountability for overpayment in acquisitions, which in turn has driven up deal volumes and investment-banking fees.[53] In the case of asset management, the recognition of fair-value changes in income has accelerated the recognition of unrealized gains as accounting profit, resulting in compensation ahead of delivered performance, a classic moral hazard situation.[54]

The growth of fair-value accounting represented a shift in the nature of GAAP, away from a contractual or realization view of transactions to a valuation or assumption-driven view. Wanda Wallace, an accounting academic and sometime member of the FASAC through the 1990s, characterized this as shift as a realignment of accounting away from auditing and auditability towards economics and finance.[55] Indeed, a longtime FASB insider James Leisenring, who has been recognized by colleagues as an enthusiastic supporter of fair-value accounting, noted in an interview in 2011, “I have no opinion on auditing anymore. I’m not qualified to answer auditing questions. I haven’t audited anything in twenty-eight years.”[56]

Fair-value accounting is consistent with the principles of reliability and verifiability to the extent that such current-value estimates are obtained from active, liquid markets and are thus objectively auditable. In GAAP, such estimates are now referred to as Level 1 fair values. However, a significant proportion of current-value estimates – for instance, for many derivatives, options, and hedges – are obtained from mathematical models using secondary data or assumptions as inputs. Such estimates are now referred to as Level 2 and Level 3 fair values, respectively. Level 2 and Level 3 do not always appear to meet the definition of reliability and verifiability, as laid out in CON 2.

The growth of fair-value accounting, from the 1990s through the early mid-2000s, and the subsequent expansion of its scope from Level 1 estimates to Level 2 and 3 estimates likely brought the nature of GAAP in contradiction with the FASB’s own conceptual framework. While that framework was statutorily non-binding, it represented an inconsistency and presented an avenue for criticism of the FASB.[57] Thus, one hypothesis for the amendments in CON 8 that abandoned reliability and deemphasized verifiability is that these amendments represented an effort by the FASB to update its “constitution” to represent its own evolved practice.

3.2. Clarifying the original intent of the conceptual framework

Whereas even some FASB board members have argued that the changes in CON 8 were precipitated by the rise of fair-value accounting, other FASB employees have been quick to refute this explanation. When presented with this hypothesis in an interview, one longtime FASB insider remarked, “I don’t think there’s any doubt that perhaps some people on the outside … would say that I can’t do fair value because it won’t be reliable… I’m not naïve enough to not believe that a good many people who wanted reliability to mean something different than what CON 2 said weren’t arguing against fair value; they were… A lot of people accused us that you guys just want to make it easier to do fair value.”[58]

As implied in the quote above, to these insiders the changes in CON 8 simply clarified the original intent of CON 2.[59] Indeed, the FASB’s official explanation for the change in CON 8, as outlined in that publication’s section on “basis for conclusions,” echoes this position. The joint FASB-IASB document argued that “neither [boards’] framework conveyed the meaning of reliability clearly.”[60] Interestingly, the FASB used the pushback it had received on its suspected misapplications of reliability over the years as a reason for the changes in CON 8, “The comments of respondents to numerous proposed standards indicated a lack of a common understanding of the term reliability. Some focused on verifiability or free from material error to the virtual exclusion of faithful representation. Others focused more on faithful representation, perhaps combined with neutrality. Some apparently think that reliability refers primarily to precision… Because attempts to explain what reliability was intended to mean in this context have proven unsuccessful, the Board sought a different term that would more clearly convey the intended meaning. The term faithful representation, the faithful depiction in financial reports of economic phenomena, was the result of that search…”[61]

Halsey Bullen, a longtime member of the senior technical staff at the FASB and sometime project manager of the CON 8 project, noted in an interview, “One of my collateral duties [at the FASB] was responding to inquiries about the conceptual framework… And the persistent inquiry that I heard most often, and the difficulty that I encountered most often, from 1983 on was kind of what the board talks about in the basis for conclusions [in CON 8]. People didn’t understand the definition of reliability… They tended to equate reliability with verifiability, in other words, ‘Is the number auditable?’… They looked at the long phrase ‘representational faithfulness’ and cruised right by it.[62] But my understanding in talking with the board members who had written and voted in favor of [CON 2] was that they always thought representational faithfulness was the lead quality in the group of qualities referred to as ‘reliability’… So this was a lingering concern.”[63]

To Mr. Bullen’s point, Robert Sprouse, a board member of the FASB for twelve years from its inception through 1985 and an author of CON 2, was known to be a strong supporter of current values in accounting. Mr. Sprouse held these views even as far back as the 1960s, when such ideas would have been considered “radical.” In an obituary of Mr. Sprouse, his successor on the FASB Robert Swieringa noted that he had coauthored in 1962 a document with Maurice Moonitz that argued “for less reliance on the realization concept and for the expanded use of current values.” Swieringa added, “They advocated the use of current replacement cost for inventories and plant and equipment, the use of discounted present values for receivables and payables to be settled in cash, and the reporting of holding gains or losses from revaluing inventories in income.”[64] Some of these proposals remain controversial to this day, despite the growth of fair values in GAAP. Robert Sprouse’s position on fair values would have likely made it difficult for him to reconcile intellectually a notion of reliability that centered on auditability rather than faithful representation.

3.3. International convergence pressures

A third explanation for the removal of reliability in CON 8 derives from the FASB’s convergence efforts with the IASB. In 2004, when the project to revise the conceptual framework was first taken on, both boards were eagerly pursuing convergence. And the IASB, with its momentum from impending European Union adoption of its standards and its vast and ambitious standard-setting agenda, was keen to revise its own conceptual framework.[65] That framework had been inherited from its much lower-profile predecessor, the IASC, and was perceived as a “Reader’s Digest” version of the FASB framework.[66]

One IASB board member from the time noted in an interview, “The timing was good because we [the FASB and IASB] were working on projects together, and it became important that the projects not get impeded by differences in the framework. And also, from the IASB’s point of view, we had been working on some projects where the existing framework was seriously getting in the way of being able to solve important problems. The way that our framework was expressed meant that, in order to reach the conclusions that we thought were the right conclusions on some of the projects, we actually had to be seen to be violating the framework… In particular, in dealing with IAS 37, the provisions on contingencies standard, and when you get to some of the financial instruments stuff, the definitions of assets and liabilities in the framework weren’t really met because of a bunch of probability terms in the definitions and the recognition criteria… So [the framework] was getting in the way, for example, for reaching conclusions on the right accounting for derivatives and guarantees.”[67]

In this context, it is plausible that the FASB conceded to the changes to reliability and verifiability in its conceptual framework as an accommodation to the IASB, in the spirit of achieving broader success by converging U.S. GAAP and IFRS.

4. A history of conceptual framework changes to “reliability” and “verifiability”

Table 2 provides a brief timeline of events concerning the conceptual framework changes on reliability and verifiability. Appendix 1 provides brief profiles of some FASB and IASB employees involved in the conceptual framework changes. These listings may help readers navigate the description of events that follows.

The first public inklings that “reliability” in the conceptual framework could be reassessed came in 2001 with the publication of a report by Wayne Upton, a senior FASB staffer (and later IASB staffer) and key author of CON 7 issued in 2000. The report argued that the term reliability was “often misunderstood.” It added that the FASB’s constituents often took the term reliability to mean “verifiability,” and the document reminded readers that the concept also meant representational faithfulness.[68]

Also that year, FASB board member John “Neel” Foster and FASB senior staffer L. Todd Johnson authored a document titled “Why Does the FASB Have a Conceptual Framework?” No reason was given for the issuance of the document, but it did assure readers that “No additional [conceptual framework documents] are currently planned although the Board has proposed a limited amendment to revise the definition of a liability and is exploring the possibility of a broader reconsideration of liabilities and their recognition.”[69] This document did give some warning of upcoming convergence pressures, noting that “the advent of the new IASB may also lead to further development of the framework. As part of its effort to achieve greater convergence of accounting standards internationally, the IASB is considering whether differences in the existing conceptual framework of the IASB and those of national standard setters—such as the FASB—may be impediments to convergence. If so, there may be need to reconsider those frameworks or at least certain aspects of them.”[70]

As a final indication of things to come, the Foster-Johnson document stated, “the FASB’s framework was, for the most part, developed two decades ago. Since then, business and financial activities have changed considerably and have become increasingly complex. As a result, many of the standard-setting issues of today are different and more complex than those that were contemplated when the framework originally was developed. For that reason, some updating of the framework may be both desirable and necessary to enable it to better cope with the issues of today and tomorrow.”[71]

A search of publicly available FASB documents suggests that no further documents on the conceptual framework were released until 2004. But in December of 2003, the FASB in conjunction with the American Accounting Association organized a conference focused on two topics: the reliability of numbers reported in financial statements and notes; and defining the unit of account. The conference was sponsored by the Big Four audit firms and the FASB itself, and it was attended by board members and senior staff, including two staff directors.[72]

Professors Laureen Maines and James Wahlen presented a paper at the conference, subsequently published in the journal Accounting Horizons under the title, “The Nature of Accounting Information Reliability: Inferences from Archival and Experimental Research.”[73] The presentation was intended to inform standard setters about the current state of archival and experimental evidence on reliability. One participant at the conference remarked, “The crowd’s reaction illuminated why we needed to discuss reliability and what it means (and does not mean) – there was a lot of disagreement/misunderstanding about what characteristics should comprise reliability. Even Bob Herz, then chair of the FASB, went on a bit of a long rant … describing reliability – but as I and others pointed out, his comments were actually more about relevance!  So even the FASB chair was guilty (in my view at least) of inadvertently commingling relevance into reliability.”[74]

Concurrent with the events described above, the FASB was engaging its formal advisory body, the FASAC, on the question of reliability in the conceptual framework. In July of 2001, Todd Johnson had led an “educational session” for FASAC members (who were themselves experts on accounting issues) on the conceptual framework. Mr. Johnson discussed the issue of reliability in the context of what was then a very new and controversial FASB standard – the decision to eliminate goodwill amortization and require fair-value based testing for goodwill impairment. “On the issue of goodwill amortization, the conceptual framework did not provide direct guidance,” he was recorded as saying. “Concepts Statement 2 states that reliability is one of the primary characteristics of useful accounting information. Many constituents cited reliability as a reason either to support or oppose amortizing goodwill.”[75]

Mr. Johnson stated that some critics of the FASB’s decision had “argued that the reliability of financial statements would be undermined if goodwill was not amortized. Others [had] argued that the reliability of financial statements would be undermined if goodwill was amortized.” Mr. Johnson explained how the discussion of reliability in Concepts Statement 2 provided some help to the Board: “Concepts Statement 2 states that precision often is confused with reliability… amortizing goodwill over a fixed period, such as 40 years, produces numbers for the income statement and balance sheet that are very precise. There is comfort in having such precision in the financial statements, especially if one is expressing an opinion on those statements. However, Concepts Statement 2 also notes that an important aspect of reliability is representational faithfulness. If goodwill is not decreasing in value, then amortizing it yields accounting information that is not representationally faithful because it depicts goodwill as decreasing when it is not.”[76]

In a remarkable foreshadowing of the changes that would eventually be captured in CON 8, Mr. Johnson explained how the board arrived at its final conclusion, “The Board observed that although goodwill amortization produced precise numbers, it would not produce reliable numbers if those numbers were not representationally faithful of the underlying economics.”[77]

In a FASAC meeting on December 2, 2002, FASB board member Neel Foster was recorded acknowledging that “the Board could be perceived as being inconsistent in the area [of fair values and reliability]. In some cases the Board simply tells practitioners to do the best they can in arriving at fair value, but in other cases the Board prohibits fair value estimates because it believes those estimates are not sufficiently reliable.”[78] One of Foster’s contemporaries on the board recalled him as being the “most passionate” board member on fair-value accounting during his tenure.[79] Just the previous year, Mr. Foster had coauthored a document with Mr. Johnson suggesting changes to the conceptual framework might be forthcoming.[80]

In June of 2003, as the FASAC membership was invited to comment on the FASB’s ambitious project to produce a taxonomy of fair-value measurements, the board experienced mixed reactions. The minutes noted, “Several Council members expressed support for the

[1] See, for example, John M. “Neel” Foster and L. Todd Johnson, “Why does the FASB have a Conceptual Framework?,” Understanding the Issues—FASB, August 2001.

[2] Financial Accounting Standards Board, “Conceptual Framework: Statement of Financial Accounting Concepts No. 8,” September 2010.

[3] Ibid., pp. 16, 26–27.

[4] Geoffrey Whittington, “Fair Value and the IASB/FASB Conceptual Framework Project: An Alternative View,” ABACUS 44, no. 2 (2008): 139–68; Kees Camfferman and Stephen A. Zeff, Aiming for Global Accounting Standards: The International Accounting Standards Board, 2001–2011 (New York: Oxford University Press, 2015).

[5] Interview with Subject No. 11, August 2015.

[6] Karthik Ramanna, Political Standards: Corporate Interest, Ideology, and Leadership in the Shaping of Accounting Rules for the Market Economy (Chicago: The University of Chicago Press, 2015).

[7] Stephen A. Zeff, “The SEC Rules Historical Cost Accounting: 1934 to the 1970s,” Accounting and Business Research 37, no. 1 (2007): 49–62.

[8] See, for example, Robert W. Holthausen and Ross L. Watts, “The Relevance of the Value-Relevance Literature for Financial Accounting Standard Setting,” Journal of Accounting and Economics 31, nos. 1–3 (2001): 3–75; S. P. Kothari, Karthik Ramanna, and Douglas J. Skinner, “Implications for GAAP from an Analysis of Positive Research in Accounting,” Journal of Accounting and Economics 50, nos. 2–3 (2010): 246–86.

[9] See, for example, Ramanna, Political Standards, and Michael Power, “Fair value accounting, financial economics and the transformation of reliability,” Accounting and Business Research 40, no. 3 (2010): 197–210.

[10] See, for example, James Kwak, “Cultural Capture and the Financial Crisis,” in Preventing Regulatory Capture, eds. Daniel Carpenter and David A. Moss (New York: Cambridge University Press, 2014), 72–98.

[11] See, for example, Paul M. Healy and Krishna G. Palepu, “The Fall of Enron,” Journal of Economic Perspectives 17, no. 2 (2003): 3–26; S. P. Kothari and Rebecca Lester, “The Role of Accounting in the Financial Crisis: Lessons for the Future,” Accounting Horizons 26, no. 2 (2012): 335–51.

[12] Carsten Erb and Christoph Pelger, “‘Twisting words’? A study of the construction and reconstruction of reliability in financial reporting standard-setting,” Accounting, Organizations and Society 40 (2015): 13–40.

[13] See §13 (b) Securities Exchange Act of 1934, Pub. L. No. 73–291, 48 Stat. 881 (1934).

[14] The Securities and Exchange Commission Historical Society, “The Richard C. Adkerson Gallery on the SEC Role in Accounting Standards Setting,” accessed September 2013, http://sechistorical.org/museum/galleries/rca/.

[15] Ibid.

[16] See, for example, Stephen A. Zeff, “The Evolution of U.S. GAAP: The Political Forces behind Professional Standards Part 1: 1930–1973,” CPA Journal 75, no. 1 (2005): 18–27.

[17] See, for example, Zeff, “The Evolution of U.S. GAAP.”

[18] See, for example, Foster and Johnson, “Why does the FASB have a Conceptual Framework?”

[19] Ibid., p. 2.

[20] See, for example, Stephen A. Zeff, “The Trueblood Study Group on the Objectives of Financial Statements (1971–73): A historical study,” Journal of Accounting and Public Policy forthcoming.

[21] Foster and Johnson, “Why does the FASB have a Conceptual Framework?,” p. 2.

[22] FASB, “Statement of Financial Accounting Concepts No. 1,” November 1978.

[23] See, for example, Stephen A. Zeff, “The Trueblood Study Group on the Objectives of Financial Statements (1971–73).”

[24] FASB, “Concepts No. 1,” p. 6.

[25] FASB, “Statement of Financial Accounting Concepts No. 2,” May 1980.

[26] Whittington, “Fair Value and the IASB/FASB Conceptual Framework Project,” p. 141.

[27] FASB, “Concepts No. 2,” p. 6.

[28] See, for example, Sudipta Basu, “The Conservatism Principle and the Asymmetric Timeliness of Earnings,” Journal of Accounting and Economics 24, no. 1 (1997).

[29] FASB, “Concepts No. 2,” p. 35.

[30] Under balance-sheet primacy, the income statement for a period simply reports the changes between the beginning-of-period and end-of-period balance sheets. Under the income-statement view, this document is prepared independently from first principles and then reconciled with the balance sheet, generally using a statement of comprehensive income. See, for example, Ilia D. Dichev, “On the Balance Sheet-Based Model of Financial Reporting,” Accounting Horizons 22, no. 4 (2008): 453–70.

[31] FASB, “Statement of Financial Accounting Concepts No. 5,” December 1984.

[32] FASB, “Statement of Financial Accounting Concepts No. 7,” February 2000.

[33] FASB, “Concepts No. 5,” p. 22.

[34] Ibid., p. 28.

[35] Robert J. Swieringa, “Robert T. Sprouse and Fundamental Concepts of Financial Accounting,” Accounting Horizons 25, no. 1 (2011), p. 216.

[36] FASB, “Concepts No. 7,” p. 4.

[37] Ibid., p. 22.

[38] Camfferman and Zeff, Aiming for Global Accounting Standards, pp. 358–59.

[39] Foster and Johnson, “Why does the FASB have a Conceptual Framework?”

[40] Camfferman and Zeff, Aiming for Global Accounting Standards, p. 364.

[41] Ibid., p. 359.

[42] L. Todd Johnson, “The Project to Revise the Conceptual Framework,” FASB Report, December 28, 2004, p.3.

[43] Abigail M. Allen and Karthik Ramanna, “Towards an Understanding of the Role of Standard Setters in Standard Setting,” Journal of Accounting and Economics 55, no. 1 (2013): 66–90.

[44] Stephen A. Zeff, “The SEC Suppresses Fair Value – Until the 1990s,” Presentation at the American Accounting Association Annual Meeting, August 2015.

[45] Ibid.

[46] Zeff, “The SEC Rules Historical Cost Accounting.”

[47] See, for example, Ramanna, Political Standards.

[48] Robin Greenwood and David Scharfstein, “The Growth of Finance,” Journal of Economic Perspectives 27, no. 2 (2013): 3–28.

[49] See, for example, Power, “Fair value accounting, financial economics and the transformation of reliability.”

[50] Allen and Ramanna, “Towards an Understanding of the Role of Standard Setters in Standard Setting.”

[51] See, for example, Rakesh Khurana, From Higher Aims to Hired Hands: The Social Transformation of American Business Schools and the Unfulfilled Promise of Management as a Profession (Princeton, NJ: Princeton University Press, 2007), and Power, “Fair value accounting, financial economics and the transformation of reliability.”

[52] Allen and Ramanna, “Towards an Understanding of the Role of Standard Setters in Standard Setting.”

[53] See, for example, Ramanna, Political Standards.

[54] See, for example, Mihir Desai, “The Incentive Bubble,” Harvard Business Review 90, no. 3 (2012): 124–33; Kothari and Lester, “The Role of Accounting in the Financial Crisis: Lessons for the Future.”

[55] Email correspondence, August 2015. Also see, for example, Power, “Fair value accounting, financial economics and the transformation of reliability.”

[56] The Securities and Exchange Commission Historical Society, “Oral History Project Interview with James Leisenring,” April 12, 2011, by James Stocker, accessed September 2015, http://www.sechistorical.org/collection/oral-histories/20110412_Leisenring_James_T.pdf, p. 47.

[57] See, for example, Ross L. Watts, “Conservatism in Accounting Part I: Explanations and Implications,” Accounting Horizons 17, no. 3 (2003): 207–21.

[58] Interview with Subject No. 15, August 2015.

[59] For a history of the notion of reliability in standard setting, see Erb and Pelger, “‘Twisting words’?”

[60] FASB, “Concepts No. 8,” p. 27.

[61] Ibid.

[62] The term representational faithfulness is equivalent to the term faithful representation. The former was the FASB’s original usage; the latter was adopted partly because it was easier to translate into other languages.

[63] Interview, August 2015.

[64] Swieringa, “Robert T. Sprouse and Fundamental Concepts of Financial Accounting,” pp. 210–11.

[65] Camfferman and Zeff, Aiming for Global Accounting Standards, p. 359.

[66] Interview with Subject No. 20, September 2015.

[67] Interview with Subject No. 19, October 2015.

[68] Wayne S. Upton, Jr., “Business and Financial Reporting, Challenges from the New Economy,” FASB Special Report, April 2001, p.96.

[69] Foster and Johnson, “Why does the FASB have a Conceptual Framework?,” p. 1.

[70] Ibid, p. 3.

[71] Ibid.

[72] FASB, “Report of the Chairman of the FASB to the FASAC,” December 31, 2003, p. 5.

[73] Laureen A. Maines and James M. Wahlen, “The Nature of Accounting Information Reliability: Inferences from Archival and Experimental Research,” Accounting Horizons 20, no. 4 (2006): 399–425.

[74] Email correspondence with Subject No. 27, August 2015.

[75] Financial Accounting Standards Advisory Council, “Minutes of Meeting” (July 10, 2001), pp. 8–9.

[76] Ibid., p. 9.

[77] Ibid.

[78] FASAC, “Minutes of Meeting” (December 2, 2002), p. 12.

[79] Email correspondence with Subject No. 28, August 2015.

[80] Foster and Johnson, “Why does the FASB have a Conceptual Framework?”