Some recent discussions of U.S. financial reporting include implicit or explicit recommendations that the U.S. abandon the current allegedly "rules-based" system in favor of a "principles-based" system, with the implication that some or all of the current difficulties facing U.S. financial reporting would be alleviated or even eliminated by such a shift. (1) In addition, Section 108 of the Sarbanes-Oxley Act of 2002 instructs the Securities and Exchange Commission (SEC) to conduct a study on the adoption of a principles-based accounting system. The study is to have four elements:
* The extent to which a principles-based accounting system exists in the U.S.;
* The time required to change to such a system;
* The feasibility of such a system (and how it might be implemented); and
* An economic analysis of the implementation of a principles-based system.
Two themes emerge from these discussions. The first theme is that the current financial reporting system in the U.S. is undesirable or inappropriate because it is rules-based, fostering an alleged current "check-box" or compliance mentality that is, in the view of some, an open invitation to financial structuring and other activities that subvert high-quality financial reporting. The second theme is that moving to a principles-based system is desirable, because such a system allows (or requires) the appropriate exercise of professional judgment. As part of these discussions, the Financial Accounting Standards Board (FASB) developed a "Proposal for a Principles-Based Approach to U.S. Standard Setting," available at http://www.fasb.org, and has held a public roundtable discussion. Relative to this commentary, the perspective of the FASB's proposal is that of standard setting, while I take a combined standard-setting and accounting-research perspective. In addition, the FASB's proposal considers several matters that I do not discuss, such as the possibility of a "true and fair override," and changes to the roles, composition, and processes of other standard setters such as the Emerging Issues Task Force.
This commentary is intended to raise several empirical questions that might be asked in discussions about the attributes, desirability, and effects of "principle-based accounting standards." I begin with an argument that U.S. GAAP is based on a recognizable set of principles derived from the FASB's Conceptual Framework, but nonetheless contains elements that cause some commentators to conclude that U.S. accounting is "rules-based." I discuss these "rules-based" elements and raise questions about how they affect the comparability, relevance, and reliability of reported numbers. I also consider the potential attributes of "principle-based standards" and, given the specifics of the current reporting system, some potential consequences if U.S. financial reporting standards were to shift so as to exhibit those attributes.
IS U.S. FINANCIAL REPORTING BASED ON PRINCIPLES?
The FASB's standard-setting activities are guided by its Conceptual Framework, as laid out in its Concepts Statements. (2) One important source of guidance is the definitions of financial statement elements in Concepts Statement No. 6. Only items that meet the definition of an element, such as asset, liability, revenue, and expense, are to be reported, and as an additional source of standard-setting discipline, the income statement elements are defined in terms of the balance sheet elements. (3) I note that, from a standard-setting perspective, it is taken for granted that an item that meets the definition of a financial reporting element is relevant to investors.
Recognition and measurement requirements of accounting standards are to be based on the qualitative characteristics of accounting information laid out in Concepts Statement No. 2. The overarching concept is decision usefulness, supported by relevance, reliability, and comparability. (4) I interpret these concepts as follows. The desire to achieve comparability and its over-time counterpart, consistency, is the reason to have reporting standards. That is, if similar things are accounted for the same way, either across firms or overtime, it becomes possible to assess financial reports of different entities, or the same entity at different points in time, so as to discern the underlying economic events. If little value is attached to having the same accounting treatment applied to identified classes of similar items, then preparers of financial reports could reasonably be left to choose the reporting that best suited their own communication strategies. (5) Once it is agreed that comparability is desirable, relev ance and reliability assist in deciding which standards to have--that is, what the requirements for recognition, measurement, and disclosure should be. To the extent U.S. GAAP is aimed at providing comparable, relevant, and reliable financial reporting, it is principles-based.
Finally, the Conceptual Framework contains criteria for recognition and measurement in Concepts Statement No. 5. However, much of this Statement "merely describes [what was then] current practice and some of the reasons that have been used to support or explain it but provides little or no conceptual basis for analyzing and attempting to resolve.. .issues of recognition and measurement...." (Storey and Storey 1998, 158.) That is, Concepts Statement No. 5 has not proved to be useful in resolving key standard-setting issues.
Some of the standard-setting difficulties in resolving recognition and measurement issues would be eliminated or at least mitigated if Concepts Statement No. 5 were thoroughly overhauled. Some of the Concepts Statement No. 5 recognition principles--the descriptions of what was then current practice--conflict with the elements definitions in Concepts Statement No. 6. For example, the application of the "completion of the earnings process" criterion for revenue recognition from Concepts Statement No. 5 sometimes causes the recording of deferred revenues that are not liabilities. In addition, the use of "earnings process" as a standard-setting concept implies that each distinct earnings process might have its own revenue recognition standard, causing the volume of accounting guidance for revenue recognition to grow almost without limit as new business models, with their attendant earnings processes, are created. One goal of the FASB's current project on revenue recognition is to eliminate the conflict between th e "earnings process" principle from Concepts Statement No. 5 and the definitions of assets and liabilities in Concepts Statement No. 6.
Examples of the Use of Principles in Setting Accounting Standards
To illustrate how principles are used in standard setting, I describe two instances involving an actual standard and a hypothetical standard. These descriptions focus on the issues faced by the standard setter in developing the principle on which a standard is to be based. I then pose several questions: Having provided a principle that clearly states the intent of the standard, how much additional explanation should be provided? How many terms should be defined, and at what level of detail? How much prescriptive explanation about how to apply the standard, such as numerical examples, should be included? My intent is to illustrate the ways in which a standard that is clearly grounded in a recognizable principle can become detailed and complex, with numerous rules, so that it appears to be rules-based and not principles-based.
In the accounting for business combinations, a key question is how to treat the difference between the purchase consideration and the book value of the acquired entity. Purchase accounting records this amount and pooling-of-interests accounting (hereafter, pooling) suppresses this amount. Comparability implies that unless there are multiple economically distinct types of business combinations, there should be only one permissible treatment. Thus, the existence of both pooling treatment and purchase treatment could be justified only if it could be shown that "pooling" combinations differ from "purchase" combinations such that the same accounting treatment could not be representationally faithful for both. Grounding its analysis in academic research, the American Accounting Association's Financial Accounting Standards Committee examined the description of a pooling of interests in Accounting Principles Board Opinion No. 16, and concluded that "although business combinations are heterogeneous transactions, none appears to have the characteristics implied by the conceptual arguments that support pooling-of-interests accounting." (AAA Financial Accounting Standards Committee 1999.)
Relevance, particularly its feedback element, implies that the accounting for business combinations should provide information that assists in evaluating the success of the business combination. Relevance supports the use of the purchase method, because the pooling method suppresses the amount of consideration paid by the acquirer. Relevance also implies that the acquired assets and. liabilities should be recorded at their fair values on the date of acquisition.
Once the purchase method of accounting is established as the only accounting treatment, and the fair value measurement attribute for acquired assets and liabilities is required, a remaining question is the accounting treatment of acquisition goodwill, the difference between the purchase consideration and the fair values of the identifiable net assets, including identifiable intangibles. Academic research (e.g., Vincent 1997) finds that investors view acquisition goodwill as an asset, in that there is a statistically reliable association with share values, but investors do not appear to view the periodic amortization of acquisition goodwill as an expense. Applying the definition of an asset from Concepts Statement No. 6, the FASB concluded that acquisition goodwill meets the definition of an asset. Combining the conceptual analysis with the empirical research evidence, we reach the conclusion in Statements of Financial Accounting Standard (SFAS) Nos. 141 and 142: acquisition goodwill is an asset with an indefi nite (i.e., unidentifiable) service life. Therefore, goodwill should be recorded at the date of the business combination, not amortized, and subjected to periodic impairment testing.
SFAS Nos. 141 and 142 are based on the concepts of comparability and relevance combined with empirical research evidence. They are principles-based in that they require a single accounting treatment for all business combinations; they require fair value measurements of acquired tangible and intangible net assets; they require the recognition of acquisition goodwill as an asset that is to be subject to impairment testing and not to periodic amortization.
Having provided a principles-based standard, what more should the standard setter include? Application of the standard requires decisions about the workings of the goodwill impairment test; for example, at what level in the organization should goodwill be tested for impairment, and how often? Since goodwill cannot be separately measured, how should the impairment test be carried out? If goodwill is found to be impaired, how should it be remeasured? The standard-setting issue is: How many of these questions should be answered in the standard, and at what level of detail. More generally, what are the costs and benefits of detailed guidance in accounting standards, and is there a level of prescriptive detail that converts a principles-based standard to a rules-based standard?
Measuring Financial Instruments at Fair Values
SFAS No. 133 states the FASB's conclusion that fair value is the most relevant measurement attribute for financial instruments. Suppose, for the moment, that the FASB wished to promulgate a principles-based standard that would require all financial instruments to be measured at fair value. Setting aside the complex issue of how to report changes in fair values, and focusing only on the measurement issue, the standard setter confronts a number of decisions. For each such issue, I describe the nature of the question, and then explain the relation between that question and the idea of principles-based standards.
Scope. As a practical matter, a threshold issue in standard setting is definitions- describing unambiguously the class of arrangements to be accounted for and the elements of the class (if any) to be excluded. (6) Definitional issues can be exceptionally complex; the definition of "derivative" in paragraph 6 of Statement No. 133 contains three specific elements, which are explained (defined) at length in the three succeeding paragraphs; there are at least 22 Derivatives Implementation Group Issues that request clarification of the definition in SFAS No. 133 (the FASB is proposing to amend the definition). Similar issues arise in defining almost any complex commercial arrangement, such as a "lease" or a "special purpose entity."
A principles-based standard that requires all financial instruments to be measured at fair value first must define the term "financial instruments," and describe which instruments (if any) meeting this definition are to be excluded from the standard. For example, the Joint Working Group of Standard Setters' Recommendations on Accounting for Financial Instruments and Similar Items (2000) contains a four-element definition of financial instrument and lists seven classes of exceptions. More generally, a financial reporting standard must define the class of arrangements--events and transactions-that is covered by the standard. Whenever an exception is allowed, the standard must also define the criteria that must be met to qualify as an exception and, possibly, explain the accounting treatment for each exception.
Definition of measurement attribute. For any measurement attribute, the standard must describe and define the empirical construct that is intended. The term "fair value" is defined as an exit amount (for assets) or settlement amount (for liabilities) in U.S. GAAP, but fair value might also be taken to mean entry value (replacement cost), net realizable value, value-in-use, or deprival value. (7)
* which of several possible prices is to be chosen;
Measurement. If the empirical construct is exit value and observable market prices are available, questions can still arise about which price to choose. Should the fair value be the bid, the ask, or somewhere between? How should sparse trading and/or stale prices be handled? What about discounts for large blocks of the instrument? In the absence of observable market prices, accountants must measure fair values using models, estimates and assumptions. The standard setter must choose the amount of prescriptive detail to include in the standard, such as:
* when should there be adjustments to observed prices and what should those adjustments be;
* which of the population of possible models are permitted; and
* which kinds of assumptions are acceptable.
WHY ARE U.S. FINANCIAL REPORTING STANDARDS VIEWED AS "RULES-BASED"?
Based on the preceding discussion, I lay out some possible reasons why commentators might find U.S. GAAP to be "rules-based," not "principles-based." Specifically, I discuss how scope exceptions, treatment exceptions, and the presence of detailed implementation guidance can be viewed as making U.S. GAAP "rules-based" even if the standard itself is discernibly based on a recognizable principle. I describe how each "rules-based" attribute of U.S. GAAP adds to the length and complexity of the standards, and how each attribute relates to one or more of the three basic elements of the Conceptual Framework: relevance, reliability, and comparability. In this discussion, I note the possibility of trade-offs among these concepts. For example, some believe there is an inherent trade-off between relevance, such as more timely reporting, which requires more estimates and judgments, and reliability, where reporting is based on transaction amounts, with little or no estimation. There may also be a trade-off between compara bility, which facilitates interfirm comparisons, and predictive ability, which facilitates calculations of intrinsic value. Predictive ability may require firm-specific reporting choices to reflect the idiosyncrasies of business models.
Scope exceptions and alternative treatments are often provided by standard setters to meet constituent concerns or, particularly in the case of scope exceptions, to avoid a conflict with a large and established body of standards; for example, insurance contracts are usually excluded from otherwise applicable standards and most employee benefit arrangements are excluded from the scope of a standard on financial instruments. SFAS No. 133 lists nine exceptions to the definition of a derivative, several of which are clearly intended to reduce costs to preparers. An example is the scope exception for normal purchases and sales.
A scope exception requires a description of the class of arrangements to be excluded. The scope exceptions in SFAS No. 133 have led to 19 Derivatives Implementation Group Issues, all of them attempting to clarify the applicability of the exceptions. Even if a standard is firmly grounded in a principle, scope exceptions add to the volume and complexity of both the standard and, in many cases, subsequent implementation guidance.
U.S. standards also sometimes provide for treatment exceptions to reduce income volatility or to achieve a specific accounting outcome for a specified class of arrangements. In some cases, exceptions and alternatives are provided in response to explicit and strongly worded constituent concerns. Examples of treatment exceptions included because of constituent concerns appear in:
* SFAS No. 115, which provides for three categories of marketable securities, each with its own accounting treatment;
* SFAS No. 133, which provides for hedge accounting for certain qualifying arrangements;
* SFAS No. 143, which provides for a deviation from fair value measurement explicitly to reduce income volatility (see paragraphs B49-B52 in SFAS No. 143); and
* SFAS 140, which includes the qualifying special purpose entity (QSPE) to assist enterprises in achieving derecognition treatment in certain arrangements involving financial assets.
The only way to provide for treatment and scope exceptions is by means of rules. Once a standard lays out a financial reporting principle, the only way to permit deviations from that principle is to include in the standard a rule or rules that provide for some combination of excluded items (scope exceptions) or items that receive special treatment (treatment exceptions). These rules add to the length and complexity of the standard, and lead to requests for explanations of the breadth of the exceptions. Because SFAS No. 123 provided for continuation of the APB Opinion No. 25 treatment of fixed plans granted to employees, numerous requests to clarify just how variable are the pricing and other terms of a fixed plan, and how expansively the notion of employee might be applied were eventually addressed in FASB Interpretation No. 44. Given the volume and complexity of guidance that results from devising and then explaining scope and treatment exceptions, it is possible that the "rules-based" descriptions of U.S. G AAP derive from the numerous scope and treatment exceptions in the standards.
However, while scope and treatment exceptions undoubtedly add to the length and complexity of standards, the empirical question remains: How, if at all, do these exceptions affect comparability, relevance, and reliability? To the extent that exceptions! alternatives cause similar items to receive different accounting treatments, comparability is reduced. Effects on relevance and reliability, however, are not so clear. It is an empirical question whether, or how much, either relevance and/or reliability are affected when a standard provides for deviations from basic recognition and measurement principles, so that the principles are applied to only part of an otherwise similar class of events and transactions.
As noted earlier, some of the detail and complexity in U.S. GAAP stems directly from requests for clarification or expansion of scope and treatment exceptions. However, a great deal of the detail also stems from explanations of precisely how to apply the standard, sometimes even in the form of numerical examples. The next section discusses several effects of including detailed guidance, either in the standard itself or in response to subsequent questions.
SOME EFFECTS OF DETAILED IMPLEMENTATION GUIDANCE
One alleged benefit of detailed implementation guidance is increased comparability. That is, specific guidance on how to apply a standard should reduce the effects of differences in professional judgment. But to the extent that the guidance is inappropriately strict, the result will be surface comparability, and dissimilar arrangements will be forced into the same accounting treatment. It is not clear whether increasing the amount of detailed guidance increases this risk.
Given the amount of detail that currently exists in U.S. accounting standards, and given that the goal of this detail is increased comparability, the key empirical issue is:
How much comparability actually exists in U.S. firms' financial reports? That is, how much comparability benefit are we obtaining from our detailed guidance? This question poses exceptionally difficult research design issues, in that answering it requires holding constant the underlying events/transactions to be accounted for. (8) However, until we can measure the amount of comparability obtained for a given level of detailed guidance, we will not be well armed with evidence on which to base discussions about the desirability of limiting the amount of detail provided in standards, when the intent of that detail is increased comparability.
A second likely benefit of detailed guidance is increased verifiability, that is, increased consensus about measurements. To the extent detailed guidance provides preparers and auditors with a common knowledge base and a common set of assumptions, the incidence of differences in measurements should be reduced.
Opportunities for Earnings Management
A third effect of detailed implementation guidance is on the opportunities for earnings management, that is, a subversive implementation of the standard, not consistent with the standard's intent. Research by Nelson et al. (2002) suggests that detailed guidance reduces earnings management achieved through management judgments and increases earnings management achieved through transaction structuring. Two empirical questions arise from this observation. First, how (if at all) does earnings management, regardless of how it is achieved, affect comparability, relevance, and reliability? For example, does earnings management make earnings less (or more) value-relevant or informative? Second, are there differential effects, depending on whether the earnings management is achieved by means of management judgments instead of transaction structuring?
Enforcement and Litigation
Two other possible effects of detailed implementation guidance are of special interest to preparers, auditors and, possibly, regulators: (1) reduced difficulties with enforcement bodies in after-the-fact disputes over a given accounting treatment, and (2) reduced incidence of litigation over allegedly defective accounting. With regard to the former, both preparers and auditors may find the cost of dealing with enforcement agencies to be much reduced if there is clear and detailed guidance as to what is expected. While enforcement actions and restatements are highly visible manifestations of these costs, there are also the less visible potential costs, which involve day-to-day interactions with the SEC staff over periodic filings and registration statements.
The SEC, the U.S. enforcement agency for accounting and financial reporting, has limited resources to devote to answering registrant queries about acceptable interpretations and applications, and limited resources to devote to detailed discussions with registrants about the choices made in preparing SEC filings. The existence of comprehensive and unambiguous statements of the expected accounting implementations simply reduces the SEC staffs cost of understanding idiosyncratic judgments of managements.
Finally, some have claimed that detailed implementation guidance not only reduces the incidence of after-the-fact disputes with enforcement agencies, but it also reduces the incidence of litigation over accounting treatments. That is, one of the arguments made against the removal of detailed guidance, or in favor of at least this attribute of a rules-based system, is that the guidance provides protection against litigation. I believe there is no systematic evidence on the question of whether detailed accounting guidance has any effect on either the incidence or the cost of litigation. To study these effects, it is necessary to analyze the specific allegations made in litigation. If the allegations involve violations of broad principle, then the presence or absence of details, and claims about adhering to the details, seem less relevant. It is an empirical question, therefore, whether a shift toward less detail in the standards would have any noticeable effect on litigation. Perhaps methods and approaches used to analyze the existing data on causes of accounting based litigation could be used to answer this question.
SOME ADDITIONAL CONSIDERATIONS IN A DISCUSSION OF "PRINCIPLES-BASED" STANDARDS
In this section, I offer some conjectures about possible effects of a shift toward standards with no (or very few) exceptions and alternatives and little detailed implementation guidance. Some of my conjectures involve effects on accounting education and accounting research, as well as effects on other participants in the financial reporting process.
Preparer and Auditor Expertise
If standards require substantially increased professional judgment about, for example, measurement, I believe that the type and amount of expertise required of both preparers and auditors will shift. Currently, U.S. standards tend to provide extensive discussions about the application of the standard and often include detailed numerical examples that illustrate how to apply the standard to specific facts and circumstances. If this guidance is removed from the standard, then individual firms and their auditors will have to work out the application details by looking to the standard's intent.
This issue seems particularly important as the number of required estimates and judgments increases. That is, in the interest of increasing relevance, financial reports prepared in accordance with U.S. GAAP increasingly are based on estimated or measured amounts and not transaction amounts. The estimated amounts in turn require a measurement objective, such as fair value, and a measurement approach, such as application of a model or application of a discounted cash flow technique. If reported numbers are not to decline to unacceptably low levels of verifiability, preparers and auditors will have to develop an appropriate level of common understanding about when and how to apply these measurement approaches.
If the emphasis on measurement expertise increases, one potential implication for accounting education is the necessity of developing the needed measurement skills within the accounting education program. While several accounting curricula offer education in valuation of whole enterprises, or subunits of enterprises, based on discounted cash flow or discounted residual income approaches, this level of training is not sufficient to develop the expertise needed to measure the amounts to be reported for specific assets and liabilities.
Volatility of Reported Income
If adoption of principles-based standards implies no (or very few) scope and treatment exceptions that are intended to smooth income (reduce volatility), one effect would be increased volatility in reported income numbers. The underlying economic volatility will not have shifted, simply because the accounting treatment changes. But to the extent preparers, investors, and others prefer accounting treatments that smooth out inherent economic fluctuations, there would be dissatisfaction with a principles-based reporting system that eliminates treatment alternatives which exist simply to smooth income.
Accounting research has pointed to contracting incentives, including both compensation incentives and incentives arising from debt contracts, as one reason why preparers might prefer low income volatility. For example, DeFond and Park (1997) report evidence consistent with the view that managers smooth earnings to affect perceptions of their relative performance over time. Accounting research (for example, Barth et al. 1999) has also documented apparent rewards, in the form of higher price-earnings multiples, to firms with persistent patterns of increasing earnings. Research, therefore, has shown that income volatility matters, so it is an empirical issue how contracting incentives and investor responses to earnings would shift if the consistent application of principles-based standards were to increase reported earnings' volatility.
Inconsistencies in Financial Reporting
The large body of U.S. GAAP that now exists contains various degrees of detail- but it is probably safe to say that the standards are very detailed relative to those likely to be produced under a principles-based system. The question then arises as to how to manage the inevitable lengthy transition, while the standard setters rewrite the existing literature to make it principles-based. In doing so, they will also of course change accounting policies, which in turn leads to over-time inconsistencies in reporting (the historical financial reports of a firm are not comparable with its current financial reports whenever accounting policies change). If these inconsistencies impair relevance, reliability and/or comparability, the quality of financial reporting will be temporarily diminished, by some unknown amount, during the transition period.
Currently, U.S. GAAP offers a variety of transition alternatives when accounting standards change. Retroactive application, which restates the financial reports for all periods presented, enhances consistency and comparability and imposes the greatest costs on preparers and auditors. Prospective application, in contrast, simply applies the new standard either on a given date or as of a fiscal year or quarter. Variants of prospective application include "grandfathering" all existing arrangements, so that the standard applies only to new events/transactions within its scope. This approach reduces consistency and comparability and imposes the least costs on preparers and auditors.
U.S. GAAP also includes the "cumulative effects" approach, which reports in a single income statement line item the effect of applying the new standard as of the beginning of the year. The cumulative effect is measured as the difference between retained earnings at the beginning of the period and the retained earnings that would have been reported if the new accounting standard was applied retroactively to all affected prior periods. Relative to the retroactive application approach, the cumulative effects approach also reduces comparability and consistency.
The IASB has tentatively determined to require retroactive application (with few exceptions). (9) A strict interpretation of consistency suggests retroactive application is the only transition approach. However, this approach also imposes the greatest cost on preparers of financial statements.
Interactions between Principles-Based Financial Reporting Standards and Corporate Governance
The recent enactment of the Sarbanes-Oxley Act of 2002 has potential implications for the adoption of principles-based financial reporting standards. I interpret this Act as providing detailed and prescriptive corporate governance guidance, at the federal level, with special emphasis on matters related to financial reporting and auditing. (10) This approach is, in my view, markedly different from the principles-based approach that has historically been taken at the state level, where the fiduciary duties of care and loyalty have formed the basis of governance guidance for officers and directors of corporations. Thus, the Sarbanes-Oxley Act is a dual departure from past governance statutes, in that it is both prescriptive and detailed, as opposed to broadly principles-based, and it is imposed at the federal level rather than the state level.
If the Sarbanes-Oxley Act opens the door to ever more detailed legal prescriptions for the behavior of officers and directors, I predict that preparers of financial reports and the audit committees charged with oversight of the process will seek ever more detailed guidance as to precisely what is expected in terms of their role in financial reporting. That is, they will seek rules. Such a demand for rules would conflict with principles-based financial reporting standards that require the exercise of professional judgment, with all its inherent ambiguity.
In this commentary, I have argued that U.S. financial reporting standards are in general based on principles, derived from the FASB's Conceptual Framework, but they also contain elements--such as scope and treatment exceptions and detailed implementation guidance--that make them also appear to be rules-based. I discuss the effects on comparability, relevance, and reliability of these rules-based elements, with special attention to the effects of detailed implementation guidance. I emphasize that detailed guidance is intended to achieve comparability and pose an empirical question: How much comparability actually exists in U.S. financial reporting? An assessment of the potential sacrifice of comparability that some believe would surely accompany the adoption of principles-based standards requires, first, an understanding of the current state of comparability." (11)
I also comment on whether the existence of detailed guidance meets institutional needs of preparers and auditors in dealing with enforcement agencies and with the threat of litigation, and I speculate about how behavior might change if U.S. standards were to eliminate most or all scope and treatment exceptions and provide a much-reduced level of implementation guidance. I note that at least some claims made about the effects of detailed guidance are at least in principle amenable to empirical investigation using the tools of accounting research.
(1.) Examples include Joseph Berardino, former CEO of Arthur Andersen Worldwide (Business Week, August 12, 2002, page 56); Walter Wriston, former CEO of Citicorp, which is now part of Citigroup (see "The Solution to Accounting Scandals? Simpler Rules," The Wall Street Journal, August 5,2002); Sir David Tweedie, current chairman of the International Accounting Standards Board (February 14,2002, testimony before the U.S. Senate Committee on Banking, Housing and Urban Affairs); Harvey Pitt, then-chairman of the U.S. Securities Exchange Commission (March 21,2002 testimony before the same Senate committee).
(2.) Some might argue that the FASB does not consistently follow its own Concepts Statements, by pointing to examples such as SFAS No. 2, which rejects internally developed intangibles as assets, or SFAS No. 13 and SFAS No. 66, which provide a series of bright lines and numerical thresholds for lease accounting and sales of real estate, respectively.
(3.) As described by Storey and Storey (1998). part of the intent of using definitions of financial statement elements was to eliminate the recording of miscellaneous debits and credits on the balance sheet. Examples include the recording of "deferred losses" as assets and the recording of "deferred revenues" as liabilities even though the "deferred revenues" do not constitute an obligation.
(4.) Both relevance and reliability are, in turn, supported by other subconcepts, as discussed in Concepts Statement No. 2.
(5.) For example, Healy and Palepu (1993) discuss the value of managerially contrived disclosure strategies based on managers' superior information. U.S. GAAP contains some examples of intentional noncomparability, in the sense that reporting treatment is based on management intent; one such example is SFAS No. 131 in which segment definitions are to be based on firm-specific organizational structures. To the extent such firm-specific decisions enhance predictive ability, even at the cost of sacrificing considerable comparability, they are not inconsistent with the Conceptual Framework, which puts forward predictive ability as an element of relevance.
(6.) In the discussion of accounting for business combinations, I ignored the question of defining the transaction or arrangement to be accounted for--in this case, a business combination. Emerging Issues Task Force (EITF) Consensus 98-3 attempts such a definition, in the context of distinguishing a business combination from a nonmonetary exchange.
(7.) Deprival value is arrived at by asking how much worse off would the entity be without the asset in question. This approach to fair value measurement has been adopted by the U.K. Accounting Standards Board, under the label of "value to the business."
(8.) Pownall and Schipper (1999) note that one can assess comparability of reporting systems in the specialized setting of Form 20-F reconciliations filed by non-U.S. SEC registrants, because the design holds constant the reporting entity and varies the applicable standards. Assessments of within jurisdiction comparability, in contrast, hold constant the reporting system and vary the reporting entities. These are essentially assessments of implementation and not of standards.
(9.) Exceptions would be expected, for example, if preparers do not have the information necessary to apply the standard retroactively.
(10.) For example, Section 301 or the Act specifies in some detail the responsibilities of corporate audit committees and Section 303 makes it unlawful to exert inappropriate influence on auditors. These matters are not left to the business judgment of governing boards operating in accordance with the fiduciary duties of care and loyalty.
(11.) A significant body of accounting research has investigated the value relevance of reported accounting numbers. Such tests have been described by, for example, Barth et al. (2001) as joint tests of relevance and reliability. In contrast, little or no research on within-jurisdiction comparability exists.
American Accounting Association (AAA) Financial Accounting Standards Committee (FASC). 1999. Methods of accounting for business combinations: Recommendations of the G4+1 for achieving convergence. Accounting Horizons 13 (September): 299-303.
Barth, M., J. Elliott, and M. Finn. 1999. Market rewards associated with patterns of increasing earnings. Journal of Accounting Research 37 (Autumn): 387-413.
-----, W. Beaver, and W. Landsman. 2001. The relevance of the value relevance literature for financial accounting standard setting: Another view. Journal of Accounting and Economics 31 (1-3): 77-104.
DeFond, M., and C. Park. 1997. Smoothing income in anticipation of future earnings. Journal of Accounting and Economics 23 (2): 115-139.
Financial Accounting Standards Board (FASB). 2002. Proposal for a principles-based approach to U.S. standard setting. Available at: http://www.fasb.org.
Healy, P., and K. Palepu. 1993. The effect of firms' financial disclosure strategies on stock prices. Accounting Horizons 7 (March): 1-11.
Joint Working Group of Standard Setters. 2000. Recommendations on Accounting for Financial Instruments and Similar Items. Norwalk, CT: Financial Accounting Standards Board.
Nelson, M., J. Elliott, and R. Tarpley. 2002. Evidence from auditors about managers and auditors' earnings management decisions. The Accounting Review 77 (Supplement): 175-202.
Pownall, G., and K. Schipper. 1999. Implications of accounting research for the SEC's consideration of International Accounting Standards for U.S. securities offerings. Accounting Horizons 13 (September): 259-280.
Storey, R., and S. Storey. 1998. The Framework of Financial Accounting Standards. Norwalk, CT: Financial Accounting Standards Board.
Vincent, L. 1997. Equity valuation implications of purchase versus pooling accounting. Journal of Financial Statement Analysis (Summer): 5-19.
The views expressed in this commentary are my own, and do not represent positions of the Financial Accounting Standards Board. Positions of the Financial Accounting Standards Board are arrived at only after extensive due process and deliberation. This commentary is based on a presentation at the 2002 American Accounting Association Annual Meeting; I thank Joel Demski for giving me the opportunity to make that presentation. I appreciate comments from Robert Freeman, James Largay, and Robert Lipe.
Katherine Schipper is a member of the Financial Accounting Standards Board.…