In September 1998, Arthur Levitt, then Chairman of the Securities and Exchange Commission (SEC), presented an address at New York University, Numbers Game. In that speech, he called attention to an escalating problem with the quality of financial reporting in filings with the SEC. The topic received a great deal of attention over the next several years, both at the SEC and in the business world at large; in the first six months of 2002 the quality of accounting reports has even been discussed in Congress. In 2000, pursuant to an SEC consent order under the Securities and Exchange Act of 1934, the American Accounting Association received a grant funding a multipart of Earnings Project. Recognizing the need to introduce future business professionals to the importance of quality early in their careers, one part of that project included funding for a special edition of Issues in Accounting Education. As stated in the call for papers to be considered for this special issue, submissions [wer e to] address some aspect of quality, and [could] be in financial accounting, managerial accounting, international accounting, auditing, systems, or tax....The overriding requirement [was] that the reader [would] be provided with ideas or materials about the quality of that he or she [could] take into the classroom. The seven cases included here are part of the response to that call. Additional manuscripts still in the review process will appear in Issues in Accounting Education, February 2003. Quality of earnings is a multidimensional concept. There are at least three distinct sets of decisions that affect the quality of earnings: decisions made by standard setters, choices made by management about which accounting methods should be chosen from a set of acceptable alternatives, and judgments and estimates made by management in order to implement the chosen alternatives. The set of accounting standards making up U.S. GAAP is generally considered to be of high quality. However, a perfect set of standards is unattainable due to the inherent tension between relevance and reliability of accounting numbers, and due to the changing nature of economic transactions over time. Even if it were possible for standard setters to agree on a single acceptable accounting alternative for every type of transaction, it might be inadvisable to do so. There may be valid reasons for allowing companies to choose among different allowed accounting alternatives. (Dechow and Skinner [2000], among others, discuss this issue in more depth.) For example, companies may enter into certain business transactions with different intents; allowing three different accounting treatments for certain investments in debt securities allows financial statements to reflect those different business objectives. Earnings management is a concept related to, but clearly not synonymous with, quality. Earnings management activities, broadly considered, can affect quality of on each of the three dimensions mentioned in the preceding paragraphs. Actions of participants in the standard-setting process that affect the quality of the allowed accounting standards could be considered management. For example, opposition to the FASB's proposed mandatory expensing of the fair value of executive stock options affected the quality of earnings. Similarly, managements' choices among allowed accounting alternatives also could be considered management. For example, decisions such as choosing the FIFO inventory cost-flow assumption in a period of increasing prices, or choosing to use straight-line depreciation, will result in earlier recognition of income, other things equal. Finally, as time passes and managements apply selected standards, when using judgments about whether criteria for recognition ha ve been met or whether estimates need revision, management may occur. …
Read full abstract