Accounting Standard-Setting Organizations and Earnings Relevance: Longitudinal Evidence from NYSE Common Stocks, 1927-93

Abstract

We provide evidence on the relevance of earnings for valuation of NYSE common stocks from 1927-93. Based on a time series analysis of the explanatory power of yearly earnings-returns regressions, we investigate whether earnings relevance has increased following: (1) the empowerment of the Committee on Accounting Procedure (CAP) in 1939 as the first U.S. standard-setting body, and (2) subsequent reorganizations of the standard-setting process which led to the establishment of the Accounting Principles Board (APB, 1959-73) and the Financial Accounting Standards Board (FASB, 1973-Present). Income measurement and disclosure has been a central focus of accounting policymakers throughout the period covered by our study. In the early 1930's, the American Institute of Accountants (forerunner of the AICPA) emphasized the cardinal importance of income as explained by the fact that the value of a business is dependent mainly on its earning capacity (see AIA [1934]). Over 40 years later, the FASB in Statement of Financial Accounting Concepts #1 adopted a similar view when it concluded that the primary focus of financial reporting is on earnings and its components. Our analyses provide little evidence suggesting that the mean and median explanatory power (i.e., adjusted R2) of yearly returns-earnings regressions are significantly higher following empowerment of the CAP in 1939 and subsequent reorganizations leading to creation of the APB and FASB. We find weak evidence of a higher median during the CAP?s tenure (1939-59) compared to the Pre-CAP era (1927-38), but this result is not robust under alternate specifications of our primary tests where either yearly rank regressions are used, losses are excluded from the sample, or operating income is used in lieu of net income in the yearly regressions. We also estimate yearly models where stock price is regressed against earnings and book value similar to other recent longitudinal studies (e.g., Collins, Maydew, and Weiss [1997] and Francis and Schipper [1997]). Results of tests examining incremental earnings relevance in price regressions are consistent with results based on earnings-returns regressions: we find no evidence indicating that the valuation relevance of earnings has significantly increased since the initiation of U.S. standard-setting in 1939. Consistent with evidence in these other studies, we document a highly significant increase in the combined relevance of earnings and book value during the FASB?s tenure compared to the APB era. However, this result is largely the artifact of abnormally low relevance in the APB era. For instance, the combined relevance of earnings and book value is lowest during the APB?s tenure (1960-73), but the combined relevance of earnings and book value in the FASB era is similar to that observed during the Pre-CAP and CAP periods.