Quantitative Measures Of the QUALITY of Financial Reporting.

• Author: Moriarty, George B.
• Position: Statistical Data Included

High-quality financial reporting is a cornerstone of the United States capital markets. In fact most observers agree that the U.S. enjoys the greatest system of raising and allocating capital in the world. Both in terms of the system's cost efficiency and returns to shareholders, our markets are unmatched in history.

Over the last 20 years, equity returns to shareholders have averaged 15.7 percent per year. The total value of equity invested in U.S. stock markets grew to $16.09 trillion in 2000, from $1.45 trillion in 1980. In the last five years alone, the value of the U.S. equity markets has increased by $7.14 trillion.

These gains have come despite increasing criticism over the quality of financial reporting during the past five years, often after a specific financial reporting failure becomes known and receives wide and dramatic media coverage. While the individual cases are important to study, the systemic implications and calls for regulation have been imputed without any accompanying context as to the depth of the problem or the effect of restatements on market value.

Recognizing the need to measure and quantify the problem, the FEI Research Foundation partnered with Min Wu, a Ph.D candidate at New York University's Stern School of Business, on a project entitled "Quantitative Measures of the Quality of Financial Reporting." The goal of the project was to identify quantifiable metrics that can track the absolute and relative quality of financial reporting over time.

This research produced two measurable metrics for the years 1977 to 2000:

* The number of announced financial reporting restatements.

* Market value losses associated with restatements as a percentage of the total market value of equity securities.

Methodology

To measure the number of restatements, Wu searched the Dow Jones Interactive and Lexis-Nexis information services, using keyword searches on variations of the word "restatement." She reviewed all results from each case before it was added to the database. Cases stemmed from irregularities or errors that led the company to restate voluntarily or comply with an external auditor's findings, as well as restatements forced by the Securities and Exchange Commission. She excluded instances of accounting methodology changes, stock splits, dividends, inflation accounting and discontinued operations.

Wu then constructed the database, extracting the following data from the restatements that met the criteria.

* Company identifiers such as stock symbol, SIC code and CUSIP

* General facts of the restatement: date announced, upward or downward earnings revision, amount, share price at close one day before and one day after

* Basis: voluntary, SEC enforced or auditor recommended

* Type of filing restated -- 10-Q or 10-K, or both

* Reason for the restatement

To determine the market value losses associated with each restatement, the study used the change in share price from one day before and one day after the restatement announcement.

Quantifying the losses presented several challenges. Some thought was given to extending the window as much as one month prior to and following the restatement. However, that posed a risk that larger market effects would skew the results. Thus, the three-day window seemed more likely to present the effect of the restatement with as little accompanying noise as possible.

Two cases, Sunbeam in 1998 and Yahoo! in 1999, illustrate the difficulty in gauging the impact of...