In recent years, a dramatic increase in the revenues big accounting firms derive from management consulting services has raised a red flag about auditor objectivity. The Wall Street Journal reported in April, for example, that just last year Sprint paid Ernst & Young $2.5 million for auditing but $63.8 million for other work, including $12 million for the deployment of a financial-information system. General Electric paid KPMG $24 million for auditing but more than three times that for other services. This growing trend triggered the Securities and Exchange Commission to seriously question whether auditors have a conflict of interest that compromises the quality of an audit. Left unchecked, such conflicts could undermine the credibility of earnings statements upon which stock market activity hinges. Until now, the audit industry has disputed these claims, in part because there has been no evidence to suggest auditors have lost their objectivity.
But in a ground-breaking study analyzing the effects of accounting firms’ consulting business on the independence of their auditors, Stanford Graduate School of Business faculty member Karen Nelson and her colleagues provide the first hard evidence showing that the provision of non-audit services impairs an auditor’s independence and dangerously stretches the bounds of accepted accounting practice. “Our motivation for doing the paper comes out of the SEC’s policy agenda of trying to crack down on so-called earnings management,” says Nelson, who is assistant professor of accounting at the Business School. “We were interested in whether public accountants really are performing their role as independent gatekeepers, or has it become a game of winks and nods between corporate management and the auditors because the auditors don’t want to lose these very lucrative consulting contracts.”
The study is relevant to SEC concerns about the increase in earnings management practices such as “big bath” charges (one-time write offs) and “cookie jar” reserves (setting aside funds to manipulate future earnings), and premature recognition of revenue. These are legal but sometimes dubious practices within the range of Generally Accepted Accounting Principles. The researchers looked to see if there was more creative accounting among companies that paid their accounting firms big consulting bills relative to auditing fees. There was.
Working with Richard Frankel at MIT’s Sloan School of Management and Marilyn Johnson at Michigan State University’s Eli Broad College of Business, Nelson capitalized on new data that has become available just since February 2001, when the SEC began requiring corporations to disclose all audit and non-audit fees paid by a corporation to its auditor. The study was based on data from the proxies of over 4,000 firms filed between February 5 and June 15 of 2001.
The study looked at the ratio of non-audit versus audit revenues paid by a corporation to its auditing firm. It found that over half of the firms paid more for consulting services than audit services, and that over 95 percent of firms purchase at least some non-audit services from their auditor. The study also tried to gauge whether companies with less independent auditors stretch earnings to meet or beat analysts’ earnings predictions and project a smooth earnings path. Corporations are under enormous pressure because failure to meet such predictions, even by a penny, can send a company stock into a tailspin.
The study found that corporations with the least independent auditors — those who paid the most in consulting fees versus audit fees — are more likely to just meet or beat earnings benchmarks, such as analysts’ expectations and prior year earnings expectations, and to report large discretionary earnings. This suggests more earnings management went on among companies in the sample that paid the highest proportion of management consulting fees to their auditors. “Taken together, our results suggest that the provision of non-audit services impairs independence and reduces the quality of earnings,” the researchers write. “This evidence is important given the ongoing debate about the effect of non-audit services on auditor independence,” says Nelson. “The SEC does not want earnings to be a fiction.”
The researchers also wanted to know if the new SEC audit fee disclosures are useful to investors. To find out, they measured the stock market reaction to earnings statements after the new SEC disclosure rules went into effect in February. The results showed a statistically significant stock price drop of four-tenths of a percent on the single day proxies were filed with the SEC among the 25 percent of corporations that paid the most in non-audit fees to their auditors. That suggests institutional investors and other stock market investors were scrutinizing and discounting the earnings of firms with potential conflicts of interest.
The remedy for conflicts and earnings management? Some accounting firms have spun off their consulting firms as separate ventures, but the best solution, says Nelson, is full disclosure so that investors can judge the quality of financial statements for themselves and so that the SEC can monitor earnings management.
