By Christine Wiedman
Illustration: Cathy Pentland
 AUDITORS AND REGULATORS CAN RESTORE THE PUBLIC'S FAITH IN THE FINANCIAL REPORTING SYSTEM
In a post-Enron environment, it is difficult to avoid reading about the crisis in financial reporting, with managers, auditors and boards all being implicated. This article shows how recent research supports the view that earnings are sometimes managed, and that auditors and boards of directors can provide countervailing pressures to significantly increase the quality of financial statements.
Earnings management: a research perspective
High-profile cases such as Enron focus on the earnings "manipulation" or fraud of companies that have been investigated by the US Securities and Exchange Commission. These cases are clear examples of manipulation because they are extreme and involve violations of generally accepted accounting principles. But researchers are also interested in less egregious cases where managers are acting in an opportunistic manner. Therefore, they examine earnings "management" defined as "a purposeful intervention in the external financial reporting process, with the intent of obtaining some private gain [as opposed to ... merely facilitating the neutral operation of the process.]"1
Researchers are faced with a challenge, as it is difficult to identify earnings management without knowing management's true intentions. Take, for example, the case where a manager lengthens the useful life of an asset to 10 years from seven, thereby reducing depreciation expense and increasing income. This change may reflect management's true belief about the productive capacity and usefulness of the asset. (Presumably, management would not have purchased the asset without feeling optimistic about its usefulness.) Alternatively, management may recognize the uncertainty regarding the useful life estimate and might more aggressively choose the highest number still within an acceptable range. Or it may increase the estimate simply to increase earnings so that analysts' expectations for the period will be met.
Academic accounting researchers cannot differentiate very easily between these various scenarios. Therefore, they must infer earnings management from patterns of reported numbers they can observe. To find persuasive evidence, they first look for contexts where earnings management is most likely to occur, recognizing that a strong motive is typically required. Second, they try to gather large samples of firms in a particular context to provide systematic evidence of earnings management across the sample.
Measuring earnings management
How do researchers establish evidence of earnings management? Many opportunities for earnings management arise from the estimates and judgments inherent in the accrual accounting system, such as the allowance for doubtful accounts, inventory valuation and depreciation expense. Thus, researchers have focused on total accruals, measured as the difference between net income before extraordinary items and cash flows from operations, and comprised of changes in noncash working capital accounts and noncash income statement items. They are interested in the portion of these accruals resulting from "purposeful intervention," so they attempt to break total accruals into two components: a non-discretionary component that naturally arises from the company's economic activities, and a discretionary, or managed, component. The assumption behind separating these components is that changes in working capital move proportionately with changes in sales and that noncash income statement items such as depreciation expense and future taxes move proportionately with the level of property, plant and equipment. When total accrual changes do not move proportionately with these items relative to other firms in the same industry and year, then the unexpected, or discretionary, portion of total accruals is assumed to be the managed portion.2 Because discretionary accruals are on average zero, large positive (negative) discretionary accruals are taken as evidence of income-increasing (income-decreasing) earnings management, even though management's true intent cannot be known.
Since this method provides a measure of earnings management for all firms, not only those under investigation by the SEC, it can be used to find evidence of earnings management on broad samples of firms in different contexts. It is by far the most-commonly used method but other measures are also being developed. For example, Beneish uses a model to predict the probability of earnings manipulation for a given company,3 which includes a broader set of variables that capture both distortions in financial statement ratios and preconditions that might prompt companies to engage in earnings management.
Motivations for earnings management
Why do firms manage earnings when there are costs associated with doing so? For example, sales that are overstated in one period result in lower sales in the next. Further, earnings management can impair the perceived quality of a firm's earnings. And in the extreme case — where a violation of GAAP does occur — the consequences to management and the company can be severe, including significant stock price declines, shareholder lawsuits, dismissal and civil and criminal penalties imposed by regulators.4
One such example is America Online's accounting for subscriber acquisition costs. In the 1990s, AOL had an accounting policy of capitalizing and then amortizing these costs rather than expensing them outright. Although this policy was considered appropriate by some and was clearly disclosed in the notes, it differed from AOL's major competitor. Moreover, many analysts viewed it as overly aggressive, resulting in significant negative press. Eventually, amid increased short-selling in the market of the company's shares, AOL was pressured to change its policy to immediately expensing its subscriber acquisition costs.5
Because of these costs, earnings management is most likely to occur in circumstances where the perceived benefits exceed the costs. Recent research has tended to focus on capital markets contexts, such as raising capital or meeting benchmarks. One incentive for earnings management is to increase the stock price before an equity offering.
Marquardt and I conducted a study on a sample of firms selling shares in secondary offerings.6 Using the prospectuses filed with the SEC, we classifed the selling shareholders as either management (executives and directors) or nonmanagement. Because managers have both the incentives and the opportunity to manage earnings, we expected that there would be greater evidence of earnings management for secondary offerings where management is selling their own shares. By increasing earnings, managers may be able to increase the stock price at the time of the offering, thereby increasing the proceeds they personally receive from the sale.
Consistent with our expectations, we found that firms with management selling shares have higher discretionary accruals on average than firms where management is not selling. These are average effects: there are still firms within the sample that show no evidence of earnings management. However, the median discretionary accruals for the management firms exceed those for the nonmanagement firms by 2.8% of total assets. This statistically and economically significant difference is taken as evidence of earnings management for the management firms in this context.
Related studies have found evidence of income-increasing earnings management prior to both initial public offerings7 and seasoned equity offerings.8 For example, Teoh, Welch and Wong find that seasoned equity issuers have higher net income growth in the issue year compared to nonissuing peer firms, but that they underperform after the issue. Further-more, in the offering year, median discretionary current accruals exceed non-issuing peers by 2.9% of total assets, suggesting the accruals are driving the at-issue peak in income. Similarly, in stock for stock mergers, acquiring firms appear to manage earnings upward in an attempt to increase their stock price prior to the merger, thereby reducing the number of shares they must use in the exchange.9
It appears that companies also manage earnings to meet different earnings benchmarks. Burgstahler and Dichev consider whether firms manage earnings to avoid reporting an earnings decrease from the previous year.10 They find that earnings changes slightly less than zero do not occur as frequently as would be expected and that earnings changes slightly greater than zero occur more frequently than expected. Therefore, a number of firms close to reporting a slight earnings decrease appear to manage earnings upward to avoid doing so. This suggests that being able to report that earnings are up is of significant value to managers. In fact, Burgstahler and Dichev find that incentives to avoid earnings decreases become even stronger if there is a run of previous earnings increases; hence, the pressure is even greater for managers to be able to report that earnings have been up, for say four years in a row, if the previous three years have had increases. In another study, Marquardt and I find that these "earnings decrease avoiders" also have discretionary accruals that are significantly positive and higher than those for a control sample matched on industry, performance and size,11 which confirms the Burgstahler and Dichev finding using the more traditional discretionary accrual measure.
Burgstahler and Dichev find a similar result for earnings levels that are close to zero. They find that many more firms report small profits than small losses, which suggests that firms manage earnings to avoid reporting a loss. Degeorge, Patel and Zeckhauser argue this threshold arises from the "psychologically important distinction between positive numbers and negative numbers (or zero)."12 Another important threshold is meeting analysts' earnings forecasts. Certainly, the business press frequently cites declines in a company's stock price after the company's reported earnings fell short of expectations. Payne and Robb find evidence consistent with firms managing earnings to meet analysts' forecasts.13 In particular, their findings suggest that managers use income increasing discretionary accruals where analysts have reached a consensus and unmanaged earnings are below analyst expectations.
Are financial statement users misled by earnings management?
Although earnings management negatively impacts earnings quality, the question remains whether it matters: that is, whether or not it has any economic consequences. Recent studies have started to examine this aspect of earnings management. In our secondary offering study, for example, Marquardt and I examine whether the earnings management impacts the relation between stock prices and earnings. We find this relation declines significantly for the firms with management selling in the year of the offering (the earnings response coefficient, or pricing multiple, on earnings drops to well below half its value for the management group). This finding is consistent with investors discounting earnings, where upward earnings management is suspected.
Is the discounting sufficient? While some of the evidence here is mixed, the research generally suggests it is not. For example, Teoh, Welch and Wong find that firms issuing equity in a seasoned offerings underperform the market following an offering and that this underperformance is related to the extent of earnings management.14 Specifically, they find that issuers with higher discretionary accruals (contributing to higher reported net income) prior to the offering have lower stock price performance and lower net income following the issuance. In a more general context, Bradshaw, Richardson and Sloan find that firms with high accruals are more likely to experience future earnings reversals.15 They further find that sell-side analysts do not anticipate these reversals in their earnings' forecasts — forecast errors are large and negative for firms with unusually high accruals, and that auditors do not communicate (through modified audit opinions, for example) the increased likelihood of future earnings declines or GAAP violations for high accrual firms. Together, these studies suggest that investors and financial statement users are misled by earnings management, at least to some extent.
Factors that strengthen the quality of financial reporting
While the research here suggests earnings management does occur, other research points to the positive role both corporate governance and the auditor can play in the financial reporting process. This also reinforces the importance of initiatives currently being undertaken by both the capital market regulators and the auditing profession.
Corporate governance
In recent years, regulators have taken steps to increase the strength of corporate governance. In 1995, the Toronto Stock Exchange issued guidelines for corporate governance and in November 2001, together with the Canadian Institute of Chartered Accountants and the Canadian Venture Exchange, formed a joint committee and issued a report calling for further improvements.16 Similarly, in the US, the SEC asked the New York Stock Exchange and the National Association of Securities Dealers to sponsor a Blue Ribbon Committee on improving the effectiveness of audit committees.17 Most recently, in June, the NYSE adopted a plan requiring its listed companies to have a majority of their directors to be independent of the firm and to create a committee of independent directors to select new board members.18
Collectively, these initiatives emphasize the need for disclosure of corporate governance practices, the importance of independence for board effectiveness, and the critical role that the audit committee plays in the financial reporting process. For example, the joint committee's final report maintains the need for the board of directors to be composed of a majority of unrelated (independent) directors, and recommends the need to designate an "independent board leader." It further recommends that the audit committee be composed only of outside directors, that all members of the audit committee be "financially literate" and that at least one member should have accounting or related financial expertise.
Board independence
Recent academic research has examined whether strong corporate governance does, in fact, improve financial reporting quality. Specifically, Klein has looked at whether corporate governance variables are related to earnings management for 687 large publicly-traded US firms.19 Like the previous studies discussed, she examines earnings management by measuring the magnitude of discretionary accruals. After controlling for other variables related to earnings management and governance, her study does not find the independence of the board of directors (measured as greater than 50% of members being independent) to be significant in relation to earnings management. However, independence of the audit committee is. When the majority, greater than 50%, of the audit committee is independent, discretionary accruals are significantly lower than when it is not. Interestingly, in light of recent recommendations by regulators, a majority appears to suffice. Her study does not find a significant difference in earnings management across firms with a 100% independent audit committee and firms with audit committees that are less than 100% independent. Klein argues, "a majority outside membership is the critical threshold for deriving a meaningful relationship between director independence and earnings manipulation." (p. 21)
The study also examines the importance of relationship investing –– where a large block-holder can take an active, interventionist role on the board. In this sample, discretionary accruals are significantly lower for firms where a large nonmanagement block-holder (who owns at least 5% of the share outstanding) sits on the audit committee, suggesting that such investors can play an active role in governance. These results support recent initiatives by Canadian pension funds and money managers to join forces in order to be more effective in influencing corporate governance.20
Audit committee expertise
The US Blue Ribbon Committee and the Canadian Joint Committee recommend that all audit committee members be financially literate and that at least one audit committee member be a financial expert (meaning that they must have past employment experience in the area finance/ accounting or have professional certification in the area).
McDaniel, Martin and Maines use an experimental research design to examine the various ways in which experts and literates differ in their evaluation of financial reporting quality in financial statements.21 They find that when identifying critical financial reporting quality issues, experts are more likely to identify those related to less prominent but recurring activities, whereas literates are more likely to raise issues prominent in the business press and nonrecurring in nature. They conclude that the inclusion of experts in audit committees is likely to result in the identification of issues that other audit committee members might have missed, but that both experts and literates provide "complementary diversity" in the evaluation of financial reporting quality. These findings reinforce the recommendations by the regulators.
The role of the auditor
Despite the negative reputation effects of accounting scandals such as Arthur Andersen and Enron, auditors do add value to the financial reporting process by reducing earnings management. Francis and Krishnan argue that auditors can curb discretionary accruals and that they can lower the threshold for issuing a modified audit report where risk exposure is higher.22 They find that auditors of high-accruals firms are more likely to issue modified opinions for asset realization uncertainties and for going concern problems. They also find that only the (then) Big Six firms demonstrate this form of conservatism, building on the research that finds that the audits of these firms are of higher quality than other firms.
Another study finds that clients of non-Big Six firms report discretionary accruals that are, on average, 1.5% to 2.1% of total assets higher than the discretionary accruals reported by clients of Big Six auditors.23 The authors conclude that higher audit quality is associated with lower income-increasing earnings management. They note one important caveat: ideally, they would like to study the unwarranted accruals prevented by each auditor group but note that this information is not observable for their study.
A recent working paper by Nelson, Elliott and Tarpley makes progress in this direction by surveying auditors to assess their role in preventing and reducing earnings management.24 Their study includes 515 auditor experiences with potential earnings management. The findings show that 44% of these earnings management attempts were, in fact, not waived by auditors, and that the earnings management was less likely to be waived if it was income-increasing. (Reasons for waiving the issue included: the client demonstrated it was complying with GAAP; there was no convincing evidence that the company's position was incorrect; or some other reason such as immateriality.) They therefore conclude that auditors do play an important role in reducing earnings management. They also find, however, that earnings management relating to unstructured transactions and imprecise standards (such as loss contingencies) is the most difficult to prevent.
Auditor independence
Perhaps the most pressing issue concerning auditors today is the extent to which the value of the audit is undermined by the conflict of interest that arises when the audit firm is not independent from its client. Auditor independence has received much attention recently, and starting in 2001, the SEC has required firms to disclose the audit and non-audit fees they pay their auditors in their proxy statements. And many others have weighed in with their opinions on the need for greater auditor independence.25
Recent research supports the importance of auditor independence for earnings quality. Using the new disclosures required by the SEC, Frankel, Johnson and Nelson study the proxy filings of more than 3000 US firms and measure the ratio of non-audit fees to total fees to assess auditor independence.26 (Examples of non-audit services include financial information systems design, tax preparation/consulting and other advisory services.) They find that, on average, non-audit fees make up approximately 50% of total fees. They also examine the association between the fee ratio and the absolute value of discretionary accruals. They find a significantly positive association, suggesting that the purchase of non-audit services is associated with lower earnings quality (as indicated by larger discretionary accruals). The study shows as well that the stock market response to the new disclosures is decreasing in the unexpected component of the non-audit to total fee ratio. It concludes that investors associate higher non-audit fees with lower quality audits. These findings support the call for greater auditor independence.
Conclusion
While earnings management has been documented in contexts where managers have the incentive to manage earnings, countervailing forces are also at work to reduce earnings management. Research suggests that efforts by regulators and auditors to further maintain high audit quality, to emphasize auditor independence and to strengthen corporate governance will have a positive impact on the quality of financial reporting, and could go a long way in restoring the public's faith in the financial reporting system.
Author's Note: I would like to thank John Friedlan, Carol Marquardt, Claude Lanfranconi, and Peter Bialo for their helpful comments. Any errors and omissions are my own.
Christine I. Wiedman, PhD, CA, is an associate professor of managerial accounting and control and the Natalie MacLean and Andrew Waitman Fellow at the Richard Ivey School of Business, University of Western Ontario
Technical Editor: John Friedlan, PhD, CA, is associate professor at the Schulich School of Business at York University
Footnotes
1. K. Schipper, "Commentary on earnings management," Accounting Horizons, Vol. 15, No. 4, December 1989, 91-102.
2. Although this technique does not separate out the discretionary component of total accruals for the non-discretionary component perfectly, it is still the most commonly used approach. For example, P. Dechow, R. Sloan, and A. Sweeney, "Detecting earnings management," The Accounting Review, Vol. 70, 1995, 193-226, discuss related measurement issues.
3. M. Beneish, "The detection of earnings manipulation", Financial Analysts Journal, Sep/Oct 1999, 24-36.
4. See, for example, P.M. Dechow, R. G. Sloan and A. P. Sweeney, "Causes and consequences of earnings manipulations: An analysis of firms subject to enforcement actions by the SEC," Contemporary Accounting Research, Spring 1996, 1-36.
5. See, for example, S. Ward, "Facing the music," Barron's, Nov 4, 1996; and J. Sandberg, "America Online, posting huge loss, writes off costs of acquiring members", Wall Street Journal, Nov 8, 1996.
6. C. A. Marquardt and C. I. Wiedman, "The effect of earnings management on the value relevance of accounting information," Working Paper, June 2002, New York University and University of Western Ontario.
7. S. H. Teoh, T. J. Wong, G. R. Rao, "Are Accruals during initial public offerings opportunistic?", Review of Accounting Studies, 3, 1998, 175-208.
8. S. H. Teoh, I. Welch and T. J. Wong, "Earnings management and the underperformance of seasoned equity offerings," Journal of Financial Economics, 50, 1998, 63-99.
9. M. Erikson and S. Wang, "Earnings management by acquiring firms in stock for stock mergers", Journal of Accounting and Economics, 27, 1999, 149-176.
10. D. Burghstahler and I. Dichev. 1997. Earnings management to avoid earnings decreases and losses. Journal of Accounting and Economics, 24, 99-126.
11. C. A. Marquardt and C. I. Wiedman, "How are earnings managed? An examination of specific accruals," Working Paper, April 2002, New York University and University of Western Ontario.
12. F. Degeorge, J. Patel and R. Zeckhauser, "Earnings management to exceed thresholds," The Journal of Business, Jan 1999, 1-33.
13. J. L. Payne and S. G. Robb, "Earnings management: the effect of ex ante earnings expectations," Journal of Accounting, Auditing and Finance, Fall 2000.
14. Teoh, Welch and Wong, 1998.
15. M. T. Bradshaw, S. A. Richardson, and R. G. Sloan, "Do analysts and auditors use information in accruals?" Journal of Accounting Research, 39 (1), 2001, 45-74.
16. Joint Committee on Corporate Governance, Beyond compliance: building a governance culture, November 2001.
17. New York Stock Exchange and the National Association of Securities Dealers, Report and recommendations of the Blue Ribbon Committee on improving the effectiveness of corporate audit committees, 1999, Stamford, CT.
18. R. Smith and S. Craig, "Critics question sincerity of corporate reforms calls," The Globe and Mail, June 7, 2002, B7.
19. A. Klein, "Audit committee, board of director characteristics, and earnings management", forthcoming, Journal of Accounting and Economics.
20. Paul Waldie, "Pension funds join forces in bid to influence firms", The Globe and Mail, June 28, 2002, B1 and B24.
21. L. S. McDaniel, R. D. Martin and L. A. Maines, "Evaluating financial reporting quality: the effects of financial expertise versus financial literacy," Working Paper, January 7, 2002, University of North Carolina, Chapel Hill and Indiana University. This paper was presented at the Accounting Review's Quality of Earnings Conference, January 2002.
22. J.R. Francis and J. Krishnan, "Accounting accruals and auditor reporting conservatism," Contemporary Accounting Research, Spring 1999, 135-165.
23. C. L. Becker, M. L. Defond, J. Jiambalvo, and K.R. Subramanyam, "The effect of audit quality on earnings management", Contemporary Accounting Research, Spring 1998, 1-24.
24. M. W. Nelson, J. A Elliott, and R. L. Tarpley, "Evidence from auditors about managers' and auditors' earnings-management decisions," Working Paper, December 2001, Cornell University. This paper was presented at the Accounting Review's Quality of Earnings Conference, January 2002.
25. For example, in June 2002, Henry Paulson, Jr., CEO of Goldman Sachs Group Inc., called for the complete separation of consulting and accounting services at accounting firms.
26. R. M. Frankel, M. F. Johnson and K. K. Nelson, "The relation between auditors' fees for non-audit services and earnings quality, Working Paper, January 2002, MIT, Michigan State University and Stanford University. This paper was presented at the Accounting Review's Quality of Earnings Conference, January 2002. Another related ongoing study finds somewhat conflicting evidence, namely that auditor independence does not appear to influence the auditors, propensity to issue going concern audit opinions for financially distressed firms. See, Mark L. DeFond, K. Raghunandan and K.R. Subramanyam, "Do non-audit service fees impair auditor independence? Evidence from going concern audit opinions.", Working Paper, January 2002, University of Southern California, and Texas A&M International University. |
