外文翻译
原文：
Does Public Ownership of Equity Improve Earnings Quality?
I. INTRODUCTION
The quality of accounting information is influenced by an array of factors, most of which stem from the demand for such information for use in contractual arrangements and from the incentives and opportunities of management to manage the reported numbers. Both the demand for quality accounting information for contractual purposes and management incentives to adjust the reported earnings are likely to be influenced by whether the equity of the company is privately held or publicly traded. In this study, we examine the differential earnings quality of private equity and public equity firms in order to shed light on how public ownership of equity affects the quality of firms’ earnings. Because earnings ‘‘quality’’ has multiple dimensions, in our tests we examine a number of attributes that have been associated by previous research with the notion of earnings quality.
The influence that public or private ownership has on the quality of accounting numbers has been examined in limited contexts by past studies. Beatty et al. (2002), Burgstahler et al. (2006), and Penno and Simon (1986) focus on the difference between public and private firms with respect to one dimension of earnings quality—the extent to which earnings are managed. The association between ownership type and another earnings attribute, conservatism, is examined by Ball and Shivakumar (2005). Because financial data of privately owned firms is generally unavailable, these studies are restricted to regulated industries such as banking and insurance for which financial reports of both public and private companies are filed with industry regulators (e.g., Beatty et al. 2002) or to countries such as the U.K. in which accounting information is available for private companies because they must publicly file financial statements (e.g., Ball and Shivakumar 2005).
The results of these studies on the differential earnings quality of public versus private companies are conflicting. Beatty et al. (2002) find that public firms have a
greater propensity to manage earnings than private firms, whereas Burgstahler et al. (2006) report the opposite. Further, to the extent that conservatism is viewed as an earnings quality trait, this finding by Beatty et al. (2002) is ostensibly in contrast to the finding by Ball and Shivakumar (2005), who use the extent of reporting conservatism to assess earnings quality. The results of these studies, while insightful, cannot easily be generalized. The examination of firms in a regulated industry provides results for a single industry with unique financial reporting issues. Further, the extent to which the results based on samples consisting of European companies generalize to U.S. firms is not clear due to differences in these countries’ reporting regimes.
Our study extends the literature on the effect of ownership type on earnings quality by examining a broader sample of U.S. firms (in non-regulated industries) and by considering several measures of earnings quality. Specifically, we compare the quality of accounting numbers between private and public equity firms along three dimensions: persistence of accruals, estimation error in the accrual process, and prevalence of earnings management. We also compare the degree of conservatism between these two groups of firms.
Our sample of U.S. companies consists of two types of public companies: those with publicly traded equity (hereafter, public equity firms) and those with privately held equity that are considered public companies because they have publicly traded debt (hereafter, private equity firms). Both types of firms are subject to identical SEC reporting and disclosure requirements. Hence, our tests control for many of the factors affecting the comparison of earnings quality across countries, such as legal institutions, tax laws, securities regulations and the extent of their enforcement, as well as reporting and disclosure requirements. We are thus able to identify more precisely how the ownership structure of the company affects its earnings quality. Relying on this unique sample of U.S. companies and on a broader set of earnings quality attributes, our study sheds light on the question of the effect of ownership type on earnings quality. Further, it resolves the question of whether he apparent conflicting results of past studies are due to differences in the examined samples or an
inherent negative association between the different measures employed to assess earnings quality.
Findings indicate that the accounting numbers produced by public equity and private equity firms exhibit different reporting attributes. While public equity firms have a lower quality of accruals in terms of their persistence and estimation error and further exhibit a greater propensity to manage earnings, they also report more conservatively (in terms of timely loss recognition) than their private firm counterparts.
This study is the first to analyze the quality of accounting information generated by firms whose debt, but not equity, is publicly traded, comparing such firms to public equity firms. We further examine the different incentives and opportunities that management of these two types of firms has to affect the reported numbers. By extending the literature on earnings quality and the differential quality between public versus private companies, this study enhances our understanding of how, and the extent to which, management incentives and investor demand for earnings quality impact financial reporting.
The study contributes to the existing research in two main respects. First, we consider a spectrum of attributes related to the concept of earnings quality rather than a single attribute of earnings quality such as earnings management. Second, by examining a unique sample of privately held public companies, the study highlights how the presence of public equity investors affects management’s reporting behavior, controlling for the regulatory environment as well as the disclosure and reporting regimes.
In the next section, the characteristics of the sample firms are described. The hypotheses are developed in Section III, followed by a discussion of the various measures used to assess earnings quality. The sample and data are described in Section V. The results are provided in Section VI. Concluding remarks are provided in the last section.
2. MEASURES OF ACCOUNTING QUALITY
The concept of earnings quality is elusive. The literature on ‘‘earnings quality’’
does not provide a clear definition of that ‘‘quality.’’ It does, however, identify different attributes that are associated with or reflective of earnings quality.
Penman and Zhang (2002, 237), while recognizing the lack of consensus on the definition of earnings quality, define the term to mean that ‘‘reported earnings ... is a good indicator of future earnings.’’ They consider high-quality earnings to be ‘‘sustainable earnings’’ and, correspondingly, deem an accounting system that produces unsustainable earnings as being of poor quality. They show that, in addition to the disruptive effect on earnings sustainability caused by changes in accounting methods and estimates, hidden reserves (such as those created by the use of LIFO or expensing of R&D) reduce the sustainability of earnings by providing more opportunity for earnings management. Richardson et al. (2005) and implicitly Sloan (1996) suggest a related dimension of earnings quality—the reliability of accruals as captured by earnings persistence. Richardson et al. (2005, 438) hypothesize and find that ‘‘less reliable accruals result in lower earnings persistence.’’ Another related measure of earnings quality, the conformity with GAAP (as captured by SEC enforcement actions), is employed by Dechow et al. (1996) and Bradshaw et al. (2001).
Dechow and Dichev (2002) suggest another aspect of earnings quality—the strength of the relation between current accruals and past, present, and future cash flows. Accordingly, they propose a model for expected accruals and interpret the deviation from this ‘‘expected’’ value as the estimation error in accruals, which they use as a measure of earnings quality. This measure is affected by firm characteristics such as the length of the business cycle as well as by earnings management. Ball and Shivakumar (2005, 84) define reporting quality in general terms as ‘‘the usefulness of financial statements to investors, creditors, managers, and all other parties contracting with the firm.’’ They view accounting conservatism in the form of asymmetric timeliness in recognizing losses versus gains as a dimension of earnings quality. However, as discussed later, equating conservative reporting with earnings quality is not universally accepted.
In summary, no single measure of accounting numbers captures all of the
dimensions of earnings quality. Previous studies have identified a number of attributes associated with different aspects of earnings quality such as accrual persistence, estimation errors in the accrual process, and the absence of earnings management. These quality traits as well as the conservatism attribute are discussed in the following paragraphs.
3、Absence of Earnings Management
It is difficult to determine the presence or absence of earnings management, since the series of unmanaged earnings is not observable. However, certain patterns in earnings are indicative of the presence of earnings management. One such pattern is the concentration of earnings numbers just above some earnings threshold (Degeorge et al. 1999). For example, earnings clustered just above zero have been interpreted as reflecting earnings management to avoid reporting a loss. Earnings growth in the current quarter relative to the same quarter the previous year of zero or a slightly positive amount may suggest that the current period’s earnings have been managed to avoid reporting an earnings decline. Similarly, earnings that result in no, or just a small, positive earnings surprise are often viewed as having been managed to meet or just beat analysts’ earnings forecasts.
In line with Burgstahler and Dichev (1997), we identify earnings management cases as those for which the observed relative frequency of earnings that are just above (just below) an earnings threshold exceed (fall below) their theoretical values. For the purpose of this analysis, we divide the distribution of the earnings measure into ‘‘bins,’’ with bin widths determined by the formula suggested by Degeorge et al. (1999). We test for the significance of the difference between the actual and theoretical frequency in a bin based on the procedure proposed by Burgstahler and Dichev (1997) whereby we calculate the standardized differences for the interval just below zero and the interval just above zero. Under the assumption of no earnings management, the expected number of observations in any given interval is equal to the average of the number of observations in the two adjacent intervals. If managers succeed in meeting the threshold, we would expect to find a shift of observations from the bins just below the earnings threshold to the bins just above that threshold.
公共股权能否提高盈余质量？【外文翻译】
