اصلاح شکایت های قانونی، تشخیص حسابداری و هزینه حقوق صاحبان سهام

In this paper, we isolate a context – the 1995 Public Securities Litigation Reform Act – where information risk (accruals quality) is likely to change, and investigate whether the increase in accounting discretion associated with litigation reform is viewed by investors as basically opportunistic (i.e., as distorting reported earnings) or as improving the ability of reported earnings to reflect economic value. We measure accounting discretion using both positive (i.e., income-increasing) as well as absolute performance-adjusted abnormal accruals. Our analysis focuses on a constant sample of firms over a 10-year (1992–2001) period, and is structured in two stages. In the first-stage, we utilize an instrumental variable technique that isolates the increase in accounting discretion associated with the 1995 Act. In the second-stage, we relate the predicted increase in accounting discretion associated with litigation reform – obtained from the first-stage regression – to the ex ante equity risk premium for Big N audit clients. Our results suggest that the increase in accounting discretion associated with the 1995 Act was viewed by investors as basically opportunistic. Further, the exogenous nature of the 1995 Act suggests that the observed increase (and pricing) of accounting discretion is related to litigation reform rather than some omitted firm-specific operating characteristic. Overall, our findings suggest that litigation reform affects firm value through managers’ exercise of accounting discretion and cost of equity capital channels.

In this study, we isolate a context – the 1995 Private Securities Litigation Reform Act – where information risk (accruals quality) is likely to change. Specifically, the 1995 Act sought to amend what was perceived to be an excessively litigious environment, by increasing restrictions on private litigation for securities frauds.1 Other things being equal, a decline in legal exposure (following litigation reform) may be expected to create a more conducive environment for managers to exercise accounting discretion. Consistent with this argument, Lee and Mande (2003) document an increase in accounting discretion by Big N audit clients in the three years subsequent to the 1995 Act.2 To our knowledge, there is little or no prior research on whether the accounting discretion implications of the 1995 Act are reflected in the cost of equity capital. In this paper, we use estimates of the ex ante cost of equity capital to investigate whether the 1995 Act-related increase in accounting discretion is priced by investors. Such a finding would indicate that litigation reform can influence firm value through managers’ use of accounting discretion and cost of equity capital channels.3 In their study, Lee and Mande (2003) implicitly assume that accounting discretion reflects managers’ opportunistic behavior (“earnings management”), is intended to mislead investors, and that it implies lower earnings quality. However, consistent with Dechow and Skinner, 2000 and Subramanyam, 1996, and Tucker and Zarowin (2006), one could argue that discretion (judgment) is essential to the practice of accrual accounting, and that “earnings management” may merely reflect the exercise of accounting discretion intended to enhance the informativeness of reported earnings by communicating private information. Moreover, to the extent that the manager’s decision to limit accounting discretion is dominated by concerns about legal liability, the drop in the incentive to curtail accounting discretion (following the 1995 Act) may actually enhance the manager’s ability to utilize accounting reports to communicate effectively with investors. In any event, it is an empirical question whether investors view the increase in accounting discretion associated with the 1995 Act as basically opportunistic (i.e., intended to distort earnings) or as improving the ability of reported earnings to reflect economic value. Prior research on the stock price consequences of the 1995 Act has largely focused on investor reactions at the time of various legislative events leading up to the Act, with mixed results. 4 For example, Spiess and Tkac (1997) and Johnson et al. (2000) report that the stock price reaction to the various legislative event-days preceding the 1995 Act was favorable, indicating that the anticipated positive effects of the Act (reduced frivolous litigation) outweighed the anticipated negative effects (potential inability to bring meritorious lawsuits or increased susceptibility to earnings manipulations). However, Ali and Kallapur (2001) report that shareholders in four high litigation risk industries reacted negatively to the various event-days leading up to the Act. By contrast, our study goes beyond the short event windows preceding the passage of the 1995 Litigation Reform Act, by examining a constant sample of Big N audit clients over a longer time frame (1992–2001) that incorporates years both before and after the Act, and focuses directly on the ex ante equity risk premium as our research metric.5 As suggested by Francis et al. (2004), the firm-specific ex ante cost of equity capital represents a summary indicator of investors’ resource allocation decisions. Consistent with Gebhardt et al. (2001) and Dhaliwal et al. (2006), we utilize the ex ante equity risk premium, i.e., the excess of the firm-specific ex ante cost of equity capital over the risk free interest rate, as our research metric to capture investor pricing. Consistent with Bhattacharya et al. (2003), several assumptions are implicit in our study. First, we assume that although investors in an efficient market can rationally anticipate the exercise of accounting discretion, they cannot “see through” it in the sense that they cannot undo the effects of earnings management to arrive at what Bhattacharya et al. (2003) refer to as the “unobservable economic earnings” number. Second, we assume that the information asymmetry created by the opportunistic use of accounting discretion is not completely resolved through other communication mechanisms such as disclosures. Third, we assume that the information risk caused by the opportunistic exercise of accounting discretion is an important factor relative to other factors that affect the equity markets, and is therefore priced.6 Finally, we assume that the exercise of accounting discretion that enhances the informativeness of reported earnings by communicating private information is also priced. As noted by Bhattacharya et al. (2003), none of these assumptions may hold. Specifically, Core et al. (2008) point out that there is no well-accepted theory to suggest that accounting information quality is not diversifiable, and question whether information risk (accruals quality) is a priced risk factor. Still, they (pp. 20–21) go on to suggest that the notion that information quality matters for the capital markets is intuitively appealing, and that (consistent with Lambert et al. 2007) the consequences of information quality may be manifested in other risk factors (such as beta) even if itself is not a distinct (separate) risk factor. Thus, whether or not the increase in accounting discretion associated with the 1995 Act is priced by investors remains an open empirical question. Bhattacharya et al. (2003, p. 650) indicate that a limitation of extant studies on earnings attributes and investor pricing is that expected earnings management is basically unobservable. In our study, we utilize the Lee and Mande (2003) model to estimate the expected level of accounting discretion. Specifically, we utilize an instrumental variable technique similar to Ittner et al. (2003) to estimate the increase in accounting discretion associated with the 1995 Act. Essentially, in our first-stage (“preparer perspective”) regression model, we parse out actual accounting discretion into three components attributable to the firm-specific control variables in the model, the 1995 Act itself (an exogenous event), and statistical noise. Then, in a second-stage (“investor perspective”) regression model, we relate the predicted increase in accounting discretion attributable to the 1995 Act (obtained from the first-stage regression) to the ex ante equity risk premium to test whether investors price the increase in accounting discretion associated with litigation reform. Our instrumental variable technique provides a refined test of the pricing of the increase in accounting discretion associated with the 1995 Act, thus potentially allowing us to provide insight into the question of whether investors take into account the effect of litigation reform on accounting discretion in setting equity discount rates. Thus, in the first-stage (“preparer perspective”) regression we examine the association between the 1995 Act and accounting discretion, where accounting discretion is measured using both positive (i.e., income-increasing) as well as absolute performance-adjusted abnormal accruals. Then, in the second-stage (“investor perspective”) regression we examine whether the predicted increase in accounting discretion associated with the 1995 Act (obtained from the first-stage regression) is priced by investors, i.e., we examine whether investors price the increase in accounting discretion associated with litigation reform. Consistent with previous research (Lee and Mande 2003), our first-stage regression results suggest that the 1995 Act was associated with an increase in accounting discretion. Also, our second-stage regression results suggest that the predicted increase in accounting discretion associated with the 1995 Act (obtained from the first-stage model) is priced by investors, i.e., the increase in accounting discretion associated with the 1995 Act increases the ex ante equity risk premium for firms. Overall, our study contributes to the literature on litigation reform, accounting discretion, and investor pricing in several ways. First, Subramanyam (1996) utilizes stock returns as his research metric and finds that investors attach value to discretionary (abnormal) accruals. Specifically, his findings suggest that managers utilize accounting discretion to improve the ability of earnings to reflect the economic value of the firm, i.e., managers utilize accounting choices to communicate value-relevant private information to outsiders based on an objective assessment that the firm’s stock is undervalued. More recently, Tucker and Zarowin (2006) suggest that the exercise of accounting discretion to smooth earnings improves the informativeness of current earnings about future earnings and cash flows. As noted previously, to the extent that a manager’s decision to limit accounting discretion is driven by concerns about legal liability, the drop in the incentive to curtail accounting discretion (following litigation reform) may increase the manager’s ability to utilize accounting reports to communicate effectively with investors. Hence, it is an empirical question whether the increase in accounting discretion associated with the 1995 Act is viewed by investors as improving the informativeness of reported earnings or as distorting firm performance. Second, the potential benefit in focusing on the 1995 Litigation Reform Act is that it sidesteps the endogeneity issues that can arise from using firm-level measures of accounting discretion. Put differently, firm-level measures of accounting discretion are subject to the criticism that an observed increase may be endogenous, i.e., attributable to changes in the firm’s operating characteristics. However, the 1995 Litigation Reform Act is an exogenous legislative event that is unlikely to be related to firms’ operating characteristics. Thus, given the exogenous nature of the 1995 Act, our evidence suggests that litigation reform is associated with an increase in accounting discretion, and that it is this increase (rather than some omitted firm operating characteristic) that is being priced by investors. Third, litigation reform is a topic of ongoing interest. For example, following the recent demise of Arthur Andersen and the rise in auditor concentration, it has been suggested that the loss of yet another large international accounting firm (due, for example, to catastrophic legal claims) would have severe repercussions for investor confidence in global equity markets (American Assembly, 2003, American Assembly, 2005, Government Accountability Office (GAO), 2003a and US Chamber of Commerce, 2006). Consequently, in the UK (and more broadly in the European Union), it has been suggested that governments allow auditors to reach agreements with their audit clients (with shareholder approval) on liability caps (Buck 2006). In the US, the American Assembly (2003), the US Chamber of Commerce (2006), and the Committee on Capital Markets Regulation (2006) have suggested that the PCAOB and the SEC be allowed to evolve arrangements whereby auditors with satisfactory quality control would be provided a measure of protection from legal liability. Our findings suggest that placing a cap on liability could be associated with an increase in accounting discretion that is likely to be viewed by investors as distorting firm performance rather than as improving the informativeness of reported earnings. Finally, the broader question of whether investors are cognizant of the valuation implications of accounting discretion is a topic of continuing interest to regulators. Although academics have “a natural tendency” to assume investor rationality and that information is correctly priced (Dechow and Skinner 2000, p. 245), investor pricing remains an unanswered empirical question. Prior research suggests that managers can manipulate reported earnings and influence stock prices to the extent that investors are unable to detect opportunistic reporting. In a recent survey (Graham et al. 2005), executives admit to smoothing earnings as a means of lowering the firm’s riskiness as perceived by investors. Along the same lines, Fields et al. (2001) suggest that the fact that rational managers engage in earnings management suggests that these executives believe that information markets are not perfect, i.e., managers believe that users of accounting information are either unable or unwilling to unravel the effects of earnings management. Other research (Bartov et al., 2002 and Skinner and Sloan, 2002) suggests that firms that meet or beat earnings expectations are rewarded with higher equity values, while firms that are unable to meet earnings expectations are severely punished by the stock market.7 From the perspective of regulators (such as the SEC) whose stated goal is investor protection, the pricing of accounting discretion is important since it suggests that investors are making informed decisions and that stock prices are informative. Consistent with this latter view, Levitt (2000) indicates that investors’ perception of earnings quality is critical to maintaining public trust and confidence in companies’ reported earnings and in maintaining the richness and depth of the US financial markets. Thus, our study is intended to inform policy and allow regulators a more complete understanding of the influence of litigation reform on firm value through the exercise of accounting discretion, perceived earnings quality, and cost of equity capital channels. The remainder of this paper is organized as follows: Section 2 describes our hypotheses. Section 3 discusses our methodology and data. The empirical findings are reported in Section 4, and Section 5 concludes the paper.

In this paper, we isolate a context – the 1995 Litigation Reform Act – where information risk (accruals quality) is likely to change. The decline in legal exposure (following litigation reform) may be expected to create a more favorable environment for managers to exercise accounting discretion and enhance their ability to utilize accounting reports to communicate effectively with investors. Alternatively, managers may utilize accounting discretion to misrepresent earnings and lower the perceived riskiness of the firm. Our study investigates whether the increase in accounting discretion associated with the 1995 Act is priced by investors and reflected in the firm-specific ex ante equity risk premium. Our findings suggest that the increase in accounting discretion associated with the 1995 Act is priced and is associated with an increase in the firm-specific equity risk premium. The evidence is similar for both income-increasing as well as absolute performance-adjusted abnormal accruals. In the context of our study, the exogenous nature of the 1995 Litigation Reform Act (i.e., a legislative event that is unlikely to be endogenously driven by firm-specific conditions) suggests that litigation reform is associated with an increase in accounting discretion, and that it is this increase (rather than some omitted firm-specific operating characteristic) that is being priced by investors. Nonetheless, since the 1995 Act occurred at the same time for all companies, we cannot rule out the possibility that some other economic event – unrelated to the Act – may be driving our findings. As noted previously, prior research (e.g., Cheng et al., 2009 and Leung and Srinidhi, 2006) suggests that the 1995 Act had a positive effect in terms of additional high quality disclosures and more effective communication with investors. Still, the 1995 Act represents a complex piece of legislation, and it is not unreasonable to suggest that the Act had both negative as well as positive consequences for investors.24 Overall, our study touches on an important problem and provides a more complete understanding of the effects of the 1995 Act by documenting an unintended negative consequence of the Act, i.e., a decrease in firm value through managers’ exercise of accounting discretion, perceived accruals quality, and cost of equity capital channels.