• Selected
• Entire List

Abstract:
This study provides evidence on the Interest Rate Risk (IRR) management activities of commercial banks including their use of derivatives. We find that (i) banks primarily focus on managing interest rate sensitivity of net income rather than the interest rate sensitivity of stock returns, (ii) the level of IRR taken by banks is directly related to liquidity, and inversely related to managerial quality and bank size, (iii) derivative users as a group have lower mean and median exposure than non-users, and (iv) for the majority of users, derivative usage reduces exposure. These findings are inconsistent with the view that derivatives threaten the viability of the banking system.

Abstract:
This study documents evidence on the valuation effects of accounting conservatism in the context of the Feltham-Ohlson [1996] model. We find, consistent with the model's predictions, that goodwill (the difference between market and book value of equity) is a function of abnormal operating earnings and beginning operating assets. Our test results support two further implications of the FO model: (1) the effect of abnormal operating earnings on goodwill increases with operating cash receipts persistence, and (2) the effect of operating assets on goodwill increases with the difference between cash receipts persistence and the depreciation parameter. Generally, the estimated coefficients from the FO model correspond to their implied theoretical values. However, we reject a paired-sample test of equality of medians between the estimated beginning operating asset weights and those implied by our persistence, depreciation, and cost of capital parameters. We interpret our results as supporting the qualitative features and the underlying intuition of the model.

Abstract:
This study provides evidence on the role of accounting conservatism in mitigating bondholder/shareholder conflicts over dividend policy. In particular, we document that firms that face more severe conflicts over dividend policy tend to use more conservative accounting. Furthermore, we also document that there is a tradeoff between conservatism and the cost of debt. Firms that choose more conservative accounting have a lower cost of debt after controlling for other determinants of the cost of debt. Taken together, the evidence is consistent with the notion that accounting conservatism plays an important role in efficient contracting.

Abstract:
This study investigates whether financial analysts correctly weight cash flows, accruals and components of accruals in forecasting future earnings. This examination is in the spirit of Sloan (1996) who documents evidence that investors do not correctly distinguish between the cash flow and accrual components of earnings. We find that analysts do distinguish between accruals and cash flows although they generally underweight the information in both accruals and cash flows. More importantly, we find that analysts do not distinguish between discretionary and non-discretionary accruals even though discretionary accruals are less persistent than non-discretionary accruals. Our findings complement and extend the findings in recent studies on analyst forecast inefficiency with respect to the information in accrual and cash flow components of earnings using alternative research designs [Teoh and Wong (1998), Bradshaw, Richardson and Sloan (2000), Barth and Hutton (2000)].

Analysts are considered to play an important role as information intermediaries in educating investors about the future prospects of firms. They are trained in analyzing financial data and have industry expertise as well as detailed firm-specific knowledge through contacts with managers. Thus, one would expect analysts to correctly incorporate the information in earnings components specifically discretionary versus non-discretionary accruals. Our evidence complements evidence in recent studies that raises questions about the ability of analysts, on-average, to correctly incorporate the information in accruals and cash flows in forecasting future earnings. This in turn implies that other outsiders are also likely to find it difficult to undo earnings management via discretionary accruals and therefore provides a rationale for the existence of earnings management.

Analysts' forecasts; Discretionary accruals; Forecast efficiency

Abstract:
This study documents evidence on the choice of loan loss provisions by bank managers. In particular, we reexamine three hypotheses investigated by prior studies. First, we examine whether the 1990 change in capital adequacy regulations affects the relation between capital and loan loss provisions. We find that the relation is less negative in the new regime consistent with a reduction in incentives to manage regulatory capital via loan loss provisions in the post 1990 capital regime.

Second, we find that there is no significant relation between earnings (before loan loss provisions) and loan loss provisions even in the new regime when the cost of smoothing via loan loss provisions is expected to be lower than in the old regime.

Third, we reexamine the signaling hypothesis which implies a positive relation between loan loss provisions and future earnings changes. We find, contrary to the signaling hypothesis, that there is a negative relation between loan loss provisions and one-year ahead change in earnings. We also investigate whether bank stock returns are positively related to discretionary loan loss provisions as implied by the signaling hypothesis. We find that stock returns are negatively related to discretionary loan loss provisions. We conclude that evidence consistent with signaling documented in prior studies is likely to be specific to the period examined in those studies.

Abstract:
This paper exploits the 1990 change in capital adequacy regulations to construct more powerful tests of capital and earnings management effects on bank loan loss provisions. We find strong support for the hypothesis that loan loss provisions are used for capital management. We do not find evidence of earnings management via loan loss provisions. We also document the reasons for the conflicting results on these effects observed in prior studies. Additionally, we find that loan loss provisions are negatively related to both future earnings changes and contemporaneous stock returns contrary to the signaling results documented in prior work.

Abstract:
We provide evidence on how investor valuation of derivative financial instruments differs depending upon whether the fair value of these instruments is recognized or disclosed. Expanded disclosures and accounting practices prior to SFAS No. 133 and mandatory recognition of derivative fair values after SFAS No. 133 provide a natural setting for comparing the valuation implications of recognized and disclosed derivative fair value information. This unique setting mitigates many of the research design problems with recognition versus disclosure studies. Using a sample of banks that simultaneously hold recognized and disclosed derivatives prior to SFAS No. 133, we find that the valuation coefficients on recognized derivatives are significant whereas the valuation coefficients on disclosed derivatives are not significant. Further, using a sample of banks that have only disclosed derivatives prior to SFAS No. 133 which are recognized after SFAS No.133, we find that while the valuation coefficients on disclosed derivatives are not significant, the valuation coefficients on recognized derivatives are significant. These results are consistent with the view that recognition and disclosure are not substitutes. Our findings suggest that SFAS No. 133 has increased the transparency of derivative financial instruments.

Value-relevance, Recognition versus Disclosure, Fair Value, Derivative Financial Instruments

Abstract:
Fudenberg and Tirole (1995) analytically demonstrate that income smoothing can arise in equilibrium if managers are concerned about job security. Consistent with their model, DeFond and Park (1997) show that managers smooth income in consideration of both current and future relative performance. We provide more direct evidence on whether job security results in income smoothing. More specifically, we hypothesize that the extent of income smoothing will vary directly with managers' job security concerns (proxied by the degree of competition in firms' product markets, product durability, and capital intensity). Our results are consistent with our predictions and add to the literature on income smoothing.

Income smoothing, Earnings management, Job security

Abstract:
Easterwood and Nutt (1999) show that analysts under-react to bad news in past earnings changes (or forecast errors) but over-react to good news in past earnings (or forecast errors) consistent with analysts exhibiting systematic optimism. We find that their results are sensitive to the cutoff used to define outliers.

More specifically, when we define outliers as observations with the absolute value of earnings-related variables exceeding 100% of stock price, we obtain results similar to those reported in their study. However, when we define (and exclude) outliers as observations with the absolute value of earnings-related variables exceeding 20% of stock price, we find evidence of under-reaction to the information in past earnings and forecast errors. However, the under-reaction to bad news is not significantly greater (in magnitude) than the average under-reaction. Furthermore, we find some evidence of overreaction to good news in past earnings changes but no evidence of overreaction to the information in past forecast errors.

With a 5% cutoff, there is no evidence of over-reaction to good news in past earnings changes or forecast errors. Furthermore, there is no evidence of over or under-reaction to the information in past returns. The last result suggests that the evidence of under-reaction to the information in past returns observed in previous studies may have been overstated. Overall, our results cast doubt on the hypothesis that analysts exhibit systematic optimism with respect to past information.

Analysts' forecasts, systematic optimism, under-reaction

Abstract:
Using three different measures of conservatism, we document that (i) the percentage of inside directors is negatively related to conservatism, and (ii) the percentage of outside directors' shareholdings is positively related to conservatism. Our results hold after controlling for industry, firm size, leverage, growth opportunities, institutional ownership, inside director ownership, and unobservable firm characteristics that are stable over time. Overall, the evidence is consistent with accounting conservatism assisting directors in reducing agency costs of firms.

Accounting Conservatism, Board Independence, Outside Directors, Corporate Governance, Agency Costs

Abstract:
Using three different measures of conservatism, we document that (i) the percentage of inside directors is negatively related to conservatism, and (ii) the percentage of outside directors' shareholdings is positively related to conservatism. Our results hold after controlling for industry, firm size, leverage, growth opportunities, institutional ownership, inside director ownership, and unobservable firm characteristics that are stable over time. Overall, the evidence is consistent with accounting conservatism assisting directors in reducing agency costs of firms.

Accounting Conservatism, Board Independence, Outside Directors, Corporate Governance, Agency Costs

Abstract:
Extreme accruals are commonly viewed as tainted by earnings management, contributing to lower quality earnings. We refer to this presumption as the earnings management/quality hypothesis. We directly examine three aspects of the presumed relation between the level of accruals and earnings management/quality: (i) the implications of earnings management for earnings persistence, (ii) the implications of earnings management for accrual mispricing, and (iii) the likelihood of greater opportunistic management to achieve earnings targets.

We find (i) no evidence that extreme income-increasing accruals have lower persistence for year-ahead earnings, (ii) no evidence that investors systematically overweight extreme accruals, and (iii) no evidence that extreme accrual firms fall into the interval of firms just avoiding losses or earnings decreases more than other firms. We do find evidence that extreme income-decreasing accruals have lower persistence, but the evidence suggests this results from poor economic performance rather than accrual manipulation. We also document that investors consistently underestimate the persistence of cash flows for all firms, and that predictable year-ahead abnormal returns to extreme total or abnormal accrual portfolios documented in prior work appear to be driven by cash flow mispricing rather than accrual mispricing. Overall, our findings cast doubt on using extreme accruals to proxy for earnings management and/or low earnings quality.

Extreme accruals, earnings management, earnings quality, persistence

Abstract:
We provide evidence on the relation between office-level client importance and abnormal accruals in a pre-SOX period and a post-SOX period. We find that in the pre-SOX period the relation is not significant for the overall sample. However, it is significantly positive for the sub-sample of firms with relatively weak governance mechanisms. More importantly, we find that for this sub-sample the positive relation between client importance and abnormal accruals remains statistically significant even in the post SOX period. Our results suggest that (i) auditor independence was not widely compromised in pre-SOX periods as presumed by the proponents of SOX, (ii) strong governance mechanisms mitigate the potentially adverse effects of client importance on auditor independence, and (iii) SOX has not been successful in mitigating the adverse effects of client importance on auditor independence for firms with weak governance.

Sarbanes-Oxley, Auditor Independence, Corporate Governance, Accounting Accruals

Abstract:
Watts (2003) and others argue that conservatism helps in corporate governance (specifically in monitoring firms' investment policies). We hypothesize that if conservatism reduces managers' ex ante incentives to take on negative NPV projects and improves the ex post monitoring of investments, firms with more conservative accounting ought to have higher future profitability and lower likelihood (and magnitude) of future special items charges. We find that firms with more conservative accounting have (i) higher future cash flows and gross margins, and (ii) lower likelihood and magnitude of special items charges than firms with less conservative accounting. Our results hold after controlling for industry, firm size, leverage, growth opportunities, prior special items charges, and stock returns. These findings are (i) consistent with conservatism mitigating agency problems associated with managers' investment decisions as predicted by Watts (2003) and Ball and Shivakumar (2005), and (ii) inconsistent with standard setters' view that conservatism is not a desirable characteristic in financial reporting.

accounting conservatism, corporate governance, agency costs

Abstract:
This study provides evidence regarding the effects of online trading on stock market reactions to quarterly earnings announcements. We test for differences in stock price and volume reactions to quarterly earnings announcements between a period with a significant amount of online trading (1996-1999) and a period without online trading (1992-1995). We conjecture that online trading has increased the proportion of naive investors in the market. Based on noisy rational expectations models of trade, we predict that this will result in larger stock price and trading volume reactions to earnings announcements. We find strong evidence in support of these predictions. The stock price results suggest that the advent of online trading has decreased average prior precision and the trading volume results suggest that online trading has increased differential belief revisions around earnings announcements. An analysis of the relation between volume reactions and price reactions in both periods suggests that the increase in differential belief revisions is primarily due to an increase in the differential interpretation of earnings announcements in the online trading period. Our findings are relevant for assessing the validity of concerns about online trading expressed by regulators and the validity of theoretical models of trade with asymmetrically informed investors.

Online trading; Noisy rational expectations; Differential interpretations; Market reaction

Abstract:
This paper documents evidence on the efficacy of maturity gap disclosures and interest rate risk policy disclosures of commercial bank holding companies in indicating their interest rate risk exposures. Using data from the Federal Reserve Y-9 reports over 1991-1996, we find a significant relation between maturity gap and future change in net interest income indicating that gap data are useful in assessing the loss potential of banks' interest rate risk positions. This is contrary to the claim of some banks that gap data are not useful indicators of interest rate risk. The finding mitigates concerns about the usefulness of the SEC (1997) market risk disclosure requirements and is pertinent to the SEC's scheduled review of these requirements.

We also find that gap is negatively related to interest rate sensitivity of banks contrary to the nominal contracting hypothesis though this relation is significant only in three out of the six years examined. Taken together, the evidence suggests that inferences about market risks based on interest rate sensitivity alone may be misleading.

Finally, we find no support for the hypothesis that qualitative policy disclosures help distinguish hedgers from speculators.

Abstract:
Prior studies find that (i) investors over-weight (or over-estimate the persistence of) past accruals and under-weight past cash flows, and (ii) analysts over-weight past accruals. We study financial analysts' and investors' assessments of the persistence of accruals and cash flows after explicitly controlling for the negative correlation between accruals and cash flows. We find that both analysts and investors under-weight past cash flows but over-weight past accruals. The accrual and cash flow effects are distinct from each other in the sense that when we control for one effect, the other effect continues to be significant. However, the cash flow effect on both forecast errors and future stock returns is much stronger, in terms of magnitude, than the accrual effect. The strength of the cash flow effect has not been generally recognized in the literature.

market efficiency, analyst forecast efficiency, accruals, cash flows

Abstract:
We contribute to the literature on Regulation Fair Disclosure (FD) in three ways. First, we provide evidence on whether FD has achieved its intended effect of leveling the information playing field by examining whether differences across investors' information quality prior to earnings announcements have declined after the pronouncement of the regulation. We find strong evidence of a decline in earnings announcement period trading volume attributable to differential prior precision after FD consistent with a more level playing field. Second, we re-examine whether FD has resulted in firms reducing or chilling their information flows (disclosures) to investors. Contrary to prior work, we find that there is evidence of an overall reduction or chill in information flows after FD relative to a "cleaner" pre-FD period than the pre-FD period used in other studies. Third, we document that while the leveling effect of FD is relatively wide-spread, the chill effect is driven by (i) relatively smaller, high technology firms and (ii) relatively larger firms with high book-to-market ratios. We interpret the latter result as evidence that firms with relatively high costs of public disclosure chose to eliminate the disclosure altogether rather than broadening access to the disclosure.

Disclosure, fair disclosure regulation, trading volume reaction, regulation FD

Abstract:
We provide evidence on the impact of SFAS No. 142 on the reliability of goodwill write-offs and goodwill balances by examining the association of goodwill write-offs with stock returns and the association of goodwill balances with stock prices in a period after SFAS No. 142 relative to a period before SFAS No. 142. We find that (i) goodwill write-offs (or impairments) and goodwill balances are more strongly associated with stock returns and stock prices respectively in the post SFAS 142 period relative to the pre-SFAS 142 period and (ii) in the post SFAS No. 142 period, goodwill write-offs and goodwill balances are more strongly associated with stock returns and stock prices respectively for firms with a high number of segments relative to firms with few segments. Our tests suggest that contrary to the relatively widespread concerns of critics about the reliability of goodwill impairments under SFAS 142, the standard has had a favorable impact on the reliability of goodwill write-offs and goodwill balances.

Goodwill, Goodwill Impairment, SFAS No. 142, Reliability

Abstract:
Prior studies document that firms using a Big 4 auditor have a lower cost of capital than other firms. We extend this literature by examining whether using an industry specialist auditor reduces cost of capital for clients of Big 4 audit firms. We document that firms that use Big 4 auditors that are industry specialists have significantly lower cost of both equity and debt than firms that use non-specialist Big 4 auditors. We further investigate whether the benefits of using an industry specialist auditor vary with the strength of alternative monitoring mechanisms. We show that using an industry specialist auditor is especially important when alternative monitoring mechanisms, such as boards of directors or institutional shareholders, are relatively weak. In other words, the benefits of using an industry specialist auditor dissipate when alternative monitoring mechanisms are strong. This evidence suggests some degree of substitutability between audit quality and alternative monitoring mechanisms.

Audit quality, Industry specialization, Financial reporting credibility, Cost of capital

Abstract:
We provide evidence on the effects of the mandatory adoption of IFRS on smoothness, conservatism, and timeliness of accounting earnings for a sample of over 1,600 firms from 21 countries that adopted IFRS in 2005 relative to a benchmark sample of firms from 17 other countries that did not adopt IFRS. First, we find that in the pre-adoption period IFRS firms exhibited less income smoothing, more conservative accruals, and more timely loss recognition than benchmark firms. Second, we find that income smoothing increased significantly, conservatism decreased significantly, and timeliness of good news increased significantly in the post-adoption period for IFRS firms relative to benchmark firms. Overall, our results suggest that mandatory IFRS adoption has not unambiguously improved accounting quality.

International Financial Reporting Standards, Income Smoothing, Earnings Timeliness, Conservatism

Abstract:
The Sarbanes-Oxley Act (SOX) was intended to protect investors by improving the accuracy and reliability of corporate disclosures. However, critics have argued that the costs of SOX far outweigh its intended benefits. Prior studies based on stock-price reactions to SOX-related events document mixed evidence on the expected impact of SOX. In contrast, we provide evidence on the impact of SOX on operating profitability by examining the net realized costs of SOX. We find that average cash flows decline by 1.3 percent of total assets after SOX. These costs are more significant for smaller firms, for more complex firms, and for firms with lower growth opportunities. Annually, these costs range from $6 million for smaller firms to $39 million for larger firms. Further, we document that net SOX-related costs are not limited to one-time expenses associated with internal control design and implementation. In aggregate, for the 1,428 firms in our sample, these costs exceed $19 billion per year or about $75 billion over the four-year post-SOX study period. Profitability is lower for up to four years post-SOX. To our knowledge, ours are the first estimates of the realized net costs imposed by SOX.

Sarbanes Oxley Act, Operating profitability, Net realized costs, Cost of compliance, Small firms

Abstract:
This study examines whether differential interpretation of earnings announcements is affected by earnings and firm characteristics. We find that Kandel and Pearson's (1995) forecast measures of differential interpretation are: 1) negatively related to earnings predictability, firm size, and price-to-book ratio, and 2) positively related to earnings surprise, negative earnings, and analyst coverage. This evidence suggests that differential interpretation of earnings announcements is decreasing in the quality of the earnings, the quality of pre-announcement disclosure, and the cost of analyzing a firm's value.

differential interpretation, earnings precision, prior information

Abstract:
This study provides empirical evidence on factors that drive differential interpretation of earnings announcements. We document that Kandel and Pearson's forecast measures of differential interpretation are decreasing in proxies for earnings quality and pre-announcement information quality. This evidence yields new and useful insights regarding which earnings announcements are less likely to generate newfound disagreement among analysts and investors. Recent research suggests that investor disagreement can increase investment risk, increase the cost of capital, and cause stock prices to deviate from fundamental value. Therefore, our results support prior intuition that increasing the quality of earnings and pre-announcement information can improve the efficiency of capital markets.

Abstract:
We provide evidence on the determinants of the choice of an industry specialist auditor and the effect of this choice on cost of equity for a sample of firms that are audited by Big N auditors. We find that firms with more severe conflicts of interest between managers and shareholders are more likely to use an industry specialist auditor. Furthermore, we document that firms that use an industry specialist auditor have a significantly lower cost of equity after controlling for firms' endogenous decision to use an industry specialist auditor.

Big N audits, ex ante cost of equity capital, financial reporting credibility, industry specialization

Abstract:
We document that Regulation Fair Disclosure has reduced differences in information quality between investors prior to quarterly earnings announcements consistent with the intent of the regulation. This reduction is driven by small firms and high technology firms, rather than the large firms targeted by the SEC, which suggests that selective disclosure among large firms may have been much more limited than what was presumed by proponents of FD. In addition, we document that FD has decreased the average information quality of investors in small and high technology firms in the period prior to an earnings announcement while having no lasting effect on other firms. Taken together these two results suggest that, for small and high technology firms, FD succeeded in eliminating selective disclosure but also lowered the average quality of information available about these firms.

Fair Disclosure, FD, SEC regulation, trading volume

Abstract:
This study investigates whether analysts correctly distinguish between discretionary and nondiscretionary accruals in forecasting eamings, and finds that they do not distinguish between the two types of accruals. The findings of this study complement and extend the findings of recent studies on analyst forecast inefficiency with respect to the information in accruals using altemative research designs.

analyst forecast inefficiency, operating cash flows, accruals

Abstract:
This study provides evidence on the effects of online trading on stock price and trading volume reactions to quarterly earnings announcements. We test for differences in stock price and volume reactions to quarterly earnings announcements between a period with a significant amount of online trading (1996-1999) and a period without online trading (1992-1995). We conjecture that online trading has increased the proportion of naive investors in the market. We predict that this will result in (i) a decrease in the average precision of investor information prior to earnings announcements implying higher ERCs, (ii) an increase in differential interpretation of earnings leading to higher trading volume reactions that are unrelated to price change, and (iii) a decrease in differential prior precision leading to a decrease in the association between trading volume and absolute price change. We find evidence consistent with all three predictions. Our findings are relevant for assessing the validity of concerns about online trading expressed by regulators and the validity of theoretical models of trade with asymmetrically informed investors.

Differential interpretations, Differential prior precision, Online trading, Stock price reaction, Trading volume reaction

Abstract:
Using both a market-based and an accrual-based measure of conservatism, we find that firms facing more severe conflicts over dividend policy tend to use more conservative accounting. Furthermore, we document that accounting conservatism is associated with a lower cost of debt after controlling for other determinants of firms' debt costs. Our collective evidence is consistent with the notion that accounting conservatism plays an important role in mitigating bondholder-shareholder conflicts over dividend policy, and in reducing firms' debt costs.

accounting conservatism, dividend policy conflicts, cost of debt

Abstract:
This paper attempts to provide empirical evidence on some of the cross-sectional implications of the Feltham-Ohlson [1995] valuation model. Overall, the evidence supports the idea that stock prices can be viewed as a combination of stock (i.e., book equity and operating assets) and flow (i.e., earnings) variables. However, (i) evidence on the role of book value of equity as articulated in the model is mixed and (ii) evidence on the positive relation between the weight on operating assets and the parameter relating operating assets to future abnormal earnings predicted by the model is weak.

Abstract:
This study investigates how discretionary accruals affect earnings valuation (or price-earnings) multiples to provide indirect evidence on whether managers use their discretion over accruals to communicate their private information or to obtain private benefits. I find that earnings multiples are significantly higher when earnings contain large income increasing discretionary accruals than when earnings contain small discretionary accruals. Furthermore, in periods of high performance earnings multiplies are significantly lower when earnings contain large income decreasing discretionary accruals than when earnings contain small discretionary accruals. These results are consistent with the communication hypothesis. The effect of large income decreasing discretionary accruals in periods of moderate performance is mixed and in periods of low performance is not significant. The study (i) extends the literature on earnings multiples, (ii) suggests that effects of discretionary accruals depend on the context in which they are reported, and (iii) suggests that reduction of managerial discretion via accounting standards is costly because it eliminates the opportunity to communicate.