Knowledge that Transforms

This study identifies conditions under which the audit risk model does, and does not, describe audit-planning (investment and pricing) decisions. In an experiment, audit partners and managers examined one of two cases where a material misstatement—error or irregularity—was discovered. The auditors assessed the elements of the audit risk model, assessed business risk and provided recommendations for the audit investment and fee. When the likelihood of an error was high, the audit risk model dominated business risk in the explanation of the audit investment, and the fee did not contain a risk premium. When the likelihood of an irregularity was high, business risk dominated the audit risk model in the explanation of the audit investment, and the fee contained a risk premium. These results suggest that the ability of the audit risk model to describe auditor behavior and the inclination of auditors to charge a risk premium depend upon the nature of the risks present in the audit. In the presence of errors, the audit risk model adequately described audit-planning decisions; in the presence of irregularities it did not.

We hypothesize that, in their annual accounting reports, propertycasualty insurers allocate premiums from multistate policies to reduce total state taxes. To test this prediction, we exploit the industry's unique state tax disclosures. We examine firm-level data, collected from the publicly available, statutory reports filed with each state government. Reported premiums at the insurer-state level, scaled by incurred losses, are regressed on state tax measures. Consistent with tax-motivated income shifting, we find the premiumloss ratio is decreasing in state tax rates. The negative relation is greatest for insurers specializing in multistate lines of business.

Prior studies of the strategic interaction between taxpayers and the tax authority have focused on reported net taxable income and on audit policies designed to discover potential misstatement of that single item. This paper extends the literature by modeling taxpayer compliance behavior and tax authority audit strategies within the context of a multidimensional report of taxable income. Specifically, the study analyzes the impact of component reporting requirements on taxpayer incentives to misstate their tax liability. It also allows the tax authority to tailor its audit policy to consider all tax return information. In particular, the model permits the tax authority to audit return components sequentially: the investigation of a second component is conditional on the results of the first component's audit. The study finds that partitioning taxable income into a multi-component report reduces overall tax evasion and increases tax authority net revenue collections relative to a singlereport model of net taxable income. However, the impact on predicted evasion is not uniform across taxpayers. While some taxpayers reduce evasion, others with multiple opportunities to evade are more likely to do so when faced with multi-component reporting requirements.

Residual income (RI) valuation is a method of estimating firm value based on expected future accounting numbers. This study documents the necessity of using linear information models (LIMs) of the time series of accounting numbers in valuation. I find that recent studies that make ad hoc modifications to the LIMs contain internal inconsistencies and violate the no arbitrage assumption. I outline a method for modifying the LIMs while preserving internal consistency. I also find that when estimated as a time series, the LIMs of Ohlson (1995), and Feltham and Ohlson (1995) provide value estimates no better than book value alone. By comparing the implied price coefficients to coefficients from a price level regression, I find that the models imply inefficient weightings on the accounting numbers. Furthermore, the median conservatism parameter of Feltham and Ohlson (1995) is significantly negative, contrary to the model's prediction, for even the most conservative firms. To explain these failures, I estimate a LIM from a more carefully modeled accounting system that provides two parameters of conservatism (the income parameter and the book value parameter). However, this model also fails to capture the true stochastic relationship among accounting variables. More complex models tend to provide noisier estimates of firm value than more parsimonious models.

Accounting researchers (and potentially others) generally select rather simple, lower-order, time-series models to develop proxies for earnings persistence. However, measures of persistence produced by such models are not related to characteristics of the firm's economic environment that are expected to influence earnings persistence. Using a sample of 162 calendar year-end New York Stock Exchange firms, we document the cross-sectional relations between a set of relatively constant, firm-specific, economic characteristics that are theoretical determinants of persistence and measures of earnings persistence derived from both lower-order and higher-order Autoregressive, Integrated, Moving-Average (ARIMA) models. When lower-order ARIMA models are used to generate measures of earnings persistence, the cross-sectional regression models measuring the association between persistence and economic determinants of persistence yield very low adjusted R2s. In sharp contrast, when differenced, higher-order ARIMA models are used to measure earnings persistence, adjusted R2s are in the 10–12 percent range. Moreover, independent variables such as capital intensity, barriers-to-entry, and product-type are all significant in the directions suggested by economic theory. Our results are consistent with Lipe and Kormendi (1994) who argue that higher-order ARIMA models do a better job of capturing the valuerelevance of current period earnings than lower-order models.

This study provides an explanation for the anomalous significantly negative price-earnings relation using the simple earnings capitalization model for firms that report losses. We hypothesize and find that including book value of equity in the valuation specification eliminates the negative relation. This suggests that the simple earnings capitalization model is misspecified and the negative coefficient on earnings for loss firms is a manifestation of that misspecification. Furthermore, we provide evidence on three competing explanations for the role that book value of equity plays in valuing loss firms. Specifically, we investigate whether the importance of book value in cross-sectional valuation models stems from its role as (1) a control for scale differences (Barth and Kallapur 1996), (2) a proxy for expected future normal earnings (Ohlson 1995; Penman 1992), or (3) a proxy for loss firms' abandonment option (Berger et al. 1996; Barth et al. 1996; Burgstahler and Dichev 1997). Our results do not support the conjecture that the importance of book value in cross-sectional valuation stems primarily from its role as a control for scale differences. Rather, the results are consistent with book value serving as a value-relevant proxy for expected future normal earnings for loss firms in general, and as a proxy for abandonment option for loss firms most likely to cease operations and liquidate.

Tax professionals provide valuable services to clients by reducing uncertainty about how clients should report transactions on their tax returns. To reduce uncertainty, tax professionals research applicable authorities (e.g., judicial precedents) and provide assessments to clients of the level of authoritative support for client-favorable positions. Tax professionals have strong incentives to make accurate assessments of the strength of client-preferred positions so that clients will understand the level of risk associated with the reporting position. Further, tax professionals must make accurate assessments of authoritative support in order to maintain compliance with tax professional standards and Federal income tax regulations. Incentives notwithstanding, psychological research on confirmation bias suggests that tax professionals' client advocacy role may inhibit professionals' ability to search objectively for relevant tax authority which, in turn, might inhibit their ability to accurately assess authoritative support. We report the results of two studies that examine causes and effects of confirmation bias in tax information search. In study 1, we find that subjects' information searches emphasized cases with conclusions consistent with the client's desired outcome (i.e., positive cases) over cases inconsistent with the client's desired outcome (i.e., negative cases), despite the fact that positive cases were no more similar to the client's facts. Additional analyses indicate that the extent of this confirmation bias was positively related to their assessments of the likelihood that a neutral court would resolve the issue in the client's favor and this in turn increased the strength with which they recommended the client's preferred tax position. Results of study 2 indicate that confirmation bias induced by client preferences can be strong enough to not only result in inaccurate assessments of authoritative support for the client-favored position, which is problematic in and of itself, but also to lead tax professionals to make overly aggressive recommendations.

This paper investigates the incentives and the penalties related to earnings overstatements primarily in firms that are subject to accounting enforcement actions by the Securities and Exchange Commission (SEC). I find (1) that managers in treatment firms are more likely to sell their holdings and exercise stock appreciation rights in the period when earnings are overstated than are managers in control firms, and (2) that the sales occur at inflated prices. I do not find evidence that earnings overstatement in these firms is motivated by concerns about debt covenant violations or the cost of external financing. The evidence suggests that the monitoring of managers' trading behavior can be informative about the likelihood of earnings overstatement. Many economists believe that insider trading is an efficient method of compensating managers for their efforts. These economists argue that reputation losses would preclude managers from making profitable trades before periods of poor corporate performance. Consequently, this paper also investigates the employment and monetary penalties imposed on managers after the earnings overstatement is publicly discovered. This evidence reveals that (1) managers' employment losses subsequent to discovery are similar in firms that do and do not overstate earnings and (2) that the SEC is not likely to impose trading sanctions on managers in firms with earnings overstatement unless the managers sell their own shares as part of a firm security offering. The evidence suggests that neither employment or SEC-imposed monetary losses are effective in preventing the managers in these firms with extreme earnings overstatements from selling their stake in their firms in the face of declining performance.

A firm with two divisions, each run by a risk-averse manager, contracts with the two managers to operate their divisions and possibly engage in interdivisional trade. Each division can increase the total surplus generated through interdivisional trade by making costly relationship-specific investments. The terms of trade are determined through negotiations between the two managers. Managerial compensation contracts are linear functions of divisional profit and firm-wide profit. If managers are compensated solely on the basis of their divisional profits, they invest less than the first-best amounts. While compensation contracts based on firm-wide profits alone can induce first-best investments, they impose extra risk on risk-averse managers. Therefore, we find that optimal linear compensation contracts will contain both divisional and firm-wide components. Our analysis also identifies a feature of negotiated transfer pricing, namely interdivisional risk sharing, and characterizes its impact on the design of optimal contracts.

The small number of full-scale adoptions of activity-based costing (ABC) coupled with ABC implementation failures have fueled a debate about the costs and benefits of ABC relative to more traditional volume-based costing (VBC) systems. ABC differs from VBC by focusing attention on activities and resources that are under the control of multiple workers. Reducing these costs often requires a coordinated effort. Therefore, incentives that motivate workers to cooperate are a prerequisite to successful process improvements based on ABC. Alternatively, when competitive incentives are combined with ABC, the result can be unexpected and negative. We examine how accounting cost system and incentive structure choices interact. We find that profits are highest when ABC is linked with group-based incentives, which provide high motivation to cooperate. In contrast, the lowest level of profit occurs when the same information-rich cost system, ABC, is coupled with tournament-based incentives. VBC, a cost system that provides a lower level of cost driver information, moderates the incentive effect. Thus, our results demonstrate that the effectiveness of ABC relative to traditional VBC is influenced by its interactive effect with incentive compensation.