Knowledge that Transforms

Treasury Regulations typically require a firm to use its manufacturing cost as one input in calculating the transfer price to be used for tax purposes. Because the characteristics of the firm's capital assets determine the realized manufacturing cost, the firm has an incentive to distort its investment choices in order to receive a more favorable tax transfer price. This represents an ex ante form of shifting income across divisions. Practical difficulties in enforcing the regulations provide the firm with another way to shift income. In particular, regulators often can identify only a range of acceptable transfer prices. As a result, the firm will not use the true “arm's length” transfer price, but will instead choose a transfer price that is more favorable for tax purposes. I find that the opportunity to shift income ex post in this manner may reduce the value of investment-based ex ante income shifting. That is, if the firm has more discretion over the ex post transfer price, then it may invest more efficiently ex ante. Although it is commonly perceived that ex post income shifting is purely deleterious, the analysis shows that it may have benefits.

When a firm practices conservative accounting, changes in the amount of its investments can affect the quality of its earnings. Growth in investment reduces reported earnings and creates reserves. Reducing investment releases those reserves, increasing earnings. If the change in investment is temporary, then current earnings is temporarily depressed or inflated, and thus is not a good indicator of future earnings. This study develops diagnostic measures of this joint effect of investment and conservative accounting. We find that these measures forecast differences in future return on net operating assets relative to current return on net operating assets. Moreover, these measures also forecast stock returns—indicating that investors do not appreciate how conservatism and changes in investment combine to raise questions about the quality of reported earnings.

This paper suggests a new measure of one aspect of the quality of working capital accruals and earnings. One role of accruals is to shift or adjust the recognition of cash flows over time so that the adjusted numbers (earnings) better measure firm performance. However, accruals require assumptions and estimates of future cash flows. We argue that the quality of accruals and earnings is decreasing in the magnitude of estimation error in accruals. We derive an empirical measure of accrual quality as the residuals from firm-specific regressions of changes in working capital on past, present, and future operating cash flows. We document that observable firm characteristics can be used as instruments for accrual quality (e.g., volatility of accruals and volatility of earnings). Finally, we show that our measure of accrual quality is positively related to earnings persistence.

This study investigates the extent to which a retailer's willingness to share internal (sales and inventory) information with a manufacturer and the reliability of the information transmission between the retailer and the manufacturer affect the total supply-chain profits resulting from two alternative inventory management systems. I present analytical models of a traditional system and a Vendor Managed Inventory (VMI) system. The VMI system is a supply-chain management technique in which the retailer delegates its inventory decisions to, and shares its internal accounting information with, the manufacturer. With VMI, the parties aim to reduce inventory-related costs and increase supply-chain profits. The theoretical analysis indicates that the system that produces higher supply-chain profits depends on the extent to which the retailer reveals its internal accounting information to the manufacturer and the ability of the manufacturer to accurately receive and use this information in its decisions. Survey data corroborate the model's prediction that manufacturers are more likely to select VMI when retailers provide more precise sales and inventory (i.e., demand) information and when the manufacturers' systems ensure reliable transmission and receipt of this information. The study's results suggest that managerial accountants should consider the retailer's willingness to share sales and inventory information with the manufacturer and the manufacturer's ability to reliably transmit this information before implementing VMI systems.

Archival studies report that analysts' annual earnings forecasts are optimistic, particularly for firms reporting recent losses. This study addresses whether forecast optimism is an unintentional consequence of analysts' reactions to the structure of information about managers' future plans. I investigate whether analysts who are forecasting earnings use processes consistent with scenario thinking: envisioning a sequence of events in which proposed actions lead to future outcomes. I study professional sell-side analysts in a 2×2 between-subjects experiment with the structure of information (scenario vs. list) and the sign of prior earnings (loss vs. profit) as independent variables. I find that analysts make more optimistic two-year-ahead earnings forecasts when provided information about a manager's future plans framed as scenarios than when provided the same information framed as lists. I also find that scenario-induced optimism is greater for a firm with prior losses than for a firm with prior profits. The results are consistent with scenario thinking causing analysts to issue optimistic forecasts.

This study compares samples of publicly and privately held bank holding companies to examine whether the high frequency of small earnings increases relative to small earnings decreases reported by public firms is attributable to earnings management. We expect public banks' shareholders to be more likely than private banks' shareholders to rely on simple earnings-based heuristics in evaluating firm performance, so we expect public banks to have more incentives to report steadily increasing earnings. Consistent with this expectation, we find that relative to private banks, public banks: (1) report fewer small earnings declines, (2) are more likely to use the loan loss provision and security gain realizations to eliminate small earnings decreases, and (3) report longer strings of consecutive earnings increases. These results suggest that the asymmetric pattern of more small earnings increases than decreases, first documented by Burgstahler and Dichev (1997), is attributable to earnings management and is not simply a reflection of the underlying distribution of earnings changes.

Questionnaire responses reported by Luft and Libby (1997) reveal that transfer price negotiators expect fairness-based price concessions that moderate the influence of an outside market price when the outside market price strongly favors one of the parties. We examine whether these expectations of fairness extend to the actual prices that result from real-cash negotiations. Findings indicate that expectations of fairness-based price concessions do not survive actual negotiations when participants negotiate over a computer network with no communication other than bids, asks, and acceptances. Conversely, both expectations and actual negotiated outcomes reflect fairness-based price concessions when participants negotiate in a face-to-face setting with unrestricted communication. Together, these results imply that the extent to which questionnaire-based judgments of social behavior generalize to actual behavior depends on the whether the competitive environment suppresses or reinforces the social presence necessary to sustain phenomena such as preferences for fairness.

I report the results of an experiment designed to investigate the influence of noncredible communications and group affiliation on auditors' formation of self-serving bias. I find that manager-subjects use noncredible communications to induce auditors to develop an unwarranted trust of managers (i.e., a biased judgment). However, the bias is neutralized when auditor-subjects belong to groups that create social pressure to conform to group norms. Thus, my finding calls into question the Bazerman et al. (1997) conclusion that auditors cannot conduct impartial audits due to self-serving biases resulting from repeated interactions between auditors and their clients.

We test the conjecture from prior research that defense contractors' excess profitability in the 1980s stemmed from their ability to shift common overhead costs to government contracts that typically allow cost reimbursement or price renegotiation (Rogerson 1992; Thomas and Tung 1992; Lichtenberg 1992). Although we confirm prior evidence that defense contractors enjoyed abnormally high profitability on their government work in the 1984–1989 period (a period of relatively low competition for defense contracts), we find no evidence that this excess profitability is attributable to cost shifting. In addition, we find no evidence that the Top 100 defense contractors (firms that likely wield above-average market power) are able to use cost shifting to exploit a lack of competition in the industry. Our results suggest that, contrary to the conjectures in prior research, the unusually high profitability reported on government contracts in 1984–1989 is more likely attributable to nonaccounting explanations than to cost shifting.

We investigate three explanations for prior studies' finding that the usefulness of earnings announcements, as measured by their absolute market responses, has increased over time. We confirm this increase for a sample of 426 relatively large, stable firms over 1980–1999. We find no evidence that this over-time increase in the magnitude of the market reaction to our sample firms' earnings announcements is attributable to increases in the absolute amount of unexpected earnings conveyed in the announcements or to increases in the intensity of investors' average reaction to unexpected earnings. To test the third explanation—an over-time expansion in the amount of concurrent (with bottom line earnings) information in earnings announcement press releases—we analyze and code the contents of 2,190 earnings announcement press releases made by 30 of our sample firms over 1980–1999. Concurrent disclosures increased significantly over this period and we find that these concurrent disclosures, especially the inclusion of detailed income statements, explain increases in the absolute market reactions to earnings announcements for our sample firms.