I examine three composite analyst forecast of earnings per share as proxies for expected earnings. The most current forecast weakly dominates the mean and median forecasts in accuracy. This is evidence that forecast dates are more relevant for determining accuracy than individual error. Consistent with previous research, I find analysts more accurate than time-series models. However prior knowledge of forecast errors from a quarterly autoregressive model predicts excess stock returns better than prior knowledge of analysts' errors. This is inconsistent with previous research, and is anomalous given analysts' greater accuracy.

The purpose of this paper is to investigate whether financial analysts with superior earnings forecasting ability can be distinguished on the basis of ex post forecast accuracy. I explore the question by estimating and comparing average accuracy across individuals, and by considering whether the observed distribution of analyst forecast accuracies differs from the distribution expected if their relative performances each year were purely random. Overall, I do not find systematic differences in forecast accuracy across individuals. Financial press coverage suggests there are superior financial analysts. For example, Institutional Investor's annual All American Research Team includes analysts rated by money managers as superior on a variety of criteria, including earnings forecasting, ability to pick stocks, and the quality of written reports. Clearly, financial analyst services other than forecast accuracy are valued by their clients. I focus on only one activity, earnings forecasting, for two reasons. First, forecast data are available, quantitative, and can be evaluated against observable earnings outcomes. Services such as insightful, well-written research reports are harder to evaluate quantitatively. Second, academic use of analyst forecasts as earnings expectations data in capital markets empir-

We examine management discretion to decide when and how much to write down an asset, in a unique case where a tracking stock provides an observable market value for the asset. We find that, despite market evidence that Inco Ltd.'s financial statements substantially overvalued the Voisey's Bay nickel mine throughout 1997 to 2000, management chose not to write down the mine until 2002. Inco management used an independent fairness opinion to justify its December 2000 redemption of the tracking stock at 25% of its initial value, indicating almost surely that Inco management was aware of the generally accepted accounting principles (GAAP) impairment. This case illustrates that GAAP's reliance on undiscounted cash flows for impairment decisions allows huge unrecorded disparities between book and market value. The management discretion exercised in this case provides a concrete example of the subjectivity inherent in fair valuation.

We study lease accounting in an international panel data set to examine how accounting outcomes vary with two features of accounting standards: the emphasis on using professional judgement to apply principles, and the presence or absence of bright-line tests. We study four countries—Australia, Canada, the UK, and the US—and companies in two lease-intensive industries—retail and transportation. Our primary study period spans the time when Australia and the UK switched from domestic to international accounting standards, and in one test, we also consider Canada’s transition to international standards. We find that neither an explicit requirement to apply a principle nor omitting bright-line tests materially increases the use of capital lease treatment among these firms. Overall, we conclude that this financial reporting outcome is relatively insensitive to these standard-setting tools.

Using hand-collected data on analyst locations, we study how geographic proximity affects analyst coverage decisions for U.S. firms that went public during 1996–2009, along with the impact of local coverage on firm visibility. Analysts are 80% more likely to cover local firms than non-local ones, and nearby non-underwriter analysts initiate coverage one to three weeks earlier than distant ones. Proximity matters most for smaller, less visible firms, for firms with less complex operations and for lower status analysts. Less visible firms may use local analyst coverage as a stepping-stone to increase visibility with other analysts and institutional investors.

Maureen McNichols, Patricia C. O'Brien, Self-Selection and Analyst Coverage, Journal of Accounting Research, Vol. 35, Studies on Experts and the Application of Expertise in Accounting, Auditing, and Tax (1997), pp. 167-199

This study examines whether investment banking ties influence the speed with which analysts convey unfavorable news. We hypothesize that affiliated analysts have incentives to respond promptly to good news but prefer not to issue bad news about client companies. Using duration models of the time between an equity issue and the first downgrade, we find affiliated analysts are slower to downgrade from Buy and Hold recommendations and significantly faster to upgrade from Hold recommendations, in both within-analyst and within-issuer tests. We also find affiliated analysts issue recommendations sooner and more frequently after an offering than unaffiliated analysts, and that unaffiliated analysts are more likely than affiliated analysts to drop coverage of sample firms. Our findings indicate that banking ties increase analysts' reluctance to reveal negative news, and that reform efforts must carefully consider the incentives of affiliated and unaffiliated analysts to initiate coverage and convey the results of their research.