Pieter T. Elgers’s research while affiliated with University of Massachusetts Amherst and other places

This study addresses the impact of firm- and time-specific attributes on the accuracy of composite forecasts of annual earnings, constructed as weighted averages of annual earnings forecasts obtained from three sources: time-series, price-based, and analysts’ forecasts. Separate evaluations are provided for smaller and larger firms, and for firms with relatively lower and higher analysts’ coverage. In addition, results are developed separately for time periods pre-dating and following the implementation of Regulation Fair Disclosure (FD). Our results using a pooled sample provide directional support for the use of composite forecasts as earnings expectations. The ensuing analyses within the size and analysts’ coverage partitions of the sample show that the forecast errors for smaller and lightly-covered firms exceed those for the larger and heavily-covered firms by factors of three or four times. Analyses within partitions for time periods before and after the Regulation FD indicate that the relative accuracy of the composite forecasts is time-specific. In the earlier time-period preceding the implementation of Regulation FD, composite forecasts significantly outperform each of the three individual forecast sources used in their construction. Moreover, the extent of improvement in accuracy of composite forecasts is significantly higher for the smaller and lightly-covered firms in the Pre-FD period. Collectively, these results suggest that the predictive accuracy of composite forecasts of annual earnings is contextual.

This paper investigates whether the market’s under-weighting of financial analysts’ earnings forecasts documented in Elgers, et al. (2001) is driven by share price scaling and/or omitted risk factors. Our empirical findings indicate that the market’s mispricing of financial analysts’ forecasts of earnings are not affected either by the incorporation of share-price covariates or by the inclusion of controls for omitted risk factors. We also document that the delayed securities returns persist in more recent years and are significant only for firms that are on the short-sale side of the profitable hedge portfolios, suggesting that the hedge portfolio strategy may be difficult to implement in practical cases. Overall, these findings indicate that neither share price scaling nor omitted risk factors underlie the evidence of the market’s under-weighting of financial analysts’ forecasts. Further research, however, is needed to address whether the economic returns available from trading on the information in price-scaled analysts’ forecasts are adequate to compensate for transaction costs and other securities market trading frictions.

This paper re-evaluates evidence in Ayers and Freeman [Ayers, F., Freeman, R., 1997. Market assessment of industry and firm earnings information. Journal of Accounting and Economics 24, 205–218] suggesting that investors anticipate industry-wide components of earnings earlier than firm-specific components, and that post-earnings-announcement drift following annual earnings announcements is due primarily to firm-specific components of earnings. Our tests indicate that post-announcement drift is entirely attributable to coefficient bias due to measurement errors in the use of realized earnings changes as proxies for unexpected earnings. Also, coefficient differences in the market's anticipation of subsequent-year industry and firm-specific earnings become insignificant when we introduce suitable controls for non-linearity in the return/earnings relation.

This paper re-examines the effect of the inventory costing method on the association between accounting risk measures (ARMs) and market risk, and extends earlier research in several respects. The groups of FIFO and LIFO firms are matched on the basis of various financial characteristics to reduce selfselection bias, and the effect of inventory costing methods on the usefulness of ARMs in predicting market risk is investigated. The findings indicate that predictions based on FIFO show an improvement on market based predictions, but predictions based on LIFO fail to show such an improvement.

Several studies have made inquiries of corporate managers concerning their motivations for undertaking stock splits and stock dividends. This paper investigates whether the factors identified by these studies are in fact associated with the actual stock distribution decisions of managers. The results are consistent with the view that managements issue large stock distributions (25 per cent or greater) in order to keep the per share price in an optimal range and to signal optimistic expectations to the market. Firms with relatively low per share prices were inclined to issue small stock distributions (less than 25 per cent); the signaling motivation also played a role here.

Recent accounting and finance literature has introduced the term returns synchronicity, the relation of firm-specific returns to market-wide and industry-wide returns. This paper evaluates the power of fundamental synchronicity, proxied by financial accounting performance measures, to explain cross-sectional variation in returns synchronicity. In addition, we also test the potential effect of fundamental synchronicity on inferences about the relation of analyst coverage to returns synchronicity, as reported in Piotroski and Roulstone (2003). We find that various fundamental synchronicity measures are incrementally important to explain returns synchronicity. Furthermore, inferences about the relation of analyst coverage to returns synchronicity are sensitive to the inclusion of fundamental synchronicity measures as control variables.

Previous studies report delayed securities returns associated with analysts' earnings forecasts, prior earnings, and prior securities returns. Although this evidence typically has been attributed to biases in investors' expectations of earnings, omitted risk factors constitute an equally plausible potential explanation of the delayed returns. For this reason, our paper evaluates whether the earnings expectations implicit in securities prices weight the information in analysts' forecasts, prior-year earnings, and prior securities returns in a manner consistent with their historical relations to earnings. We find that securities prices reflect a significant over-reliance on historical earnings and a substantial under-reliance on both analysts' forecasts and prior-year securities returns. Composite earnings forecasts that are based on the historical relations of realised earnings to analysts' forecasts, prior year earnings and prior year returns have significantly lower forecast errors than do composite earnings forecasts that are based upon inferred investor weightings of these same information variables. Moreover, significant size-adjusted returns are achieved with hedge portfolios based upon errors in investor earnings expectations that are predicted from differences in the historical and investor weights of these complementary information variables. A disproportionate part of these delayed returns occur in earnings announcement months. Overall, these findings lend credibility to the interpretation that biases in investors' earnings expectations, rather than omitted risk factors, underlie at least a portion of the delayed securities returns.

This paper examines whether financial analysts’ forecasts of annual earnings reflect an over-weighting of working-capital accruals that is comparable to the over-weighting implicit in securities prices documented in Sloan [Account. Rev. 71 (1996) 289] and Bradshaw et al. [J. Account. Res. 39 (2001) 45]. Our results indicate that the over-weighting of working-capital accruals in analysts’ earnings forecasts is less than one-third of the over-weighting by investors that is implicit in stock prices. Moreover, we are able to attribute less than 40% of the delayed securities returns associated with working-capital accruals to subsequent errors in analysts’ annual earnings forecasts, for firms with lower analyst coverage. These findings suggest that securities market inefficiencies that are unrelated to financial analysts’ earnings forecasts underlie at least part of the accruals-related anomaly.