• Selected
• Entire List

Abstract:
In this paper we investigate the usefulness of analysts' earnings forecast revisions in the allocation of funds to different industries and countries. In particular, we ask whether a post analyst revision announcement drift in prices can be exploited to guide an asset allocation strategy based on industry, or country, selection. Using monthly consensus I/B/E/S-First Call analysts' earnings forecasts for companies listed on the main European stock markets over the period January 1987 to December 2001, we find a significant post revision announcement effect for individual companies. However, the abnormal returns evaporate away as we move from an individual company level to an industry or country level. We provide two kinds of evidence which seem to cast doubt on the analysts' ability to fully incorporate industry and country specific information into their forecasts. First, we show that returns are driven more by common components than earnings forecast revisions. Second, we find that company specific news reflected by the revision signal dominates industry or country news.

Earnings forecasts, asset allocation

Abstract:
This paper investigates the effects of takeovers on workers' employment prospects and wages in the UK for the years 1987-1995. We address directly the idea that takeovers involve a 'breach of trust' with employees. Our results provide no support for the breach of trust hypothesis and rather suggest shareholders and workers in the post-acquisition joint entity are locked in a form of 'equal misery' following the execution of the takeover.

There already an exist a wide range of event studies documenting the effect of takeovers on shareholders and a smaller number of studies discussing the impact of takeovers on employees. The contribution of the present study is to relate the separate effects of acquisition on these two groups to each other. By doing so we seek to test directly the proposition that takeovers reallocate rents from workers to target shareholders, via the bid-premia paid on acquisition.

Abstract:
We revisit the debate on the interpretation given to prior-year earnings changes in predicting analysts' future forecast errors. We advance a new specification of this relation that distinguishes between earnings reversion and momentum. For a large UK dataset for the years 1990-1996, we find substantial underreaction, particularly in situations of earnings momentum. We find that underreaction is further increased for cases of downward earnings momentum when the analyst's merchant bank acts as a broker to the company. We interpret this as a reporting bias caused by an analyst's response to bad news being compromised.

Abstract:
This paper recasts an old debate in accounting, that of the appropriate means of revaluing assets in place, within a new theoretical and empirical framework. The new theoretical framework is to regard the current movement of current cost accounting indices as reflecting inter-industry arbitrage of both assets in place and investment. This new perspective results in a new empirical development which draws upon recent advances in time-series econometrics, namely, error correction representations of the trajectory of related time-series. We construct error correction representations of both the current-cost and the current purchasing power method of valuing assets in place and compare them according to their ability to forecast the future value of assets in place. The results suggest that the time path of specific price indices has a parsimonious error-correction representation involving other related price indices. The bivariate and multivariate tests of forecasting accuracy we conducted show that the error correction model clearly outperforms the random walk benchmark model.