UK Takeovers and Acquiring Company Wealth Changes: The Impact of Survivorship and Other Potential Selection Biases on Post-Outcome Performance

Abstract

A number of studies of takeover activity, both in the UK and elsewhere, have documented a pattern of long-run negative post-bid returns to shareholders of acquiring companies. One possible interpretation of the phenomena is that it represents a delayed market reaction to overpriced takeovers. An alternative explanation is that the phenomena is illusory and caused by methodological errors in the identification of long-run returns. These errors may arise through choice of inappropriate control models or through the introduction of some element of bias into the chosen sample, including the use of inappropriate tests of statistical significance. Potential sources of bias include survivorship and prior performance bias and the aggregation and testing of non-normal returns' distributions. In the current paper we demonstrate that the pattern of long-run negative returns to acquirers in UK acquisitions over the period 1977-1990 is not a function of survivorship bias. Nor do benchmark controls using both portfolios and matched pairs, with matching on book-to-market and size, provide results inconsistent with long-run negative wealth effects to shareholders of acquiring firms. However the application of test statistics that reflect the non-normal properties of distributions of long-run returns do reduce the significance of results based on parametric statistics. Our results also confirm that a number of characteristics of the acquiring firm can explain some of the cross-section distribution of wealth changes. In particular cash-financed acquisitions consistently outperform acquirers who rely on a method of financing based on equity. Cash-financed acquirers do not experience significant wealth changes when measured over a five-year post-acquisition period. We are unable to conclude that the negative wealth effects experienced by equity-financed acquirers is associated with long-run returns reversal, as tests that use benchmarks of companies matched on the basis of prior performance do not materially alter the pattern of negative long-run wealth losses. Large acquirers also suffer less-negative wealth effects than their smaller counterparts. In addition we also find that regular acquirers suffer no significant wealth losses, irrespective of the control model adopted, and consistently outperform single acquirers. This last finding does suggest that management do at least tend to learn from previous experience. In multiple regression tests, coefficients representing method of financing and experience of acquirers are the only variables that are significant. We find no evidence on any additional 'glamour versus value' effect after controlling for book-to-market and size.

Keywords: Takeovers, acquisitions, wealth effects, survivorship, selection biases

JEL Classification: G34