Differential Market Reaction to Pooling and Purchase Methods
[In this study I reexamine the impact of merger accounting method by using a sample of tax-free mergers drawn from a different time period and using more refined cumulative average residual (CAR) methodology. As in the Hong et al. study, the purchase method sample exhibits significant positive CARs over the entire period, which appear to originate in the interval preceding the announcement. The pooling method sample does not generate any significant residuals or CARs. To identify variables for which merger accounting method may be proxying, covariance analysis is used to determine that variables omitted from the capital asset pricing model (CAPM) do not influence the abnormal returns. Furthermore, probit analysis is used to investigate potential indirect cash-flow effects such as managerial income manipulation. A positive relationship between leverage and income-reducing policies is observed. This suggests that firms with high financial risk (leverage) may prefer purchase accounting to reduce reported earnings and regulators' attention. Tax characteristics of the acquired firms (net operating loss and investment tax credit carryforwards) are also examined as they can result in indirect cash-flow effects, such as reduced postmerger corporate income taxes. The limited data available indicate that tax factors have little impact on the CARs. Finally, a variable representing the relative bargaining strengths of the merging firms is tested for association with accounting method used. A statistically significant relationship is observed, and two implications follow from this result: (1) acquiring firms in a strong bargaining position are more likely to use the purchase method, and (2) as the bargaining strength of the acquiring firm decreases, the consideration given to the target firm increases while the magnitude of the CARs decreases. The contributions of the study are as follows. First, the abnormal returns to tax-free purchase method mergers, first observed by Hong et al., are found to persist and appear to originate during the preannouncement period. Second, other CAPM-omitted variables do not appear to influence these results. Finally, two tests of the association between potential indirect cash-flow effects and the CARs are significant and provide a partial explanation for the observed abnormal returns.]