Factors Affecting the Probability of Being Acquired: Evidence for the United States
基于英国研究框架，分析美国公司被收购概率的影响因素，发现估值比率等变量在统计上显著但判别能力有限。
Within the merger literature, most attention has been paid to the demand side of the market for corporate acquisitions and the effects of mergers. Relatively less attention has been paid to the acquired firm and the factors influencing the probability of acquisition. The most important studies have been done for the United Kingdom (Singh (I97i and I975) and Kuehn (I975)). This paper presents evidence, consistent with the studies done for the United Kingdom, on the factors affecting the probability of acquisition for American firms. The work by Singh and Kuehn is based on the Marris (i 964) growth maximising model. As applied to acquisitions this model suggests the relative likelihood of firms with certain characteristics being acquired. The managerial objective function is maximised subject to a security constraint expressed in terms of the valuation ratio. This is defined as the ratio of the stock market value of a firm's equity to the book value of its net assets. In the Marris framework, managers obtain security (protection from takeover) by maintaining the valuation ratio at or above some minimum level consistent with managerial security. When the valuation ratio falls too low, the firm becomes a candidate for acquisition. In the strong form of Marris' security constraint there is a level of the valuation ratio above which security from takeover is assured and below which takeover is a certainty. The weak form of the constraint says only that the probability of being acquired varies inversely with the valuation ratio. Kuehn (I 9 75) and Singh (I 97 I) use different methodologies to test the Marris hypothesis. Singh used discriminant analysis (with ten potential individual discriminators') to test whether the valuation ratio (alone or in concert with other variables) is a good discriminator between acquired and non-acquired firms.2 He finds that several of the variables have average values for acquired firms which are statistically different from those of non-acquired firms, but the degree of overlap between the two groups is too great for any variable to be a good discriminator. 'The results indicate that in the cases of most of the... discriminators 40 % or more of the firms would be expected to be misclassified (vs. 50 % on random allocation), if any of them was used for classifying firms into taken-