167 A Summary of Merger Performance Studies in Banking, 1980-93, and an Assessment of the "Operating Performance" and "Event Study" Methodologies Stephen A. Rhoades July 1994

Summary: Mergers reached record levels in the banking industry as well as in the industrial sector in the second half of the 1980s. The general economic conditions of the period and changes in the enforcement of the antitrust laws regarding mergers may have eased the way for some combinations, but there is good reason to believe that the increased merger activity is likely to persist on its own and result in a restructuring of the industry.

The effect of mergers on firm performance is the subject of ongoing debate, and studies of the question have been growing in number. To assess the current state of knowledge, the present work examines the thirty-nine studies found to have been published from 1980 to 1993 on the effects of bank mergers on efficiency, profitability, or stockholder wealth. The first of these studies appeared in 1983; most of them have been published since 1987. This recent burgeoning of research is reminiscent of the period around 1970. At that time, passage of the 1970 amendments to the Bank Holding Company Act and liberalization of bank holding company laws by many states, particularly those with unit banking laws, set off a substantial increase in bank holding company formations, acquisitions, and expansion. That activity in turn stimulated many studies of the performance effects of bank holding company affiliations and acquisitions. By 1980, however, the holding company movement had slowed, and, through the mid-1980s, bank mergers generated little research interest. Then another combination of legislative and marketplace developments led to a resurgence of interest in the performance effects of bank mergers.

This overview is intended to determine whether, in the aggregate, the research since 1980 permits any general conclusions regarding the performance effects of bank mergers. It is not intended to be a study-by-study critique of the research. However, about half of the thirty-nine studies published in the 1980-93 period used a fundamentally different methodology than the other half: Nineteen used the "operating performance" (or "observed performance") approach, which observes the financial performance of a firm following a merger; and twenty-one were "event" studies, which measure the reaction of the stock price of acquirers and targets to a merger announcement (one study used both methods). Hence, after presenting what, on balance, appear to be the conclusions represented by the entire body of studies in the period, the present work concludes with a broad assessment of the two methodological approaches. An appendix summarizes, in a table, the methodological details and results of each study.

Findings of the operating performance studies are generally consistent. Almost all of these studies that find no gain in efficiency also find no improvement in profitability if they include both measures. In contrast, the six studies that show at least some indication of a performance improvement do not obtain consistent efficiency and profitability results, or they are unique in some respect, or both. In general, despite substantial diversity among the nineteen operating performance studies, the findings point strongly to a lack of improvement in efficiency or profitability as a result of bank mergers, and these findings are robust both within and across studies and over time.

In contrast, some inconsistency exists in the main findings of the event studies. For example, seven studies find that a merger announcement has a significantly negative influence on the returns to stockholders of the bidding firm. Seven other studies find no effect on bidder returns, three studies find positive returns, and four find mixed effects on bidder returns. The differences in findings are not readily explicable from differences in approach or the years covered by the analysis. In contrast to the frequently negative or neutral returns to bidders from merger announcements, eight of nine studies that analyze the merger announcement effect on the target bank find a positive return to target stockholders, and one study finds no abnormal return.

In general, the basic findings from the event studies of the announcement effects of mergers provide generally consistent evidence that there are gains to stockholders of target firms. However, the evidence regarding returns to bidders, as well as that regarding the net returns to bidders and targets combined, is too inconsistent to permit any clear conclusion. On balance then, evidence from the event studies does not provide much support for the hypothesis that mergers are expected by the financial markets to improve bank performance because of efficiency gains or other factors.

Two factors seriously undermine the usefulness of event studies relative to operating performance studies. First, and least subject to debate, is that the financial market response to a merger announcement, in terms of abnormal returns to stockholders, reflects expectations about all of the elements (not only efficiency) that may influence the general performance results of a merger as well as differences in expectations between investors and bidders.

Second, event studies are based on short-term movements in stock prices, which may reflect speculation by sophisticated investors who seek short-term trading gains by outguessing other sophisticated market players. To the extent that stock price changes surrounding a merger announcement reflect short-run trading as opposed to long-term investments, abnormal returns would appear to be of limited use for assessing the performance effects of mergers.

On balance, the problems inherent in the event study methodology with regard to the effects of bank mergers appear to be substantially more troublesome than those inherent in the operating performance methodology; one is therefore justified in giving greater weight to the findings of the operating performance studies. Those findings indicate consistently that bank mergers do not generally result in gains in efficiency or general operating performance.

Three caveats attend these findings. First, the findings do not imply that no bank mergers yield efficiency gains. Second, almost all the mergers analyzed occurred before 1989, and mergers after that time might yield different results. And third, the findings by economists that bank mergers generally do not result in efficiency gains is not necessarily inconsistent with the argument by some bankers that mergers lead to significant cost cutting. The difference in the apparently conflicting positions may arise from the fact that economists focus on the efficiency effects of mergers, which are typically measured by an expense ratio such as total expenses to total assets, whereas bankers typically focus on the dollar volume, or percentage, of costs that will be cut. Nonetheless, from the standpoint of public policy and the real long-term performance of the industry, an efficiency measure is the relevant benchmark.

Full paper (126 KB PDF)