منابع سود در ادغام های افقی: شواهد از مشتری، تامین کننده، و شرکت های رقیب

We investigate the upstream and downstream product-market effects of a large sample of horizontal mergers and acquisitions from 1980 to 1997. We construct a data set that identifies the corporate customers, suppliers, and rivals of the firms initiating horizontal mergers and use this data set to examine announcement-related stock market revaluations and post-merger changes in operating performance. We find little evidence consistent with increased monopolistic collusion. However, we do find evidence consistent with improved productive efficiency and buying power as sources of gains to horizontal mergers. The nature of the buying power gains, i.e., rents from monopsonistic collusion or improved purchasing efficiency, is also investigated.

Managers of firms undertaking horizontal mergers and acquisitions often cite improved productive efficiency as the primary source of anticipated gains to mergers.1 On the one hand, improved efficiency in production or distribution or both could arise from greater realization of economies of scale, elimination of overlapping facilities, etc. On the other hand, antitrust authorities frequently posit that horizontal mergers enable the merging firms to gain at the expense of customers and suppliers by allowing the merging firms an opportunity to engage in anticompetitive collusion. For instance, merged firms could more easily collude with rival firms to restrict output to monopoly levels and raise prices at the expense of their customers (see Stigler, 1964). Similarly, merged firms could collude with rivals to restrict aggregate purchases to monopsony levels and thereby lower input prices at the expense of their suppliers (see Robinson, 1933). Merging firms could gain from pooling their purchasing, without necessarily engaging in anticompetitive monopsonistic collusion with rivals. For instance, merging firms could realize purchasing efficiencies if they reduce their input usage or obtain quantity discounts from suppliers. Merging firms could also be able to use their combined purchasing to induce suppliers to compete on price to sell to the combined firm. Further, if the supplier industry is not already competitive, then pooled purchasing and the competition that it induces could allow the merging firms to effectively countervail anticompetitive practices upstream, as noted by Galbraith (1952) and Snyder (1996). If the most efficient suppliers are likely to win such price competitions, then the result is a more efficient allocation of industry resources post-merger. This is in clear contrast to the efficiency implications of anticompetitive collusion. The gains from increased buying power or anticompetitive collusion are not mutually exclusive of improved productive efficiency. However, gains from buying power or monopolistic collusion are distinct from efficiency gains in that they are expected to affect not only the post-merger performance of the merging firms but also the performance of other firms that share a product-market relationship with the merging firms. In this paper, we investigate the relative importance of buying power (both the anticompetitive monopsonistic variety and the efficiency increasing variety) and monopolistic collusion as sources of gains to mergers. We construct a data set that identifies the important corporate customers, suppliers, and rivals of a large sample of firms that announced horizontal mergers and acquisitions between 1980 and 1997. This approach, originally suggested by Eckbo (1983), allows us to examine both the valuation impact at announcement and subsequent changes in operating performance for firms both upstream and downstream of the merged firms.2 To our knowledge, no previous paper has examined the effects of horizontal mergers on actual corporate customers and suppliers in a large sample setting. As in previous studies that examine anticompetitive collusion as a motivation for mergers, we find evidence that rival firms experience positive abnormal returns at merger announcements but do not experience negative abnormal returns when antitrust authorities challenge the mergers. This pattern of abnormal returns appears inconsistent with the conjecture that the market views gains to mergers as largely stemming from anticipated increased collusive activity (of either the monopolistic or monopsonistic variety).3 We also report insignificant stock price reactions for customer firms at merger announcements and negligible post-merger changes in customer operating performance. These results hold even for those corporate customers that appear to be particularly reliant on the merging firms for their purchases. Taken together, the findings for rivals and customers suggest increased anticompetitive collusion is not a significant source of gains to the mergers in our sample. We find that, on average, suppliers experience significant declines in cash-flow margins immediately subsequent to downstream mergers. This result suggests that some form of buying power is an important source of gains in horizontal mergers. However, this finding alone does not allow us to determine whether the gains are related to anticompetitive or to efficiency-increasing manifestations of buying power. To gain additional insight into the nature of these purchasing gains, we examine the effects of mergers on different types of suppliers (e.g., retained versus terminated) and in different industry contexts (e.g., concentrated versus fragmented). This analysis yields several interesting results. We find that the net effect of a merger on a particular supplier appears to depend heavily on the supplier's ability to retain its product-market relationship with the merged entity. Specifically, those suppliers that are terminated subsequent to a customer merger experience negative and significant abnormal returns at the merger announcement and significant cash-flow deterioration post-merger. In contrast, those firms that are retained as suppliers experience significant gains in market share and insignificant abnormal returns or changes in operating performance. These results suggest that merging firms could realize gains by pitting their preexisting suppliers against one another in a price competition to remain suppliers post-merger. Given that winning suppliers do not appear to suffer, we interpret these results as more consistent with efficiency-increasing buying power than monopsonistic collusion given that all suppliers would likely be expected to suffer negative effects under the formation of an effective monopsony. Multivariate analysis confirms the above asymmetric effects of mergers on suppliers’ operating performance and also reveals other interesting cross-sectional patterns in the effects of mergers on supplier firms. In particular, suppliers that are relatively more reliant on the merging firms for sales revenue (i.e., suppliers that face potentially higher costs of switching output to another industry or customer) experience significantly larger reductions in cash flow margins subsequent to the merger. In addition, buying power effects are more pronounced when the merging firms operate in industries that are relatively concentrated. Moreover, greater concentration in the suppliers’ industry is also associated with larger reductions in cash-flow margins for suppliers. The regressions also confirm anecdotal evidence that, in our sample period, buying power effects are particularly important in mergers among retailers. Finally, for the subsample of deals in which the merging firms operate in concentrated industries, we find that supplier post-merger cash-flow reductions are significantly related to merging firm post-merger cash-flow increases (and cost of good sold reductions), which is consistent with increased buying power representing a significant source of gains to mergers in these settings. The above patterns are strongly evident in supplier operating performance changes but not nearly as evident in announcement period abnormal returns. One explanation for this seeming inconsistency is that the market anticipates that the observed operating performance changes will be temporary in nature. That is, the market anticipates that, in the long run, upstream firms might undertake strategic actions such as mergers of their own to counteract the buying power effects of downstream consolidation. Consistent with this explanation, we generally find the strongest evidence of supplier operating performance changes in the year immediately following the merger, with less significant changes in subsequent years. This paper proceeds as follows. In Section 2, we develop the hypotheses to be tested. We describe our sample construction and methodology in Section 3. We report results in Section 4 and offer a concluding discussion in Section 5.

We investigate the upstream and downstream product-market effects of a large sample of horizontal mergers and acquisitions from 1980 through 1997. We construct a data set that identifies the corporate customers, suppliers, and rivals of firms initiating horizontal mergers, and we use this data set to examine announcement-related stock market revaluations and post-merger changes in operating performance. For the merging firms themselves, we find positive and significant abnormal returns at merger announcement. Also, several subsamples of merging firms are found to significantly increase post-merger cash flows and reduce cost of goods sold. These findings are consistent with mergers generating gains for the merging firms. However, as detailed in Table 1, these findings alone do not allow us to identify the nature of the gains. Thus, we also examine the effects of mergers on customers, suppliers, and rivals. We find that customers experience insignificant stock market reactions at announcement and negligible changes in industry-adjusted operating performance subsequent to upstream mergers. These results continue to hold for mergers in concentrated industries and for mergers that result in a significant increase in concentration. Further, the results are similar for customers that ex ante appear particularly reliant on the merging firms for inputs and thus would be expected to face the greatest switching costs. Finally, as in previous studies, we also find evidence that rival firms experience positive abnormal returns at merger announcements but do not experience negative abnormal returns when antitrust authorities challenge the mergers. Taken together, the customer and rival results are strongly inconsistent with the monopolistic collusion hypothesis. The evidence for the merging firms and their suppliers is generally consistent with some form of increased buying power being an important source of gains to mergers. For instance, several subsamples of merging firms are found to reduce their cost of goods sold to sales post-merger while their suppliers as a group experience negative and significant reductions in cash flow to sales immediately subsequent to a downstream merger. In addition, a sign test reveals that significantly more individual suppliers experience negative announcement period returns than positive. Suppliers that are relatively more reliant on the merging firms for sales revenue experience significantly larger reductions in cash-flow margins subsequent to the merger. Furthermore, buying power effects are more pronounced when the merging firms operate in industries that are relatively concentrated. While all of the supplier results summarized above are consistent with the general notion of buying power being a potentially important source of gains to horizontal mergers, these results do not allow us to determine the nature of these gains; i.e., monopsonistic collusion or purchasing efficiencies/countervailing power. Because the two buying power hypotheses have somewhat similar predictions, we rely on somewhat subtle but important distinctions between the hypotheses to further identify the nature of the buying power gains. The monopsonistic collusion hypothesis suggests all potential suppliers could suffer from a downstream merger, and the purchasing efficiencies/countervailing power hypothesis predicts potentially asymmetric effects on suppliers. Specifically, suppliers that are retained post-merger might not be harmed on balance by the merging firms’ increased buying power; i.e., they could sell a higher quantity but at a lower price. Those suppliers that lose a bidding competition might suffer. Thus, while some individual suppliers might be expected to experience net gains in response to merging firm purchasing efficiencies or the exercise of countervailing power or both, it is relatively less likely that any suppliers would benefit under monopsonistic collusion. Empirically, we do find that the net effect of a merger on a particular supplier appears to depend heavily on the supplier's ability to retain its product-market relationship with the merged entity. Specifically, those firms that are terminated as suppliers subsequent to a merger experience negative and significant abnormal returns at the merger announcement and significant deterioration in cash flows post-merger. In contrast, those firms that are retained as suppliers experience significant gains in market share and insignificant abnormal returns and changes in operating performance. Following the reasoning above, we interpret these results as somewhat more consistent with efficiency-increasing buying power than monopsonistic collusion. We also find that suppliers that operate in concentrated industries experience larger performance declines after a downstream merger. This result is consistent with an efficient form of countervailing power lowering rents in the supplier industries. Further, we find that customers experience insignificant announcement returns. This finding is more consistent with purchasing efficiencies/countervailing power than monopsonistic collusion because monopsonistic collusion can often negatively affect customers as well as suppliers. In general, our overall results suggest that horizontal mergers’ effects on suppliers and merging firms appear to be somewhat temporary in nature. Specifically, merging firm improvements in operating performance and supplier operating performance deterioration appear to be strongest in the first year after the merger. One plausible explanation for this finding is that, in the longer term, upstream firms can undertake strategic actions of their own to counter the increased buying power of their customers; e.g., mergers to consolidate the supplier industry. Empirically validating this conclusion is a potentially fruitful topic for future research. Finally, we believe our methodology to identify product-market relationships could be of independent interest to other researchers and that this data could be useful in answering many types of questions beyond those addressed in the present study.