We find that an intermediary merger is more likely to result in lower output prices if three conditions are met.
First, the upstream input market must not be very competitive and the downstream market must be highly competitive. If the upstream input market is not very competitive and the downstream market is highly competitive, the merging intermediaries will not have a lot of bargaining leverage with input suppliers. The merger, therefore, may increase the merging intermediaries’ bargaining leverage—an effect that puts downward pressure on the input price. A reduction in the input price, in turn, puts downward pressure on the output price.
Second, the merging intermediaries must not be close competitors. The closeness of competition between the merging intermediaries can be measured by “diversion,” the fraction of consumers who would substitute from one merging intermediary to the other merging intermediary if the consumer’s preferred intermediary were not available. Diversion between the merging intermediaries leads to direct upward pressure on output prices and direct downward pressure on input prices. The latter effect leads to indirect downward pressure on output prices. For a given amount of diversion, however, the direct upward pressure on output prices is greater than the indirect downward pressure on output prices. As a consequence, an intermediary merger is more likely to result in lower output prices if diversion between the merging intermediaries is low, all else equal.
Third, for the consumers who would substitute from one merging intermediary to the other merging intermediary if the consumer’s preferred intermediary were not available, the merging intermediaries and non-merging intermediaries must not be close competitors for these consumers. We refer to the fraction of these consumers who would not switch to a non-merging intermediary if neither of the merging intermediaries were available as “diverted diversion.” If diverted diversion is low, an intermediary merger will increase the merging intermediaries’ bargaining leverage with input suppliers more than if diverted diversion were high. The greater increase in the merging intermediaries’ bargaining leverage will, in turn, result in greater downward pressure on input prices and, therefore, greater downward pressure on output prices.
It is unlikely that all three of these conditions will be satisfied in many markets. For example, if the upstream input market is not very competitive and the downstream market is highly competitive, we might expect low diversion between the merging intermediaries and high diverted diversion from the merging intermediaries to other intermediaries.
We also find that an intermediary merger may result in higher input prices even if the merger increases the merging intermediaries’ bargaining leverage with input suppliers. The upward pressure on output prices from the loss of downstream competition, as well as the increased surplus created by operational efficiencies, means that the merging intermediaries and their input suppliers will have more profits to bargain over. As a result, input suppliers may benefit from a larger pie, even if their share of the pie shrinks.
Our model sheds new light on the recently litigated Anthem-Cigna merger. Specifically, our model shows that the merger likely would have led to higher output prices in most (if not all) places where Anthem and Cigna compete. In defending the merger, economic experts for the merging parties claimed that the merger would allow Cigna to pay lower input prices and that the resulting downward pressure on output prices would more than offset any upward pressure on output prices caused by a lessening of competition. The government countered that the input-price reductions did not meet the standard of cognizability in the Horizontal Merger Guidelines. For efficiencies to be cognizable, the Guidelines require that they be verifiable and merger-specific and that they not arise from anticompetitive reductions in output or service. In Anthem-Cigna, the district court agreed with the government that the savings from input-price reductions were not verifiable or merger-specific, but left open whether lower input prices resulting from an increase in upstream bargaining leverage could be cognizable.
If the input-price reductions were cognizable efficiencies, an analysis of the merger’s impact on output prices would have needed to account for these input-price reductions. The challenging question is how to account for these potential input-price reductions in a model. Anthem presented an ad hoc and highly idiosyncratic way to account for input-price reductions, a model that the government challenged on theoretical and empirical grounds. Methods in the academic literature offered a way to incorporate input-price reductions, but these methods were either based on complicated procedures that were impractical for the Anthem-Cigna case—indeed, impractical for almost any litigation—or based on potentially unrealistic assumptions about consumer substitution patterns.
In this article, we offer a workable model for measuring the effects of input-price reductions. We use this model to reconsider the effects of the proposed Anthem-Cigna merger. We find that even with input-price reductions, the merger likely would have led to higher output prices in all of the markets contested by the government.
Continue reading the full text of this article in PDF format.
