The Whole Foods case illustrates an important lesson about the treatment of marginal and inframarginal customers in merger analysis. Too many investigations turn into free-form inquiries as to whether there are certain customers who, after the merger, could be worse off because — for one reason or another — they would not switch to competing suppliers. Those are then identified as the “inframarginal customers,” whose welfare is at stake. That’s fine as far as it goes. However, discussing the inframarginal customer welfare question in the abstract is putting the cart before the horse. The initial (and in many cases, only) question is whether a post merger price increase vis-a-vis all customers would be profitable. Break-even (or critical loss) analysis helps answer that question. If the price increase vis-a-vis all customers would not be profitable (because the percentage of profits from marginal customers is too great), then the next question is whether the firm has the ability to better target its infra-marginal customers after the merger. Relevant inquiries are whether the firm can identify the infra-marginal customers (e.g., the way that airlines identify business travelers) and, if so, whether the firm can set a different price only vis-a-vis those customers (e.g., through rebates). Given the technological savvy of many firms’ marketing departments, staff often concludes that profitable price discrimination is, in fact, practically feasible. As a result, the merger is identified as potentially harmful.

What’s wrong with that line of reasoning? Primarily that there is usually nothing to connect the feasibility of price discrimination to the merger in any way. In other words, the firms seem to be leaving money on the table in the pre-merger world by not already discriminating on price. What should one make of such evidence of seemingly irrational behavior? First, if price discrimination is uncommon in the industry practice, then there should be a strong presumption for rationality. In other words, chances are that management got it right. It is usually a problem for a theory if reality doesn’t conform to it, not vice versa. Second, there should be a heavy burden on the government to show specifically why as a result of the merger the combined firm is likely to introduce new and different pricing practices. There may well be instances where the post-merger incentives for price discrimination are significantly different from the incentives in the pre-merger world, e.g., if the target is somehow committed to uniform pricing and thus, as a practical matter, prevents price discrimination by the buyer. However, in all other cases where the incentives for price discrimination are the same before and after the merger, we should require highly specific evidence of why there is an increased likelihood of a substantial lessening of competition after the merger.

Technorati Tags: antitrust, merger, price discrimination

Comment feed: RSS 2.0. You can skip to the end and leave a response. Pinging is currently not allowed.