In the comments to this post, Peter Swire (Ohio State) points to some recent comments (see also here and here) he submitted to the Federal Trade Commission on how to incorporate privacy into conventional antitrust analysis. The privacy and antitrust link appears to be something that will receive quite a bit of attention in the coming months and years. The basic argument in favor of incorporating privacy into antitrust analysis under appropriate circumstances is not too controversial:
- Antitrust exists to protect against the exercise of market power that reduces consumer welfare
- Reductions in non-price competition can reduce consumer welfare
- Privacy can be a form of non-price competition in some markets
- Ergo, antitrust analysis ought to be concerned with privacy concerns
The first three bullet points are easy to understand. I agree with Swire’s comments that to the extent that privacy amenities (or services or rights) can be an important dimension of non-price competition, antitrust analysis must be flexible enough to incorporate those concerns. Indeed, each of the Commissioners evaluating the Google/Doubleclick merger agreed that privacy concerns are part of the consumer welfare analysis.
What seems to me to be missing in this discussion is a theory of how a particular merger will change the incentives of the firm to provide privacy amenities as a form of non-price competition. Modern merger analysis, especially in the unilateral effects context which seems most relevant here, focuses on the question of how the pricing incentives of the post-merger firm change after the merger. There is a substantial economics literature now which has increased our understanding of how mergers might impact pricing incentives. It is generally no longer sufficient in merger cases to point to an increase in concentration by itself as support for the assertion that consumer welfare will be harmed. An agency challenging a merger must present a compelling competitive effects story. Here, the competitive effects are going to be privacy-related. It seems to me that to move forward from “privacy should count in antitrust analysis because it is a form of non-price competition” to “this merger will reduce privacy and harm consumers” one must have a theory that explains: (1) why the specific merger changes the firms incentives to provide privacy amenities above and beyond a showing that the merger increases concentration, and (2) if the merger creates market power, why the firm will exercise that power in the form of reducing privacy rather than increasing the price.