A second point, which the Klein excerpt gets at is that if we accept true consumer surplus as the absolute goal of antitrust, we do get into the micro-management of the economy. That is, we know that firms with very small shares price discriminate in highly competitive markets (think grocery store coupons, movie theater tickets, etc.). This is related to the first point.

Finally, a third point is related to the welfare economics of price discrimination. There is a substantial literature that suggests that the welfare effects of price discrimination, when one accounts for both static and dynamic effects (e.g. increased returns to monopolist spur innovation), are less ambiguous and more likely to be positive. In any event, they are part of the normal competitive process in highly competitive markets.

It is very curious to someone relatively new to the field to hear that increasing consumer surplus is not at least a distal goal for competition regulatory authorities. Who are we protecting? Perhaps the answer is new entrants, the presence or absence of which might be easier to measure. Regardless of the answer, I’m all for modesty when it comes to the practical value of our theories and measurements of something like consumer surplus.

You’re right that the point of procompetitive instances of RPM is to induce retailers to provide demand-enhancing point of sale services so as to maximize sales of the manufacturer’s product. This might occur by, as you suggest, limiting price competition among retailers so as to induce them to compete on non-price amenities. In addition, service enhancement may occur — regardless of the presence or absence of inter-retailer competition — if the manufacturer is using RPM to guarantee retailers the sort of retail mark-up (profit margin) that will induce them to promote the manufacturer’s product over those of its competitors. Klein seems to be emphasizing this latter function of RPM.

As for the second point that you raise — i.e., can we be sure that using RPM to enhance point-of-sale services actually enhances consumer surplus? — I think the issue should be irrelevant to the antitrust analysis. A number of theorists, most prominently William Comanor (see Vertical Price-Fixing, Vertical Market Restrictions, and the New Antitrust Policy, 98 Harv. L. Rev. 983 (1985)), have argued that even if RPM enhances total sales it may not enhance total consumer surplus. I summarized their argument in my forthcoming William & Mary Law Review article:

To the extent RPM increases total sales of a manufacturer’s products, it does so by inducing services that make the product at issue more desirable to the consumers who are “on the fence” as to whether to buy the product or not. If the incremental services resulting from RPM increase those marginal consumers’ willingness to pay for the product at issue more than it raises the product’s price, then RPM will result in a greater number of total sales, despite the higher prices. But the additional services that attract new buyers may not be of value to many consumers who value the product more than marginal consumers and would be willing to pay more than the market clearing price, even without the additional services occasioned by RPM. For those “inframarginal” consumers, RPM results in a higher price that is not offset by additional valuable services. This means that their consumer surplus — their wealth gain from buying the product at issue — is reduced by RPM.

Thus, it is theoretically possible for demand-enhancing RPM to reduce total consumer surplus.

It’s really beyond the competence of antitrust to determine whether any particular instance of RPM has enhanced or reduced consumer surplus. (To determine that matter, one would need to know the relative quantities of marginal and inframarginal consumers in a market and the reservation prices of the inframarginal consumers.) Accordingly, I would argue that any output-enhancing instance of RPM should be deemed per se procompetitive and immune from liability.

Klein seems to agree with this analysis. He writes:

[T]he essence of the competitive process is that some consumers gain and others may lose. For example, there are many competitively supplied costly retailer services that increase price which are not consumed by all customers, such as free delivery or intensive sales assistance. The fact that one customer tries on twenty different pairs of shoes over an hour-long period while another customer purchases the same pair in five minutes without trying them on does not mean that we should prohibit retailers from supplying free sales assistance, and prohibit manufacturers from compensating retailers for supplying such service. Although there may sometimes be positive welfare effects from such a prohibition because inframarginal consumers who do not use intensive sales assistance may be better off as a result, it does not make the prohibition procompetitive. The provision of free retailer services, such as salesperson service, is part of the normal competitive process undertaken by firms without market power. Therefore, rather than attempting to regulate this competitive process to protect inframarginal consumers, antitrust policy should leave it up to the competitive market to determine which of these free services are supplied by retailers, often with the financial assistance of manufacturers.

If that’s the argument, then the issue would seem to be whether the consumer surplus captured by the offer of higher-quality services and customized goods is large enough to offset the deadweight loss of consumers priced out by the RPM *and* the longer-term effects that pure price-competition might have on how the retail market evolves (with new retail experiments being more difficult to finance without RPM).

But how do you measure these kinds of quantities? I’m asking naively. I haven’t seen this issue litigated.