In his post on mixed bundling, Hanno distinguished three categories of mixed bundling: cases in which price is below (average variable) cost for both products, cases in which the competitive product is priced below cost after the bundling discount has been applied to price of the competive product, and cases in which the bundling discount will leave the price of the competitive product at or above cost, even after the entire discount has been applied to the competive product. This last category is the interesting one, and the one on which the debate around the LePage’s decision focuses.

Hanno points out two lines that an argument against the per-se legality of above-cost mixed bundling would have to follow. First, that the long term effects may be to limit consumer choice, and second, that a focus entirely on price competition in the discussion jeopardizes consumer choice.

Here is a third line along which an argument against the per-se legality of above-cost mixed bundling could proceed: a distinction between mixed bundling and predation. Looking at the relation of price and average variable cost (as a proxy for marginal cost) makes sense in predation cases, where we need as clear an indication of anticompetitive effects as we can get, given the oddity of an antitrust prohibition against low prices, and given the experience that predatory pricing seems to rarely, if ever, lead to a sustainable monopoly and a recoupment of lost profits. The method of applying the entire bundling discount to the competitive product translates the bundling story into a predatory-pricing story. But an important element is different: The likelihood and timing of recoupment.

The situation in above-cost mixed bundling is the following: A & B are competing makers of widgets, called Wc. A also makes widged Wm and holds a monopoly on Wm. Customers for widgets typically buy both Wm and Wc. Assume that the average variable cost for A and B are the same: $10 per widget. The competitive price for Wc is $12. A sells Wm at $15. A has been losing Wc market share to B. In response, A decides to introduce a mixed bundle, and begins to sell its widget bundle Wm+c for $25.50 (instead of $27 for Wm + Wc). B rapidly loses sales of Wc to A and in response to A’s bundle lowers the price of Wc to $10.50, in effect meeting A’s bundle price for customers buying Wm from A and Wc from B.

B’s margin on Wc has declined from $2 to $.50. In a perfect competition model, price will equate marginal cost, of course. In practice, a price approaching short-run average variable cost will not be enough to sustain B in the industry. In pricing below average variable cost, it is clear that prices cannot be maintained indefinitely at such a level. Whether prices are above average variable cost but below a level of profitability is difficult to tell and will depend an many industry-specific factors. And let’s remember that the Supreme Court in Brooke Group did not state that any price above (an appropriate measure of) cost is pro-competitive; the court found that for predatory pricing (or primary-line cases under Robinson-Patman) prices above cost are “beyond the practical ability of a judcial tribunal to control without courting intolerable risks of chilling legitimate price cutting.”

Above-cost mixed bundling permits instant and certain recoupment, unlike predatory pricing, where recoupment is deferred and uncertain. A’s ability to offer the bundled price depends on A’s margin for Wm, and the mixed bundle constitutes a cross-subsidy from from the monopoly product to the competitive product. This observation is a lever with which an argument for the per-se legality of above-cost mixed bundling could be unhinged. Why should such cross-subsidies be permissible? If cross-subsidies from monopoly to competitive markets are considered potentially anticompetitive, a rule against mixed bundling should be based not on a comparison of price and cost, but on the market power in the bundled-product markets. The less likely it is that A is earning monopoly profits in the market for one bundled product, the less concern there is that a mixed bundle could have anticompetitive effects. A claim under Section 2 would have to show market power in one of the bundled markets, a foreclosure effect in the other bundled market, but not below-cost pricing for the bundle.

Comment feed: RSS 2.0. You can leave a response, or trackback from your own site.