As documented by the article in this symposium coauthored by Margaret Levenstein and Valerie Suslow, cartel members often rely on various types of vertical restraints to implement their collusive agreements. Yet while cartels and vertical restraints have both been the object of intense scrutiny, generating a very substantial number of theoretical and empirical analyses, research at the intersection of these two strands of the economics literature remains remarkably scarce. This symposium thus constitutes a welcome first step toward filling this gap. As illustrated by the diversity of the situations and practices tackled by the articles that follow, and by the light that they shed on firms’ business motives and strategies, this topic raises many issues and challenges but also provides rich and fruitful research avenues.
I first briefly review the main lessons from the economics literature dedicated to the interplay between vertical restraints and horizontal cartels. As we will see, this literature has so far mostly focused on the role of provisions related to prices or information as “facilitating practices” that ease the monitoring or the enforcement of tacit or explicit collusion, either upstream or downstream. I then discuss several promising next steps for addressing the key issues and challenges. For this, I draw on some of the lessons of the existing literatures on vertical restraints and on collusion, as well as the insights brought by the following articles.
The literature thus far has focused on the role of vertical restraints in enabling or facilitating collusion at either the upstream or downstream stage of the vertical production-distribution chain. I first focus on purely horizontal collusive scenarios; I then briefly discuss more involved scenarios.
Some vertical restraints operate directly to eliminate competition between downstream firms selling the same brands (“intrabrand competition”), and therefore might be used to support a cartel among these firms to fix prices, limit output, or divide geographic or customer markets. As collusion can be threatened by entry, vertical restraints may also help keep potential entrants out, or exclude actual rivals.
For example, when cartels in Switzerland were legal but monitored and regulated, in response to pressure from the regulator to reduce the prices of German books, bookstores decided to abandon their cartel status and turned to a single importer to handle all trade with German publishers—with resale price maintenance (RPM) as part of the vertical arrangement.
Blatant examples of downstream cartels can still be found in some jurisdictions. Ajay Bhaskarabhatla’s article documents such an example in the Indian pharmaceutical industry, highlighting the role of maximum RPM in a context where this practice is not only legal but mandatory. Through the threat of boycotts, pharmacists have succeeded in coercing manufacturers into collusive arrangements aimed at raising pharmacy margins, by increasing maximum pharmacy prices while maintaining the wholesale prices charged to pharmacists.
That vertical restraints, and most particularly price restrictions, may facilitate horizontal collusion among suppliers has long been recognized, not only in the academic literature but also by courts. For example, in GTE Sylvania the U.S. Supreme Court overturned previous case law and adopted a rule of reason approach to non-price restrictions, while maintaining the per se illegality of price restrictions, quoting Richard Posner as recognizing that “industry-wide resale price maintenance might facilitate cartelizing.” Likewise, in Business Electronics the Supreme Court maintained the per se illegality of price restrictions based on the concern that they “might assist horizontal price fixing at the manufacturer level (by reducing the manufacturer’s incentive to cheat on a cartel, since its retailers could not pass on lower prices to consumers).”
Bruno Jullien and I provided the first formal analysis showing how manufacturers with distinct retail networks—as in the case of franchising or monobrand car dealers—could use RPM to facilitate the monitoring and enforcement of collusive agreements. In the absence of any restriction, retail prices respond not only to manufacturers’ wholesale prices but also to local variations in cost and demand conditions. This is likely to be efficient but it also makes deviations in one manufacturer’s pricing policy hard to detect. As a result, as Edward Green and Robert Porter show, collusion is both inefficient and difficult to enforce: low prices may be induced by local conditions but may also be triggered by manufacturers’ deviations. Hence, low prices need to be “punished” by some form of retaliation, which reduces the profitability of the collusive scheme. This, in turn, makes collusion less sustainable, as the gain from short-term deviations is more likely to be greater than the long-term benefit of continued collusion.
Against this background, manufacturers can use RPM to impose more uniform retail prices. The lost responsiveness to local conditions reduces the efficiency of manufacturers’ and retailers’ bilateral relations but enhances market transparency by making unilateral deviations much easier to detect. This, in turn, facilitates collusion among manufacturers and enables them to sustain higher prices. Overall, the benefit to the manufacturer of higher collusive prices may be greater than the harm from price rigidity. More generally, manufacturers may want to limit retail price flexibility, even without eliminating it entirely, to facilitate collusion and sustain higher prices and profits. As consumers and society favor prices that respond to costs (but not to demand fluctuations), RPM harms them whenever local variations mainly affect retail costs. When, instead, demand is the main source of local fluctuations, RPM is again likely to harm consumers and society when manufacturers find it profitable to use it in order to sustain higher collusive prices.
Continue reading the full text of this article and citations in PDF format.
The author thanks the Phelps Centre for the Study of Government and Business for financial support, and Tom Brown, Margaret Levenstein, and Valerie Suslow for helpful comments.
