Many antitrust violations require proof of market power. Historically, the way antitrust litigants and courts have estimated power is by determining a market share of a properly defined “relevant market” of substitutable products. However, many firms sell more than a single product and, frequently, sell non-competing products. The possibility of aggregating noncompeting products or services leads to the creation of “cluster markets” consisting of noncompeting goods.
Antitrust claims are often made regarding such clusters. It then becomes important to ask when it is sensible to locate power in the cluster itself, rather than in the simple presence of any particular item.
This paper, available here, argues that clustering noncompeting products into a single market for purposes of antitrust analysis can be valuable, provided that the limitations of such an approach are understood. Clustering contributes to market power only when it is found, cumulatively, that: (1) many customers need or at least prefer the convenience of receiving the defendant’s grouping of products rather than any single one, or (2) economies of joint provision (economies of scope) make joint distribution of the cluster cheaper per good than distribution of each separately, and (3) entering into competition with the cluster is difficult.
However, often the best way to address the cluster market problem is to avoid market definition altogether. Digital markets are particularly susceptible to direct measurements of market power that do not depend on a market definition.
Section 2 looks at cluster markets in antitrust cases.
Both the Supreme Court and lower courts have recognised antitrust cluster markets several times, often implicitly. Examples include ‘commercial banking’ (a cluster of various types of accounts, loans and other financial services); ‘central station property protective services’ (a cluster of burglary alarms, fire alarm service, and flooding alarms); or ‘hospital services’ (a cluster of very different medical services). All these were recognised as being relevant markets, even though some firms offered only some of those individual goods or services.
Often the “clustering” problem serves to refocus our attention on the precise input that is being monopolised – e.g. not the individual service, but the underlying ‘facility’. Sometimes, the clustering occurs because it is administratively convenient to consider the markets together.
Section 3 considers network externalities and cluster markets.
Large digital platforms often provide numerous products or services. Can these be clustered into a single relevant market for purposes of antitrust analysis? The same criteria that delineate cluster markets in traditional technologies outlined above also apply to digital platforms, but there is also an additional one – network effects. “Direct” network effects make a particular platform more valuable as the number of users increase, although that fact alone does not necessarily provide a rationale for clustering diverse services. Indirect network effects can do the same thing on two-sided markets, making the platform more valuable as the number of participants on the other side increases. The extent to which network effects operate as a substantial entry barrier has been widely debated. Network effects can sometimes operate as a significant entry barrier, although mainly vis-à-vis new entrants attempting to enter with an identical product.
Which individual services are contained in a firm’s cluster could be relevant in a private competitor lawsuit alleging harm that is focused on a particular product or service. In a government suit, however, the only query is whether the cluster as a whole is a meaningful aggregation capable of exercising power. In two-sided markets, it is not uncommon that firms exercise power on one side while they obtain their revenue on the other side. Often, the question will be whether a company’s market power places it in a position either to charge anticompetitive prices or impose unreasonably exclusionary practices on businesses with whom it deals on the revenue-raising side of the market.
Section 4 considers cluster markets without market definition.
Clustering is one way to approach the market power problem in a case involving a digital platform. It is very likely not the best way. After all, in most cases defining a relevant market and computing a market share does not permit us to quantify market power, but only to draw a relatively general inference that it exists. By contrast, “direct” proof of market power relies on estimates of firm elasticity of demand, evidenced mainly by output responses to price changes. Economists favour these methodologies because they are capable of giving more accurate measures of market power. They have the additional advantage that they can slice through the clustering problem by taking aggregated supply or demand as given. Direct measurement has the further advantage that the data that it relies on are usually aggregated so as to reflect the total value that customers place on sellers’ offerings when they are sold in packages.
In a few cases, courts have looked to both cluster market definitions and direct measurement in order to assess power. Typically they regard these as alternative methodologies for answering the same question.
The paper is as thoughtful and insightful as you would expect from the author. My main comment is that its content struck me as being mainly descriptive of how courts and agencies make use of multi-product markets in practice. Personally, I would have appreciated a more analytical dissection of these matters.
For example, I would have liked to see a bit more discussion of how are cluster markets different from other ‘product’ markets where the product can be broken down into different constituent elements (e.g. all types of aftermarkets, particularly if integrated into the product). I would also have liked to see a bit more discussion on how to measure market power directly in zero priced markets, since the discussion focused mainly on the paying-side of platform businesses.