Once SPRB coal is mined, it is processed, cleaned, and then transported to power producers, typically by rail. Customers usually purchase SPRB coal at the mine and then arrange delivery to their plants directly with the help of rail carriers. The price of SPRB coal, therefore, typically consists of the mine-mouth price of the coal plus a transportation cost.
One key feature of different energy sources is the efficiency to produce energy. Coal produces energy using the heat it produces when burning. The higher the heat, the more efficient is the coal. The heat produced by burning coal is typically measured in Btu “per pound” (though per pound is generally excluded in favor of just Btu). The coal found in the SPRB has thermal content of between 8,400 and 9,400 Btu. There are a series of mines in the SPRB that produce coal with a Btu at the upper end of this range (i.e., approximately 8,800 Btu and above). These mines are typically referred to as Tier 1 mines. The largest Tier 1 mines are Peabody’s North Antelope Rochelle mine and Arch Coal’s Black Thunder mine.
A Primer on Market Definition
A common first step when analyzing a proposed transaction’s likely effect on competition is to focus the inquiry on a subset of an industry that includes the products most likely to be affected by the transaction. This process is known as “market definition,” and the subset of an industry that results is known as the “relevant market.” Market definition is described in detail in the 2010 DOJ-FTC Horizontal Merger Guidelines (“Merger Guidelines”). Given the importance of product market definition in the two SPRB coal cases, it is worth discussing it in more detail. Below, we summarize some of the key precepts of market definition established by the Merger Guidelines and discuss how they are applied in practice.
Market definition is concerned with the preferences of customers. It identifies (1) a group of products that customers view as being close substitutes and (2) the geographic area in which customers will look to purchase these products. The group of products that is identified is known as the “product market.” The area in which customers would look to purchase the group of products is known as the “geographic market.” Together, the product market and the geographic market provide an analytically appropriate scope in which to analyze the effects of a joint venture on competition to supply a product to customers.
The Merger Guidelines explain that “[w]hen a product sold by one merging firm (Product A) competes against one or more products sold by the other merging firm,” it is appropriate to “define a relevant product market around Product A to evaluate the importance of that competition.” The product market is complete when one has identified a group of products that customers view as being reasonable substitutes for Product A.
When defining a product market, the key question is where to draw the line between which substitute products should be added to the market and which should not. If one adds fewer products, one may exclude an important substitute for Product A. If one adds products that are more distant substitutes, one is likely to obscure competition between Product A and the products that are closer substitutes for it. For example, including cars in a motorcycle market will tend to underestimate the intensity of competition between motorcycle producers.
The Merger Guidelines employ a well-accepted method to determine which products may be excluded from a market: the hypothetical monopolist test. The hypothetical monopolist test is designed to ensure that important substitute products for Product A are not excluded from the market. Conceptually, it tests whether the competition between a group of products is valuable, where “valuable” means that eliminating that competition would result in a small but significant price increase. In particular, the hypothetical monopolist test asks whether a hypothetical profit-maximizing monopolist of the products in a proposed market would increase the price of at least some products in that market by a small but significant and non-transitory amount, often taken to be 5 percent.
Suppose a hypothetical profit-maximizing monopolist would impose at least a small but significant and non-transitory increase in price (a “SSNIP”). In that case, the proposed product passes the hypothetical monopolist test (i.e., it includes the key substitutes for Product A). If, however, the hypothetical profit-maximizing monopolist would not impose a small but significant and non-transitory increase in price, it means that some important substitutes have been excluded from the market, and the proposed market fails the hypothetical monopolist test. In this case, the market must be expanded to include at least one previously excluded substitute product, and the newly expanded market must be tested again.
It is important to understand that a market need not—and typically will not—include every reasonable substitute for Product A. As the Merger Guidelines put it, “[G]roups of products may satisfy the hypothetical monopolist test without including the full range of substitutes from which customers choose.” Indeed, a group of products may pass the hypothetical monopolist test “even if customers would substitute significantly to products outside that group in response to a price increase.” In the motorcycle hypothetical mentioned earlier, for example, two motorcycle brands could have two-thirds of their lost sales in the event of a price increase divert to cars and still be a product market (i.e., cars may compete with motorcycles and yet not be in the product market because the elimination of competition between motorcycle brands would not be offset by competition with cars). A product market is not a catalog of all substitutes for Product A—rather, it is a collection of products whose competition is valuable (i.e., the removal of competition between the suppliers of those products would lead to higher prices than would prevail if the current level of competition were maintained).
Product Market Definition in the Arch Coal Case
Each of the parties in the Arch Coal case proposed two product markets. The FTC proposed relevant markets of (1) all SPRB coal and (2) a narrower market of 8,800 Btu SPRB coal. The defendants agreed that all SPRB is a properly defined product market, but also argued for an all PRB coal market. This market would have included NPRB coal in addition to SPRB coal.
Faced with these three possible markets, the court first ruled that the evidence supported defining an SPRB coal market. The court noted explicitly in making this determination that both the plaintiff and defense experts had implemented the hypothetical monopolist test and concluded that SPRB coal was a relevant product market.
Turning next to the FTC’s narrower market of 8,800 Btu coal, the court rejected the FTC’s narrower market for two reasons. First, the plaintiffs’ own expert was reluctant to endorse this market. Second, there was “significant evidence” of interchangeability between 8,800 Btu SPRB coal and other types of SPRB coal. There is no evidence that the plaintiffs’ expert performed a hypothetical monopolist test in support of an 8,800 Btu SPRB coal market.
The court rejected firmly the defendants’ contention that there is a broader PRB market. It wrote: “Defendants’ half-hearted argument for a market of all PRB coal is totally unpersuasive.” In rejecting this broader market, the court appealed to the “‘narrowest market’ principle” and described the methodology as the following: “the analysis begins by examining the most narrowly-defined product or group of products sold by the merging firms to ascertain if the evidence and data support the conclusion that this product or group of products constitutes a relevant market. If not, the analysis shifts to the next broadest product grouping to test whether that is a relevant market. This process continues until a relevant market is identified.” This principle is also embodied in the 2010 Merger Guidelines, though in less mechanistic terms: “defining a market broadly to include relatively distant product or geographic substitutes can lead to misleading market shares.”
It is perhaps not surprising that the Arch Coal court accepted an SPRB coal market given that both the plaintiffs and the defendants proposed it and that both economic experts performed hypothetical monopolist tests that supported it. The court rejected both a narrower and broader market than SPRB coal. The court determined that the FTC’s narrower market did not align with the qualitative evidence on substitutability, and this proposed market does not seem to have been supported by a hypothetical monopolist test. The court also found the defendants’ argument that a broader market could be relevant when the narrower all SPRB coal market was a properly defined market to be unavailing.
Product Market Definition in Peabody-Arch
The parties in the Peabody-Arch litigation differed sharply on whether SPRB is a product market. The FTC’s economic expert argued that SPRB coal was the relevant product market, based on multiple applications of the hypothetical monopolist test, each of which found that SPRB coal passed.
To implement the hypothetical monopolist test, the FTC’s economic expert used the critical elasticity method. This method is regularly used to define markets and has been accepted by courts. To apply the method, one first calculates the critical elasticity for a product (here SPRB coal). The critical elasticity captures how willing consumers must be to switch away from the product in the event of a price increase to make a monopolist of that product unwilling to implement that price increase. Next, one calculates the actual price elasticity of demand for the product. This captures how willing consumers actually are to switch away from the product in response to a price increase. Finally, one compares the critical elasticity to the actual elasticity—if the actual demand is less elastic (i.e., less willing to switch products) than the critical elasticity (i.e., the critical willingness to switch), then the hypothetical monopolist test is satisfied, and the group of products is a properly defined product market.
The FTC’s expert calculated margins using the parties’ accounting data and then calculated the critical elasticity for a 5 percent price increase. He then estimated the actual price elasticity of demand for SPRB coal using five distinct approaches. First, he estimated the price elasticity of demand using data on SPRB coal shipments and prices over time. Second, he estimated it using an ordinary-course study of the impact of railroad costs on demand for SPRB coal taken from Peabody’s documents. Third, he estimated it using data on retirements of plants that burn SPRB over time. Fourth, he estimated it using data on how demand for SPRB coal changes as the profitability of SPRB power plants changes in response to the price of SPRB. Finally, he estimated it using the demand for SPRB coal forecasted by an electricity consulting firm under different SPRB coal and natural gas price scenarios.
For each of the FTC’s expert’s five methodologies, the actual price elasticity was less elastic than the critical elasticity, which means that SPRB coal passed the hypothetical monopolist test and is a properly defined product market. Since these estimated price elasticity of demand estimates incorporated the competitive pressures from other potential substitutes like natural gas, they showed that the elimination of competition between SPRB coal producers would likely lead to higher prices, notwithstanding competition between SPRB coal and other energy sources.
The defendants’ economic experts rejected the idea that SPRB coal is a properly defined product market, stating that “commercial realities have changed significantly in the sixteen years since the FTC last litigated—and the federal courts last adjudicated—a merger challenge in this industry.” The defendants’ experts presented three reasons that they believed that SPRB coal was no longer a product market: (1) the availability of low-cost natural gas, (2) increasing generation of electricity by renewable sources, and (3) the risk of coal plant retirements that prevented SPRB coal producers from raising prices. The crux of the defendants’ economic experts’ arguments was that SPRB coal customers would directly switch from using SPRB coal to using other generation sources in large enough numbers that SPRB coal was no longer a product market.
Crucially, though, none of the defendants’ experts calculated an alternative estimate of either the actual price elasticity of demand or the critical elasticity of demand. This failure to produce any alternative estimate supporting a broader relevant market, in the face of the FTC’s expert’s five estimates, may well have been decisive.
The Peabody-Arch court ruled that the FTC had established that the relevant market was for SPRB coal and ruled in favor of the FTC, granting a preliminary injunction. In its opinion, the court highlighted the importance of market definition in its decision stating that “if the FTC does not provide sufficient evidence for its proposed market definition. . . . it will be very difficult to justify its request for a preliminary injunction against the JV.”
The court accepted the hypothetical monopolist test as defined in the Merger Guidelines as an appropriate method to evaluate market definition, writing that “the FTC has presented substantial legal authority supporting the use of the [hypothetical monopolist test] for questions of market definition” and that the defendants “have not persuaded the Court that the energy industry is so different from all other industries that a standard, well-accepted analytical tool like the [hypothetical monopolist test] must be discounted entirely, or that the Court should favor Defendants’ less scientific approach to market definition.”
However, the court’s opinion does not turn a blind eye to the interchangeability of product substitutes. In fact, the court recognized the role of substitution in its thinking about product market definition by citing past case precedents such as Brown Shoe and Process Controls Int’l, Inc. v. Emerson Process Mgmt. In particular, the court addressed product substitution head-on by stating that it is “indisputable… that coal competes with natural gas and renewables in a broader energy market.”
The court ultimately granted its preliminary injunction not because SPRB coal does not compete with other fuels, but rather because “the FTC has presented more than sufficient evidence that there is also a distinct competitive market among SPRB coal producers that satisfies the applicable criteria for market definition.” If alternative energy sources were sufficiently close substitutes for SPRB coal, then the market definition tests performed by the FTC’s expert would have found that the demand for SPRB coal is significantly more elastic.
In the end, then, the Peabody-Arch decision was fully consistent with the Arch Coal decision with respect to the relevant product market, both in general and in the specifics. Both opinions defined SPRB coal as the relevant product market and rejected other markets. Both decisions also confirmed the primacy of the Merger Guidelines approach in general, and the hypothetical monopolist test in particular. And both decisions emphasized the importance of the facts, of using data to implement the hypothetical monopolist test. As such, both decisions argue in favor of neither narrow nor broad markets per se, but rather markets that are defined in accordance with the Merger Guidelines and comport to the facts of the case.