In vertical merger investigations, elimination of double marginalization (EDM) is not credited as an efficiency unless it is considered merger specific. For example, in its challenge to the acquisition of Grail by Illumina, the Federal Trade Commission has alleged that even if the claimed EDM benefits of the merger are quantifiable, the parties cannot prove they are merger specific. The FTC alleges an “alternative reality” in which the merging parties can realize EDM by an arm’s-length contract without a full-fledged merger.
Although the FTC’s publicly available (redacted) pretrial brief does not provide enough information about what type of “complex multi-part pricing” contract it refers to, a simple and practical form of multi-part contract that potentially can serve to realize EDM at arm’s length is a two-part pricing contract. A two-part pricing contract is characterized by two elements: a per-unit price for each unit of input, and a fixed fee that is invariant to the number of units of input purchased by a downstream firm. For a two-part pricing contract to realize EDM, the per-unit price of the input must be equal to the upstream firm’s marginal cost, thus avoiding the markup that the upstream firm might otherwise charge. The second element of the two-part price—the fixed fee—can be used to reimburse the upstream firm for its fixed costs and provide it a profit.
The 2020 Vertical Merger Guidelines provide guidance about how merging parties (and the government) can determine whether such EDM-realizing arm’s-length contracts are practically feasible and not merely a theoretical possibility. The Vertical Merger Guidelines add, “The Agencies will generally take the same approach to evaluate the likely contractual arrangements absent the transaction as the one they use when evaluating raising rivals’ costs or foreclosure.” Put differently, a proper evaluation of a merger’s true net effect requires that a consistent set of contractual possibilities be used to evaluate the merged entity’s incentives to foreclose rivals as is used for the evaluation of whether EDM is merger specific. One way to accomplish this consistency when pre-merger prices are characterized by double marginalization is to first imagine the but-for as one in which two-part pricing contracts (say) are used by the upstream firm to sell to downstream firms and then evaluate the harms and benefits of the merger relative to the imaginary but-for.
The fixed fee in a two-part pricing contract offers the upstream firm (and the merged entity following a merger) an additional lever, over and above the per-unit price, with which to earn profits from supplying inputs to downstream firms. Availability of this additional lever can alter the incentives of a stand-alone upstream firm, as well as those of the merged entity, with regard to the per-unit price.
In this article, I address two questions related to the pricing incentives of an independent upstream firm and those of a vertically integrated firm when two-part pricing is practically feasible. First, absent the merger, would an independent upstream firm that has market power have the incentive to offer two-part pricing contracts that eliminate double marginalization? Second, how does the use of two-part pricing affect the merged entity’s incentive to disadvantage rivals relative to when two-part pricing is not feasible? Answers to these questions are essential to determining whether two-part pricing can allow EDM to be realized without a merger but at the same time incentivize a merged entity to disadvantage rivals to the detriment of consumers.
EDM refers to the phenomenon in which, in a vertical merger, the merging firm that produces the downstream product (the “relevant product”) gets access to an input (the “related product”) at cost. Typically, vertical mergers raise concerns when rival downstream firms do not have adequate alternatives to the related product sold by the merging upstream firm. The lack of alternatives gives the producer of the related product the ability to charge a markup over its cost. A vertical merger between the producer of the related product and a downstream firm eliminates the markup on the newly internal sales, thus reducing the merging downstream firm’s cost of production. Although a vertical merger may also create other types of efficiencies, EDM is a prominent form of potential efficiency that is relatively easy to quantify using information that is routinely submitted by the parties during a merger investigation.
EDM can occur after a vertical merger when the related product is sold pre-merger using, say, linear pricing contracts. A linear pricing contract specifies a uniform price for each unit of input. Such linear pricing is common in many vertically related industries. For example, cable distributors pay for network content on a per-subscriber basis. Similarly, computer OEMs that license technologies may pay patent licensors a uniform royalty fee for each computer they produce.
Real-world vertical contracts for scarce inputs often involve linear pricing due to factors that are labeled “transaction costs.” An example of a transaction cost is incomplete information. For example, if demand for a downstream firm’s product is uncertain at the time it writes a contract with the upstream firm, neither firm will know how many units of the related product will ultimately be purchased by the downstream firm. Without this information, the two firms may not be able to agree on the amount of fixed fee (as part of a two-part pricing contract) to be paid by the downstream firm to the upstream firm. In such cases, they may settle for a linear pricing arrangement involving a markup above the upstream firm’s marginal cost.
The linear price of the related product is part of the marginal cost for a downstream firm. A vertical merger, by eliminating the markup previously paid by the downstream merging firm, reduces its marginal cost. The reduction in cost may lead the downstream merging firm to reduce the price of its product and expand output, thus creating procompetitive benefits. Unlike other forms of efficiencies that may be created by a vertical merger, EDM is the internalization of a pricing externality that puts downward pressure on prices. Conceptually, EDM is similar to the well-understood idea of internalization of competitive pricing externalities in a horizontal merger that are considered to create upward pricing pressure.
In two-part pricing, the per-unit price and the fixed fee perform different roles in the upstream firm’s profit maximization. The per-unit price determines the pricing and output decisions, and thus gross profit (profit before the fixed fee is deducted), of the downstream firm. On the other hand, the fixed fee does not affect the pricing decisions of the downstream firm. (Much like fixed cost efficiencies are not considered to affect pricing in a horizontal merger, the fixed fee in a two-part pricing contract does not affect downstream firms’ pricing incentives or their gross profits.) The fixed fee is simply a lump-sum transfer by which the upstream firm, depending on the extent of its market power, can extract some or all of the gross profit earned by a downstream firm. The total profit earned by the upstream firm (before taking into account its own fixed production costs, if any) is the fixed fee from each downstream firm plus any margin embedded in the per-unit price for each unit of input purchased by downstream firms.
When it can use two-part pricing, an upstream firm’s profit maximization boils down to a two-step process. The first step is to choose a per-unit price that serves to maximize the total surplus in the supply chain across the upstream and downstream firms. The total surplus is the total revenue earned by downstream firms minus all costs incurred at the two stages of the supply chain except the cost of inputs paid by the downstream firms and earned by the upstream firm. A larger rather than a smaller surplus can make all parties to the vertical contracts better off when they can divide that surplus in any fashion by appropriately choosing the fixed fees. Once the surplus is maximized, the fixed fee is then chosen for each bilateral contract between the upstream firm and each downstream firm to divide the surplus in a manner that reflects each firm’s relative bargaining power vis-à-vis the other.
It is useful to begin by considering a vertical merger in which the relevant market has only one firm (that is, the merging downstream firm is a monopolist). But for the merger, the upstream firm would first seek to maximize the total surplus in the supply chain by choosing the per-unit price. It can do so by setting the per unit price as its marginal cost. Taking its per unit input cost as given, the downstream firm would maximize its profits by choosing the monopoly price for the relevant market. The monopoly price is the surplus-maximizing price for the hypothesized bilateral monopoly. Thus, the upstream firm would have an incentive to avoid double marginalization by setting the per-unit price of the input at its marginal cost when the downstream market structure is a monopoly.
But what are the incentives of an independent upstream firm when there is competition in the relevant market? Competition between the downstream firms will erode some of the surplus that would otherwise have been generated had the downstream market been a monopoly. To prevent the competitive erosion of surplus, the independent upstream firm will find it surplus-maximizing to set the per-unit price of the input at a level at which each competing downstream firm finds it profit-maximizing to mark that price up to the monopoly price in the relevant market. Put differently, to prevent the downstream firms from competing away some of the surplus, an independent upstream firm will find it optimal to set a high enough per-unit price that competition in the downstream market yields the monopoly outcome. The fixed fees in each bilateral contract can then be used by the upstream firm to divide the surplus between itself and the downstream firms in accordance with each firm’s relative bargaining power.
This per-unit price paid by the downstream firms but for the merger will be higher than the marginal cost of the upstream firm. The more competitive the downstream market, the higher the markup in the surplus-maximizing per-unit input price set by the independent upstream firm. Thus, but for the merger, left to its own profit-maximizing devices, the upstream firm will not have the incentive to set the per-unit price in arm’s-length contracts with downstream firms at the level of its marginal cost. In other words, in the alternative reality of two-part pricing, elimination of double marginalization will not occur so long as there is competition in the relevant market. A merger, on the other hand, will serve to eliminate double marginalization between the upstream firm and the merging downstream firm.
Whether a pre-merger switch from linear to two-part pricing will reduce (i.e., partially eliminate) the extent of premerger double marginalization depends on whether the premerger linear price of the input exceeds the per-unit price in the alternative reality of two-part pricing. Determination of the amount of premerger double marginalization that can be avoided by switching to two-part pricing likely requires a complex simulation using estimated demand and cost functions. Data necessary for such estimation may be hard to come by and the results of such simulation may be sensitive to modelling assumptions.
A prominent and quantifiable form of potential harm from vertical mergers is partial foreclosure through an increase in the price of the input to rival downstream firms (“raising rivals’ cost”). The theory of raising rivals’ cost is rooted in the increased leverage over rival downstream firms that a vertical merger can give to the merged entity. If post-merger, the merged entity and a rival downstream firm fail to reach an agreement on the terms of input sale, the rival—left without access to the input—will lose some or all of its sales of the relevant product. Some units of its lost sales will be diverted to the downstream merging firm, thus adding to the downstream merging firm’s profits. Premerger, the (independent) upstream firm does not benefit from this diversion, but post-merger, the merged entity (of which the previously independent upstream firm is a part) stands to benefit. The merged entity monetizes this additional leverage by raising the price of the input to rival downstream firms.
Competitive harm due to raising rivals’ costs can occur if rivals, faced with an increase in input cost, raise their own downstream prices. Increases in rivals’ downstream prices cause harm to consumers of the rivals’ product.
When pricing is linear, the only price-related way to monetize an increase in leverage is by raising the linear price. On the other hand, when the upstream firm (merged entity) can use two-part pricing, it also can exercise its leverage by raising the fixed fee in lieu of raising the per-unit price. The fixed fee does not affect rivals’ pricing decisions in the relevant market. Thus, with two-part pricing, the merged entity can exercise an increase in its leverage without causing rivals to raise their prices on the relevant product. In other words, two-part pricing allows the merged entity to exercise its leverage without causing harm to consumers in the form of higher prices.
Raising the fixed fee is not just an alternative to raising the per-unit price; it also can be a more profitable way for the merged entity to exercise its leverage. If the input price increase causes rivals to raise their prices of the relevant product (as may be the case when the per-unit component of two-part prices is increased), then, all else equal, that price increase will also cause some diversion from rivals to the merged entity. In other words, even though bargaining disagreement may be avoided, some amount of diversion from rivals will likely still occur if there is an increase in the per-unit component of two-part prices.
In that case, not all of the marginal units of sales lost by rivals may divert to other products in the relevant market. Some of the rivals’ lost customers may cease to purchase (or reduce their purchases) rather than divert to another downstream product. (These are customers whose next best alternative to the rival’s product is some other product that is not in the relevant market.) For customers of the rival that cease to purchase the relevant product or reduce their purchases, the merged entity loses the margin on the associated units of input that are no longer purchased by the rival. If instead the merged entity raises the fixed fee of its two-part prices, rivals do not raise their downstream prices and none of their customers cease to purchase or reduce their purchases. The presence of such customers—those whose next best alternative is not in the relevant product market—can make it profitable for the merged entity to hold per-unit prices unchanged and instead increase only the fixed fee.
A proper economic analysis of foreclosure and EDM requires that the same set of contracting possibilities be assumed for both. If the government alleges that but for the merger, EDM can be realized by a nonlinear arm’s-length contract between the parties, then the assessment of competitive harm from disadvantaging rivals should also be undertaken in a context in which, premerger, firms use such nonlinear contracts. An analysis of two-part pricing contracts, a simple and practical form of nonlinear contract, shows that two-part pricing is likely neither to eliminate double marginalization but for a merger nor to create consumer harm due to passthrough of higher input costs paid by rivals.
The only type of merger for which two-part pricing can fully eliminate double marginalization is one in which the downstream merging firm is a monopolist with no rivals to disadvantage anyway.
If the government alleges that EDM is not merger specific because it can be realized through arm’s-length nonlinear contracts, it needs to clearly describe what type of nonlinear contract it has in mind. Whether it is the government’s burden or that of the parties, several underlying questions then need to be addressed for a court to reach an opinion that is properly grounded in economic principles: Is the nonlinear contract described by the government profit-maximizing for the independent upstream firm? Is it simple and practical enough to be of use to businesses? If so, why has the upstream firm chosen not to use such contracts premerger? What are the implications of such a contract with regard to the merged entity’s incentive to disadvantage rivals?
All views expressed in this article are solely those of the author and do not necessarily represent the views of Charles River Associates, its other consultants, or clients.
