In this post, I want to talk a bit about vertical integration and some cool papers I’ve read on the topic.
To begin with, we can think of a vertical structure being a situation where we have upstream and downstream firms. Upstream firms tend to be manufacturers, or other firms high up in the production process, whilst downstream firms are retailers or firms lower down in the production process. An industry where manufacturers and retailers are non-integrated (separate) can lead to an incentive problem, whereby both firms want/need to make a profit but end up setting too high a price.
Let’s think about this more carefully. In most (product) industries, the manufacturer produces a good, at a cost, which it then sells to retailers at a wholesale price. The retailer purchases this good at the wholesale price and may have other retail costs before it sells this to a consumer for a retail price. When the manufacturer makes a decision as to what wholesale price to sell at, it wants to maximise its profit. It therefore sets the wholesale price at the maximum profit price. The retailer makes the same decision when setting its retail prices, it sets it at the maximum profit price. The problem is, there are now two profit margins which means the final retail price is high, which reduces consumption (law of demand). If either the manufacturer or retailer were to reduce their wholesale/retail price, then there would be more consumption. However, neither has an incentive to do so, when setting prices individually.
We can consider this double marginalisation problem to be an externality problem. The manufacturer, when setting the wholesale price, ignores the externality that a high wholesale price imposes a larger cost on the retailer. The retailer, when setting the retail price, ignores the externality that a high retail price results in fewer sales for the manufacturer.
To solve this problem of double marginalisation, economists, tend to recommend vertical integration. This is the merger of upstream and downstream firms in an industry, which results in the elimination of double marginalisation. Why? Now the two firms are one firm and they internalise the externality and when maximising profits only need to “add” one unit of profit. They realise that by setting a slightly lower retail price, they can sell more products and benefit from more sales on these units. Furthermore, the vertical merger might lead to benefits from economies of scope, enhancing relationship-specific investments and eliminating contracting costs (Kwoka and Slade, 2019).
This sounds great, so what’s the problem! Well vertical integration can cause other competition problems, with concerns around foreclosure arising. There are two forms of foreclosure: input foreclosure and customer foreclosure. Input foreclosure occurs when a vertically integrated firm refuses to supply rival downstream retailers a crucial input into the production process. Customer foreclosure occurs when a vertically integrated firm refuses to purchase the inputs of a rival upstream firm. In the 1980s, proponents of the Chicago School, such as Robert Bork believed that vertical mergers would have no competitive concerns, as they argued that foreclosure would not happen (why lose profitable customers) and that such mergers were highly efficient (Salop, 2018). Modern economic analysis has found foreclosure to be a real economic harm, particularly in oligopolisic markets, which can cause competitive issues through allowing vertical mergers. Vertical integration can lead to increased market power for the merged entity, which may decide to foreclose downstream competitors to reduce competition and increase prices.
Another problem arises when the vertical merger involves a multi-product firm, a firm which sells multiple products, as found in a neat empirical paper by Luco and Marshall (2020). These authors point out that in the context of a multi-product firm, vertical integration for one product but not another in the product portfolio of the firm has two effects on pricing:
– Efficiency effect: products with eliminated double margins are cheaper to sell, putting downward pressure on prices.
– Edgeworth-Salinger effect: products with eliminated double margins earn a relatively higher profit margin giving the firm an incentive to divert sales towards these products and away from other products in the firm’s portfolio. This diversion happens by increasing the price of products which do not benefit from the elimination of double margins.
I find this such a fascinating paper as the theory is pretty clear (or at least, the authors make it so) and the authors use an incredible source of empirical variation to study these pricing effects. They look at the carbonated beverage industry in the US and look at vertical integration between manufacturers (such as Coke, Pepsi and Dr Pepper) and downstream firms (bottlers). They exploit variation coming from real-life mergers between manufacturers and bottlers, where in some cases the bottlers only produced one product whilst in other cases the bottlers produced multiple. The Edgeworth-Salinger effect will only be seen in mergers where the bottler produced multiple products, whilst the efficiency effect will be seen in both types of mergers.
To put it more precisely, Coke and Pepsi bought out bottlers in different parts of the US. In some cases, the bottlers were producing just Coke or Pepsi (single-product firms), whilst in other cases, bottlers were producing Coke and Dr Pepper or Pepsi and Dr Pepper (multi-product firms). The Edgeworth-Salinger effect would predict that the price of Dr Pepper would go up following a vertical merger when the merger involved a multi-product firm.
The results: the authors find that both the efficiency and Edgeworth-Salinger effects are observable in real-life. The price of Coke/Pepsi went down by around 1% following a vertical merger (efficiency effect), whilst the price of Dr Pepper went up by around 1.3% (Edgeworth-Salinger effect).
So we have seen that foreclosure might be an issue and there might be problems when the merger occurs with a multi-product firm. What to do? Well, Kwoka and Slade (2019) point out that the problem of double marginalisation can be solved without the need to resort to a vertical merger. Vertical contracts (or restraints) can be used to align incentives and eliminate double marginalisation. One such contract is a two-part tariff, whereby the manufacturer sets the wholesale price to be the cost and then uses a franchise fee to extract the surplus profit that the retailer receives. Does the manufacturer or the retailer get the lions-share of the profit? De los Santos, O’Brien and Wildenbeest (2018) show that the answer to this question depends on the relative bargaining power of the manufacturer and retailer.
References
De los Santos, Babur, Daniel P. O’Brien, and Matthijs R. Wildenbeest. “Agency pricing and bargaining: Evidence from the e-book market.” Kelley School of Business Research Paper 18-90 (2018).
Luco, Fernando, and Guillermo Marshall. “The competitive impact of vertical integration by multiproduct firms.” American Economic Review 110, no. 7 (2020): 2041-2064.
Kwoka, J. and Slade, M., 2019. Second thoughts on double marginalization. Antitrust, 34, p.51.
Salop, S.C., 2018. Invigorating vertical merger enforcement. The Yale Law Journal, pp.1962-1994.