تبعیض قیمت از طریق بسته بندی معاملات: در مورد monopsony

This paper shows that for a price setting monopsony, offering to transact in a mixed bundle of goods of uncertain quality is profit enhancing. The magnitude of this enhancement relative to no bundling is greater the smaller the gap in the degree of quality uncertainty between the two goods purchased is. Moreover, contrary to coventional wisdom, the use of mixed purchase bundling by a monopsonist is trade enhancing. There is more room for a dramatic improvement in the volume of trade in a good with a low degree of quality certainty if its purchase is combined with a good of a substantially higher quality certainty.

The purpose of this paper is the theoretical analysis of price discrimination through the strategy of mixed bundling exercised by a buyer setting prices under conditions of monopsony power. Mixed bundling takes the form of offering either to trade separately in different goods at specific prices, or to trade in a package of goods at an aggregate price which includes a premium. Mixed bundling is relative to no bundling both profit, for the price setting firm, and trade enhancing.1 An example of bundling is the case of procurement where contracts are combined by the US government agencies. The result of our analysis is that the US government should ensure that it is of a mixed format.2 Such an approach is reflected in the recent pieces of legislation concerning defence procurement where it is required that both bundled and unbundled purchasing arrangements have been considered before a decision is reached.3 Mixed bundling can also be used to deal with information asymmetries. If the firm is unable to determine the quality of the goods offered by a trading partner, it may find it profitable to bundle its purchases. This feature introduces what we refer to as partner preference (e.g. adverse selection). By bundling its purchases and offering a premium for doing so, the firm can reduce adverse selection problems by enhancing its ability to successfully identify trading partners and increase profits. Mixed purchase bundling by the buyer attracts sellers with low costs in either or both goods, leaving out only firms with high costs in both goods. Hence increasing the volume of trade in each good relative to no bundling. As a result, in the case of US government procurement practices trade will increase. This is discussed in Dassiou and Glycopantis (2006). From the point of view of the price setting buyer, under t ransactions bundling he offers a bundled price which is h igher than the sum of the two separate prices.4 In our model package purchasing reduces the buyer’s problem which arises from the dispersion in the sellers’ individual costs.5 Offering a combined price which is higher than the sum of the individual prices he attracts also sellers with a high cost in one of the goods but a reasonable overall cost for the bundle. On the other hand, the separate prices are set to extract surplus from those producers who might find it more profitable to trade in only one good. Hence, by offering the opportunity of mixed bundling the firm increases its ability to identify trading partners and thus both enhances its profits and increases the trade volume. This reason behind purchase bundling by a monopsonist is minimally, if at all, discussed in the literature. In our model there are no complementarities between the goods in the bundle. The argument is that it is profitable for a monopsonist to offer a bundled purchase price which is higher by a premium than the sum of the individual prices on offer and this does not rely on the existence of complementarities between the goods. If anything the presence of these would add a strategic incentive for the monopsonist to bundle (Nalebuff, 2004), especially if he is faced with potential entrants. Section 2 discusses the monopsony model, Section 3 proves that the use of mixed bundling is both profit enhancing for the price setting monopsonist and increases the volume of trade in both goods. Section 4 summarises the conclusions of our analysis.

We have shown that mixed purchase bundling in a two product market is locally optimal. For specific values of the parameters for quality certainty we determine the optimal value of the premium, prices, the surplus and trade volume. Mixed bundling allows the monopsonist to profitably exploit the producers average willingness to sell by offering the option of a purchase premium for a package sale. This is shown to be profitable, without the need for the existence of any complementarities between the goods, as the problem of the dispersion in the sellers’ valuations (costs) of the two goods is reduced. The volume of trade in goods with high quality uncertainty is increased if the transactions for such goods are packaged. This increase is higher when such a good is bundled with one with a substantially lower degree of quality uncertainty. A deduction from our model is that if it is thought that pure bundling is trade restricting as in the case of US government procurement, then the right move is a shift towards the use of mixed bundling, rather than banishing bundling altogether.