Vertical Integration and Market Foreclosure

In addition, we shall see that exclusive dealing, even if it is feasible, is not generally a perfect substitute for integration.In particular, if U 's supply costs are finite rather than infinite, then it is no longer optimal for an integrated U -D^pair not to supply D at all.Instead U -D will want to offer D the same amount that U would offer D , but at a slightly lower price (see Section 3) .An exclusive dealing contract will not achieve this.Moreover, a contract that limits the amount that U can sell D" may be very difficult to enforce: given that U is supplying D anyway, it may be hard for D. to verify that supplies equal 100, say, rather than 200.Integration avoids this problem: profit-sharing between U.. and D means that U automatically finds it in its interest to supply the profit-maximizing level (and quality of service) to D.. In extensions of this first variant,, we consider the possibility that it is not known in advance whether U or U is the more efficient supplier, and that the upstream and downstream firms must make ex-ante industry-specific investments prior to trading ex-post.We show that the more efficient (in a stochastic sense) upstream firm will have a greater incentive to merge to monopolize the market ex-post.Also, if U -D merge, D"'s profits will typically fall since if U turns out to be the more efficient firm ex-post, U will channel supplies towards D at the expense of D-. This fall in D 's profits may cause it to stop investing (or exit the industry).To the extent that exit by D. reduces U" ' s profit by lowering the total demand for its product, U" may have an incentive to rescue D by An analysis of exclusive dealing contracts is contained in Appendix 3.The enforcement problem becomes even greater if I' vjants to commit itself not to supply D with quality of service above U 's.