Strategic Delegation By Unobservable Incentive Contracts

Abstract

Many strategic interactions in the real world take place among delegates empowered to act on behalf of others. Although there may be a multitude of reasons why delegation arises in reality, one intriguing possibility is that it yields a strategic advantage to the delegating party. In the case where only one party has the option to delegate, we analyse the possibility that strategic delegation arises as an equilibrium outcome under completely unobservable incentive contracts within the class of two-person extensive form games. We show that delegation may arise solely due to strategic reasons in quite general economic environments even under unobservable contracts. Furthermore, under some reasonable restrictions on out-of-equilibrium beliefs and actions of the outside party, strategic delegation is shown to be the only equilibrium outcome. A partial list includes applications in industrial organization (Vickers (

Strategic incentives for market share

“…See, for example, Vickers (1985) for more discussion of this distinction. 13 More recently, Koçkesen (2004) and Koçkesen and Ok (2004) point out that even completely unobservable contracts can have commitment value in some delegation games, although the assumption of no renegotiation here remains crucial. 14 These results receive some support from experimental studies.…”

Agency Problems and Commitment in Delegated Bargaining

“…Katz (1991), Fershtman and Kalai (1997), Corts and Neher (2003), Kockesen and Ok (2004), and many others have addressed the issue of whether unobservable contracts still can serve as a credible commitment device. Depending on the other party's belief, these papers find that unobservable contracts still can have commitment value when some equilibrium refinements are used.…”

Dynamic commitment games, efficiency and coordination

“…30 Delegation games can also deliver uniqueness results like those in Section 3.3, but require extraneous equilibrium selection arguments to do so. As discussed in Katz (1991), Fershtman and Kalai (1997) and Kockesen and Ok (2004), the observability of contingent payment contracts plays an important role in delegation games. This is true for dynamic commitment games too and particularly evident in equilibria in which commitments occur over multiple rounds, such as those discussed 27 Lipman and Wang (2000) in a related study of revisions with switching costs, allow for simultaneous revisions but get a much weaker uniqueness result.…”