Implications of Breach Remedy and Renegotiation for Design of Supply Contracts

A manufacturer writes supply contracts with N buyers. Then, the buyers invest in product development and marketing, and the manufacturer builds capacity. Finally, demand is realized, and the ...rms renegotiate the supply contracts to achieve an e¢cient allocation of capacity among the buyers. The court remedy for breach of contract (speci...c performance versus expectation damages) a¤ects how the ...rms share the gain from renegotiation, and hence how the ...rms make investments ex ante. The objective is to design supply contracts that will induce the buyers and manufacturer to make the ...rst best investments. When the manufacturer is dominant and the breach remedy is expectation damages, the ...rst best is achieved with simple advance purchase contracts. In contrast, with a dominant manufacturer and speci...c performance, the ...rst best may not be achieved. A “tradable options” clause is needed to increase buyers’ incentive for investment. When the buyers have signi...cant bargaining power and the breach remedy is expectation damages, simple advance purchase contracts will always result in underinvestment in capacity and overinvestment in innovation by the buyers. However, if the separability condition proposed in Edlin and Reichelstein (1996) holds and the breach remedy is speci...c performance, then the ...rst best is achieved with simple advance purchase contracts. Finally, building on Maskin and Moore (1999) we describe more complex contracts that induce the ...rst best where simple advance purchase contracts fail. Subject Classi...cations: renegotiation, biform games, bargaining, contract manufacturing, capacity pooling, quantity ‡exibility contracts Preliminary and Incomplete Draft Comments Welcome April 2004 We would like to thank Larry Thomas, Contract Manager at Lonza and Jim Doran, Vice President of Memory Technology Development at AMD, for helpful discussion about renegotiation of advance purchase contracts. We would also like to thank Stefan Reichelstein and our anonymous referees and associate editor for transformative comments on preliminary model formulation.