This paper studies the effect of contract duration on the incentive to invest in a relationship when its parties are rational and have perfect information, and contracts are complete, except that short-term contracts specify only current-period actions. Then, short-term contracting distorts investment decisions only when the efficient plan involves mainly sunk-cost investment and the relationship plays a consumption-smoothing role. There is a general, but not universal, tendency to underinvest.

This paper studies the settlements generated by several variants of a compulsory-arbitration scheme called ``final-offer arbitration'' (FOA). Some of these are now in use in several states, and one has been recommended by Clifford Donn as an improvement on existing schemes. Under reasonable economic assumptions, two versions of FOA now being used are equivalent to imposing z̄, the settlement the arbitrator would impose in conventional compulsory arbitration would impose in conventional compulsory arbitration, a result contrary to the intent of the FOA statues. However, a simple modification of Donn's proposal leads to a scheme that generates Pareto-efficient settlements that are at least as good for each agent as z̄. This suggests that substantial gains in welfare could be realized by a simple change in existing FOA statues and possibly also by using the new procedure in situations where compulsory arbitration is not now prescribed by law.

Harstad and Selten's article in this forum performs a valuable service by highlighting the dominance of optimization-based models over boundedly rational models in modern microeconomics, and questioning whether optimization-based models are a better way forward than boundedly rational models. This article complements Rabin's response to Harstad and Selten, focusing on modeling strategic behavior. I consider Harstad and Selten's examples and proposed boundedly rational models in the light of modern behavioral economics and behavioral game theory, commenting on the challenges that remain and the most promising ways forward.

The effect of conventional and final-offer compulsory arbitration on negotiated settlements is characterized, using Nash's variable-threats bargaining solution, with particular attention to the interaction between arbitral risk and bargainers' risk preferences. When there is no arbitral risk, both schemes are shown to have the same effect on negotiated settlements under very general conditions, even when bargaining is over several issues. Whether or not there is arbitral risk, negotiated settlements favour a bargainer more if his opponent is more risk-averse. On the other hand, increases in arbitral risk need not improve the position of the less risk-averse bargainer. As a by-product of the analysis, Kihlstrom, Roth and Schmeidler's risk-sensitivity results for the Nash and Raiffa-Kalai-Smorodinsky solutions are generalized from fixed-threats to variable-threats bargaining.

Journal Article A Self-administered Solution of the Bargaining Problem Get access Vincent P. Crawford Vincent P. Crawford University of California, San Diego Search for other works by this author on: Oxford Academic Google Scholar The Review of Economic Studies, Volume 47, Issue 2, January 1980, Pages 385–392, https://doi.org/10.2307/2296999 Published: 01 January 1980 Article history Received: 01 January 1978 Accepted: 01 May 1979 Published: 01 January 1980

Journal Article A Game of Fair Division Get access Vincent P. Crawford Vincent P. Crawford University of California, San Diego Search for other works by this author on: Oxford Academic Google Scholar The Review of Economic Studies, Volume 44, Issue 2, June 1977, Pages 235–247, https://doi.org/10.2307/2297064 Published: 01 June 1977

Journal Article Relationship-Specific Investment Get access Vincent P. Crawford Vincent P. Crawford University of California, San Diego Search for other works by this author on: Oxford Academic Google Scholar The Quarterly Journal of Economics, Volume 105, Issue 2, May 1990, Pages 561–574, https://doi.org/10.2307/2937801 Published: 01 May 1990

This paper proposes a model of the process by which players learn to play repeated coordination games, with the goal of understanding the results of some recent experiments.In those experiments the dynamics of subjects' strategy choices and the resulting patterns of discrimination among equilibria varied systematically with the rule for determining payoffs and the size of the interacting groups, in ways that are not adequately explained by available methods of analysis.The model suggests a possible explanation by showing how the dispersion of subjects' beliefs interacts with the learning process to determine the probability distribution of its dynamics and limiting outcome.