M. Jensen and K. Murphy (1990,J. Polit. Econ.98, 225ï¾–264) argue that the observed payï¾–performance sensitivity of CEO compensation is too low to be consistent with formal agency theory. This paper uses a dynamic agency model to offer a resolution of the Jensen and Murphy puzzle. We show that the dynamic agency model can predict either a positive or a negative payï¾–performance sensitivity, depending on the parameter values of the model and the distribution of the CEOs' initial expected discounted utilities. For a large variety of parameter values and for properly chosen distributions of initial CEO expected discounted utilities, our model is capable of generating data where the payï¾–performance sensitivity is significantly positive but very small, as in Jensen and Murphy's data. The key to our result is a compensation rigidity that is created endogenously by the optimal dynamic contract.Journal of Economic LiteratureClassification Numbers: C63, D82, G30. *1 This paper was motivated by a conversation with Narayana Kocherlakota to whom I am also indebted for his advice. I thank Steve Williamson for his guidance and support. I thank the associate editor and three anonymous referees for their useful comments and suggestions. I am also grateful to Dean Corbae, Ed Green, Andreas Hornstein, Peter Howitt, Arthur Robson, Steve Spear, and seminar participants at the Universities of Iowa, Western Ontario, Rochester, Queen's, Chicago, Federal Reserve Bank of Richmond, Iowa State, Toronto, Simon Fraser, Carnegie-Mellon, and Illinois for discussions and comments. Financial support from the Social Science and Humanities Research Council of Canada is acknowledged. *2 E. PrescottN. Wallace, Eds.
