The Generalized Informativeness Principle

Working Paper: CEPR ID: DP10279

Authors: Pierre Chaigneau; Alex Edmans; Daniel Gottlieb

Abstract: This paper shows that the informativeness principle, as originally formulated by Holmstrom (1979), does not hold if the first-order approach is invalid. We introduce a "generalized informativeness principle" that takes into account non-local incentive constraints and holds generically, even without the first-order approach. Our result holds for both separable and non-separable utility functions.

Keywords: Contract Theory; Informativeness Principle; Principal-Agent Model

JEL Codes: D86; J33


Causal Claims Network Graph

Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.


Causal Claims

CauseEffect
first-order approach invalid (C69)original informativeness principle does not hold (D83)
non-local incentive constraints bind (D10)value of a signal may be zero (C29)
first-order approach violated (C69)incentive constraints are not local (D10)
generalized informativeness principle holds generically (D83)set of parameters with measure zero (C25)
signal affects likelihood ratio between principal's preferred effort and all other efforts (D79)generalized informativeness principle necessary (D83)
any signal affecting all likelihood ratios has positive value (C32)generalized informativeness principle generically holds (D83)

Back to index