Asymmetric‐information allocation to avoid coordination failure

• Author: Fumitoshi Moriya,Takuro Yamashita
• DOI: http://doi.org/10.1111/jems.12329
• Date: 01 January 2020
• Published date: 01 January 2020

J Econ Manage Strat. 2020;29:173–186. wileyonlinelibrary.com/journal/jems © 2019 Wiley Periodicals, Inc.

|

173

Received: 3 November 2017

|

Revised: 5 August 2019

|

Accepted: 21 August 2019

DOI: 10.1111/jems.12329

ORIGINAL ARTICLE

Asymmetric‐information allocation to avoid coordination

failure

Fumitoshi Moriya

1

|

Takuro Yamashita

2

1

Kobe City University of Foreign Studies,

Kobe, Hyogo, Japan

2

Toulouse School of Economics,

University of Toulouse Capitole,

Toulouse, France

Correspondence

Takuro Yamashita, Toulouse School of

Economics, University of Toulouse

Capitole, 21 Allée de Brienne, 31000

Toulouse, France.

Email: takuro.yamashita@tse-fr.eu

Funding information

Japan Society for the Promotion of

Science, Grant/Award Number: 25780189;

H2020 European Research Council,

Grant/Award Number: 714693; JSPS

KAKENHI, Grant/Award Number:

25780189; European Research Council,

Grant/Award Number: 714693

Abstract

In the context of team production, this paper studies the optimal (deterministic

and stochastic) information allocation that implements desired effort levels as

the unique Bayesian equilibrium. We show that under certain conditions, it is

optimal to asymmetrically inform agents even though they may be ex ante

symmetric. The main intuition is that informing the agents asymmetrically can

be effective in avoiding “bad”equilibria, that is, equilibria with coordination

failure.

KEYWORDS

asymmetric‐information allocation, moral hazard, unique implementation

JEL CLASSIFICATION

D21; D23; D86

1

|

INTRODUCTION

This paper considers a simple team‐production problem as in Holmstrom (1982), with an additional element that the

environment depends on some exogenous state variable. Specifically, two (risk‐neutral) agents either work or shirk in a

project, and the project either succeeds or fails. The success probability of the project depends not only on the agents’

total efforts, but also on the binary state of the world (high/low). Given everything else equal, we assume that an agent

has a higher incentive to work in the high state than that in the low state (“state effect”). Therefore, naturally, not only

the bonus contract offered by the principal but also the agents’information structure about this state variable would

affect the agents’incentives. Our goal is to obtain some insights about the optimal bonus scheme and optimal

information structure.

More specifically, our main interest is in the environments in which the efforts are complementary for success

(or more precisely, the success probability is a supermodular function of the agents’effort choices), and failure means a

hazardous event. For example, in a nuclear power plant, one agent’s shirking drastically increases the probability of a

serious and tragic accident. Similarly, in a chain store, shirking by a manager in one store can significantly damage the

brand image of the entire chain. These situations arise when efforts exhibit complementarity. In such situations, we

believe that it is a reasonable assumption that the principal desires to make sure that none of the agents shirk,

regardless of the state realization, as the unique equilibrium.

1

We characterize the optimal bonus scheme and optimal information structure that implements a “full‐effort”

profile, such as the unique equilibrium. It is obvious that different bonus schemes affect the agents’incentives

differently; however it should be explained how an information structure affects the agents’incentives. Imagine

that the principal can decide which agent (none, only one, or both) can observe the realization of the state, at the