Persistent and snap decision‐making

• DOI: http://doi.org/10.1111/jems.12395
• Published date: 01 February 2021
• Author: Tomoya Tajika
• Date: 01 February 2021

J Econ Manage Strat. 2021;30:203–227. wileyonlinelibrary.com/journal/jems © 2020 Wiley Periodicals LLC

|

203

Received: 24 June 2018

|

Revised: 8 July 2020

|

Accepted: 8 July 2020

DOI: 10.1111/jems.12395

ORIGINAL ARTICLE

Persistent and snap decision‐making

Tomoya Tajika

Department of Law and Economics,

Hokusei Gakuen University, Oyachi‐nishi

2‐3‐1, Sapporo, Japan

Correspondence

Tomoya Tajika, Department of Law and

Economics, Hokusei Gakuen University,

Oyachi‐nishi 2‐3‐1, Sapporo 004‐8631,

Japan.

Email: t-tajika@hokusei.ac.jp

Abstract

To avoid unfavorable inferences about her ability, an expert might cling to her

original opinion and ignore valuable new information in formulating sub-

sequent opinions. Conceivably, the expert might decline an initial opportunity

to offer an opinion, delaying the opinion formation until more accurate in-

formation has arrived. However, we show that reputational concerns often

lead an expert to express an opinion at the first opportunity, thereby making a

snap decision.

1|INTRODUCTION

It is widely observed that experts often cling to their original opinion and that such consistency is valued by others.

1

For

example, when a politician makes a statement, it is often seen as a pledge and it is difficult for her to renege on it

without being subject to criticism. Likewise, economists who predict economic trends that are inconsistent with their

previous predictions tend not to be trusted. However, remaining consistent could lead an expert to ignore valuable new

information in formulating subsequent opinions, which often results in bad outcomes.

Many studies examine reputation concern as a force driving consistent behavior (Falk & Zimmermann, 2017;

Li, 2007; Trueman, 1990). In these studies, an expert recommends a choice to an evaluator. The expert has two

opportunities to make a recommendation and, before each opportunity, she receives an independent signal conditional

on the true state of the world. The accuracy of the signals positively correlates with the expert's ability, and this

accuracy increases over time, such that the second signal is more accurate than the first. The expert has reputational

concerns about her ability. In this setting, although a recommendation based on the second signal is preferable for the

evaluator, the expert has an incentive to cling to the recommendation based on the first. This is because an expert with

high abilities would receive highly accurate signals and is less likely to receive inconsistent signals. Hence, the expert

behaves consistently to convey the impression that she received consistent signals because she has high abilities.

Although this explanation is intuitive, one may argue that if consistent behavior is inevitable and leads to a less

valuable recommendation, the expert would remain silent during the initial opportunity to avoid revealing any in-

consistency between the received signals. If this is accurate, unless the expert is forced to make a decision at the initial

opportunity, such consistent behavior resulting in less valuable recommendations would not be observed. In the leading

example, the politician should be careful not to commit herself in a speech because she is better off waiting for

additional information, rather than declaring an intention immediately. However, in reality, people often make hasty

decisions. Nutt (2002) discusses how hasty decision‐making is a major cause of poor decisions using several case studies

of companies. Consider the following case as an example. Shell UK decided to undertake a deep‐sea disposal of a spar (a

floating oil storage facility). Before the disposal process began, Greenpeace activists began a public campaign opposing

it, which damaged Shell UK's public image, and forced it to find another way to dispose of the spar. As Nutt (2002)

discusses, the Shell officials did not need to rush in making their decision and could have waited until a public debate

about the best disposal method had occurred.

In this study, we seek to answer the question of why experts convey their earlier opinions rather than waiting for

new information, even when it is well known that this leads them to seek unproductive consistency of behavior, which,

in turn, leads to bad outcomes. In our model, as in previous studies, the expert has two opportunities to make a

recommendation. However, this study takes a further step by examining a case where the expert has two options during

the first opportunity; that is, the expert can either reveal or conceal the information about the signal that she has

received at the initial opportunity. We find a sufficient condition under which the expert has an incentive to reveal the

information during the first opportunity, even if it leads to unproductively consistent behavior. Therefore, even when

the expert is not forced to make a decision at the initial opportunity, consistent behavior can be observed. We refer to

the behavior of breaking silence during the first opportunity and ensuring consistent behavior subsequently as snap

decision‐making.

The intuition is as follows. The expert considers that by remaining silent during the first period, the evaluator will

believe that the expert's final decision is based on the second signal. Conversely, if the expert makes a decision during

the first period and employs the strategy of behaving consistently, the evaluator will believe that the expert's decision is

based on the first signal. If the expert waits for a more accurate signal and matches her recommendation to the true

state, this will be attributed partly to the improvement of the signal and less to the expert's ability than is the case with

acting on the first signal. By contrast, if the recommendation does not match the state when the expert acts on the

second signal, this is attributed to ability more than is the case with the first signal. Therefore, if the expert remains

silent, her ability is discounted regardless of whether her recommendation matches the state. This reveals the advantage

of snap decision‐making.

By contrast, if the expert remains silent initially, her ultimate recommendation is more likely to match the true state

than is the case under snap decision‐making. Because the expert's reputation is higher when her recommendation

matches the state, this indicates the disadvantage of snap decision‐making.

In our model, we find that under symmetric information between the expert and evaluator concerning the expert's

ability, if the expert's utility function for the assessment is strictly convex

2

and some additional assumptions are

satisfied, an equilibrium with snap decision‐making exists, which results in higher expected utility than does remaining

silent. Conversely, there is also an equilibrium where the expert reveals no information about the first signal, and

makes an ultimate recommendation based only on the second signal.

3

However, making a snap decision results in

higher utility than does remaining silent and, thus, there is an incentive to reveal the first signal if this is possible. This

implies that the equilibrium revealing no information about the first signal violates a natural equilibrium selection

criterion, called the neologism‐proofness (Farrell, 1993). In this sense, the snap decision is the unique prediction of our

model.

Our study suggests that, focusing on the neologism‐proofequilibria,evenwhentheexpertcanremainsilent,

she may decide to engage in unnecessary snap decision‐making, leading to a poor decision. To avoid such

unproductive recommendations, our study suggests that the client should not only refrain from listening to

earlier opinions but also silence the expert during the first period by imposing a cost. For example, in the context

of decision‐making in organizations, creating a norm that encourages deliberation may be a solution. In political

contexts, electorates should criticize politicians if they comment on a future political decision without waiting

for new information. Conversely, an intentionally noncommittal (and thus ambiguous) statement should be

praised.

The remainder of the paper is organized as follows. Section 2reviews the related literature. In Section 3, we explain

our model. Section 4addresses the case where the expert is forced to recommend a choice in each period and provides a

sufficient condition under which the expert behaves consistently. In Section 5, we analyze the case in which the expert

can remain silent in the first period. Section 6includes several extensions for analyzing the consistent behavior, and

Section 7draws conclusions. All proofs appear in the appendices.

2|RELATIONSHIP TO THE LITERATURE

Reputation concern distorts an expert's decision‐making.

4

The relation between reputation concern and persistency in

behaviors has been investigated by several studies, such as Kanodia, Bushman, and Dickhaut (1989), Trueman (1990),

Boot (1992), Prendergast and Stole (1996), Levy (2004), Majumdar and Mukand (2004), Ferreira and Resende (2007), Li

(2007), Sabourian and Sibert (2009), and Falk and Zimmermann (2017). Except for Prendergast and Stole (1996), these

studies deal with binary‐state cases, as in our model. Prendergast and Stole (1996) deal with a continuous state case in

which the true state and signals are drawn from normal distributions. They show that the reputation‐concerned expert

becomes conservative over time.

5

204

|

TAJIKA