ENTRANTS' REPUTATION AND INDUSTRY DYNAMICS

• DOI: http://doi.org/10.1111/iere.12226
• Date: 01 May 2017
• Author: Felipe Zurita,Bernardita Vial
• Published date: 01 May 2017

INTERNATIONAL ECONOMIC REVIEW

Vol. 58, No. 2, May 2017

ENTRANTS’ REPUTATION AND INDUSTRY DYNAMICS∗

BYBERNARDITA VIAL AND FELIPE ZURITA1

Ponticia Universidad Cat´

olica de Chile, Chile

This article analyzes entry–exit decisions in a market where reputation determines the price that firms

may charge, within a rational-expectation model of competition in a nonatomic market under heterogeneous

reputations. The analysis focuses on the class of name-switching reputational equilibria, in which a firm discards

its name if and only if its reputation falls below the entrants’ reputation. The main technical result is the

existence of a unique steady-state equilibrium within this class, in which the entrants’ reputation is endogenous.

The resulting industry dynamics is largely on agreement with the findings in the empirical literature.

1. INTRODUCTION

It has long been recognized that consumer trust, or reputation, constitutes a valuable—even

determinant—asset to a firm (Tadelis, 1999). Firms with good names are able to charge higher

prices, expand more easily, and live longer. Conversely, a bad name may be an obstacle to firm

survival and expansion (Cabral and Hortacsu, 2010), and, as such, it may be worth changing

(McDevitt, 2011, Wu, 2010).

This article presents a theoretical model of a competitive market in which reputation is the

driving force behind entry, exit, name changing, and pricing decisions. Firms with better repu-

tations charge higher prices; entrants, in turn, are willing to sell below cost with the expectation

that building a reputation will allow them to recover those initial losses. When reputation

matters, free entry has an additional consequence besides keeping prices low; namely, if the

entrants’ reputation is high enough some existing firms may wish to discard their names by

exiting the market and reenter under a new name. The explicit consideration of this option is

the distinctive feature of our analysis.

The model predicts that older firms have stochastically larger reputations than new firms, that

their exit rates are smaller, and their prices higher as well. At the industry-wide level, it predicts

that industries in which ability can be lost more easily will have higher entry-level reputations.

Yet, higher entry-level reputations will not necessarily translate into higher turnover ratios.

The predominant explanation for firm dynamics in the literature is based on technological

shocks. Firms exit either because others drove them out through product and process innovation,

as in the creative destruction hypothesis (see the seminal paper by Hopenhayn, 1992), or because

∗Manuscript received January 2014; revised August 2015.

1We are grateful to the editor Hanming Fang and the three anonymous referees for their careful revision

and insightful comments. We are also grateful to Axel Anderson, Lu´

ıs Cabral, Francesc Dilme, Juan Dubra,

Mehmet Ekmekci, Juan Escobar, Hugo Hopenhayn, David K. Levine, C´

esar Martinelli, Steve Thompson, Ro-

drigo Wagner, and Federico Weinschelbaum, to conference participants at the Association for Public Economic

Theory, the European and Latin American meetings of the Econometric Society, the European Association for

Research in Industrial Economics annual conference, International Industrial Organization Conference, Interna-

tional Workshop of the Game Theory Society, Jornadas de Economa del Banco Central del Uruguay, Jornadas Lati-

noamericanas de Teor´

ıa Econ´

omica, Midwest Economic Theory Meetings, Sociedad de Econom´

ıa de Chile,SING

conference, World Congress of the Game Theory Society, and seminar participants at Universidad de Santiago,

Universidad de Chile,andPonticia Universidad Cat´

olica de Chile. Financial support from FONDECYT grant

#1121096 is gratefully acknowledged. Please address correspondence to: Felipe Zurita, Instituto de Econom´

ıa,

Pontificia Universidad Cat´

olica de Chile, Avda. Libertador Bernardo O’Higgins 340, Santiago, Chile. E-mail:

fzurita@uc.cl.

529

C

(2017) by the Economics Department of the University of Pennsylvania and the Osaka University Institute of Social

and Economic Research Association

530 VIAL AND ZURITA

of adverse shocks that increased their production costs. Although such shocks are certainly a

key factor in the observed dynamics, some evidence suggests that reputation-driven exit is also

at play. For instance, there is evidence that exit is more likely subsequent to poor consumer

reviews or complaints in some industries (Cabral and Hortacsu, 2010; McDevitt, 2011). The

literature also finds that a significant number of firms of different sizes change their names

instead of leaving the market. As changing names is a strategy aimed at affecting consumer

beliefs instead of controlling costs, the interplay between reputation and firm decisions seems

worth exploring. A reputational theory of industry dynamics must also confront the fact that

there is considerable heterogeneity across industries. McDevitt (2011) finds that about 8% of

residential plumbing services firms in Illinois changed names within one year, whereas Wu

(2010) finds that this frequency is much smaller among CRSP-listed companies: about 0.5% per

year on average in the 1925–2000 period.

To address these issues, we develop an adverse selection model with imperfect public moni-

toring along the lines of Mailath and Samuelson (2001). There are two types of firm: competent

and inept. There is perfect competition in the sense of Gretsky et al. (1999): There is a contin-

uum of price-taking firms and a continuum of consumers, and entry is free. The incumbents’

reputation is the Bayesian update of a common prior given an observed history of (imperfect,

public) signals; the entrants’ reputation μEis also a consistent belief. The equilibrium price

function is increasing in the seller reputation.

In equilibrium, firms exit when their reputations fall below μE. As competent firms obtain

stochastically larger signals than inept firms, their expected present value is higher at each

reputation level. As a result, competent firms always want to participate, either with their old

names or new ones. In the steady state, there may be exit and entry flows even if the industry

as a whole is stagnant. Exit probability is found to depend on firms’ characteristics, like type

and age. In particular, the reputation of competent firms stochastically dominates that of inept

firms, and the reputation of older firms dominates that of younger firms as well. Yet, there is

always heterogeneity both within and between cohorts; in fact, the reputation distributions for

all cohorts have full support.

The entry-level reputation provides an endogenous lower bound on reputations, which is

jointly determined with the steady-state reputation distribution. The support of the reputation

distribution at the trading stage is affected by the entrant’s reputation. Reciprocally, the fraction

of competents among entrants depends on how many competent firms choose to change their

names, which is determined by the reputation distribution. Using a fixed-point argument, we

show that there is a unique pair consisting of a mutually consistent entry-level reputation

and a steady-state reputation distribution. The relationship between the entry-level reputation

and the equilibrium exit rate is nontrivial: A change in a parameter that increases μEmay

also shift the reputation distributions such that firms fall below μEless often, reducing the

equilibrium turnover rate. Thus, treating μEand the reputation distribution as exogenous may

be misleading. The endogenization of the entry-level reputation has proven to be important in

other contexts; for instance, Atakan and Ekmekci (2014) and Jullien and Park (2014) find that

entrants’ reputation affects, respectively, the equilibrium assignment in a search market with

bargaining and the equilibrium announcement strategy in a reputation model with pretrade

communication. On the other hand, μEemerges as a determinant of the equilibrium price level,

as potential entrants that would enjoy a reputation μEif they enter must be indifferent between

entering or staying out.

Firms are subject to the possibility that their types change exogenously. Our analysis shows

that a key determinant of entrants’ reputation is the type-change probability. Indeed, a smaller

probability that a competent firm becomes inept implies a lower entry-level reputation. The

explanation is purely informational: If competence is more persistent, then the fraction of

competents among entrants is reduced since the likelihood that their histories of signals

are bad enough to motivate an exit from the market is lower, as they have probably been

competent—obtaining larger signals than inept firms in a stochastic sense—for a long period

of time.