Learning quality through prices and word‐of‐mouth communication

• Author: Carla Guadalupi
• Date: 01 March 2018
• DOI: http://doi.org/10.1111/jems.12230
• Published date: 01 March 2018

Received: 8 January 2016 Revised: 31 July 2017 Accepted: 23 August 2017

DOI: 10.1111/jems.12230

ORIGINAL ARTICLE

Learning quality through prices and word-of-mouth

communication

Carla Guadalupi

Business Department, Universitat de les

Illes Balears, Palma de Mallorca, Spain

(Email: carla.guadalupi@uib.es)

Iwould like to thank Anna Bayona, Luis

Cabral,Gonzalo Cister nas, Juan PabloMon-

tero,Carlos Ponce, and Carlos Serrano for

helpfulcomments; seminar par ticipants at

the International Workshopon Game Theory

(SãoPaulo), SECHI (Viña del Mar), LACEA-

LAMES(São Paulo), Industrial Organization

Workshop(Universidad de Chile, Santiago),

andespecially Nicolás Figueroa, Joaquín

Poblete,and Gastòn Llanes for discussions

andsuggestions. Special thanks are given to

Conicytfor the financial support in the form of

DoctoralScholarship.

Abstract

This paper studies the effect of word-of-mouth communication on the optimal pric-

ing strategy for new experience goods. I consider a dynamic monopoly model with

asymmetric information about product quality, in which consumers learn in equilib-

rium from both prices and other consumers. The main result is that word-of-mouth

communication is essential for the existence of separating equilibria, wherein the

high-quality monopolist signals high quality through a low introductory price (lower

than the monopoly price), and the low-quality one charges the monopoly price. The

intuition is simple: low prices are costly, and will only be used by firms confident

enough that increased experimentation (and therefore communication among con-

sumers) will yield good news about quality and increased future profits. Additional

results are the following: for the high-quality seller, the expected price (quantity) is

increasing (decreasing) over time; whereas for the low-quality one, the opposite is

true. Moreover, signaling becomes more difficult when consumers pay less attention

to their peers' reports and more attention to past prices. Finally, word-of-mouth com-

munication improves consumer welfare.

1INTRODUCTION

Word-of-mouth (WOM) communication is a powerful instrument to disseminate product information prior to purchase. When

consumers face uncertainty about the quality of a new product, they are able to learn from their peers and, specifically, tend to

pay attention to what their fellow consumers observed after their ownpurchases (Ar ndt, 1967; Chen & Xie, 2008; Zhu & Zhang,

2010). This information, communicated via WOM or eWOM,1is likely to be more credible, and thereforemore persuasive than

advertising or other seller-created information, because it is based on personal experience and usage (Bughin, Doogan, & Vetvik,

2010; Dellarocas, 2003; Mayzlin, 2006). As such, WOM can be an important component of a firm's marketing strategy,and may

also act as a substitute for advertising, as it promises more impact than any other communication channel (Chen & Xie, 2008;

Godes & Mayzlin, 2004). After all, “Marketers may spend millions of dollars on elaborately conceived advertising campaigns,

yet often what really makes up a consumer' s mind is not only simple but also free: a word-of-mouth recommendation from a

trusted source” (Bughin et al., 2010).

Firms understand the importance of WOM and use it strategically. When firms are better informed about the characteristics

and quality of the product, prices may also provide valuable information to consumers, as they can be used as signals of quality

(Bagwell & Riordan, 1991; Milgrom & Roberts, 1986). Thus, prices become a signaling instrument to encourage or discourage

experimentation and thereby control the diffusion of information. In particular, low prices that generate WOM buzz can signal

high product quality. Low prices (and higher sales) lead to increased product exposure,which amplifies the good (or bad) news

generated through WOM communication.

J Econ Manage Strat. 2018;27:53–70. © 2017 WileyPeriodicals, Inc. 53wileyonlinelibrary.com/journal/jems

54 JOURNAL OF ECONOMICS & MANAGEMENTSTRATEGY

This is particularly true in the case of experience goods, whose quality is unobserved at the time of purchasing, and only

becomes known after consuming and “tasting” the product.2Travelers, for example, would be remiss to reserve an Airbnb

property without first consulting online reviews to ensure that previous guests confirm the host-reported quality. For their part,

hosts would be equally remiss not to incorporate this fact into the pricing strategy. Here, eWOM can bridge the information

gap between past consumers and potential consumers and help turn views into purchases. Indeed, Aerosolve, the dynamic open-

source algorithm used by Airbnb to provide hosts with date-specific price suggestions, prioritizes the capture of a first and second

review via penetration pricing. The fact that a property has no reviews is a negative “feature”, which puts significant downward

pressure on the suggested price. Even in a machine-learning environment (as in the Airbnb case), prices provide a signal from

which consumers can infer product quality, even as they influence the rate of information transmission via interpersonal com-

munication.

Empirical evidence shows the positive effect of eWOM on sales for experience goods3(Chevalier & Mayzlin, 2006; Cui,

Lui, & Guo, 2006; Duan, Gu, & Whinston, 2008; Liu, 2006; Lu, Ba, Huang, & Feng, 2013). Moreover, eWOM is shown to

be particularly influential for recently introduced goods: the volume of reviews has a significant effect on new product sales in

the early period and decreases over time (Liu, 2006; Zhu & Zhang, 2010). On the other hand, extensive economic literature has

explored the optimal pricing strategy and dynamics for new products of initially uncertain quality.Nevertheless, most signaling

models have ignored learning from others and the question of whether and how interpersonal communication across consumers

affects or is affected by prices. It is therefore interesting to study how a high-quality seller can take strategic advantage of

experimentation and WOM when choosing his pricing strategy. This paper addresses these questions by providing a simple

model in which WOM communication is combined with strategic pricing and signaling.

Specifically, I consider a two-period model, in which a long-livedmonopolist introduces a new experience good of unknown

quality to consumers. The monopolist is privately informed about his type (either high or low), and a high-type seller is more

likely to produce high-quality products. Consumers are short-lived and the satisfaction they receivefrom buying is a function of

both objective quality and personal preference (or taste). First-period consumers can only relyon pr ices to inferthe monopolist's

type (and therefore expected quality).Second-per iod consumers, on the other hand, learn through the observation of noisy signals

of past prices and observed quality. Notably, past prices are not directly observable, as public price information usually refers to

the posted price, which does not accurately reflect the final transaction price because sales, discounts, and/or geographic price

variations are not taken into account. Furthermore, consumers have been shown to have imprecise memories when it comes to

the final price paid for the product.4Still, past consumers report through WOM communication the satisfaction obtained from

the product, if purchased. Reports from past consumers are noisy signals of the observed quality as higher satisfaction can come

either from high quality or from strong personal preference. Moreover, communication between past and present consumers

is incomplete because social interaction remains limited by time, geography, and other constraint.5Nevertheless, information

gathered through WOM communication is more precise when the fraction of consumers who bought in the past is high.

I focus on the role of prices as signals of quality when consumers learn in equilibrium through both prices and WOM com-

munication. First, I characterize the benchmark case without WOM communication among consumers, and show that there is

no separating equilibrium. I then show that, when allowing for WOM communication, there exists a separating equilibrium in

which prices signal quality. Given a sample information structure, I study how the separating equilibrium responds to variation

in the signal precision and compare consumer welfare in the benchmark and WOM case. Finally, I present model extensions.

The main result is that WOM communication is essential fort he existenceof separating equilibria. Moreover, in any separating

equilibrium, the high-quality monopolist signals quality through a low introductory price (lower than the monopoly price).

The intuition is simple: low prices are costly, and will only be used by firms confident enough that increased experimentation

(therefore communication among consumers) will yield good news about quality and increased future profits. The low-quality

monopolist, on the other hand, avoids information transmission by choosing a higher price (the monopoly price), which limits

sales and the possibility of learning through WOM. An additional implication: for the high-quality seller, the expected price

(quantity) is increasing (decreasing) over time; whereas for the low-quality one, the opposite is true. Since the second-period

price is a function of beliefs, it will be higher (lower) if good (bad) news is revealed in the first period when the product

is introduced. Moreover, the (low) separating price charged by the high-quality monopolist in equilibrium decreases as the

impact of signaling on consumers' beliefs increases: a high-quality seller must work harder to discourage the low-quality one

from mimicking its behavior if consumers pay less attention to their peers' reports. Finally, consumers are better off with WOM

communication. As a robustness check, I showthat WOM communication and advertising can be either substitute or complement

instruments for the firm. In particular, when allowed to use WOM, the firm will never use “money-burning” advertising as a

signal of quality. On the contrary, when advertising is demand-enhancing, firms might use both instruments to signal quality.

Finally, the equilibrium is robust to reporting bias by first-period consumers.