STRATEGIC ADVERTISING AND DIRECTED SEARCH

• DOI: http://doi.org/10.1111/iere.12235
• Author: Benoit Julien,Pedro Gomis‐Porqueras,Chengsi Wang
• Published date: 01 August 2017
• Date: 01 August 2017

INTERNATIONAL ECONOMIC REVIEW

Vol. 58, No. 3, August 2017

STRATEGIC ADVERTISING AND DIRECTED SEARCH∗

BYPEDRO GOMIS-PORQUERAS,BENOIT JULIEN,AND CHENGSI WANG1

Deakin University, Australia; UNSW Australia, Australia; University of Mannheim, Germany,

and Monash University, Australia

Imperfect observability and costly informative advertising are introduced into a standard directed search

framework. Capacity-constrained sellers send costly advertisements to direct buyers’ uncoordinated search by

specifying their location and terms of trade. We show that the equilibrium advertising intensity is nonmonotonic

in the buyer–seller ratio. In addition, we also find that price posting dominates auctions since both mechanisms

yield the same expected revenue, but the latter results in higher advertising expense. Finally, we find a positive

comovement between market transparency and price for low market tightness when the measure of informed

buyers is endogenous.

1. INTRODUCTION

In many markets, sellers have a limited number of goods (services) to offer, and buyers search

and compete for them. These include the labor market, where firms’ vacancies at one point in

time are limited, the procurement market where governments only announce a small number of

projects to be allocated, the real estate market, where landlords/real estate agents sell indivisible

housing units, and the online dating market, where men and women seek partners, just to name

afew.

How are goods allocated in these markets where search and capacity constraints are relevant?

The literature has emphasized several factors that affect the matching process, such as infor-

mational frictions, trade protocols, search costs, and search intensities.2This article highlights

one particular feature—informative advertising—that informs buyers about trading opportuni-

ties. Such an aspect has not been fully explored in the literature, yet advertising nowadays has

become increasingly important in organizing trades.3We study a setting in which sellers who

face capacity constraints use informative but costly advertising to compete for uncoordinated

buyers.

∗Manuscript received January 2014; revised September 2015.

1We would like to thank the editor and the two anonymous referees for their insightful comments and suggestions.

We also thank Chris Bidner, Inga Deimen, Jan Eeckhout, Pieter Gautier, Thanasis Geromichalos, John Kennes, Philip

Kircher, Stephan Lauermann, Andy Mclennan, George Mailath, Roger Myerson, Andras Niedermayer, Mike Peters,

Guillaume Roger, Nicolas Schutz, Sandro Shelegia, Ron Stauber, Peter Vida, Liang Wang, John Wooders, Charles

Zheng, and seminar participants at Australian National University, University of New South Wales, University of Essex,

University of Hawaii, University of Queensland, University of Technology Sydney, University of Western Ontario,

and ESAM 2011 (Adelaide), EARIE 2012 (Rome), SAET 2012 (Brisbane), Summer Workshop on Money, Banking,

Payments and Finance 2012 (Chicago), SFB/TR15 internal meeting 2013 (Bonn), and EEA 2015 (Mannheim) for helpful

input. Chengsi Wang gratefully acknowledges research funding from MaCCI and Deutsche Forschungsgemeinschaft

through SFB-TR 15. The usual disclaimer applies. Please address correspondence to: Chengsi Wang, University of

Mannheim, L7, 3–5, 68131 Mannheim, Germany. Email: chewang@mail.uni-mannheim.de.

2We refer to Burdett et al. (2001) and Julien et al. (2000), among others, for more on this issue.

3For instance, Monster.com, a leading online job-matching Web site, had 63 million monthly jobseekers, posted

over 1 million job advertisements (hereafter ads), and raised $US 996 million in revenue in 2008, mainly obtained from

advertising. Similarly, according to Kantar Media in 2013, restaurants spent $US 6,449 billion on measured media, a

5.2% increase over 2012 spending.

783

C

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

and Economic Research Association

784 GOMIS-PORQUERAS,JULIEN,AND WANG

Unlike most advertising models in which sellers can supply the entire market, we study sellers

who have binding capacity constraints.4More precisely, we build on a directed search frame-

work where capacity-constrained sellers attract uncoordinated buyers through price posting or

auctions, as in Burdett et al. (2001) and Julien et al. (2000), respectively. Sellers send costly

advertisements (hereafter ads) that contain information regarding their location and terms of

trade. As in Butters (1977), ads’ reach is stochastic so that ex ante homogeneous buyers en-

dogenously become heterogeneous after ads have been sent.5Since buyers are uncoordinated,

they have to select strategically which seller, among the ones they are aware of, to trade in order

to avoid rationing. As advertising intensities affect how many competitor they will face, buyers

generally care about how many ads each seller sends and have to form beliefs about advertising

intensities.

We characterize the unique symmetric equilibrium with positive advertising intensity and

determine the equilibrium advertising patterns that emerge in large frictional markets. We

also derive testable implications regarding the relationships between advertising intensities and

market fundamentals such as market tightness and trade protocols. Finally, we explore the

relationship between the measure of informed sellers and equilibrium prices.

We find that the equilibrium advertising intensity has an inverted U-shape with respect to

market tightness. This is the case as each seller only needs a few buyers to sell her inventory. They

have weaker incentives to advertise when there are too few or too many buyers. In particular,

with a small buyer–seller ratio, fierce competition among sellers weakens sellers’ incentives to

advertise. Similarly, with a large buyer–seller ratio, a small amount of ads almost guarantee

sales, and therefore sellers do not advertise much either. Thus, the equilibrium advertising

intensity reaches its maximum when the number of buyers and sellers is relatively balanced.

Other than market tightness, even when dealing with large markets, we find that the specific

terms of trade are critical in determining equilibrium advertising intensity. To this end, we

compare the equilibrium that emerges under auctions and price posting. We find that for any

market tightness, the equilibrium advertising intensity under auctions is always higher. This

is because, under auction, sellers have the opportunity to exploit ex- post surpluses whenever

they meet multiple buyers. Thus, sellers using auctions have stronger incentives to advertise.

Sellers’ expected revenues, however, are equal regardless of the trade protocol they use. As

a result, sellers are thus worse off with auctions, as they incur higher advertising costs. We

can then conclude that price posting is a superior trade protocol in large markets whenever

sellers face costly and imperfect advertising. Finally, we study the comovement between market

transparency and price. Even when the measure of informed buyers is endogenous and buyers

and sellers trade in large markets, we find a positive comovement whenever the market tightness

is low.

The rest of the article is organized as follows: We first review the related literature. Section

2 describes the model setting. Section 3 characterizes the competitive search equilibrium un-

der auctions. Section 4 characterizes the competitive search equilibrium under price posting.

Section 5 discusses various implications of our results. Section 6 considers alternative specifica-

tions of buyers’ beliefs and market structures. Section 7 concludes.

1.1. Related Literature. The main ingredients of our framework are capacity constrained

sellers and informative advertising. The literature has analyzed both of these features but not

jointly. Our exercise to study them jointly not only captures important features of relevant

markets, such as labor and procurement markets, but also helps us understand the role that

capacity constraints have on strategic advertising. Thus, our article complements the literature

both on informative advertising and directed search.

4We refer the reader to Bagwell (2007) for a comprehensive survey on advertising.

5In the context of homogenous goods markets, Butters (1977) formalizes the idea that advertising reach is both

costly and stochastic.