Poaching in media: Harm to subscribers?

• Date: 01 June 2018
• Author: Elias Carroni
• DOI: http://doi.org/10.1111/jems.12238
• Published date: 01 June 2018

Received: 13 September 2016 Revised: 9 June 2017 Accepted: 9 November 2017

DOI: 10.1111/jems.12238

ORIGINAL ARTICLE

Poaching in media: Harm to subscribers?

Elias Carroni

Dipartimento di Scienze Economiche Alma

Mater Studiorum Università di Bologna 2,

piazza Antonino Scaravilli, 40126 - Bologna,

Italy

Abstract

Two media platforms compete for heterogeneous users bothered by commercials and

sell advertising spaces to firms. In a two-period model, media are allowed to condi-

tion subscription prices on the past behavior of users. Within-group price discrimina-

tion intensifies media competition on the firms' side, as some firms advertise only on

one media outlet (single-home), where they can meet early users and switchers. As a

consequence, advertising revenues are reduced and this puts an upward pressure on

subscription prices. However,pr ice discrimination also induces stronger within-group

competition to poach the rival's users. Depending on the balance between these two

forces, conditioning subscription prices on past behavior might be beneficial or detri-

mental to users, whereas it is always detrimental to platforms. In relation to within-

group uniform pricing, total welfare might increase or decrease, as the lower adver-

tising intensity may entail either underprovision or a mitigation of overprovision of

advertisements.

1INTRODUCTION

When a firm knows the identity of its customers, it often decides to charge new clients with a lower price in order to capture

new demand. There is strong evidence of the fact that this strategy is used in the market of media subscriptions. As pointed

out by Taylor (2003), price discrimination based on past purchases, called behavior-basedpr ice discrimination (BBPD), is very

common in subscription markets. In these markets, because transactions are never anonymous, a firm knows the identity of

current subscribers and can thus propose discounts to those who did not subscribe in the past. Discounts take different forms

such as low introductory prices and free trial memberships. More specific to media, Caillaud and Nijs (2014) report how a new

subscriber for three months to the French newspaper, “Le Monde,” pays50 euros whereas a previous customer is charged 131.30

euros. Similar offers can be found in many traditional media (TVs, newspapers, and magazines) as well as on online platforms

such as Spotify and Deezer, which offer free trial memberships to access their contents.

These strategies have captured the attention of many economists.1Theirmain concer n has been the study of the consequences

of such practices on firms' profits, consumer surplus, and price levels. Roughly speaking, there is a consensus on the conclusion

that BBPD reduces firms' profitability, as they compete fiercely to poach the rival's customers, with a consequent benefit for

consumers in relation to uniform pricing. This conclusion is quite important from a policy viewpoint. Indeed, the access of firms

to data on consumers is something that, from a strictly economic perspective, is ultimately positive for consumers themselves.

The idea is that the more information firms possess, the more fiercely they compete, and this goes all at the benefit of consumers.

I am grateful to the editor Ramon Casadesus-Masanell and three anonymous referees for helpful reviewson the paper. I also wish to thank Eric Toulemonde,

Paul Belleflamme, Marc Bourreau, Lidia Carroni, Marco Delogu, Vincenzo Denicolò, Rosa Branca Esteves, LucaFer rari, Andrea Mantovani,Gabr iella Mezei,

Antonio Minniti, Leo Mocciola, Dimitri Paolini, Giuseppe Pignataro, and Simone Righi. I am indebted to the participants to the Doctoral Workshop 2012

at UCLouvain, 2013 Ecore Summer School—Governance and Economic Behavior (Leuven), 3rd GAELConference—Product differentiation and innovation

on related markets (Grenoble). This research was conducted as part of the project Labex MME-DII (ANR11-LBX-0023-01). I acknowledge the “Programma

Master & Back - Regione Autonoma della Sardegna” for financial support. All remaining errors are my own.

J Econ Manage Strat. 2018;27:221–236. © 2017 WileyPeriodicals, Inc. 221wileyonlinelibrary.com/journal/jems

222 JOURNAL OF ECONOMICS & MANAGEMENTSTRATEGY

However,media markets have the peculiarity of advertising. All articles studying BBPD neglect the presence of cross–group

externalities that typically characterize media markets. It turns out that traditional media as well as new—online—media have

the common feature that users (subscribers)2are not the only customers, as their profits also come from the advertisers. In

economic jargon, these markets are run by two-sided platforms allowing the interaction between different groups of customers

linked to each other by cross-group externalities. Namely, the utility that a user enjoys by subscribing to the service decreases

with the number of commercials (nuisance) present in the platform, whereas advertisers are more satisfied if the number of

users increases.

Because of externalities, one of the distinctive features of these markets is the pricing rule, which is different from the gen-

eral rule that applies in a one-sided framework (i.e., market without externalities). Indeed, the subscription price (advertising

intensity) affects not only the demand of subscriptions (advertisement revenues), but also the well-being of advertisers (sub-

scribers) who join the platform. Hence, platforms offer a low (often below-cost) price to the group whose participation entails

a larger (reduced) marginal benefit to the other group, which becomes the profit-making segment.3Hence, media make use of

two different kinds of strategies. On the one hand, they sort customers according to their externalities (i.e., cross-group price

discrimination). On the other hand, once they know the identity and the behavior of the users, they also engage in BBPD within

the group of subscribers.

This paper provides a model of two-sided competing media. On one side of the market, heterogeneous subscribers receive a

utility coming from contents and suffer the presence of advertisements. On the other side, advertisers want to sell their products

and the platforms are a means to reach consumers. In a two-period model, after the first round of subscription decisions, the

platforms are allowed to discriminate users on the basis of their past behavior. The aim is to contribute to the literature of two-

sided media as well as to the one of pricing under customer recognition. On the one hand, the paper shows that subscribers'

switching affects the optimal behavior of firms and ultimately reduces the provision of advertisements. On the other hand, it

demonstrates that, even if platforms are always worse off in the discriminatory case, subscribers might be worse or better off

when they can be recognized and discriminated.

The advertisement is interpreted as in the broadcasting models of Anderson and Coate (2005) and Peitz and Valletti (2008).

Firms pay broadcasters to advertise their products in order to meet potential consumers among viewers. Advertisements are

purely informativeand create some s ocial value,given the fact that people exposed to commercials become aware of the existence

of products they may like. However, wheneversubscr ibers are repeatedlyexposed to the same commercial, there is overprovision

of advertisements. Under uniform subscription pricing, there are no movements of subscribers over time. The users' market is

a competitive bottleneck as in Peitz and Valletti (2008), as firms must advertise their products on a given platform to reach that

platform's subscribers. This leads to multihoming and gives an important power to media outlets, which become monopolistic

in the eyes of advertisers. Differently,under BBPD, the switching of subscribers reduces t his monopolisticpower, as a firm may

prefer to advertise its products only on one platform rather than multihome. Indeed, this choice allows for reaching a sufficient

number of subscribers (early subscribers plus switchers) meanwhile saving on the fee paid. This intensifies media competition

on the advertisers' side, thus reducing the incentives to make profits on that side. Consequently, the two media setan amount of

advertising lower than the one they would have set under uniform pricing.

The mechanisms just discussed allow for distinctive results in terms of subscriber surplus, confirming the fact that the theo-

retical conclusions and the implications offered by two-sided models can be different from the one that one would find using a

one-sided logic.4This is because the analysis of any strategy used on one side of the marketalone does not take into account what

this strategy provokes on the other side. In one-sided oligopolies, price discrimination has often been proven to have a positive

impact on consumer surplus. This is due to the fact that firms are very aggressive in order to attract individuals more inclined to

buy the rival's product.5More specific to the BBPD literature, under repeated purchase, this business-stealing effect often drives

toward a consumer-benefiting outcome. In Chen (1997), Villas-Boas (1999), Fudenberg and Tirole (2000), and Esteves (2010),

the negative consequences on firms profits of late business stealing alwaysoutweigh the (possible) mitigation of early competition

resulting from consumers' anticipation of future discounted prices. Recent articles have demonstrated how BBPD can actually

be (partially) detrimental to consumers. In a context where firms have incomplete information about consumers' purchase histo-

ries, Colombo (2016) shows an inverse U-shaped relationship between consumer surplus and information accuracy. Moreover,

BBPD boosts firms' profits at the detriment of consumers in the presence of a weak over-time correlation between consumers'

preferences (Chen & Pearcy, 2010) or when consumers are sufficiently myopic in anticipating the future (Carroni, 2016).

In the present model, pricing subscribers according to their past behavior causes switching which, in turn, changes the optimal

decisions of advertisers. This finally leads to a different level of advertising intensities at equilibrium, which has an impact on

the final subscription price. In particular, three forces are at play. First, platforms competemore severely in the second period in

order to poach the rival's subscribers. Second, the first-period competition is weakened by the users' anticipation of advantageous

offers they will receive in the switching stage. Finally, on top of these two within-group forces, there is also a cross-group force,