When Does a Platform Create Value by Limiting Choice?

• DOI: http://doi.org/10.1111/jems.12052
• Published date: 01 June 2014
• Author: Ramon Casadesus‐Masanell,Hanna Hałaburda
• Date: 01 June 2014

When Does a Platform Create Value by Limiting

Choice?

RAMON CASADESUS-MASANELL

Harvard Business School, Strategy Unit

Boston, MA 02163

casadesus@gmail.com

HANNA HAŁABURDA

Harvard Business School, Strategy Unit

Boston, MA 02163

hhalaburda@gmail.com

We present a theory for why it might be rational for a platform to limit the number of applications

available on it. Our model is based on the observation that even if users prefer application variety,

applications often also exhibit direct network effects. When there are direct network effects, users

prefer to consume the same applications to benefit from consumption complementarities. We

show that the combination of preference for variety and consumption complementarities gives

rise to (i) a commons problem (to better satisfy their individual preference for variety, users have

an incentive to consume more applications than the number that maximizes joint utility); (ii) an

equilibrium selection problem (consumption complementarities often lead to multiple equilibria,

which result in different utility levels for the users); and (iii) a coordination problem (lacking

perfect foresight, it is unlikely that users will end up buying the same set of applications). The

analysis shows that the platform can resolve these problems and create value by limiting the

number of applications available. By limiting choice, the platform may create new equilibria

(including the allocation that maximizes users’ utility); eliminate equilibria that give lower

utility to the users; and reduce the severity of the coordination problem faced by users.

1. Introduction

Platforms such as computer operating systems (Windows), video game systems (Nin-

tendo), betting exchanges (Betfair), stock exchanges (NYSE), or online gaming sites

(Kaixin001) are institutions that facilitate users’ access to applications (defined as op-

portunities to fulfill users’ particular purposes—such as writing documents, playing

We thank two anonymous referees and the coeditor for helpful comments that have helped improve

the paper. We also thank Kevin Boudreau, Kenneth Corts, Jim Dana, Eric Darmon, Chris Dellarocas,

Nicholas Economides, Joshua Gans, Andres Hervas-Drane, Doh-Shin Jeon, Elon Kohlberg, Matt Mitchell, Jo˜

ao

Montez, Emre Ozdenoren,Al Roth, Mike Ryall, Catherine Tucker,Dennis Yao, Pai-Ling Yin, Feng Zhu, seminar

participants at HEC Paris, the Second Annual Searle Center Conference on Internet Search and Innovation,

IIOC 2011, INSEAD, Wharton, Universitat Aut`

onoma de Barcelona, HBS Strategy Unit Research Day and

Brown Bag seminars, MIT IO lunch, NET Institute conference, TorontoRotman, London Business School, the

2011 North American Summer Meetings of the Econometric Society in St. Louis, the 2011 European Sum-

mer Meetings of the Econometric Society in Oslo, the Second Annual Searle Center Conference on Internet

Search and Innovation in Chicago, and the 9th ZEW Conference: The Economics of Information and Com-

munication Technologies, Mannheim. We also thank Adrianna Lohnes for expert assistance. Yusuke Norita

provided excellent research assistance. Wegratefully acknowledge financial support from the NET Institute

(www.netinst.org)and the HBS Division of Research and Faculty Development.

C2014 Wiley Periodicals, Inc.

Journal of Economics & Management Strategy, Volume23, Number 2, Summer 2014, 259–293

260 Journal of Economics & Management Strategy

games, betting money, or investing capital).1Among the many governance choices that

platforms make, they determine the number of applications users have access to (e.g.,

how many games to offer by a given online gaming platform, how many firms to list by

a given stock exchange, and so on). In this paper, we study the relationship between the

number of applications available on a platform and users’ equilibrium utility. We find

that narrow choice often increases utility and thus creates value.

Platforms are characterized by the presence of indirect network effects: the larger

the number of users the more firms are willing to join, thus increasing the diversity of

applications available, which in turn raises users’ valuation of the platform. For example,

firms’ desire to list their shares in the New York Stock Exchange grows with the number

of investors who are expected to trade there; likewise, the larger the number of firms

expected to be listed in the NYSE, the more willing the investors are to invest there

(Cantillon and Yin 2011). Naturally, indirect network effects induced by users’ preference

for application variety have played a prominent role in models of platforms, beginning at

least from the pioneering work of Church and Gandal (1992) and Chou and Shy (1996),

and spanning to recent contributions such as Hagiu (2009) and Weyl (2010).

When the value of a platform increases with the number of applications offered,

common wisdom dictates that platforms should provide as many applications as possi-

ble. Indeed, suboptimal exploitation of indirect network effects may have dreadful con-

sequences; superior platforms (better technology,better capitalized, early movers...) may

perish in their competition against second-rate alternatives. Arthur (1990), for example,

describes how Sony lost its battle against JVC in the 1980s whose VHS standard was in-

ferior to Betamax, due largely to lesser movie availability on Sony’s standard. Likewise,

it is widely believed that Apple lost its battle against the PC in the late 1980s because

of a dearth of applications. Although Microsoft aggressively evangelized independent

software vendors and provided them with tools and support, Apple based its approach

on in-house development of a small number of applications. By the early 1990s, the

number of applications available for the Mac was a small fraction to that for the PC.

Given the wealth of evidence suggesting that maximizing application variety is a

good idea, it is puzzling that successful platform providers such as Betfair, Nintendo,

or Kaixin001 appear to have actively limited the number of applications available on

their platforms. Betfair provides an electronic platform that allows its customers to back

teams to win in sports such as soccer or horse races, but also to lay odds for others

to bet on. The company began operations in the United Kingdom in 2000 as a second

mover after Flutter.com. Although Flutter was the first mover and had better access

to capital—its initial funding was $43.7 million versus £1 million for Betfair—Betfair

won over the market.2A key difference between the two betting exchanges was that

while Flutter would allow users to bet on any event they wished to create (such as next

week’s weather), Betfair adamantly restricted the number of events (applications) on

which users could bet. Interestingly,the platform that offered fewer applications ended

up faring better.

Similarly, in the late 1980s, Nintendo restricted the number of games that devel-

opers were allowed to release each year for the Nintendo Entertainment System (NES)

to five. The company also restricted the number of developers who could sell games for

1. Examples of applications include: word processors or spreadsheet programs (in the case of computer

operating systems), games (in video game systems or online gaming sites), sports events (in betting exchanges),

and listed companies (in stock exchanges).

2. Betfair acquired Flutter in December 2001 and become the dominant betting exchange in Europe. See

Casadesus-Masanell and Campbell (2008).

When Does a Platform Create Value by Limiting Choice? 261

the NES. Nintendo went on to become the dominant player (market share and profit)

for the 8-bit generation.3Likewise, the leading online social networking site, Kaixin001,

provides a limited number of games for users to engage in (e.g., Parking Cars and Steal-

ing Crops) when many more could be offered. The site offers the smallest number of

social games among the top social networking sites in China and lags behind its com-

petitors in making its platform open to third party application developers;4however,

the site has the most highly active users among them.5These examples run counter to

the conventional wisdom that when considering application variety in platforms “more

is always better.”

In this paper, we ask: why might it be rational for a platform to limit the number

of applications when indirect network effects are at play? Our answer is that by limiting

the number of applications the platform may resolve three problems faced by users: a

commons problem, an equilibrium selection problem, and a coordination problem. When

the platform resolves these three problems it creates value because users achieve higher

utility. Thus, our analysis focuses on how limiting the number of available applications

affects the value created.6

Our theory is based on the observation that even when platforms enjoy indirect

network effects, applications often exhibit direct network effects, that is, users are better

off using the same applications as other users due to consumption complementarities. For

example, Cantillon and Yin (2010) demonstrate that there are important direct network

effects in derivatives’ trading. Specifically, as the number of traders for a particular

derivative increases, so does liquidity. Similarly, the richness of gameplay in massively

multiplayer online games (MMOG), such as World of Warcraft, is based on the number

of interactions between players; MMOGs are not fun if played alone. When users have

limited resources (such as finite time to enjoy applications or an income constraint) and

there are many applications available, they must pick and choose which ones to use. If

direct network effects are at play, users are better off by purchasing and consuming the

same limited set of applications.

We show that when users prefer application variety but also benefit from con-

sumption complementarities, three issues may arise. First, the number of applications

that maximizes users’ utility may not be part of an equilibrium as each user may find it

optimal to unilaterally deviate to consume more applications so as to better satisfy her

craving for variety. This is because agents do not internalize the negative externalities

they impose on others: a single user suffers almost nothing from the decrease in con-

sumption complementarities when he increases the number of applications consumed

while, in aggregate, the loss of consumption complementarities is much greater. Sec-

ond, multiple equilibria often arise. With the usual assumption that users have perfect

foresight, any one of those equilibria could, in principle, be selected. Although some

equilibria lead to higher user utility than others, nothing guarantees that the equilib-

rium yielding the highest utility will be selected. Third, if users lack perfect foresight on

each others’ choices in equilibrium, it is unlikely that they will end up purchasing and

3. The NES was the leading second-generation (8-bit) game console. Nintendo’s global market share for

8-bit consoles in 1990 was greater than 90%. See Brandenburger (1995).

4. http://www.nth-wave.com/wordpress/?p=32985.

5. http://www.nytimes.com/external/venturebeat/2010/04/07/07venturebeat-chinas-top-four-social-

networks-renren-kaixi-55248.html.

6. We do not consider how the platform may capture that value through prices (access prices and/or

royalties). We note, however, that generating larger user surplus allows the platform to charge higher access

fees to users. A detailed analysis of platform pricing (i.e., value capture) is beyond the scope of this paper and

left for future research. See Section 4 for a more detailed discussion of this issue.