Investment Incentives in Open‐Source and Proprietary Two‐Sided Platforms

• Author: Ramon Casadesus‐Masanell,Gastón Llanes
• DOI: http://doi.org/10.1111/jems.12089
• Published date: 01 June 2015
• Date: 01 June 2015

Investment Incentives in Open-Source

and Proprietary Two-Sided Platforms

RAMON CASADESUS-MASANELL

Harvard Business School,

Strategy Unit Boston, MA 02163

casadesus@gmail.com

GAST ´

ON LLANES

Pontificia Universidad Cat´

olica de Chile,

Santiago, Chile

gaston@llanes.com.ar

We study incentives to invest in platform quality in open-source and proprietary two-sided

platforms. Open platforms have open access, and developers invest to improve the platform.

Proprietary platforms have closed access, and investment is done by the platform owner. We

present five main results. First, open platforms may benefit from limited developer access. Second,

an open platform may lead to higher investment than a proprietary platform. Third, opening one

side of a proprietary platform may lower incentives to invest in platform quality. Fourth, the

structure of access prices of the proprietary platform depends on (i) how changes in the number

of developers affect the incentives to invest in the open platform, and (ii) how investment in the

open platform affects the revenues of the proprietary platform. Finally, a proprietary platform

may benefit from higher investment in the open platform. This result helps to explain why the

owner of a proprietary platform such as Microsoft has chosen to contribute to the development of

Linux.

1. Introduction

Although open-source and proprietary platforms have coexisted since the early days

of the computing industry, competition between these two modes of development has

intensified dramatically following the surge of the Internet in the mid-1990s. Prominent

examples include Linux versus Windows, Open Office versus Microsoft Office, Firefox

versus Safari, Apache versus MS Internet Server, and more recently, Google’s Android

versus Apple’s iOS.

The open-source development model is characterized by two distinctive features:

open access (the freedom to use the software free of charge) and open investment (the

freedom to modify the source code).1Proprietary development, on the other hand, has

We thank two anonymous referees and the Editor for helpful comments that have resulted in a much better

paper. We also thank Jay Pil Choi, Jacques Cr´

emer, Andrei Hagiu, Hanna Hałaburda, Andr´

es Herv´

as-Drane,

Elon Kohlberg, Francisco Ruiz-Aliseda, YaronYehezkel, David Zvilichovsky, and seminar participants at the

Third Annual Searle Center Conferenceon Internet Search and Innovation (Northwestern University), the 39th

Annual Conference of the European Association for Research in Industrial Economics (Rome, Italy), the 7th

Bi-annual Conference on The Economics of Intellectural Property,Software and the Internet (Toulouse School

of Economics), and Pontificia Universidad Cat´

olica de Chile. Casadesus-Masanell gratefully acknowledges

financial support from the HBS Division of Research. Llanes gratefully acknowledges financial support from

FONDECYT Initiation to Research Grant No. 11110043.

1. Open access and open investment are complementary but do not always go hand in hand. For example,

MS Explorer is an open-access program, but it does not allow for open investment as the source code is not

made available to users.

C2015 Wiley Periodicals, Inc.

Journal of Economics & Management Strategy, Volume24, Number 2, Summer 2015, 306–324

Investment Incentives in Two-Sided Platforms 307

closed access and closed investment: the platform owner sets access prices and invests

centrally to improve its quality. The coexistence of these two diametrically opposed

modes of platform governance has sparked a thriving literature on open source ex-

amining why individuals and profit-maximizing firms might choose to contribute to

open-source development (see Lerner and Tirole, 2005; von Krogh and von Hippel,

2006; Fershtman and Gandal, 2011, for recent surveys).

Although insightful and enlightening, theoretical developments on the economics

of open source have fallen short of fully embracing the modeling breakthroughs offered

by the literature on two-sided platforms of the past decade (e.g., Caillaud and Jullien,

2003; Rochet and Tirole, 2003, 2006; Armstrong, 2006; Hagiu, 2006a; Spulber, 2006; Weyl,

2010). Likewise, while the literature on two-sided platforms has studied some aspects of

open platforms, the most distinctive feature of open source (i.e., open investment) has

not been considered.2

In this paper,we bring together these two streams of work to address the following

questions: (i) How are the incentives to invest in platform quality affected by the degree

of platform openness? (ii) What is the relation between access and investment strategies?

and (iii) How are access prices and incentives to invest in platform quality moderated

by competition between open-source and proprietary two-sided platforms?

We set up a model of a platform that brings together users and developers of ap-

plications. Users are heterogeneous in their willingness to pay for access to the platform.

Developers are heterogeneous in that they bear different costs for developing applica-

tions. A proprietary platform chooses how much to invest in platform quality and sets

access prices for each side of the market. An open platform may be accessed for free

and developers may invest in improving its quality.3After users and developers have

accessed the platform, developers compete to sell applications to users. Users prefer

product variety but consider applications as interchangeable. Along with the case of

substitute applications whose marginal value decreases with the number of applications

available, we study the mirror case of complement applications.

Wedivide the analysis into two parts. First, we examine models of proprietary and

open monopoly platforms; that is, we consider incentives to invest in proprietary and

open platforms in isolation from each other and compare equilibrium outcomes. In the

second part, we study a mixed-duopoly model with direct competition between both

types of platforms.

We obtain the following results. First, open platforms may benefit from limited

developer access. The intuition is that the level of investment by developers depends

on the expected profit that these investments generate. If the effect of a change in the

number of applications on developer revenue decreases with investment in platform

quality, a lower number of developers may result in higher investment in the open

platform.

2. To the literature on two-sided platforms, an open platform is one that offers open access, and a pro-

prietary platform is one that has closed access (the platform sets access prices, positive for at least one side).

Thus, this literature is silent about the investment side of platform openness: an open-source platform not

only offers open access but it also allows users and developers to invest in platform quality by modifying the

source code.

3. For concreteness, we focus on developer investment in open platforms and assume away user invest-

ment. Recent empirical evidence suggests that a large share of investments in open source are made by firms

rather than by users. For example, a 2012 report by the Linux Foundation (Corbet et al., 2012) states that 75%

of all Linux kernel development is done by developers who are being paid for their work, and that the top ten

organizations sponsoring Linux development are Red Hat, Intel, Novell, IBM, TexasInstruments, Broadcom,

Nokia, Samsung, Oracle, and Google. Extending the model to include user innovation would only strengthen

our results.