ECONOMIC PERSPECTIVES ON OPEN SOURCE

• Date: 27 April 2004
• DOI: https://doi.org/10.1016/S1048-4736(04)01502-4
• Published date: 27 April 2004
• Pages: 33-69
• Author: Josh Lerner,Jean Tirole

ECONOMIC PERSPECTIVES

ON OPEN SOURCE

Josh Lerner and Jean Tirole

ABSTRACT

There has been a recent surge of interest in open source software de-

velopment, which involves developers at many different locations and

organizations sharing code to develop and refineprograms. To an economist,

the behavior of individual programmersand commercial companies engaged

in open source projects is initially startling. This paper makes a preliminary

exploration of the economics of open source software.We highlight the extent

to which labor economics, especially the literature on career concerns’, can

explain many of these projects’ features. Aspects of the future of open source

development process, however, remain somewhat difficult to predict with

off-the-shelf’ economic models.

1. INTRODUCTION

In recent years, there has been a surge of interest in open source software

development. Interest in this process, which involvessoftware developers at many

different locations and organizations sharing code to develop and refine software

programs, has been spurred by three factors:

The rapid diffusion of open source software.A number of open source products,

such as the Apache web server,dominate their product categories. In the personal

Intellectual Property and Entrepreneurship

Advances in the Study of Entrepreneurship, Innovation and Economic Growth,

Volume15, 33–69

© 2004 Published by Elsevier Ltd.

ISSN: 1048-4736/doi:10.1016/S1048-4736(04)01502-4

33

34 JOSH LERNER AND JEAN TIROLE

computeroperatingsystemmarket,InternationalDataCorporationestimatesthat

the open source program Linux has between seven to twenty-one million users

worldwide,witha200%annual growthrate.Manyobserversbelieveit represents

a leading potential challenger to Microsoft Windows in this important market

segment.

The significant capital investments in open source projects. Over the past two

years, numerous major corporations, including Hewlett Packard, IBM, and Sun,

have launched projects to develop and use open source software. Meanwhile, a

number of companies specializing in commercializing Linux, such as Red Hat

and VA Linux, have completed initial public offerings, and other open source

companies such as Cobalt Networks, Collab.Net, Scriptics, and Sendmail have

received venture capital financing.

Theneworganizationstructure. The collaborative nature of open source software

development has been hailed in the business and technical press as an important

organizational innovation.

Yet to an economist, the behavior of individual programmers and commer-

cial companies engaged in open source processes is startling. Consider these

quotations by two leaders of the open source community:

The idea that the proprietary software social system – the system that says you are not allowed

to share or change software – is unsocial, that it is unethical, that it is simply wrong may come

as a surprise to some people. But what else can we say about a system based on dividing the

public and keeping users helpless? (Stallman, 1999).

The “utility function” Linux hackers is maximizing is not classically economic, but is the intan-

gible of their own egosatisfaction and reputation among other hackers. [Parenthetical comment

deleted] Voluntarycultures that work this way are actually not uncommon; one other in which I

have long participated is science fiction fandom, which unlikehackerdom explicitly recognizes

“egoboo” (the enhancement of one’s reputation among other fans) (Raymond, 1999b).

It is not initially clear how these claims relate to the traditional view of the

innovative process in the economics literature. Why should thousands of top-

notch programmers contribute freely to the provision of a public good? Any

explanation based on altruism1only goes so far. While users in less developed

countries undoubtedly benefit from access to free software, many beneficiaries

are well-to-do individuals or Fortune 500 companies. Furthermore, altruism has

not played a major role in other industries, so it would have to be explained why

individuals in the software industry are more altruistic than others.

This paper seeks to make a preliminary exploration of the economics of open

source software. Reflecting the early stage of the field’s development, we do

not seek to develop new theoretical frameworks or to statistically analyze large

samples. Rather, we focus on four “mini-cases” of particular projects: Apache,

Economic Perspectives on Open Source 35

Linux, Perl, and Sendmail.2We seek to draw some initial conclusions about the

key economic patterns that underlie the open source development of software.

We find that much can be explained by reference to economic frameworks. We

highlight the extent to which labor economics, in particular the literature on

“career concerns,” and industrial organization theory can explain many of the

features of open source projects.

At the same time, we acknowledge that aspects of the future of open source

development process remain somewhat difficult to predict with “off-the-shelf”

economic models. In the final section of this paper, we highlight a number of

puzzles that the movement poses. It is our hope that this paper will have itself an

“open source” nature: that it will stimulate research by other economic researchers

as well.

Finally, it is important to acknowledgethe relationship with the earlier literature

on technological innovation and scientific discovery. The open source develop-

ment process is somewhat reminiscent of the type of “user-driveninnovation” seen

in many other industries. Among other examples, Rosenberg’s (1976) studies of

the machine tool industry and von Hippel’s (1988) of scientific instruments have

highlighted the role that sophisticated users can play in accelerating technological

progress. In many instances, solutions developed by particular users for individual

problems have become more general solutions for wide classes of users. Similarly,

user groups have played an important role in stimulating innovation in other

settings: certainly, this has been the case from the earliest days of the computer

industry (e.g. Caminer et al., 1996).

A second strand of related literature examines the adoption of the scientific

institutions (“open science,” in Dasgupta & David’s, 1994 terminology) within

for-profit organizations. Henderson and Cockburn (1994) and Gambardella (1995)

have highlighted that the explosion of knowledge in biology and biochemistry

in the 1970s triggered changes in the management of R&D in major pharma-

ceutical firms. In particular, a number of firms encouraged researchers to pursue

basic research, in addition to the applied projects that typically characterized

these organizations. These firms that did so enjoyed substantially higher R&D

productivity than their peers, apparently because the research scientists allowed

them to more accurately identify promising scientific developments (in other

words, their “absorptive capacity” was enhanced) and because the interaction with

cutting-edge research made these firms more attractive to top scientists. At the

same time, the encouragement of “open science” processes has not been painless.

Cockburn, Henderson and Stern (1999) highlight the extent to which encouraging

employees to pursue both basic and applied research led to substantial challenges

in designing incentive schemes, because of the very different outputs of each

activity and means through which performance is measured.3