When the money is gone: Survey study on post-reward effects of financial incentives in OSS development

Søren Iver Kryhlmand Knudsen & Thor Martin Jensen

Student thesis: Master thesis


Businesses in the Internet and software related industries are increasingly reshaping their business models, building tools and platforms that enable them to harness collaborative open source (OS) communities and benefit from productive social relations. Moreover, investments by commercial firms in open source software (OSS) development are on the rise. This growing trend has raised concern among prominent members of the OS movement, which speculate that by paying some developers to work full- or part-time on projects they will lose interest in contributing to projects when they are not being paid. Theorists in the realm of social psychology and behavioral economics have shown that financial rewards can have a detrimental effect on intrinsic motivation, effort and productivity. However, contrary to findings in other experiments and studies, financial rewards have so far been found not to negatively affect intrinsic motivation in OSS development. Drawing on theories of intrinsic and extrinsic motivation, which underlies most of these studies, we challenge this assumption. Inspired by motivation crowding theory and the findings in Edward Deci‘s widely cited Soma experiment from 1971 we advance nine hypotheses, which predict that the detrimental effect of financial incentives on intrinsic motivation, effort and productivity occurs once payments are discontinued – an area that has not been explored by previous research. The empirical study is approached with a quantitative method. A self-completion web-based survey is sent to 2000 active OSS developers and responses are gathered from 456 individuals. Respondents are then categorized according to their payment status (unpaid, paid or previously paid), and a Mann-Whitney U test is used to examine the hypotheses on post-reward crowding effects. Interestingly enough, all hypotheses are rejected. We do not find any crowding-out of intrinsic motivation between active OSS developers who have never been paid and those who have previously been paid. The results remain consistent when the focus is narrowed to OSS developers who have previously received completion contingent rewards or those who perceived financial rewards to limit their autonomy during the payment phase. Further, we find a substantial crowding-in of productivity among OSS developers who have previously been paid. This crowding-in effect partially disappears when controlling for OSS experience.

EducationsMSc in Management of Innovation and Business Development, (Graduate Programme) Final Thesis
Publication date2010
Number of pages235