by Eric von Hippel
Cambridge, Mass. : MIT Press, 2005.
The shifting boundary between scientists and lay people in the production of scientific knowledge is just one among many changes in the division of social labour that we are witnessing. These changes involve many more sectors of the economy alongside scientific research and they share some fundamental characteristics and may be due to similar drivers. In particular, the cost of sharing information is rapidly decreasing as a result of widespread and fast digital networks. This decrease has released unexpected forms of production, where some crucial distinctions that we once used to understand economic activity and the reproduction of social life are being reshuffled or scrapped altogether: those between producers and users, experts and non-experts, even supply and demand. Also, the emergence of these sharing economies and peer-to-peer forms of production defies some entrenched assumptions about productive behaviors. People often distribute for free their creations in sharing communities and networks.
Science may be pre-adapted to these changes, insofar scientists do not generally orient their efforts according to price signals but to reputation signals – as members of peer-to-peer networks tend to do – and because scientists understood quite early the idea of recursive refinements of knowledge permitted by fast exchange of results. The Mertonian ideal of the scientific community is — for some — as much a model for human polity as it is a model for emerging productive endeavours (1). In the scientific domain as in other domains, peer-to-peer production is not replacing other modes of organizing research and the economy. It may well be that some scientific disciplines and other sectors of the economy will at the end turns out particularly unfit for widespread and meaningful forms of peer-to-peer organization. However, to understand the most novel particularities of post-Fordist sciences and economies (I use ‘post-fordist’ for lack of a better term, with the caution that ‘fordism’ stays here for many other Xs for which we might be in a post-x condition), it is important to look at what all these changes have in common, what are their drivers, and whether they are desirable. The classic von Hippel’s ‘Democratizing innovation’ does all these things, and does it at an abstract level that permits to extrapolate hypotheses to other domains – as he does himself in the concluding chapters.
Eric von Hippel is professor of economics at MIT and a leading economist studying user-led innovation. His models seek to explain when users will develop their own innovations, why this is expected to increase social welfare, and how policy-makers could facilitate this process. User-innovation is something that we all do when we purchase a product and we use it – as it were – off-label, but von Hippel believes that this is of the utmost significance for contemporary economies. While his language and argumentative strategies are squarely situated in contemporary economics, including welfare economics, the book is full of references to other literatures, from business management to social studies of science – which provide a useful compass to the reader.
To understand the concept of user-led innovation one has to look at what it replaces: manufacturer-centric innovation development systems, ‘the mainstay of commerce for hundreds of years’. In such systems, a user’s role is to have needs, needs that manufacturers then identify and try to satisfy by offering their own products developed at the point of production. These manufacturers will also use patents and copyrights to prevent free-riders taking advantage of their discoveries without contributing to the effort of developing them. However, manufacturer-centric systems are at all times just one part of innovation systems: much innovation happens at the point of use, both productive use of firms and consumption use of individuals. Individuals and firms give meaning and adjust the products that they purchase.
Much innovation is done by manufacturers, which may have some particular advantages related to superior know-how and economies of scale. Why then would a firm (or a private person) decide to carry out itself R&D on the technologies it employs instead of outsourcing it to specialized producers? Von Hippel’s model addresses this key question A firm will engage in user-led innovation when the costs of doing so are lower than the costs associated with delegating the innovation process to another agent. These costs include the standard costs involved in any principal-agent relationships: costs of sharing information about the needs of the principal to its agent, and conflict of interests between users (principals) wanting personalized solutions and producers (agents) preferring one-size-fits-all solutions that will satisfy a broader market niche. It must then be true than if the needs are too localized (at the limit: if only one user wants a particular solution), there won’t be any firm offering it. In this case, users have an incentive to carry out their own innovations. Yet this is inefficient: many users will develop independently from each other similar solutions, with a waste of time and energy. Von Hippel’s models basically suggests that any user-led innovation system that overcomes this inefficiency will be socially beneficial.
Fortunately, user-led innovation systems are already in place in the form of sharing communities, much facilitated by digital networks. Sharing communities existed before digital networks, and Von Hippel tries to account for the puzzling phenomenon of agents giving away innovations that have been expensive to create. However, sharing itself has become inexpensive, and this opens up a much broader space for user-led innovation. This is a main driver of the emergence of peer-to-peer innovation and production. Von Hippel is hence able to explain the changes from which we started with a simple and elegant model. However, Von Hippel also believes that user-led innovation systems will not thrive everywhere, nor they will thrive independently from the choices of policy-makers, who should at least be neutral in respect to manufacturer- and user-centric innovation systems. In theory, any legal provision that hinders the circulation of knowledge, including patents and copyrights, is at odd with user-led innovation systems – and indeed the free source community has by-passed regulation by creating their own licences, which are unsuitable for a manufacturer-centric system (this is much debated, however. Cf. Boldrin & Levine (2) believe that such impediments hinder innovation in any innovation system).
This book can and should be read for a variety of reasons. I conclude by pointing out a reason that may have gone unnoticed so far. A key normative concern of science and technology scholars have been the challenge of devising research systems and innovation systems where values, needs and expectations of the citizenry at large are appropriately taken into account. They diagnosed this social concern arguing that earlier forms of delegate decision-making in science and technology were no longer appropriate, if they ever were – as it was shown by prominent cases where the trust relationship between citizens and their delegates, both politicians and scientists, was broken. While sceptical toward earlier forms of expertocratic and delegated forms of decision making in science and technology, most STS scholars also stayed away from market-based solutions and opted for technologies of participation (science shops, citizen consensus conference) that situated the dialogue between citizens and scientists at a relatively early stage of the research systems, e.g. allocative decisions, ethical constraints (3). It is intriguing that the key normative concern of STS scholars is profoundly mutated in these emerging research and innovation systems, at least because it is increasingly unclear where to draw the line between citizens and R&D experts. User-led innovation systems seem to resolve – as it were – endogenously some of the problems related to the alignment of citizen values and technology. It would be hasty to conclude that this is a revolutionary scenario for the aforementioned social concern, at least because we do not know to what extent the handful of successful cases can be extrapolated elsewhere. This is at any rate a reason to argue that what von Hippel’s call democratizing innovation can be taken at face value.
(1) Delfanti, A. (2013) Biohackers: the Politics of Open Science. Pluto Press.
(2) Boldrin, M. & D.K. Levine (2008). Against Intellectual Monopoly. Cambridge Univ. Press.
(3) Callon M. et al. (2001) Acting in an uncertain world. The MIT Press.