Knowledge Spillovers

By knowledge spillovers, economic geographers usually refer to technical, scientific, or organizational knowledge that, once discovered, easily comes to the attention of other individuals due to its nonrival and (partially) nonexcludable nature. Nonrivalness implies that additional users bear no extra cost and do not prevent others from possessing and using the knowledge. Nonexcludability indicates that it is also impossible, or at least very difficult, to prevent others from gaining the knowledge, whether they have contributed to its production or not. The term spillover is intended to highlight the difficulty of controlling knowledge and preventing its unintentional dissemination in the economy. Now regarded as the central factor for economic growth, knowledge spillovers and their underlying circumstances have received increased scholarly attention.

A common assumption in the geography of innovation literature is that knowledge often takes the form of know-how or tacit knowledge, which is hard to convey by means other than repeated face-to-face interaction. When spillovers are thus locally bounded, firms tend to locate near their sources, such as universities, research and development (R&D) intensive firms, or skilled labor. Many empirical studies have therefore examined the impact of universities or labor mobility on regional growth, using various indicators. Patents and patent citations are now most commonly used in this respect, and in the case of scientific knowledge, various bibliometric methods are generating increased interest.

The most influential analytical framework to study the impacts of knowledge spillovers on urban growth was developed in 1992 by Glaeser, Kallal, Scheinkman, and Shleifer, who distinguished among three different perspectives:

• 1.
  MAR spillovers, referring to the original insights of the economists Alfred Marshall, Kenneth Arrow, and Paul Romer
• 2.
  Porter spillovers, named for the famed management scholar Michael Porter
• 3.
  Jacobs spillovers, named after the urban theorist Jane Jacobs

The first two approaches stress spillovers with in a sector (intraindustrial spillovers) and the importance of geographical economic specialization, whereas Jacobs spillovers occur between sectors (interindustrial spillovers) and are therefore more abundant in a more diversified local economy. In addition, the MAR perspective favors local monopolies, whereas the other two see strong local competition as a better incubator of innovative behavior. Although much debate surrounds the issue, Jacobs's perspective has gathered most support from researchers.

While the empirical literature on geographically localized knowledge spillovers is extensive, many central questions remain unsettled. The main criticisms revolve around the issues of tacitness—spillovers as externalities—and the actual nature of spillover processes. For example, tacit knowledge was originally viewed as implying that individuals know more than they can tell. It was therefore not merely difficult but impossible to convey this through formal communication. In the more recent literature, a case is now sometimes made that the main difficulty in transmitting much expert knowledge has often more to do with other factors, such as cognitive rather than geographical distance.

Another problematic issue is that knowledge spillovers usually imply that the diffusion is unintended and that the innovator does not receive full compensation. Some scholars therefore use the term knowledge externalities interchangeably. In many cases, however, the characteristics of positive externalities are not satisfied, or this proves difficult to verify. Finally, the numerous [Page 1664]empirical studies based on knowledge spillover proxies do not document or even prove their existence, and other explanations or factors might be more significant.

...