Research Policy 31 (2002) 125–144 Measuring knowledge spillovers in manufacturing and services: an empirical assessment of alternative approaches Ulrich Kaiser ∗ Department of Industrial Economics and International Management, Centre for European Economic Research, P.O. Box 10 34 43, D-68034 Mannheim, Germany Accepted 13 December 2000 Abstract In this paper it is tested which of the various alternative approaches for constructing knowledge spillover pools suggested in existing literature measures the extent to which a firm can costlessly receive external knowledge best. Since knowledge spillovers are unmeasurable, a ‘goodness of fit’ measure is constructed using innovation survey data. It turns out that measures of the uncentered correlation of firm characteristics seem to fit actual knowledge spillovers best. Direct measures constructed from innovation survey data appear to work reasonably well while measures of the Euclidean technological distance and of the geographical distance lead to counterintuitive results. Empirical evidence is provided for both the German service sector and the manufacturing sector. © 2002 Elsevier Science B.V. All rights reserved. JEL classification: O31; C12; C35 Keywords: Knowledge spillovers; Technological distance; Geographical distance; CIS-II data; Kernel density estimation; Ordered probit estimation with sample selection 1. Introduction Economists have demonstrated that the social returns to innovation exceed the private returns to in- novation if the knowledge produced in an innovation process is not fully appropriable by the innovating firm. As a consequence, spillovers may lead to Pareto– inferior Nash equilibria of innovation efforts as demonstrated by, e.g. Kamien et al. (1992), Mowery and Rosenberg (1989), and Suzumura (1992). Arrow (1962) was among the first to notice that a firm’s in- centive to invest in innovation decreases if knowledge generated by its innovation efforts is involuntarily ∗ Tel.: +49-621-1235-292; fax: +49-621-1235-333. E-mail address: kaiser@zew.de (U. Kaiser). transmitted to competitors. 1 In another early contribu- tion, Schmookler (1966) articulated that technologi- cal progress achieved by a firm may not solely be a result of its own research efforts but also from other firms’ research results. The impact of knowledge transmission between firms on the firm’s propensity to invest in innovative activity has been reemphasized by Spence (1984), who shows that the appropriability problem leads to a reduction of firms’ incentives to invest in R&D. This basic result is shared by other authors using different model setups such as D’Aspremont and Jacquemin (1988, 1990), De Bondt et al. (1992), Kamien et al. 1 Firms may find it profitable, however, to deliberately transfer knowledge to downstream users in special cases. See Geroski (1995a) and Harhoff (1996) for fuller treatments of this issue. 0048-7333/02/$ – see front matter © 2002 Elsevier Science B.V. All rights reserved. PII:S0048-7333(00)00159-1