Evaluation of Regional Specialization Effects
Two predominant contradictory concepts of specialization are underlying the subsequent evaluation of regional specialization effects: First, the Marshall- Arrow-Romer (MAR) externalities according to Marshall (1890), Arrow (1962) and Romer (1986) and second, the Jacobs’ (1969) externalities.
As put forward in Glaeser et al. (1992), the concept of MAR, on the one hand, focuses on intra-industry knowledge spillovers, i.e. between firms in an industry. Hence, benefits due to specialization within a particular industry and due to regional concentration are seen as main drivers for innovative activities and the growth of industries (Glaeser et al. 1992, pp. 1127-1128). On the other hand, Jacobs (1969) points out the importance of the diversity of industries and complementarity of knowledge of the agents for innovation and growth (Glaeser et al. 1992, p. 1128).
However, the literature remains ambiguous about whether MAR specialization externalities or Jacobian diversification externalities are the driving forces for innovativeness. Numerous studies find empirical evidence for either MAR or Jacobian externalities or both (Fritsch and Slavtchev 2008; Greunz 2004; van der Panne 2004; Paci and Usai 1999; Feldman and Audretsch 1999). For instance, Paci and Usai (1999) find for the case of Italy that innovativeness in a local industry is positively affected by both MAR externalities associated to production specialization in the same sector and Jacobs’ externalities regarding the degree of diversity of the whole local system (Paci and Usai 1999, p. 389). Similarly, Greunz (2004) observes both kinds of externalities for the European regions using patent applications as a proxy for innovation (Greunz 2004, p. 584). Evidence for Jacobs’ externalities has been found by Feldman and Audretsch (1999). They point out the stimulating effect of diversified industries (within a specific location) sharing a common science base (Feldman and Audretsch 1999, p. 427). For the case of the Netherlands, Van der Panne (2004) depicts that regional specialization towards a certain industry tends to increase the innovative performance in that particular industry which is support for the concept of MAR externalities (van der Panne 2004, p. 603). Findings of Fritsch and Slavtchev (2008) suggest that regional specialization in a certain industry stimulates innovative activities only to a certain degree. As a performance indicator they use the efficiency of regions in generating new knowledge. They indicate that a relationship between specialization and performance of a region is inversely U-shaped. In other words, specialization is conducive only to a certain level, after that it rather hinders knowledge generation and therefore innovativeness (Fritsch and Slavtchev 2008, p. 20). Below we address the significance of the above stated findings in the case of knowledge-intensive sectors for the example of the Vienna life sciences sector.
In a meta-analysis Beaudry and Schiffauerova (2009) survey numerous contributions to this debate and depict a diverse picture of possible conditions and circumstances under which each kind of externality could be at work. Their findings show that the results depend on the choice of measurement and methodology rather than on actual differences in the strength of agglomeration forces across the industries, countries or time periods (Beaudry and Schiffauerova 2009, p. 318). Analyzing 67 studies, they identify certain factors that determine the likelihood of finding either MAR or Jacob’s externalities: the level of industrial classification, the fact whether low, medium of high technology sectors are examined, the choice of independent and dependent variables—including different performance measures like economic growth, productivity or innovation. In most studies a diverse environment (i.e. in favor of Jacob’s externalities) is more beneficial to innovation than a specialized industrial base at a medium and detailed industrial classification. In contrast, evidence for MAR externalities is more likely to be found if a coarse industrial level was examined (Beaudry and Schiffauerova 2009, pp. 321-325).
In the following, different specialization scenarios are conducted and analyzed referring to the ambiguous effects of the above stated types of externalities on innovativeness.