شبحی در اپرا: امکانات فرهنگی، سرمایه انسانی و رشد اقتصادی منطقه

We analyze the extent to which endogenous cultural amenities affect the spatial equilibrium share of high-human-capital employees. To overcome endogeneity, we draw on a quasi-natural experiment in German history and exploit the exogenous spatial distribution of baroque opera houses built as a part of rulers' competition for prestigious cultural sights. Robustness tests confirm our strategy and strengthen the finding that proximity to a baroque opera house significantly affects the spatial equilibrium share of high-human-capital employees. A cross-region growth regression shows that these employees induce local knowledge spillovers and shift a location to a higher growth path.

The connection between the presence of exogenous local amenities, such as weather conditions, and spatial differences in wages and rents has long attracted a great deal of interest.3Roback (1982) shows in the framework of a spatial equilibrium model that exogenous local amenities that are valued by employees are capitalized into rents and wages. In an extension of this model, Moretti (2004) distinguishes low-human-capital employees from high-human-capital employees, assuming that only the latter value the local amenity. This model suggests a spatial equilibrium, with a larger share of high-human-capital employees in the high amenity location. In this paper, we focus on consumptive amenities and argue that especially high-human-capital individuals cherish the availability of cultural amenities, a hypothesis that is supported by reality. For example, a large survey of about half a million individuals in Germany finds that the highly educated full-time employed respondents who moved in the previous 10 years in fact rank “cultural offerings and an interesting cultural scene” among the top five reasons (out of 15 possible reasons) for their location choice.4 The challenge in analyzing the effect of local cultural amenities on high-human-capital employees is that cultural amenities – as compared to natural amenities like weather – are not exogenously determined. Initially, the endogeneity issue arises from local high-human-capital individuals' ability and willingness to pay for cultural services (Glaeser, 2005). Furthermore, local governments might compete for creative individuals by subsidizing cultural services. For instance, German municipal governments subsidized “music and theatre” with more than 35 Euro per capita in 2006 (Destatis, 2008). Given the local government's budget constraint, subsidies for cultural services, other local public spending, e.g., for education and local taxes are to a degree simultaneously determined (Gyourko and Tracy, 1991). This gives rise to a second endogeneity problem when analyzing the effect of local cultural amenities on high-human-capital employees. In both cases, it is not possible to refer an observed share of high-human-capital individuals in a region to the cause of cultural amenities and, consequently, it is not guaranteed that expenditures for cultural amenities do indeed attract high-human-capital individuals. To overcome these endogeneity problems and identify a causal relationship between cultural amenities and the spatial concentration of high-human-capital employees, we exploit a quasi-natural experiment in German history. During the Baroque era, absolutistic courts and churches competed with each other for musical talent.5 This competition was especially fierce in what is now Germany, which at that time and up until industrialization was politically fragmented into several hundred princedoms. Vaubel (2005) illustrates this nicely when he mentions that Thuringia alone, where the composer J. S. Bach grew up, contained 22 separate courts. Music was so highly regarded that “every local court (Hof) worth its salt had its own orchestra or band (Kapelle or Harmonie), and the more affluent courts maintained opera houses” ( Scherer, 2001a, p. 719). These opera houses, many of which still exist, acted as tangible symbols of their builders' prestige. However, the presence of one of these opera houses does not necessarily mean that the surrounding region was wealthy and prosperous enough to afford it; very often the rulers incurred vast debt and engaged in deficit spending in their quest for grandeur ( Duchhardt, 1992 and Vierhaus, 1984). In other words, we argue that baroque opera houses do not indicate regional wealth or predict future prosperity. Accordingly, proximity to these opera houses, which were the result of a competition for prestige between kings, dukes, and princes, should be exogenous to the distribution of high-human-capital employees that originates from the period of and after the Industrial Revolution. We find that proximity to a baroque opera house is a strong predictor of a region's equilibrium share of high-human-capital employees. In several robustness tests, we discuss confounding factors that might be correlated with both proximity to a baroque opera house and the local share of high-human-capital individuals. Including control variables that are measured at the baroque opera house location and that are likely to capture productive spillovers from that location to the region being investigated does not change our result. To further validate our argument that proximity to a baroque opera house is a valid explanation for the effect of cultural amenities, we construct counterfactual opera house locations. Using propensity score matching, we determine from the pool of locations not having a baroque opera house those locations that are the “historic twins” of locations that did have an opera house. We then test the impact of proximity to these counterfactual places on a location's share of high-human-capital employees. These counterfactual locations differ significantly from the actual opera locations in the size of their cultural scene today but not in terms of the quality of infrastructure or the degree of urbanization. Eventually, we do not find a significant effect of distance to the closest counterfactual baroque opera house location on a location's share of high-human-capital employees. Local governments, however, are interested not only in whether local cultural amenities attract high-human-capital individuals but also in whether these high-human-capital individuals generate some form of local knowledge spillover. In the absence of positive spillovers, it is difficult for a local government to justify its use of taxpayer money to subsidize cultural amenities. Thus, we exploit the exogenous variation in the local share of high-human-capital individuals stemming from proximity to a baroque opera house and analyze the effect of the local level of human capital on a location's growth path. Endogenous growth theories suggest that in the presence of some form of local knowledge spillovers, the local level of human capital should have a positive effect on a location's growth path. Applying an instrumental variable approach, we find that it is the local level of high-human-capital employees who value their proximity to a baroque opera house that shifts a location to a higher growth path. The remainder of the paper is organized as follows. In Section 2, we provide a detailed description of the quasi-natural experiment in German history that provides us with an exogenous spatial distribution of cultural amenities, i.e., baroque opera houses. In Section 3, we analyze the effect of proximity to a baroque opera house on the spatial equilibrium share of high-human-capital employees and discuss a variety of confounding factors that might bias our estimates. In Section 4, we exploit this exogenous variation in the local share of high-human-capital individuals to test the impact of the local level of human capital on a location's growth. We conclude, in Section 5, by discussing some of the implications of our findings.

In this paper, we exploit exogenous variation in the spatial distribution of endogenous cultural amenities. Based on predictions derived from spatial equilibrium models with exogenous amenities, we show that the equilibrium share of high-human-capital employees is larger in regions with a high level of cultural amenities than in regions with low cultural amenities. More precisely, we show that, all else equal, for every 10 km nearer to an opera house, the NUTS3 region's equilibrium share of employees with a tertiary degree increases by about 0.2 to 0.4 percentage points. Robustness tests give us confidence that this finding can be interpreted as a causal effect and not as the result of confounding factors that are correlated with both the proximity to a baroque opera house and the local share of high-human-capital employees. To justify subsidizing cultural amenities, local governments want to know if the high-human-capital individuals who will be attracted to the region by these amenities will be the source of some form of local knowledge spillovers. According to our estimations, an increase of the local share of employees with a tertiary degree by one standard deviation increases average annual growth of regional GDP per employee by about 1.0 to 2.1% age points. The positive effect of the regional level of human capital on a region's growth path is a clear indication of local knowledge spillovers induced by the presence of high-human-capital employees. Given the quasi-experimental design of our study, we can isolate the positive effect of cultural amenities on the regional share of high-human-capital employees and, subsequently, on regional growth. We exclude possible simultaneous effects, such as reduced spending for education and infrastructure or increased local taxes. Our advice to local policymakers is to be aware of the value of cultural amenities when competing for high-human-capital individuals. However, local policymakers should also carefully consider the possibility of unwanted side effects from redistributing resources to cultural amenities at the expense of other public spending or increased taxes.