The Role of Amenities in the Location Decision of Households and Firms

Traditionally, cities have been regarded as centres of production: Firms produce goods in cities because agglomeration economies make them more productive. People live in cities because firms provide jobs and income. The existence of cities and closely related location decisions by households and firms have often been explained by these production linkages. Density is thought to offer agglomeration benefits on the production side but negative congestion effects on the consumption side. More recently, urban scholars have departed from this view and considered cities not only as centres of production but also of consumption (Glaeser et al., 2001). Workers have arguably become more heterogeneous in terms of taste, more educated and their incomes have risen. Firms have become more footloose and goods more diverse (Brueckner et al., 1999; Kolko, 1999; Kotkin, 2000; Glaeser et al., 2001; Florida, 2002; Dalmazzo & de Blasio, 2011; Glaeser, 2011; Bauernschuster et al., 2012; Suedekum et al., 2012; Ahlfeldt, 2013). These developments have left workers with more leisure time and income to spend on the consumption side so that workers and firms are no longer expected to only respond to classic production links but to a wide range of amenities. This dissertation intends to shed further light on the consumer city idea. Motivated by a lack of empirical evidence, I contribute to this young field of research by investigating different aspects of the role of amenities in the location decision of households and firms. The work is structured around the classification of urban amenities as suggested by Glaeser et al. (2001). This classification as well as related literature is presented in Chapter 2. The comprehensive literature review is intended to provide a background for the analyses carried out in this work. Moreover, it shows the field’s methodological development which is characterised by a move from correlations to establishing causality. Chapter 3 investigates the role of cultural amenities in the location decision of firms. I define cultural amenities as localised goods and services, which is one of the four urban amenity categories defined by Glaeser et al. (2001). The idea is that innovative service firms are highly footloose and mainly rely on qualified labour as input factor. At the same time, highly qualified and “creative” individuals have a strong preference for a rich social and cultural life (Florida, 2002). It is therefore expected that firms, following its workers, act as amenity-maximising agents. I empirically test this hypothesis by estimating a location choice model for internet start-ups in Berlin. The identification of the cultural amenity effect is based on the fall of the Berlin Wall which is interpreted as a quasi-natural experiment. Amenities are found to positively impact on the location of web firms. A comparison with other service industries moreover suggests that amenities are significant to the location choice of creative sectors, whereas no effect can be observed for non-creative firms. Chapter 4 is centred around the amenity role of aesthetics and physical setting, and on heritage preservation in particular. Heritage designation is considered to solve an externality problem thus providing benefits to home owners, in terms of a reduction of uncertainty regarding the future of an area, but at the costs of development restrictions. The chapter proposes a simple theory of the designation process, in which it is postulated that the optimal level of designation is chosen so as to Pareto-maximise the welfare of local owners. The implication of the model is that a) an increase in preferences for historic character should increase the likelihood of a designation, and b) new designations at the margin should not be associated with significant house price capitalisation effects. The empirical results are in line with these expectations. In Chapter 5 a third type of urban amenities according to the Glaeser et al. (2001) definition is investigated in further detail, i.e. speed of transportation. Transport’s important role in economics is beyond controversy. The estimation of the impact of a new infrastructure project is, however, not entirely straightforward as the relation between transport and economic development is plagued by a notorious simultaneity problem. The allocation of transport is not completely random and may respond to demand as infrastructure projects usually require large investment costs. Conventional causal inference has approached this problem by only focusing on the uni-directional effect of transport provision. I therefore propose a method, which is well established in macroeconomics, to explore the structure of mutually related endogenous variables. In particular, I run bivariate Panel VAR models using unique historical data for Berlin during a dynamic period when most of today’s public rail network was established (1881-1935). Results do indeed suggest a simultaneously determined relation between transport and urban development. Chapter 6 extends the previous analysis of the Berlin rail sector by applying the Panel VAR methodology to the city of Chicago, Illinois, and the development of the ‘L’ train over a period of over 100 years (1910-2010). The analysis can be interpreted as both an additional robustness test and a comparative study. Results are in line with the findings for Berlin. The dissertation ends with the conclusion in Chapter 7, where I summarise the main findings and stress important contributions to the literature.