1# RECORD 01e2f1bf2915117a013dff8a924c07e2ed16f1522# TITLE: Living apart together: The economic value of ethnic diversity in cities3# YEAR: 2018.0 VENUE: Environment and Planning B Urban Analytics and City Science DOI: 10.1177/239980832210821124# AUTHORS: J. Bakens; R. Florax; H. D. de Groot; P. Mulder5# PAGES: 16 CANDIDATE TABLE PAGES (1-based): [5, 6]6# ABSTRACT: In consumer cities, the presence and location of immigrants impacts house prices through two channels, which both can be valued positively as well as negatively: (i) their presence and contribution to population diversity and (ii) the creation of immigrant-induced consumer amenities like those associated with ethnic restaurants in terms of both their quantity as well as diversity. We hypothesize that these two mechanisms create a trade-off in which city dwellers want to live apart yet consume together. We use unique microdata of house prices and ethnic restaurants in the city of Amsterdam over the 1996–2011 period and a generalized propensity score (GPS) matching technique for multiple continuous treatments to estimate the trade-off between consumers’ love for ethnic goods and their variety on the one hand, and ethnic residential composition on the other hand. We find evidence for the existence of such a trade-off: proximity to ethnically diverse restaurants indeed compensates for a negative correlation between the presence of immigrants and house prices. This trade-off mostly holds for neighborhoods featuring both a diverse population and a high level of amenities in terms of restaurants.789===== PAGE 1 (score 6, numeric-paren 0, table-words 0, var-words 6) =====1011Manuscript12EPB: Urban Analytics and City Science132022, Vol. 0(0) 1–1614© The Author(s) 202215Article reuse guidelines:16sagepub.com/journals-permissions17DOI: 10.1177/2399808322108211218journals.sagepub.com/home/epb19Living apart together: The20economic value of ethnic21diversity in cities22Jessie Bakens23Research Centre for Education and the Labour Market, Maastricht University, Maastricht, The Netherlands24Raymond JGM Florax25Department of Agricultural Economics, Purdue University, West Lafayette, IN, USA26Department of Spatial Economics, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands27Tinbergen Institute, Amsterdam, The Netherlands28Henri LF de Groot29Department of Spatial Economics, Vrije Universiteit Amsterdam, Amsterdam, The Netherlands30Tinbergen Institute, Amsterdam, The Netherlands31Peter Mulder32TNO, Amsterdam, The Netherlands33Abstract34In consumer cities, the presence and location of immigrants impacts house prices through two35channels, which both can be valued positively as well as negatively: (i) their presence and con-36tribution to population diversity and (ii) the creation of immigrant-induced consumer amenities like37those associated with ethnic restaurants in terms of both their quantity as well as diversity. We38hypothesize that these two mechanisms create a trade-off in which city dwellers want to live apart39yet consume together. We use unique microdata of house prices and ethnic restaurants in the city40of Amsterdam over the 1996–2011 period and a generalized propensity score (GPS) matching41technique for multiple continuous treatments to estimate the trade-off between consumers’ love42for ethnic goods and their variety on the one hand, and ethnic residential composition on the other43hand. We find evidence for the existence of such a trade-off: proximity to ethnically diverse44restaurants indeed compensates for a negative correlation between the presence of immigrants and45house prices. This trade-off mostly holds for neighborhoods featuring both a diverse population and46a high level of amenities in terms of restaurants.47Corresponding author:48Jessie Bakens, Research Centre for Education and the Labour Market, Maastricht University, Tongersestraat 53, Maastricht496211LM, The Netherlands.50Email: j.bakens@maastrichtuniversity.nl515253===== PAGE 5 (score 20, numeric-paren 0, table-words 3, var-words 12) =====5455In addition to the density of immigrants and restaurants, we look at their diversity. Diversity is56measured with the broadly used Hirschman-Herfindahl index, which allows to measure both57richness (the number of groups) and diversity (the distribution across groups).58Analogously to the density measure, we use the relative diversity index, that is, the diversity59relative to the overall diversity of the immigrant population and restaurant diversity in Amsterdam.60A summary of all variables included in the analysis and descriptive statistics can be found in Tables611 and 2.62The majority of the observations in our data lie between a treatment intensity of 1 and 5 for63immigrant density, between a treatment intensity of 0.9 and 1 for immigrant diversity, and below a64treatment intensity of 1 for restaurants. As Table 3 shows, the correlation between immigrant and65restaurant density is negative and low, but statistically significant (the coefficient is 0.14), which66indicates that within areas there is no strong correlation between the size of the immigrant pop-67ulation and the number of restaurants. Correlations between the treatments do show that restaurants68are more diverse in areas with more immigrants. The positive correlation between restaurant di-69versity and immigrant density is not investigated in this paper, but used as an assumption. The70statistically significant correlation points towards the fact that immigrant-induced diverse amenities71arise with the presence of immigrants because of a supply as well as demand effect of immigrants of72ethnic goods and services. The data also reveal a positive correlation between restaurant density and73diversity indicating that areas with a higher restaurant density on average have a higher diversity of74restaurants. The same holds for immigrant diversity. Figures 1 and 2 show the spatial distribution of75the treatments, specifically the treatment intensities. From these maps, it can be seen that the areas76that have both a large immigrant population share and many restaurants (in addition to high77immigrant and restaurant diversity)) are limited to the neighborhoods just outside the city center—78where, compared to the city center, immigrant density starts to increase, restaurant density starts to79decrease, and restaurant diversity is highest. Insofar as we would find a trade-off between immigrant80density and restaurant density it is most likely to happen in these neighborhoods because the joint81Table 1. Descriptive statistics.82mean83st. dev.84min.85max.86Price per square meter in e873435881000897709012,19091Relative immigrant densitya924.05932.16940.059513.0596Relative restaurant densitya971.28981.47990.081009.11101Relative immigrant diversitya1020.851030.151040.141051.02106Relative restaurant diversitya1070.911080.161090.181101.20111Share social rent pc61120.241130.3211401151116Share owner occupied pc61170.431180.2911901201121Population density km212210,930123434612432212520,510126Distance to Amsterdam CS station (meter)12736171282,011292501301162131Distance to top10 (meter)b13232,50013318,16713411,000135108,000136Distance to nearest metro station (meter)1371660138925139211404889141Data source: OIS and NVM.142aWe use the diversity and density relative to the overall diversity and density of the immigrant population and restaurants in143Amsterdam. Because of this scaling, the distribution of the measures is not normally distributed around 1, which would have144been the case if we had used the measures relative to the density or diversity in the sample. The immigrant density and145diversity are measured in a 250 m radius around the dwelling. The restaurant density and diversity is measured in a 1000 m146radius around the dwelling.147bThe ten main tourist attractions are based on visitor numbers. We use total distance from the dwelling to the ten attractions.148The ten tourist attractions used are the Rijksmuseum, Van Gogh Museum, Anne Frank Huis, Artis Zoo, Stedelijk Museum,149Madame Tussauds, Heineken Experience, Nemo Science Center, Hermitage, and Venustempel Museum.150Bakens et al.1515152153154===== PAGE 6 (score 27, numeric-paren 6, table-words 4, var-words 13) =====155156occurrence of both restaurants and immigrants is necessary for a trade-off. A more detailed dis-157cussion of the distribution of the treatments and the joint spatial occurrence of multiple treatment158intensities can be found in the Supplemental Material.159Method and diagnostic tests160Following the approach of Hirano and Imbens (2004) for GPS and the modification of the estimation161for multiple continuous treatments (Egger and Von Ehrlich, 2013) we estimate the dose-response162Table 2. Dummy variables.163Category164Observations165Inside A10 highway ring166Yes16732, 621168No1697, 519170Garden171Yes1722, 852173No17437, 288175Transaction year17620061777, 78517820071797, 64018020081816, 38518220091836, 17918420101856, 27918620111875, 872188Number of rooms1893 or less19027, 98119141927, 5051935 or more1944, 654195Type of dwelling196Simple house197153198Canal house199209200Bungalow20129202Villa20394204Ground floor apartment2055, 640206Upper floor apartment20724, 732208Gallery apartment2091, 457210Maisonette2111, 464212Portico flat2133, 035214Ground and upper floor apartment215285216Single-family house2172, 266218Town house219734220Rest category22142222Observations22340, 140224Data source: OIS and NVM.225Table 3. Correlations between relative treatments.a226Immigrant density227Restaurant density228Immigrant diversity229Restaurant density2300.14***231Immigrant diversity2320.45***2330.55***234Restaurant diversity2350.36***2360.29***2370.04***238Data source: OIS and NVM.239aThe statistical significance of the correlation coefficients is indicated by ***, **, * for the 0.01, 0.05, and 0.1 significance240levels, respectively.2416242EPB: Urban Analytics and City Science 0(0)243244245===== PAGE 7 (score 11, numeric-paren 0, table-words 2, var-words 7) =====246247function in three steps. A detailed description is given in the Supplemental Material. In the first step,248the conditional distribution of the treatments given the covariates is modeled. In the second stage,249we estimate the conditional expected house prices given the multivariate treatment intensities and250the estimated GPS. In the final stage, the average potential outcome at different treatment level251combinations is estimated.252The covariates in the selection equation explain about 53 percent of the variation in immigrant253density and 81 percent of the variation in restaurant density.9 The R2-values of the immigrant and254restaurant diversity estimations are relatively low, 25 percent and 37 percent, respectively. Although255the spatial distribution of the treatment intensity of restaurant and immigrant diversity does not256indicate the nature of the omitted variables, there is likely an underlying pattern explaining res-257taurant and immigrant diversity that we cannot perfectly capture with our data. Restaurant diversity258might also be related to market factors such as competition and quality, while immigrant diversity259might relate to other spatial aspects, that we do not observe. For the GPS to obtain unbiased260estimates of the dose-response function, the covariates in the first step need to be normally dis-261tributed. A Henze-Zirkler multivariate normality test of a random 10 percent sample of the errors of262the GPS estimation shows that the conditional treatments are not multivariate normally distributed.263The univariate Q-Q plots of the residuals in the Supplemental Material show that the non-normality264in the data is present for all treatments. The Q-Q plots show high kurtosis for all treatments,265especially in the left tails, but the effect is most apparent for restaurant diversity.266In the second stage, we estimate the conditional expected house prices given the multivariate267treatment intensities and the estimated GPS. The likelihood ratio test shows that including the GPS leads268to a better model specification for estimating the dose-response function, with a χ2 of 1941. We reject the269null hypothesis that the GPS does not improve the specification, with a p-value less than 0.1 percent.270Figure 1. Average per square meter house prices between 2006–2011 in constant 2088 prices.271Bakens et al.2727273