The Effects of Cultural Diversity on Regional Labour Markets in Germany

Transcription

1 The Effects of Cultural Diversity on Regional Labour Markets in Germany Jens Suedekum*) Katja Wolf Uwe Blien University of Duisburg-Essen IAB Nürnberg IAB Nürnberg Abstract In the last decade there have been marked changes in the composition of the non-native workforce in the German labour market. Firstly, there has been a notable increase in the group of high-skilled and a decline in the group of non-university trained foreign workers. Secondly, the diversity of the nationalities of which the foreign workforce is composed has risen considerably. In this paper we show that there are positive effects of cultural diversity for the native German employees at the local level. When considering the effects of skilled foreign workers, we find that both their total share and their diversification into different national groups raise local wage and employment levels for natives. This implies a positive effect on regional productivity. For the share of unskilled foreign workers we find negative wage and employment effects, which implies a negative effect on regional productivity. However, for a given share of unskilled foreign workers we still find positive productivity effects if this group is culturally diversified. Keyword: JEL-Class. Regional Labour Markets, Cultural Diversity, Immigration, Spatial Equilibrium R23, J21, J31 November 2008 Preliminary version! Do not quote! *) corresponding author: Jens Suedekum, University of Duisburg-Essen, Mercator School of Management, Lotharstraße 65, Duisburg, Germany - Phone: +49/(0)203/ jens.suedekum@uni-due.de The authors thank Holger Bonin, Fabrice Defever and seminar participants in Aarhus, Kiel, Mannheim and Nürnberg for very helpful suggestions on an earlier draft of this paper. The authors also thank Anette Haas and Van Phan thi Hong for assistance. All remaining errors and shortcomings are solely our responsibility. 1

2 1.) Introduction Germany has one of the largest shares of workers with foreign nationality across all European countries. This share has substantially increased during the early 1990s, mainly due to the fall of the iron curtain and the massive immigration from Eastern European countries. It reached a peak in the years after the German reunification, but since 1995 the total share of foreign workers in (Western) Germany is roughly stable or even slightly decreasing. Figure 1a focuses on full-time employees integrated in the social security system over the period and shows that about 8% of these so-called regular employment relationships are filled with workers with foreign nationality in the most recent year of the observation period. Figure 1a: Share of foreign workers in total full-time employment, West Germany, Figure 1b: Diversity within the group of foreigners (Herfindahl-index), West Germany, all foreign workers low-skilled foreign workers high-skilled foreign workers Yet, over the same time period, there have been marked changes in the composition of the non-native workforce. Firstly, particularly since 1998 there has been a notable increase in the group of high-skilled foreign workers with completed tertiary education, and a decline in the group of non-university trained foreigners. Secondly, the diversity of the nationalities of which the foreign workforce is composed has risen considerably. Figure 1b depicts the time trend of a Herfindahl-index over 180 different foreign nationalities in the total population of foreign workers with a regular employment relationship. 1 This figure suggests that the group of foreign workers in Germany has become substantially more heterogeneous in terms of national and, therefore, linguistic and cultural backgrounds. The labour market effects of immigration to Germany have been widely studied in the literature. For the immigration wave prior to reunification, De New and Zimmermann (1994) 1 See sections 2.3. and 3 for a description of the data and a formal definition of the Herfindahl-index. 2

3 find some evidence for displacement and adverse wage effects for native workers, whereas Pischke and Velling (1997) and Bonin (2005) find little evidence for such negative impacts for the late 1980s and for the period , respectively. Glitz (2007) focuses on ethnic German immigration after reunification and finds some displacement but no wage effects. Ottaviano and Peri (2008) use a structural approach and differentiate the foreigners and the natives by their skill and experience level. They argue that negative labour market effects arise, if at all, only for previous migrants but not for the native workers. 2 Little attention has been devoted to the labour market effects of the diversity of the foreign workforce for the native workers, however, despite the fact that the increase in heterogeneity is an important empirical fact as shown above in fig.1b. In this paper we use a panel of 326 German regions over the time period and analyze the wage and employment effects associated with cultural diversity of the foreign workforce at the local level. We condition our estimates on the total regional shares of foreign workers, in order to disentangle the effects of the fractionalization into different nationalities from pure size effects of the foreign workforce. The regional perspective that we adopt not only allows us to exploit the huge degree of intra-national variation of wage, employment and foreign participation levels in the data. It is also useful, because the theoretical arguments why cultural diversity might affect labour market outcomes of natives rest on the logic of small, local labour markets. Diversity may affect labour markets through different channels, and the net impact could be positive or negative (Ottaviano and Peri 2005, 2006). Firstly, workers with different cultural backgrounds embody complementary skills and problem-solving abilities. When these workers interact, productivity may rise due to knowledge spillovers or other forms of externalities. On the other hand, when the variety of the national backgrounds becomes too diverse, fractionalization may also imply excessive transaction costs for communication and, in fact, lower productivity. Secondly, the effects of diversity may also operate through a different channel, namely by affecting the quality of life in different locations. A tolerant native population may value a multicultural atmosphere as an attractive feature of a city, for example because of a more diverse supply of ethnic goods. Yet, diversity may also be perceived as an unattractive feature, e.g., if natives fear that social conflicts between different foreign nationalities are imported into their own neighbourhood. Notice that all of the aforementioned mechanisms, whose theoretical foundations are discussed in greater detail below, are far more likely to be relevant and visible on a local than on a national level. 2 An overview of the issues of German immigration is provided by Zimmermann et al. (2007). Dustmann and Glitz (2005) and Friedberg and Hunt (1995) survey the literature on the labour market impacts of immigration. Further important contributions to that literature include Borjas (1994) and Card (2001). 3

4 To address these various possible effects we use an estimating framework that is based on the spatial equilibrium model by Roback (1982). With this framework we recover the direction of the net impact of cultural diversity and the channel through which it affects natives from regional wage and employment regressions. The works that are most closely related to ours are two studies by Ottaviano and Peri (2005, 2006) who present comparable analyses for the US. They find that a diversity index, which describes the fractionalization of the local population into different linguistic groups, is robustly positively related with average wages, land rents and employment density among US metropolitan areas. This suggests that positive regional effects of diversity outweigh potential negative ones. Furthermore, the main channel through which cultural diversity benefits cities appears to be a rise in local productivity. Our paper adds value to this recent literature in at least three respects. First of all, one purpose of this paper is to validate if Ottaviano and Peri s conclusions for the US are representative for other developed countries. One has to keep in mind that the labour markets in Germany and the US are organized quite differently, so that the effects of (the diversity of) foreigners on native workers may also differ across countries. The United States are the classical immigration country. Their major cities like New York or Los Angeles always have been melting pots and the most favourite destinations for migrants from all over the world. For many European countries, including Germany, immigration laws have set quite different priorities for a long time, and migration flows have been considerably smaller (see Zimmermann 2005 for an overview). Furthermore, the institutional arrangements in the US labour market are more flexible than in Germany. 3 Given these major differences, it is not clear if the positive effects of cultural diversity that Ottaviano and Peri have found for the US are similarly present in a continental European country. The second contribution of our paper is to provide an analysis that pays closer attention to the skill composition of the foreign workforce. In their theoretical and empirical analysis, Ottaviano and Peri (2005, 2006) implicitly treat all non-english speaking workers as equally skilled and do not distinguish between skilled and unskilled foreigners. It is likely, however, that the labour market effects of skilled foreign workers are quite different from those of unskilled foreigners. Inter-cultural learning and knowledge spillovers potential candidates that may explain why diversity raises productivity reasonably require a certain education on the part of the foreigners. Frequent public debates about the principles of immigration laws also suggest that the attitude of the native population is quite different towards skilled than towards unskilled foreigners. In this paper we therefore analyze if the total stock of skilled 3 Schmidt et al. (1994) show that the effects of immigration into labour markets with high unionisation rates are quite different from the effects of immigration into flexible labour markets. 4

5 foreign workers and their cultural diversification have different effects on the natives than the stock and the diversification of unskilled foreigners. Taking into account the changes in the skill composition of the foreign workforce that have occurred in Germany (see fig.1a), it becomes even more important to explicitly make this distinction. Our third and final contribution is on the methodological side. Our benchmark results come from conventional panel estimation with regional fixed effects. This specification already partly takes care of the major econometric concern that arises in this type of study, namely that foreign workers may self-select into particular German regions if there are returns to cultural diversity. 4 With regional fixed effects all permanent regional attributes that vary only little over time, such as the presence of nation-specific migration networks that imply a permanent overrepresentation of certain nationalities in certain regions (the Japanese in Düsseldorf etc.), are filtered out by taking deviations from the mean. Still, to further address reverse causality, we turn to instrumental variable estimation where we instrument current with previous local diversity levels and also experiment with other exogenous variables. A second major concern is that local cultural diversity might capture occupational heterogeneity. Many industries which are prone to hire a diverse body of (specialist) foreign workers are likely to be concentrated in major cities, whereas rural hinterlands are unlikely to be very diverse in terms of nationalities. By conditioning our estimates on local industrial structures and other controls, we try to take this issue of omitted variable bias into account. Briefly previewing our main results, we find that positive labour market effects of cultural diversity do in fact exist in Germany, particularly when considering skilled foreign workers. Both, their total share and their diversification raise local productivity. For the share of unskilled foreign workers we find negative wage and employment effects, which implies a negative effect on regional productivity. However, for a given share of unskilled foreign workers, we still find positive effects if this group is diversified. Hence, the most adverse labour market effects are to expect if a region hosts a large and culturally homogeneous group of unskilled foreign workers. The rest of this paper is structured as follows. In section 2 we discuss some theoretical background and present an organizing framework that guides the estimations. Section 3 describes the data. Our main results are introduced in section 4. Section 5 is devoted to robustness checks. Section 6 concludes and draws some policy implications. 4 Glitz (2007) exploits the fact that some immigrants, namely the ethnic Germans from Eastern Europe, have been randomly assigned to different locations in Germany. This unique setting allows him to view this immigration wave as a quasi-natural experiment. In many other instances the location choice of foreigners within Germany is not random, however, which forces us to worry about self-selection and simultaneity bias. 5

6 2.) Theoretical background and estimation framework 2.1. Theory Our approach to identify the regional effects of cultural diversity is based on the seminal spatial equilibrium concept by Roback (1982). Consider a country that consists of r =1,2,,N locations. There are two goods in this model: a freely tradable good Y which is produced under constant returns to scale, and a non-tradable stock of land R which is in fixed supply in every region and owned by absentee landlords. Firms in the Y-sector use labour and land in production and are perfectly mobile across space. Workers are also perfectly mobile across space and consume residential land and the tradable consumption good. For simplicity it is assumed that consumer preferences are homothetic, so that expenditure patterns are independent of income. All markets are perfectly competitive and always clear in equilibrium. The price of the tradable good Y is used as the numéraire and is normalized to one. The price of land in region r is denoted as p r. The total number of workers in this economy is given by L ( L ) =. A crucial assumption is that the locations offer different productive and consumptive amenities, respectively affecting the location decisions of firms and workers. The utility level C of a worker in region r is denoted by Ur Ur( wr, Lr, Rr, A r ) =. With homothetic preferences, this utility level can be specified such that it depends on the nominal equilibrium wage ( w ), on the supply of land ( R r ), on the number of workers located in that region ( L r ), and on the C local consumptive amenity of region r ( A r ). Similarly, with constant returns in production, P the profit level of a firm in region r can be written as πr πr ( wr, Lr, Rr, A r ) =, where denotes the level of the productive amenity of location r. 5 We can then define a spatial equilibrium as follows: r r r P A r Definition 1: Spatial equilibrium A spatial equilibrium is a sequence of prices { } 1,..., N w w, { 1,..., N } workers { L1,..., L N }, such that, given the local supplies of land { 1,..., N } the amenities { C 1,..., C P A A N } and { 1,..., P N } A A : p p and an allocation of R R and the values of 5 An alternative identification strategy would explicitly solve for the equilibrium price of land p r, and would specify utility and profit levels as functions of wages w r, land prices p r and amenities. We use an identification approach that uses wage and employment regressions, because the available German data for housing and land prices are less reliable than our very accurate official employment data. 6

7 1.) Utility levels are equalized across space U ( ) = U ( ) = U k, l [ 1,..., N] 2.) Profit levels are equalized across space π ( ) = π ( ) = π, [ 1,..., ] 3.) All markets clear. k l k l kl N Figure 2 illustrates the spatial equilibrium. Think of one particular city r =1. We can invert P the profit function π π1( w1, L1, R1, A1 ) = and derive a labour demand schedule L D that is P downward sloping in the (w,l)-space with π, R 1 and A 1 as background parameters. This function shows the combinations of local wages and employment such that the firms in this location are just indifferent between staying and moving to a different region, given the exogenous stocks of land and the amenity levels of all regions in this economy. Similarly, the C utility function U U1( w1, L1, R1, A1 ) = can be inverted to yield an upward sloping labour supply schedule L S in the (w,l)-space, with U, R 1 and A 1 as background parameters. The point X in figure 2 is consistent with a spatial equilibrium: wage and employment (density) in city 1 is endogenously determined as w 1 and at L 1, respectively, given the profit and utility level that could be obtained in other locations (π and U ). Now consider an P increase of the productive amenity of region 1 ( A1 ). At any employment level firms in that region could pay higher wages without being induced to leave: the L D -curve shifts outwards. To restore spatial equilibrium, both wage and employment in city 1 must go up. This is illustrated by the movement from point X to point Y in figure 1. Since land supply is fixed, there is also an increase in the equilibrium price of land as more firms and workers move in. C An increase in the consumptive amenity A 1, on the other hand, would cause the L S schedule to shift down: For any level of employment individuals are willing to work at lower wages, due to the higher quality of life in that location. The new equilibrium would involve a lower wage in city 1, but a higher employment level (and thus, higher land rent). This is illustrated by the movement from point X to point Z in figure 2. Hence, both types of (positive) amenities increase employment (and thus capitalize in higher land prices), but productive amenities imply higher wages whereas consumptive amenities imply lower wages. An equivalent thought experiment is the following: Consider point X as the level of wages and employment in city 1. Now suppose we observe a second city (r=2) that has an identical supply of land but a higher employment density (and thus, higher land C 7

8 rent). Figure 2 suggests that this can happen for two reasons. Either city 2 has a higher productive, or a higher consumptive amenity. The former case is associated with higher wages (point Y), the latter case with lower wages in city 2 (point Z). Wage and employment regressions can be used to recover productive and consumptive amenities across cities. Figure 2: The identification approach w L S w 1 Y w 1 X w 1 Z L D L 1 L 1 L Obviously, the interesting question is: what precisely determines the productive and consumptive amenities of a location? Various candidates such as weather, landscape, local infrastructure, etc. have been considered throughout the literature. The discussion about the effects of cultural diversity can be fitted into this debate by hypothesizing that the cultural diversity of a city is a location characteristic as well. It is, however, not at all clear if diversity mainly affects business or private location decisions, and whether it is actually a positive amenity or rather a negative (dis-)amenity. Sensible stories can be told for each of these cases, and there is no other way but empirical analysis to sort out their real world relevance. A) Diversity may be a positive productive amenity: A culturally diversified stock of foreigners may raise local productivity, because it enriches the variety of culture-specific skills or problem solving abilities, which in turn fosters innovation (Florida, 2002). In the age of globalization the local availability of knowledge about specific conditions in foreign markets may be particularly important. 8

9 B) Diversity may be a negative productive amenity: At the same time, diversity may also lower productivity because it increases communication costs which naturally arise if people of different cultural backgrounds have to interact and to work together on projects. The trade-off of cultural diversity as a productive (dis-)amenity has been described in formal models by Berliant and Fujita (2008) and by Ottaviano and Peri (2005). In Berliant and Fujita (2008) knowledge creation or innovation requires interaction between individuals. The potential innovators need some non-overlapping prior knowledge, because people with identical background cannot learn from each other. At the same time the individuals also need some overlapping stock of knowledge since they otherwise have no common basis for communication. Berliant and Fujita (2008) do not explicitly distinguish between foreigners and natives, but the trade-off described in their model also applies to the productivity effects of cultural diversity at the local level: Some diversity may raise productivity because it goes hand in hand with sufficiently differentiated initial stocks of knowledge. Too much diversity, on the other hand, may lower productivity because communication becomes too costly. Ottaviano and Peri (2005) assume that cities are composed of different cultural groups, who interact in the production of a final good. At the same time, they also assume a transaction cost of cultural interaction. Provided this transaction cost is not too large final goods production exhibits increasing returns in the number of intermediate (nation-specific) inputs. This changes if transaction costs are sufficiently large, hence, the net effect of diversity on local productivity is ambiguous. Turning to the second possible channel, diversity may also be a location characteristic that mainly affects private location decisions of individuals: C) Diversity may be a positive consumptive amenity: A diversified multicultural environment may be perceived as an attractive regional attribute by a tolerant native population. Consumers may, for example, value a larger variety of ethnic goods that comes along with a culturally diversified population, such as a larger variety of restaurants or cultural events. D) Diversity may be a negative consumptive amenity: Diversity may, however, also be perceived as an unattractive location characteristic, because the natives fear foreign infiltration that gives rise to social conflict between too many different nationalities, etc. 9

10 This trade-off of diversity as a consumptive amenity is also entailed in the model by Ottaviano and Peri (2005), because foreigners do not only supply differentiated intermediate inputs, but they also produce differentiated non-tradable goods that are consumed directly. Using a CES-specification and monopolistic competition generates a love-for-variety effect. I.e., more cultural diversity leads to more consumption variety at the local level. At the same time Ottaviano and Peri again assume a transaction cost in the consumption of the ethically diversified final goods. If this cost is large enough, a negative net effect of diversity on utility is implied. Alesina and La Ferrara (2005) and Lazear (1999) also discuss the pros and cons of cultural diversity, referring mainly to its impact on consumers. The rationale for distinguishing between skilled and unskilled foreign workers relates directly to this theoretical framework. It is well conceivable that skilled and unskilled foreign workers affect regional productive and consumptive amenities differently. A positive impact of diversity on productivity may require a certain skill level of the foreign workers. Similarly, the consumption value of a multicultural environment may also depend on how educated the group of foreigners is Estimation framework In sum, cultural diversity has an ambiguous effect on regional development, and it may work through different channels. The theoretical framework outlined above shows how to disentangle between these various possibilities and gives rise to the following specification for the empirical model to be estimated: ln ( ) wage = α + α + β div + γ X + ε (1) wage wage wage wage wage r t emp emp emp emp emp ( emp ) = αr + αt + β div + γ X + εrt, ln (2) emp rt is total city employment and time t. w rt is the average wage for native workers in city r and div rt measures cultural diversity and refers to the foreign-born workers in region r. The precise specification of div rt for the empirical analysis is discussed below. X rt are additional control variables (such as the firm size, industry and skill composition of the native population), the α s are time and region fixed effects, and the ε s are error terms. 6 6 Notice that we could have specified employment density (emp r,t /area size) as the dependent variable in (2), which is equivalent since time-invariant area sizes are captured by the regional fixed effects. We have also controlled more explicitly for the regional supply of housing, as our theoretical framework assumes a fixed supply of land. 10

11 wage emp The central coefficients of interest are β and β. If diversity is a positive production wage emp amenity we should find positive wages and employment effects ( β > 0, β > 0 ). A negative production amenity would imply negative signs of both coefficients. If diversity is a positive consumption amenity we should find positive employment and negative wage effects wage emp ( β < 0, β > 0 ). If it is a negative consumption amenity, there must be a compensating wage emp wage differential ( β > 0 ) and negative employment effects ( β < 0 ). In the estimation we include regional fixed effects α and α wage to control for timeinvariant city characteristics that permanently affect wages or employment, respectively. With this approach we take a first cut at the issue of reverse causality, since permanent regional characteristics that influence wages and employment and, thereby, migration flows and the composition of the foreign workforces are filtered out by the region-specific intercept terms. In section 5 we address self-selection of foreigners into high wage locations, by treating as an endogenous variable and using instrumental variable techniques. emp r r div 2.3. Measurement of diversity In the estimations we typically include two variables related to the foreign workforce. Firstly, we control for the share of foreign workers in total area employment, i.e. s = foreigners emp (3) This variable measures the stock of foreign-born individuals who work in region r at time t. 7 The second type of variable measures the degree of diversification of the stock of foreigners into different nationalities. We use a standard Herfindahl-Hirshman index ( hhi ), which is defined in the following way: rt hhi 2 K foreigners k, = 1 k= 1 foreigners (4) Group k = 1,2,..., K indexes the different foreign nationalities. This index takes values between 0 and 1 and stands for the probability that two randomly drawn foreign employees 7 This type of variable has been widely used in previous studies on the effects of immigration. In our study we cannot explicitly distinguish if this total share in region r changes over time because of immigration from abroad, because of internal migration of foreigners within Germany, or because of naturalization of foreigners. Below we discuss the advantages and the limitations of our data set in greater detail. 11

12 belong to two different nationalities. If all foreigners in region r have the same nationality we would have hhi = 0, and the index then increases in the degree of diversity. 8 The correlation between s and hhi in the data turns out to be rather modes ( ρ 0.2 ), which allows us to control for both variables at the same time and, thus, to separate fractionalization and size effects of the foreign workforce. More formally, in the equations (1) and (2) we typically use the specification β wage divrt = wage wage, β1 srt, + β2 hhirt, and wage emp emp β, = β1 srt, + β2 hhirt,. When we explicitly distinguish the group of foreign div rt workers in region r by their level of qualification, we replace s by the share of skilled skilled s = skilled foreigners emp ; (unskilled) foreign workers in total employment (i.e., ( ) unskilled s ( unskilled foreigners ) emprt, = ). 9 Furthermore, we then measure the diversity index hhi for the sub-population of skilled (unskilled) foreign workers in region r separately. 3.) Data set and descriptive overview 3.1. Data issues The data basis for this study is provided by the Institute for Employment Research (IAB). It includes the complete population of full-time employment relationships subject to social security (i.e. excluding civil servants and self-employed individuals), aggregated at the level of the 326 Western German NUTS3-districts ( Landkreise and kreisfreie Städte ). Several remarks are in order about this data set. First of all, it contains information drawn from the official employment statistics by the German Federal Employment Agency (Bundesagentur fuer Arbeit) which is used in the administration of the German social security system. The data is therefore highly reliable as unit non-response and measurement error are reduced to a minimum. It is available as a balanced panel on an annual basis ( ). For every region and every year we also have some additional detailed information about structural characteristics of the local workforces, such as the qualification, firm size and industrial structure of regional employment. This information will be used to construct further control variables. 10 Still there are a few limitations that we have to face. Firstly, we measure regular employment at the regional level. I.e., foreigners as well as natives enter our data set only if they work full 8 For robustness checks we have also considered an entropy index as an alternative diversity measure. The estimation results for the entropy index turn out to be very similar as for the Herfindahl-index. 9 We have also considered a different specification, where we calculate the share of skilled (unskilled) foreign workers over the total number of natives and foreign skilled (unskilled) workers. Results turn out to be insensitive to this alternative specification. 10 Further information about this data set can also be found in Blien and Suedekum (2007). 12

13 time and pay social security contributions. On the one hand this seems to be appropriate as we are interested in the labour market effects of cultural diversity. Yet, some of the above mentioned theoretical mechanisms may actually depend on the size and composition of the local foreign population, rather than on the sub-population of foreigners who work full time. Since foreign population and employment are strongly correlated, however, we believe that the use of our data set does not lead to any systematic errors or biases. The second data issue concerns the definition of native and foreign workers. Notice that we proxy culture with the recorded nationality of an individual to begin with. We can measure the latter in a very detailed way, as more than 180 different foreign nationalities can be distinguished. Nationality is still only an imperfect measure of culture. First of all, there is naturalization of immigrants who obtain German citizenship subject to certain restrictions. Naturalization, which is particularly relevant for the ethnic German immigrants, implies that these individuals are recorded as native workers, even if they may not be culturally assimilated in any meaningful way. Second-generation migrants also often have German nationality although they may be culturally closer to their parents country of origin. A response to both caveats would be to measure culture by the primary language spoken at home, or by a similar variable. This would mean abandoning the use of official employment data, however, which typically does not include such information. We would have to use survey data instead, which is of substantially lower quality in other respects. Furthermore, it is known that overall naturalization rates in Germany are quite low (see Ottaviano and Peri 2008, table 1). Hence we believe that our measure of cultural diversity is reasonable and that the associated data problems are not too severe. The next data issue concerns the wage information for native workers, which will be used as one of our dependent variables. For every region and every year we compute the average daily wage income per employee, including all bonuses and extra payments subject to social security. The underlying micro-data for individual wage earnings have the problem that income levels which exceed the threshold for social security contributions are reported with this value. The data therefore understates the true degree of wage dispersion in Western Germany. There have been experiments at the IAB to estimate and to impute true wage earnings for individuals at the income threshold, but it turns out that this procedure has only minor effects on the distribution of regional average wages. Since our study uses average regional wages only, we are confident that our data is reasonably accurate also in this respect. The final issue we would like to discuss is the classification of regions in our data set. We use administrative regions (NUTS3) which are not defined according to economic criteria. This 13

14 raises the concern that some of our results may be artefacts of the choice of these units. 11 However, below we also present results where we use 204 local labour market units ( Arbeitsmarktregionen ) instead, which are functional spatial units defined in order to minimize cross-regional commuting flows Descriptive overview Before turning to the estimation results we give a brief descriptive overview about the participation of foreign workers in this section. In Table 1 we report the ten nationalities with the largest employment shares in Western Germany for the years 1995 and 2006, respectively. Table 1: Largest foreign nationalities in Western Germany (in %) Rank Turkey Turkey frm. Yugoslavia frm. Yugoslavia Italy Italy Greece 5.58 Greece Austria 3.95 frm. Soviet Union France 3.35 Poland Poland 3.01 France Portugal 2.51 Austria Spain 2.41 Portugal Netherlands 1.90 Spain 1.91 Note: Table 1 shows the ten largest share of nationality k in the total population of foreign full-time employees for Western Germany and for the years 1995 and 2006 (in per cent). As can be seen, the Turkish employees form the largest foreign group, followed by employees from the former Yugoslavia, Italy, and Greece. This ranking is very persistent over the years, and reflects longer historical incidents that can be traced back to the mid-1950s. The first large wave of immigration after WWII was caused by a shortage of labour that characterised the post-war economic boom. Various industries obtained the permission to hire foreign 11 The problem that the choice of the spatial units may crucially affect empirical findings is known in economic geography as the modifiable area unit problem (MAUP), see Briant et al. (2007). 14

15 guest workers, in order to fill low-paid positions for which it was difficult to find applicants. Turkey, Yugoslavia, Italy and Greece are the countries where most guest workers have been recruited. Originally the intention has been to issue only temporary work permits, and that the guest workers return to their countries after a certain period, but this was never strictly enforced. Quite the contrary, many former guest workers became assimilated in Germany and reunified with their families. Turning to the regional level, the map in figure 3 illustrates the total shares of foreign employees in The largest shares are found in the metropolitan areas in the southern part of the country (around Stuttgart, Frankfurt and Munich) and in the Rhine-Ruhr area in the West (Cologne and Düsseldorf). The share of foreign workers is below 1% in several other, more rural areas in the North. In table 2 we report further information for the seven largest cities in Western Germany (in terms of total regional employment). Except for Hannover, all large cities have above-average total shares of foreign workers. Table 2 also suggests that the foreign workforces in large cities tend to be relatively highly skilled compared to the West German average, Hannover again being the exception. Finally table 2 reports the diversity index of the foreign workforce in the selected large cities. Notice at first that the high-skilled foreign population is more diverse than the foreign population in general. This is due to the fact that foreigners of certain nationalities (particularly Western European and Northern American) are likely to be high-skilled if they work in Germany. Furthermore, urban foreign workforces tend to be more diverse than rural ones, since the total diversity index tends to range above the average for the large cities. This is also true when focussing only on the diversity of the skilled foreign workforce. Table 2: Labour market participation of foreigners in selected cities 2006 City Total employment Total share of foreign workers (%) Diversity index, foreign employees (native + foreign) all high-skilled all high-skilled Hamburg 633, Munich 563, Frankfurt 396, Cologne 369, Hannover 335, Duesseldorf 293, Stuttgart 286, Ø WESTERN GERMANY

16 Figure 3: Total share of foreign workers, 2006 Note: Map shows the shares of foreign employees in total regional full-time employment for the 326 Western German NUTS3-regions ( Kreise and kreisfreie Städte ) for the year ) Results Table 3 presents the benchmark results for the wage and employment regression, eqs. (1) and (2). In the first two columns of the wage regression we only control for the total share of foreign born workers ( s ), for their degree of cultural diversification as measured by the Herfindahl-index ( hhi ), and for area and time period fixed effects. In these specifications we leave out all additional control variables that pertain to structural characteristics of the native workforces. TABLE 3 ABOUT HERE We find highly significant negative wage effects from the total share of foreign-born workers. The diversity index on the other hand is positively associated with average native wages. That 16

17 is, wages are lower in German regions with a large share of foreign-born workers, but for a given share wages are higher if the foreign workforce is diversified into many nationalities. In column 3 we include an index of industrial diversity in region r, more precisely a Herfindahl-index of regional employment shares across 28 different industries. This index is included to control for the possibility that cultural diversity of the foreign workforce may capture regional occupational heterogeneity. By conditioning our estimates on the local industrial structure in this way, we decrease the concern of measurement error or omitted variable bias. As can be seen, the findings for the central coefficients of interest remain robust. Although wages tend to be higher in regions with industrial diversity, there is still an independent and significantly positive effect of the cultural diversity of the foreign workforce. Finally, in column 4 we add several other additional regional control variables, such as the qualification and firm size structure of the native workforces and some further variables that describe the global industry structure of a region. The estimated coefficients for these characteristics all have the expected sign and are precisely estimated. With respect to our main coefficients of interest, we find that the negative wage effect for the total share of foreign workers diminishes, which suggests that some correlation with the regional control variables for native workers exists. The overall effect remains negative and is highly statistically significant, however. The effect of the diversity index on native wages remains positive and highly significant, and the size of the effect is also roughly the same. Turning to the employment regression, we find positive and highly significant effects arising from both the total share of foreign workers and their degree of cultural diversity hhi. Both results are robust regardless of whether or not we include additional regional control variables, such as the index of regional industrial diversity (see column 6). Taken together, we find clear-cut evidence for a positive impact of cultural diversity on regional productivity, since we find positive wage and employment effects. For any given size of the foreign workforce productivity is larger if that group consists of members from various national/cultural backgrounds, and this effect is passed on to and benefits the natives. As for the negative wage effects associated with the total share of foreign workers, they can in principle have two interpretations according to our estimation framework. Either the share of foreign workers is a negative regional production amenity or a positive consumption amenity. emp To differentiate between these two possibilities one has to refer to the sign of β 1. As this coefficient is positive, our framework suggests the presence of a consumption amenity. We believe, however, that our results have a different economic interpretation, namely that the 17

18 coefficients for the total share of foreigners capture a net effect that masks more specific impacts of foreign-born workers with different qualification levels. In table 4 we turn to the wage and employment regressions where we control separately for the share of skilled and unskilled foreign workers, respectively. We use the same skilled specifications as in columns 4 and 6 in table 3 and only exchange s with s (table 4a) or unskilled with s (table 4b). Furthermore, we now use the diversity index hhi for the subpopulations of skilled (unskilled) foreigners only. For expositional purposes we only report the results for our central control variables in table 4 and omit the other estimated coefficients for brevity, since they turn out to be very similar as before. TABLE 4 ABOUT HERE For the group of skilled foreign-born workers with completed tertiary education we find that both their overall share in total regional employment and their diversification are associated with positive wage and employment effects. All estimated coefficients are highly significant at the 1% level. These results clearly suggest that a skilled foreign population is a productive amenity for a German region. The larger is the share of skilled foreign workers, the higher is regional productivity and the higher are, thus, average regional wages and employment for the natives. This positive impact is reinforced if the population of skilled foreigners is heterogeneous in terms of nationalities and cultures. With respect to the group of unskilled foreign workers results are different. The larger is the share of unskilled foreign employees in a region the lower is the average wage and the lower is employment for the natives. Hence, the size of this group of unskilled foreign workers can be interpreted as a negative production amenity. Still, for any given size of the unskilled foreign population we still find positive effects associated with cultural diversification of this group. This means that a region with a large share of foreign workers (call it region 1 ) need not be less productive than an otherwise similar region with a smaller share of foreigners ( region 2 ), provided the foreign population in region 1 is sufficiently heterogeneous and the foreign population in region 2 is sufficiently homogeneous. The most adverse labour market effects for native workers are to expect if they work in regions with a large and culturally homogenous group of unskilled foreign workers. Setting our estimated coefficients into perspective, we find that the involved magnitudes of the implied wage changes are modest but not negligible. With our point estimates one can calculate that an increase of the share of skilled foreign workers by one standard deviation 18

19 (roughly percentage points) raises the average monthly wage of a representative German region by about 2.89 Euro. An increase of the share of unskilled foreign workers by one standard deviation would reduce average monthly wages by slightly more than 5.82 Euro. Raising the Herfindahl index of cultural diversity for skilled (unskilled) foreign workers by one standard deviation implies a wage increase of 2.37 to 4.74 Euro per month, for given shares of foreign-born workers. Summing up, the main message of this section is that skilled foreign workers exert quite different effects on regional labour markets than unskilled foreign workers. Whereas skilled foreign workers can be seen as a positive production amenity from a regional perspective, unskilled workers are a negative production amenity. The diversification of the group of foreign workers tends to raise productivity regardless of the qualification level of the group of foreigners, i.e., it can also be interpreted as a positive productive amenity. 12 In the remainder of this paper we address the validity of these conclusions with several robustness checks. 5.) Robustness checks 5.1. Self-selection into high wage regions It is possible that foreign-born workers do not causally affect regional productivity and wages, but that foreigners instead endogenously self-select into particular locations. 13 Particularly with respect to the wage regression our estimations may suffer from a problem of reverse causality, as workers may sort into high-wage locations. A simple pooled regression model would be clearly ill-equipped to distinguish between the effect of diversity on wages and the opposite effect of wages on migration flows, and it is likely to overstate the importance of the former channel. Introducing regional fixed effects reduces these concerns because wage changes are less likely to cause contemporaneous changes in cultural diversity given the generally low mobility of labour in Germany. Still, to further address the direction of causality we turn to instrumental variable estimation. The principal difficulty of this approach is to find variables that are correlated with the geographical distribution of foreign workers but not with current wages or productivity. One approach that we take in this paper is to instrument current levels of cultural diversity with previous levels. Using lagged variables as instruments is standard in the dynamic panel data 12 It should be noted that our empirical framework identifies the net effects of cultural diversity. A large skilled foreign population may also have a consumption value for the native population. If this is so, this consumption value capitalizes in lower wages, and the observed positive wage effect even underestimates the true productivity effect. Likewise, the diversification of the foreign population may also have an effect on the consumption side. 13 For certain types of foreign immigrants (e.g. asylum seekers) it is in fact the case that they are exogenously distributed across German regions. For the lion s share of the foreign-born workers in our data set, however, the assumption is plausible that they can decide on their location within Germany. 19

20 framework. Because of our fixed effects specification such instruments are valid only asymptotically for long time periods. In our case we could use twelve points in time, which seems sufficiently long for using such lagged variables as instruments. 14 In addition we also experiment with two other exogenous instruments. The first one are deeply lagged regional employment shares of classical guest worker industries. The industries where these first guest workers were typically employed in the 1950s have been mining and building&construction. Family members of the immigrants often tended to work in gastronomy and other basic service industries. We construct combined regional employment shares of these four industries for the earliest year for which we have such information: the year 1977, i.e., 18 years before the observation period of our main data set starts. These deeply lagged industrial structures which determined historical location patterns of immigrants are likely to be correlated with the current location of families with migration background, due to the persistence of migration networks for example. At the same time these historical local industry compositions are less likely to have a bearing on current economic performance, because the importance of the classical guest worker industries for aggregate economic activity has rapidly diminished over time in the process of structural change. Even those regions whose economic structure was strongly characterised by these industries have tended to considerably change their industry compositions over time through a process of industry churning (see Findeisen and Südekum 2008). 15 The second type of additional instrument that we try out are the regional vote shares of the Green party in the national elections (Bundestagswahlen) which may be seen as a proxy for the tolerance of the native populations towards foreigners, but per se is likely to have little effect on the labour market performance of native workers. Table 5 summarizes the results for the instrumental variable estimation of the wage regression. Again we focus on the main coefficients of interest while leaving out the other regional controls for brevity. In the first column we report the results of the standard fixed effects estimation as a benchmark. In the second and third column we instrument the qualification-specific shares of foreign workers and their degree of diversity with time lags values of the same variable. More precisely, in column 2 we use a time lag of only one year, whereas in column 3 we increase the number of instruments and include two lag periods. 14 A more detailed discussion of these issues in the context of regional labour market analysis can be found in Martin et al. (2008). 15 In the comparable US analyses Ottaviano and Peri (2005) use the geographical distance of metropolitan areas to major immigration hubs as the instrument for the share of foreign-born workers, since most immigrants have entered the US through one of these hubs and tended to locate close to these places. The main intuition for our instrument is somewhat close in spirit, as we use information about where the foreign-born workers first entered the German labour market decades ago. 20