Distance from diasporas and immigrants’ location choice: evidence from Italy

ABSTRACT

Diasporas play a fundamental role in explaining the location choice of new immigrants. We investigate the spatial dimension of diaspora externalities focusing on immigrants in Italian local labour market areas (LLMAs). We show that the net pull effect of diasporas spills over an estimated average distance of 82 km. We find evidence of negative spatial spillovers at greater geographical distances, suggesting a competition effect from neighbouring diasporas. Ethnic-specific labour markets and ethnic consumption externalities are important channels through which the distance–decay effects of diasporas take place. We also find that the spatial effects of diasporas are highly heterogeneous across gender and origin countries.

KEYWORD:

• community networks
• immigration
• gravity models
• spatial dependence

JEL:

• C14
• C21
• F22
• J61
• R23

ACKNOWLEDGEMENTS

We are grateful for the valuable comments received from the participants at the DISEA-CRENoS Seminar organized by the University of Cagliari, 15 May 2020, and at the 2020-SIE online conference.

DISCLOSURE STATEMENT

No potential conflict of interest was reported by the authors.

Notes

1 For a detailed discussion of the literature and of our theoretical hypotheses, see Appendix A in the supplemental data online.

2 The only partial exception is Nowotny and Pennerstorfer (2019), who analyse the role of migrant networks in neighboring regions using individual-level data from the European Union Labour Force Survey (EU-LFS). They show that networks in adjacent regions (up to a second order of contiguity) matter in driving new immigrants’ location decisions. Although Nowotny and Pennerstorfer have the merit of showing the relevance of neighboring diasporas in shaping immigrants’ location decision, they leave unexplored the high degree of heterogeneity observed in their data, as well as in most studies on immigrants clustering (see Zavodny, 1999; and Chiswick et al., 2001, for the United States and Australia, respectively). Individuals as well as group-specific characteristics might explain these differences.

3 We thank an anonymous referee for suggesting the role played by family reunification.

4 The spatial ubiquity of job opportunities for female workers in Italy – in particular, those from Eastern European countries – implies that these immigrants are more likely to trade-off the proximity to local diasporas (and hence access to externalities the diasporas generate) with access to jobs in other localities that are more distant to them.

5 Annual flows are measured only at the time of the official registration which frequently occurs with a substantial lag – even years – after the actual entry into the country. The use of panel data would also strongly complicate the IV procedure adopted to reduce the bias due to the presence of the foreign population on the right-hand side of our econometric model as a proxy for diaspora.

6 Nuisance spatial dependence can also occur when the boundaries of the local markets where positive network externalities are expected (labour, housing or marriage markets, markets for ethnic goods or services, etc.) differ considerably from the boundaries of the spatial unit of analysis.

7 Recent contributions to spatial econometrics documented the presence of spatial dependence in conventional gravity models (e.g., Lesage & Pace, 2009; Lesage & Fischer, 2010).

8 This model is reported in Section D of the Appendix in the supplemental data online.

9 All regressions are estimated using the restricted maximum likelihood (REML) estimation method.

10 Equation (1)(1) $\begin{array}{rl}& {\eta }_{ij}=\ln ({\mu }_{ij})={\mu }_i+{\mu }_j+\delta \ln d_{ij}+\alpha \ln (1+M_{ij})+\sum _{\lambda \in \mathrm{\Lambda }}{\beta }_{\lambda }\sum _{l\ne j}{w}_{jl}^{\lambda }\ln (1+M_{il})\\ & {\mu }_{ij}=E(N_{ij})N_{ij}\sim Negbin(\mu ,\theta )\end{array}$(1) represents a finite distributed spatial lags-in-X variable model, that is, a model that incorporates a certain number – indefinite a priori – of spatial lags of the regressor. This approach allows us to directly estimate the overall spillover effects. Indeed, the structural (and also reduced) form of this model tends to mimic the reduced form of the [AQQ12] Spatial Autoregressive Model ($y=\lambda Wy+X\beta +\epsilon$), that is, $E[y|X]=X\beta +\lambda WX\beta +{\lambda }^2{W}^{2}X\beta +{\lambda }^3{W}^{3}X\beta +\dots$. If the hypothesis of a rapid distant-decay effect is valid, we should empirically observe a zeroing (or even a sign change) of the spatial spillover parameter after the first few spatial lags. An advantage of the finite-distributed spatial-in-X variable model in our application is that standard maximum likelihood methods can be applied to estimate the negative binomial model for count data, and also standard IV approaches can be used to control for the endogeneity of our main explanatory variable and its spatially lagged values.

11 We use two standard weight matrices that are purely exogenous. Often these choices are used as alternatives and, given the absence of a clearcut criterion for strongly preferring one over the other, as a robustness we decided to use both. We avoided using other weights matrices that might include elements of endogeneity in the estimation (e.g., a weights matrix based on commuting data).

12 We performed robustness analysis using different thresholds on the minimum size of diasporas (1000, 3000, 5000 and 10,000) in 2007. The results are qualitatively identical and are available in Section F of the Appendix in the supplemental data online. The parameter $\alpha$ associated with $ln(1+M_{ij})$ is quite stable across the different subsamples, indicating that the exclusion of small source countries does not affect the results. Moreover, the value of the Akaike information criterion (AIC) is lower when we select the subsample on the basis of the threshold of at least 5000 foreign citizens in Italy from the same origin country in 2007.

13 By a-spatial model, we mean a model that only includes the localized diapsora.

14 The diagnostic tests confirm the endogeneity of the network variable and of its spatial lags and the relevance of the instrumental variables used (Table 4).

15 The different magnitude of the localized diaspora effects with respect to Jayet et al. (2010) may depend on several dimension on which the two studies diverge: time period, spatial unit of analysis, model specification and estimation method. However, the main conclusions about the existence of a robust and sizable network effect are consistently corroborated in both studies.

16 For example, in the case of male immigrants, the CD test statistic for strong dependence (Pesaran, 2004) is 0.591 (p = 0.554), while the CD test statistic for weak dependence (Pesaran, 2015) is 1.840 (p = 0.065) (for the test of weak cross-sectional dependence, we have used the spatial contiguity matrix). In the case of female immigrants, the corresponding values of the statistics are 0.964 (p = 0.335) and 2.077 (p = 0.038).

17 The measure of cultural distance is that proposed by Del Gatto and Mastinu (2016, 2017); it is a weighted average of three distance measures: genetic, linguistic and religious. Economic distance is computed either as the ratio of the GDP per capita of the origin country and that of Italy, or as the ratio of the employment rate (number of employees/working age population) of the origin country and that of Italy.

18 Appendix G in the supplemental data online reports the results based on subperiods in order to take – at least partly – this limit into account. The analysis confirms the robustness of the results reported in the paper.