Skip to main content

Migrants’ Influence on Firm-level Exports


We examine the role of migrants in trade using a firm-level approach. We exploit a new employer–employee panel for Sweden, which encompasses close to 600,000 full-time employees, approximately 12,000 firms and data for 176 countries for the period 1998–2007. The resulting analysis provides novel firm-level evidence on the trade-migration relationship. Foreign-born workers have a positive association with firm exports. However, immigrants do not have an unconditional positive impact on firm trade. Mainly small firms gain from hiring foreign-born workers, and migrants need to be skilled and recently arrived to have a clear positive impact on firm export performance.

This is a preview of subscription content, access via your institution.


  1. 1.

    Egger et al. (2012) provide a review of the empirical literature on the trade-migration relationship. Overall, the link has been found to be larger for trade between dissimilar countries, trade with countries with weak institutions and the trade in differentiated products. Genç et al. (2011) conducted a meta-analysis based on 48 studies and found that, according to the existing research, a ten percent increase in a country’s immigrant stock was associated with one to two percent increase in foreign trade. Studies have also shown a strong positive correlation between provincial migration stocks and foreign trade, as well as between out-of-province stocks and in-province trade (e.g., Herander and Saavedra 2005). Previous studies are displayed in the accompanied web-appendix, which also includes further descriptive statistics, empirical analysis, and details on the construction of the dataset.

  2. 2.

    A handful of firm-level studies of the migration-trade relationship have started to emerge only recently (Koenig 2009; Bastos and Silva 2012; Pennerstorfer 2012; Hiller 2013). Unlike this study, however, previous firm-level attempts to analyze the trade-migration relationship have typically combined country or regional migrant stocks with firm export data within a gravity model framework, which is problematic because country and/or regional migrant stocks are inadequate proxies of migrants employed in private business.

  3. 3.

    The largest immigrant groups by source country are Finland, Iraq, Poland, Serbia/former Yugoslavia, and Iran. The web-appendix includes a complete list of Sweden’s largest immigrant groups, their respective size and share of population.

  4. 4.

    Bastos and Silva in turn draw on, inter alia, Chaney (2008), Eaton et al. (2011), and Crozet et al. (2012). It can be noted that in the neoclassical framework, trade and migration are substitutes. Relaxing underlying assumptions—allowing for non-identical technologies across countries, for example—it is possible to achieve complementarity. Rauch (1991) expanded this approach in a Heckscher–Ohlin model.

  5. 5.

    The variable imstock j is related to Crozet et al. (2012), according to which foreign variation in tastes for goods may explain heterogeneous trade outcomes.

  6. 6.

    This is in line with Bastos and Silva (2012), in which the variable is interpreted as representing firm-specific migrant networks.

  7. 7.

    In Eq. (3), fixed effects are incorporated in the spirit of Mundlak (1978), as explained in Wooldridge (2002). Specifically, we define \( {\mu}_{fj}={\displaystyle {\sum}_k{\overline{Z}}_{kf}}+{\overline{imstock}}_j+{\overline{GDP}}_j+{\overline{POP}}_j \)

  8. 8.

    In initial estimations of Eq. (4), a small constant is added to the dependent variable to avoid truncation.

  9. 9.

    We cluster standard errors by firm-country and adopt the Huber/White/sandwich variance–covariance estimator to adjust for downward bias, serial correlation, and heteroscedasticity. This approach is also motivated by the fact that zeros account for 94 % of the observations in our data. Wooldridge’s (2002) selection test also indicates that disregarding selection would bias the results.

  10. 10.

    In the first step, we evaluate Eq. (3) for each year of the panel using probit estimation. We then compute the yearly inverted Mill’s ratios (IMR) based on predicted probabilities. In the second step, we interact the IMR with year dummies in OLS estimation of the target Eq. (4), while adjusting for the addition of an estimated parameter using cluster-robust standard errors. If such differences exist (which seems plausible), but are ignored, estimation of Eq. (4) will be biased.

  11. 11.

    As specified in footnote 10 above.

  12. 12.

    A Hausman test confirms the appropriateness of specification of fixed effects over random effects.

  13. 13.

    Alternative procedures for dealing with zero trade have been considered, such as the Poisson Pseudo-Maximum Likelihood (PPML) estimator with fixed effects recommended by Santos Silva and Tenreyro (2006). We also considered a Tobit estimator. These estimators are not advantageous, however and the PPML estimator can result in severely biased estimates when zero-trade values are frequent (e.g., Gómez-Herrera 2013; Bratti et al. 2014). Regarding the Tobit, a Hausman test indicated that it would be inconsistent. With respect to missing trade, our panel excludes very small firms. Furthermore, running estimations separately for non-EU and EU member countries, where cut-offs exist, did not alter the results.

  14. 14.

    A detailed account of the construction of our measure can be found in the web-appendix.

  15. 15.

    Observations are lost for a few countries due to a mismatch in country coverage between the datasets used.

  16. 16.

    The strategy performs well, allowing identification. The regulatory measure affects the propensity to trade, while the conventionally used variable (common religion) measure affects the intensity of trade.

  17. 17.

    Reasons used to support the proposition that the direction of causality runs from migration to trade generally include a) descriptive macro data show no distinct support for the idea that trade precedes migration; b) in the literature on international migration and development, bilateral trade is not put forward as an important determinant of migration; and c) sociological and anthropological studies that have investigated migration decisions at the individual and household level have not found that preexisting trade between the host country and the source country influences the decision to migrate (Gould 1994).

  18. 18.

    A survey among Swedish managing directors and human relations managers shows that the attitudes of firms toward diversity are becoming more positive, and among large firms, a majority of managing directors view diversity as a potential means to improve profitability.

  19. 19.

    The survey, which included a set of questions proposed by the authors, was conducted by the Swedish Federation of Business Owners (SFBO), Sweden’s largest independent business association. It was carried out during March and April of 2013. Approximately 4,000 firms were asked to participate in the survey, and about 1,200 did, which is a normal response rate for a survey conducted by the SFBO. The questions that were proposed by the authors focused on foreign trade, workforce composition, and hiring decisions. One question (translated from Swedish) was as follows: “Based on what qualities does your firm decide to hire foreign-born worker(s)?” The questionnaire asked the respondents to rank various qualities from 1–5 based on importance (1 being “Not at all important” and 5 being “Very important”). Only 3 % answered that knowledge of foreign markets and foreign contacts was a very important quality, and only an additional 6 % considered this knowledge somewhat important. Conversely, 52 % answered that this knowledge was not at all important.

  20. 20.

    For full descriptive statistics, see the web-appendix.

  21. 21.

    Before using these data we had to take into account numerous and substantial changes to the nomenclature. Pierce and Schott (2012) showed that failure to do so can substantially inflate trade expansion at the extensive margin of trade and deflate it at the intensive margin of trade. We constructed a detailed concordance of the CN8 between 1998 and 2007 that was matched with trade data. This concordance was created by applying the algorithm developed by Pierce and Schott (2012) for the 10-digit US nomenclature to the EU context.

  22. 22.

    In line with the literature, other factors positively associated with firms’ trade include having a larger workforce, being a multinational enterprise, having foreign trade experience, having higher labor productivity, and having physical capital intensity. In addition, the partner country’s GDP has a positive and significant relationship to firm trade, as expected.

  23. 23.

    The yearly selection equation results and subsequent the target equation outcome with the estimated coefficients of the interactions between the yearly inverted mills ratios and year dummies are displayed in the web-appendix.

  24. 24.

    Our result for exports are approximately 20 and 80 % lower than in the only other employee–employer study for main and selection-adjusted results, respectively (Hiller 2013). We attribute our lower estimates to the inclusion of the immigrant stock, imstock jt , which is likely to bias results upward when omitted.

  25. 25.

    Previous (macro-level) evidence for Sweden has also found a positive link (Hatzigeorgiou 2010b; Hatzigeorgiou and Lodefalk 2015).

  26. 26.

    Bastos and Silva (2010) confirmed the validity of such a binary classification using firm-product-country data for Portugal in 2005. Export unit values differ substantially more within a product category at the 8-digit CN level for differentiated goods.

  27. 27.

    In 2013, 57 % of residence permits granted to immigrants were allocated to refugees, asylum seekers or their relatives.

  28. 28.

    Based on the literature, we also expected that the correlation is stronger between migration and trade for countries farther away from the host country, but this had not been examined at the firm level. To consider this, we divided countries into nine regions and reran the within-firm-country estimations of Eq. (4). The results are available in the web-appendix.

  29. 29.

    Skill is likely to be positively related to occupational position, for which we unfortunately have not been able to obtain data.

  30. 30.

    However, Piperakis (2011) found no impact of high-skilled for the EU-15, using education level in the source country as a proxy.

  31. 31.

    Partridge and Furtan (2008) found that length of stay affects exports positively but imports negatively.

  32. 32.

    The studies examined shipments of Spanish provinces in eight-digit products, four-digit product margins for Switzerland, and two-digit margins for Denmark. Additionally, Coughlin and Wall (2011) studied industry margins for US states’ foreign trade and found a positive immigrant impact on trade due to expansion at the intensive margin.

  33. 33.

    The result of Peri and Requena-Silvente (2010) is likely to be upwardly biased, using province shipments instead of firm trade. As illustrated in Hiller (2013), sample selection may also affect the results on the margins of trade.

  34. 34.

    Admittedly, even at the eight-digit product level, there is substantial heterogeneity within products included. Unfortunately, statistics at a more detailed level are not reported in the EU.

  35. 35.

    The reported panel selection estimates are omitted here because of space limitations but are found in the web-appendix.

  36. 36.

    These results are provided in the web-appendix.

  37. 37.

    These results and related descriptive statistics are available upon request. We are grateful to one referee for suggesting this investigation.

  38. 38.

    For brevity, the results are presented in the web-appendix. Because the panel selection approach is very demanding and because we focus on export intensity, the analysis is mainly carried out using within-estimation of Eq. (4).


  1. Aguiar A, Walmsley T, Abrevaya J (2007) Effects of bilateral trade on migration flows: the case of the United States, Draft paper presented at the 10th Annual Conference on Global Economic Analysis, Purdue University, USA

  2. Aleksynska M, Peri G (2014) Isolating the network effect of immigrants on trade. World Econ 37(3):434–455

    Article  Google Scholar 

  3. Andrews M, Schank T, Upward R (2006) Practical fixed-effects estimation methods for the three way error-components model. Stata J 6(4):461–481

    Google Scholar 

  4. Armenter R, Koren M (2009) Economies of scale and the size of exporters, Federal Reserve Bank of Philadelphia, Working Paper: 09–15

  5. Bastos P, Silva J (2010) Identifying vertically differentiated products. Econ Lett 106(1):32–34

    Article  Google Scholar 

  6. Bastos P, Silva J (2012) Networks, firms, and trade. J Int Econ 87(2):352–364

    Article  Google Scholar 

  7. Becker G (1957) The economics of discrimination. University of Chicago Press, Chicago

    Google Scholar 

  8. Bernard A, Bradford Jensen J (1995) Exporters, jobs, and wages in U.S. manufacturing: 1976-1987, Brookings Papers on Economic Activity, Microeconomics, pp. 67–119.

  9. Bernard A, Bradford Jensen J (1999) Exceptional exporter performance: cause, effect, or both? J Int Econ 47:1–25

    Article  Google Scholar 

  10. Bratti M, De Benedictis L, Santoni G (2014) On the pro-trade effects of immigrants. Rev World Econ 150(3):557–594

    Article  Google Scholar 

  11. Carlsson M, Rooth D-O (2007) Evidence of ethnic discrimination in the swedish labor market using experimental data. Labour Econ 14(4):716–729

    Article  Google Scholar 

  12. Casella A, Rauch J (2002) Anonymous market and group ties in international trade. J Int Econ 58(1):19–47

    Article  Google Scholar 

  13. Chaney T (2008) The intensive and extensive margins of international trade. Am Econ Rev 98(4):1707–1721

    Article  Google Scholar 

  14. Chaney T (2011) The network structure of international trade, NBER, Working Paper: 16753

  15. Coughlin C, Wall H (2011) Ethnic networks and trade: intensive versus extensive margins. Econ Lett 113(1):73–75

    Article  Google Scholar 

  16. Crozet M, Head K, Mayer T (2012) Quality sorting and trade: firm-level evidence for french wine. Rev Econ Stud 79(2):609–644

    Article  Google Scholar 

  17. Dunlevy J, Hutchinson W (1999) The impact of immigration on American import trade in the late nineteenth and early twentieth centuries. J Econ Hist 59(4):1043–1062

  18. Eaton J, Kortum S, Kramarz F (2011) An anatomy of international trade: evidence from french firms. Econometrica 79(5):1453–1498

    Article  Google Scholar 

  19. Egger P, von Ehrlich M, Nelson D (2012). Migration and Trade. The World Economy, 35(2): 216–241

  20. Felbermayr GJ, Jung B (2009) The pro-trade effect of the brain drain: sorting out confounding factors. Econ Lett 104(2):72–75

    Article  Google Scholar 

  21. Genç M, Gheasi M, Nijkamp P, Poot J (2011) The impact of immigration on international trade: a meta-analysis, Norface Migration Discussion Paper: 11–20

  22. Gómez-Herrera E (2013) Comparing alternative methods to estimate gravity models of bilateral trade. Empir Econ 44(3):1087–1111

    Article  Google Scholar 

  23. Gould D (1994) Immigrant links to the home country: empirical implications for U.S. bilateral trade flows. Rev Econ Stat 76(2):302–316

    Article  Google Scholar 

  24. Greif A (1989) Reputation and coalitions in medieval trade: evidence on the Maghribi traders. J Econ Hist 49(9):857–882

    Article  Google Scholar 

  25. Hatzigeorgiou A (2010a) Migration as trade facilitation: assessing the links between international trade and migration. The BE J Econ Anal Policy 10(1), 24

    Google Scholar 

  26. Hatzigeorgiou A (2010b) Does immigration stimulate foreign trade? Evidence from Sweden. J Econ Integrat 25(2):376–402

    Article  Google Scholar 

  27. Hatzigeorgiou A, Lodefalk M (2015) Trade, migration and integration – Evidence and Policy Implications. World Econ 38(13):2013–2048

  28. Head K, Ries J (1998) Immigration and trade creation: econometric evidence from Canada. Can J Econ 31(1):47–62

    Article  Google Scholar 

  29. Heckman J (1979) Sample selection bias as a specification error. Econometrica 47(1):153–161

    Article  Google Scholar 

  30. Helpman E, Melitz M, Rubinstein Y (2008) Estimating trade flows: trading partners and trading volumes. Q J Econ 123(2):441–487

    Article  Google Scholar 

  31. Herander M, Saavedra L (2005) Exports and the structure of immigrant-based networks: the role of geographic proximity. Rev Econ Stat 87(2):323–335

    Article  Google Scholar 

  32. Hiller S (2013) Does immigrant employment matter for export sales? evidence from Denmark. Rev World Econ 149(2):369–394

    Article  Google Scholar 

  33. Jansen M, Piermartini R (2009) Temporary migration and bilateral trade flows. World Economy 32(5):735–753

    Article  Google Scholar 

  34. Javorcik B, Özden Ç, Spatareanu M, Neagu C (2011) Migrant networks and foreign direct investment. J Dev Econ 94(2):231–241

    Article  Google Scholar 

  35. Koenig P (2009) Immigration and the export decision to the home country, Paris School of Economics, Working Paper: 2009–31

  36. le Grand C, Szulkin R (2002) Permanent disadvantage or gradual integration: explaining the immigrant-native earnings gap in Sweden. Labour 16(1):37–64

    Article  Google Scholar 

  37. Lopéz R (2009) Do firms increase productivity in order to become exporters? Oxf Bull Econ Stat 71(5):621–642

    Article  Google Scholar 

  38. Melitz M (2003) The impact of trade on intra-industry reallocations and aggregate industry productivity. Econometrica 71:1695–1725

    Article  Google Scholar 

  39. Mundlak Y (1978) On the pooling of time series and cross section data. Econometrica 46(1):69–85

    Article  Google Scholar 

  40. Mundra K (2012) Immigration and trade creation for the US: The Role of Immigrant Occupation, IZA, Discussion Paper: 7073

  41. OECD (2013) International migration. OECD, Paris

  42. Partridge J, Furtan H (2008) Immigration wave effects on Canada’s trade flows. Canadian Public Policy 34(2):193–214

    Article  Google Scholar 

  43. Pennerstorfer D (2012) Export, migration, and costs of market entry: evidence from Central European firms, WIFO, Working Paper: 405

  44. Peri G, Requena-Silvente F (2010) The trade creation effect of immigrants: evidence from the remarkable case of Spain. Can J Econ 43(4):1433–1459

    Article  Google Scholar 

  45. Phelps E (1972) The statistical theory of racism and sexism. Am Econ Rev 62(4):659–661

    Google Scholar 

  46. Pierce JR, Schott PK (2012) Concording US harmonized system categories over time. J Off Stat 28(1):53–68

    Google Scholar 

  47. Piperakis A (2011) The impact of immigration on EU-15 trade. Int Trade J 25(5):558–580

    Article  Google Scholar 

  48. Rauch J (1991) Reconciling the pattern of trade with the pattern of migration. Am Econ Rev 81(4):775–796

    Google Scholar 

  49. Rauch J (1999) Networks versus markets in international trade. J Int Econ 48(1):7–35

    Article  Google Scholar 

  50. Rauch J, Trindade V (2002) Ethnic Chinese networks in international trade. Rev Econ Stat 84(1):116–130

    Article  Google Scholar 

  51. Santos Silva J, Tenreyro S (2006) The log of gravity. Rev Econ Stat 88(4):641–658

    Article  Google Scholar 

  52. Vézina P-L (2012) How migrant networks facilitate trade: evidence from swiss exports. Swiss J Econ Stat 148(3):449–476

    Google Scholar 

  53. White R (2007) Immigrant-trade links, transplanted home bias and network effects. Appl Econ 39(7):839–852

    Article  Google Scholar 

  54. Wooldridge J (2002) Econometric analysis of cross section and panel data. MIT Press, Cambridge

    Google Scholar 

  55. World Bank (2013) Doing business surveys, Available at: [Accessed March 2013]

Download references


The authors are grateful for useful comments and suggestions from Martin Andersson, Peter Egger, Zouheir El-Sahli, Joakim Gullstrand, Pär Hansson, Sune Karlsson, Håkan Nordström, Stephen Redding, Selen Savsin, Fredrik Sjöholm, Robert Stern, Dean Yang, and from participants at presentations at the 2011 European Trade Study Group meeting, the Swedish National Board of Trade, Lund University, Örebro University, and The Ratio Institute, as well as the editor and four anonymous referees of the Journal of Industry, Competition and Trade. The authors are thankful to the Swedish Federation of Business Owners for including questions concerning the hiring decisions of Swedish firms in one of their surveys for the specific purpose of this study. The findings, interpretations, and conclusions expressed are entirely those of the authors.

Author information



Corresponding author

Correspondence to Andreas Hatzigeorgiou.

Electronic supplementary material

Below is the link to the electronic supplementary material.


(PDF 734 kb)

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Hatzigeorgiou, A., Lodefalk, M. Migrants’ Influence on Firm-level Exports. J Ind Compet Trade 16, 477–497 (2016).

Download citation


  • Trade
  • Migration
  • Firm-level analysis
  • Trade costs
  • Networks
  • Information
  • Trust

JEL Classification

  • F 22
  • D 83