Comparing the immigrant-native pay gap: A novel evidence from home and host countries

Cupák, Andrej; Ciaian, Pavel; Kancs, D'Artis Working Paper Comparing the immigrant-native pay gap: A novel evidence from home and host countries JRC Working Papers in Economics and Finance, No. 2023/3 Provided in Cooperation with: Joint Research Centre (JRC), European Commission Suggested Citation: Cupák, Andrej; Ciaian, Pavel; Kancs, D'Artis (2023) : Comparing the immigrantnative pay gap: A novel evidence from home and host countries, JRC Working Papers in Economics and Finance, No. 2023/3, European Commission, Ispra This Version is available at: https://hdl.handle.net/10419/283091 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ Comparing the Immigrant-Native Pay Gap A novel evidence from home and host countries Cupák, A., Ciaian, P., Kancs, D. 2023 JRC Working Papers in Economics and Finance, 2023/3 This publication is a Working Paper by the Joint Research Centre (JRC), the European Commission’s science and knowledge service. It aims to provide evidence-based scientific support to the European policymaking process. Working Papers are pre-publication versions of technical papers, academic articles, book chapters, or reviews. Authors may release working papers to share ideas or to receive feedback on their work. This is done before the author submits the final version of the paper to a peer reviewed journal or conference for publication. Working papers can be cited by other peer-reviewed work. The contents of this publication do not necessarily reflect the position or opinion of the European Commission. Neither the European Commission nor any person acting on behalf of the Commission is responsible for the use that might be made of this publication. For information on the methodology and quality underlying the data used in this publication for which the source is neither Eurostat nor other Commission services, users should contact the referenced source. The designations employed and the presentation of material on the maps do not imply the expression of any opinion whatsoever on the part of the European Union concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Contact information Name: d'Artis Kancs Address: European Commission, DG Joint Research Centre Email: d'[email protected].eu Tel.: +32 229-59203 EU Science Hub https://joint-research-centre.ec.europa.eu JRC125375 Ispra: European Commission, 2023. © European Union, 2023 The reuse policy of the European Commission documents is implemented by the Commission Decision 2011/833/EU of 12 December 2011 on the reuse of Commission documents (OJ L 330, 14.12.2011, p. 39). Unless otherwise noted, the reuse of this document is authorised under the Creative Commons Attribution 4.0 International (CC BY 4.0) licence (https://creativecommons.org/licenses/by/4.0/). This means that reuse is allowed provided appropriate credit is given and any changes are indicated. For any use or reproduction of photos or other material that is not owned by the European Union/European Atomic Energy Community, permission must be sought directly from the copyright holders. How to cite this report: Cupák, A., Ciaian, P. and Kancs, D. (2023). Comparing the Immigrant-Native Pay Gap: A Novel Evidence from Home and Host Countries, Joint Research Centre, European Commission, Ispra, 2023, JRC125375. Executive summary The literature has robustly documented a negative migrant-native wage gap in developed economies. Yet empirical evidence of pay differences has been elusive for developing countries. We approach this question by leveraging internationally harmonised microdata with 1.5 million individuals from 6 transition and developing countries and 15 OECD economies spanning from 1995 to 2016 and employ counterfactual decomposition techniques which allow us to control for individual-productivity and job-specific characteristics, and explain up to 72% of the observed immigrant-native wage gap. The Blinder-Oaxaca baseline results indicate that, vis-à- vis workers born in developed economies the pay for workers born in transition and developing economies is discounted both in their home country labour markets and – if migrating – also in developed host country labour markets. The estimated Blinder-Oaxaca wage differentials suggest the opposite holds for workers born in developed countries – their wages are higher not only in developed countries but for migrants also in developing host countries. These results are novel and have not been reported for developing countries in a cross-country setup. Our estimates also show that in the developed country sub-sample, the mean immigrant wage disadvantage has remained nearly unchanged over the last two decades both in terms of the trend and variance. The magnitude and growth rate of the mean wage gap for the transition/developing economies subsample is similar to developed economies though with the opposite sign – native-born workers in developing countries systematically receive lower wages than foreign-born workers. During the two decades, the unexplained wage gap – attributable to the labour market discrimination, differences in unobserved job characteristics, variation in unobserved skills and the institutional framework of labour market – has remained at a non-trivial magnitude. Complementing the quantitative microanalysis, we have also provided a narrative evidence of the unexplained gap of native-born wages vis-à-vis immigrants and attempted to relate potential explanations to the key sets of factors identified in the literature: group differences in the labour force attachment due to labour market discrimination, differences in unobserved job characteristics, and differences in unobserved skills. Correlation analyses suggest that labour market discrimination is not of a first-order importance in the link between cross-country variation in the unexplained wage gap – in contrary to an often speculated determinant. A more important factor driving the cross-country variation of unexplained native-to-migrant wage gaps appear to be the cross-country variation in unobserved job characteristics (e.g., distribution between native-born and foreign-born workers by skill levels, temporary contract, complementarities between immigrant and native workers and/or wage competition) and unobserved skills among foreign-born and native-born workers (e.g., imperfect transferability of migrants’ skills, language proficiency, literacy skills, numeracy skills or problem solving skills, time spent in the host country by migrants). These results provide a robust cross-country evidence strengthening previous literature findings based on single country data. From a policy perspective, our findings hint to an untapped potential of economic gains at the aggregated level. In addition to ethical and social considerations, lower demand and lower wages for equally productive foreign workers results in a waste of valuable human capital resources. Our findings contribute to the growing body of literature that shows that eliminating distortions in the allocation of talent can result in sizeable productivity and welfare gains. For example, Hsieh et al. (2019) estimate large gains for the U.S. between 1960 and 2010 – their study focuses on race- and gender-based distortions. Kancs and Lecca (2018) find that although the immigrant integration (e.g., by the providing language and professional training) is costly for the host country budget, in the medium- to long-run, the social, economic and fiscal benefits can significantly outweigh the short-run immigrant integration costs in the EU. Contents 1.Introduction . ....... ....... ........ ....... ........ ....... ........ ....... ........ ....... ........ ....... ........ ....... ....... ........ ....................................................................................................................5 2.Empirical strategy ............................................................................................................................................................................................................................8 2.1.Decomposition analysis .................................................................................................................................................................................................8 2.2.Unexplained wage gaps and institutions ........................................................................................................................................................9 3.Data and variable construction ............................................................................................................................................................................................9 3.1.Definitions ........... ....... ....... ........ ....... ........ ....... ........ ....... ........ ....... ........ ....... ....... ........ ....... ........ .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... . 10 3.2.Luxembourg Income Study data ....................................................................................................................................... ... .. ... .. ... .. ... .. ... .. ... .. ... . 10  3.3.Variable construction .................................................................................................................................. ... .. ... .. ... .. ... .. .. ... .. ... .. ... .. ... .. ... .. ... .. ... ..... .. ... . 11  3.4.Institutional characteristics........................................................................................................................................ ... .. .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... . 11 4.Results and Discussion ............................................................................................................................... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... . 13 4.1.Raw wage gaps between native-born and foreign-born workers .......................................................................................... 13 4.2.Unexplained wage gap: controlling for productivity differences .......................................................................... ............... .... 17 4.3.Robustness and further analysis .................................................................................................................................. .. .. ... .. ... .. ... .. ... .. ... .. ... .. ... . 21  4.4.Unexplained wage gaps and institutions ....................................................................................................................................... ... .. ... .. ... . 22 5.Conclusions .................................................................................................................................... ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... . 25  References ........................................................................................................................................ .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... . 27  List of abbreviations and definitions ........ ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... . 31 List of figures ....................................................................................................................................... ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... . 32  List of tables ............ ..... ..... ..... .... ..... ..... ..... ..... ..... ..... ..... ..... ..... ..... ..... ..... ..... ..... ..... ..... ..... ..... ..... ..... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... . 33  Appendix ........................................................................................................................................ .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... . 34  Appendix A: Data and variable construction .......................................................................................................................................... .. ... .. ... .. ... . 34  Appendix B: Validity of migrant share in the LIS database ...................................................................................................................... 39 References for appendix ....... ....... ........ ....... ........ ....... ........ ....... ........ ....... ........ ....... ....... ........ ....... ........ ..... ....... ... .. ..... ... .. ... .. ... .. ... .. .. ... .. ... .. ... .. ... .. ... .. ... .. ... .. ... . 41 Comparing the immigrant-native pay gap: A novel evidence from home and host countries Andrej Cupák * Corresponding author National Bank of Slovakia Imricha Karvaša 1 813 25 Bratislava Slovakia and University of Economics in Bratislava Dolnozemská cesta 1 852 35 Bratislava Slovakia E-mail: [email protected] Phone: 00421 902 408 990 Pavel Ciaian European Commission Joint Research Centre Via Enrico Fermi, 2749 21027 Ispra (VA) Italy E-mail: [email protected] d'Artis Kancs European Commission Joint Research Centre Via Enrico Fermi, 2749 21027 Ispra (VA) Italy E-mail: d'[email protected] ACKNOWLEGEMENTS We gratefully acknowledge financial support received from the Slovak Research and Development Agency under the contract No. APVV-20-0359 “Rural development and agricultural employment: the role of policies, globalisation and climate changes.” The authors are grateful for helpful comments from participants at the Econometric Network meetings. The authors acknowledge helpful comments from the editor Peter Benczur as well as two reviewers for helpful comments. The authors are solely responsible for the content of the report. The views expressed are purely those of the authors and may not in any circumstances be regarded as stating an official position of the European Commission or the National Bank of Slovakia. All remaining errors are ours. Comparing the immigrant-native pay gap: A novel evidence from home and host countries Abstract The literature has robustly documented a negative migrant-native wage gap in developed economies. Yet empirical evidence of pay differences has been elusive for developing countries. We approach this question by leveraging internationally harmonised microdata with 1.5 million individuals from 6 transition and developing countries and 15 OECD economies spanning from 1995 to 2016 and employ counterfactual decomposition techniques which allow us to control for individual-productivity and job-specific characteristics, and explain up to 72% of the observed immigrant-native wage gap. The Blinder-Oaxaca baseline results indicate that, vis-à- vis comparable workers born in developed economies, the pay for workers born in transition and developing economies is discounted both in their home country labour markets and – if migrating – also in developed host country labour markets. However, the unexplained native-to-migrant wage gap remains sizeable in most countries even after controlling for productivity differentials (28% and more). Cross-country correlation analyses contribute a direct empirical support to the link between variation in unobserved job characteristics and skills among foreign-born and native-born workers and wage gap, while the labour market discrimination environment is of a second-order importance. Keywords: Labour market, wage gaps, immigrants, decomposition. JEL codes: D31, J15, J7. 1. Introduction In 2020, there were around 276 million international migrants comprising a continuously growing share of the world’s population (United Nations 2022). The majority – estimated 169 million – have relocated to another country for a work purpose, more than half of them from developing countries (IOM 2022). Whereas 37.6% of migrants residing in a developed country were from another developed country, only 11.7% of migrants to a developing country were from a developed country; the remaining originating from developing countries. Typically, in developing economies the immigrants’ contribution to value added exceeds their population share in employment (OECD 2018). For developed economies, the labour income ratio of foreign- to native-born workers is more nuanced and differentiated by the migrant home country, skill level and the sector of activity, indicating possibly differing levels of productivity (Lemieux 2006). The empirical literature has robustly documented substantial employment and earnings differentials between immigrants and native-born, suggesting that the foreign labour is often treated as an imperfect substitute for native-born in host labour markets. Studies on developed countries report a significant wage advantage for native-born compared to immigrant workers (e.g., Lehmer and Ludsteck (2011) for Germany; Van Kerm et al. (2016) for Luxembourg; Long hi et al. (2013) for the UK; Abbott and Beach (1992), Huffman (2004), Ruist (2013), Bertrand and Mullainathan (2004), and Smith and Fernandez (2017) for the US and Canada). The relative employment disadvantage for foreign-born workers is similarly persistent. For instance, in a review of 36 studies Zschirnt and Ruedin (2016) find that a median call-back rate for minorities relative to native-born whites is only 67% in OECD countries, implying that employers tend to set a significantly higher bar for foreign workers, or avoid hiring it altogether. In contrast, the scarce evidence available for developing economies suggest that the relative labour income ratio of foreign- to native-born workers exceeds unity. The price for a comparable foreign work is found substantially higher, for example, in Kyrgyzstan (22%-25%), Rwanda (12%-15%) and Ghana (12%) (OECD/ILO 2018). In South Africa, newly arrived immigrant workers are found to increase the negative wage gap between native-born and immigrant workers. Gerard et al. (2021) estimate an ethnic wage gap between whites and non-whites natives in Brazil in a range of 27% to 33%, disproportionately disadvantaging the nonwhite native population. An important limitation of these studies is that country-specific data and differences in methods employed do not allow to assess how robust and comparable are these estimates with those for developed countries. The present report approaches these cross-country and over-time comparability issues by leveraging internationally harmonised microdata – the Luxembourg Income Study (LIS) – with 1.5 million individuals for 6 transition/developing economies and 15 OECD countries spanning from 1995 to 2016. The LIS microdata have been used for cross-country studies before, though in different contexts of migration (e.g. Anastossova and Paligorova 2006; Birinci et al. 2021). We employ counterfactual decomposition techniques to compute the levels of wage differentials and inequality trends of foreign-born and native-born workers. The Blinder- Oaxaca decomposition technique has been extensively used in the empirical labour literature to study gaps in wages and employment across different groups (e.g., Oaxaca and Ransom 1994). Controlling for individual-productivity and job-specific characteristics allows us to explain up to 72% of the observed immigrant-native pay gap. The Blinder-Oaxaca estimates show that, vis-à-vis comparable workers born in developed economies, the pay for workers born in transition and developing economies is discounted both in their home country labour markets and – if migrating – also in developed host country labour markets. Benchmarking these estimates to those for workers born in developed countries – in line with literature – the estimated wage differentials are negative. Further, in the developing economy sub-sample, the mean immigrant wage differential has remained nearly unchanged during the last two decades in terms of both the trend and variance. For comparison, the magnitude and growth rate of the mean wage gap for the developed countries sub-sample is similar to developed economies though with the opposite sign – comparable nativeborn workers in developing countries systematically receive lower wages than foreign-born workers at the mean. These cross-country and over-time comparable estimates are novel for developing countries. Interestingly, the wage gap shrinks as the immigrants live longer (e.g., more than 15 years) in the host developed countries. The opposite holds true for transition and developing host countries: the longer the immigrants live in these countries, the more they earn compared to the similar native population. Our report is related to the large body of the inequality literature, showing that a significant part of the observed raw differences in labour market outcomes between heterogeneous groups of workers can be explained by productivity differences (Dustmann and van Soest 2002; Ferrer et al. 2006; Hellerstein and Neumark 2003; Bratsberg and Ragan 2002). Two sources for productivity differentials have been studied in the literature: intrinsic productivity differences between immigrants and native-born, and segregation into labour market categories with a differentiated productivity (García-Pérez et al. 2014).1 A step-by-step decomposition shows that observed productivity differences alone cannot explain fully why heterogeneous groups of workers receive different wages for an otherwise comparable work when being employed. Even when controlling for individual and work characteristics, including industry and employment type, education and skills intensity, etc., 28% and more of the total native-to-migrant wage gap still remains unexplained across the 21 countries covered in this report. The result on shrinking wage gaps based on the years of residence in the host country ties in with the previous literature on immigrants’ assimilation (e.g., Izquierdo et al. 2009). The literature attributes a large fraction of the unexplained wage gap to differences in the labour market discrimination between migrants and natives (e.g., Lehmer and Ludsteck 2011). Other frequent causes of the unexplained wage gap studied in the literature are unmeasured/unobserved (omitted in the data) productivity differentials between immigrants and natives and factors affecting it, particularly, when conducting estimations at the individual level. Tang et al. (2020) decomposition shows that in the US the within-job inequality accounts for more than 80% of the wage inequality between 1983 and 2013. Similar results were found for the immigrant wage inequality by García-Pérez et al. (2014) and Izquierdo et al. (2009), Himmler 1 Whereas intrinsic productivity effects capture differences between natives and immigrants within the same category (e.g., unequal productivity between immigrants and natives within the same occupation), sorting refers to differences in the distribution of natives and immigrants between categories that each encompasses a distinct level of productivity (e.g., over-representation of immigrants in occupations with lower productivity/wage) (Autor and Katz 1999; Lemieux 2006). can alter wage and employment outcomes which ultimately dependents on particular country market conditions (Katz and Autor 1999). The variation of unobserved skills (or those not available in LIS database) among foreign-born and native-born workers across countries may be, among others, due to the crosscountry variation in the distribution of the transferability, comparability and recognition of migrants’ experience and education to the host country (e.g. over-qualification or under-qualification effects), language proficiency, literacy skills, numeracy skills or problem solving skills (Dustmann and van Soest 2002). To account for unobserved skills we use the following variables in the correlation analyses: (i) the ratio of foreign-born to native-born over-qualification rates form OECD, and (ii) the share of immigrants born in a high-income country from OECD (iii) the ratio of foreign-born to native-born in literacy, numeracy and problem solving indicators from the Programme for the International Assessment of Adult Competencies (PIAAC) database of the OECD, (iv) the ratio of immigrants not speaking the host-country language to those that do, (v) the share of immigrants that are multilingual native speakers from OECD (for more details, see Appendix A.3). 4. Results and Discussion 4.1. Raw wage gaps between native-born and foreign-born workers The Blinder-Oaxaca wage differentials (i.e., total wage gaps including both explained and unexplained parts) are reported in Table 1, where the earnings differentials between native-born and immigrant workers are presented in percentage points. First, notice a significant heterogeneity in the native-born/immigrant earnings differentials between countries in the harmonised and hence directly comparable sample. For example, whereas in Luxembourg on average native-born workers receive one third higher salary than migrant workers (+34.34%) (migrants are disadvantaged), in Brazil on average native-born workers are paid only half of what migrant workers are paid (-48.72%) (migrants are advantaged) (column ‘1995-2016’ in Table 1). Second, the Blinder-Oaxaca wage differentials are strikingly consistent within the two country groups (‘developed’ and ‘transition/developing’). The total observed pay gap between native born and migrant workers is positive and statistically significant for all OECD economies in our sample implying that, on average, immigrant workers face a wage disadvantage in advanced economies (column ‘1995-2016’ and top panel in Table 1). These results are in line with previous estimates for developed countries which tend to find positive native-to-migrant wage gap (e.g., Baker and Benjamin 1994; Chiswick and Miller 2008; Lehmer and Ludsteck 2011; Van Kerm et al. 2016; Longhi et al. 2013; Ruist 2013, Bertrand and Mullainathan 2004; Smith and Fernandez 2017). Table 1: Raw native-to-migrant percent wage gap Mean wage difference, % 1995-2016 1995-2000 2001-2010 2011-2016 Developedeconomies Austria 22.13 23.55 25.24 17.61 Canada 11.55 12.38 12.88 9.41 Czechia 2.40 7.05 3.22 -3.06 Estonia 17.16 28.15 23.33 Germany 7.53 0.74 10.21 11.63 Greece 30.76 34.63 28.16 29.50 Iceland 10.61 11.88 19.94 Ireland 8.82 5.38 5.29 15.79 Israel 12.53 21.94 15.66 Italy 18.83 5.98 22.33 28.17 Luxembourg 34.34 35.91 31.06 36.06 The Netherlands 6.27 9.39 9.43 Spain 18.82 23.47 32.98 Switzerland 4.42 7.11 6.14 The United States 10.21 10.21 11.38 9.05 Transitionanddevelopingeconomies Brazil -48.72 -51.00 -46.45 Chile -23.53 -29.43 -17.63 Guatemala -36.90 -42.53 -31.28 India -7.69 -9.66 -5.72 Paraguay -23.85 -18.28 -30.35 -22.91 South Africa -19.21 -20.37 -18.06 Notes: Missing values imply no LIS data are available for the specific country-period. Source: Estimated results based on Luxembourg Income Study data. In contrast, Blinder-Oaxaca wage differentials are negative and statistically significant for all transition and developing economies in the LIS sample, implying that, on average, the relative mean wages of immigrant workers are higher than those of native-born workers (bottom panel in Table 1). The wage disadvantage for native-born vis-à-vis immigrant workers ranges from -7.7% in India to -48.7% in Brazil. In all six analysed transition and developing economies the wage differentials have been narrowing slightly during the last two decades (see last two columns in Table 1). These results are novel, they are striking though not necessarily surprising when compared to single-country studies. For example, OECD/ILO (2018) have estimated that in South Africa, newly arrived immigrant workers increase the wage gap between native-born and immigrant workers. Gerard et al. (2021) have estimated negative wage gaps between non-whites natives and whites in Brazil in a range of 27% to 33%. The methodological consistency of the LIS harmonised data across countries and over time allows us to comparably assess both the inter-national and inter-temporal dimension of labour earnings inequality by immigration status. Columns 3-5 in Table 1 and bold lines in Figure 1 report the development of the relative mean wages of immigrant workers vis-à-vis native-born workers during the last two decades. The differencein-differences perspective suggests that the mean wage gap of immigrant workers vis-à-vis native-born workers has remained of the same order of magnitude in most developed economies (top panel in Table 1). A similar pattern can be observed for most transition and developing economies in our sample – the mean wage differential of immigrant workers vis-à-vis native-born workers has changed (narrowed) little (bottom panel in Table 1). Figure 1: Raw native-to-migrant percent wage advantage in developed economies and transition/developing economies Notes: Positive wage gap indicates the percentage by which the wages of native-born workers exceed those of the foreign-born. Source: Estimated results based on Luxembourg Income Study data for wage gaps and the UN Population Division, Trends in Total Migrant Stock data for migrant population weights used to calculate the mean wage gap for the two country groups. Capturing both dimensions, Figure 1 plots a weighted average of these inequality trends between foreignborn and native-born workers across developed economies (solid line) and economies in transition and developing economies (dashed line). Indeed, the average wage inequality trends (solid and dashed lines in Figure 1) and the cross-country wage dispersion (shaded area in Figure 1) have changed insignificantly during the last 15 years. It is a well-established in the literature that the average earnings of immigrants differ from those of natives, among others, depending on the migrant country of origin and time spent in the host country (e.g., Adsera and Chiswick 2007). In order to investigate the impact of the length of immigrant experience in the host country, we split our sample into three cohorts: migrants having lived in the host country less than 10 years, 10-15 years and more than 15 years.13 The estimated native-born/immigrant wage differentials for each cohort are reported in Table 2. Table 2: Time in the host country and the percent native-to-migrant wage gap Mean wage difference, % <10 years 10-15 years >15 years Developedeconomies Austria 20.96 18.16 12.93 Canada 26.87 18.74 1.00 Estonia 4.61 8.64 26.53 Germany 28.98 20.75 2.35 Greece 50.29 33.59 19.07 Ireland 12.22 14.16 4.56 Israel 53.89 32.42 0.71 Italy 35.57 28.27 15.01 Luxembourg 37.05 35.10 25.54 Switzerland 5.36 6.80 7.27 The United States 26.12 15.27 -4.24 Transitionanddevelopingeconomies Chile -12.11 -20.19 -34.45 Guatemala -4.06 -44.36 -14.35 South Africa -35.05 -0.62 -16.84 Notes: Missing data on the immigrant time in the host country for Brazil, Czechia, Iceland, India, the Netherlands, Paraguay, Spain. Source: Luxembourg Income Study data. The wage differentials by the length of experience in the host country suggest a sizeable heterogeneity in the LIS sample (Table 2). Both the sign and magnitude of the impact of the time spent in the host country on wage differentials between native-born and migrant workers differ substantially between sample countries. At the same time, we can observe a remarkably consistent pattern within the developed country sub-sample (Austria, Canada, Germany, Greece, Ireland, Israel, Italy, Luxembourg, the United States), where the relative mean wages of immigrants vis-à-vis the native-born decrease in the time spent in the host countries. Estonia and Switzerland are the only developed economies in our sample where the native/immigrant wage differentials are widening – even after longer time periods spent in the host country the immigrant wage disadvantage remains substantial, suggesting that integration may be more challenging for foreigners in Estonia (mainly Russian-speaking immigrants opting not to integrate for ideological reasons, see, e.g. Kielyte 13 Note that information on years of residence is available only for a subset of countries. and Kancs 2002) and Switzerland (which is known for its tough stance on immigrants) compared to other developed economies (see, e.g. Hainmueller and Hangartner 2013).14 In contrast, the wage gap does not seem to be considerably decreasing in the time immigrants have spent in the host transition and developing economies (Chile, Guatemala, South Africa). 4.2. Unexplained wage gap: controlling for productivity differences A significant part of differences in labour market outcomes between heterogeneous groups of workers can be explained by productivity differences. In the Blinder-Oaxaca decomposition, we control for two sources of productivity differentials between immigrants and native-born: intrinsic productivity differences between immigrants and native-born within the same category (age, gender, education, experience, family composition) and sorting into labour market categories with a differentiated productivity (sector of employment, occupation). Controlling for these sources of productivity differences yields a robust estimate of the explained part of the Blinder-Oaxaca wage differentials. The residual (unexplained) wage gaps between native-born and immigrants and its development over time, after controlling for the observable intrinsic and segregation related characteristics are reported in Table 3. Figure 2 displays the size (share) of the unexplained wage gap relative to the total wage gap and in comparison with the explained gap. The Blinder-Oaxaca decomposition results suggest that, after controlling for productivity differentials, the native-to-migrant percent wage advantage becomes smaller in most countries. The exceptions are Iceland, the Netherlands, Estonia and Israel where the productivity differentials (the explained wage gaps) tend to magnify the immigrant worker wage advantage relative to the native-born workers, while the unexplained wage gap remains the major source of the total wage gap (compare Table 1 and Table 3; Figure 2). Overall, the unexplained wage gap remains sizeable in most countries even after controlling for productivity differentials. Its share in the total wage gap varies between 34% and 127% in developed countries and between 28% and 78% in transition and developing economies. With few exceptions (i.e., Germany, the United States, Paraguay), the share of the unexplained wage gap in the total wage gap is greater in developed economies than in transition/developing economies (Figure 2). 14 Hainmueller and Hangartner (2013) document the immigrant discrimination and foreigner integration difficulties in Switzerland using a natural experiment. Table 3: Native-to-migrant percent unexplained wage gap after controlling for productivity differentials Mean wage difference, % 1995-2016 1995-2000 2001-2010 2011-2016 Developedeconomies Austria 16.81 19.34 17.92 13.18 Canada 9.37 10.49 10.64 6.97 Czechia 1.90 4.95 2.01 -1.25 Estonia 21.42 23.40 19.44 Germany 2.58 -0.49 3.03 5.20 Greece 23.05 31.31 20.71 17.14 Iceland 13.52 9.83 17.20 Ireland 6.50 2.46 4.48 12.56 Israel 14.70 17.41 11.99 Italy 11.39 4.88 11.74 17.56 Luxembourg 23.67 19.36 21.96 29.68 The Netherlands 7.83 7.88 7.78 Spain 16.33 15.52 17.14 Switzerland 3.16 3.10 3.22 The United States 3.43 2.95 3.54 3.79 Transitionanddevelopingeconomies Brazil -21.13 -20.35 -21.91 Chile -11.64 -14.81 -8.47 Guatemala -21.83 -28.53 -15.13 India -2.13 -2.08 -2.17 Paraguay -18.69 -13.29 -23.27 -19.51 South Africa -6.07 -1.04 -11.09 Notes: Missing values imply no LIS data are available for the specific country-period. Source: Luxembourg Income Study data. We observe a sizeable heterogeneity in the magnitude of the unexplained wage gap across the LIS sample countries (column ‘1995-2016’ in Table 3), the cross-country heterogeneity being persistent both between and within country groups. The unexplained wage gap remains positive for most developed economies and negative for all studied transition and developing economies. In other words, the unexplained factors cause mean wages of immigrant workers to be lower (higher) than those of the comparable native-born workers in developed countries (transition and developing economies). Figure 3 plots the weighted average unexplained inequality trends between foreign-born and native-born workers for developed economies (solid line) and transition and developing economies (dashed line). The unexplained wage inequality trends (lines in Figure 3) and the cross-country wage dispersion (shaded areas in Figure 3) have increased slightly during the last decade. Figure 2: Distribution of explained and unexplained wage differentials (at the mean) between natives and immigrants across countries (1995-2016, total gap = 100%) Note: Distribution of wage gaps are sorted according to the size (share) of unexplained wage gap within the two country groups. Source: Luxembourg Income Study data. Further, we observe a decrease in the unexplained wage gap in the time spent by migrants in the host countries in most developed countries with the exception of Canada and Switzerland. In contrast, the unexplained wage gap does not seem to be considerably decreasing with the immigrants’ time spent for the host transition/developing economies (Chile, Guatemala, South Africa) (Table 4). Overall, results reported in Table 3, Table 4 and Figure 3 suggest that vis-à-vis workers born in developed economies, the pay for workers born in transition and developing economies is discounted both in their home country labour markets and – if migrating – also in developed host countries. The opposite is true for workers born in developed countries – the estimated Blinder-Oaxaca unexplained wage differentials are positive (negative) vis-à-vis workers born in developing economies in home country (in transition and developing economies). Figure 3: Native-to-migrant percent unexplained wage gap after controlling for productivity differentials in developed economies and transition/developing economies Notes: Positive wage gap indicates the percentage by which the wages of native-born workers exceed those of the foreign-born. Source: Luxembourg Income Study data for wage gaps and the UN Population Division, Trends in Total Migrant Stock data for migrant population weights used to calculate the mean wage gap for the two country groups. Table 4: Time in the host country and percent unexplained wage gap Mean wage difference, % <10 years 10-15 years >15 years Developedeconomies Austria 12.93 9.30 6.59 Canada 16.34 11.88 0.56 Estonia 2.89 4.36 16.61 Germany 15.97 12.72 1.59 Greece 27.76 17.77 12.89 Ireland 6.17 7.12 2.84 Israel 28.62 18.71 0.50 Italy 18.75 17.98 8.09 Luxembourg 21.27 19.97 17.70 Switzerland 3.47 4.06 4.26 The United States 15.83 10.26 2.90 Transitionanddevelopingeconomies Chile -8.40 -13.55 -20.67 Guatemala -2.38 -23.47 -9.70 South Africa -19.84 -0.41 -9.41 Notes: Missing data on the immigrant time in the host country for Brazil, Czechia, Iceland, India, the Netherlands, Paraguay, Spain. Source: Luxembourg Income Study data. 4.3. Robustness and further analysis The analyses in sections 4.1 and 4.2 have focused on computing results obtained from the decomposition of wage differences at the mean. As already introduced in Section 2, we further explore the robustness of our results beyond the mean and decompose the wage gaps at different parts of the wage distribution, namely: p25, p50, and p75. A summary of findings from the quantile decomposition is presented in Figure 4.15 15 Full results obtained from the quantile decomposition are available from authors upon request. Figure 4: Native-to-migrant percent unexplained wage gap at different parts of distribution after controlling for productivity differentials Notes: A positive wage gap indicates the percentage by which the wages of native-born workers exceed those of the foreign-born. Source: Luxembourg Income Study data for wage gaps and the UN Population Division, Trends in Total Migrant Stock data for migrant population weights used to calculate the mean wage gap for the two country groups. Results from the quantile decomposition confirm the baseline results from the mean decomposition: vis-à-vis workers born in developed countries, the pay for workers born in transition and developing economies is discounted both in their home country labour markets and – if migrating – also in developed host countries. However, we can observe two opposite trends between developed and developing/transition economies. While in developed countries we observe larger (unexplained) gap at the bottom of the wage distribution (p25), developing/transition economies exhibit a large (negative) gap in the upper part of the wage distribution (p75). Furthermore, we can observe some divergence over the time, especially among developing/transition countries. 4.4. Unexplained wage gaps and institutions As established in the previous section, the unexplained wage gap varies significantly between countries even within a sample of economies with a comparable wage structure, implying that other sources of wage differentiation must be present given the persistence of the unexplained wage inequalities. Here we provide a suggestive evidence of cross-country differences in the labour market discrimination, unobserved job characteristics, and unobserved skills. Our correlation analyses of discrimination contribute a nuanced empirical support to the link between labour market discrimination against migrants and wage gap. The discrimination and violence against minorities exhibit a negative correlation with the unexplained wage gap in the developing country sub-sample, while the Huffman, M. L. (2004). More pay, more inequality? The influence of average wage levels and the racial composition of jobs on the Black–White wage gap.Social Science Research,33(3), 498–520. ILO (2020). The migrant pay gap: Understanding wage differences between migrants and nationals. International Labour Office, Geneva. IOM (2022). World Migration Report 2022. International Organization for Migration, Geneva. Izquierdo, M., Lacuesta, A., and Vegas, R. (2009). Assimilation of immigrants in Spain: A longitudinal analysis. Labour Economics, 16(6), 669–678. Jann, B. (2008). The Blinder-Oaxaca decomposition for linear regression models. The Stata Journal, 8(4), 453– 479. Kampelmann, S. and Rycx, F. (2016). Wage discrimination against immigrants: measurement with firm-level productivity data.IZA Journal of Migration,5(1), 1–24. Kancs, D. and Lecca, P. (2018). Long-term social, economic and fiscal effects of immigration into the EU: The role of the integration policy. World Economy, 41(10), 2599–2630. Katz, L. F. and D. H. Autor. (1999). Changes in the wage structure and earnings inequality. Handbook of Labor Economics 3, 1463–1555. Kielyte, J. and Kancs, D. (2002). Migration in the Enlarged European Union: Empirical Evidence for Labour Mobility in the Baltic States, Journal of Baltic Studies, 33(3), 259–279. Lee, D.S. (1999). Wage inequality in the United States during the 1980s: Rising dispersion or falling minimum wage? Quarterly Journal of Economics, 114(3), 977–1023. Lehmer, F. and Ludsteck, J. (2011). The immigrant wage gap in Germany: Are east Europeans worse off? International Migration Review, 45(4), 872–906. Lemieux, T. (2006). Increasing Residual Wage Inequality: Composition Effects, Noisy Data, or Rising Demand for Skill? American Economic Review, 96(3), 461–498. Lemieux, T., MacLeod, W.B., and Parent, D. (2009). Performance pay and wage inequality. Quarterly Journal of Economics, 124(1), 1–49. Leuven, E. and Oosterbeek, H. (2011). Overeducation and mismatch in the labor market.Handbook of the Economics of Education,4, 283–326. LIS (2020). Luxembourg Income Study Database. www.lisdatacenter.org (2019–2020). Longhi, S., Nicoletti, C. and Platt, L. (2013). Explained and unexplained wage gaps across the main ethnoreligious groups in Great Britain. Oxford Economic Papers, 65(2), 471–493. Neumark, D., Salas, J.M.I., and Wascher, W. (2014). Revisiting the Minimum Wage- Employment Debate: Throwing Out the Baby with the Bathwater? ILR Review, 67(3), 608– 648. Oaxaca, R. (1973). Male-female wage differentials in urban labor markets. International Economic Review, 14(3), 693–709. Oaxaca, R. and Ransom, M. (1994). On Discrimination and the Decomposition of Wage Differentials. Journal of Econometrics, 61(1), 5–21. OECD/ILO (2018). How Immigrants Contribute to Developing Countries’ Economies. OECD Publishing, Paris. Plasman, R., Rusinek, M., and Rycx, F. (2007). Wages and the bargaining regime under multilevel bargaining: Belgium, Denmark and Spain. European Journal of Industrial Relations, 13(2), 161–180. Ruist, J. (2013). Immigrant–native wage gaps in time series: Complementarities or composition effects? Economics Letters, 119(2), 154–156. Smith, W. C. and Fernandez, F. (2017). Education, Skills, and Wage Gaps in Canada and the United States. International Migration, 55(3), 57–73. Tang, R., Tang, Y., and Wang, P. (2020). Within-Job Wage Inequality: Performance Pay and Job Relatedness. NBER Working Papers 27390, National Bureau of Economic Research. Valentine J., Rycx, F. and Volral, M. (2021). Wage Effects of Educational Mismatch According to Workers’ Origin: The Role of Demographics and Firm Characteristics, GLO Discussion Paper Series 966, Global Labor Organization. Van Kerm, P., Yu, S., and Choe, C. (2016). Decomposing quantile wage gaps: a conditional likelihood approach. Journal of the Royal Statistical Society: Series C (Applied Statistics), 65(4), 507–527. Visser, J. (2019). ICTWSS Data Base on Institutional Characteristics of Trade Unions, Wage Setting, State Intervention and Social Pacts, 1960-2018 (version 6.1). Amsterdam: Amsterdam Institute for Advanced Labour Studies AIAS. World Bank (2020). World Development Indicators. Database. Washington, DC: The World Bank. Available at: https://data.worldbank.org/indicator. Zschirnt, E. and Ruedin, D. (2016). Ethnic discrimination in hiring decisions: a meta-analysis of correspondence tests 1990–2015.Journal of Ethnic and Migration Studies,42(7), 1115–1134. List of abbreviations and definitions B-O Blinder-Oaxaca EU European Union ILO International Labour Organization LIS Luxembourg Income Study OECD Organisation for Economic Co-operation and Development UK United Kingdom US United States List of figures Figure 1: Raw native-to-migrant percent wage advantage in developed economies and transition/developing economies .................................................................................................................................................................................................................... 15 Figure 2: Distribution of explained and unexplained wage differentials (at the mean) between natives and immigrants across countries (1995-2016, total gap = 100%) ..................................................................................................... 19 Figure 3: Native-to-migrant percent unexplained wage gap after controlling for productivity differentials in developed economies and transition/developing economies .......................................................................................................... 20 Figure 4: Native-to-migrant percent unexplained wage gap at different parts of distribution after controlling for productivity differentials ............................................................................................................................................................................. 22 Figure 5: Correlations between the unexplained wage gap and discrimination (A, B) and job characteristics (C, D) ....................................................................................................................................................................................................................................... 23 Figure 6: Correlations between the unexplained wage gap and unobserved skills ............................................................ 24 List of tables Table 1: Raw native-to-migrant percent wage gap .............................................................................................................................. 14 Table 2: Time in the host country and the percent native-to-migrant wage gap ............................................................... 16 Table 3: Native-to-migrant percent unexplained wage gap after controlling for productivity differentials ........ 18 Table 4: Time in the host country and percent unexplained wage gap..................................................................................... 21 Appendix Appendix A: Data and variable construction Table A.1: Country sample used in the empirical analysis Country Wave IV (~ 1995) Wave V (~ 2000) Wave VI (~ 2004) Wave VII (~ 2007) Wave VIII (~ 2010) Wave IX (~ 2013) Wave X (~ 2016) Austria (AT) AT 97 (1,778/189) AT 00 (1,011/44) AT 04 (3,154/499) AT 07 (4,048/759) AT 10 (4,575/843) AT 13 (4,158/812) AT 16 (4,343/866) Brazil (BR) BR 06 (71,379/170) BR 09 (75,380/168) BR 13 (71,694/204) Canada (CA) CA 00 (5,298/1,396) CA 04 (5,444/1,368) CA 07 (5,178/1,413) CA 10 (4,681/1,380) Chile (CL) CL 06 (43,143/407) CL 09 (39,495/420) CL 13 (42,719/1,045) CL 15 (53,314/1,639) Czechia (CZ) CZ 96 (25,925/80) CZ 04 (3,549/44) CZ 07 (7,564/73) CZ 10 (6,887/69) CZ 13 (6,162/68) Estonia (EE) EE 07 (4,179/903) EE 10 (3,668/637) EE 13 (4,282/753) Germany (DE) DE 95 (4,615/1,079) DE 00 (8,403/1,308) DE 04 (7,939/1,194) DE 07 (7,776/1,022) DE 10 (12,200/1,699) DE 13 (10,079/1,898) DE 15 (9,097/2,942) Greece (GR) GR 95 (1,998/96) GR 04 (2,165/275) GR 07 (1,102/110) GR 10 (1,292/141) GR 13 (1,959/156) Guatemala (GT) GT 06 (4,934/43) GT 11 (7,416/36) GT 14 (12,799/42) Iceland (IS) IS 04 (3,781/220) IS 07 (3,410/268) IS 10 (3,296/211) India (IN) IN 04 (34,328/330) IN 11 (37,902/433) Ireland (IE) IE 96 (1,567/107) IE 00 (1,595/86) IE 04 (3,000/391) IE 07 (2,915/393) IE 10 (2,166/523) Israel (IL) IL 07 (2,784/1,903) IL 10 (3,383/1,915) IL 14 (5,162/2,466) IL 16 (5,659/2,390) Italy (IT) IT 95 (4,431/99) IT 00 (4,338/148) IT 04 (4,033/305) IT 08 (3,909/475) IT 10 (3,763/487) IT 14 (3,375/476) Luxembourg (LU) LU 97 (1,173/850) LU 00 (975/982) LU 04 (1,268/1,739) LU 07 (1,217/2,827) LU 10 (2,279/3,153) LU 13 (1,706/2,258) The Netherlands (NL) NL 04 (3,504/201) NL 07 (4,448/223) NL 10 (4,184/244) NL 13 (4,132/225) Paraguay (PY) PY 00 (5,882/376) PY 04 (5,700/289) PY 07 (3,985/141) PY 10 (4,283/163) PY 13 (5,095/156) PY 16 (8,475/323) South Africa (ZA) ZA 08 (3,087/121) ZA 10 (3,748/43) ZA 12 (4,319/79) Spain (ES) ES 04 (6,750/508) ES 07 (10,385/1,083) ES 10 (7,730/680) Switzerland (CH) CH 07 (4,923/1,665) CH 10 (4,643/1,721) CH 13 (4,598/1,471) The United States (US) US 97 (48,445/8,158) US 00 (81,076/13,761) US 04 (75,339/13,858) US 07 (73,880/14,962) US 10 (68,487/14,881) US 13 (47,018/11,022) US 16 (64,115/14,835) Note: Under each country we show the number of native/foreign born population. In our empirical analysis, the total number of natives across countries sums to 1,453,344 individuals, while the total number of immigrants across countries sums to 154,916 individuals. Source: Luxembourg Income Study data. Table A.2: Variable definition used in the decomposition analysis Variable LIS code Description Immigration status IMMIGR All persons who have that country as country of usual residence and (in order of priority):  whom the data provider defined as immigrants  who self-define them-selves as immigrants  who are the citizen/national of another country  who were born in another country Hourly wage GROSS1/NET1 Gross/net basic hourly wage rate for the main job. Overtime payments, bonuses and gratuities, family allowances and other social security payments made by employers, as well as ex gratia payments in kind supplementary to normal wage rates, are all excluded from the calculation of the basic gross hourly wage Employment status EMP Indicator of an employment activity in the current period Industry INDA1 Industry classification of the main job into 3 categories:  agriculture  industry  services Years of residence YRSRESID Cumulative number of years of residence in the country Education EDUC Highest completed level of education:  low: less than secondary education completed (never attended, no completed education or education completed at the ISCED levels 0, 1 or 2)  medium: secondary education completed (completed ISCED levels 3 or 4)  high: tertiary education completed (completed ISCED levels 5 or 6) Gender SEX Classification of persons according to their sex Age AGE Age in years. Note that when original data provide age in intervals, values given are the lowest value of the interval. For example, the intervals 10-14 and 15-19 are coded as 10 and 15, respectively Children NCHILDREN Number of own children living in household Source: Luxembourg Income Study data. Table A.3: Institutional macro-level variables Variable Source Period covered Country groups covered Definition Labour market discrimination Discrimination and violence against minorities index Social Progress Imperative, Social Progress Index Average 2004-2016 over available years DC and TDC The index captures discrimination, powerlessness, ethnic violence, communal violence, sectarian violence, and religious violence, measured on a scale on 0 (low pressures) to 10 (very high pressures) Tolerance for immigrant score Social Progress Imperative, Social Progress Index Average 2004-2016 over available years DC and TDC The percentage of respondents answering yes to the question, “Is the city or area where you live a good place or not a good place to live for immigrants from other countries?” It takes values between 0 (=low tolerance) and 100 (=high tolerance). Job characteristics Ratios of foreignborn to nativeborn in low skill employment ILOSTAT, International labour migration statistics (ILMS) (https://ilostat.ilo.org/topics/labourmigration/) Average 2010-2016 over available years DC Skill levels considered represent occupation categories based on the International Standard Classification of Occupation (ISCO) as follows. Skill level 1 (low): elementary occupations. Skill levels 3 and 4 (high): legislators, senior officials and managers; professionals; technicians and associate professionals (ILOSTAT 2020) Ratios of foreignborn to nativeborn in high skill employment Ratio of foreignborn to nativeborn workers with a temporary contract for low educated workers OECD (2015a) Average 2012-2013 over available years DC The ratio is calculated as foreign-born to native-born workers with a temporary contract represented as percentages of total employment, (persons aged 15-64 not in education) Ratio of foreignborn to nativeborn workers with a temporary contract for highly educated workers Share of migrant stock in the total population United Nations Population Division, International migrant stock 2019 Average 1995-2015 DC and TDC International migrant stock as a percentage of the total population (both sexes) Unobserved skills Ratio of foreignborn to nativeborn overqualification OECD (2015a) Average 2012-2013 over available years DC Ratio of foreign-born to native-born overqualification rates among 15-64 year-olds who are not in education. Overqualification rate is defined as the share of people with tertiary-level qualifications who work in a job that is classified as low- or medium-skilled by the International Standard Classification of Occupations (OECD 2015a) Share of immigrants born OECD (2015a) Average 2010-2011 DC and one TDC Percentage immigrant populations aged 15 to 64 years old and born in a Variable Source Period covered Country groups covered Definition in a high-income country over available years high-income country of the total immigrant population Ratio of foreignborn to nativeborn in literacy, numeracy and problem solving indicators OECD, PIAAC 2012 DC and one TDC The ratio of foreign-born to nativeborn in literacy, numeracy and problem solving is calculated as a simple average over the individual foreign-born to native-born ratios of indicators for literacy, numeracy and problem solving Ratio of immigrants not speaking the hostcountry language to those that do OECD (2015a) 2012 DC The ratio of the share of immigrants not speaking the host-country language at home or are monolingual native speakers to the share of immigrants who host-country language most often spoken at home Share of immigrants that are multilingual native speakers OECD (2015a) 2012 DC The share of immigrants who are multilingual native speakers Notes: DC: developed countries; TDC: transition and developing countries. Source: own processing based on existing data sources. Proxies constructed to control for unobserved job characteristics: (i) the ratio of foreign-born to native-born in low skill employment and high skill employment from ILOSTAT, (ii) the ratio of foreign-born to native-born workers with a temporary contract for low-skill workers and for high-skill workers from OECD and (iii) the share of migration in the total population from the United Nations Population Division. The first set of variables provide an additional description of job characteristics in terms of tasks and duties associated with occupations which is more detailed compared to the occupation variable used in the B-O estimations which controls for the sector of employment (i.e., industry classification). The second set of variables account for inferior (non-standard) forms of employment, which typically feature lower pay and fewer benefits and is more widespread among migrants than natives. The non-standard forms of employment include, among others, temporary employment, part-time work, temporary agency work, seasonal work and dependent selfemployment (Hotchkiss and Pitts 2007; ILO 2015, 2016; OECD 2015b; OECD/ILO 2018). With the share of migration in the total population variable we attempt to proxy the complementarity effect between immigrants and native workers in production. The complementarity effect emerges when immigrants and natives are imperfect substitutes in the production process, e.g., due to different skills, occupation segregation, etc., which may lead to raise in demand for complementary production tasks and skills of natives and thus enhance their wage or may rise price competition among migrant workers and exercise a downward pressure on their wages (D’Amuri et al. 2010; Manacorda et al. 2012; Ottaviano and Peri 2012). Proxies constructed to control for account for unobserved skills: (i) the ratio of foreign-born to native-born over-qualification rates form OECD, and (ii) the share of immigrants born in a high-income country from OECD (iii) the ratio of foreign-born to native-born in literacy, numeracy and problem solving indicators from the Programme for the International Assessment of Adult Competencies (PIAAC) database of the OECD, (iv)