The last few decades have seen unprecedented growth in international migration to higher-income countries. Many European countries nowadays rely on the foreign-born for sustained population and economic growth (Arce et al. 2025). The integration of immigrants – immigrants’ economic mobility and social inclusion – is the object of contentious public debates. Immigrants’ labour market success is a central dimension of integration, with implications for aggregate productivity (Caselli et al. 2024) and the design of fiscal and social insurance policies (Dustmann and Frattini 2014). In parallel, it has been argued that immigration regulations (e.g. employer-sponsored visas) can limit immigrants’ job mobility and hinder immigrants’ labour market outcomes (Wang 2021).

Economists have long documented – at least since Chiswick (1978) – immigrants’ labour market outcomes, their evolution over time, and convergence with natives. Nonetheless, many potential drivers of immigrants’ long-term prosperity and aspects of immigrants’ labour market experiences have proved elusive to document, including questions about the importance of job mobility and firms. How important is job-switching and climbing the firm ladder towards better jobs for immigrants’ labour-market integration and convergence with natives? Answering this requires unusually rich data: following immigrants from arrival, across employers, and over many years.

In Arellano-Bover and San (2026), we study a migration episode with unique institutional features and data availability: the mass migration of nearly one million Jews from the former Soviet Union to Israel during the 1990s, following the unexpected lifting of Soviet emigration restrictions. There are three main advantages of studying this episode. First, these immigrants were automatically granted Israeli citizenship upon arrival, thus facing no differential regulatory restrictions compared to natives. We can then study what we call unconstrained assimilation (unconstrained by legal status).

Second, immigrant self-selection based on labour market prospects was not prevalent due to push factors related to the desire to escape Soviet persecution and political turmoil. Lessons from this setting can speak to policy-relevant large migration waves driven by conflict or climate change, such as the one initiated by the war in Ukraine (Rapoport 2023).

Third, Israel allows researchers to access detailed administrative records that merge tax records with the population registry. These population-wide records enable us to track immigrants’ employers, wages, and job moves year by year – from the moment of arrival and for nearly three decades thereafter.

Figure 1 Migration flow: Former Soviet Union to Israel, 1948–2019

Notes: Number of immigrants arriving in Israel from the former Soviet Union, by year.
Source: Israel Central Bureau of Statistics.

Why might firms matter for immigrant–native wage gaps in the first place? The labour market can be thought of as a job ladder: workers gradually move from lower-paying to higher-paying firms as they search for jobs and receive offers. Even workers with identical skills can earn very different wages depending on which firm employs them, because firms differ in productivity, wage-setting policies, and the rents they share with employees.

Newly arrived immigrants enter this job ladder from the bottom. They lack local networks, information about employers, and a history of past job search. If they start in lower-paying firms simply because they have not yet had time to search and receive better offers, then wage gaps may partly reflect limited time in the host-country labour market rather than permanent productivity differences. In that case, job mobility – how often immigrants switch firms and how large the steps they take – becomes central to understanding wage convergence.

We find that on the employment margin, male immigrants from the former Soviet Union found their first job remarkably quickly: within a year of arrival, they were slightly more likely to be employed than comparable native men. This aligns with these immigrants’ lack of savings and limited pecuniary assistance from the Israeli government – finding a job quickly was a necessity. Female immigrants instead experienced a sizeable employment gap that persisted for roughly five years before closing.

Wages tell a different story and suggest that immigrants’ initial jobs were likely of poor quality. Male immigrants earned about 57% less than comparable natives in their first year in Israel, while the gap for women was about 47% (Figure 2, solid markers). Convergence was substantial but gradual. Wage gaps narrowed steadily over time, yet full parity took nearly three decades.

Figure 2 Immigrant-native wage gaps, overall and within-firm, by gender

Notes: Immigrant–native log monthly wage gaps, by immigrants’ years since arrival in Israel, adjusting for age and year effects. The hollow markers further adjust for employer fixed effects.

To understand how firms contribute to these wage gaps, we build on the Abowd–Kramarz–Margolis (1999) framework, which decomposes wages into worker effects and firm effects. In this setup, a worker’s wage reflects both their individual characteristics and the pay premium of the firm that employs them. Crucially, we estimate these firm pay premiums separately for immigrants and natives. This allows us to construct what we call the gap in firm pay premiums – the difference between the average firm-level pay boost received by immigrants and that received by natives. We then decompose this gap into two components: a between-firm (sorting) component, which captures the fact that immigrants and natives work at different firms, and a within-firm (pay-setting) component, which captures differences in how the same firm pays immigrants and natives. This decomposition lets us distinguish whether assimilation is primarily about gaining access to overall better firms, or firms that treat immigrants and natives more equally.

Figure 3 shows how firm pay premium gaps evolve for male immigrants. The solid line represents the total gap in firm pay premiums – the average wage boost that firms grant to immigrant men relative to comparable native men. On arrival, immigrant men receive firm pay premiums that are about 16–17% lower than those of natives. This gap shrinks steadily over time.

Figure 3 Immigrant-native gap in firm pay premiums, for males, by years since arrival in Israel

Notes: ‘Total gap’ corresponds to the overall gap in firm pay premiums, which can be decomposed into ‘differential sorting’ (between firms) and ‘differential pay-setting’ (within firms).

The figure also breaks this total gap into two components. The open circles capture sorting across firms: immigrant men initially work in lower-paying firms, but over time they move toward firms that offer higher pay premiums. The crosses capture within-firm pay-setting: even inside the same firm, immigrants initially receive roughly 6% lower firm pay boosts than natives. At the beginning of the assimilation process, sorting and within-firm pay-setting account for roughly equal shares of the total gap. Over time, sorting continues to improve and eventually turns positive, while the within-firm pay gap declines more slowly and remains modestly negative even in the long run. Women feature similar dynamics in overall convergence, though within-firm pay-setting plays a larger role early on for female immigrants.

As opposed to the wage gaps in Figure 2, immigrants can only close the firm-premium gaps in Figure 3 through job mobility. Were immigrants from the former Soviet Union more likely to switch jobs than natives? Conditional on doing so, did they take greater ‘steps’ up the firm ladder? Figure 4 shows that they did both. In the early years after arrival, immigrant men were about 18–19 percentage points more likely than comparable natives to change employers (about 16 percentage points for women). Although this differential declines over time, it remains positive even three decades later. As such, these immigrants, who faced no regulatory barriers to their job-switching, remained persistently more mobile than natives. Moreover, conditional on moving, immigrants climbed faster: in the first years, movers gained roughly 1%–5% larger firm-level pay boosts than comparable natives. Convergence in firm pay premiums was therefore driven by immigrants switching jobs more often and making larger upward moves when they did.

Figure 4 Immigrant–native gaps in job mobility and firm-ladder climbing

Notes: Left panel: immigrant–native gap in the probability of changing employers in a given year, where overall probabilities for natives are 0.13 and 0.10 for men and women, respectively. Right panel: immigrant–native gap in the size of the firm-premium change, conditional on changing jobs in a given year, where average change for natives is equal to 0.04 for men and 0.02 for women.

Wages, however, are not the only dimension of job quality. Using a revealed-preference measure of employer desirability based on worker flows (Sorkin 2018), we find that immigrants initially sort into firms that offer lower overall job utility, with non-pay amenities differentials amounting to 0.4 and 0.9 native standard deviations, for men and women, respectively. Immigrants are thus initially employed in firms providing lower utility in the form of non-wage amenities, exacerbating immigrant–native pay gap disparities. Gaps in non-pay amenities do close faster than pay gaps, typically within a decade. As such, immigrants move not only toward higher-paying firms, but toward more desirable jobs overall.

What are the broader implications of our findings? Our key result is that firms – and immigrants’ ability to climb the firm ladder – were central to integration in an unconstrained assimilation setting, undistorted by immigration regulation. The substantial wage gains we document would not have been possible without recurrent employer changes. Policies that tie immigrants to a single employer or delay unrestricted work authorisation may therefore slow access to better firms and delay convergence. Reducing search frictions – through rapid work permits, language training, and job search support – can accelerate movement up the firm ladder.

More broadly, these mobility patterns are not only about immigrants’ own success. They can be interpreted as a gradual reallocation of a large share of the workforce toward more productive firms, with potential aggregate productivity gains. In that sense, policies that shape immigrants’ job mobility may influence not only individual earnings trajectories, but also how efficiently labour is allocated across firms.

References

Abowd, J M, F Kramarz, and D N Margolis (1999), “High wage workers and high wage firms”, Econometrica 67(2): 251–333.

Arce, O, A Consolo, A Dias da Silva, and M Weissler (2025), “Foreign workers: a lever for economic growth”, The ECB Blog, European Central Bank, 8 May.

Arellano-Bover, J, and S San (2026), “The role of firms and job mobility in the assimilation of immigrants: Former Soviet Union Jews in Israel 1990–2019”, CEPR Discussion Paper 21185.

Caselli, F, A G Dizioli, and F Toscani (2024), “Macroeconomic implications of the recent surge of immigration to the EU”, VoxEU.org, 14 October.

Chiswick, B R (1978), “The effect of Americanization on the earnings of foreign-born men”, Journal of Political Economy 86(5): 897–921.

Dustmann, C, and T Frattini (2014), “The fiscal effects of immigration to the UK”, The Economic Journal 124(580): F593–F643.

Rapoport, H (2023), “Labour market access for Ukrainian refugees”, VoxEU.org, 9 January.

Sorkin, I (2018), “Ranking firms using revealed preference”, The Quarterly Journal of Economics 133(3): 1331–93.

Wang, X (2021), “US permanent residency, job mobility, and earnings”, Journal of Labor Economics 39(3): 639–71.