Skilled immigration is of great importance to the US, representing 16% of US workers with a bachelor’s degree, and 29% of the growth of this labour force over the period 1995-2008. Presently, proponents for increased skilled immigration argue that it supports economic growth, and some have gone so far as to say that the alternative would be paramount to ‘national suicide’. Opponents believe skilled immigration is already too high and hurts the outcomes of citizen workers. In particular, Bill Gates has stated that Microsoft hires four additional employees to support each skilled worker brought into the US through the H-1B visa program. Matloff (2003) instead argues that US companies use skilled immigrants to displace older citizen workers that have higher salaries.
Our research proposes a more subtle relationship than that suggested in the public discourse on immigration or from more aggregate models of international worker flows. We study the way in which the hiring of skilled immigrants affects the employment structures of US firms using employer-employee data from the US. While OLS and IV specifications find increases in total skilled employment by the firm with increases in skilled immigrant employment, we find evidence that employment expansion is greater for younger natives than their older counterparts.
Our focus on the firm is, at the same time, both rare and important. There is little tradition for considering the firm as the unit of analysis, yet many of the features of our analysis would be obscured with more standard approaches. Our approach allows us to account for a greater level of heterogeneity, and at the same time, is intuitive given that many skilled immigrant decisions in the US are driven by firms themselves. We think the development of a number of new employer-employee datasets like the one used in our study presents a significant opportunity for better understanding the impacts of immigration.
Our analysis deviates from a long line of previous research. Card (2001), Kerr and Lincoln (2010), Hunt and Gauthier-Loiselle (2010) and Peri et al. (2013) employ a conceptual model that defines labour markets as local areas, where an immigrant to an area most directly impacts the opportunities of natives in the same area. Borjas (2003) employs a second model that emphasises competition among immigrant and native labour with similar education, age, and experience profiles. In such a model, an immigrant to one area may directly affect the employment opportunities of a similar native in another area of the same country relative to a dissimilar native in the same area. A third model considers labour markets to be specialised fields of study or expertise – such as mathematics research – which often involves institutional settings different than those of firms. The book chapters by Kerr, Kerr, and Lincoln (2014) and Kerr (2014) provide a broader review of the empirical literature on skilled immigration.
Conceptual framework and firm panel
While the approach of modelling immigration as shifts in the supply of workers to a market may be appropriate for studying the effects of unskilled immigration, the immigration of skilled labour is distinct in that firms often actively select and recruit skilled immigrants to enter the country.
- We deviate from past work by analysing how firms’ decisions regarding hiring skilled immigrants affect their employment structures.
Our research capitalises on data regarding the H-1B visa program and employer-employee records maintained by the US Census Bureau in the confidential Longitudinal Employer-Household Dynamics (LEHD) database.
We construct a panel of 319 firms over the 1995-2008 period to examine the effect of skilled immigration on the employment patterns of top patenting and employer firms in the US. We focus the analysis on large US employers and major US patenting firms as the current public debate about skilled immigration focuses on employment and innovation outcomes. Of the 29 states participating in the LEHD database, we limited the sample to the subset of 18 states that provided data starting in 1995. Firms are included in the sample if they are:
- Listed as a top 100 employer by the US Census Bureau’s Longitudinal Business Database in any year from 1990 to 2008;
- Ranked as a top 100 US firm for worldwide sales or employment in Compustat over the full 1990-2008 period; or
- Included in the Fortune 200 companies in 2009.
The sample further includes firms that account for more than 0.05% of all US patents applied for during 2001-04.
In order to ensure that our firms operate primarily in the states covered in our sample of workers, we restrict the analysis to firms that have at least 25% of their employment in the core 18 LEHD states. Consistent with the skewness of the firm size distribution, the resulting panel accounts over one-third of US patents, and 20-30% of all workers in the LEHD are employed at one point or another by the companies during the sample period. The sectoral composition of the panel is approximately:
- 30% manufacturing;
- 25% wholesale and retail trade;
- 30% finance, insurance, real estate, and services; and
- 15% other sectors.
Firm-level ordinary least squares employment analysis
Ordinary least squares (OLS) estimation finds a link between the expansion in a firm’s young skilled immigration employment and growth in the employment of the firm’s overall skilled workforce. We estimate that a 10% increase in a firm’s young skilled immigrant employment correlates with a 6% increase in the total skilled workforce of the firm. Increases in native worker employment are found to be mostly balanced for older and younger skilled workers. Similar conclusions are found when considering effects on hiring as well. We find the OLS estimates to be robust when considering a firm-state approach with the full set of 29 LEHD states, increasing the LEHD employment ratio criterion, splitting the sample by long-term firm growth rates, and when using different definitions of skilled workers.
Firm-level instrumental variable analysis
Ordinary least squares estimation is potentially biased due to omitted factors and issues of measurement error. To further the purposes of this research, we use an instrumental variable (IV) approach that utilises national changes in the H-1B visa program’s size interacted with how important H-1B workers are for each firm. We consider a variety of instruments in our paper and focus here on ones that use Labour Condition Applications made by firms as first step in the H-1B process to measure firm dependency. The IV estimations are consistent with the OLS estimations in that we see expansions in firms’ skilled workforces with the increased employment of young skilled immigrants. However, this expansion disproportionately favours young skilled native workers and older skilled immigrant workers as opposed to older native workers.
We next consider extensions to this analysis by matching occupation data on over 25,000 workers from the Current Population Survey with records from the LEHD. We find that departure rates are higher for older workers in STEM occupations with greater young skilled immigration into the firm, especially for workers earning over $75,000 a year. However, although we find larger effects for older workers in STEM occupations relative to older workers in non-STEM occupations, the difference in the coefficient estimates is not economically large or statistically significant.
Our analysis provides a multi-faceted view of the impact of skilled immigration on the employment structures of US firms. Our findings consistently show that the hiring of young skilled immigrants is tied to the greater employment of skilled workers overall, a larger share of the firm’s workers being skilled, a larger percentage of immigrants amongst skilled workers, and a lower proportion of skilled workers being over the age of 40. Unlike with other worker groups, we find evidence that older native employment expands very little with increased skilled immigration. Overall, our findings paint subtle picture of the employment effects of high skilled immigration, and point to the importance of considering the role of firms in immigration.
Authors' note: Any opinions and conclusions expressed herein are those of the authors and do not necessarily represent the views of the U.S. Census Bureau. All results have been reviewed to ensure that no confidential information is disclosed.
Borjas, G (2003), “The Labor Demand Curve Is Downward Sloping: Reexamining the Impact of Immigration on the Labor Market,” Quarterly Journal of Economics 118(4), pp. 1335-1374.
Card, D (2001), “Immigrant Inflows, Native Outflows, and the Local Labor Market Impacts of Higher Immigration,” Journal of Labor Economics 19(1), pp. 22-64.
Hunt, J and M Gauthier-Loiselle (2010), “How Much Does Immigration Boost Innovation?” American Economic Journal: Macroeconomics 2(2), pp. 31-56.
Kerr, W (2014), “US High-Skilled Immigration, Innovation, and Entrepreneurship: Empirical Approaches and Evidence,” in C Fink and E Miguelez (eds.) Intellectual Property, the Migration of Skilled Workers, and Development (Cambridge, UK: Cambridge University Press), forthcoming.
Kerr, S P, W Kerr and W Lincoln (2014), “Skilled Immigration and the Employment Structures of US Firms,” Journal of Labor Economics, forthcoming.
Kerr, S P, W Kerr and W Lincoln (2014), “Firms and the Economics of High-Skilled Immigration,” in W Kerr, J Lerner and S Stern (eds.) Innovation Policy and the Economy Volume 15 (Chicago, IL: University of Chicago Press), forthcoming.
Kerr, W and W Lincoln (2010), “The Supply Side of Innovation: H-1B Visa Reforms and US Ethnic Invention,” Journal of Labor Economics 28(3), pp. 473-508.
Matloff, N (2003), “On the Need for Reform of the H-1B Non-Immigrant Work Visa in Computer-Related Occupations,” University of Michigan Journal of Law Reform 36(4), pp. 815-914.
Peri, G, K Shih and C Sparber (2013), “STEM Workers, H-1B Visas and Productivity in US Cities,” Working Paper, University of California, Davis.