Share this:

Ben Rissing 80x108Emilio Castilla 80x108Employment based visa programs offer a way for hundreds of thousands of foreign individuals to work in the U.S. every year. But is there any bias in who gets approved and who does not? In new research that examines nearly 200,000 labor certification applications, Ben A. Rissing and Emilio J. Castilla find that foreign workers from Latin America are 23 percent less likely than Canadians to be certified to work in the U.S., and that Asians are 13 percent more likely to be approved than Canadians. This said, they find no statistically significant differences in approval outcomes by immigrant world region during government evaluations of audited applications – which are reached using detailed employment-relevant information. To address unequal outcomes in these assessments, Rissing and Castilla suggest that the foreign worker citizenship field within the labor certification application be removed during government evaluations.

Each year in the United States, employers seek to legally hire hundreds of thousands of immigrant workers through America’s employment-based visa programs.  Unlike standard hiring processes, employment of these foreign workers is also often contingent upon the work authorization decisions of U.S. government agents.  In new research, we examine the key first stage of one such work authorization process, the U.S. labor certification program, which is required for the granting of most employment-based green cards in the United States.  We find that there is substantial variation in approval outcomes associated with foreign workers’ country of citizenship.  Notably, 66.8 percent of foreign workers from Latin America are approved by government agents, while 90.5 percent of workers from Asia are approved.  These approval disparities exist after controlling for foreign workers’ offered salary, job title, job skill level requirement, location, industry, and prior visa, among other key application characteristics.

In the United States, Equal Employment Opportunity laws and immigration acts mandate the equitable evaluation of individuals, regardless of their country of origin.  Along these lines, the U.S. labor certification process is intended to be merit-based, and contains no evaluation criteria pertaining to foreign workers’ origin country or demographic characteristics.  This program specifically seeks to ensure that a described foreign worker is qualified for a given job opportunity and that no qualified U.S. workers are available for the position.  Randomly-assigned government decision makers act similarly to human resource managers during their application assessments, reaching decisions based on fields in a submitted application and without ever meeting, or communicating with, a described immigrant worker.  This said, information on each described immigrant worker’s country of citizenship is visible early in the labor certification application, a field that these government decision makers are instructed to disregard during their training.

We quantitatively examined the entire population of over 198,000 labor certification applications that were approved or denied between June of 2008 and September of 2011, as determined by a small team of government decision makers working in a single Atlanta, Georgia U.S. Department of Labor processing center.  We find unequal approval outcomes associated with immigrants from a variety of countries.  Notably, foreign workers from Latin America are 23 percent less likely to receive approval than Canadian individuals (the study’s reference category), even with controls accounting for employer, occupation, and immigrant worker characteristics.  Asian individuals, in contrast, are 13.3 percent more likely to receive approval relative to Canadians, all else equal.  These results suggest inequality in this U.S. labor certification process, which affected employment outcomes at over 68,000 organizations during the time period associated with this study.

Credit: Kathryn Decker (Flickr, CC-BY-2.0)

Credit: Kathryn Decker (Flickr, CC-BY-2.0)

The majority of labor certification applications are filed by U.S. employers seeking to place a foreign worker into a high-skill and high-paying job.  In this regard, 40 percent of applications are related to computer and mathematical occupations, followed by architecture and engineering occupations (9 percent of all applications), and management occupations (9 percent).  The median salary for immigrant workers described in labor certification applications during this 40 month period was $73,000 USD.  This said, when government decision makers have limited information with which to assess an application, available demographic data may consciously or unconsciously shape decision outcomes.

Research has shown that U.S. natives tend to regard Asian immigrants as highly competent, Canadians as moderately competent, and Latin American immigrants as having low competency.  With specific regards to the attitudes of government decision makers, their evaluation outcomes could be influenced by publicized aggregate processing statistics for other U.S. immigration programs.  Along these lines, U.S. Department of Homeland Security data indicates that of all immigrants seeking U.S. entry which were regarded as inadmissible (or not valid), 32 percent were from Mexico.  Moreover 93 percent of all U.S. deportations targeted immigrants from eight Latin American countries, and 60 percent of unauthorized U.S. immigrants are estimated to be from Mexico.  These aggregate immigration statistics, for instance, could affect government decision makers’ work-relevant evaluations when limited information is available during their work authorization assessments.

The question of how available employment-relevant information may shape evaluation outcomes is at the heart of our recent research.  Notably, through the government’s labor certification program, a portion of applications are audited each year (13 percent during our 40 month period of study).  When audited, detailed employment information is collected from the sponsoring employer and used to inform a government agent’s labor certification assessment.  We find no statistically significant differences in approval outcomes by immigrant world region during these evaluations of audited applications.  These results highlight the importance of detailed information in reaching equitable employment-based evaluations of workers.  This research also has important implications outside of immigration programs, as many organizations conduct initial employment screenings of potential workers using limited information (for instance, during assessments of candidates’ resumes or submitted job application forms).  The findings of this research thus highlight the unintended consequences of employment-based evaluations that rely on limited information.  Without detailed employment information, decision makers’ judgments may be consciously or unconsciously shaped by applicants’ available demographic data.

In order to address these unequal approval outcomes, we recommend the audit of all labor certification applications; such that these requests might be assessed using detailed employment-relevant information.  However, we acknowledge the costs and administrative burden associated with such widespread auditing activity.  An alternative and practical low-cost solution may be to mask all foreign worker demographic characteristics during government decision makers’ application review.

Immigration reform has returned to the forefront of political debate in the United States as a result of U.S. President Barack Obama’s executive order last November.  This said, proposed immigration reform measures have to date not attended to the process by which immigrant applicants are assessed.  Many aspects of U.S. immigrant evaluation systems are opaque and discretionary.  As such, we recommend that future U.S. legislation or administrative actions concerning immigration reform to attract and retain highly-qualified foreign workers could, and should, materially address the processes by which potential immigrants are evaluated.

This article is based on the paper ‘House of Green Cards: Statistical or Preference-Based Inequality in the Employment of Foreign Nationals’, in the American Sociological Review.

Please read our comments policy before commenting.           

Note:  This article gives the views of the author, and not the position of U.S.App– American Politics and Policy, nor of the London School of Economics.

Shortened URL for this post: http://bit.ly/1wRDfep

 _________________________________ 

About the authors

Ben Rissing 80x108Ben A. RissingBrown University
Ben A. Rissing is the Pearson Visiting Assistant Professor in the Department of Sociology at Brown University. His research interests are in the areas of organizations, work, and employment. His most recent work has been published in the American Sociological Review, British Journal of Industrial Relations, and Encyclopedia of Global Human Migration. Ben received his doctorate in management from the Institute for Work and Employment Research (IWER) at the MIT Sloan School of Management and previously he conducted research as a Postdoc with the Brown University Watson Institute for International Studies and as a Wertheim Fellow with the Harvard Law School Labor and Worklife Program.

Emilio Castilla 80x108Emilio J. Castilla – MIT Sloan School of Management
Emilio J. Castilla is an associate professor of management at the MIT Sloan School of Management (Behavioral and Policy Sciences Area), where he teaches courses in organizational behavior and strategic human resource management. He is a member of the Institute for Work and Employment Research at MIT; and also a research Fellow at the Wharton Financial Institutions Center and at the Center for Human Resources at the Wharton School. His research primarily focuses on the sociological aspects of work and employment. He is particularly interested in examining how social and organizational processes influence employment outcomes over time, and he tackles these questions by examining different empirical settings with unique longitudinal datasets, at both the individual and organizational level.

 

 

 

Print Friendly