Unlike other developed countries such as Australia and Canada, the U.S. government simply does not collect and analyze enough information on the labor market outcomes of immigrants who are issued visas that grant them legal permanent resident (LPR) status. Especially when it comes to longitudinal data that track the same immigrants over time in order to see how their situation has changed. This dooms policymakers to make uninformed decisions based on assumptions or that are influenced by lobbying from interest groups. That’s why new (and as-of-yet unpublished) research from Professors Mark Rosenzweig (Yale University) and Guillermina Jasso (New York University) is groundbreaking: it analyzes data from Princeton University’s New Immigrant Survey (NIS) and offers a glimpse at how immigrants in the United States are performing in the labor market over time and by visa category.

The NIS “is a nationally representative multi-cohort longitudinal study of new legal immigrants and their children to the United States,” and one of the most useful data sets available on U.S. immigrants. The first cohort of immigrants participating in the NIS were interviewed in 2003, and follow-up interviews were conducted from 2007 to 2009. Rosenzweig recently presented their research based on these survey data at an informative conference on family immigration hosted by the U.S. Department of Homeland Security and the Organisation for Economic Cooperation and Development (OECD). He showcased some of the first analyses of the NIS’s longitudinal data on the labor market performance of immigrants who came to the United States through the Diversity Visa (DV) lottery and the employment-based (including spouses) and family-based immigrant visa categories; the latter of which represent two-thirds of all immigrant visas issued (by far the highest share in the OECD), and allow foreign citizens to join a spouse or parent who is a U.S. citizen or LPR, or to join the son, daughter, or sibling of a citizen already in the United States. (Lindsay Lowell of Georgetown also presented notable new research and findings documenting the growth of the family-based visa category in the United States since the 1970s.)

Some of Rosenzweig and Jasso’s main findings include: Unemployment rates are “very high” for family-based immigrants, but after four to six years they are low for those in all the family-based categories, including the spouses of both DV lottery and employment-based immigrants. Immigrants in all of the visa categories also had high rates of growth in earnings (although they were not compared with earnings growth for natives), the highest being DV lottery winners, who initially struggle when admitted in terms of high unemployment and low wages (likely because they do not necessarily arrive with a job or the support of family members and are only required to have a high-school education). They also found that a high share of all immigrant groups are self-employed (consistent with other research).

Those in the visa category for siblings of U.S. citizens had lower earnings than other family-based immigrants, and after four to six years had not made significant gains. Rosenzweig and Jasso suggest this is because “they have less schooling and U.S. experience, are older and come from poorer countries,” and also showed that those obtaining LPR status in the sibling category were the most likely to subsequently petition for an immigrant visa for other relatives. The principal immigrants in the employment-based category were the most likely to be employed and least likely to be unemployed, and were the highest earners. This is the least surprising finding, since these immigrants are usually recruited and hired by U.S. employers who select them for their skills. In fact, employment-based immigrants and their spouses earn about twice as much as family-based and DV immigrants.

These findings will be important to keep in mind once the U.S. Senate and House of Representatives finally get together to negotiate reforms and improvements to the permanent immigrant visa categories. The big-picture takeaway is that most immigrants who attain LPR status—to different extents—are doing fine and integrating into the labor market quite well over time. But if members of Congress decide they need to set priorities among these categories, at least now they can make decisions that are grounded in evidence and data.

Congress could, for example, prioritize those immigrants who are more likely to be skilled, employed, and who have greater earnings potential. In his presentation Rosenzweig did not mention any policy implications or discuss contemporary immigration politics, but in my opinion these findings support the approach taken in the Senate’s 2013 comprehensive immigration bill, which creates a points system that increases the number of employment-based visas and allocates them based on skills and experience, and eliminates the family-based sibling category and the DV lottery. But this is ultimately a political decision, and another policy option is to prioritize the importance of the extended family in contrast to purely economic factors, for instance, by keeping in place or expanding the number of visas available for siblings and adult parents and adult children of U.S. citizens, and by clearing family-based visa backlogs that keep families apart for many years (the Senate bill also clears the backlogs). Or Congress could decide to fall somewhere in-between, and decide how many visas to allocate to each category based on a balancing of these priorities. These are tough issues but luckily the debate has been enriched by Rosenzweig and Jasso’s analysis of the NIS data.