The 2019 edition of the Organization for Economic Cooperation and Development’s (OECD) annual International Migration Outlook report included a new chapter, “Capturing the ephemeral: How much labour do temporary migrants contribute in OECD countries?” It’s a good question, and one that had not yet been answered.

There is a dearth of data on temporary labor migration programs (TLMP) or schemes—aka guestworker programs, where migrants are employed temporarily in a country outside their own—and it hinders the ability of policymakers to make informed decisions. The OECD declared TLMPs are “a core concern in the public debate across OECD countries” but warns that their impacts are “understudied.” This information deficit exists despite the fact that TLMPs are controversial and make up an increasing share of labor migration, and in the United States in particular have been at the center of debates about how to reform the U.S. immigration system.

Why are TLMPs controversial and at the center of public debates? First, their size. One of OECD’s key findings is that the 4.9 million temporary labor migrants that entered OECD countries in 2017 is “almost as many… as permanent migrants in all categories combined.” Ignoring temporary labor migration in the OECD means ignoring nearly half of all migration.

Many employers want larger TLMPs and fewer regulations governing their use. But there are high economic, social, and psychological costs for the migrant workers who participate in temporary programs, including frequent human rights violations suffered in both countries of origin and destination. Further, some abuses that are technically legal are facilitated by the legal frameworks of TLMPs. In most TLMPs, employers control the visa status of their temporary migrant employees or “guestworkers”—which means getting fired makes them deportable. In part, that’s why TLMPs have been called things like “The New American Slavery.”

TLMPs raise technical issues that are not easily resolved. For example: Which industries are permitted to hire migrant workers? How will appropriate numerical limits in TLMPs be determined? What rights will migrants have once they’ve been admitted into receiving country labor markets? Can they bring their families? Will migrants be tied to one employer or be allowed to change jobs and employers? How will receiving country governments ensure that migrants return after their employment contracts end, or will migrants be allowed to become permanent residents? Do citizens in receiving countries have first preference for jobs that employers want to fill with migrants? Will migrants be paid the same wages as similarly situated local workers?

Answers to these questions require honest discussions about the trade-offs that are inherent in migration policymaking, including whether labor migration is an appropriate response to employer claims of labor shortages. (Raising wages or increasing training may be a better response in many cases.) But these questions are particularly difficult—if not impossible—to answer satisfactorily without reliable data for evidence-based policy making.

The United States illustrates the problem of trying to make policy without reliable data. Comprehensive immigration reform bills adopted by the Senate in 2006 and 2013 were based on a “three-legged stool” of more enforcement, legalization for unauthorized immigrants, and more temporary labor migration. Thinking about the future impacts of temporary labor migration required analysis about existing TLMPs and predictions based on that analysis. But U.S. government-collected data on temporary work visas are inadequate, generally of poor quality, recorded in an inconsistent manner across federal agencies, and the most useful data are not published regularly or systematically. The result was a lack of reliable estimates and assumptions—instead, policymakers and the public had only claims made by advocates and employers to rely on.

The OECD “aims at closing” research gaps on labor migration “by providing the first estimates of the total employed temporary migrant population in full-year equivalent for 20 OECD countries.” This is an especially useful finding, since the United States government has no official corresponding estimate of the number of temporary migrants employed in the labor market. The U.S. Department of Homeland Security (DHS) estimates what it refers to as the “resident nonimmigrant population”—temporary migrant workers are issued “nonimmigrant” visas that authorize employment—but it does not estimate how many nonimmigrants were employed in the U.S. labor market or calculate the number of full-time equivalent (FTE) jobs, nor does it provide numerical estimates by individual TLMP or visa classification.

OECD includes in their estimates “all categories of temporary migrants who may participate in the host country labour market” such as “international students, cultural exchange programme participants, service providers such as [European Union/ European Free Trade Association-]posted workers, as well as cross-border workers.” There are hundreds of thousands of temporary labor migrants in the United States, and many more around the world, who are employed despite the fact that their visas are ostensibly for other purposes, like attending university or participating in a trainee or cultural exchange program. But while DHS’s provides a population estimate for international students and cultural exchange visitors, it does not provide an estimate on how many of each group were employed.

Before the OECD’s report, previous EPI research estimated that there were 1.4 million nonimmigrants employed in the United States during some portion of 2013, accounting for approximately 1% of the labor force. The OECD found 1.6 million FTE jobs filled by temporary labor migrants in the United States in 2017, accounting for 1.04% of the labor force.

OECD reported that 1.8 million temporary residence visas were issued in the United States in 2017, accounting for a third of all temporary residence visas issued across all 20 OECD countries. Next on the list were Australia, Japan, Canada, New Zealand, France, and South Korea. While most of the visas counted in this estimate permit employment, not all of them do, and across the OECD, one-quarter of the visas issued were renewals rather than newly issued visas. OECD also found that in the United States, less than half of temporary visas were issued to migrants in traditional TLMPs; over 20% went to accompanying family members, one-quarter to international students, and 6% to exchange visitors.

The OECD charted the distribution of the maximum allowed duration of stay of temporary visa holders, finding that 14.5% allowed a stay of over five years and 13.7% had no maximum duration, while just over one-fifth allowed a stay of 13–24 months and one-third allowed a stay of two to five years. Only 12.1% authorized a stay of less than 12 months.

These findings raise questions about what “temporary” means and whether the jobs migrant workers fill are truly temporary: If nearly nine in 10 temporary visas authorize a stay longer than one year, are migrants really filling temporary jobs? Can a job that lasts more than one year be considered temporary? Many of these jobs may be permanent jobs that have been misclassified as temporary—perhaps because employers prefer precarious workers over whom they can exert more control if they hold job-contingent temporary visas. Shouldn’t governments instead consider issuing permanent immigrant visas that lead to citizenship for migrants who are filling de facto permanent jobs?

The OECD has spotlighted an important gap in migration research—one we have tried to call attention to in the past—that leads to important questions about the current trend in labor migration governance toward more temporary workers instead of permanent immigrants who have equal rights in destination countries. We hope that the OECD’s new findings revealing the importance and size of temporary labor migration will spur improved data collection by governments and more research on TLMPs.