Cutting Legal Immigration Won’t Help Low-Skilled American Workers
by Alex Nowrasteh on February 9, 2017 at 10:17 AM
Senators Tom Cotton (R-AR) and David Perdue (R-GA) recently introduced the Reforming American Immigration for Strong Employment (RAISE) Act. If it were to become law, RAISE would cut legal immigration by 50 percent over the next ten years by reducing green cards for family members of U.S. citizens and lawful permanent residents, slashing refugees, and eliminating the diversity visa lottery. These goals are in line with President Trump’s stated objective to cut legal immigration in most categories.
The RAISE Act’s goal is to increase wages for lower-skilled Americans by reducing the supply of lower-skilled immigrants. Their press release argues that the “generation-long influx of low-skilled labor has been a major factor in the downward pressure on the wages of working Americans, with the wages of recent immigrants hardest hit.” Under such a worldview, only a drastic reduction in green cards and the supply or workers can raise American wages - and it’s not crazy.
The National Academy of Sciences’ (NAS) exhaustive literature summary on the economic effects of immigration concluded that: “When measured over a period of 10 years or more, the impact of immigration on the wages of native-born workers overall is very small. To the extent that negative impacts occur, they are most likely to be found for prior immigrants or native-born workers who have not completed high school—who are often the closest substitutes for immigrant workers with low skills.” Although the effect is small, RAISE seeks to take advantage of the finding in the academic literature by inferring that if an increase in the supply of workers slightly lowers some wages then a decrease in that same supply will do the opposite.
It might seem odd then that RAISE doesn’t target employment-based green cards but that category is for highly skilled workers while the categories this bill would cut are more likely to allow in lower-skilled workers who have fairly high labor force participation rates. I’ve rebutted Senator Cotton’s poor economic arguments for immigration restrictions before but recent research is even more compelling.
A recent paper by economists Michael Clemens, Ethan Lewis, and Hannah Postel seems tailor-made to test what would happen if a bill like the RAISE Act were to become law. The paper studies the effectiveness of an immigration policy “designed to raise domestic wages and employment by reducing the total size of the workforce.” The U.S. government’s 1964 termination of the Bracero program for Mexican farm workers provides a natural experiment for their paper which is comparable to what would happen if RAISE becomes law. Senators Cotton and Perdue will be disappointed to discover that this new research found that ending lower-skilled migration for farm workers had little measurable effect on the labor market for Americans who worked in those occupations.
Figure 1 from their paper compares the quarterly average real farm wages by states where Braceros made up more than 20 percent of their seasonal agricultural labor (black line), states where Braceros were fewer than 20 percent of the workforce (gray line), and states where there were no Braceros at or very negligible numbers (dashed line). Clemens et al write that “[t]he figure shows that pre- and post-exclusion trends in real farm wages are similar in high exposure states and low-exposure states. It also shows that wages in both of those groups rose more slowly after bracero exclusion than wages in states with no exposure to exclusion.”
How can that be the case, shouldn’t a leftward shift in labor supply increase wages? Not necessarily as farmers had other options not usually contemplated by those who only think about the supply and demand for labor in isolation. Instead of hiring more American workers or raising their wages, farmers turned to machines and altered the crops they planted to take account of the new dearth of workers. Instead of planting crops that required labor-intensive harvesting or care, they planted other crops that required many fewer workers. Farmers turned to machines like tomato pickers and changed methods for planting and harvesting other crops to take account of the newer wages they would have faced had they stuck with the Bracero-era farm techniques.
The farmer’s actions in response to Bracero’s cancellation were economically inefficient and presumably raised the costs of production relative to their employment of legal workers under the Bracero program. But since the Bracero program wasn’t available anymore, using more machines and changing techniques were still the cheapest options available. Those options did not include hiring more Americans or raising their wages. The detailed empirical work in this paper considers many other possibilities but convincingly answers them, such as pointing out that nearly all of the Braceros went home instead of staying on illegally and that the flow illegal Mexican farm workers did not pick up immediately.
The similarities between the end of the Bracero program and the RAISE Act are enough to make this new research a compelling reason to reject the RAISE Act out of hand. The Bracero program allowed in half a million workers a year before it was eliminated which is about the same number of green cards that RAISE would cut. Bracero workers were lower-skilled and many of those that would be cut by RAISE are also low-skilled. Braceros were concentrated in some states and not others just like the new immigrants who would not be allowed under RAISE.
The ending of the Bracero program was a policy shift very similar to that proposed RAISE Act. Ending Bracero didn’t raise wages for American farm workers and we shouldn’t expect the RAISE Act to do the same for other low-skilled Americans who are suffering for myriad reasons that have nothing to do with immigration. Senators Cotton and Perdue may intend to raise the wages of lower-skilled Americans, but their bill is more likely to line the coffers of firms that manufacture machines that can substitute for them.