Barriers along America's southwestern border significantly reduce illegal immigration, especially by low-skilled migrants, a new study argues.
The analysis, by University of Illinois at Chicago economist Benjamin Feigenberg, provides empirical evidence that the construction of border barriers—in this case, the 700 miles of border fencing authorized by Congress in 2006—cuts migration both in the areas where they are constructed and in adjacent territory.
In total, Feigenberg estimates, a fully fenced border would deter some 86,000 people from crossing the border every three months, at a cost of less than $2,000 per person deterred.
These findings provide robust empirical support for the idea of constructing a barrier along America's often-porous border with Mexico—a proposal long-supported by Republicans, and President Donald Trump in particular, but vociferously opposed by Democrats. The failure to support more border fencing, once a bipartisan goal, may have helped drive the recent wave of immigration that led to a crisis at the border.
To reach his conclusions, Feigenberg examined the results of the 2006 Secure Fence Act, a bipartisan bill signed into law under then-president George W. Bush. The bill called for the construction of some 700 miles of fencing along the Mexican-American border; a subsequent appropriations bill put the exact location of the fencing at the discretion of the secretary of homeland security.
Using a novel dataset, Feigenberg argues that the locations at which the fencing was constructed were effectively randomly distributed. That allows him to exploit the randomness in variation in construction timing to measure the effects of building a border fence on migration in adjacent Mexican territories using a robust collection of U.S. and Mexican government data.
The results are pronounced. The paper finds that fence construction in a given municipality reduces average migration by 27 percent. In addition, fence construction in an adjacent municipality has a spillover effect, reducing migration by an additional 15 percent.
The total effect, Feigenberg writes, is to alter the illegal immigrant population in the United States: "Fence construction significantly decreases the size of the likely undocumented, Mexican-born population on the US side of the border and in the US interior."
That reduction is particularly concentrated among low-skilled migrants, with the data indicating that the probability of migration becomes less biased towards those with less education after border fencing is constructed. Feigenberg is not able to conclude what drives this change, but speculates that increased crossing costs' differential effect on low-earners, limited access to credit, and "differences in access to legal migration opportunities offer potential explanations."
This last finding is significant because of the particular role that low-skilled migration plays in the debate over immigration restriction. Proponents of increased immigration argue that the complementary effects of immigrants increase national wealth on net. But opponents argue that low-skilled migration lowers wages and job opportunities for the lowest-skilled Americans, in turn fomenting social unrest.
The overall effect of border fencing, Feigenberg writes, is likely relatively easy to explain, attributable to "the increase in the risk of apprehension for those who would attempt to climb over new fencing," as well as "the increase in mortality (and perhaps apprehension) risk faced by migrants who choose to go around the fence and cross in remote, topographically inhospitable border segments where border fencing was more likely to be delayed."
Feigenberg's is just the latest contribution to a literature that argues that immigration enforcement can substantially reduce levels of illegal immigration. One 2018 study, for example, found that more aggressive sanctions on repeated illegal border crossings did significantly deter reentry.
At present, illegal immigration has returned to near-record lows, with just 23,000 individuals apprehended illegally entering in May compared with 144,000 in May of 2019. That decline is likely attributable to the novel coronavirus, which has spread as virulently throughout Central America as elsewhere. Feigenberg's findings, however, suggest that future border fencing construction may help mitigate any surge like the 2019 wave of illegal immigration.