Following the deaths of two children in Border Patrol custody, President Trump made his pitch on how to solve the humanitarian crisis at the U.S.-Mexico border. He spoke about “the dangerous trek up through Mexico” and pledged to end “the cycle of human suffering” at the border. His solutions were a border wall and ending the chance to apply for asylum, which he called a “loophole.” Yet the evidence indicates that border fences have made the journey far more dangerous—even deadly—and that asylum made the border safer.

In fact, asylum and other humanitarian relief programs appear to have already saved about 1,300 lives along the border since 2013. By contrast, increased enforcement—including the fence—appears to have resulted in about 4,600 more deaths from 1999 to 2019.

The Rise of Border Deaths

Border Patrol reports the number of dead bodies and other deaths of migrants that it finds along the border. Overall, Border Patrol has recorded 7,529 deaths from FY 1998 to FY 2019. This death count excludes hundreds of people who local authorities discover, according to separate investigations by CNN and the Arizona Republic. This means that overall Border Patrol figures are undercounts, but the general trends up or down from year to year are still useful to determine whether more migrants are dying in the harsh conditions along the border.

Figure 1 shows the official number of deaths identified by Border Patrol each year. The grey bars show total deaths, and the black line shows the number of deaths per 100,000 apprehensions—a proxy measure for how likely any particular crosser was to die crossing the border in that year. The absolute number of deaths increased from 263 to a peak of 492 in 2005, and it remained at about that level until 2013, after which it fell back to the lowest levels since the 1990s. The rate of death per 100,000 apprehensions, however, underwent an even greater eight-fold increase from 17.3 in 1998 to 132 in 2012 before dropping back to just 25.5 so far in FY 2019—the lowest level since 2000.

So what made the journey so much more dangerous and what reversed the situation? First, increased enforcement at the border, including the construction of nearly 600 miles of fences, caused migrants to cross in more remote areas. The length of the border fence predicted about 75 percent of the variance in the death rate from 1998 to 2012. From 1998 to 2006, the length of the fence grew 210 percent, and the rate of deaths grew 144 percent. Then, the cycle repeated. From 2006 to 2012, the death rate grew 212 percent, and the fence length grew 218 percent.

The rapid rise in the death rate reflects the rapid construction of the border fence and the doubling of the Border Patrol. The number of Border Patrol agents predicted about 85 percent of the death rate from 1998 to 2013. The agency’s manpower grew along the southwest border from 7,357 in 1998 to 18,611 in 2013.

The result of increased enforcement was inevitable. As the Government Accountability Office found in 2001:

INS’ phased approach to implementing its strategy has included several operations in which INS allocated additional Border Patrol agents and other resources—such as fencing, lighting, night vision scopes, sensors, cameras, vehicles, and aircraft—to targeted locations along the Southwest border… The strategy assumed that as the urban areas were controlled, the traffic would shift to more remote areas where the Border Patrol would be able to more easily detect and apprehend aliens entering illegally.

The strategy also assumed that natural barriers such as rivers, mountains, and the harsh terrain of the desert would act as deterrents to illegal entry. However, INS officials told us that as the traffic shifted, they did not anticipate the sizable number that would still attempt to enter through these harsh environments. A study of migrant deaths along the Southwest border concluded that while migrants have always faced danger crossing the border and many died before INS began its strategy, the strategy has resulted in an increase in deaths from exposure to either heat or cold.

The GAO repeated its conclusion in 2006, just as Congress voted on the Secure Fence Act—which mandated the construction of 700 miles of fences. It should have surprised no one that the rate of deaths increased as enforcement grew in urban areas.

The Fall of Border Deaths

If the increase in border security caused an increase in deaths, why did the death rate suddenly reverse course and ultimately plummet back down? The answer is that the flow shifted away from Mexico to Central America, and the immigrants who came applied for asylum or other humanitarian relief. Asylum provided an alternative to sneaking into the country.

Because the Trump administration turns asylum seekers away at ports of entry, most asylum seekers end up crossing illegally by going around the fence, but rather than trying to hide, crossers in recent years have immediately sought out Border Patrol agents to turn themselves into. As Figure 4 shows, the likelihood of dying while crossing started to decline in 2013, as the number of asylum claims suddenly jumped by nearly threefold. Central American families and unaccompanied children who receive similar processing by Border Patrol also shot up during this time period.

The death rate briefly ticked back up in 2017—possibly fueled by the unfounded impression that President Trump would end asylum—but once the reality sank in that Trump did not end humanitarian relief, the death rate continued its downward plunge in 2018. In Fiscal Year 2019—beginning in October 2018—there were 32 recorded deaths along the border as of early December, on pace for just 167 for the year. Though it is based on only two months, it would be the lowest figure on record. It also amounts to a death rate of just 25.5, the lowest rate since 2000.

Some people believe that people sneak into the United States illegally and only request asylum as a backup if they are caught. But a fascinating recent report by the Institute for Defense Analyses (IDA)—which was commissioned by the Department of Homeland Security (DHS)—demonstrates that this is not the case. It used nonpublic DHS data of apprehensions of asylum applicants (as well as unaccompanied children or parents with children who are subject to similar processing) at interior checkpoints versus elsewhere. The IDA authors explain:

The large majority of Northern Triangle migrants (including asylum seekers) enter the United States in southern Texas, and specifically in the USBP’s Rio Grande Valley sector. This sector has a network of interior traffic checkpoints that are major chokepoints in the road network north of the immediate border region. Migrants who are intent on evasion and succeed in making it through the immediate border area will sometimes be apprehended at or near the interior traffic checkpoints. Those who present to law enforcement, however, will do this near the border and not travel many miles north to present at a traffic checkpoint.

We can thus evaluate whether there are systematic differences in the ratio of interior (checkpoint) apprehensions to total apprehensions for asylum-seeking groups and a control group of those believed to be evaders (adults who do not claim credible fear). If the ratios of interior to total apprehensions of the asylum-seeking group and the “traditional” evader group are similar, the two groups are likely choosing to evade versus self-present at similar rates. If, however, the ratio of interior to total apprehensions of the asylum- seeking group is much less than the other group, the asylum-seeking group is likely choosing to evade at a much lower rate.

Figure 5 shows the share of apprehensions in the Rio Grande Valley that occur at checkpoints by type of apprehension from FY 2011 to FY 2018. As it shows, the share of asylum seekers caught in the interior plummeted from 14 percent in 2010 to 2 percent in 2018. Non-Mexican children and parents saw similar declines. Even the share of single adults apprehended in the interior declined from 14 percent to 10 percent. This may reflect the fact that many single adults move with groups of families, asylum seekers, or children who present themselves to Border Patrol.

IDA concludes, “The rate of evasion by asylum seekers fell dramatically after 2012, and the rate at which they presented to border enforcement authorities rose to a very high level at the same time.” Either the new arrivals never tried to evade when they started arriving in 2013 or they discovered that they did not need to evade in order to enter the country around that time.

The result is that journeys into the country shortened considerably across all Border Patrol sectors. Unfortunately, Border Patrol doesn’t regularly publish data on where apprehensions occur, but a 2017 Government Accountability Office (GAO) report provides numbers through FY 2016. Figure 6 shows that from 2012 to 2016, the share of apprehensions that occurred within a mile of the border almost doubled from 33 percent to 58 percent, while those occurring more than 20 miles from the border fell by nearly half from 27 percent to 15 percent. While the GAO doesn’t provide the exact figure, it appears that the average crossing declined from 11 miles to 6 miles from 2012 to 2016. It likely continued to fall after 2016.

Far too many people are dying at the border still, but it is clear that the availability of asylum has saved the lives of many border crossers. If the number of deaths per 100,000 apprehensions had not come down from FY 2012 to December 2018, about 1,336 more migrants would have died at the border—an increase from about 1,961 to 3,297. On the other hand, if the number of deaths per 100,000 apprehensions had not increased since 1998—when the fence and agents started rapidly growing—4,601 fewer deaths would have occurred at the border through December 2018.

President Trump has repeatedly called asylum and similar programs that provide legal channels for immigrants “loopholes” and has blamed them for the “humanitarian crisis” at the border. But the evidence demonstrates that in fact, the border fence caused the humanitarian crisis and led to numerous deaths along the border, while asylum has made the border safer and more secure. From a security perspective, immigrants turning themselves in to agents is certainly better than them trying to evade capture. In other words, the humanitarian procedures are a win-win for security and immigrants.