The US Administration Wants to Control Immigration by Slashing Aid. Here’s What They Need to Know

June 04, 2018

The US is going to use aid to shape migration. That’s at least how the president’s remarks seem to have laid it out at an immigration roundtable last month, when he announced his White House is “working on a plan to deduct a lot of aid” from countries whose nationals arrive at the US border. “[W]e may not just give them aid at all.”

The target of these proposed cuts is clear. He was speaking at an event about gangs originating in El Salvador. The president’s words follow a recent series of remarks focused on the gang’s convicted and alleged crimes here in the US, purportedly the result of criminal entities sending members across the border as and among unaccompanied child migrants.

But slashing assistance would miss an opportunity to effectively and cost effectively shape migration through aid. Cooperation in Central America has the potential not only to meet this administration’s goal of reducing illegal migration, but simultaneously to extend more security and opportunity to children in the communities that child migrants are leaving. To achieve this, the administration’s strategy to shape migration through aid needs to done right. If evidence isn’t behind the president’s efforts, this policy will at best do nothing to deter illegal migration from El Salvador, and at worst will encourage it.

If what the United States wants to do is prevent irregular child migration in a way that works and is cost effective, it should not do what it has traditionally done—spend ten times as much on border enforcement trying to keep child migrants out as it spends on security assistance to the region. In fact, smartly packaged security assistance is the only things that have been shown to reduce violence effectively and cost effectively.

This is based on the initial evidence we have: that well-considered expansion of security assistance may reduce child migration. There is a lot of evidence we don’t have—namely, zero evidence that slashing aid will reduce migration.

Let’s take stock of what we do know:

First, cutting aid to the Salvadoran government will not make it stop promoting illegal migration by gang members, because the Salvadoran government is not doing that. Emigration from El Salvador is a choice made by individuals, not a policy of the government. And the Salvadoran government, like every other government, is barred under international law from restricting emigration of its nationals.  

Second, enhanced US assistance can meet the goal of reducing illegal migration, including by Salvadoran youths. This is because projects financed by US aid have been shown to reduce violence in the region, and that violence is a major driver of illegal migration.

We know this from independent and scientifically sound evaluations of such projects. The best example comes from a rigorous multi-year evaluation of USAID-funded crime prevention programming, under USAID’s Central America Regional Security Initiative (CARSI), including in El Salvador, Honduras, and Guatemala. CARSI deployed a package of community-based programs aimed at violence prevention, also encompassing tools such as vocational skills building, engagement of at-risk youth, efforts to boost employment, and community-based policing. The evaluation is reliable and transparent because it was carefully designed to compare randomly-chosen “treatment” and “control” communities, like a pharmaceutical trial. The package was shown to reduce reports of homicide and extortion by half.

Third, reduced violence acts directly to suppress illegal migration by youths. My research showed this through an unprecedented statistical analysis. I analyzed confidential government data on all 179,000 Unaccompanied Alien Children from El Salvador, Honduras, and Guatemala apprehended by US Customs and Border Protection over a six-year period, linking this to detailed data on violence in their communities of origin.

I find that a decline of ten homicides in an average municipality of this region caused six fewer children from there to be apprehended at the US border. I don’t just mean that less violence was associated with less child migration. I find that declines in violence caused less child migration, regardless of those Central American municipalities’ geographic location, size, urbanization, ethnic composition, or extent of prior migration.

Putting this together implies that cutting off US assistance to Central American governments in their fight the gangs and cartels could drive more youths to the desperate choice of emigration. That would miss a big opportunity to help those kids find the safety they deserve.

But it will also miss an opportunity to act in the direct national interest of the United States. Assistance that reduces the number of people moving in desperation takes revenue away from the transnational criminal networks that prey on those people.

Beyond this, unaccompanied child migration from Central America is a major burden on every taxpayer in the United States. Hannah Postel and I estimate that each apprehension of a child migrant at the US border costs taxpayers at least $50,000. Averting just one homicide per year in the region between 2011 and 2016 would have prevented about four child migrant apprehensions—a savings to US taxpayers of $200,000. So if a violence prevention program can stop just one homicide per year, and that program costs less than $200,000 (which is quite likely), that would mean a savings to US taxpayers.

Strategically designing foreign aid programs to effectively reduce violence, building programs on the evidence we have and piloting other ideas of what might work, can help shape the migration flows—including deterring UACs from leaving home.  Greater cooperation with Northern Triangle partners can advance the US national interest. Leaving Central America to its fate will do the opposite.


CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.