Does America face a refugee or an immigration problem?

Separating children from their parents at the southwest border ignited a broad  emotional outcry not often seen in the country’s long-running immigration debate. After weeks of defending the policy both as a deterrent and necessary to make the border legally meaningful, President Donald Trump and his administration reversed course.

The policy switch was itself a rarity for Trump, and it prompted the host of NBC’s Meet the Press Chuck Todd to ask the guests on his June 24 show, "Is this more of a refugee crisis than an immigration crisis?"

There is no definitive answer, but to frame the issue in terms of refugees instead of immigration redefines the problem.

"It becomes more of a humanitarian problem, than a jobs or economic one," said Sarah Pierce, an immigration analyst with the Migration Policy Institute.

We looked at what the data have to say.

For decades, the bulk of people crossing the southwest border without permission were Mexican. They worked on American farms, built American homes, processed American pigs, beef and chickens, and did other manual work.

The people showing up at the border today are more likely to come from El Salvador, Guatemala and Honduras, a cluster known as the Northern Triangle. In 2017, non-Mexicans stopped at the southwest border outnumbered Mexicans by 50,000.

Not only are those nations poor, they are also particularly violent. The highest murder rates in the world are found in El Salvador and Honduras.

As the places people came from changed, so did the demographics of who was caught at the border.

Between 2013 and 2017, the number of families apprehended by Customs and Border Patrol agents rose five-fold, from 14,855 to 75,622.

Minors (anyone under 18) went from 7 percent of all people stopped at the border in 2011 to 27 percent in 2017 – nearly a four-fold increase. In the same period, the fraction of females, both children and adults, doubled from 13 percent of all apprehensions to 26 percent.

"The recent crisis brings very different immigrants from those who were undocumented and came in the 1990s," said economist Giovanni Peri at University of California, Davis. "They are younger, more vulnerable and more similar to refugees."

Peri, along with most researchers, emphasizes that migrants have a blend of motivations. In the case of the Northern Triangle countries, escaping poverty and fleeing violence go hand in hand.

Teasing out the relative weight of each is tricky.

Fortunately for us, one economist took a stab at it.

Around 2012, a surge of teenagers without their parents began arriving at the southwest border. By 2013, for the first time, the unaccompanied minors from El Salvador, Guatemala and Honduras outnumbered those from Mexico. In 2014, they accounted for 75 percent of arrivals at the border.

U.S. authorities ended up collecting detailed information on 178,825 minors. Critically, that data included the towns and cities they had left behind.

Economist Michael Clemens at the Center for Global Development took that location data and ran it against the homicide data specific to those places. Violence, Clemens noted, isn’t spread evenly. It concentrates in hot spots.

He also folded in figures for local unemployment, income, poverty and school enrollment.

Clemens found that murders drive migration, but it wasn’t as simple as people fleeing the areas with the highest murder rates. What mattered most was a rise in homicides.

Clemens determined that on average, an increase of 1.08 homicides per year over four years in a community led to one additional minor from that community at the American border each year. To further complicate the picture, a jump in murders had a snowball effect, driving migration numbers up for several years going forward, even if the violence leveled off.

Clemens also found that poverty alone didn’t drive migration. Teenagers – and it was mainly 16-and-17-year-olds – tended not to come from the poorest places, but from places where there was enough money to pay smugglers to shepard minors through Mexico and across the U.S. border.

At the same time, there was a strong connection between sustained poverty and migration.

At the end of the day, Clemens gave about equal weight to violence and economic drivers.

The impact of "short-term increases in violence is roughly equal to the explanatory power of long-term economic characteristics like average income and poverty," Clemens wrote.

Put another way, the surge of minors at the U.S.-Mexico border is both an immigration and refugee issue.

A key caveat to Clemens’ work is that it looked only at unaccompanied minors, and it’s possible that the picture would change if adults and families were included. But based on studies of family migrants, Clemens said it likely would look the same.

"The decision for movement by unaccompanied children is one of safety and opportunity for children, and that is also heavily in the minds of family units," Clemens said. "The two kinds don’t self-report substantially different motives."

Clemens also noted that in the 2011-16 period, changes in U.S. policy had little impact on migration trends from the Northern Triangle.

For decades, a simple correlation tied illegal immigration from Mexico to the United States.

"When the U.S. unemployment rate went down, migration went up," said Jeffrey Passel, senior demographer at Pew Research. "If the unemployment rate went up, migration went down."

That held true through 2010. Then, the pattern stopped.

"After 2010, the unemployment rate dropped dramatically," Passel said. "If this model had held, Mexican migration would have gone up. Instead, Mexican migration went down."

Passel said the rise in numbers from other Central American countries, mainly the Northern Triangle group, is not a simple swap of one group for another. The demographic mix is entirely different.

"There’s something else going on that’s causing the Central American increase," Passel said. "The economy plays a role but it’s not solely determinative."

Passel and other researchers point to several reasons for the breakdown in the traditional pattern of illegal migration. Studies show a drop in the number of young men in Mexico, a group that was always the main source of border crossers. In some areas of Mexico, jobs are more available. And importantly, American border security has gone steadily up since about 2005.

"Just about everybody gets caught now," said Passel. "That seemed to work as a deterrent of Mexican migration, but those same factors don’t seem to be affecting Central American migration."

Peri, the UC Davis economist, suggested that the hardening of the border itself changed the mix of people caught trying to enter illegally.

"The very strict enforcement keeps away people who are motivated mainly by jobs, and selects only very desperate Central Americans, including many kids."

To the extent that dynamic is at work, then the primary solution to an immigration problem fueled a humanitarian one.

Legally, American laws and regulations count someone as a refugee by granting them asylum on the grounds that their personal safety is at risk if they stayed in their home country. As we’ve reported, the number of asylum requests has skyrocketed, but the fraction granted has held steady at about 20 percent.

"If they do not qualify for asylum either then they are not ‘refugees’ in the border sense of someone fleeing for their life and in need of resettlement," said the director of the Center for Immigration Studies Steven Camarota. The center promotes lower and more selective immigration.

"If someone has gone 1,500 miles through Mexico, and did not apply for asylum there, then it implies that they are not so much fleeing for their life, which is suppose to be the case for an asylee, but rather they are looking for a better life in the United States," Camarota said.

Passel at Pew Research noted that ties to family and friends play a large role when migrants pick a destination. Many people stopped at the border have an immediate or extended family member already in the United States. Migrants aim for places where community connections will help them once they arrive.

"Safety is one thing, but you have to live, too," Passel said.