PhD Partnership in Global Social Protection Transnational Studies Initiative – Harvard University

Starting in January 2014, TSI will host one or two PhD or post-doctoral students working on related topics who would like to partner with us on this intellectual journey. These individuals will work with Professors Levitt and Viterna on developing these ideas, participate in all TSI activities, and be welcome to take part in all open University seminars and public conferences. Please download the announcement for details.

Immigration reform predictions are mathematical and personal

BY CINDY CHANG   JULY 30, 2013

Tom Wong sat in the parking lot of a San Diego McDonald’s, scarfing a double cheeseburger and listening to the Senate’s roll-call vote on immigration as it live-streamed over his iPhone.

Landrieu, aye. Leahy, aye. Lee, no.

Just as he had predicted.

Finally, the 100th name: Wyden, aye. Relieved and smiling broadly, he called his wife with the good news. Not only had the bill passed, but his statistical models had worked nearly perfectly. He was right about all but a few senators.

As the immigration battle shifts to the House, word has spread among activists that Wong might be the Nate Silver of immigration reform — the go-to data geek with the crystal ball.

But Wong doesn’t just want to predict the future. He also wants to change it, by giving immigrant-rights advocates the statistical ammunition they need to influence lawmakers.

The UC San Diego assistant professor recently led a conference call with about 30 advocacy groups, including Mark Zuckerberg’s Fwd.us. Still more pro-reform organizations joined a second phone call to talk about his projections.

This time, he would like to be proven wrong. His models show a range of measures, including a path to citizenship for the 11 million immigrants in the country illegally, flaming out in the House.

“Ultimately, if these groups are effective, then hopefully some of these ‘nos’ become ‘yeses,'” said Wong, 31. “Then, my final tally might be off, but in a way that might make me happy.”

With his crew cut and compact, athletic build, Wong projects a cheerful efficiency, parrying criticisms with a dose of self-deprecation. Outwardly, nothing sets him apart from other ambitious young academics. But his past drives nearly everything he does.

Once, he was one of the 11 million.

A graph on Wong’s computer screen, with 100 dots representing all U.S. senators, helps him predict their votes on immigration reform. (Don Bartletti / Los Angeles Times)

King Bun and Shiao Ping Wong came to Southern California with their two young sons on tourist visas in 1984, fearing for the future after British-ruled Hong Kong was reclaimed by China.

Their college degrees worthless because they couldn’t speak English, they worked at a Chinese restaurant, hoping to get legal status through their employer. The restaurant closed.

On weekends, they sold toys at swap meets. One employer after another promised a green card, then went out of business. They put off telling the boys — surely, a solution would come along eventually.

When Wong was 16 and wanted a driver’s license, his parents said no. They were just being strict, he thought. He asked to go to a basketball tournament in Canada. That’s when they told him the truth: He wasn’t an American citizen or even a legal resident. The tournament was out of the question. If he tried to cross the border without papers, he might be detained.

He was a toddler when he left Hong Kong, and had never given a thought to his immigration status. He assumed he was American, like everyone else he had known since kindergarten.

“Everything I thought I was, was completely shattered,” Wong recalled. “I felt broken.”


In his spartan third floor office in the weeks before the Senate vote, Wong tapped rapidly on his laptop keyboard. A sequence of dots and lines emerged — his prediction for a border enforcement measure. The solid “nos” clustered in the lower left, the solid “yeses” in the top right. Each line represented a lawmaker’s chance of switching sides.

For those hoping to influence the outcome, the sweet spot is the middle of the graph — the “lean yeses” and “lean nos.”

In the mold of Silver, who is famous for his election predictions, Wong bridges the gap between equations and shoe-leather politics, said David Damore, a political science professor at the University of Nevada, Las Vegas and a senior analyst for Latino Decisions, a political opinion research group.

Activists already have an idea of which lawmakers to target, but Wong gives them an extra edge. He can generate a custom analysis for, say, who might be receptive to an argument based on religious faith. With the House likely to consider separate measures rather than a comprehensive bill, Wong covers every permutation.

“In the House, everybody’s in their own unique geopolitical context,” Damore said. “What he’s doing is very, very useful.”

The equations Wong uses are familiar to many political scientists. So are his raw materials: each lawmaker’s past votes and the ethnic composition of his or her district. But no one else appears to be applying those tools to immigration in quite the way Wong does.

Neema Singh Guliani, a legislative counsel for the American Civil Liberties Union, which is part of the CAMBIO coalition seeking immigration reform, said: “It helps to know where you’re starting from, that this is a member on the fence, or who has had the population in the district change dramatically and so might have a different approach to immigration than in the past.”


King Bun Wong, right, and Shiao Ping Wong came to Southern California in 1984 from Hong Kong with their sons David, left, and Tom. (Wong family photo)

After his parents revealed their secret, Wong took it out on them, yelling at them in broken Cantonese, switching to English for the curse words.

“If you knew it was going to be like this, why’d you bring me here?” he remembers saying.

At his Riverside high school, he had been a popular jock with good grades. But what was the point when he couldn’t go to college, when he would end up working at the swap meet like his parents? He stopped studying and barely graduated.

Wong’s girlfriend, Rose Bloomberg-Rissman, wanted him to get back on track.

She had been dating him only for five months, but she was sure she wanted to spend her life with him.

When she proposed, Wong was so shocked that he mumbled something noncommittal. The relationship was going well, but so soon? He was only 19. He had long resisted the idea of marrying for a green card. But Rose was the one. They were only speeding up the timetable.

The next day he said yes. They were married in Las Vegas on March 24, 2001. They sailed through their immigration interview, and Wong began classes at UC Riverside.

More than a decade later, they’re still together, the parents of 2-year-old triplet boys Soul, Pace and Ever.

Wong is haunted by his experience and inspired by it. He mentors young immigrants without legal status — the so-called Dreamers, kids who remind him of himself. He funds scholarships so they can apply for a new federal program and receive temporary work permits.

“I saw school as one of the things that was taken away from me,” Wong said. “So I had a chip on my shoulder. I wanted to show everybody that I could actually do this and do it well. So that’s what gave me the drive to go through undergrad, graduate magna cum laude, get my PhD, get an academic position. After I get tenure, I’m going to law school.”


With House Republicans signaling their reluctance to support a path to citizenship, the pro-immigrant forces are looking for every vote they can get.

For a comprehensive bill, Wong predicts, at most, 203 likely yes votes, when 218 are needed. Many House members are from districts with few immigrants, insulating them against Latino and Asian voter backlash.

The House is unlikely to pass even a measure awarding citizenship to the Dreamers, according to Wong’s analysis.

Hurried fast-food lunches have become the norm for Wong as he fields requests from dozens of groups hoping to coax legislators into the “yes” column. He works mostly pro bono, with the occasional stipend. So far, no one from the anti-legalization side has approached him. His answer to them would be no, though they are free to use the data on his blog.

“If there’s actually legislation that goes through that takes the rough edges off some of our immigration policies, and I can say I played a role in that, that’s motivation enough for me,” Wong said.


The curses he once hurled at his parents make Wong cringe.

Now he knows how byzantine the American immigration system is, that sometimes no matter how hard people try, a green card remains out of reach. His brother also married a U.S. citizen. For his parents, the road was almost two decades long.

Tom was angry when he found out about his immigration status, but understandably so, his parents recall. Their pride in him is obvious. “He’ll be governor one day,” his father interjected at one point.

“We had bad luck, and it affected our kids. I feel very guilty,” his mother said in Mandarin, which she and her husband speak in addition to Cantonese. “Everyone hopes things will go smoothly for their children. We really tried.”

Wong desperately wants to apologize to them. But he can’t, not yet. It’s painful to dig up those dark days. And there is the language barrier — his Cantonese remains stalled at a kindergarten level, while his parents never learned English.

For now, he speaks through others.

“I will tell my story to any stranger, but I’m still not comfortable enough with my family and closest friends to get deeply into it,” Wong said. “There are still wounds those conversations might open up. This is so much easier — I know my parents can hear my story through what you write.”

Read the article ».

Border security faults may be result of poor analysis

BY BOB ORTEGA   JULY 20, 2013

AGUA PRIETA, Sonora – Three times in the past month, Paloma Flores Lopez has been caught by the Border Patrol crossing into the U.S. illegally. She’s spent $8,000 on “coyotes,” or guides, to take her across. Now Flores, 26, has to decide: Will she try again?

She knows the dangers. But she says she sees a chance for a better life in the United States that she doesn’t see in Michoacan, where she’s from, or in Tijuana, where she lived briefly before trying to cross in Agua Prieta, where an 18-foot-tall steel fence looms between Mexico and Douglas, Ariz.

Arizona delegation demands information

So, said Flores, flipping her pony tail, “I have to try.”

As Congress weighs whether to pin immigration reform on reaching a threshold of border security, the measure most often cited would call on the Department of Homeland Security to stop 90 percent of illegal border crossings. Doing that means figuring out how to persuade people like Flores not to try again and stopping others headed for el norte from slipping over the border.

That, in turn, hinges on solid answers to such questions as: How many people actually get through? Where do they get across? When they’re caught, do they give up or keep trying until they make it?

Homeland Security officials don’t fully know the answers to those questions. And the reason, say leading migration researchers, is that DHS officials don’t want to know, and don’t want the public to know, either.

“There is zero interest in that kind of analysis among DHS’ leadership,” said economist Bryan Roberts, who served as the agency’s assistant director of the Office of Program Analysis and Evaluation until 2010. “There was no interest when I was there, and there still isn’t any.”

Among the kinds of findings outside researchers are seeing that they say the DHS ought to be examining are:

Three-quarters or more of migrants who set out to cross make it. Even when they’re caught, detained for days or weeks, or deported hundreds of miles from where they crossed, most simply keep trying until they succeed.

Tough “consequence” programs, imposing more and tougher criminal penalties on crossers, are very expensive and don’t seem to be an effective deterrent.

The billions of dollars spent on fences, thousands more agents, and high-tech surveillance have pushed migrants into more remote regions, driving up the number who die each year. But when it comes to stopping migrants or getting them not to try, the fences and agents pale compared with factors such as how the U.S. economy is doing, or how scared migrants are of the drug cartels in northern Mexico.

Over the past four years, more undocumented migrants have left the United States than have entered.

Roberts and several other researchers said that the DHS doesn’t have the answers because it doesn’t jointly analyze data from the Border Patrol, which works between ports of entry; Customs and Border Protection, which works at ports of entry; and Immigrations and Customs Enforcement, which works across the interior of the country.

Not looking at the big picture makes it harder for the DHS to figure out whether, say, to build more fences or focus on interior enforcement, the researchers said. And there’s little pressure on the DHS to work with outside experts or better analyze its data to figure out what has worked and what hasn’t.

In response to written queries from The Republic, DHS spokesman Peter Boogaard issued a statement last week that “DHS measures success utilizing many metrics, each of which paints a different portion of the overall border security picture and each of which informs tactical decision making.”

Boogaard noted that the Obama administration has made “significant investments in border security” and that “dozens of metrics we use every day clearly demonstrate significant progress and improved quality of life at the border.” Boogaard would not address any specific questions.

Controversial findings

It is clear that the past seven years have seen an unprecedented investment in border security, topping $106 billion. There’s also no dispute that apprehensions, often seen as a rough proxy for crossings, have fallen by 66 percent since 2006. And the Border Patrol, its parent agency Customs and Border Protection and the department they’re part of, Homeland Security, gather reams of data on migrants they apprehend or spot. But researchers say that the DHS and its agencies seem to take a don’t-ask, don’t-tell approach to data that might lead to politically troublesome conclusions.

The immigration bill passed last month by the Senate, now languishing as the House weighs its own measures, would evaluate border security using an “effectiveness rate” based on data that the DHS and its agencies won’t release to the public. The bill would provide more than $46 billion and nearly double the size of the Border Patrol by adding 19,200 agents along the Mexico border over the next eight years, to help the agency reach a 90 percent effectiveness rate.

That rate is meant to show how effectively the Border Patrol prevents illegal crossings. To estimate the number of illegal crossings, the Border Patrol adds up three figures: apprehensions; “turn-backs” (people spotted starting to cross who turned back to avoid getting caught); and “got-aways” (people detected by agents or surveillance equipment but not caught). A 90 percent effectiveness rate means nine apprehensions and turn-backs for each got-away.

The Border Patrol doesn’t release information on turn-backs or got-aways to the public, just apprehensions; and it admits that the effectiveness rate is a flawed yardstick. Among other gaps, it can’t account for crossers whom agents don’t see. And because the Border Patrol works between ports of entry, its rate doesn’t include those who cross illegally at ports of entry, either hidden in vehicles or using false documents.

The Government Accountability Office, in a report last December, published previously unreleased data showing that the Border Patrol’s effectiveness rate for fiscal 2011 was 84 percent. By that yardstick, getting to 90 percent isn’t a huge stretch, noted former DHS official Roberts.

The Border Patrol hasn’t released turn-back or got-away data for fiscal 2012, and hadn’t responded by deadline to The Republic’s request for that information.

Outside researchers say efforts to come up with a better approach to accounting for undocumented migration run smack into Homeland Security’s unwillingness to let academics analyze its data.

Last year, for example, a panel of leading statisticians, economists and demographers at the National Academy of Sciences conducted a study on illegal immigration at the request of Homeland Security. But the DHS refused to provide the panel key apprehension data, such as coded fingerprint figures that would identify precise numbers of repeat crossers. The DHS had demanded that researchers promise not to disclose that data to the public. Panel members said keeping the information classified would impair the quality of their work; they declined, and didn’t get the data.

That study, which included data from Mexican governmental sources and previous U.S. academic studies, suggested that about three-quarters of those who decide to cross keep trying until they make it. Other outside studies have found 85 or even 90 percent make it.

“Almost everybody who really tries eventually gets in,” said Jeffrey Passel, a member of the panel and a senior demographer at the Pew Hispanic Center, a nonpartisan research organization in Washington, D.C., that studies the U.S. Hispanic population.

The National Academy of Sciences study essentially was ignored in presentations that the DHS gave to the Senate earlier this year during the immigration-reform debate, said the study’s panel members.

They said the DHS was not eager to draw attention to the study’s findings even though it paid for the report. “In a sense, it throws a monkey wrench into the discussions on immigration. I’m totally for immigration reform, but this report would make Republicans giddy and Democrats go, ‘Oh, crap,’ ” said Alicia Carriquiry, a professor of statistics at Iowa State University and a co-author of the study.

The massive buildup in border security may make some difference in reducing illegal immigration, the researchers said. For example, independent research suggests that many migrants who might have tried to cross in the past now look at the dangers and don’t set out in the first place. Those people don’t show up in Border Patrol or DHS data.

But Carriquiry and several other researchers said they’re frustrated that enormous decisions about future spending on border security are being made with little or no reference to research.

“The effectiveness rate is pretty deceptive,” Carriquiry said. “There is a lot of data one could use to get a much better idea of the probability of successful crossing, and those data are being held very close to the vest by DHS.”

Seemingly ignored

Researchers say similar data was ignored in the last go-round on immigration reform, in 2006 and 2007, when a bill passed the Senate but failed in the House. Instead, lawmakers ramped up border security: building fences, hiring thousands of new Border Patrol agents, and acquiring billions of dollars in new technology.

At that time, CBP asked the Homeland Security Institute, a federally funded research center, to study border-crossing recidivism and the likelihood of apprehending crossers. The study found that, from 2001 to 2005, when border security and the consequences imposed on crossers were both relatively slight, the likelihood of being apprehended on any crossing attempt was about 35 percent, according to sources familiar with the study. But to this day, that study, completed in 2007, remains classified as “law-enforcement sensitive.”

“The department doesn’t want to release it, and they have the final say,” said Joe Chang, the author of the report and a corporate fellow at what is now called the Homeland Security Studies and Analysis Institute. Chang said he worked almost continuously on expanding and updating that study year after year; but all of that work remains classified, too.

It isn’t clear what impact, if any, that study had on the DHS’ border-security strategy during the buildup, which created 652 miles of fencing and pushed migrants into more remote and dangerous areas.

More than two years ago, the DHS promised Congress that it would create a new Border Conditions Index to measure more accurately the impact of the additional agents, fences and other infrastructure on a range of border-security issues. But in March, as the immigration-reform debate got under way, officials told Congress that they couldn’t report any progress on the index. At the time, DHS spokesman Boogaard said that declining apprehension numbers showed that “we’re heading in the right direction.”

Apprehensions have dropped dramatically in recent years, from more than 1 million in 2006 to fewer than 365,000 last year.

Meanwhile, since about 2008, at least as many undocumented immigrants have gone back to Mexico as have come north, said Passel, the demographer, for a net flow of zero.

Other researchers see a stronger flow south. Roberts estimates that, based on apprehension data and surveys of migrants in Mexico, between 250,000 and 270,000 undocumented migrants likely entered the United States last year. By comparison, Customs and Border Protection deported just under 410,000 people last fiscal year.

That balance likely will change, say some researchers, as the U.S. economy slowly recovers. So far this year, the Border Patrol has reported a 13 percent increase in apprehensions of illegal border crossers compared with a year ago, though numbers remain near historic lows. And that rise doesn’t necessarily mean overall crossings are up to a similar degree — because many apprehensions are repeats, the same person is being caught several times.

Out of 365,000 Border Patrol apprehensions last fiscal year, barely half, 183,000, were of people being apprehended for the first time, according to data obtained through a Freedom of Information request by the non-profit news organization Center for Investigative Reporting in Berkeley, Calif. The rest were people caught at least for the second time, and 21,000 had been caught six or more times, the center said.

To discourage crossers, the Border Patrol uses what it calls consequence programs. These include more criminal prosecutions, tougher penalties, and moves to make it harder to come right back, by transporting those caught hundreds of miles either to another part of the border or back south to their hometowns in Mexico.

But the programs have proven expensive, and the prosecutions have clogged federal courts. Last year, immigration cases made up more than 40 percent of all federal prosecutions nationwide, according to the Department of Justice. In high-traffic border sectors such as Tucson, prosecutors have had to limit immigration prosecutions to no more than 70 a day.

As a result, many migrants don’t get hit with these penalties. Paloma Flores Lopez, for example, said she and her sister were held 24 hours, then sent right back across to Agua Prieta all three times they were caught trying to cross last month.

That’s common. One recent morning, at the modest Migrant Resource Center in Agua Prieta, less than 100 feet from the border fence, volunteers passed out homemade burritos and cups of coffee to eight migrants, sweating and fanning themselves in a room that has no air-conditioning. One woman hadn’t crossed yet. Five of the other seven had been caught trying to cross at least twice in prior weeks.

“My husband wants me to try again; I haven’t seen him for eight years, since he went north,” said Lilia Garcia, from the Mexican state of Guerrero.

She’d been caught two times in eight days, she said. Both times, she walked at night with a group several hours out of Agua Prieta, crossed the fence, and then headed to Douglas. “I think two times is enough. My husband wants me to keep trying, but I want to go back home,” she said.

Keep trying to cross

Garcia’s inclination to give up is unusual, say researchers.

David FitzGerald, a sociologist at the Center for Comparative Immigration Studies at the University of California-San Diego, is part of a project that since 2005 regularly surveys three towns in central and southern Mexico that send large numbers of undocumented migrants north.

“Consistently, the vast majority keep trying until they get across,” often paying for door-to-door service from coyotes, FitzGerald said. “In our most recent survey, in January 2013, about 85 percent of those who try to cross are getting across.”

At the same time, he says, the surveys show evidence of what he calls “remote deterrence,” people deciding not to set out. The top two issues people cite, he said, are a belief the U.S. economy is still bad (making it harder to find work), and fear of running afoul of the drug cartels that control northern Mexico.

“People aren’t afraid of being caught by the Border Patrol; they’re afraid of physical harm and death,” FitzGerald said. “Fear of crime along the border is the most important fear. The terrible things that have happened are very well publicized even in the smallest villages.”

And people fear dying of exposure or thirst in the desert.

“To the extent that U.S. policy has any effect, it’s based on the fact that our current policy is channeling migrants into remote areas where they’re dying at the rate of more than one a day,” he said.

Their research also suggests that the Border Patrol “consequences” programs make little difference. Even the long bans imposed on migrants who sign voluntary-departure orders, barring them from legally entering the United States for five or 10 years or longer, have little effect, said Wayne Cornelius, founder of the Center for Comparative Immigration Studies and a political-science professor at UC-San Diego. He has researched Mexican migration for 38 years.

“Factors like family reunification trump fears of incarceration or prosecution. If you have children or spouses in the U.S., that’s going to trump everything else. If you have a desperate economic situation in Mexico, and you can’t feed your family, that’s going to trump any fears of enhanced consequences,” Cornelius said.

“U.S. policy has consistently underestimated the sheer determination, the sheer tenacity of Mexicans once they have made that decision. They will find a way to rationalize the costs and the risks and to borrow the money to make the trip, and to persist until they succeed.”

Where researchers find a deterrent effect, there are significant trade-offs.

Pia Orrenius, a senior economist at the Federal Reserve Bank of Dallas, has studied Operation Streamline, a controversial policy calling for criminally prosecuting nearly all undocumented crossers. First-time crossers receive sentences of up to six months; second-time crossers are subject to felony prosecutions and much longer sentences before being deported. That program began in 2005 on part of the Texas border, and the next year in Arizona’s Yuma Sector, but has since been extended to most of the border.

As the policy was rolled out across more and more of the border, the deterrence effect vanished.

“Once most of the sectors implemented the policy, there was no overall negative effect on apprehensions,” said Orrenius. She said that having to serve jail sentences does seem to matter, “but it’s also the most costly for the U.S. government.”

Last fiscal year, the U.S. spent about $2 billion detaining undocumented immigrants, at an average cost of $164 per day per person, according to the DHS.

A study published in May by the Congressional Research Service detailed how aggressively the U.S. government has pursued its consequence programs: In fiscal 2005, more than three-fourths of those apprehended were granted voluntary returns, which allows them to avoid criminal charges. Last fiscal year, fewer than one in seven was granted a voluntary return — with seven out of eight getting some combination of criminal prosecution, formal removal or being transported hundreds of miles from their original crossing point.

Even as apprehensions along the border fell by 58 percent from 2007 to 2012, the absolute total number of criminal prosecutions and other enforcement actions taken against migrants climbed by 60 percent, to nearly 453,000. (Some people were subject to more than one type of enforcement action.)

But researchers continue debating the impact of these programs, with many suggesting that, in the end, they are not proving particularly effective at discouraging crossers.

“Look at it this way,” says Bryan Roberts, the economist. “Imagine if you could triple your salary by emigrating illegally to Canada. … We have a wage gap of at least three to one with Mexico. It’s very hard to fight against that.”

One can hear echoes of that notion in conversations with would-be crossers at the Migrant Resource Center in Douglas.

“It’s hard because one comes to escape from Mexico. There’s a lot of poverty, corruption, narco traffic,” says Ivan Pacheco, 28, from Teotihuacan, in central Mexico, puffing on a cigarette outside the shelter. “I want to get ahead honestly, not by getting on with the mafia. But I don’t see a way to get ahead here.”

Pacheco says he’d like to join his brother in Utah. He crossed nine years ago, in Tijuana, without any problems. Pacheco has been apprehended twice by the Border Patrol trying to cross in recent days. The first time he signed a five-year voluntary deportation. The second time he says he had to sign a 20-year ban. He’d never been in any trouble with the law before, he said. But he sees little choice but to try again.

“I’ve been without work eight months,” he says, with a shrug. “Before that I worked in a paper factory. I earned 700 pesos a week ($55), working six days a week. You’re working to eat, nothing more.”

Republic reporter Brenna Goth contributed to this article.

Read the full article (subscription may be required).