$250 million question: Where'd foreigners go?

Pricey U.S. effort to track exiting visitors is a bust

By AUBREY COHEN
P-I REPORTER

The U.S. Department of Homeland Security has spent $250 million over the past four years to find a way to track when foreign visitors leave the country. And there's still no effective system.

While government officials hold out hope for technology that doesn't exist yet and insist they are making progress, others say they should stop throwing good money after bad.

"They just keep putting more money into it and hoping that will solve the problem," said Lee Tien, a senior staff attorney with the Electronic Frontier Foundation, a San Francisco non-profit that focuses on citizens' digital rights. "But there's just no evidence it will."

The U.S. Visitor and Immigrant Status Indicator Technology program, or US-VISIT, is aimed at tracking visitors as they enter and leave the country without harming cross-border trade. The department has spent about $1.3 billion on US-VISIT since 2003 -- including the $250 million on exit tracking. It now has a system that takes photographs and biometric data, such as fingerprints or iris scans, from foreign visitors entering the country at most air, sea and land crossings, and feeds those details into federal databases.

But it has not met the original goal of showing when foreigners overstay their visas.

"Without better definition and justification of its future exit efforts, the department runs the serious risk of repeating its past failures," wrote Randolph Hite, the U.S. Government Accountability Office's director of information technology architecture and systems issues, on Aug. 31, in the most recent of a series of critical reports.

Border crossings historically were set up to screen people entering -- not leaving -- the country. Congress first mandated a better system to track visitors in 1996 after learning that a plotter in the 1993 World Trade Center bombing was a student who overstayed his visa. In 2000, it called for that system at all airports and seaports to be in place by the end of 2003, at the 50 busiest land crossings by 2004 and at remaining land crossings by 2005.

The idea is if the government knows who has overstayed their visas, federal agents can try to track them down.

Failed pilot projects

Homeland Security officials committed in January 2004 to have a biometric exit system in place at land border crossings by the end of 2005.

However, they concluded in January 2005 the only way to make such a system work at land crossings would be to create booths where all cars would stop as they left the country -- and that the potential trade impact from traffic backups plus the cost, estimated at more than $3 billion, made that unfeasible.

That's when the department turned to radio frequency identification tags, launching a pilot project in 2005 at land crossings in Blaine and elsewhere along the northern and southern borders. Officers gave foreign visitors paper forms with tags, which machines could detect when visitors drove or walked across the border.

But, without a biometric component, the system only could ensure the piece of paper -- not necessarily the person -- had left, critics noted. As it turned out, it didn't even do that.

In one-week test periods, the system read 14 percent of the documents leaving the border through Blaine's Pacific Highway crossing and 4 percent of those at the Thousand Island Bridge in Alexandria Bay, N.Y. -- well short of the target success rate of 70 percent. For pedestrians leaving the country at the Nogales-Mariposa crossing in Arizona, the rate was 67 percent, lower than the 95 percent target.

These results provided some vindication for people like Greg Boos, a Bellingham immigration lawyer who has been fighting increased border requirements and criticizing proposed security measures for years.

"We always said it wouldn't work," Boos said earlier this month.

US-VISIT Director Robert Mocny acknowledged in an interview Thursday, "The numbers weren't great."

Homeland Security officials ran into similar problems with kiosks and mobile systems used in pilot projects at Sea-Tac Airport and 13 other air and sea sites, from 2004 through May 2007. On average, just 24 percent of those whose visas required them to use these systems did so. The highest success rate was 36 percent.

In total, the government spent about $64.2 million on the land pilot project and $92 million on the air and sea pilots.

Officials now plan to incorporate US-VISIT into airline check-in processes and propose regulations for the air system by the end of the year. But airline and airport officials say this would cause delays and hinder passengers who increasingly are checking in before they get to the airport.

The prospects for land crossings are dimmer.

Homeland Security officials have talked about adding biometric verification to the radio chips by making visitors activate the chips with a thumbprint while driving across the border.

Phillip Bond, president and chief executive of the Information Technology Association of America, held out hope for the radio chips at a January hearing of the Senate Judiciary subcommittee on homeland security.

"This kind of technology works," he said, adding that chips with biometric capabilities are "on the horizon."

Tien said Homeland Security officials seem to have tailored criteria for the land-exit solution specifically for radio tags.

"The general picture that I, as an outside observer, get is they started with a solution and tried to implement it, and when they discovered the solution wasn't going to work, they hit a wall," he said.

Companies working on the technology have no incentive to say it won't work, Tien said. "They'll keep spending taxpayer money to do something they know can't be done."

Mocny said the pilot test showed the radio chips do not work well while cars are moving at near-highway speeds. The thumb-activated chips are still too big and expensive to work, he said, adding that it might not be a good idea anyway to make drivers take a hand off the wheel to press their thumb to a chip while driving across a border at 45 mph.

"An at-speed departure with biometrics doesn't seem to be ready for prime time," Mocny said.

Instead, Homeland Security officials plan to talk with their Canadian and Mexican counterparts about having them collect information as foreigners enter those countries from the U.S.

Such a system could range from having border officers collect paper forms, as sometimes happens informally now, to installing radio-chip readers at Canadian and Mexican entry stations, where cars already stop, Mocny said. But, he said, negotiations with foreign governments take years and, because Mexico doesn't have stations at all southern crossings, such a system might require the U.S. to build facilities in Mexico.

No solution?

But even if it worked, a system that successfully showed which foreigners had left and, therefore, who overstayed their visas wouldn't really be that useful, Boos said.

"How do we know where to find them?" he asked. "This is a big country. People can get lost awfully easy."

Richard Stana, director of homeland security and justice issues for the GAO, noted at January's Senate hearing that the government now has 200 to 300 agents searching for an estimated 5 million people who have overstayed visas.

"If you consider the numbers, you would see they are not going to get very far," he said.

Some critics say the real answer is to give up on a biometric exit system.

"Right now, there is no good solution," Stana said at January's hearing. He suggested expanding trusted-traveler programs and non-border measures, such as better work-site enforcement, beefed-up intelligence collection and extra agents to search out visa overstays.

Boos said the real solution would be a system to keep undesirable people from entering in the first place.

Officials should weigh the costs of an exit system against its potential benefits, particularly given that the U.S. has thousands of miles of lightly guarded land and sea borders, Tien said.

Likening the situation to cars, he said, "We generally accept that the price of an automobile society is some number of automobile fatalities and injuries."

But Mocny noted that members of Congress have not responded favorably to the notion of giving up on biometric land exit.

At January's Senate hearing, for instance, committee Chairwoman Dianne Feinstein, D-Calif., said: "We have left a gaping hole in our country's border. ... By failing to address exits at all ports, we are providing a blueprint to those who wish to harm the United States."

Homeland Security officials, Mocny said, are emphasizing the challenges in discussions with congressional staffers and their plan for next year's budget talks about why a biometric land exit system can't be put in place yet.

But that doesn't mean he agrees with Stana, Boos and Tien on the need to give up on such a system.

"I think it's worth it," Mocny said. "We want to make sure that when an individual leaves the country we can close that book on that individual."

So how long will an effective biometric land exit system take?

"I don't have a good answer for you," he said.
P-I reporter Aubrey Cohen can be reached at 206-448-8362 or aubreycohen@seattlepi.com.

http://seattlepi.nwsource.com/local/332 ... source=rss