The Visa Overstay Problem and the Exit Tracking Congress Ordered in 1996

The wall gets all the cameras. Nobody films the arrivals hall, where there is no wall at all, just a counter, an officer, a few questions, a stamp or an electronic record, and a date by which you are supposed to leave. It is not dramatic, which is exactly why it gets ignored. Then you walk out into the sunshine and the government, for all practical purposes, takes your word for it.

That is where some unlawful presence begins. Not in the dark, not across a river, but under fluorescent lights with a valid visa and a plane ticket. If you only ever argue about the border between ports of entry, you are arguing about half the map. This post is about the other half: the visa that ran out, the departure nobody counted, and the law Congress passed in 1996 telling the government to count it. Thirty years later, the argument is still about whether it did.

I will keep this to the law, the mechanism and the proposals. I will not give you a single statistic that I cannot source, and I will tell you where to look for the real ones.

What an overstay is, and what it is not

First, the definition, because the word gets used loosely. A person is admitted to the United States on a visa or under the Visa Waiver Program for a specific period. The record of that period used to be a paper card stapled into the passport. Today it is mostly an electronic record, and the traveler can look up their own admission record on the government's I-94 website. If the person stays past the date on that record, they have overstayed.

An overstay is a civil violation of immigration law. It is not a crime. The person has violated the terms of their admission, and the law makes them removable for failing to maintain their status, but ordinarily they have not committed a criminal offense by staying. Compare that with entering the country without inspection, which is a federal misdemeanor on a first offense under 8 U.S.C. 1325, and reentering after removal, which is a felony under 8 U.S.C. 1326. I go through those distinctions in Is Crossing the Border Illegally a Crime? Eleven Answers From the U.S. Code, and the words matter, as you will see in The Immigration Words You Should Use Correctly Before You Argue.

That distinction between civil and criminal is not a loophole someone discovered. It is how Congress wrote the law, and it is a reason enforcement against overstays works differently from enforcement against illegal entry. A person who overstayed cannot be arrested at the border, because they are already inside. They are found, if at all, through other contacts: a traffic stop, a records match, a tip, an employer check, a later application.

The consequences the statute already imposes

Overstaying is not free, even if no one comes to your door. Congress built penalties into the statute in 1996, in the same law that ordered exit tracking. If someone accrues more than 180 days but less than one year of unlawful presence and then leaves, they are barred from returning for three years. If they accrue a year or more of unlawful presence and then leave, they are barred for ten. Those bars are in section 212(a)(9)(B) of the Immigration and Nationality Act. In addition, a visa is voided by an overstay under INA section 222(g), so the person generally has to apply for any new visa at the consulate in their country of nationality.

Think about what that does. The statute creates a penalty box for people who leave late, but a penalty box only works if the government knows they were late. That is the problem with the whole enterprise. A bar that depends on a departure record you do not keep is a bar you cannot enforce.

What Congress ordered in 1996

Enter the law. In September 1996, Congress passed the Illegal Immigration Reform and Immigrant Responsibility Act, the law behind so many of today's fights, which I described in The 1996 Immigration Law Behind Most of Today's Border Fights. Section 110 of that law told the Attorney General to develop an automated entry and exit control system. The system was to collect a record of arrival of every alien arriving in the United States, collect a record of departure of every alien leaving, and then match the two to identify anyone who stayed beyond their authorized period. The law set a deadline in October 1998 for airports and seaports and later dates for land borders.

It was a straightforward idea. You count the people who come in, you count the people who go out, and the difference is a list of people to look for. Any grocery store does inventory this way.

The deadline passed. The government did not build the system. Why not? The honest answer, which is the most important thing in this post, is that the land borders were the problem. Hundreds of thousands of people cross our land borders every day, many of them local, lawful and routine. Imposing an exit check on every crossing from the United States into Mexico or Canada would have required infrastructure that did not exist, and it would have clogged commerce and local life. Business groups and border communities objected loudly. Congress responded by softening the mandate, and in 2000 it passed the Immigration and Naturalization Service Data Management Improvement Act, which replaced the original system with a more flexible "integrated entry and exit data system" and shifted the deadlines.

What the 9/11 Commission added

On September 11, 2001, nineteen hijackers entered the United States and carried out the worst attack in the country's history. The 9/11 Commission, in its final report in July 2004, examined how they moved through the immigration system. It found that several of them had violated the terms of their admission, and it treated border security and immigration control as parts of counterterrorism. Among its recommendations was a biometric entry-exit screening system, so that the government would know who came in and who left, with fingerprints and photographs rather than a name that could be misspelled or changed.

Congress followed in the Intelligence Reform and Terrorism Prevention Act of 2004, which, in section 7208, called for a biometric entry and exit data system. Two things had already begun. In 2004, the Department of Homeland Security launched a program called US-VISIT, which collected fingerprints and photographs from visitors on entry at the ports. And exit pilots were tried at a number of locations. The exit side was always the weak half. Entry is easy to build because the traveler has to stop and talk to an officer. Exit is hard because most travelers leave without stopping.

How exit is counted today, and what is missing

Here is where I need to be careful and tell you what I can say with confidence.

For air and sea travel, the government has had a good record for years, because airlines and cruise lines are required to send passenger manifest data to the government before departure. That data lets the government match the departure to the arrival record for most people who leave by plane. It is not biometric in every case, though the government has tested and expanded facial comparison technology at some airports, and the Customs and Border Protection website describes the current status of that program. I leave the details there because the program changes.

For land travel, the picture is different. At the northern border, the United States and Canada agreed some years ago to treat each other's entry records as exit records, so that a person who enters Canada by land is recorded there, and that record can be matched with the United States' departure record. At the southern border, there is no comparable system for people leaving the United States into Mexico. If you want to see where the gap remains, read the annual overstay reports that the Department of Homeland Security has published for several years. They are public, and they report the numbers that I am refusing to guess at. Go look, and notice how the report explains what it can and cannot count.

There is also one tracking system that works well and that is rarely mentioned: SEVIS, the Student and Exchange Visitor Information System. It was authorized by the 1996 law after the 1993 World Trade Center bombing, and it requires schools to report on foreign students and exchange visitors. When the government wants to know whether an F-1 student is where he says he is, it has a database. The existence of SEVIS shows that the government can track people when the law requires a third party, such as a university, to report. That is a model worth thinking about.

The other side's best argument

Let me state, as its supporters would, the best case that overstay enforcement is a distraction.

"Overstays are mostly a paperwork problem. The people involved came through the front door with a visa. They were screened. Many are working or studying or married to Americans, and many will leave on their own. Spending billions on exit tracking will not find them, and the real issues are elsewhere. A modern economy needs a flow of people in and out, and any system that makes that flow slow punishes lawful travelers and the businesses that depend on them."

There is something to it. Overstays are not a single group. A student whose paperwork lapsed during a renewal is not the same as a person who disappeared. A system that treats them the same wastes resources. And the cost of a heavy exit check on every land border would be real.

But the argument proves less than it seems. The reason to count departures is not to punish the student with the lapsed paperwork. It is to give the government a reliable measurement. If you do not know how many people stayed past their dates, you cannot judge whether the visa programs are working, you cannot tell which countries' travelers overstay at higher rates, and you cannot give a consular officer data to use when deciding whether to issue the next visa. A measurement is not a punishment. It is the thing that lets you set policy by evidence rather than anecdote.

What conservatives propose, without inventing numbers

The fixes that come up most often in conservative policy discussions are these:

1. Finish the biometric exit at airports and seaports. This is the most feasible, because the traveler is already at a controlled point and the airlines already send the data.
2. Make the land-border solution real. Use data-sharing with neighboring governments where it works, and invest in technology at the major land ports instead of promising a single national checkpoint that will never come.
3. Use the visa itself as leverage. The State Department has authority under INA section 243(d) to stop issuing visas to countries that refuse to take back their own nationals who have been ordered removed. A country whose citizens overstay and then cannot be returned has a reason to cooperate if visas are at stake.
4. Tie the Visa Waiver Program to performance. Countries in the program get easier travel for their citizens in exchange for meeting security and compliance standards. It is reasonable to ask that the standards include cooperation on overstays.
5. Verify employment. If employers check work authorization, a person who overstayed cannot easily hold a regular job, and the incentive to stay shrinks. See E-Verify Explained, the Free Federal Check Most Employers Skip.
6. Prioritize. The government cannot find everyone. Decide who matters most, those with criminal records or security flags, and say so openly.

Notice what is not on the list. Nobody serious proposes to stop tourism or student visas. The proposal is to make the system report what it already promises on paper.

Why it matters to a family

You may be wondering why a farm family in Iowa or a retired couple in Arizona should care. Here is the answer. If you believe the law should be enforced fairly, the overstay is the test of whether it is. A system that is strict at the river and loose at the airport is a system that people will see through. The family waiting years for a green card sees the difference. So does the traveler who follows the rules and returns on time.

It matters for a second reason, too. Visa policy ought to rest on what we know. For decades Congress has legislated about the visa programs without a reliable count of who leaves. Imagine running a farm without a count of the cattle. You might do fine for a while, and then you would not.

And a third: security. The 9/11 Commission treated the entry-exit system as part of the answer to a deadly attack. We have an obligation to ask whether the answer was built.

For the larger picture of how the doors into the country are organized, and who holds the keys to each, see The Four Doors Into America and Who Holds the Keys to Each.

The bottom line

• An overstay is a civil violation, not a crime, but it carries real penalties: a three-year bar after more than 180 days, a ten-year bar after a year or more, and a voided visa.
• Congress ordered an automated entry and exit system in 1996. The 9/11 Commission and the 2004 intelligence reform act pushed for a biometric one. The air and sea side is mostly built. The land side is not.
• Count first, then punish. A reliable departure record lets Congress judge visa programs on evidence, not anecdote.
• Support the practical fixes: finish airport exit, use data-sharing at the land borders, tie visas to a country's cooperation on removals, and verify employment.
• Look up the Department of Homeland Security's annual overstay report and the Customs and Border Protection pages yourself. They have the current figures, and I would rather send you there than guess.