Not so realistic: why some would-be immigration reformers don't have the answer.

THE Senate is again considering various proposals to address our massive illegal-alien problem, and the competing bills have one thing in common: They claim to offer "realistic" solutions to the supposedly unrealistic desire to enforce the law. Writer Tamar Jacoby, perhaps the most energetic salesman of the McCain-Kennedy amnesty bill, used some form of "realistic" ten times in her testimony at a July Senate hearing. Senators Kennedy, Cornyn, Brownback, and Feingold all touted the realism of their preferred solutions at the same hearing, and the New York Times and Washington Post have done the same in numerous editorials.

The problem, of course, is that no one has checked whether our very real immigration bureaucracy is capable of implementing any of these proposals. And this is no trivial concern: The success of any proposal depends on registering and screening millions and millions of illegal aliens within a short period of time--a daunting task.

It is therefore necessary to look first at the administrative mandates of the two most popular immigration bills. The McCain-Kennedy bill would offer amnesty to the approximately 11 million illegal aliens in the United States, by re-labeling them "temporary" workers, and after six years, it would grant them permanent residence. The Department of Homeland Security would have to determine that the person was, in fact, an illegal alien on the date of the bill's introduction; that he hadn't left the United States in the meantime; that he was employed in the United States at the time of the bill's introduction--"full time, part time, seasonally, or self-employed"; that he has remained so employed, which he can prove with records from the government, employers, labor unions, banks, remittances, or "sworn affidavits from nonrelatives who have direct knowledge of the alien's work"; that he, if not employed, was a full-time student; that he has "not ordered, incited, assisted, or otherwise participated in the persecution of any person on account of race, religion, nationality, membership in a particular social group, or political opinion"; and that he is not a security threat, a criminal, a polygamist, or a child abductor. This required background check is to be conducted "as expeditiously as possible"--for potentially 11 million people.

Now consider the better of the two major bills before the Senate, the Cornyn-Kyl bill, which has stronger enforcement provisions and isn't quite an amnesty. (It instead requires illegals to return home and sign up for its version of a "temporary" worker program from abroad, and there is no permanent-residence offer.)

A central component of the bill is "Deferred Mandatory Departure," which would give illegals who register with the government five years to get their affairs in order and leave, something you might call an "exit amnesty." The bill instructs the immigration service to determine many of the same things that McCain-Kennedy requires, including "the alien's physical and mental health, criminal history and gang membership, immigration history, involvement with groups or individuals that have engaged in terrorism, genocide, persecution, or who seek the overthrow of the United States government, voter registration history, claims to United States citizenship, and tax history." And the Cornyn-Kyl bill has even more specific deadlines: All applications are to be processed within one year of its enactment, and at that time the Department of Homeland Security also must have ready a new document for applicants that would be machine-readable, tamper-resistant, and have a biometric-identification component.

The point is not that these requirements are inappropriate; if you're going to be registering illegal aliens, you'd certainly want to know about their health status and involvement with terrorism. But the most pressing question remains: Is it achievable? What would happen, in the real world, if one of these "realistic" solutions were to become law?

Two words: "fraud" and "paralysis."

WILL WE EVER LEARN?

We've attempted much smaller programs like this in the past, and the story has been the same each time: The lack of administrative capacity, combined with intense political pressure, causes the immigration service to drop everything else to meet impossible deadlines-and ineligible applicants get through anyway.

The closest parallel is the administration of the 1986 Immigration Reform and Control Act by the old Immigration and Naturalization Service. Some 3 million illegal aliens (out of 5 million total) applied for amnesty, and 90 percent were approved. Fraud was omnipresent, especially in the farm-worker portion, where applicants presented fake proof of employment--sometimes as flimsy as a handwritten note on a scrap of paper--and gave ludicrous stories, like, "I've picked watermelons from trees," or "harvested purple cotton."

A report from the inspector general's office of the Department of Justice noted years later that "given the crush of applications under the program and the relative fewer investigative resources, INS approved applications absent explicit proof that they were in fact fraudulent." Two of the fraudulent applications that were approved were from Egyptian brothers in New York: Mahmud and Mohammed Abouhalima, who went on to participate in the first World Trade Center bombing.

We've seen paralysis before, too. On several occasions during the past decade Congress briefly opened a window for illegal aliens to apply for green cards without first returning home (by getting married to a citizen or legal resident, for instance, or by being sponsored by an employer). Hundreds of thousands applied under the program, and INS had to scramble to redirect resources. As a result, the backlog of unresolved cases ballooned to around 6 million in 2003, and wait times at some immigration offices were two years or more--even for relatively straightforward matters.

Unrealistic mandates like these are largely to blame for the backlog of immigration applications, which is now "only" 4.5 million. Since it can't work through these anytime soon, the immigration service often issues work permits and travel documents to many green-card applicants right when they submit their forms, knowing that it will be years before anyone actually reads the application. This is the present situation at the agency that is supposed to carry out extensive, complicated new responsibilities under the proposed immigration bills.

So, if an immigration package anything like McCain-Kennedy or Cornyn-Kyl were to pass, the following would almost assuredly occur: Immigration offices would be deluged by millions of applications that would need to be approved under a tight deadline; harried DHS employees would be forced to put aside their other duties to meet the onslaught; candidates for citizenship--foreign spouses of Americans, refugees, skilled workers sponsored by employers--would effectively be pushed to the back of the line; political pressure would force DHS to cut corners in adjudicating the applications; and huge numbers of ineligible applicants would be approved (in addition to the huge numbers of eligible applicants).

The workload created by any such program would be larger than the annual number of visas currently issued worldwide by the State Department (approximately 5 million), and many times larger than the annual number of green cards issued by DHS (around 1 million). Delays, mistakes, and fraud may not be what the supporters of the immigration bills have in mind, but they are nonetheless the guaranteed outcome. None of this is to pin blame on the bureaucrats charged with implementing congressional mandates. Rather, the proposals themselves are the problem, measures animated by an almost utopian spirit, one that seeks to solve an enormous and long-brewing problem with one swift masterstroke.

Things don't work that way in the real world. Instead, the illegal population needs to be decreased via muscular, across-the-board immigration enforcement over a long term. Rather than wait for a magic solution, we can implement an attrition strategy right now, using available resources. We could, for instance, immediately reject fake Social Security numbers submitted by employers on behalf of new employees (the government currently looks the other way). Or the Treasury Department could instruct banks that the Mexican government's illegal-alien ID card is no longer a valid form of identification. Or a small portion of enforcement resources could be devoted to random raids at day-labor gathering spots. This has an added advantage: As more resources become available--be they monetary or technological--they could easily bolster the attrition approach, as opposed to current proposals, which from the get-go require vast and untested programs.

An attrition strategy would adopt the conservative goal of shrinking the illegal-alien population over time by making it unappealing to be an illegal alien in the first place. As a result, fewer people would come here illegally, and those already here would be more inclined to deport themselves. Over the space of several years, what is now a crisis could be reduced to a manageable nuisance.

Now that's realistic.

Mr. Krikorian is executive director of the Center for Immigration Studies.