The public debate over immigration reform, which all too often devolves into emotional rhetoric, could use a healthy dose of economic realism. As Congress and the White House fulfill their recent pledges to craft immigration-reform legislation in the months ahead, they must ask themselves a fundamental question: can we afford any longer to pursue a deportation-only policy that ignores economic reality?

At a time when the budgets of federal, state, and local governments contain more red ink than revenue, in the midst of the worst recession since the Great Depression, what can we realistically afford to do with the roughly 12 million unauthorized-immigrant men, women, and children whom the Pew Hispanic Center estimates now live in the United States—plus the four million U.S.-born, U.S.-citizen children who have an unauthorized-immigrant parent? Even more to the point in the present economic climate, how can we best tap these millions of unauthorized workers, consumers, and—yes—taxpayers as a force for economic recovery?

There are three possible strategies for moving beyond the status quo: the fanciful “deport them all” approach still advocated by fringe anti-immigrant groups; the nebulous “attrition through enforcement” approach now advocated by mainstream anti-immigrant groups; and the creation of a program under which unauthorized immigrants could apply for legal status—an approach advocated by a wide array of groups on pragmatic, economic, and humanitarian grounds. In contrast to the first two options, the legalization approach represents an acknowledgment that enforcement measures alone cannot fix a problem that was caused in large part by a decades-long mismatch between legal limits on immigration and the actual labor demands of the U.S. economy. Just as importantly, though, legalization also acknowledges the fact that, since the U.S. economy is now in recession, incorporating currently unauthorized immigrants into our strategy for economic recovery makes far more fiscal sense than spending untold billions of dollars, in the middle of multiple budget crises, in a quixotic quest to force them all out of the country.

The “deport them all” scenario is, obviously, the most unrealistic. Leaving aside the daunting logistical, legal, and civil-rights issues involved in raiding homes, schools, and worksites around the country in search of unauthorized immigrants, what would a mass-deportation campaign cost? Julie L. Myers, head of Immigration and Customs Enforcement (ICE) during the Bush administration, told Senators during her confirmation hearing in 2007 that it would cost at least $94 billion; a figure which ICE subsequently noted did not include the cost of actually finding unauthorized immigrants, nor the court costs associated with deporting them. The Center for American Progress (CAP) released a more comprehensive estimate in 2005 that put the price of mass deportations at somewhere in the range of $206 billion to $230 billion over five years.

Suffice it to say that deporting 12 million-plus people would not be cheap. And neither the ICE nor the CAP estimates even begin to account for the economic impact on numerous businesses of suddenly losing the workers who make their products or the consumers who buy them. A 2008 report from The Perryman Group estimated that, were all unauthorized workers and consumers somehow removed from the country, the United States would lose $551.6 billion in annual spending, $245 billion in annual economic output, and more than 2.8 million jobs. Moreover, federal and state treasuries would lose the revenue they now receive from unauthorized taxpayers. (Contrary to popular opinion, between half and three-quarters of unauthorized immigrants pay federal and state taxes, according to the 2005 Economic Report of the President.)

As opposed to the brute force of mass deportations, the goal of “attrition through enforcement” is to make life in the United States so difficult for unauthorized immigrants that they choose to leave, or “self-deport.” This approach involves several types of immigration enforcement (including the construction of an expensive and ineffective U.S.-Mexico border fence, which Congress still seems intent upon pursuing). But the centerpiece of the effort would be the mandatory, nationwide implementation of the federal government’s now-voluntary “E-Verify” system, through which employers check new hires against the databases of the Social Security Administration (SSA) and Department of Homeland Security in an attempt to ensure that they are authorized to work in the United States.

In addition to the question of how much it might cost to make E-Verify mandatory for all employers, there is also the question of whether or not it stands a good chance of actually working. In other words, how likely is it that E-Verify would make it so difficult for unauthorized immigrants to find a job that they would leave the country? Reports by the federal government’s own researchers do not bode well with regard to either the effectiveness or cost of a nationwide E-Verify program.

The Government Accountability Office (GAO) reported to Congress in 2005 that E-Verify cannot detect identity fraud in which an unauthorized worker presents an employer with either valid identity documents belonging to another person, or reasonably well-made counterfeit documents containing valid information about another person. The SSA Inspector General reported to Congress in 2006 that the Social Security records of about 12.7 million native-born U.S. citizens probably contain errors that would “result in incorrect feedback” to employers as to their identity or authorization to work. And CBO estimated that the mandatory E-Verify system called for in the SAVE Act of 2007 would have cost at least $12 billion over 10 years to implement, and probably would have also decreased federal revenue by $17.3 billion over the same period as more workers were paid under the table, outside of the tax system.

In other words, implementing E-Verify nationwide would cost tens of billions of dollars, would not detect identity fraud, would incorrectly flag millions of U.S. citizens as not being who they say they are, and would result in less tax revenue being collected from unauthorized workers than is now the case. None of these outcomes seems particularly desirable at a time of high unemployment and gaping budget deficits. Nor does this seem to be a promising means of persuading unauthorized immigrants to self-deport.

The third option—creation of a legalization program—would require unauthorized immigrants to pass criminal background checks and pay fines, fees, and any back taxes they might owe. The relative cost-effectiveness of this option is apparent in a CBO analysis of the Comprehensive Immigration Reform Act of 2006—which preceded the onset of the current recession and therefore included increases in legal limits on future immigration as well as a legalization program for unauthorized immigrants already in the country. CBO estimated that the bill would have generated $66 billion in new revenue over 10 years, primarily from income and payroll taxes paid by both new and newly-legalized immigrants. This revenue would have more than offset the anticipated $54 billion increase in spending for refundable tax credits, Medicaid, Medicare, Social Security, and food stamps for newly eligible immigrants and their families during the same period.

Given the current condition of the U.S. economy and the federal treasury, what makes more sense? Spending tens of billions of dollars trying to force, or forcefully persuade, millions of unauthorized dishwashers, nannies, and gardeners to uproot millions of U.S.-citizen children and leave the country? Or create a legalization program which weeds out the very few unauthorized immigrants who might be dangerous criminals, and then ensures that the rest are paying into the tax system? Does it make more sense to take unauthorized consumers and taxpayers out of the economy, or—by allowing them to earn legal status—enable them to earn higher wages and therefore spend more and pay more in taxes?