Does It Pay to Enforce the Law?

By Steven A. Camarota and Steven A. Camarota on September 22, 2010

Recently a blog posting at the American Enterprise Institute web site included an excerpt from a new AEI book by Gordon Hanson. The book is entitled, Regulating Low-Skilled Immigration in the United States. The excerpted passage attempts to make the case that making illegal immigrants return home is probably not good idea. The passage itself uses percentages of GDP to make it case. However, in the discussion below I have put the actual numbers into the calculation, to make it clearer to most readers. Here is Hanson's argument:

First he assumes a U.S. GDP of $15 trillion, which is correct. He also argues that illegal immigrants create a "surplus" of $4.5 billion a year. This figure represents the net gain that comes from having illegal immigrants in the labor market. Wages are lowered for less-educated natives and legal immigrants who compete with illegals. This in turn creates benefits for owners of capital, consumers, or more skilled workers. It can be shown economically that the wage losses for low-wage workers are smaller than the gain for everyone else, and he estimates the net gain to natives is $4.5 billion dollars. Put a different way, the $4.5 billion is the difference between what native-born low-wage workers lose and what all other natives gain from having illegal immigrants in the labor market.

It should be noted that the nation's GDP overall is something like $100-$150 billion larger because of the labor illegal aliens provide. But almost all of this goes to the illegals themselves in the form of wages and benefits – as it should, since they are the ones doing the work. Again this is just a standard economic model. The $4.5 billion is the net benefit or surplus that goes to natives and legal immigrants. Hanson does not mention the wage losses suffered by the poorest American workers in order to generate this gain. But it must be some loss for a surplus to exist. None of this is particularly controversial among economists.

Perhaps most important, Hanson's estimated surplus is trivial relative to the overall economy (GDP), as are almost all other serious estimates of the surplus. It is just 0.03 percent of GDP – three one-hundreds of one percent, or expressed as a decimal, .0003 of the overall economy. Please note, these are his estimates of the benefit from illegal immigration, not mine. Certainly you could argue that illegal immigrants should return home, secure in the knowledge it would do no significant, or even measurable, damage to the economy.

In the next part of his analysis he doubles the enforcement budget and assumes that half of the illegal population goes home, to assess the utility of enforcement. Actually he is skeptical they would leave, but he assumes it anyway for his thought experiment. He argues that we spent $15 billion on immigration enforcement annually (ICE + CBP). This is not correct. The $15 billion includes substantial sums spent not on immigration but on customs enforcement, since both enforcement agencies have dual responsibilities. Next he assumes that if half of the illegals leave, half the surplus they create would leave with them. So natives and legal immigrants would lose $2.25 billion (half of the estimated $4.5 billion surplus). To this $2.25 billion he adds the $15 billion in extra enforcement expenditures (the result of his assumed doubling of such spending). Now he also assumes that illegal immigrants create $15 billion in net fiscal costs, so with the departure of half of the illegal population, $7.5 billion of the fiscal costs they create leave with them. Add this all together – fiscal savings from ½ of illegals leaving minus extra enforcement spending minus half of immigrant surplus – and you get the following math: 7.5 - 15 - 2.25 equals a net loss of $9.75 billion. Thus, making illegals going home seems like a bad idea, or so Hanson suggests. Of course, relative to the size of the economy the supposed loss is trivial.

The first thing to observe about Hanson’s analysis is that it is entirely assumption-driven. Make different assumptions, and you get an entirely different answer. The immigration surplus he uses is reasonable. However, his enforcement discussion is not reasonable in my view. Again, he does not seem to understand that we do not spend $15 billion on immigration enforcement, so the extra $15 billion in spending he assumes from a doubling of the budget is incorrect. The Center for Immigration Studies has estimated the total additional five-year cost (on top of the existing level of spending) for an effective attrition strategy would be $2 billion. This is very different from the extra $15 billion Hanson assumes for just one year.

His discussion of the net fiscal cost of illegal immigrants (taxes paid minus services used) is also not reasonable. In a personal correspondence he states that he got the $15 billion net costs estimate from a study I did using 2002 data, which showed net costs of $10 billion annually at the federal level only. He then assumes that the costs have grown to $15 billion, with the increase in the illegal population. But my study was very clear that the cost estimate was only at the federal level, and net costs at the state and local levels are much larger. A recent study by FAIR estimated $113 billion in net cost annually at all levels of government. Robert Rector at the Heritage Foundation estimates that the average household headed by an unskilled immigrant created a net drain of about $20,000 a year. Of course, Rector did not focus just on illegals, and only about half of illegals lack a high school education. But given the education level of illegals immigrants, his work implies a fiscal drain of over $50 billion a year.

So let us follow Hanson's formulation but with more reasonable assumptions. To be conservative, let's assume a net fiscal drain of only $50 billion a year, less than half the FAIR estimate. So if half of illegals go home, the savings would be $25 billion. Let's also assume twice the increase in enforcement costs CIS estimated (from $2 billion to $4 bilion) and let's further assume that all of the enforcement costs are concentrated in just one year rather than spread out over five years. Now the numbers look like this: 25 - 2.25 - 4, which produces a net gain for natives and legal immigrants of $18.75 billion, assuming half the illegals leave. In my view this is a much more reasonable way of looking at the issue. Even with $15 billion in additional enforcement costs there would still be a net gain.

Perhaps best of all, the wage losses that unskilled natives and legal immigrants suffer because illegal immigration increased the supply of workers is partly mitigated by the departure half of this population. This means the poorest American workers are made somewhat better off.

One final point worth making: Hanson seems very skeptical of enforcement efforts. He does not seem to understand the effects of having every president for more than a decade, especially Bush and Obama, repeatedly making strong public announcements in favor of an amnesty/legalization. Moreover, legislation designed to legalize illegal immigrants is continually introduced. All of this is extensively covered in Spanish-language media. Illegals would be a foolish to leave if legalization is a real possibility.

In contrast, if the president would say, "There will be no legalization – if you are here illegally you must go home," it would cost nothing and do more to make enforcement effective than any single thing I can think of. Also it should be pointed out that although we have spend a lot at the border, we still do not have a fully functioning entry/exit system to keep track of visa overstays, we have no mandatory employment verification system, the Social Security Administration and the IRS continue to knowingly accept false Social Security numbers, we have 500,000-plus deportation absconders (people ordered deported who we were never follow up on), and the Treasury Department has implicitly told banks that illegals are allowed to open accounts. The list of things not done goes on and on.

Arguing that we have spent a lot on border enforcement and it has not worked is like arguing that the expensive new lock on the front door of your house is a failure, when at the same time you leave your back door and windows wide open.