Robert Rector and Jason Richwine have a new report out for the Heritage Foundation on 'The Fiscal Cost of Unlawful Immigrants and Amnesty to the U.S. Taxpayer," which is introducing pro-immigration right-of-center groups (Cato Institute, American Action Network, Americans for Tax Reform) to the cesspool that is Heritage economic analysis.
The study starts by simply ignoring large swathes of the bill. There's no W Visa program here. No replacement of the Diversity Visa Lottery with a new points-based program. No expansion of H1-B, no reform of the treatment of spouses of skilled green card holders. There's nothing but amnesty for undocumented workers presently residing here. They tally up the taxes likely to be paid by the typical undocumented worker (low because he's poor) and compare them to the cost of public services associated with each person. This latter is high because Heritage mixes and matches its methodology. When it comes to means-tested benefits, they do an individualized analysis at the margin so one extra low-income person costs however much it would cost to sign up an extra person for SNAP. But when it comes to general public services, they do a population average method. So if per person policing costs are $X, the marginal immigrant is associated with an extra $X in spending. But even though they refer to "police, fire, highways, parks, and similar services" as "population based services," this is clearly not how things work. Running a full bus is not twice as expensive as running a half-empty bus. A city that loses 5 percent of its population cannot lay off 5 percent of its cops and leave the remainder of the population equally well-protected. Most of these services have fixed costs alongside marginal ones, experience some economies of scale, and are often more efficient to provide in denser areas.
What's more, Heritage pretends that you could make every single unauthorized immigrant vanish with no relevant macroeconomic impact. Heritage's conservative detractors are trying to link this to the "dynamic scoring" debate about tax policy, but it's really pretty different. The relevant issue in fiscal terms is that the employer of the immigrant is also paying taxes, and most likely at a substantially higher marginal rate. Last, but by no means least, the assumption here is that granting legal status to unauthorized workers will have no economic value to the workers. If that were true, it would be hard to understand why we'd even be having this debate. A big part of the reason that people want amnesty is that living illegally in the United States is a huge pain. There are lots of jobs you can't get. You can't get a bank loan to start or grow a business. You often can't get a driver's license. You may not be able to go to college. All that stuff depresses incomes and educational attainment. Personally, I would not put a ton of stock in point estimates of the economic value of legal status, but the value can't be $0.
But it illustrates an underappreciated point in Washington, namely that even ideological think tanks do their movements a disservice when they do bad work. As Republican members of Congress ponder what to do about immigration, having accurate information about its fiscal impact would be very useful to them. You actually want to have a team of people "on your side" who you can trust to do good work. Heritage is not that team.