Charles Kenny of the Center for Global Development, home of Michael Clemens, one of the leading intellectual advocates of liberalizing global labor migration, has written a new Bloomberg Businessweek column on why the U.S. immigration system should move from quotas to tariffs:
Back in the 1960s and ’70s, one of the early and hugely important steps taken by GATT—the predecessor to the World Trade Organization—was to move the world from a reliance on trade quotas to a system based on tariffs. As a rule, quotas are less efficient than tariffs. If the market changes, and there’s more demand for an import, quotas mean that none of that extra demand is met. And someone has to chose who gets the imports that are let in. That process is, at best, bureaucratic, and at worst, rife with corruption and favoritism.
Tariffs, on the other hand, are applied equally to all imports, are flexible in the face of changing demand, and don’t favor the connected few. And governments get the benefit of tariff revenue to pay down the national debt or fund anything from flood relief to research on gun crime.
The same logic that applies to importing goods applies to importing people. The current immigration system involves a bunch of different quotas: for skilled workers from particular countries, for family members, for lottery winners—the list goes on. The process of deciding who gets in under the quota is expensive and chronically inefficient. And the government doesn’t even make any money from it.
And so Kenny proposes a $50,000 tariff:
So, how much could—or should—we charge for the right to live and work in the U.S.? Becker suggested the U.S. should let in anyone who can pay $50,000 to Uncle Sam and pass a criminal background check. That may seem like a lot of money, but Miao Chi and Scott Drewianka of the University of Wisconsin estimate (PDF) that, allowing for factors including age and education, the average recent Mexican immigrant with a green card (permanent resident status) earns roughly $20,000 a year more than the average Mexican immigrant without one (on a more limited visa or undocumented). So, allowing for education, the average immigrant from south of the border would recoup that $50,000 in less than three years.
Money generated from immigrant tariffs could be used to support low-income native workers through initiatives like the Earned Income Tax Credit—reducing political opposition to migration among those who see themselves most at risk. A $50,000 tariff applied to 1 million migrants (about one-third of 1 percent of the U.S. population) would be enough to almost double the size of the EITC program.
The problem with Kenny’s proposal, in my view, is that if we are going to set a tariff, $50,000 is almost certainly not the “correct” price. Kenny’s concern is that the price might be too high, yet the findings of Miao Chi and Scott Drewianka suggest otherwise. Moreover, his thought experiment stipulates that a $50,000 tariff would lead to an influx of 1 million, but of course we don’t know what the market-clearing price would be. In the first year of the new system, in light of pent-up demand, the $50,000 tariff might lead to far more than 1 million immigrants, which in turn might lead to a backlash against immigration tariffs.
This is why Derek Khanna’s proposal for H-1B visa auctions is closer to the mark:
In the current H-1B visa system, the price of the visas does not rise or fall in response to market conditions. The entirely predictable consequence is that companies scramble for visas and often fail to get enough of them. A company whose need for the visas is strong receives the same treatment as do companies whose need for them is weak. It’s a failed market model.
We should greatly increase the number of H-1B visas and put them up for competitive bidding. There are many models for this. One is the spectrum auction, whereby telecommunications companies bid on government licenses to transmit signals over specific bands of the electromagnetic spectrum.
Competitive bidding would enable the companies that value visas the most to pay for them. And it would help small and medium-sized businesses, which often have the most difficulty filing their paperwork on time and often lose out to big businesses in the competition for the few available visas. Competitive bidding for visas would mean a rational market governed by supply and demand.
Khanna’s proposal is an incremental reform that would apply only to H-1B visas, presumably on the grounds that Congress would find extended the logic of competitive bidding to all immigration would be distasteful. It is worth observing, however, that if we moved from allowing extended family reunification, etc., to competitive bidding for virtually all visas, including spousal visas, U.S.-based families could still bring relatives to live with them. One difference, however, is that families would be able to choose which relatives, and indeed which friends, matter most. Most proposed reforms of family reunification call for restricting eligibility to spouses and minor children, but of course it is easy to imagine a U.S. citizen who really wants to bring over her second cousin. Competitive bidding makes room for her to do just that, provided she is willing to pay for the privilege.
Family reunification also requires sponsorship, in which the sponsoring family demonstrates that it has the resources to support the newly arriving family member at 125 percent of the poverty line. My sense is that this requirement is only loosely enforced. But competitive bidding would mean that families would have to think seriously about whether they could afford to take on the potential burden of supporting a family member. Generally speaking, competitive bidding would tend to skew the immigrant influx towards those most able and willing to pay, which will greatly reduce the likelihood that these new arrivals will represent a net fiscal burden.