June 13, 2019 12:32PM

Market‐​Based Visas: Problems, Criticisms, and Solutions

Steven Kopits of Princeton Energy Advisors wrote a few criticisms of our proposal to sell Gold Cards through a market-based program that I’ve called an immigration tariff.  An immigration tariff is an attempt, based largely on Nobel Prize-winning economist Gary Becker’s idea to sell visas, to create a market-based visa that accounts for many of the most trenchant criticisms of liberalized immigration  The idea is simply to create a new visa called a Gold Card without numerical quotas or caps.  The Gold Card would supply permanent legal residency and work permission, but cannot be used to naturalize –like a green card lite.  The government would then sell those Gold Cards for a price set with three goals in mind.  Those goals are to make sure that the Gold Card is a net-fiscal windfall for the federal government, to outcompete human smugglers, and to create a more flexible immigration system that responds to the U.S. market demands. 

The prices that I’ve recommended for the Gold Card are designed to more than offset the worst-case scenarios as outlined by the National Academy of Sciences fiscal cost projections by age of entry and education level for individual immigrants.  This is an earnest and direct answer to a forceful conservative criticism of immigration based on its supposedly negative fiscal effect.  Although I dispute the notion that immigrants are a net fiscal drain, that criticism is still politically popular and must be addressed in any final policy.  

In most cases the prices for Gold Cards would compete favorably with human smugglers and drive most illegal immigrants and many asylum-seekers into the Gold Card market.  A legal Gold Card visa guarantees legal work and residency, so long as the migrant doesn’t commit a deportable offense, while smuggling is an expensive and risky chance at potentially working illegally in the United States with the ever-present risk of deportation and abuse in the black market.  Furthermore, the prices charged by human smugglers are not so much lower than Gold Cards in my mock tariff schedule.    Lastly, an immigration tariff is the most market-friendly visa that still offers some protections for the U.S. labor market, the last point being politically necessary.

Kopits also combines some criticisms of Cato’s immigration tariff with that of the IDEAL immigration plan that charges $2500 a year for a work permit.  The IDEAL plan has a lot of merit and it’s worth examining in detail, but I’m going to focus on the points that Kopits raises about Cato’s immigration tariff proposal here.  I take Kopits’ criticisms seriously because he is a proponent of creating a market-based visa system for the United States, he has written thoughtfully on this and other immigration issues, and he’s a smart and respectable individual.  Kopits’ criticisms are in quotes below and my responses follow.


The fee is too low and the market is too big

Both the Cato and Ideal fees are too low, particularly as they offer permanent residency in some form.  From the conservative perspective, the idea is to limit or reduce the number of immigrants and reduce tenure of migrants in the US.  A low fee will not only encourage immigration, it will crush the system.”

The Gold Card is a form of specially designed permanent residency that is not a pathway to citizenship.  The Gold Card does not preclude its holder from earning a green card through one of the existing legal channels and eventually naturalizing that way, but it does not provide a special path toward citizenship.  The non-immigrant feature of the Gold Card is intended to cater to conservative worries that Gold Card purchasers will all naturalize and vote against them, just like they fear new immigrants and those legalized in an amnesty for illegal immigrants will vote against them.  Anecdotally, this fear is an important reason why many Republicans and conservatives are skeptical of increasing legal immigration.  I think their fear has all of the hallmarks of a self-fulfilling prophecy, but it’s probably come true to some extent because of Republican positions on the issue of immigration, so the Gold Card is designed to deal with that.

“Consider Cato's $15,000 fee for young adults without high school educations.  How big is the potential market?  If we apply the offer to 133 low income countries globally -- to Latin America, East and South Asia and Africa -- then we are speaking of a gross population of 6.5 bn.  If just 0.1% decided to take up the offer . . . well, it would be game over in a matter of days.”


If 0.1 percent of 6.5 billion people paid $15,000 for a Gold Card in one year, the tariff fees by themselves would be $97.5 billion.  That’s a lot of money paid directly to the U.S. Treasury from Gold Card migrants on their first day of arrival in the United States – and it doesn’t even include the taxes that they’ll pay once they start working. 

Furthermore, 0.1 percent of 6.5 billion is 6.5 million Gold Card migrants, a number equal to about two percent of the current U.S. resident population.  In other words, this inflow would be equal to about the annual inflow of immigrants to New Zealand as a percentage of their population in 2016.  If the rest of the U.S. immigration system remains intact after a Gold Card is adopted, that would translate to an annual inflow equal to about 2.3 to 2.4 percent of the current U.S. resident population annually.  That’s a big increase in legal immigration, but close to the flow in some other OECD countries with very successful merit-based immigration systems.  The daily additional average flow of Gold Card migrants in such a scenario would be about 17,808 and they’d pay over $267 million directly into the U.S. Treasury each day – assuming they were all high school dropouts under the age of 25.  I doubt this would be game over for such a system, especially after a few days.

Furthermore, this doesn’t even consider the large numbers of lawful present on other less-flexible work visas like the H-1B or OPT who would pay a lower price for a Gold Card to skip the insanely long line for green cards that afflicts so many of them – especially Indians.  Green cards are preferable for these folks, but Gold Cards would be an improvement over the current situation they’re in and immediately raise revenue for the U.S. Treasury. 


“If the number of visas is capped, let's say at 1 million to choose a number well beyond the political tolerance of Republicans, then the 150 million citizens of Mexico and Central America who represent illegal immigration across the southwest border would gain about 1% of all offered visas, with the rest going to countries like China, India, Pakistan, Indonesia and Nigeria, among many others.  At the current pace, that would deter illegal immigrants from the southwest border for all of three days.  Put another way, Cato's proposed fee would either have little impact on illegal immigration or blow up the legal immigration system on contact.”

As I make clear in my policy analysis on the immigration tariff, I do not favor capping the number of Gold Cards and auctioning them.  One of the benefits of a tariff whereby quantity can adjust but the price is fixed, compared to an auction where quantity is fixed but prices adjust, is that the quantity of immigrants is more flexible and can respond to changes in the U.S. labor market.  This is the important benefit of a market-based visa.  If the unemployment rate is low and the economy is booming, more migrants would be willing to purchase the Gold Card, but fewer will do so when unemployment is higher.  After all, the big benefit of immigration is having more of them working here in the United States where they are so much more productive than in their home countries.  Under an auction, the price will fall along with the U.S. employment rate, but it would probably be filled each year and provide less revenue as a result – exactly when political arguments against immigration are more likely to be salient.  This seems like a more unstable political equilibrium.  If the number of Gold Cards is capped annually then an auction is the best way to distribute them, but there should not be a cap on the number and a price as the mismatch would produce surpluses or shortages.

Under a numerically uncapped Gold Card, human smuggling would decline radically.  According to the RAND Corporation, smuggling fees from Central America to the United States are typically between $6,000 and $10,000 but can range much higher.  Most of those people would fall into the $15,000 price bracket for a Gold Card and virtually all of them would choose to come legally by paying a slightly higher price for a Gold Card with the certainty of being able to legally work and live in the United States.  That certainty is worth a few thousand dollars more compared to the uncertainty of the smuggling route where fraud and coercion are rampant, contracts cannot be enforced regularly, the migrant still has only a chance of making it to the United States, and he may be detained for months, forced to wait in Mexico, and eventually deported without a refund – to say nothing of avoiding Mexico’s security services.  Combined with the fact that observationally identical illegal immigrant workers initially have wages about 11.3 percent lower than legal immigrant workers, the certainty, security, ease, and higher wages offered by a Gold Card are preferable to human smuggling for most people currently being smuggled to the border. 

The current headlines about the border are focused on the 676,315 people detained by Border Patrol or at ports of entry this fiscal year.  From the perspective of an immigration tariff, I think the headlines would be quite different if those 676,315 people paid $10.1 billion directly to the U.S. Treasury, an average of $15,000 each, this fiscal year.  This is another big advantage of selling visas: It changes the political narrative surrounding immigration.  The rhetoric of invasion on the right and charity on the left gets replaced with government revenue and a shrunken black market.  The disorder on the border shrinks dramatically overnight.  Other than a handful of diehard restrictionists, everybody else comes out ahead on this.

Congress is incapable of managing visa market conditions

Cato sees Congress setting visa volumes and tariffs.  There is zero chance that the US Congress would post tariffs for minimum wage migrant workers at a 35% effective tax rate.  Politically, that is not feasible.  If Congress is to post a tariff schedule, the rates will be much lower.  But, of course, this would flood the market with new entrants, prompting Republicans to put a cap on numbers at low levels -- the historical pattern -- which in turn would prevent the southwest border from closing.  An auction would not change this outcome per se.  If the number of visas auctioned is materially below the demand for labor, then the auctioned visas will be sold at some price, but employers will continue to take any labor deficit from the black market, just as they have since at least 1965.  Cato's price/volume mechanism is not viable, because Congress would have no more competence in managing visa market conditions than it would in setting, say, gasoline prices.” 


Just to be clear, my favored realistic market-based visa design is setting a price for the Gold Card and letting the quantity fluctuate.  But Congress should absolutely NOT set the price for Gold Cards and set the quantity.  That would lead to shortages or surpluses.  In my paper, I pitch the idea as a tariff or an auction but I don’t support combining the worst aspects of both into a nonfunctional visa where price and quantity are set by Congress.

Congress has a monopoly on the issuance of visas and the marginal cost of the government’s production of them is near zero.  It can issue as few or as many as it wants under any conditions or price that it sets.  The negative consequences of setting the price and letting quantity adjust based on changing market conditions are more economically beneficial than setting the quantity and letting price adjust – similar to how tariffs on trade goods are preferable to quotas even though free trade is preferable to both.   

I don’t see any political reason why Congress wouldn’t charge poor immigrants a one-time $15,000 tax.  Congress imposes much higher taxes on poor migrants that it blocks and, frankly, the welfare of migrants or potential migrants is rarely relevant in political debate.  My great fear with an immigration tariff is that Congress would charge much higher prices for each visa, not lower prices.  After all, Gary Becker proposed a fee of $50,000 back in 2011.  If the prices are too high then few will come and the big economic gains from the immigration tariff relative to the current system, which comes from increased legal immigration and not the extra tariff revenue, would not be realized. 

A price-based system is incompatible with permanent residency

For H2 class migrants -- unskilled workers typically lacking English language skills -- a price-based system is incompatible with permanent residency.  On an hourly basis, permanent residency -- even if promised fifteen years into the future -- is worth about $4 / work hour.  If we assume that migrants earn $10 / hour on average and the right to work in the US by itself is worth $3.50 / work hour, then a work visa with a residency option is worth $7.50 / hour, leaving $2.50 / hour in wages.  Of course, Mexicans cannot afford this, but a residency option will force them to make wrenching trade-offs between current consumption and future residency.  It will lead to penury for migrants, and by extension, to claims -- readily visible to the public -- that visa fees are exploitative and are impoverishing migrants.  In such an event, the system will fail, just as the Bracero system failed in 1965, and for similar reasons.  For this reason, residency cannot be included in a price-based program.  Work visas should not be conflated with formal immigration mechanisms.”

From the migrant’s perspective, the difference between real expected future income in the United States compared their home country is most relevant in determining whether to purchase a visa.  According to work by economists Michael Clemens, Claudio Montenegro, and Lant Pritchett, the lower-bound real annual wage gain adjusted for purchasing power parity (PPP) for a migrant with 9-12 years of education is $13,600.  For a Mexican migrant, the real annual gain is $10,523.  That wage increase beats the sheepskin effect of additional education.  It’s a very good deal for the Mexican migrant to pay $15,000 upfront for an immediate annual increase in real wages of $10,523 that will continue so long as he works in the United States.  They win big on this deal and, being the rational economic calculators that they are, will take advantage of it to a mutually beneficial extent. 

Furthermore, this seems like a kitchen sink argument to me.  The first criticism was that too many people will purchase the Gold Card while this criticism seems to be that too few will purchase a Gold Card because the price is so high and those who do purchase it will be reduced to penury.  Compared to the current immigration system, this is a big improvement for migrants – to say nothing of the extra economic benefits to the United States.

There are many different ways to design a market-based visa system for the United States, either as an addition to the current immigration system or as a replacement for it.  They range from IDEAL to auctions, immigration tariffs, and formulas.  Political feasibility is an important consideration, but the most important consideration is whether the proposal will achieve its goals and benefit the United States more than the current system does.  When those goals are met, then political feasibility should be considered in a very detailed way – but that needs to be tested against what the new system proposes and less on the current problems.  In this way, a market-based visa system that raises substantial government revenue through visa sales has an attractiveness to policymakers and the public that the current system does not. 

Economics is not the driving reason why Americans support or oppose immigration, but what if it was?  Consider, for a moment, the scenario that Kopits lays out where about 6.5 million people in the first year under a Gold Card system individually pay the $15,000 tariff for a combined revenue increase of $97.5 billion for the U.S. Treasury.  If that revenue was used to fund a tax cut or refund for the approximately 135 million native-born American workers, that would amount to about a $722 refund per American worker.  Let’s ask Americans this: “If your taxes were guaranteed to fall by $724 per year by allowing in more immigrants, would you support allowing in more immigrants?”  I suspect the answer would be quite different from a question that asked merely about increasing the numbers without any tax cut or refund.