David R. Henderson  

Posner and Weyl on Sponsoring Immigrants

Never reason from an interest ... Is Education Worth It? My Open...

University of Chicago professor Eric Posner and Microsoft employee Glen Weyl wrote last week about their intriguing proposal for immigration. It's titled "Sponsor An Immigrant Yourself," Politico, February 13, 2018.

I like parts of it but I don't totally understand it.

Here's their reasoning for why the proposal makes sense to them:

The problem posed by migration is that the benefits are not evenly distributed. They flow to the migrants themselves and the corporations that hire them. Consumers do receive better products and lower prices, but ordinary people don't really perceive these benefits. And working-class people may suffer a decline in their wages (some or many of them, depending on which economist you ask, but most agree the decline is not large), or (certainly, in most cases) believe that immigration undercuts their wages and threatens their cultural values.

So, immigration expands the economic pie but gives too meager a slice to ordinary people. The goal must be to retain, and in fact expand, immigration while ensuring that its benefits are distributed fairly. The current system does the opposite: channeling the benefits of migration to immigrants and domestic elites. Right now, special classes of citizens--mostly corporations (and in practice, big corporations) and family members--can sponsor temporary or permanent migrants, benefiting shareholders mainly, as well as ethnic enclaves.

Notice that they go from some pretty good reasoning in the first paragraph above to a conclusion in the first sentence of the next paragraph that's a non sequitur. If they are making the point that "ordinary people" don't perceive these benefits, a point they do make in the first paragraph above, they may be right. But they jump from this alleged lack of perception of benefits to the conclusion that ordinary people don't get many benefits. They may be right even there, but they do nothing to establish that claim.

Here's the gist of their proposal:

This system should be wiped away and replaced with a system of citizenship sponsorship for immigrants that we call a Visas Between Individuals Program. Under this new system, all citizens would have the right to sponsor a migrant for economic purposes.

Here's how the program would work: Imagine a woman named Mary Turner, who lives in Wheeling, West Virginia. She was recently laid off from a chicken-processing plant and makes ends meet by walking and taking care of her neighbors' pets. Mary could expand her little business by hiring some workers, but no one in the area would accept a wage she can afford. Mary goes online--to a new kind of international gig economy website, a Fiverr for immigrants--and applies to sponsor a migrant. She enters information about what she needs: someone with rudimentary English skills, no criminal record and an affection for animals. She offers a room in her basement, meals and $5 an hour. (Sponsors under this program would be exempt from paying minimum wage.) The website offers Mary some matches--people living in foreign countries who would like to spend some time in the United States and earn some money. After some back and forth, Mary interviews a woman named Sofia who lives in Paraguay.

Sofia, who grew up in a village, has endured hardships that few Americans can imagine. She is eager to earn some money so that she could move to her nation's capital city and get some vocational training. A few weeks later, Sofia arrives in Wheeling, after taking a one-week training course on American ways. If things don't work out, the agency that runs the website will find a new match for Sofia, and Mary will find someone new as well.

Wiping away a system in which corporations can hire immigrants is extreme. We would lose a lot of benefits from doing so: benefits that now are captured by corporations, their customers, the workers who complement these immigrants, and, of course, the immigrants themselves.

So I think it's better to (1) not wipe out the current system and (2) implement theirs too. It's good economically and it's good politically. Economically, for the reasons above, and politically so that corporations won't fight their proposal.

So how does this benefit the Mary Turners who hire immigrants? I had thought this was straightforward: gains from hiring cheap labor. But no, that's not the main way. Posner and Weyl write:

According to our calculations, a typical family of four could boost its income by $10,000 to 20,000 by hosting migrants. The reason is that migrants to the United States usually increase their wages many times, allowing them to pay as much as $6,000 to hosts for sponsorships (and our average family could sponsor up to four visas, one for each member).

Hold on. Where did this $6,000 come from? I agree with the authors that this could easily be a minimum estimate of the gains to the immigrants. But why would they voluntarily pay their hosts for sponsorships? Is their a price set by the government? Do the sponsors price it on their own? The authors have left out an important step that's more than a detail.

I could see immigrants coming here, figuring out they're in the wrong job, and then getting a better job. There's nothing wrong with that. The authors seem even to countenance that possibility. It would just be nice to see them spell out better the particular structure that they are proposing.

They also anticipate the following:

Others might try, in entrepreneurial fashion, to find foreign workers for American businesses--which would not be allowed to sponsor migrants under our proposal--taking a cut in the process. Google and Exxon would need to pay people like Mary to find migrants for their businesses.

Notice the completely unnecessary step they introduce because of their proposal to forbid corporations from hiring immigrants directly. They purposely advocate making the labor market less efficient.

Does this mean I oppose their proposal? Not necessarily. I would need to see the basis for their pricing of sponsorships and also how many people they would allow under their program. Allowing, say, 2 million people a year through this system rather than say the approximately 1 million people now would probably be an improvement. But if we get the same number of immigrants under their proposal that we get now, it's probably worse.

Comments and Sharing

COMMENTS (17 to date)
john hare writes:

I don't know where the $6,000.00 comes from, though there are a couple of instances I know of with immigrants spending money to stay here. One was a Columbian woman I dated for a while that mentioned that her sister had paid a man $10,000.00 to marry her so she could stay in the states. One of my employees from Mexico has spent about $20,000.00 on various legal fees to make his family permanent residents.

To me, costs of this type are a dead weight loss to the immigrants and the productivity of the country.

Miguel Madeira writes:

For what I understand from the cited passages, what they are proposing will be a kind of semi-serfdom, where immigrants will be a kind of "owned" by the sponsors, and be expelled from the country if they did not do what the sponsors demand (namely, paying a tribute).

Thaomas writes:

I agree with Henderson that anything that got us a significant increase in (mainly) high skilled immigrant would be an improvement, but I doubt the political economy of being able to buy the support of opponents by allowing them to sponsor and rent out the immigrant so sponsored would work.

In general I think the best way of "buying" the support of those who are or think they would be harmed by potentially pareto-improving policies - more immigration, freer trade, lower minimum wages, deregulation -- is a fast growing economy where the incomes of lower income people are growing in pace with the economy if not higher. What combination of infrastructure investments, investment in human capital, targeted compensation, and redistribution through taxes and expenditures would best do this is up for discussion.

Hazel Meade writes:

People can already sponsor immigrants via the employment sponsorship route, but that path requires Mary to prove that there is no American available who can do the job. For low skilled jobs, that is essentially impossible. Mary COULD hire an American, but she would have to pay more. The Dapartment of Labor would, consequently, deny the visa request.
This is why it is nearly impossible for unskilled or low-skilled immigrants to obtain visas to immigrate legally. It's not that nobody is willing to sponsor them. It's that the laws governing employment sponsorship effectively prevent any such visas from ever being approved.

Hazel Meade writes:

This is to some extent already done ... via "sham marriages". There are plenty of people out there willing to marry foreign nationals, for a fee. There's some trouble getting past the CIS to prove that the marriage is legitimate. A co-worker of mine actually moved to Spain because he couldn't convince the US government to give his fianace a visa - he couldn't convince them the relationship was legit. Obviously, laws would have to change to legalize paying US citizens a fee in exchange for visa sponsorships.

David R Henderson writes:

@Hazel Meade,
On your first comment above, I know that that's true. Were you thinking that I didn't know that?

Hazel Meade writes:

@David, I wasn't sure if you were aware from the text. These laws are pretty hard to change, and nobody is proposing repealing the labor certification requirement for employment sponsorship, as much as I wish they would.

Hazel Meade writes:

My point is ... there is nothing stopping people from doing this right now. There is no law preventing an individual from sponsoring another individual via the employment route. You could set up this website today. But it wouldn't matter, because none of those visas would get approved.

David R Henderson writes:

@Hazel Meade,
These laws are pretty hard to change, and nobody is proposing repealing the labor certification requirement for employment sponsorship, as much as I wish they would.
I agree that they're hard to change. I think that Posner and Weyl are advocating that they be changed. But their proposal is insufficiently spelled out.
Re the marriage option, that's not such a great way. I once was at a dinner party with a former INS official who had had as one of his tasks figuring out whether a marriage was legitimate. I asked him how he did it. One key indicator they looked for was whether the couple shared a joint bank account. Think about how few people would do that if the marriage weren't legitimate. So it's probably a good test.

Mark writes:

Why not just allow employers (any of them, including corporations) to offer employment (and the prospect of citizenship tied to, say, a few years of consistent employment) to foreigners? Then the migrants will pay payroll taxes, and there’s your compensation for ‘natives.’

Or just do as Gary Becker suggested and auction citizenship, thenredistribute the money to the unemployed?

Joseph E Munson writes:

I think its a great idea to make the gains from immigration real to the lower classes.

The proposal seems better than the status quo but people seem to hate seeing that kind of inequality, up and close even if it reduces total inequality. Current immigration policy seem to be basically homeless spikes at the moment.

With that in mind, here are a couple ideas to make economic gains from immigration more real to people and increase support for immigration.

1. slowly create free labor agreements with rich white countries, slowly expand to poorer browner countries. This way people can become surrounded by immigrants who are less different/scary and hopefully this will cause an increase in support for immigration.

2. add a Immigration rebate given to low income (or perhaps even middle class income) american citizens to "compensate" them for possible lost wages due to competition from immigrants. Make this rebate very obvious, fund it via a entry tax or something. This way people won't actually have to see the immigrants and wont be bothered by the fact that the immigrants are going to be poor by american standards.

Matthias Görgens writes:

A much simpler system to directly benefit the masses: auction off work visas to the highest (cash) bidders. Distribute the proceeds equally among all citizens. Done.

Adam writes:

Return to indentured servitude? Suppose Sophia has a child. Does the hypothesized "economic sponsorship" law bind the child to same servitude?

H1b visa is also coercive servitude: "If a foreign worker in H-1B status quits or is dismissed...the worker must...find another employer...or leave the United States" (Wikipedia).

Mark writes:

Adding to Mattias Görgens's idea, if an employer wants to hire a foreign employee, they could bid on their behalf for the prospective immigrant.

And American citizens may then want to increase immigration, because, assuming sufficient demand, expanding the defined supply of visas being auctioned off would increase the revenue being distributed to them. Coase theorem in action.

James writes:

Instead of sponsoring an immigrant why not just swap seats. You want one to come here, then you go there.

Matthias Goergens writes:

Mark, indeed. And I would also insist for people to pay upfront, but otherwise be treated for taxation etc as if they are locals.

That's the simplest system, and avoids any surplus to go to 'arrangers'. And it avoids the government having to chase after people with sob stories.

For people who are projected to be income rich but currently can't afford the visa, there are either loans or their employer can bid on their behalf. (For simplicity again, make the debt between employer and sponsored employee purely a private thing. Ie if an employer pays, and the immigrant switches jobs without paying off their bond, fine by the government, but just a breach of private contract.)

Matthias Goergens writes:

One more point: compassionate migration like for putting family's together can either keep its own pot of visas/green cards to hand out---even a flexible pot. Or can be structured to officially have to buy the visas at the prevailing price from the general pot.

The latter is just moving money from one government account to another one, and doesn't cost any real money. But presenting the figures like that might have interesting political consequences.

By the way, the idea of coupling some proposed change with payout of the proceeds to all citizens seems simple and generally applicable, and can take the wind out of many objections to economically efficient policies.

So why haven't we seen someone propose this setup for eg a carbon tax? (The land value tax people are somewhat interested in coupling their lvt with such a payout.)


Return to top