A while back, Daniel Costa of the progressive Economic Policy Institute, a think tank that advocates for the interests of poor Americans, did an interesting blog post titled On International Migrants Day, remember that guest worker programs aren’t the solution for immigration reform. This met with a lot of pushback on Twitter, and Costa followed up with a related blog post describing what he considered to be extensive abuses in guest worker programs in the United States (the H2 visa program). Among the stories that Costa linked to were Filipino teachers being conned by a recruiter, workers reporting exploitation by a seafood company in Louisiana that supplied to Walmart, and a negotiated settlement about people under J-1 visas being exploited at work. Based on these and other incidents, Costa is understandably very skeptical of proposals that go by the name of guest worker programs.
I will not commit the moral and strategic error of shrugging off the problems of current H2 visa programs in the US with “not as bad as” trivialization. I think that problems and abuses at guest worker programs, while not the worst thing in the world, are definitely worth putting in the balance when proposing the expansion of guest worker programs. However, I think that Costa’s prescriptions don’t necessarily follow from his observations.
Guest worker programs: tied to an employer?
Most open borders advocates view the keyhole solution of guest worker programs as a half-way compromise, not a desirable ideal (in the jargon of this blog post, they tend to have a (1) > (2) > (3) preference ordering: open borders preferable to expanded guest worker programs preferable to the status quo).
As the guest worker programs page on this site describes, there are many different parameters whose values can be fiddled and adjusted while still staying within the broad category of guest worker programs: the time duration of the program, flexibility in terms of jobs, extent of legal rights, eligibility for citizenship, deportation conditions, and eligibility for welfare benefits being the parameters listed on the page. Of these, the first three (time duration, flexibility in terms of jobs, and extent of legal rights) are the most relevant for considering the problem of worker abuse. The kinds of guest worker program solutions that open borders advocates typically propose are those with essentially unlimited (or periodically renewable) time duration, the ability to switch jobs at will (i.e., not tied to any particular employer), and full legal rights (however, some proponents of these programs oppose some labor regulations per se, like the minimum wage, for natives as well as foreigners). Further, as a general rule, people coming at guest worker programs from an open borders angle oppose quantity caps on the amount of guest worker labor that can be used.
I think that these key elements will lead to abuses of the kind that Costa sees in current guest worker programs becoming more rare. With the status quo in the US, guest worker programs are heavily time-limited and tied to specific jobs. There is also a pretty severe quantity restriction on these programs. This makes it extremely hard for workers to “shop” between employers, both at the time of applying for a visa, and once they are in the US. Their main element of discretion is in whether they choose to come the next year. Even in the status quo, reputational effects and the need for good worker morale check some worker abuses. But with fewer quantity caps, fewer time limits, and the ability to switch between jobs, worker abuses are likely to be lower as workers can “shop” better.
A quick analogy might help. Suppose a particular factory is the main employer in town, and it pays its workers very low wages and has demanding working conditions. Now, a competing factory wants to open up in the same town. Should the residents of that town fear the new factory, based on the rule that factories exploit their workers? Or should they welcome the new factory, in that the competition between the two factories may improve conditions for workers? While the details vary from case to case, I would suspect the latter.
Now, admittedly, the cases aren’t quite parallel, because in the analogy I gave, the population of the town was not changing. But if the creation of a new town just attracts more labor from outside the town, then the effect on wages in the original factory may be smaller (though probably still positive). Even here, though, unless you discount completely the welfare of people who move to the town, the net effect on wages is still expected to be positive.
If you think these difficult jobs are bad for Mexicans, think about this: 85% of the NCGA’s Mexican seasonal employees last year were repeat employees. They came the previous season, and they chose to come back the following season. It is inappropriate and unfortunate that some labor advocates call H-2 visa jobs “close to slavery.”Slaves had no such choice, and would not have happily gone back to the plantation that owned them. Furthermore, the H-2 visa holders who work for the NCGA are not tied to a single farm: their visa allows them to work throughout the 700-farm network, so that there are opportunities to move if any given farm violates labor standards. Any shortcomings of the H-2 program are not the fault of migration itself; they can be fixed by fixing the program.
I don’t have independent corroboration of these statements, but it does seem to make sense that workers whose visa allows them to switch employment between a lot of farms in a huge network would be less susceptible to the problems of worker abuse. Repeat seasonal migration also creates incentives for employers to treat workers fairly and honestly.
Note: I read through the Red Card solution website, which offers a detailed guest worker program proposal for the US, and I was disappointed to see that the proposal did not address the issue of how this proposal would accommodate the possibility of workers changing employers — would they need to return to their home country to re-apply, or could they change their authorization while still in the US? On the other hand, this proposal is overall a “no-questions-asked” guest worker program proposal, so even with a need to return home to re-apply, that would still be a fixed overhead cost as opposed to an uncertainty of whether or not one could switch employers. With respect to duration (essentially unlimited) and legal rights, the proposal is pretty good.
What if the “tied to one employer” condition is non-negotiable? Should open borders advocates still favor expansion of guest worker programs?
Suppose the idea of guest worker programs with complete freedom of movement between employers is politically infeasible, so that the only guest worker programs on the table are those tied to a specific employer. Even in this case, if employer networks are allowed to sponsor visas in lieu of individual employers (as described above), this could provide a more flexible de facto employment market that might reduce the potential for worker abuse and improve options for both workers and employers in matching each other. But even setting aside the possibility of employer networks, I think that the benefits of guest worker programs to workers who choose to use their programs through “revealed preference” are likely to be greater than the costs to those specific workers.
That said, although I express guarded support for the expansion of inflexible, time-limited, and employer-tied guest worker programs as better than closed borders, these are a very minuscule step towards open borders — so minuscule that open borders advocates would be misleading themselves in thinking of the expansion of such programs as a significant step towards open borders.
The managed migration solution: decide number of visas by area, don’t tie visas to employers
Costa’s solution, which he says is also preferred by many progressives and by labor unions, focuses on “managed migration.” Here is how he puts it:
The best way to determine the future flow of immigrant workers is through the creation of a permanent, independent commission, like the one proposed by EPI or the Migration Policy Institute, an approach endorsed by the Council on Foreign Relations’ bipartisan-led Independent Task Force on U.S. Immigration Policy, and the nonpartisan Brookings-Duke Immigration Policy Roundtable. A commission on foreign workers would figure out ways to improve immigration data collection and analysis, provide lawmakers with advice on how make the system more rational, transparent and responsive to changing labor market conditions, and promote broadly shared prosperity by working to ensure that new immigrant workers complement U.S. workers. This would increase productivity and wages, rather than displace U.S. workers by flooding the labor market in particular occupations in order to put downward pressure on wages and working conditions.
Setting aside the citizenistic connotations of the criteria Costa proposes for the time being, a proposal that ties visas to worker skills rather than specific employers is good in the sense of allowing for a more flexible labor market with workers and employers better able to match each other. This could also reduce the potential for worker abuse. However, the idea that a commission could figure out the number of immigrants needed seems to me to be hubristic central planning. I do agree that in many cases, commissions can figure out broad trends — as even uninformed observers can. It’s easy to see, for instance, that Silicon Valley needs more immigrants. But there’s a difference between figuring out overall general directions and fixing precise quantities. Letting employers sponsor workers allows for local knowledge of how much labor is actually needed to play a bigger role.
It may be the case that the status quo probably combines the worst features of both systems: on the one hand, workers are tied to employers. On the other hand, the government does restrict and regulate the number of visas severely and arbitrarily based on not-quite-transparent heuristics of whether the jobs are really needed.
Should the same minimum wage be applicable to migrant workers?
There is considerable debate surrounding what the appropriate value of the minimum wage should be. By and large, open borders advocates do not support double standards on minimum wages. While some of them want to get rid of minimum wage laws, they do not usually argue that migrants should be exempted from these laws. In his video on immigration restrictions, Bryan Caplan suggested that exempting immigrants from the minimum wage might be an acceptable keyhole solution if getting rid of the minimum wage entirely is not acceptable. Co-blogger David Henderson pointed out the flaw in this:
Bryan’s other error is to suggest that enforcing a minimum for native-born people but not for immigrants would help the native-born. No way. Would United Airlines want a pricing restriction that doesn’t let it cut fares but lets Southwest do so?
Caplan responded in a subsequent blog post:
If I said this, I misspoke. But it’s more complex than David lets on. If (a) labor is homogeneous, (b) natives are subject to a minimum wage, and (c) immigrants aren’t, then employers will only hire immigrants, and natives clearly lose. But in the real world, as I emphasized in my talk, native and immigrant labor are heterogeneous, so the welfare effects of selective minimum wages for workers are not clear-cut. Price floors that apply to wheat farmers but not corn farmers can make wheat farmers better off. Still, on balance I think David’s right.
Path to citizenship: neither here nor there
Costa also suggests that any immigration reform must be combined with a clear-cut “path to citizenship” not just for those already in the United States but also for any future migrants. Since guest worker programs typically do not come with a path to citizenship (which is what distinguishes them from immigration tariffs and proposals like my co-blogger Nathan Smith’s DRITI) this is another deal-breaker. For those who are concerned about a path to citizenship per se, this is indeed a deal-breaker. However, I don’t think that the specific problem of worker abuse is directly related to a path to citizenship. I suppose it could be argued that if politicians knew that immigrants would be future voters, they might crack down more harshly on abuses of immigrants. While this is plausible, I think that long time lag between migration and achievement of citizenship, and the lack of direct culpability of politicians in any worker abuse by a specific employer (who has no direct public connection with a specific politician) makes this argument relatively weak.