H1B visas provide employers every incentive to lie to import cheap talent and work them hard.
There are ways to meet the very real need to import the best brains from around the world. But H1B isn't set up to do this very effectively.
Off the top of my head, a better option would be to allow immigration with a $60k bond to the government that has to be paid down at $1k/month by the current employer. With the employee in question able to transfer jobs at any time with the bond transferring to the new employer.
Really, we should just enforce the original intent of the H1B.
1. Companies that import labor when they had a perfectly acceptable local candidate should face severe fines, and Justice should be given a real budget for running these investigations as a matter of course. And the burden of proof to demonstrate you needed to import labor should be pretty high. If you don't do this, then H1B abuse will happen.
2. Companies should be forced to pay H1Bs significantly above prevailing wage, where "prevailing wage" is interpreted liberally instead of conservatively. H1Bs for software engineers making less than $200K (maybe 150K) in a major metro area is a pretty obvious violation of the intent of the H1B, imo.
NB: this will have the effect of making H1Bs a lot harder to get and keep. But I think this is the least evil of all possible worlds. In addition, we could also independently argue for increased immigration in general. But midskill immigration to fill jobs for which local qualified candidates exist should not be done through the H1B program. That's not what it's designed for, and that mismatch between program design and actual use is the source of a lot of H1B mistreatment.
I actually don't believe that ending the tie to an employer will help. I mean, we should do it anyway, but the abuse will continue.
Companies that import labor when they had a perfectly acceptable local candidate should face severe fines...
Who defines acceptable local candidate?
Anyone who has interviewed candidates knows that are always a lot of local candidates who think that they are qualified for the job who you don't want for any of a variety of reasons. I don't want the government telling me that I can't hire the sysadmin I want because there is a Java programmer who thinks he can administer Unix systems. Because computers are computers, right?
Alternately I don't want to be told that I have to hire someone with a particular degree rather than the high school dropout who is self-taught and whose open source projects demonstrate ability to me. Government tends to be very impressed by certificates that industry insiders know well don't mean squat.
Companies should be forced to pay H1Bs significantly above prevailing wage, where "prevailing wage" is interpreted liberally instead of conservatively.
A modest proposal. If H1Bs are at the quota and someone else wants to hire an H1B worker and pay more than you're paying, you should get the choice between giving up your worker's spot (they can continue working for you remotely) or upping their pay. Without setting any particular dollar figure this would create salary competition among H1Bs that would take spots away from the worst of the sweatshops.
I don't think that would be much of a fix. But it would be an improvement.
I actually don't believe that ending the tie to an employer will help. I mean, we should do it anyway, but the abuse will continue.
I both believe that it would help, and that the abuse would continue. I see value in limiting how much abuse employers can get away with.
> Who defines acceptable local candidate?... sysadmin... java
I think your question already demonstrates the deviation of the H1B from its original intent.
Bringing in a generic sys admin or generic java programmers on an H1B is visa abuse. Sorry, but neither of those is a special skill and neither of those jobs is an appropriate use of the H1B. There are more than enough sys admin and java programmer candidates in the USA above a certain (quite low) price tag.
When it comes to H1B's intended purpose, I would expect something like: "we need someone who is expert in this aspect of the Java garbage collector and knows the internals of the X framework. Our candidate has made significant open source contributions to both. There is no one in this country with expertise in both who is currently looking for a position." Or, "we need someone who has extensive experience working in the insurance sector and also has experience being the senior developer on a large project written in Haskell. We can't find anyone who has passed the first actuarial exam and who has also been a senior engineer at a firm using Haskell". That sort of thing. The H1B was designed for __skilled immigration__. Not warm body java school kids.
I get your point that this is difficult to evaluate. That's why one of my proposals was to provide Justice with plenty of funding so that they have the resources to spot-check cases with high accuracy.
NB: I absolutely would support a new midskill immigration program for things like sys admins and generic java app developers. But the H1B program was never really meant for those sorts of jobs. The H1B program has radically diverged from its initial intent, and that is why the abuse of H1B workers happens. And, in fact, this is why even within H1B you don't see nearly as much of the abuse that happens in tech within other sectors (academia, medicine). CS has been democratized over the past 20 years and our H1B criteria in CS needs to catch up.
Oh, and BTW, H1B abuse also fucks over actually skilled immigrants. I've been trying to hire a PhD with huge amounts of high-quality open source contributions, a bunch of publications, and there's a whole class of very particular algorithms that literally no one except him understands. He's literally the only person in the world who is qualified from day 1 to lead this heavy-R R&D project. This is what the H1B is for. But apparently his qualifications are equivalent to low-level Accenture drones doing sys admin and java app dev. I just don't buy that a properly equipped Justice wouldn't be able to hire someone who knows the difference...
If we want a midskill immigration program, we should just create it from scratch. That'll be easier than morphing the H1B into 80% of what a midskill program should look like.
Have you ever seen the difference between a top notch sysadmin and the average one? There are lots of sysadmins who are interchangeable cogs, just as there are lots of developers who are just warm bodies.
Personally, I don't think it matters much. So what if we allow more software engineers from other countries? We haven't done anything special to deserve a job more than they have. My only concerns with it are about their well-being and freedom under the current system.
I agree with you, actually. But we have to be honest about the intent of the H1B program if we want to fix it.
TBH I would support throwing away H1B entirely and creating a much more permissive midskill/lowskill immigration visa whose intent is more in-line with the actual use of the H1B. But if you try to get there by reforming H1B, it's never going to work well. So, if we keep H1B, we need to start using it as intended. And then fight for another visa class. I'm just fundamentally skeptical that the H1B will ever not be abusive until we align the original intent and practice of the policy.
So what if we allow more software engineers from other countries?
We do! The US takes about 1.2 million immigrants legally into the country every year, and they are welcome to become software engineers.
Now, if you ask "so what if we create a visa so that high tech employers can bestow, control, and revoke the US residency rights of people who work in a few narrow fields, including software engineering", I can take a crack at explaining how this can lead to harmful market distortions, to say nothing of labor and basic human rights abuses.
I mean, I don't really want a job in a country paying a fraction of what I make now? It's true though, there is not an existing reciprocal program afaik, but so what? We have things much better economically. And who knows, maybe we could arrange a labor trade deal of sorts with these countries.
By the logic presented earlier , it would only be fair if Americans could compete for jobs in foreign countries the same way their workers compete for jobs in ours .
Sure, I think that would be reasonable. I have two points though. Someone else doesn't have to do the right thing for us to do it. While it would be better if other countries reciprocated, it doesn't have to stop us from doing the right thing. Secondly, from a practical point of view the demand for people to emmigrate from the US and work in a country with lower wages is not nearly as high as the demand the other way.
With the high number of jobs being offshored and outsourced I would not be surprised if many Americans would gladly leave to increase their job prospects. It may be the case that people in the United States can no longer sustain a living here due to the high cost of living. Perhaps we will reach that point.
The United States had a very high employment rate leading up to the covid crisis. Lower than it had been during the 90s, but that was an anonymously propserous decade: https://tradingeconomics.com/united-states/employment-rate#:.... It stood at roughly 62%.
According to the same website India's current employment rate is ~50% (part of this likely due to gender imbalances) and you can bet that wages are much lower. The relevant metric is purchasing power parity adjusted household median income. It's roughly 50 times higher in the United States than in say India (less so for China and Russia).
I don't understand how you can see countries like China or India as presenting more opportunity with local wages. There are of course cultural reasons to want to move there, but I don't think there's any reasonable case that there's more economic opportunity. Some things will be cheaper for sure (rent and labor, namely) but consumer goods, food, electricity, and the like, will not be commensurately cheaper wrt the drop in income.
Yes, but having a job and staying in the middle class in any country is much better than entering a lower class after your position has been either eliminated, outsourced, or off-shored and you cannot find another.
However, maybe there is an argument that the middle class in a different country is worse off in terms of purchasing power than the lower American class. I'm not sure if that is the case, but either way it would be better to be gainfully employed in an industry of a person's choosing than to be lower class in an industry a person does not want to work in.
The threat of this happening in America, as I perceive it anecdotally, is much more real than the statistics suggest. I could see a scenario where only the top 20% of skilled professionals in major industries survive in the United States, while the demand for the bottom 80% is met by outsourced or off-shored professionals.
> We haven't done anything special to deserve a job more than they have.
By being born to an area you have done something special. You have received the special benefit of local education and infrastructure, you have familial and social connections to the people of that local area. You may even have a shared sense of local obligation to uplift the people you live near.
The problem with importing foreign labor isn't that they can't develop feelings of community or form social ties to an area. It's that someone local is harmed or excluded (after significant investment in them) while a foreign worker is underpaid and overworked in a job that exists because of the local's tax dollars.
We're allowing greed to create a permanent underclass of locals who can not climb the social hierarchy while also importing highly-educated quasi-slaves into the country. The whole thing reeks of economic opportunism and, to me, represents the very worst of capitalism.
Yes, I don't think the current implementation is great. Hence my last sentence. That said, it may be better for the people coming in than not having it at all. Although the net harm maybe be greater than without it or a similar program (corporations benefit, locals hurt some, immigrants benefit). It's not clear to me what the magnitude of the harm to locals is though.
Why not just stack rank h1b applicants by salary instead of using a lottery system? That forces the income levels up which should benefit Americans workers as well.
If you did that you'd favour profitable industries (say computing) over areas with actual shortages (say Arabic translators in Detroit). You'd also make it impossible for areas with low cost of living (Detroit) to be allocated any H1B visa. It's one of those perennial proposals that always come up but are really poorly thought out when you look into it.
You could just raise the salaries of translators to $150k to ensure they get a visa. Looking at h1b data it seems they are all employedin the medical field. What do doctors make and why? There’s a shortage of medical professionals. Since translators doesn’t requiring accreditation when you’ll get American workers who will switch into the role from different professions when you increase salary. Making h1bs salary ranked makes the market more efficient and helps adjust incomes in the employees benefit.
If there were "actual shortages" of Arabic translators in Detroit as you suggest, market forces would drive wages up until enough people with the skills were sufficiently motivated to move to Detroit.
In the case of a long-term, widespread shortage, rising salaries would lead to a widespread increase in people training to enter the industry. This is exactly what's been happening with software the past several years.
> If there were "actual shortages" of Arabic translators in Detroit as you suggest, market forces would drive wages up until enough people with the skills were sufficiently motivated to move to Detroit.
I don't know about the specific case of Arabic translators, but that's far from the only possible outcome. It's also likely that you won't find one at any price that your poor Arabic-speaking parents can possibly afford to pay to attend your parent-teacher conference meetings, and you end up with no translator at all.
An actual shortage would, by definition, mean that, at least in the short term, no amount of price pressure would meet the need, because of a supply constraint that prevents response to price level changes. For jobs that need to be done on site and take significant time to acquire necessary skills, national borders can be a supply constraint for labor if there is an absence of local talent sufficiently qualified but there is foreign qualified talent. That's the problem H-1B notionally addresses.
If you have static demand and it is possible to train to meet the need, you shouldn't have long-term shortages, but dynamically increasing needs can produce a long-term gap, where you are continuously playing catch up.
If H-1B was well adapted to real shortages, the continuous use in tech would indicate such a persistent dynamic shortage. I don't think that's the real condition, and I don't think H-1B is well adapted to serve only real shortages, but the idea of real, including persistent, shortages that price signalling alone doesn’t suffice to close is not to be dismissed.
> An actual shortage would, by definition, mean that, at least in the short term, no amount of price pressure would meet the need, because of a supply constraint that prevents response to price level changes.
There are over a million Arabic speakers in the US, hundreds of thousands of whom are legally able to do the work. I strongly suspect that a number sufficient to handle the city of Detroit's translation needs could be brought in at under (likely far under) the rates proposed above. Having done translation work myself, I can say the pay is often startlingly low!
In the unlikely case that weren't possible, the necessary salary would exceed the H1B qualifying threshold, and translators could be imported to do the work.
What exactly are you arguing here? In what situation would a shortage truly require importing labor at low wages? As they are now, H1Bs suppress wages, especially immigrant wages.
> There are over a million Arabic speakers in the US, hundreds of thousands of whom are legally able to do the work. I strongly suspect that a number sufficient to handle the city of Detroit's translation needs
That's an argument that there is not an actual shortage (which, I suspect, is correct), not that an actual shortage could be addressed by bidding higher with wages (which it could not, by definition.)
> What exactly are you arguing here?
Your incorrect statement about an actual shortage being addressable by bidding higher with wages.
> In what situation would a shortage truly require importing labor at low wages?
I wasn't arguing that would occur, but since you ask, a sudden supply constraint or demand surge (perhaps from a debilitating epidemic that spreads particularly well in conditions associated with a particular job) in an essential, common, but not completely unskilled job might require that to avoid massive economic disruption.
> As they are now, H1Bs suppress wages, especially immigrant wages.
I'm not sure I agree with the “especially” part, but, yes, I've said elsewhere in the thread that that is what H-1Bs do, and by design even if it's not the sales pitch.
The thing with H-1B visas is they tie the immigrant employee to a specific employer, thus giving the employees little ability to negotiate for raises. It also results in H-1B visa holding immigrants being paid significantly less for their contributions.
In contrast, adrr's suggestion above to "stack rank h1b applicants by salary instead of using a lottery system" would result in a system where H-1B holding immigrants are well-paid and "shortages" result in rising wages over time.
Re: sudden shocks, I agree with you. Market pricing will tend to fix imbalances over time (barring extreme regulation or other distortions), but it does take time. In a truly extraordinary situation such as you suggest where huge swaths of the population are suddenly dead or incapacitated, you'd have a reasonable argument against market pricing. I don't believe the H-1B was ever intended for such catastrophic scenarios, though.
Isn’t that the same case with hiring America workers. I’d also argue that high pay attracts better talent. This why FAANG companies dominate the market since they offer compensation packages higher than anyone else and attract better talent. You could argue it was anti-competitive from a consumer standpoint but it’s beneficial from an employee standpoint. Immigration regulations are designed to protect American employees.
The employee would need to be skilled enough, to generate enough excess value, to make it worth it for everyone (the employee, the company, ...) over hiring a local worker.
The first purpose is to pay for background checks and any costs that the immigrant might incur in the USA. (For example it would pay for deportation if the immigrant broke the law.)
The second is to serve as proof that the immigrant is enough in demand that the government should believe we get a net positive from their admission to the USA. And not exactly onerous proof - it is only a fraction of what has to be paid in salary and benefits.
Also note that this is only a partially baked proposal. I am sure that there are much better options than this, just as this is would be likely to work better than the current H1B program does.
Yup this was my experience at a very large financial corporation in the US. As a Team Lead I had to interview and hire for multiple positions across my org. Most of our jobs ended up being filled by H1B candidates, and not necessarily because we couldn't find local people. We paid them peanuts, and worked them to death. A lot of the money went to the staffing company that provided them.
Predictably, some people are going to claim racism for this move by Trump, but I think it's a big problem that needs to be addressed.
There are ways to meet the very real need to import the best brains from around the world. But H1B isn't set up to do this very effectively.
Off the top of my head, a better option would be to allow immigration with a $60k bond to the government that has to be paid down at $1k/month by the current employer. With the employee in question able to transfer jobs at any time with the bond transferring to the new employer.