Wow, that's a really important point. I'm so in the tech-news bubble that I'd never heard anybody say that before - thanks for pointing that out!

The point of H-1B is when salary is not the problem.

PS: Remember, someone on an H-1B can leave your employment. Paying market rates is supposed to keep that from happening, thus the intent is to remove a shortage not lower pay.

Median personal income in the US is $30k, so I don't think it's fair to assume every job is like tech and has market rates in anywhere near the same place.

The entire point of capitalism is not everyone get's what they want and price how that is decided.

But you don't appear to have data to back up that particular number for all jobs with no specialization whatsoever, and as far as I can tell the administration doesn't either.

That's the point of around a six figure wage floor. It prevents people from understating the requirements then bringing in a H-1B. I could just Advertise for a secretarial job, then only accept people that happen to speak Farsi.

Why would the optimum H1B wage floor for every industry that suffers from a lack of local talent just happen to be the one that makes sense for tech? And why would it make sense for it to be the same everywhere when wages and living expenses vary so much with location? You're proposing actively ignoring the market wage, not taking it into account.

The H-1B program is a tiny fraction of the overall workforce and if you want it to stick around you need to justify it. Saying industry X needs workers for wage Y, falls on deaf ears. No, it really does not because for the most part it get's by without them as there is just not a lot of H1B workers in any industry.

Having said that, at the societal level we can talk about gaps that a H-1B program can fill in the short term. If there is a sudden demand for people in machine learning then giving the US access to a wider talent pool has value. But, at the social level each industry needs to compete with every other industry or your talking about soviet style planned economies which don't work.

How do you separate companies wanting cheaper workers from gaps society is better off fulfilling? Well wage is by far the strongest indicator. If the oil drilling industry goes though a boom and can pay 200k for people working oil jacks for a few years then they should get most of the H1B pool as they may need it more than every other industry combined.

However, as the demand is met internally wages will fall then some other industry may have a larger demand.

Now the US is a horribly corrupt society with groups with a lot of power use it to get more power. Thus, the actual H1B program looks very different from what I described. But, again I am only saying there is a justification for an H1B program with a very strong salary floor not anything like our current system.

PS: Anyway, understating the requirements and advertising a job that's below market rates is one the main problems with the H1B system that's most often gamed. I have seen more than one advertisement for a junior job title with 5 years of experience and a masters degree. Lying on a form is easy, lying with money is harder.

You mean protected capitalism - protected by artificial minimum wages put in place by government?

You can even get an Indian worker in Mexico and import the results free of charge. But, limiting immigration is one of the fundamental functions of government, and it's not going away any time soon.

PS: I also somehow doubt you would be OK with al qaeda corp setting up shop an then importing large number of people.

https://en.m.wikipedia.org/wiki/Page_Act_of_1875

  You can even get an Indian worker in Mexico

We live in a capitalist society where price is how you communicate demand. Now, if you are willing to pay X times median wage in the US and still can't find people then perhaps there is an argument. But, saying you can't find people when you are not willing to go that far suggests you are not willing to pay market wages.

PS: And yes I chose an undefined X, because there is no clear point when that happens. However, a reasonable lower bound for that X is probably ~3-5.

And then what? Yes, an Alabama community collage can't pay for a language teacher, that's a fairly normal problem for community collages. It does not suggest there is somehow a market failure.

PS: I think you miss understood my upper bound. I doubt people want an immigration policy based on filling any jobs that pays more than ~30k/year.

I don't see what "market failure" has to do with the H1B at all. Just because the market worked correctly doesn't imply that the outcome was desirable.

Some distant collage may chose to pay for such a teacher and then gain students from the community collage who want that instruction. So, in that context your community collage bringing in the H-1B may end up costing an American a job. It could also depress wages for teaching that language discouraging other students from learning it thus extending the shortage over time. Alternatively, the demand may simply not be there for the language at which point the collage is better off paying for a different type of instruction that more students want.

In the end your H-1B is clearly a boon for that collage. However, if may end up hurting the country overall.

The H-1B system short circuits that mechanism. The salary doesn't rise, and young people are savvy enough to realize once an industry starts using foreign workers it never will.

Voila! Permanent shortage of Americans willing to do that job.

It might be a better outcome for native translators who see higher demand increase their salaries. The market has to be fair for all participants. Salary will never increase with demand, if increases in demand are always undermined by bringing in H1B workers.

The supply of gold is huge if I am willing to pay above market rates. If I want a conductive metal I am going to chose something other than gold in the vast majority of cases because of price. In the case of gold the price is based on both demand and resource extraction costs.

Moving to the workforce, students pick jobs in part because of what they will pay. Over time this feedback loop combines with demand and other factors to set a clearing price for the industry. If a job pays less than 100k that's a very big sign that people are choosing to do something else because of pay not the jobs inherent difficulties.

Now what happens if you try and subsidize an industry with H1B's. Let's say you add 50% as many H1B as people working in the field for a huge effect. Well in the short term wages fall and people either find something else, but more importantly students study something else. Fast forward 20 years, the market price is a little lower but not by that much even though lots of H1B's are now doing that job' you still need to entice a lot of US workers. Meanwhile close to 1:1 with those H1B's, US students have moved into other fields.

Thus, unless you are going to have most people in an industry be H1B's trying to help out an industry shortfall with some H1B quota is not that useful and simply subsidizes an industry for minimal benefit.

PS: Even defining things based on job is tricky. I need a Doctor what's the price for that, type: surgeon, type of surgeon: cosmetic. Now a hospital that can fill out generalist one level up can get a discount. Even industry gets tricky as a school may need a doctor for example.

You can't explain things based on an entire economy without taking into account the entire economy, and all it's parts and interactions; much like, say, blood-sugar levels in the body involve multiple organs.

If you focus on one area of the economy, you must avoid those 'non-local' attributes, unless you are willing to do this. If you analyse the attributes of any local economic part enough, you will eventually hit upon a non-local attribute; Hence, any analysis eventually hits this roadblock.

Back to context, the specific effect of H1B visas, on national salary is complicated, because salary not only is non-local, but involves the silent interactions of supply vs demand across and outside the nation. You'll hit all sorts of GM-like fallacies if your analysis is too shallow.

This is why it is necessary to take it, almost as 'faith', in the mechanics of the market; A true analysis of market principles is a complex process indeed, most people can only take them on faith at some level. Even mathematicians accept some (personally) unchecked axioms.

[1] https://en.wikipedia.org/wiki/Parable_of_the_broken_window

If, on the other hand there is no difference, then the visa is justified.

Uh, or you know, you can't afford it?

IMO, that's the crux of the issue. Yes, the clearing price for some jobs mean people are expensive; making some business nonviable. But, that does not inherently mean there is a shortage.

The CEO of the theoretical company is always welcome to put in an extra 30 hours a week and reap the extra economic value.

The first cultural, linguistic, or otherwise geographically-constrained job in the list is #84, "Foreign Language And Literature Teachers." "Interpreters and Translators" is down at #131, with a grand total of 268 applicants - practically a rounding error after 300,000 tech applicants. Whatever_dude's example seems so appropriate, that it's shocking to see it so far down the list.

Where does that page say anything about large companies?

The problem with a hard salary floor is that it doesn't account for differences in cost of living at all. It would very quickly become the visa of coastal cities and little else.

If the aim of the H1B program was to bring in the highest possible paid workers, you'd be correct. But it isn't. These programs are designed in such a way that all states can benefit from them, and a variety of professions.

It is to make it possible for foreigners to provide work to US organizations when no local person is available.

I have hired many many people on H1s. It was always a hire of last resort: when we could not find someone local. It always cost us a lot more than a local person: in both legal fees and procedural fees (you need to verify that the salary you are paying is not lower than the prevailing wage, which is totally fair) not to mention that it takes a while to get the person so while you're messing about you have nobody to fill the job. Some people are unusually distinguished (perhaps expert in a particular machine or language) but lack the appropriate degree, and they are even harder to bring in.

I have a hard time complaining about that (and I'm an immigrant myself). I think it's totally fair that the system should be biased towards locals. But the system does need an escape mechanism when nobody local can do the work.

There are problems. The big outsourcing companies (not just Indian ones but IBM) flood the lottery early. Also somehow they apparently/allegedly pay below market rate, which I don't understand. As usual it's the startups that get screwed.

Note this H-1 is to bring someone here. If I could outsource I would (to Alabama or Amadebad, it would be all the same to me). When we can get away with a distributed team we hire people where they already are.

I submit that there is no work in the US that can't be done by a US citizen. The problem isn't "finding someone" the problem is "finding someone at the below-market-rate wage we want to pay"

In another thread to this article I gave an example of a teacher for a German school -- native speaker, experienced in teaching the German state curriculum. A few such people have moved here and indeed we hired them, but mostly they had to come in on H-1.

There are certainly medical specialties for which there is a shortage of available doctors or nurses, in particular outside the cities.

And you really believe that there is a US citizen expert in every possible discipline in every possible location?

> The problem isn't "finding someone" the problem is "finding someone at the below-market-rate wage we want to pay"

In the very comment to which you were replying I just said we pay more for H-1 workers, in both time and money (and salary -- the labor conditions verification makes sure that's the case. Why do you assume they come in at below market wages? Have you ever hired one? In my experience it's impossible to do, at least for a startup.

I don't have a problem with hiring foreigners as long as they pay a fair sum to the taxpayers for thier diminishment of citizen's welfare. This is because immigrant is a potential drain on the welfare state and also uses resources that otherwise would be enjoyed by citizens.

Also I do believe most h1bs are paid below market rate because they are hired by large IT contractors like Infosys. They have the people and presence in India to make hiring in H1b cheap- they contract to US firms and pocket the difference.

For small business in the US I would agree that h1b is expensive.

discipline, yes, but location is the same as wage - people can move if offered enough incentive.

"The H-1B program applies to employers seeking to hire nonimmigrant aliens as workers in specialty occupations or as fashion models of distinguished merit and ability. A specialty occupation is one that requires the application of a body of highly specialized knowledge and the attainment of at least a bachelor’s degree or its equivalent. The intent of the H-1B provisions is to help employers who cannot otherwise obtain needed business skills and abilities from the U.S. workforce by authorizing the temporary employment of qualified individuals who are not otherwise authorized to work in the United States."

https://www.dol.gov/whd/immigration/h1b.htm

You will note that the word "salary" does not appear in that description.

Unless fashion model is a legal euphemism for someone at the top tier of new tech or skills? edit: Like, some sexy new JavaScript framework or some ancient functional language.

I think they're referring to real fashion models, though, amazingly.

For example, Disney of Florida replaced 250 American IT workers in US using immigrants with H1-B Visas. They in theory could have saved more money by sending that operation overseas but they chose not too.

Computer operations today are way too mission critical to most organizations to risk problems by moving tech overseas at least in operational settings.

I think that's the problem here - those on H1-B visas aren't immigrants.

It is a dual intent visa and most outsourcing firms keep it strictly in the non-immigrant category.

Until they get a DoL clearance & get into an I-140 approval process, they're "temporary workers" who can be dispensed with any day.

Perhaps that's Trump's plan.

Not every company operates like that. I know it's hard to believe, or even comprehend, in the tech world.

Thats the POINT. Today it would be absurd, but a successful multipronged plan to bring both manufacturing as well as constricting supply of people with specialized skills, would raise the salary demands of the whole US population.

The H1B salary part of the system hasn't been modified since 1989. $60,000 in 1989 is equivalent to $120,000 in today's dollars. Does that influence your opinion at all?

Congress created a rationale back then, the industry merely used the tools available since then. Congress' rationale did not change, they are both adjusting for inflation as well as responding to the direction the industry actually went. The industry's adaptation is incongruent with the will of Congress.

In 1989 salary increases were still keeping up with the cost of housing. Or were still expecting to. For example. And a protectionist America really could do that again.

Also: the majority (by far) of H1-B visas are for technology work. It may not technically be "tech only" but in practice it may as well be.

Though obviously a single number is less work.

There's more about specialization than just the area. To stay with the example, there may be many reasons why someone would want to hire an interpreter for a given language, and for someone to be willing to do the job, without it ever making sense to pay twice what a good average living wage is on the area. The "pay higher if it's high demand" excuse is cutthroat capitalism, IMO, and ignores the fact that the employer might not actually be trying to exploit someone but actually bring something that their customer base/community needs, something they wouldn't get otherwise. It's not about being "desperate enough" to hire a foreigner - in many cases, as in certain niche skills, a foreigner might be the best option.

Also remember the H1B is supposed to be a temporary. It's classified as a non-immigrant visa. Temporary need for a foreigner also exists.

> Also: the majority (by far) of H1-B visas are for technology work. It may not technically be "tech only" but in practice it may as well be.

That's exactly my point, and it's one of the many problems with H1B. Tech workers flood the request and suddenly everyone is just thinking about a salary floor cap because tech has higher salaries than other roles. But there are other roles and businesses that could actually benefit from foreign workers, but they're all crushed by how large tech is instead.

Technically, it is a "dual-intent" visa.

> Even though the H-1B visa is a non-immigrant visa, it is one of the few temporary visa categories recognized as dual intent, meaning an H-1B holder can have legal immigration intent (apply for and obtain the green card) while still a holder of the H-1B visa

It means you can apply for immigrant status while holding an H1B. Differently from most other visas, which means you lose them automatically as soon as you apply for residency.

Why should Facebook's ad tech needs trump the needs of a poor refugee community? Budget size should not be the only virtue signal.

Lower cost of living does not mean the area is undesirable to live. Building adequate housing supply (I know, unheard of in the Bay, but some cities do it) can cause cost of living to stay low.

A major weakness of this policy is that it doesn't scale by regional median salary. Under exigency a company in SF could easily pay $130k to bring in a foreign worker with a needed rare skill. A company in Alabama could not.

Usually when I here someone critique the salary floor it's in reference to poorly paid scientists or biologists, as if it's a law of nature that these people should be poorly paid.

For your example the employer could easily show a US worker was not being displaced and could not be recruited. Your example may also be exempt from the H1B numerical cap if the role is with a institution of higher education, a related or affiliated nonprofit entity, a nonprofit research organization, or a governmental research organization.

[1] https://www.dol.gov/whd/regs/compliance/FactSheet62/whdfs62Q...

Edit: This is how H1Bs should operate according to their stated purpose. edge17's link to http://www.myvisajobs.com/Reports/2017-H1B-Visa-Category.asp... shows that it's not being used in this way at all.

Supply and demand curves are (a) curves, not two points on a line (b) curves as a function of price. You can use them to pick a price, but not to come up with one from amounts of X ex nihilo. Even if we're sticking with the most primitive micro-economic theory here, you've discarded a whole dimension.

If you made the floor $100K, say goodbye to non-PhD (and some PhD) scientists! And since the demand for PhDs isn't that high (you often have 2-3 non-PhDs reporting to a PhD), say goodbye to a lot of scientists using the H1B.

That said I do agree that there should be some kind of exception for highly skilled individuals in STEM fields (I believe Japan does something like this).

For example require the company bringing the worker in to post a bond, make it trivial for the imported worker to switch companies, and make the original company to pay the difference between what the worker's salary winds up being and what their new salary is after 2 years if they can't hold on to said worker.

If you're a company that really needs that skill, really can't hire it, and is paying above market wages, then this will not slow your hiring. But if you're trying to hire below market, it would backfire badly on you.

I'm not sure if there's a way to prevent global labor arbitrage. Multinational corporations will shift labor to countries with the lowest wages.

Perhaps they should put a multiplying factor on payroll taxes that's proportional to the total non-US labor wages paid (directly or indirectly) for producing a product. All expenses should be default considered as foreign labor-based unless proven otherwise. It would introduce a ton of paperwork initially, but then everyone will be incentivized to produce documentation showing US labor-produced products to prevent the increase in taxes.

ex: PRICEWATERHOUSECOOPERS LLP

TAX ASSOCIATE - 200 Records, Median Salary $56,181

ASSOCIATE - 243 Records, Median Salary $53,000

ASSURANCE ASSOCIATE - 700 Records, Median Salary $59,000

What exactly do these associates do?

There are many immigrant/refugee communities in the US that are in dire needs of certain services. No matter how well intended the members of that communities are, they run into many a lot of problems during integration. One I've heard about from a friend (who's a teacher) is that once their children starts going to school, because of the language difficulty and even some cultural differences, they usually fall behind and are treated as "special needs" because teachers don't want to deal with them. This creates a vicious cycle.

They wanted to hire a teacher with knowledge of the community and the language, to serve as much as a liaison as to make the community more comfortable. They can't immediately hire anyone from the community because of the requirements for the job - they'd need a degree and about 2 years training. Now, he knows for a fact there's capable people in the country the community came from willing to reallocate to help with integration. But they can't hire them, even as a part of any special program; the uncertainties about the visa process, its cost, and its duration make it impossible. Everybody loses.