Hacker News new | past | comments | ask | show | jobs | submit login
America’s Need for Skilled Immigrants Isn’t Going Away (bloomberg.com)
232 points by petethomas on Oct 24, 2018 | hide | past | favorite | 379 comments

I do not typically post, but these suggestions seem like genuine improvements to the existing system:

"Kerr suggests allocating H-1Bs not by lottery as they are now, but by salary — the more that a company is willing to pay for a foreign worker, the quicker they can get a visa."

"A second change would be to allow H-1B workers to apply for permanent residency green cards on their own, without having to be sponsored by their employers."

Both of these would ensure that we can get smart and talented individuals from outside the US and would drastically improve the ability for the US to compete while continuing to provide broad-based wage increases.

> Both of these would ensure that we can get smart and talented individuals from outside the US and would drastically improve the ability for the US to compete while continuing to provide broad-based wage increases.

As an immigrant, I'm all for better and easier immigration system. As a parent of an 11 year old, I can tell you that what your country really needs is not more smart and talented immigrants, but better education.

When I was brought to work here on H-1B visa, I couldn't understand why the US would need so many immigrant software developers. Now that I've had 5 years to see what your schools are teaching kids, this is no longer a mystery to me. Your education is appalling and is the main reason why I'm planing to leave the country eventually.

America's education is a paradox.

The U.S. produces many of the world's great intellectuals, inventors and engineers. So how does it do it if the education system is that bad? The answer is that it isn't. This is what I've observed:

Its K-12 education system isn't that great in general, but it's good enough.

Its colleges however do a good job filtering the non-academically inclined and catching up those remaining to the rest of the world in 4 years. U.S. college graduates are no worse than say UK college grads in terms of ability, even in technical disciplines. Elite U.S. colleges arguably produce students of equal or higher quality and the elite universities in most countries.

Taken broadly, U.S. grad schools are on par with the rest of the world. Its elite grad schools however are heads over heels better -- which is why everyone wants to go to grad school at a good US university. The funding is way better too.

Certain European countries like France, and funneling societies like India and China, produce more highly trained technical types at the college level, but these advantages wash out at the graduate level.

Most of the world optimizes education at the K-12 levels. The U.S. optimizes education at the tertiary level.

Echoes David Goodstein's "Paradox of Scientific Elites and Scientific Illiterates" from the early 90's[0]:

"The answer, I think, is that in education and in science, as in fast food and popular culture, America is not really worse than the rest of the world, we are merely a few years ahead of the rest of the world. What we are seeing here will happen everywhere soon enough. Our colleagues abroad can take what scant comfort they can find in the promise that our dilemmas in science and education are on the way, along with Big Macs and designer jeans ... Most elementary school teachers are poorly prepared to present even the simplest lessons in scientific or mathematical subjects. In many places, Elementary Education is the only college major that does not require even a single science course, and it is said that many students who choose that major do so precisely to avoid having to take a course in science. To the extent that is true, elementary school teachers are not merely ignorant of science, they are preselected for their hostility to science, and no doubt they transmit that hostility to their pupils, especially young girls for whom elementary school teachers must be powerful role models. Even those teachers who did have at least some science in college are not likely to be well prepared to teach the subject."


You should hear how much elementary school teachers complain about the math portion of their required test. Quite a few fail to pass it on the first time (or second, even) and the test doesn't even get as complex as exponentials/logarithms, if I remember correctly.

I can agree that the graduate level in the US is good, and sometimes the best. But as someone who studied in Europe and taught undergrads in the US, they were appallingly poor students, entitled, and lazy. The university was and is considered (roughly) in the top 15% in national academic rankings, and one of the top for its region. No way they were graduating at the same level as many other countries.

Yeah, I can only compare with Canadian undergrad and they were still far behind by 3rd and 4th year, even though the rankings show the universities higher. The rankings have little to do with quality of education and more with impact of research I presume (or they're entirely made up).

At the graduate level (Master/PhD) I'd say it becomes comparable, but there were very few Canadian students compared to Chinese and Iranians.

The comparisons tend to only work well nationally. See https://www.usnews.com/education/best-colleges/articles/how-...

>>America's education is a paradox.

That is really not that hard to explain. Once you solve hunger, clothing, mobility and housing. You basically take away the most important of human instincts which drive a person to work insanely hard for success.

At that point a person has to call upon his internal human enterprise to succeed. That's hard.

America is just a victim of its own success. This is also called decadence.

Nice anecdote. Here's another theory: American kids are being priced out of universities with good CS programs because those universities have burdensome pension programs that can only be paid by importing international students with rich parents.

Everyone who focuses mostly on universities is forgetting that you have years of education to go through before getting to the university.

Well, as long as we are speaking in hyperbole, everyone needs to realize that K-12 CS education needs CS teachers, who are disproportionately not those international students who either get cushy jobs in SV or go home.

You're getting at the root of the problem: teaching in the U.S. is a low-paid, low-status position. Teaching in other countries is a decently-paid, high-status position. This directly translates into job desirability and ability to attract qualified candidates.

I had some good teachers growing up, as well as a number of bad ones. The good ones all fell into one of these categories: 1) they were young and idealistic 2) they were doing a sabbatical or gap-two-years before going back to their day job as a children's book author, radio astronomer, or medical student 3) they were an independently-wealthy trust fund kid on a mission to do good. I also had some Google coworkers that taught CS in elementary school; they also fell into the "independently wealthy and semi-retired" category.

I come from a family of 3 generations of teachers, but when my starting salary as a computer programmer was within $3K of my mom's retirement salary as a teacher and I made more in 7 years of programming than she did in 39 years of teaching, it's very difficult to justify entering the profession.

(My mom also has stories about the cultural differences in how people treat teachers. The parents of her Asian students would bring her gifts, always treat her with respect, and effusively compliment her on the difference she was making in their child's lives. The parents of many - not all - of her native-born students would treat her like a glorified babysitter, yell at her, threaten legal action, and generally take a complete disinterest in their children's lives. You get what you value.)

I'm a current teacher (looking to get out), and I can completely agree with the cultural differences in how people treat us. I teach in the rural South, and, for one, there's not a culture of education here. People actively disdain schools, even high school, and don't understand the importance because they (and their parents) were able to get jobs without it; this gets passed on to their kids as well.

When they show up at parent-teacher conferences, they're usually mad that little Johnny is failing...even when little Johnny has literally turned in nothing all year, even classwork you gave them time in class to do. And hasn't retaken that test you've given him three times to do it at home. But it's my fault that I just don't get him a straight passing grade, despite all this. Or, their parents don't come/get mad when the kids cheat and get marked for it, etc. And, of course, often the parents of the best students do come, when they're the ones I need to talk to the least!

But, contrast them with my Hispanic students (particularly the Guatemalan ones, and the Mexican ones; the Puerto Rican ones are different). The Guatemalan students themselves are always so focused and wanting to pay attention, and the two Mexican girls usually are. They get frustrated at the Puerto Ricans, who goof off a lot.

Now, I don't see a lot of their parents, but that's more because they all work night shift (as do my students). However, if I offered (extra) credit for them to email/call, or show up to parent teacher conferences, every single one of them would be there. And usually they contact us, often through their kid since they don't speak much English, if their grades drop asking what the kid can change. It's very different in the attitude they have towards teachers compared to my American students (in general), and I think anyone who deals with a lot of them daily and gets to know them (one class is solely Hispanic students, so I don't have to worry about balancing language issues between them and American ones).

A fourth category of good teachers from my childhood was Biology teachers. All of mine had done research but then found that there were not very many jobs for them other than in teaching.

The best high school science teacher I had was an entomologist. Teaching was the best paying job he could find.

Which countries with decent state-funded universities generally don't have because you've decided that the free market "works" in providing what is basically an essential service for the growth of society.

US colleges used to be affordable before the federal government started pouring money into them with guaranteed loans. It's not a free market, but one distorted by government intervention.


Please don't post like this here. If another comment is wrong, feel free to provide correct information, so we all learn something; or simply don't comment.

Edit: You've posted uncivilly to HN before. If you'd please review https://news.ycombinator.com/newsguidelines.html and follow the rules from now on, we'd appreciate it.

Which countries? Here in Canada our universities are very well funded (relatively speaking) and yet the pensions programs are still screwed from 2008.

Never mind the issues with union staff you can't fire despite their rampant ineffectivity...

It’s not the pension programs, it’s the administration, executives, and luxury accommodations. College should be cheap considering the number of students to professor ratio. Keep admin costs low and pass that on to students. No where seems to do that anymore though.

A story of underfunded pensions coming to a local municipality near all of us.

that's not why tuition costs are out of control

The fact that CS programs / education largely start at a college / university level is I think a bigger problem. A growing set of countries have such courses as required curriculum for all students, starting earlier.

Paul Graham pointed out that the US has 5% of the global population. But, it has been a magnet for talent, allowing it to punch above its weight in technology.

You do make a good point. Perhaps america is only using 2.5 of it's 5%. Half, in other words.

But that's still tiny compared to the sheer amount of foreign tech talent america has been able to benefit from.

Your opinion is obviously contradicted by the facts. The American education system is a major reason people immigrate here in the first place. Many of the best universities and research centers in the world happen to be in America. You may have some gripe about your kid’s teacher or whatever, but if you feel that a single data point from a parent’s biased perspective is more informative than the facts and statistics, then there’s not much point trying to have a reasonable conversation with you on this topic. You are choosing innumeracy and anecdote over easily accessible data.

Maybe it's my fault, but I haven't found "easily accessible data" on the quality of elementary and middle school education, which is something a lot of people seem to be ignoring in this discussion.

Sure, people might be motivated to immigrate here by your universities. Even when I was a teenager growing up in an Eastern European country, I had heard of MIT and dreamed of studying there. But there's a long road to go down before you arrive at your university's doors.

Another thing I would like to point out is that an important aspect of an education system is the access to it. It's kind of like your health care: some of the most advanced treatments are available here, but there's a huge inequality preventing vast numbers of people from having access to those treatments.

Bottom line is that the overall quality of your education system should not only be measured by the quality and prestige of your universities, but a myriad of other factors that influence the final outcome -- that outcome being an apparent shortage of your own skilled citizens.

> that outcome being an apparent shortage of your own skilled citizens.

The interesting thing is even when we had millions of people out of work who worked in technical fields, SV and other major tech hubs were pushing for even more H1-B's.

It really has nothing to do with skilled citizens and more to do with the salaries they demand which tend to be on the scale of 1-3 times more expensive to hire and keep then the off-shore people they're hiring.

Just an anecdote for you:

I was hired as a contractor on a financial application at a major health care provider here in the states. I was the only US born developer on the entire team. That included developers, project managers, and scrum mangers, On a team of 18 people, I was the only American. Several times I overheard team members talking about salary and what they were getting paid. It was shocking to me that I was getting almost twice as much as they were per hour for virtually the same work.

It's not the skills or education that are lacking, it's the cost of the labor which is considerably lower for off-shore talent or H1-B contractors.

From my personal experience and observation, a large amount of employers that say that they can't fill their open positions, skilled or unskilled, have unrealistically high standards. They would rather leave a position unfilled than risk making a bad hiring decision. To give them a benefit of doubt, they might have more to loose from a bad hire than a successful one.

Of course, I don't have any direct experience with the few companies that are famous for pushing the boundaries of the H-1B visa.

If you want your children to be educated well in most regions of the United States, you either need to do it yourself or through private schools. There are a small handful of public schools that have the funding and leadership to do it well, but they are rare - and usually located in the richest counties.

You can't pay attention to achievement metrics or school ranking systems either. They are almost all relative to American schools only and fail to measure well across different global education models. You may find your local public school as an "A" grade from some rating agency, but it means nothing. The only reliable metric is how many of the graduates are accepted into top tier liberal arts colleges.

Frankly, I'm not even trying any of that mess with my children. We teach them ourselves.

A shortage of skilled citizens is another way of saying costly labor. Better access for US citizens to universities and healthcare will not be gained by importing ever cheaper highly skilled labor, especially labor that like you is here to take what they can, selfishly, before they jump ship.

The k-12 numbers are easy to find. Google Pisa.

> Better access for US citizens to universities and healthcare will not be gained by importing ever cheaper highly skilled labor

I believe I already stated that more immigration is not the solution, in my first comment in this thread.

> especially labor that like you is here to take what they can, selfishly, before they jump ship.

What's with the hostility and personal attacks? I'm paying the same taxes are everyone around me, but without the right to influence the system because I can't vote. So I can't see what makes me selfish, except the fact that I plan to "jump ship", as you put it. I came to this country with high hopes and expectations, with the idea that I can not only enrich my life and that of my family, but also work on marvelous tech and make the world better. What my actual experience turned out to be is out of scope of this discussion, but would certainly be a prerequisite for judging me fairly.

I’m not judging you. You came here because you had personal desires that, noble or not, were not fulfilled. And you will leave for that same reason. That seems like pretty common human behavior to me and so I make no judgements about your morals, but I will point out that a less selfish way of seeing the world might lead you to ask yourself how you can have a positive impact and make your new home a better place. Perhaps that sense of personal responsibility is an American affectation. As one of our leaders once said, ask not what your country can do for you, ask what you can do for your country. That seems even more apropos for those who have made the conscious decision to adopt this place as their home.

> Perhaps that sense of personal responsibility is an American affectation

for all the talk of "liberal snowflakes", i do find that americans possess a decent amount of personal responsibility, insofar as they will take action to improve their own lives. as for improving their home country, i don't really see many americans doing much other than talk. the only people i see taking action in this way are members of marginalized groups working to improve things for their group and my white liberal friends taking selfies at protests.

> I’m not judging you.

Again, I might be wrong, but every time I've read or heard the word "selfish", it was always used as a judgment.

> ask yourself how you can have a positive impact and make your new home a better place

I was under the impression that having a discussion about how to improve things with a group of people who might be in a better position than average to help improve them was also a way of working towards that goal.

> Perhaps that sense of personal responsibility is an American affectation.

This is the third country I've lived in, including my own country of birth. My experience is limited, but it's still more than the average person gets. From that experience, it seems to me to be a generational thing, much more than cultural. It can be seen in different attitudes that different generations adopt towards their employers: generally older generations see their employer as part of their life and a kind of a "family", whereas younger generations view it as a business transaction with severe imbalance of power. I expect same thing happens with patriotism.

There does seem to be at least one cultural aspect to it, though. Americans, as well as citizens of other countries where capitalism has advanced to similar extremes, seem to view things with a sense of "personal responsibility" rather than collective or communal responsibility.

> As one of our leaders once said, ask not what your country can do for you, ask what you can do for your country. That seems even more apropos for those who have made the conscious decision to adopt this place as their home.

Why would it seem "even more apropos" for immigrants? I honestly don't understand this logic: why would you expect guests in your home to take better care of it than you do yourself? And I use the word "guest", rather than "fosterling", because not all immigrants come here immediately convinced that this will be their new home.

Here's a truth about a lot of us immigrants: we come here with a hope, but not with a conviction. We would have preferred to live a happy, fulfilled life in our own country and we tried, but for whatever reason we couldn't and we decided to leave. The definition of insanity, as they say, is doing the same thing over and over again and expecting different results.

Besides, we're not talking about what the country can do for me. We're talking about what the country can do for its future generations. I already have an excellent education. But I have a much bigger responsibility to my own son, than to your country or even my own.

> as well as citizens of other countries where capitalism has advanced to similar extremes, seem to view things with a sense of "personal responsibility" rather than collective or communal responsibility.

that's wrong, what you are referring to is "individualism vs collectivism", there's definitely very developed countries who favor collectivism, the perfect example is Japan.

Google "PISA scores"

Private, elite prep schools and colleges in the US are the best. The average college or high school is dismal compared to the rest of the world. We consistently score below the average of OECD countries.

These stats are thrown off by a number of factors that it would be imprudent to get into here. Control for demographic samples similar to the other countries and you'll see a very different picture.

Singapore is a multi-racial country, and it does better than the US on PISA. So is Canada and it does better.

And if your point isn't race or culture, but of economic disparity, that is sort of my point.

"Multi-racial" is not a homogenous term you can just throw around as though it means the same thing everywhere.

Singapore is full of Chinese people. Now compare Singapore Malays to Chinese-Americans.

So you're saying poor people in the US don't matter because they bring down the average?

I don't think that's what I'm saying. Of course they _matter_, as much as any of us do anyhow.

What I'm saying is that if you actually compare people with similar demographically-expected IQ and socioeconomic situations, Americans stack up fine. It's the extenuating circumstances that drag the overall averages down.

All that really means is that overall averages across 300 million people aren't that useful.

They matter, but they don't negate the statistical extremes.

For what it's worth, I didn't go to any elite universities and I make less than 35,000 per year, so I don't have a horse in this race.

Do the comparison again controlling on country of origin. E.g., how do US citizen students with families from Japan compare with students still in Japan? Instead of Japan, substitute Canada, India, Pakistan, the Philippines, South Korea, England, Germany, Finland, Sweden, Norway, the Baltic countries, the Balkan countries, Austria, Switzerland, Israel, Poland, Russia, France, Spain, Italy, Singapore, Turkey, Iraq, Iran, the Gulf States, North Africa including Egypt, West Africa, Central Sub-Saharan Africa, East Africa, South Africa, Mexico, Brazil, ....

The US is a uniquely diverse country, and such a diverse country necessarily will have a tough time competing in average performance scores with better countries that are not diverse and, instead, have a quite uniform culture. So, right away on average scores, the US will have a tough time competing with much of Europe and the better Asian countries.

Also understand that the famous, prestigious US universities -- Harvard, Princeton, Yale, Cornell, MIT, Johns Hopkins, NYU, Columbia, most of the SUNY schools, most of the Big Ten, CMU, Chicago, U. Washington, most of the University of California campuses, Stanford, Caltech, University of Texas, Rice, Georgia Tech, etc. are RESEARCH universities and do not concentrate on technical or vocational training for software developer positions where H1Bs are common.

The US is not uniquely diverse, Canada and Australia among other countries have a higher percentage of foreign born population.

The OP was in part discussing education in the US and claiming that the education is inferior to the education in several other countries. Some of the evidence is some test scores on some tests given internationally.

Okay: We're considering education quality and using test scores as evidence.

Well, it is Social Science 101 (my wife's Ph.D. was in mathematical sociology, and my brother's Ph.D. was in political science) to control on other likely explanatory variables.

So, an obvious candidate alternate explanatory variable is culture of the country of the recent ancestors of the students.

So, to compare education in the US with that in, e.g., Japan, we should see how (A) US students of recent Japanese descent compare with (B) Japanese students in Japan. So, now are holding culture reasonably constant and, thus, doing better comparing the schools and whatever else we have not controlled on.

Not to control on culture would be grade F in Social Science 101 and give total garbage results.

I.e., no fair blaming the education in the US when the real cause of the differences is culture and not the education.

Foreign born doesn’t necessarily mean diverse.

Why are you picking a fight with the parent poster when they are describing k-12 education (which is rotten; many studies have shown this) and using our universities (which are good) as your counterpoint?

May be, it is that the academic level after high school is much higher while the academic level in high school and below is very low.

How are U.S. schools failing students? What should they be doing/teaching instead?

Where did you go to school? How did your education differ from what you've seen in America?

It really depends on the region and city, the US educational system has zero consistency across the nation. It also varies considerably by income, where rich school districts have educational standards above European levels, and where poor districts have standards on par with Latin America or even Africa.

It's not that we don't produce educated people, it's we don't consistently produce educated people across the board and your location and income is a huge factor in your future success.

Those living in West Virginia have almost no chance, and those living in California have much better opportunities. The US, like many countries, falls prey to sweeping generalizations that are at best partly true.

Some state schools in not well known areas of the country are excellent and very affordable (IU, Purdue, Case Western, Carnegie Mellon).

Of those, only IU and Purdue are "state universities". Case Western and Carnegie Mellon are private universities and aren't particularly affordable, although certainly more affordable than Harvard or Yale, etc.

Carnegie Mellon is neither a state school nor what I would call affordable.

why wouldn't you call it affordable? CMU has a $72k sticker price (which only 30% of students actually pay in full) and has an average financial aid award of $44k. if your parents are actually destitute, you will pay very little. the elite universities in america do not cost much more than a typical state school if you are poor.

Education in this country echos everything, it's a shadow of the extreme wealth inequality.

You are spot on.

Used to be, at least in my region of the USA, companies made local investments to build the talent pool needed.

Tektronix is a stellar example.


People literally walked in off the street, got jobs, got education and went on to do great things.

It costs more to do that, but the benefits are easily seen.

Today, we see much less, and demand for people, largely from places that do continue to invest in people, has grown dramatically.

I just wish it was reciprocated. I would like to work in Spain (no really). If I work in Spain, the US should open a spot for someone from Spain to work in the US. You want to hire someone from Turkey? No problem, find someone in the US to move to Turkey. You could allow companies to even grease the wheels, "Work from Turkey and get a 30k / year extra free bonus stipend!! yay!!"

Not on a 1-for-1 basis, but something like this is the argument for allowing free movement between comparable countries. If the UK, Canada & Australia opened up to each other, there'd be no massive net flow, but many individuals could work for the most suitable company / near their in-laws / in a climate they like.

This was I think also what was envisaged when the EU first wrote free movement into law -- general mixing, rather than one-directional floods.

EU folks can work any where in the EU right? I wonder how it is working out

Places to rent in cities in Germany are getting harder and harder to find because we have so many migrants, most of them from the EU. This is about the only problem we have with it locally, but it is getting annoying. The housing market is not very flexible and takes a while to catch up.

Migration is limited because almost every EU country has its own language.

Mostly... brexit is how it worked out.

Only after an influx of economic migrants. Without that influx of migrants, I don't think most of the pro brexit people would have been concerned about the free movement of people within Europe.

Doesn't free movement include economic migrants? That said, they have also had a lot of non-EU immigration.

But my point was that free-movement between near-peer countries seems like almost pure benefit. And this was where the EU started, knocking down walls in benelux... you would struggle to get many people upset about that.

Integration all the way to Romania etc. is more recent, really quite a different thing. The hope was for rapid convergence, and the evidence is mixed I guess.

I was responding to a very short post by coding123. My point is that the free movement of people within Europe may have been a necessary condition for brexit, but it was not a sufficient condition. The free movement of Europeans did not lead directly to brexit. It was only after a massive influx of migrants from Africa and the Middle East that we had brexit

I do not disagree with your claim about near peer countries.

Personally I would like to see freedom of movement in trade agreements. Goods and capital can already flow freely - letting labor do so as well would be more free and have potential to make the workforce something competed for more as well. Companies can already outsource so letting people flow with less artificial resistance (as opposed to natural like language and cultural barriers) seems both fair and efficient.

Granted I acknowledge that has potential to be messy in reality without a whole lot of work to get a framework - let alone one that would be accepted. Perhaps some sort of "social service fee" minimum for benefits in taxes paid to prevent free rider issues for instance.

You just invented the European Union :)

Can we try it again except this time without the power of an organization to push awful copyright laws down the throats of every country[1]? It’s like if you consolidate power it becomes easier to corrupt.

1. https://www.eff.org/deeplinks/2018/10/effs-letter-eus-copyri...

I dont know about Turkey, but at Spain 30K/year wont be your bonus. Will be you salary. And consider it good salary.

I have no idea why this comment was killed but yes, 30K/year is a pretty good salary in a lot of Europe.

Especially in eastern Europe, for example in Romania you could live like a king on $30k per annum

I wont say that about Spain. You will have a decent life. If you want to save money or invest, that's it. No fancy dinners nor trips during your PTO. You will also have to work until you are at least 65-70 years old to have a regular retirement. I wont call that living like a King. Just like a regular working class person. Decent? Sure. But that's it.

It's a figure of speech, living like a king is overrated anyway, you got no privacy. I'd say that you'd live a comfortable life nonetheless

Yeah well rural Europe is a lot of Europe, but 30k in a capital where the 30k jobs are give you is just mediocre.

This would be the extension of an already bad idea in the US immigration system that almost no other country shares. I can't think of another country that has this notion of per-country quotas.

Spain isn't interested in getting an American immigrant just because one of its nationals moved to the US.

There’s obvious brain-drain problems. Turkey doesn’t want it’s smartest people moving and working here. Nor is it in the interest of the US for high skilled workers to go abroad.

I'd argue that it is in our (U.S.) best interest for highly skilled worked to go abroad for a short period of time (2 years?). Being exposed to different ways of living can do wonders for creative thinking.

This kinda implies that countries own people.

That's lunacy. How many Americans can you find that want to work in Africa, or India or China?

Its going to be a infinitesimally small fraction compared to the number of people from those regions that want to work here.

How much will it pay? Americans work in UAE and KSA where their life looks like it was plocked from a dystopian novel.

But if your doing that as a westerner your on a full ride expat ticket with a considerably higher salary and can save a lot.

For really dodgy areas not soft postings like UAE / KSA you also get a premium. I remember a colleague who was on +80% for Angola (he got robed twice on the trip from airport to town)

I'd love to work in China. So there's one.

Underrated idea

The problem is that pay isn't uniform across skilled labor demand, and the purpose of the H1B program is to fulfill demand in the economy.

If you set a quota and give to the highest bidder, you're going to see private sector executives and financial roles being filled in cities with high costs of living and competitive wages. Salary is a function not only of demand but also location and industry.

This will discriminate against rural and middle America by depriving them of the skilled labor that they need to survive.

> This will deprive rural and middle America on the skilled labor that they need to survive.

That's kind of already happened. Since the recession almost all new jobs created were in cities and metro areas, and mostly to educated workers.

Rural and middle America have this problem all on their own: skilled labor does not want to live in rural and middle America. Especially young skilled workers. There's even the concept of rural brain drain, as the top 5 % are sure to leave right after they're done high school.

> skilled labor does not want to live in rural and middle America.

I'm unsure about this, and think it's kind of a chicken and the egg problem. I know tons of people who are from the midwest that would prefer to be there, but tend to flock to the coast simply because it's hard to find as rigorous of a professional environment back home.

It's not a chicken and egg problem. Cities are much more efficient for labor and companies. You have a larger pool of talent, a larger pool of freshly educated replacements for those who retire or leave, and often more services available for said talent.

Edit: Now that I'm thinking about it a little more I can maybe sympathize a little bit. I prefer to live in rural areas in some ways due to my hobbies, but I'm never going to as the economics of a tech company, or even a company that needs skilled, educated workers makes no sense in rural areas. So there's something to the want vs. choice angle, but what one fantasizes about has little to do with economic realities. Plenty of people want high paying low skilled manufacturing jobs to be a thing again, that doesn't mean it's realistically going to happen.

Another factor is the 2-body problem. If most of the workers are looking for a city with enough jobs for two different careers, then that's where employers will need to be to hire them.

I don't disagree with what you're saying, but one thing about less dense areas including parts of the midwest is you can commute from rural areas into the downtown of a city in under an hour (where these are parts of metro areas that can contain over 2 million people).

I'm not suggesting that a headquarters needs to be in these smaller cities either, but often times right now, there isn't even a satellite office with any engineering.

s/want/choose/ in the original, I think.

That's an entirely different problem, and switching H1B to an auction will only exasperate it without solving the task that H1B is intended to solve.

> If you set a quota and give to the highest bidder, you're going to see private sector executives, lawyers, and financial roles being filled in cities with high costs of living and competitive wages.

Yes, that's the point. Skilled labor immigration is for skilled labor.

It should not by a subsidy to companies that cannot afford to pay prevailing wages and/or are unwilling to invest in their workforce.

> This will deprive rural and middle America of the skilled labor that they need to survive.

Middle America will be fine; Chicago, KC, Pittsburgh, etc. can and do compete with the coasts for skilled labor!

That said, I could get behind a rule that adjusts for cost of living so that e.g. 100k in Pittsburgh or KC == 200k in SF. That seems fair.

As for rural America, there's a reason these areas can't attract companies or talent.

  > As for rural America, there's a reason these areas 
  > can't attract companies or talent.
Look, I don't want to live in a remote shit-hole either, but...

The fact that the concerns of rural communities are summarily dismissed when making policy was one of the big contributing factors to the election victories of both Brexit and Trump. (Possibly more due to resentment rather than as a rational response, but that's irrelevant.) Urban vs Rural is becoming one of the more significant divisions in this country, as wealth consolidation favors urban.

And, more fundamentally, there are people who live there. The actions of a democratic government are supposed to be for the benefit of its people. That includes the people who live in rural communities. Policy that lets the economic advantage of urban areas snowball more is something that should be given more thought than "it wouldn't matter anyways, those places suck."

Maybe there isn't a solution, or maybe this policy in particular isn't the best place to address it, but it deserves at least a modicum of thought rather than a dismissal that is, frankly, disdainful. Those places do suck, but we should treat that as a problem to be addressed rather than avoided.

[edit]: Let me clarify: I am not saying the urban-rural divide is caused specifically by the lack of high-skill immigrant labor. I am saying the urban-rural divide is exacerbated generally by dismissing the concerns of rural communities when making policy. I don't think this policy suggestion in particular is the highest-value one, but I also think that no individual snowflake is responsible for the avalanche.

Sorry, but this doesn't scan. "The concerns of rural communities" are overrepresented in American politics. Twenty percent of the population is represented by the Senate majority and are able to, at minimum, gridlock the American system of government (and, at maximum--well, quite a lot, given SCOTUS and what we've seen over the last few months).

This is, of course, an intentional behavior of our system of government. You can even make an argument that it's a positive one. (I would disagree.) It is not, however, "democratic", nor does it point at some kind of underrepresentation.

And actually the problem is worse - some of those places don't suck, and causing them to RAPIDLY grow in population, turning them into Urban areas that really shouldn't be.

It would be much better if more small towns had similar offerings such that the 10% aren't overrun by people looking to get away from the city.

Of all of the reasons for Trump winning the election, rural areas not having enough immigrats is not one of them.

I am sure that many of these places would be very happy to have less H1Bs living there.

> The fact that the concerns of rural communities are summarily dismissed when making policy was one of the big contributing factors to the election victories of both Brexit and Trump

Do you think that an influx of underpaid skilled labor is going to ease the tensions that created Brexit and Trump?

Your assertion here appears to be that Trump voters were pissed off that more immigrants weren't making it to their hometowns. I find that assertion... highly implausible.

> Policy that lets the economic advantage of urban areas snowball more is something that should be given more thought than "it wouldn't matter anyways, those places suck."... Those places do suck, but we should treat that as a problem to be addressed rather than avoided.


1. IMO they don't suck. They have many advantages. But they aren't particularly good locations for doing highly skilled labor.

2. There are a lot of policy levers that can be used to help rural areas. Trying to turn them into immigration-fueled discount tech/med hubs is doomed to fail and is therefore definitely not the right lever.

3. It's also worth noting that even rural economic development tends to be geographically concentrated. Just as wealth concentrates in cities with 10M people compared to cities with 100,000 people, wealth also concentrates in cities with 100,000 people compared to cities with 1,000 people. And BTW, those 100,000 person towns where skilled labor would invariably concentrate are in many cases already doing quite well.

Even rural development -- and ESPECIALLY rural development based on skilled labor immigration -- isn't going to help the rural places that have been hardest hit in the past two decades, which tend to be in the 1,000 to 10,000 range.

The argument isn't that policy should benefit urban America. The argument is that trying to bootstrap tiny local economies with skilled labor immigration is doomed to fail.

>underpaid skilled labor

H1B visas are only granted to people who are paid more than the prevailing wages of citizens working the same SOC code job in the same area. There have been consultancies in the past who have gamed this, but USCIS has cracked down severely on this in the past year or so.

> Middle America will be fine; Chicago, KC, Pittsburgh, etc. can and do compete with the coasts for skilled labor!

These cities may as well be "coastal" relative to their regions. There are a lot of hurting small PA towns less than 100 miles away from Pittsburgh experiencing a brain drain, for example.

> These cities may as well be "coastal" relative to their regions.

IF KC and Pittsburgh aren't "middle america", then the term ceases to have much of a meaning and we might as well stick with "rural".

KC and Pittsburgh metro areas have populations over 2M.

There are a lot of smaller metro areas that are far rural, but aren't big cities - for example, my hometown - Erie, PA, which has a metro population of just over 275k - is very much a city - but is too far from Pittsburgh to benefit from their economic growth. Pittsburgh is fine, but Erie is in decline.

If "middle america" refers to every non-rural city between the coasts, then it is a meaningless term. There is too much economic variation to lump them all together for a single analysis.

I don't see rural and middle America crying out for more immigrants of any type. They won't be complaining, and it isn't proper to complain on their behalf due to whatever you judge to be best for them.

That leaves nobody to complain.

If you insist on deciding what is best for rural and middle America, I sure hope you don't object when they return the favor, deciding on your behalf that cities would be best off without any immigration.

The supply / demand in the economy is specifically determined by the salary. If the demand grows high enough so that salaries go up, then you can obviously compete with other highly paid professionals. There is no reason Tata and Infosys should be hiring 70k contractors to do shitty jobs.

>The supply / demand in the economy is specifically determined by the salary.

Salary is a function of location, industry, and demand. Discriminating against American businesses that need skilled labor and are not in the most competitive industries or locations is not the purpose of the H1B program.

>There is no reason Tata and Infosys should be hiring 70k contractors to do shitty jobs.

This is a different problem, which is being solved as USCIS is cracking down on consultants who are registered in one region but work in another.

Arguably, the salary (and price in general) is measure of marginal utility. Hence if one job pays 140, 000 over 70, 000 we can argue that the one that pays more provides more marginal utility to society. That is, the United States as a whole gains more value from that job.

Or you could argue that the U.S. gains more from allowing 20 lower-paid civil engineers than allowing 1 private sector executive buy their H1B.

The point is that switching to an auction with the H1B causes more problems without really solving any, at least within the scope of the program's purpose.

Immigration policy has to balance two concerns: letting in unassimilated immigrants too quickly, and missing out on the economic value they bring. Therefore, it should optimize for economic value added per unassimilated immigrant. The harm of adding to the unassimilated population must be justified by the value brought.

Bringing in a lot of foreign civil engineers may be more total economic value brought in, but in a way that doesn't justify the cost in unassimilated people.

An "auction" by how much employers are willing to pay is indeed much closer to what the policy is trying to optimize for.

The lower-paid high-skill jobs should not be using the scarce H1-B spots, but should raise wages to draw more Americans, since the premium is much lower than you'd need to hire an American for the higher-paid jobs.

But the choice is not between 20 low paid engineers vs 1 highly private sector employee.

The choice is instead between 20 low paid engineers and 20 high paid executives/engineers/employees.

This makes the answer to this choice much more clear.

Which comes back to location and industry discrimination, since higher pay is a function of these different factors.

Again, auctioning H1B with quotas doesn't solve a problem, it just creates new ones.

I don't know.

If you're paying 20x what a civil engineer gets then you might as well go with E5 Investor Visas. Its true that auctions have issues, but we could go with (for example) second price (aka Vickrey) auctions for salaries.

How you would think the E5 is a substitute for an H1B is beyond me. A business isn't going to pay their executive to found and work at an entirely different business, especially if it jeopardizes the business's ability to sponsor future visas with USCIS.

EB-5 has nothing to do with founding and working at a different business. USCIS allows applicants to invest into EB-5 Regional Centers, which are essentially funds that would not require you to work there or start any new business. If an investment is made into a TEA, the capital requirements drop to $500, 000. Its effectively a for sale green card. If you are a sufficiently important executive, $500, 000 is chump change to ensure you can stay in the United States indefinitely.

I'm not sure why you are turning to hostilities. Please review the legislation.

The business then needs to cover the executive salary on top of EB investments into other businesses - that is not a substitute for H1B. This fact along with addressing the tone makes it seem like you're fishing for a red herring.

> This will discriminate against rural and middle America by depriving them of the skilled labor that they need to survive.

Rural and middle America needs to get off its collective ass. I have seriously considered moving to one of those places -- alas, while I can get a house cheap and have what I need to survive inexpensively, I cannot get high speed internet even if I pay $1,000.00 per month.

That's not the fault of most of those communities. Many of them have tried to set up their own municipal wifi/ISPs and many (most?) have been stopped cold by legislation and back room deals at the local and state levels by the telco and cable companies. Just read some of the stories about what Google, one of the richest companies on Earth, ran into when attempting to bring high-speed Internet to various cities.

I don't care. I have money. I want a service. They can't provide it. This means they do not get to complain that my six figure buying power chooses a city.

> This will discriminate against rural and middle America by depriving them of the skilled labor that they need to survive.

What jobs are rural American companies hiring H1Bs?

You could index it with cost of living. So an H1B in a California company would expect to see a higher salary than one in Kansas.

The easier way is to just have ratios. Companies that already employ lots of Americans should be able to easily hire foreigners. Companies that hire predominantly foreigners should have a difficult time getting more. A few rules to prevent gaming the system and you've got a system that works well for everyone involved and doesn't need CoL adjustments. Plus, ensuring that foreigners work closely with Americans helps both integrate those foreigners into American culture as well as helping them learn from those foreigners.

The H1-B system seems designed to allow big companies to exploit foreign labor to drive down wages and hold those foreigners hostage while they're here. Ratios better allow mobility of foreign workers and allow them to seek market rate salaries. If I were in control of immigration policies, I'd set a 3:1 American-to-foreigner ratio based on job classification/salary bands and make it apply to any company with more than 50 employees. Any immigrant who passes a DHS background check and can get an offer from an eligible company should be able to get a work permit, regardless of their home country, with no time limit when they'd be sent home. Only failing to find employment at an eligible employer within, say, 6 months should force someone to leave. And after a 5-year working period, they should be able to apply for citizenship. At worst, it becomes a jobs program that forces employers to employ American citizens to get the workers they need.

Banning any body shops from the H1B program all must be direct employees - would help I think.

That still doesn't consider pay differences between two completely different jobs.

Then it would be working as intended. Jobs where labor isn't so scarce as to push prevailing wages up really high don't need extra labor brought into the market.

The remaining concern is distorted markets like foreign language teachers (brought up elsewhere in the thread), where it may be politically infeasible to raise compensation to match the value workers create.

The purpose of the H1B program is to fulfill skilled labor demand where local labor is unable to fill it.

The purpose is not to give visas to the wealthiest bidder in the wealthiest locations and industries.

Prevailing wage requirement already solves the problem of wage velocity, so this is just creating more issues without solving any.

"Unable to fill" is itself a function of the salary. If a local business is unable to fill it at what they're willing to pay, how about paying more to attract said labor from other parts of the country?

H1B and similar programs should be used to fill in the holes that are on the national level - i.e. where there's no labor force anywhere else in the country that could be enticed to move.

"Unable to fill" for skilled graduate labor, not all labor. Key difference.

Perhaps this could be offset by using something like the GS Locality Pay[0] index?

[0]: https://www.federalpay.org/gs/locality

Your arguement is true, but it is also, IMO, the intended behavior.

Having higher paid workers push out the lower paid ones, is kinda the whole point. I want more skilled labor to move to where they are in demand.

Does rural and middle America rely on H1B to fill its labor pool today?

"Kerr suggests allocating H-1Bs not by lottery as they are now, but by salary — the more that a company is willing to pay for a foreign worker, the quicker they can get a visa."

Would need to be artfully worded so that sign-on bonuses and short-term employment is not used to game the system by permitting averaging down over time.

I think I agree with you (although I bet for different reasons) but, Why?

What do you think is actually is the problem if highly compensated employment is used as a fast track to immigration and perhaps even citizenship?

rrggrr is worried companies will say "1 year contract, $10,000 per day first week, $0 per day subsequent 51 weeks" and jump to the top of the stack as they're paying $10,000 per day.

Presumably the salary they bid would have to be maintained throughout the visa validity (e.g. three years for H1-B).

If the salary ever drops the visa is invalidated and becomes available for bidding by other companies.

You can require them to pay an average salary over the whole period of employment.

Most employment doesn't have a planned end point though

Why is a planned end point necessary? An average of say $100k per year isn't impacted whether you pay that person $50k for 6 months or $200k for two years.

It's to prevent you hiring an H1B for a one-month, $50,000 contract, then hiring them for a separate 11-month, $50,000 contract after. The company will argue the initial hire is a $600k/yr rate.

You could argue that they would get a one-month visa and top priority for that first month. The employer would have to go through the visa process again for the second contract.

They're already here, renewal is a completely separate process. The biggest hurdle is getting them off the plane, if they're employed, sponsored, and here it's much easier to keep them here.

Most visas do though, so it could be worked out.

Why do you need a planned endpoint?

In the UK, the immigration agency checks that you are still being paid above the minimum salary for a work visa. I see no reason why they couldn't check on this sort of thing.

I always thought H1B should simply be $100k or $120k+ plus jobs only.

I was looking through online records and found folks being brought in for sub 50k jobs that looked like low level ... customer service roles.

That's what I thought too, then I saw my company use an H1B to fill a junior web developer job at 70k in the DC/Metro area, which is basically the area's going rate for a 0 experience, straight out of college software developer.

Man, if someone needs an entry level web dev fast... there are so many bootcampers available. Granted the quality varies wildly (I'm a bootcamper myself, I know those people) but they are easily found.

That seems fishy, the minimum required salary to apply for h1b was 60k/year till the beginning of the year, and there was a bill that was introduced which proposed raising it to 90k/year.

Straight up it was sub 50k listed in the database...

I'll see if I can find it later.

you can get the same effect on salary by making it easier for a H1B recipient to leave the company that brought them over. Right now its very hard for a visa recipient to move companies so they can be given sub market rate salaries, if you make it so once they show up they can move to get a market rate salary then the companies who make a pratice of hiring foreign talent and paying them below market rates will end. The companies will incur all the cost of bringing in an H1B and some other company can just pay them more and receive the benefit.

This also addresses the concerns about salary differences between different industries and regions.

In theory it is not very complicated for a H1B to change employer. It simply requires the employer to file a petition to sponsor the H1B. It doesn't count towards the "seats" available yearly, and it's almost guaranteed to be granted. There is a small cost, maybe a few grands. If you think about it, it's way cheaper than paying recruiter fee fo example.

In practice you're right though. Most companies have limited experience with this process and just don't want to get into it.

If you intend on applying for a green card you need to stick with the same employer until your I-140 is approved. Just getting there with a single employer can take years, and if you don't have an approved I-140 when your H1B expires your out of luck. Also up until 2017 even if you had an I-140 you would still lose your spot in line if your employer revoked it.

oh man. This chain of responses must be literally on hackernews like a million times. Maybe someone can extract QA site from hackernews comments.

Two problems:

- The green card process doesn't completely transfer over.

- Filing a petition to transfer an H1B puts the employee at risk of getting a denial. Every petition sent to USCIS is risking an audit, denial, and at the very least, many headaches.

Much of the green card process has to be redone though.

Can you elaborate? I'm not very familiar with GC via employment.

Word on the street from hiring managers is that it now takes > 6 mo to process such a visa transfer and you can’t get premium processing to speed it up.

People are making offers to candidates with undefined start dates many months in the future. The candidate has to commit to switching jobs when the transfer is submitted. If current job gives them a promotion in the interim and they decide to stay, they can’t just cancel the transfer. Their current employer (who didn’t know a transfer was happening) has to file a reverse transfer with the same delay that puts them in some kind of limbo.

It’s a complete clusterfuck and is making it really hard to hire for some positions.

I think most of this comment is wrong. A "transfer" is actually not a thing. When another employer files a petition for a worker who already has an H1B, it doesn't "cancel" the first H1B. It simply makes it possible for this second employer to hire the worker. It is completely possible for the applicant to change their mind and decide to stay with the first employer.

I went through 2 H1B "transfers" (so in total I have been employed by 3 different employers as a H1B) 7-ish years ago so I'm very familiar with this process. At least, the way it was working at that time.

Has anyone had success with an H1B visa transfer in the past few months? If so, I’d be very interested to hear and it would add more to the discussion than a downvote.

A nice idea but would make it impossible, for example, for my kid’s foreign-language school to bring in qualified teachers.[+]

It’s difficult already as teachers switch jobs over the summer which over the past couple of decades has been when the supply is already used up.

[+] English teachers are recruited from the California system. Unfortunately CA schools are so terrible that if the kids go back to their jobs,encountry they are frequently behind in English!

These sorts of "highest salary wins" rules should typically be crafted in an industry/sector-specific way. So schools have to compete with each other on salary but not with tech/finance/med. And then there are caps by sector.

The problem with this approach, of course, is that you need a judge/executive who follows the spirit of the law so that I can't get a bunch of cheap tech workers by saying "oh, job ABC in town XYZ is its own industry/sector and not comparable to SWEs at Google".

I'm not sure I agree with this. I work for a university, and I am stunned with how talented and accomplished many post-docs are considering the level of low pay and career uncertainty they continue to face.

The RAND institute, among others, concluded that the reason US citizens have turned away from graduate degrees in the sciences is a rational market response to poor pay and working conditions in favor of fields that offer better salaries and career stability. In short, there is no shortage, just a desire to pay extremely low salaries and establish exploitive working conditions - which you can do when you control someone's right to live and work in the US when these rights are difficult to come by.

Not only do I think that there's no reason to prop up a system that suppresses salaries outside tech/finance/med (or within those fields, that happens too, especially in tech), I think there's real harm in the market distortions these schemes produce.

There's always a trade-off. You can have cheap steel, or well paid steel workers. You can have cheap medical care, or well paid doctors.

Arguably by making postdocs underpaid, the US as a nation is getting cheap scientific advancement, in exchange for dangling the promise of eventual American citizenship (and the American dream) in front of said postdocs.

I don't think that's worth the overhead. If you end up only importing medical professionals then that's going to help the US economy quite a bit more than importing 'highly skilled' bank tellers.

Money is a very good proxy for demand after all. If company's can easily place people for 350+k jobs that's going to lower the pay of those 350k jobs.

PS: Effectively, your describing a planned economy where someone picks the level of expected pay for each job with more foreign workers available for overpaid jobs.

It sounds like foreign-language teachers ought to have a different type of visa. This is a case where being foreign is actually a benefit, unlike the situation with engineers.

How about allocating some large number of 6 year work visas by a pure auction? The visa goes with the person, they can work anywhere, and after the visa is up can be a US citizen with minimal hassle. Employers could pay the fee for people to come and work for them, if they like, with a claw-back clause, if necessary. It would be interesting to see what the price would be for such a visa.

> The visa goes with the person, they can work anywhere, and after the visa is up can be a US citizen with minimal hassle.

You basically described a green card (except it's nominally five years to citizenship, not six).

Yes. A green card auction would be great. Expand it to some large number and get rid of all the huge variety of visas and paths to citizenship. Maybe half or more of all immigration to the US could go this route.

After some years of high demand and high prices, I imagine that the price would drop to some stable level. When that happened, we might have an opportunity to bring some sanity to the illegal immigration problem in the US. In some more sane world, both parties might agree that a person that can prove they have been in the US for some number of years can pay for a green card at the auction rate (gov financing available).

As a former H1B, I fully support the second change yes. IIRC, an H1B can be renewed once and last a total of 6 years (apologies if regulations changed in the past 6 years, I haven't followed since I got a GC). 6 years is a long time and should allow application for GC without sponsorship.

I'm not sure the first change is a good idea. Wouldn't it give a huge advantage to bigger companies? Haven't really thought this one through.

I think the first is a great idea. A $450k/year neurosurgeon should get a massive preference over a $45k/year Sharepoint contractor. The highest paid are tautologically in the shortest supply.

Realistically though, a neurosurgeon would probably be able to immigrate on another type of visa. Same goes for most executives, lawyers, partners at VCs or hedge funds...

An auction based allocation would favor ad selling companies that can pay juniors 200k to do A/B testing.

Then sure, it would partially solve the abuse from cheap IT consultancies, and that's better than nothing. But there would definitely be adverse effects.

How is that different than now? Larger companies can already afford to pay people more and smaller companies have to compete in salary.

Outside of the west coast and FAANG companies, salaries aren’t that crazy. I haven’t seen any difference in comp based on company size in most of the US. They all have to compete for the same talent pool.

> ...would drastically improve the ability for the US to compete...

of course "the US" is not a monolith. it's a collection of competing interests. big corporations, consumers, and the existing stock of US-based workers each claim to be "the US."

While the salary thing would help me gravely, not all industries/states pay in a similar way. It would help software companies in CA/NY a lot, but a biological research center in Midwest would be tough out of luck.

They'd definitely need to account for region, rather than just looking at the dollar amount.

This would result in most H1B allocations on east/west coast, where salaries (and cost of living) are relatively higher. I think this is already the case, but enforcing it by salary will result in a wider gap. Companies operating in low "cost of living" areas, will not be able to get skilled workers on H1Bs.

There is already a “prevailing wage” thing you have to meet, that varies by region, so it could be done as percentage above the prevailing wage.

I"m not sure this should be a consideration. If that midwest company cannot compete, it should not exist. That is tghe essence of capitalism. You should not subsidize it by making it not compete fairly for H1B applicants.

It’s a bit of an odd argument to make against a state sanctioned limit to labor that it should be driven by pure capitalism/free market principles.

The premise of the article was to use a capitalist supply and demand metric (salary) as a way to fairly allow merit based immigration. I think that strategy makes a lot of sense, and was making the point that our hypothetical midwest company should not be allowed to scuttle the non-gameable merit-based proposal in favor of something more political and squishy. Our hypothetical midwest company can either pay enough wage to get the employees it needs, or it can't, and if it can't, capitalism has a solution for that. (creative destruction)

The problem I have with this argument is that salaries don't have to be comparable to be competitive, because cost of living is different across the country. I make a fair amount less working in Ohio then if I moved to California, but I would also pay a lot more to live there.

So, making hiring these employees based on salary only means that places with a higher cost of living will be able to hire these employees for more typical rates, where-as companies in lower cost of living areas will have to pay well above average for the same employee. If actual salary wasn't a requirement, then midwest companies could still be competitive even while offering lower salaries.

That would only make one more thing you can get better with more money.

So you want only the rich and privileged to be allowed into the USA? I find this highly negative.

Having people immigrate who are guaranteed to be able to make their own way is not only NOT negative, it's written into the immigration policies of virtually every nation.

This is a separate argument from allowing economic refugees asylum into a country.

So you want to hire managers and other upper level people but not skilled technicians or artists? If we are going by the whoever has the highest salary idea, we obviously won't get real skilled workers. Just more managers. I am not talking about economic refugees.

The H-1B topic strikes a personal tone with me. After the dot-com crash in the early 2000's, IT jobs in California were hard to come by. I almost got one job, but the company decided to go with an H-1B worker instead. I had a new young family, but had to go out of state to find work, leaving my family behind.

I'm NOT summarily against H-1B's but wish to insist they be used ONLY to fill in true skill shortages instead of those manufactured via job ad word-smithing.

> I almost got one job, but the company decided to go with an H-1B worker instead.

This scenario is explicitly forbidden by the provisions of the visa program and it is illegal to select a guest worker in preference to a domestic candidate. That is why there is a public notification requirement for all jobs open to guest workers. Obviously all it takes is one minor reason to deem you unqualified. The Disney IT incident shows that there are no consequences for abusing American employees.

"Public notification requirement" = job posting you leave up and internal email you circulate to CYA in case you get audited. It's a ruse. This can't be enforced.

They can go get hired elsewhere I suppose. Does it rise to 'abuse'? Who is entitled to a particular job anyway?

Who is "entitled" to work in America is defined by the law, which disallows replacing American workers with H1B recipients.

Are you suggesting that everyone in the world should have the right to live and work in America?

I think it will not change anything to be honest. I work for a relatively large US company we have offices in CA, VA, NC, Ireland, Germany, India. If it becomes harder to higher in US they will just grow teams in other offices.

But not every organization is internationally distributed like that. If physical location of employee really didn't matter, then your org would have most of its IT staff in India, because the labor rates are lower and that's the largest pool of English-speaking IT talent.

I agree there is some fluidity to IT labor, but there is a limit to it also.

Besides, the stated on-paper purpose of H-1B visas is to fill (alleged) skills shortages. After the dot-com crash there was a big surplus of labor in the area, yet us citizens were competing with H-1B's still. The stated goal was missing the target during that time.

I'd say the ones employing majority of programmers are and that will affect the market at large.

As someone who has worked on H1b in US, I can understand what you are saying.

Primary reason an H1b with a net hourly rate same as a US citizen is more attractive to hire, is because the costs of firing that employee are too low. Unlike a citizen, especially if the citizen is older.

With the suggestion made in the Bloomberg article, it’d have probably been a positive outcome for you if H1b was distributed by way of highest salary first, instead of lottery (which I think didn’t exist during the dot com boom).

Don't mean to sound harsh but anyone who can't compete with an H1B to the extent they can't find a job in the entire state of CA should take a hard look at their skills and expectations and change one or both.

Please elaborate. The time-period was after the dot-com crash. Companies were laying off IT employees left and right, often because the companies died. CA was flooded with excess IT workers. I even put an ad on Craigslist around 2002 that I'd do development for minimum wage. No takers. What else do you want? I nearly left IT. Okay, I'm not the cream of the crop I admit, but are you saying average developers should get out?

But companies in CA were still hiring, as you mentioned. If the labor pool was flooded with laid off workers, then it's a buyers market. You would think the companies can just hire from that pool for whatever they wanted to pay (cuz in such a climate, they would have takers), without bothering with the H1B search and paperwork hassle.

Hiring isn't as simple as paying the least possible in the current climate. Significant variance in salary among people with similar jobs will be an issue (people talk). Additionally, labor surpluses don't last forever. Low-ball salaries in that situation will lead to turnover at the worst time - when the economy is recovering.

If they are thinking medium or longer term, then the short-term slump may weigh less in their decision process. A visa worker is less likely to be ABLE to go elsewhere if/when economy or an industry improves. They are almost like indentured servants.

Theres a ton of talk about skilled immigrants and the american education system, and as a blue-collar automotive mechanic, I have no problem with immigrant doctors and engineers. Where I see the real problem is unskilled immigrants.

Take for example a job posting I had up for more than 3 months. I needed a solid mid-level mechanic. oil/air/tires/suspension and electrical a plus. I dont want my apprentices doing this work because theyre learning to do manifold rebuilds and such. And i dont want my old timers doing it because it will cost a fortune and i need them on harder problems.

So for weeks on end, I get at least 3 interviews a day from grossly unqualified immigrants. Why immigrants? nobody wants a kid who grows up to be a successful mechanic i guess. Everyone "goes to college" now. just ask my alignment tech with a masters in english.

- Ive had a slew of random tourists from europe backpacking around in the US who want a job, and cant seem to understand the gravity of a visa violation. They all want three times what I'm paying and health insurance.

- Ive got college drop-outs from china, russia, you name it, who either blew through all their money or were never qualified for college in the first place, and need something to keep afloat while they fake college.

- and finally, perhaps most controversially, I have central and south americans. Theyre either trapped here because of american immigration policy, or theyre fresh off the boat with a slip of paper written by a friend dictating what they will work for in cash. Most of them speak about 8 words of english. They are radioactive as far as im concerned; you would be insane to hire them.

We're always going to need doctors and engineers. not everyone grows up to be a rocket scientist, and thats okay. But im frustrated at how Americans have turned trade jobs (plumbing, auto work, electricians, linemen, pipefitters, etc...) into some kind of field that no kid can ever enter lest they be disowned.

update: I should clarify. These applicants are upset they must wait 90 days for benefits. we do provide health/dental/401k, just not on day one.

So you're complaining about unskilled immigrants applying for your job opening, and apart from being unqualified for the job, they apparently can't even legally work in the USA anyway.

What about if you had skilled immigrants applying for the job? Say, some qualified mechanics, whether from Germany or China or Mexico?

The problem with America's immigration system (from an outsiders perspective) is that skilled immigration visas (i.e. H1B) are only worthwhile for high skill white collar jobs, and in particular for larger companies who can be bothered with the visa lottery.

I see no reason why skilled immigration for lower paying, but still skilled jobs, like auto mechanics shouldn't be possible. In most developed countries, they run a points system. If you want to move here to Australia and become an electrician or any other high demand job, it's relatively easy as long as you have a qualification, good health, and can speak English, and I think that's the way it should be. Obviously long term you should be trying to train more mechanics, but that's a 5-10 year process, and doesn't fix the problem that you're facing right now with a lack of mechanics.

It sounds like you have a bigger problem than simply a lack of qualified candidates. Do you have any unqualified Americans applying for your job? How little are you offering that European backpackers (who in my experience will work for a pittance) are asking three times what you're offering?

> Do you have any unqualified Americans applying for your job? How little are you offering that European backpackers (who in my experience will work for a pittance) are asking three times what you're offering?

Bingo. I think you nailed OP’s problem right there. If you’re offering to pay in bananas, don’t be surprised when you get a line of monkeys out your door.

I respect everything you're saying, but you kinda lost me at, "They all want ... health insurance".

Yes, indeed... as long as a soul-crushing job sitting in a chair and staring at a screen comes with health coverage, and a job in the trades does not, then my hope as a father is that my kids will take the health coverage route. Until we fix our crazy system, blue collar work it is a very frightening prospect.

> But im frustrated at how Americans have turned trade jobs (plumbing, auto work, electricians, linemen, pipefitters, etc...) into some kind of field that no kid can ever enter lest they be disowned.

I think the main reason is that you can easily find an office job where you won't be doing much other than showing up and participating in the office politics, that pays comparably to the jobs you offer.

Your salary constraints are obviously dictated by supply/demand (i.e. what others agree to work for and what the customers agree to pay for the repairs), so it's not really your fault, but I see the same trend in many industries - plenty of bullshit jobs pay more (or comparable) to the "real" jobs.

Frankly speaking, it spooks the hell out of me because it reminds me of the last decades of USSR before the big crash happened.

It seems more of a prestige thing than anything really. Americans seem to have a low opinion of blue collar workers.

>They all want three times what I'm paying and health insurance

Is health insurance not something you give out? If that's common enough in the trades to be a complaint about employee desires, then no wonder no one goes into the trades. There is definitely a stigma attached to blue collar work, but part of that is it not paying. All I had to do was read a few tutorials online and a few years later and the only way I'd be making as much money in the trades is by owning my own business or in super specialized niches.

Why would anyone looking at a career pick the trades over that?

If no one wants the job, then you're not paying enough or offering enough benefits. Period.

The fact that you name health insurance as a problem is a big enough warning sign.

Strange that the government acts with zealous intervention and unconventional ideas to solve worker "shortages" but does essentially nothing to solve extreme housing costs and shortages.

The Government isn’t sending out airplanes to pick up workers and bring them here. The legal immigration system is neither easy to get through nor available to everyone. Stop blaming the Government for all your imaginary problems.

>all your imaginary problems

A personal attack as your response? Am I on 4chan?

I agree with what Kerr is suggesting. It makes a lot of sense.

I also agree with a many Americans who are anti-H1b that it suppresses wages, hence makes it attractive to higher foreign workers, quite often for jobs that are not truly state of the art, but do require the technical know how.

This is why it makes sense for the H1-B to be a highest paid gets first kind of a visa. That way there's a strong incentive to not hire lowest paid employees.

In fact, I think a better way would be to give two 3-year temporary green cards. if at the end of 6th year if the person still has job, it becomes a permanent job. This removes any control that an employer has to depress the wages, and be fair to the American as well as immigrant labour. Such a system can be made available to at least the undergrad/grad/doctoral degree holders from the US universities.

By limiting the immigrant talent, US has already ensure China is taking a huge leap in terms of tech advances. India will figure out how to make the leap soon.

All the while, Canada is making the most of USA's anti high skilled immigrant situation, but inviting them in droves. We'll see a different Canada in the next 5-10 years, mainly due to the huge number of tech immigrants (of Chinese or Indian origin) moving to Canada.

>>In fact, I think a better way would be to give two 3-year temporary green cards. if at the end of 6th year if the person still has job, it becomes a permanent job.

No, the entire reason companies get away with paying H1Bs below market rate is because the visa is tied to employment.

It needs to be decoupled so that visa holders can freely and easily switch jobs without worrying about losing their status.

Or both?

ie grant them to people being paid the highest wages first, and limit the h1b holder to mandatory 1 year employment with the sponsoring company then 5 years of free to switch at will.

We'd still be willing to sponsor as we aren't keeping h1bs via the visa; we keep employees because they want to work here. The pain with sponsoring isn't the cost; it's the time and uncertainty. Sponsoring costs compare favorably to the standard 25% a contingency recruiter charges.

>> grant them to people being paid the highest wages first

I'm against that because it would give FAANG a massive advantage, since they can afford to pay really high salaries compared to, say, a startup.

average salary of h1b holder in bay area is $91k [1]. If startups can't beat that, they don't deserve employees.

[1] https://www.businessinsider.com/h-1b-visa-workers-earn-six-f...

You seem to have not comprehended the parent poster's suggestion. He wants the highest salaries to be given precedence. This has nothing whatsoever to do with whether startups can afford the average salary.

My point -- communicated poorly -- is the average salaries h1b startup employees are being paid is quite low. There's lots of room for them to grow.

Further, why shouldn't we allocate visas to the most valuable, ie highest paying, uses?

>>Further, why shouldn't we allocate visas to the most valuable, ie highest paying, uses?

For this question to be valid, you first need to demonstrate that someone's value can accurately be determined by their salary and their salary only, and that any difference in salary between two people is the result of differences in the value they offer (as opposed to, say, being able to interview well, or knowing someone high up in the company).

Though it should be divided by industry. Otherwise, highly paid software engineers will take all the visas and other industries will struggle to match the salary required get a visa. How to divide industry is a very messy business though. Some of our immigration laws divide by industry and those were written 30 years ago. A lot of jobs we hire now didn't exist 30 years ago.

So what? Wouldn't that just make wages rise in those other industries, pushing more Americans into those apparently underserved fields?

I don’t know whether to agree or disagree with that.

Fields that’s get affected negatively could be offered some kind of respite wrt quota. Eg a medical professional on h1 probably won’t be subject to quota.

I think Eric Weinstein covers this question in a relatively academic way in his article:


"During the late 1990s I became convinced that in order to orchestrate lower wages for scientists, there would have to have been a competent economic study done to guide the curious policy choices that had resulted in the flooded market for STEM PhDs."

The supply of technical workers is artificially limited by Leetcode style interviews. There doesn't seem to be an actual supply side limitation, especially not at the entry level.

Maybe on the west coast but here in the Midwest we often have issues finding candidates that can write fizzbuz.

We recently had a candidate who's main language was python but failed to even run his code on the command line. Saddly this isnt overly rare.

I use to work in the midwest before moving to the west coast.

We had a guy come in for a programming interview. He claimed to have 3 years experience with a programming language we commonly use. We gave him a simple challenge (connect to a database using a database connector, read some fields and update another field based on data from the read fields). We told him it's fine to use google, and left him alone for four hours with a computer.

He asked to expand his interview time at the end of the four hours, we grant him an additional two hours. At the end he didn't have code that compiled, much less connected to the database.

This story is not atypical from my experience.

First I thought thats not straightforward to make a DB update because you know, not everything is loaded into our brains working memory like which package to use, exact syntax etc etc. But then I read that you guys gave 4 hours + Google and I am shocked. BTW are you hiring :)

Out of curiosity, what language was it?

Actually even in the west coast we have a hard time finding candidates that can write even fizzbuzz level of code. Its not even a recent problem.

The worst I've hard was a person with 20+ years of consulting experience for top companies in firmware could not code 3 lines of C code to traverse a linked list. Literally didn't know the syntax to dereference pointers!

Sounds like you need to pay more, have a better working environment, else work on your recruiting pipeline. A phone interview should be able to sniff that out quickly.

Anyone with a CS degree can write a fizzbuz

Wow, didn't realize this was a comedy site.

Speaking from personal experience, the number of people with multiple years of experience and multiple degrees in CS (usually BS and MS) who can't pass FizzBuzz is really high.

That's amazing to me. Not arguing you're wrong, I'm not interviewing people, just amazes me that anyone with degrees in CS would not be able to solve such a basic problem.

I don't understand this either. There is an interesting blog post that suggests maybe some programmers get tripped up because they want to find a more clever way to do it:


It doesn't sound like this is the problem, though, in all the cases. It sounds like someone who majored in CS was unable to write this.

Which seems inconceivable to me. And yeah, "inconceivable" is often based on wrong assumptions (it was the basis of one of the oft-repeated lines from cinema: you keep using that word. I do not think it means what you think it means).

But... is there a chance that maybe "CS" just means different things from different programs. I feel like it should be... well, impossible for someone who graduates from a proper CS program to fail to write fizz buzz.

I know, I know, the "no true scotsman" is already getting typed into the response field. But here's the thing, I'm not defining a "proper" program as one that results in an ability to write fizz buzz. I'm defining it as a program that contains, at a minimum, basic algrothmns and data structures. Do people graduate with CS degrees without coding a linked list, binary tree, doing basic sorting, and set permutations?

I'd say if you take that one required class, it should be more or less impossible to not write fizz buzz. I'd be interested in knowing what the "CS" curriculum was for people who can't write this.

You don't even need algorithms or data structures -- it's a cs101/151 problem at best. When I first saw the problem the solution appeared self evident to me -- I've been coding since the 80s, but still. It's just a simple problem and writing the pseudocode for a question like it ought to be almost second nature to anyone who has done any coding at all.

I know people confuse CS and IT/CIS, but even at that, those majors generally require at least one basic CS class.

I agree, you don't need data structures for fizz buzz. In fact, I think it would be tough to pass data structures if you weren't at the level where you could write fizzbuzz, so it's more or a pre-req. I have trouble imagining what is taught in a CS program that doesn't have a course in data structures and algorithms. How can it be CS?

That said... I keep hearing about fizzbuzz, and I'm not sure what I think of the complaint. I hear something along the lines of "We need to do technical screening because you wouldn't believe how many people can't do fizz buzz", but the questions aren't fizz buzz, they're "find all matching subtrees in a binary tree", often at the whiteboard, in 45 minutes. I'm not saying that this is (or isn't) unreasonable, just that there's an awful big difference between fizz buzz and the kind of questions I've encountered in a typical tech interview.

Both candidates are in abundance. I know plenty of “engineers” who can’t do fizzbuz. I also know talented developers who can do fizzbuz for you in 60 seconds, but might not be able to invert your binary tree on a whiteboard in that amount of time. Neither of them are getting past today’s ultra picky interview-hazing rituals. One probably shouldn’t get hired while the other should. Companies seem to have a hard time distinguishing them.

You'd be shocked how many people don't know about the modulus operator.

Even so... try dividing. If it looks like division truncates, check the product. If it looks like floating point division, check if rounding/truncation changes the value.

Or keep two extra counters that you increment every iteration and reset when they hit 3 or 5.

Not knowing modulus or bitwise ops etc isn't a good sign, but it isn't disqualifying. Not being able to get fizzbuzz working should be disqualifying, though, unless you're willing to teach someone to program completely.

That's a fair point, but not even knowing about the concept?

I agree with the parent comment. So many people just can't code the most basic program correctly. And this is with a generous amount of time and forgiving syntax errors.

Well, it is a little bit more complicated than that.

I am sure that any CS major couos write fizz buzz, on their computer, for the purpose of their homework assignment or whatever.

Whether or not they can do the same thing, in 15 minutes, on a whiteboard in front of someone is a different story.

The moral of this story is that writing code, on a blank slate, while having to have all the syntax memorized, and having everything conpile without having access to a compiler, is a real skill that many people haven't even attempted to do.

If you're applying for a job as a full-time software developer, you should be able to write something that is comprehensible without having a syntax reference.

I straight-up tell people I don't care if you use the wrong method name (in Python is it ends_with() or endswith? toupper, to_upper, or upcase?). Heck, I even tell them I don't care what language it is as long as it's not too weird.

A few months back I interviewed someone who was currently employed as a software engineer at a major company in Silicon Valley, who couldn't even figure out how to write a tiny little program which kept track of state across method calls. I told him that he could store the state in global variables, make an object to track the state, or anything else he wanted; still couldn't make any progress outside creating a single stateless function. I was truly puzzled.

Is the correct and most efficient way to solve it?

  #include <iostream>
  using namespace std;

  int main()
      for (int i = 1; i <= 100; i++)
          if ((i % 3 == 0) && (i % 5 == 0)) cout << "FizzBuzz" << endl;
          else if (i % 3 == 0) cout << "Fizz" << endl;
	  else if (i % 5 == 0) cout << "Buzz" << endl;
	  else cout << i << endl;

Off the top of my head, you can do a single mod 15 for the first case. I'm sure there are other micro optimizations, but really being able to write exactly what you did is sufficient to move on to real questions that actually have something to do with the job you're interviewing for.

Technically, this would be more efficient:

  #include <iostream>
  using namespace std;
  int main() {
    for (int i = 1; i <= 100; i++) {
       bool t = (i % 3 == 0);
       bool f = (i % 5 == 0);
       if (!t && !f) { cout << i << endl; continue; }
       if (t) cout << "Fizz";
       if (f) cout << "Buzz";
       cout << endl;
But most interviewers are interested in the basic logic of the process.

What is the explanation? Presumably, to get a CS degree you had to solve problems harder than FizzBuzz, no? Are we suggesting these people cheated their way to a CS degree?

Anyone with a CS degree can write a fizzbuz

That's not even close to true. I've interviewed people with Ph.D.'s in Computer Science and a documented 10+ years of experience as a "developer" of some sort, who can't write Fizzbuzz.

Decades ago tech companies like IBM would hire entry-level software developers with zero coding experience based on the results of aptitude tests. Then they put the new employees through internal boot camps. There's no reason a similar approach couldn't work today. Employers just generally got lazy about workforce development while the economy was slower and they had an unlimited supply of applicants.

It's not just software engineers. This also happens with skilled labor, like machining.

You used to be able to get hired and taught on the job, now the employers expect you to pay your own way through a technical school for several years before applying for a minimum-wage job. So you've got (a) lower wages (b) you need to pay to educate yourself (c) you need to pay to support yourself during those years of training and (d) you're guessing which job skills are going to be in demand after N years, so after spending the time training and taking on debt there might not even be a job waiting for you.

In the uk I often see posts for graduate programmer positions with essentially this arrangement: words to the effect of “candidates need not have experience coding but should show a willingness to learn and deep interest in tech” etc.

that's how I see it too.

I feel like the whole question here is simply: how much commitment to the existing US labor pool should US laws/regulations require from US employers?

Over the last decade, the answer has been "not much." But, now, because of political and economic realities, that answer might change.

This. If there was a real piece of missing demand these companies would change their interview process, but things are only moving in the opposite direction. The shortage is a myth to flood the software engineering job market with cheap workers. It needs to stop

Piggybacking off the fizzbuzz comment, but at my job we we don't ask hard whiteboard questions - just simple ones that prove people can code, and 75% of them can't do it. If you can't do fizzbuzz, you can't do this job. It's an effective filter.

Had one guy who insisted on using C++ in favor of C for a problem despite being hinted to use the latter. He was a self proclaimed "expert". All was good until forgetting to use delete[] instead of delete.

That and a complete unwillingness to educate and mentor your workforce. I don't understand why the government should step in to help employers because they want training to be somebody else's problem.

Companies don't train their workforce because said workforce then leaves and joins another company, which then benefits from that training.

...that's how it was explained to me by an executive at my former job (a big part of the reason why it's my former job).

California is an at-will state and all that, but surely executives and business consultants can structure some sort of incentive package to cause workers to stay for at least some minimum duration?

If a company was willing to train me and provide me decent salary and benefits, I'd totally stay. If there was room to move up in the company, and they kept offering good salary and benefits, I'd probably never seriously consider leaving unless I hated the job or just got an amazing offer (barring events like needing to take care of parents, etc, of course). I feel like a lot of people would prefer the stability, even, especially if the company is investing in them.

The workforce leaves because the company doesn't raise the salary to account for new training.

You can't have it both ways.

That by definition is called a “market force”. They should pay them more once trained.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact