I mean it is economics 101 that H1B's depress wages. I'd say especially since the demand for technology is relatively more inelastic that the increases on labor supply. There is a lot of money involved in masking this fact, but let's agree on something. There is always a tension between what part of economic surplus workers get and what part capital (shareholders) get. The sky high returns on capital in tech are possible in no big part due to the relative ease of bringing in new people. Successful companies will still make money but the actual return on the invesment will depend more on how high of a salary they have to pay. Take for instance FB, they make about 18Bn in net income, have 52K employees. About 350K per employee, imagine suddenly they were facing hiring engineers, they could easily spend a lot more on keeping them, but there's no need. Thanks to their lobbying through FWD.us among others, they can always import more people.

I mean they have desperately tried to get more people into CS, but even that has had little results. Biggest bang for the buck is lobbying and marketing for more H1Bs. What's a couple hundred million here and there

It's not actually obvious in "Economics 101" that wages are depressed by the best and the brightest coming to the US.

What you're not taking into account is the relative efficiencies of scale. When there's more talent in a particular geography than somewhere else, high value global companies can grow in a way they can't elsewhere. A precondition of that is lots of raw material - people - who can be repurposed by the highest bidder. Limit the raw material, and the industry might not even be located on the West Coast.

If supply was enough to depress wages, the US would have the world's lowest cost developers. In fact, it has the most expensive, because of the valuable companies that are able to grow in SV, which each bid up wages.

Economics teaches us about winner effects. The best attract outsize rewards, far larger than their proportional increment in ability or effort. SV attracts the best of the best because it's the best place for such people to be rewarded. It's a virtuous circle in multiple, compounding dimensions.

People are a fixed cost in the economics of software development. That means scale is king; it's the divisor on those costs. Scale isn't just on the sales side, though; it's on the talent side too.

Reducing skilled immigration is the kind of shortsighted approach that kills the golden goose in the long term.

Exactly, even though the economies of scale were making the economic value of the company much higher, this extra will just keep disproportionately going to the shareholders. In fact, the bigger the companies, the more power they hold in wage negotiations. Think about an efficient market with many, many buyers and sellers. What kind of equilibrium does it reach? Now think of a few big companies, hiring from an ever increasing supply. The companies grow bigger, but the wage that laborers receive, will decrease. At some point it becomes this question, why should American workers give up an ever bigger part of their wage to shareholders and other immigrants?

In spite of their best efforts, most of the software development still occurs in the US. Think Microsoft hasn't tried to offshore as much as possible to India and China, building dev centers there? Or Google and FB?

Even then, just look at what kind of workers end up being imported through this programs. Not always the best and brightest. Many do entry level jobs not even in software development, for outsourcing companies.

I'm not saying outright ban immigration but better control is needed. (And saying this as a non-American living in the US). Something that struck me as particularly interesting and President Trump recently commented, is how about making companies bid for a fixed number of visa slots. Whoever pays the highest salary, gets the visa. This would certainly make companies be able to bring the brightest, while raising wages in general.

> This is such a lie. My brother has looking for around a year and cannot get his first position.

Anecdotes are not evidence. Maybe your brother is bad at interviewing. Getting the first position is always the hardest.

That's a great angle for looking at - do we have data on the average age of h1bs? How can a young person already have the specialized skill they're supposedly brought in for?

You've highlighted the problem with public discourse in the country. "I'm going to choose to ignore all scientific evidence because my brother can't get a job and it feels like H-1B is the cause" - sadly applicable to any policy topic where politicians & media are taking advantage of people's emotions.

I am not ignoring scientific evidence. I read the paper, it is not convincing. This is not a double blind experiment, it simply looks at data and asserts causes. Take this paragraph:

"An increase in the share of workers with an H-1B visa within an occupation, on average, reduces the unemployment rate in that occupation. The results indicate that a 1 percentage point increase in the share of workers with an H-1B visa in an occupation reduces the unemployment rate by about 0.2 percentage points. The findings suggest the presence of H-1B visa holders boosts employment among other workers in an occupation. The results provide no evidence that the H-1B program has an adverse impact on labor market opportunities for U.S. workers"

There is nothing scientific about this. No experiment was performed. They simply looked at the data, saw there was lower unemployment in occupations with higher percentage of h1b, and then asserted that means there is no evidence that the program has an adverse impact on us workers. They are reversing cause and effect.

You didn't read the methodology. They segmented the data by occupation, and then examined the data per occupation by year.

Do you have a study that shows the adverse impact of H1-B visas on American workers?

> There is nothing scientific about this

Soft sciences are harder than hard sciences to do.

> No experiment was performed. > This is not a double blind experiment

Do you really think experiments are a reasonable approach to political questions?

I too dislike wish-washy science, but there really isn’t much other choice for many real world situations.

Someone above accused him/her being unscientific and they responded accordingly.

I think an equally big problem is that a lot of "scientific evidence" like this doesn't pass sanity checks, and definitions of "net economic benefit" can vary and might not preclude strongly negative impacts to the average US citizen.