This is actually a fantastic experiment to evaluate whether people are employed on H1-Bs because of a shortage which can't be met locally, or because indentured

This is actually a fantastic experiment to evaluate whether people are employed on H1-Bs because of a shortage which can't be met locally, or because indentured workers are cheaper. Ideally there would be another cohort, allocated randomly, where the employee is not tied to any one employer, like a temporary green card. It is fallacious to say that with decreased supply, demand will dictate wages will go up. It is quite possible that demand is elastic at a particular price point, but at higher prices demand falls to zero (aka set up an office in Canada). There is no reason to assume linearity in any of this. What's missing is that, even with the potential for high wages after graduation, it is still a big risk for an individual to borrow for a college education. These employers should be taxed, perhaps 10% of their H1-B expenditure, with that tax subsidising education in that field.