The naive interpretation is that these companies would pay higher wages without access to H1B workers.
Studies like TFA suggest that is not actually true. Companies in all sectors are constrained by the availability of skilled labor and to some extent are paying what they can afford.
Reducing the number of H1B visas would likely hurt employers to the extent that their workers would suffer. It is not sensible to pretend that all of those jobs would still exist if you cut the H1B program.
One major issue I did not touch on above is that it is unequivocally true that immigration grows GDP. What is not unequivocal is that immigration grows GDP per capita, and I think this matter a lot more than the top line figure. American immigrants usually grow GDP per capita, but it depends on the migrant group and reason for migration. Refugees are usually a net loss, for example. To point: most of the gains in GDP growth have been accrued to the top 1-10% over the last few decades. Meaning that high immigration is great for business and asset owners. As someone who is displaced by immigration, unless you own a lot of assets, you will bear the cost of immigration without experiencing any of the gains.
This is the premise of a fascinating paper by [Eric Weinstein from 2001.](https://www.ilo.org/publications/migration-benefit-all-towar...) He proposed an "immigrant dividend." Since the costs are usually borne by the poor, and the gains usually accrued to the rich, there should be some mechanism to rebalance this. His proposal is a fee which companies need to pay to the government for hiring an immigrant. For the sake of discussion, let's say this is $1M. This money is then used to compensate the people most affected. It could be money split evenly between everyone, or targeted for those most impacted.
This aligns multiple negative incentives with public good. For starters, there is a widespread belief that companies like Microsoft regularly fire (or fail to hire) good local employees because they can hire much cheaper (and desperate) foreign workers, who will not complain if they are asked to work 70 hour weeks. Secondly, it compensates people affected by these practises, so they can be made whole. Third, it forces companies to only hire immigrants when there is a genuine skills shortage. Fourth, it encourages companies to upskill employees (this is a major issue right now) - or at least partake in improving education across the nation.
If companies were forced to pay an immigrant dividend, I think we would immediately discover that very few of those "genuine" skill shortages were real, and instead it was really a shortage of people willing to work for low pay and poor conditions.
The question is whether jobs in low-margin industries would exist at all if the immigrant workers weren't there to do it. It's not cut and dried, but the research tends to suggest that cutting immigration would hurt both employers and the workforce.
My last employer (manufacturing sector) flat out refused to consider paying $100,000 for a non-resident H1B. Employers that did fork out were likely hiring top level or very scarce talent. An "immigrant dividend" paid by employers would simply restrict immigration to highly paid work.
As it is, H1B visas already go primarily to highly paid (mostly tech) workers. The problem is that most immigrants arrive in the US by different avenues. Who pays for spouses and dependents? Asylum seekers?
As for Microsoft: it is not a low-margin business. If Microsoft is suppressing labor costs to juice profits, that is the sign of a company that does not know how to use the talent it has at its disposal.
What happens when immigrants force people out of the highest-skilled job tiers, and there is no higher tier to upskill into? Here in the US, with the H1-B program, we're being told that these immigrants are only for the very highest levels of skill, which the American workforce lacks.