Think of rents in Sunnyvale: $2K per 1b/1b. Still people save, even on $65K salary, because one lives with 4 other roomies.
Maybe we can make it a policy that companies make such information public when they employ more than N number of H1-B holders. With mean, median, standard deviation and error bars, you should be able to get an idea of whether or not this is happening.
Besides these numbers for salary, it would also be interesting to see the same statistics for years of experience programming of domestic labor versus imported labor. I would expect to see a significantly greater mean and median years of experience among H1-B holders versus domestically sourced talent.
For example, the first petition I clicked on was very fishy: https://icert.doleta.gov/index.cfm?event=ehLCJRExternal.dspC...
$96k-116k offered for an engineering position that requires a master's degree and where the prevailing wage is quoted at $92k!?
I want to see real numbers her.
For example, there are loopholes in both salary-calculation and also in terms of crafting hideously specific (and not entirely honest) requirements and then claiming nobody local can fill them.
If you cite a job with high requirements and a relatively low associated wage, you're essentially describing a job that only foreign talent is going to be able and willing to fill. I don't know a single person that meets the criteria for that position and isn't earning way more already. What self-respecting domestic candidate with the freedom to change jobs freely and that knows his worth would ever apply for such a position?
http://www.uscis.gov/eir/visa-guide/h-1b-specialty-occupatio...
Thus, when I hear people complain of companies who hire H1Bs only because they want to pay them less, I always want to ask, "why haven't you reported them?"
Either they don't actually know that the companies are exclusively paying H1Bs less, or they don't believe that the government will enforce the salary floor. Either way, raising the salary floor from 1x to 1.5x probably won't address their concerns.
edited for clarity