I looked at the correlation between the relative supply of H-1B workers and the wages of IT workers. Surprisingly, I found no significant correlation between the presence of H-1B workers and labor market outcomes. However, surprisingly, if you look at the impact at the impact on the wages of only male workers, then there is a slight but a very clear (statistically significant) 'impact' on their earnings. Even more surprisingly, if you also look at the correlation between the earnings of female domestic workers and the relative supply of H-1B workers, then there is a POSITIVE impact on their earnings.
Of course, I concede that there could be a A LOT of problems with the methodology I used and with the data employed in this study. My methodology was basically constrained by whatever data I had access to. However, if we assume for a moment that the data and methodology were more or less reliable, then I suspect that what's happening is that the IT labor market is somewhat segregated by genders (someone needs to test this hypothesis). E.g. the female workers tend to be employed in occupations that are complimentary to occupations that are dominated by male workers (e.g. QA, testing, etc). If this assumption is correct, then the H-1B workers (who are predominantly male) might indeed depress the earnings of male domestic workers a little bit, but at the same time the increase supply of male workers boots the demand for occupations that tend to employ female IT workers. So, if you look at the overall effect on earnings, there is no relationship, but there is clearly something going on once you break down the earnings data by genders.
Slashdot Mirror
Last year, I took an undergraduate Labor Economics seminar and wrote a paper on H-1B program's impact on domestic labor force. You can find it here:
http://panic.berkeley.edu/~akopps/paper/paper.pdf
I looked at the correlation between the relative supply of H-1B workers and the wages of IT workers. Surprisingly, I found no significant correlation between the presence of H-1B workers and labor market outcomes. However, surprisingly, if you look at the impact at the impact on the wages of only male workers, then there is a slight but a very clear (statistically significant) 'impact' on their earnings. Even more surprisingly, if you also look at the correlation between the earnings of female domestic workers and the relative supply of H-1B workers, then there is a POSITIVE impact on their earnings.
Of course, I concede that there could be a A LOT of problems with the methodology I used and with the data employed in this study. My methodology was basically constrained by whatever data I had access to. However, if we assume for a moment that the data and methodology were more or less reliable, then I suspect that what's happening is that the IT labor market is somewhat segregated by genders (someone needs to test this hypothesis). E.g. the female workers tend to be employed in occupations that are complimentary to occupations that are dominated by male workers (e.g. QA, testing, etc). If this assumption is correct, then the H-1B workers (who are predominantly male) might indeed depress the earnings of male domestic workers a little bit, but at the same time the increase supply of male workers boots the demand for occupations that tend to employ female IT workers. So, if you look at the overall effect on earnings, there is no relationship, but there is clearly something going on once you break down the earnings data by genders.