Date: Sun, 22 Jan 2012 22:47:39 -0800 From: Norm Matloff To: Norm Matloff Subject: more on PPIC article To: H-1B/L-1/offshoring e-newsletter As I mentioned earlier today, I don't have time right now to post long analyses, but I did want to correct and amplify my January 13 posting of the recent Lofstrom and Hayes study at PPIC; see http://heather.cs.ucdavis.edu/Archive/LofstromHayes.txt for what I wrote then. Before beginning, it's crucial that I review my frequent terms, Type I and Type II salary savings. Type I savings are incurred when employers hire H-1Bs at lower wages than comparable Americans. Type II savings are incurred by hiring younger, thus cheaper, H-1Bs instead of older, thus more expensive, Americans. I believe that almost every reader of this e-newsletter is well aware of the Type I/II distinction that I make. Most readers should also recall, at least when I bring it up, that I've said repeatedly that I believe that Type II, the age-related salary savings, is the more important of the two types, the place where the big money is to be saved. Indeed, I've often said, "H-1B is fundamentally about AGE." Unfortunately, that is often forgotten, and people focus on Type I savings, much to my chagrin. By presenting age data regarding the H-1Bs, the PPIC paper does confirm my point about Type II. (It was hardly the first to do so, though. See last year's GAO study, and my earlier work, to name two earlier sources on age distribution of the H-1Bs.) But again, the PPIC authors are completely focused on Type I, so let's discuss that. The key word in my above statment, "Type I savings are incurred when employers hire H-1Bs at lower wages than comparable Americans," is COMPARABLE. In assessing the Type I savings issue, it is crucial to correct for variables such as age, education, job type and so on, so that we compare likes to likes. In my January 13 analysis, I had asserted that the PPIC student did not correct for job type within IT. However, I had overlooked a line in their IT analysis table, showing that they did in fact have some job type variables. Nevertheless, the fact remains that they still did include in their analyses ANY worker in the very broad "IT" area, including for technician types of jobs. That point is important, as H-1Bs are generally not hired into technician jobs. And since those pay less, there is the danger that lumping everyone together like this would artificially make the H-1Bs look like they are paid more, relative to Americans. This should be the case even though the PPIC study does have some variables for different IT job types, because of the skills issue, for instance. As most of you know, the industry lobbyists claim that employers hire H-1Bs because they have special skill sets, which in my experience is largely true. (Though I've also written extensively that the employers are misguided in emphasizing skill sets so much.) These skill sets command a premium wage in the open market. Today, for instance, a hot skill is Android programming, and employers pay a salary premium for it. A quick check of the Web just now, for instance, turned up http://startupcafe.co.uk/2011/10/30/silicon-milkroundabout-the-jobs-fair-matching-developers-with-startups/ which quoted a 20% premium at the new grad level. Having several hot skills in combination--seen commonly in job ads--commands an even greater premium. The problem is that data on skill sets is not available, and thus such variables are not included in anyone's H-1B analysis. For those who find that H-1Bs are underpaid, such as me, this means that the problem is even worse than what their findings indicate; wrong magnitude, but correct algebraic sign. But for those whose analysis actually finds that the H-1Bs are paid slightly more than Americans, say 5-6% more as found by PPIC, their findings would actually indicate that the H-1Bs are paid LESS than COMPARABLE Americans, once the skill set premium is accounted for. Again, I've made this point before, e.g. in my analysis of the Mithas and Lucas paper, It should be noted that although Mithas and Lucas repeatedly say that their data measures skill sets, it definitely does not. What they are actually referring to is the number of years of IT experience. This is a key point, because real skill sets command much more of a salary premium than the 2 or 3% ML found. Remember, the industry claims that the H-1Bs are hired primarily because they have special "hot" skill sets. (ML say this too.) These skills generally command a salary premium of 15-25% (see data in my University of Michigan Journal of Law Reform article), so even if one were to take Mithas and Lucas' 2-3% as accurate, that would indicate that H-1Bs are subject to net UNDERPAYMENT relative to Americans. (See http://heather.cs.ucdavis.edu/Archive/MithasLucasPublished.txt Of course, the ML posting had myriad other serious problems, as I pointed out there.) In closing, I must reiterate a point I made in my original posting on the PPIC study: After they're hired, the H-1Bs have limited mobility in the labor market, and if they're being sponsored for a green card, they have essentially NO mobility. If you can't move around in the labor market, then you can't have employers bid against each other for your services. In other words, you can't swing as good a deal for yourself as you would if you were a citizen or permanent resident--so the H-1Bs are on average underpaid. The point is that fundamental economic theory (or common sense, if you will) shows that Type I salary savings is real. The H-1Bs ARE paid less than comparable Americans. If you want to know how MUCH they're underpaid, then you can try regression analyses and the like, which is fine, but you MUST understand the effects of the missing variables. Norm