To: H-1B/L-1/offshoring e-newsletter Sun Nov 17 08:06:08 PST 2013 I need to clarify my posting of yesterday, at http://heather.cs.ucdavis.edu/Archive/NewGradsAffected.txt First, I had referred to what I have called Type I and II salary savings that accrue to employers of H-1Bs: Hiring H-1Bs at lower wages than are paid to comparable Americans, and hiring younger H-1Bs in lieu of older Americans, to save money that way. Afterward, a reader asked (again) why I didn't mention what he called "Type 0"--the negative impact on wages of the swelling of the labor market under the H-1B program. Actually, I've brought this up many times, and have often referred to the (unquantified) statement in the 2001 NRC report that there is a "Type 0" effect. However, the question at hand in my posting yesterday, is a little different, which I should have clarified then. My posting concerned a Computerworld article reporting that an increase in the H-1B cap would lower wages for new American grads. I downplayed the issue, saying that the job market for new computer science graduates is quite good these days. That would seem to beg the question, so let me go into more detail on this. As you know, the computer-related H-1Bs fall into two main categories--foreign students hired from U.S. university campuses, and workers brought in directly from abroad. Let's call these Category A and Category B, respectively. Note by the way that under the Senate bill that passed (but which the House does not want to take up), Category A is considered "good H-1B" while Category B is viewed as the bad part of the visa program, a distinction that I have strongly disputed, including in my posting yesterday. Though I don't have data on this, in my experience the primary group that competes with new American grads is Category A, not B. On one level, the competition between the American and foreign new grads is literal--both groups are signing up for the same on-campus employment interviews, attending the same campus job fairs, relying on the same faculty connections with industry, and so on. But it's also true on a more indirect level, but the Category B workers, though young, are not viewed as new grads; they generally have some experience, at least in the form of an intensive job-like training course in some specialized computer language or platform. Yes, there is still an impact on wages of new American new grads from Category B, but it is less direct, and I believe it would be small if the Senate bill were ever enacted. As I said yesterday, I believe the largest impact on Americans would be on the older U.S. workers. Well, then what about the impact of Category A on new U.S. grads? The answer centers on the Optional Practical Training feature of the F-1 student visa. Any STEM foreign student has the right to work for 29 months after graduation--almost 2.5 years. Contrary to President Obama's rhetoric, "We train the foreign students in our schools but then send them home after graduation," we DON'T send them home. I work with a lot of CS foreign students, and I've never seen a single instance of one returning home in spite of a wish to stay here. I read various foreign worker blogs, and haven't seen any instances there either. Surely such cases exist, but I believe they are rare. Accordingly, the effect of the Senate bill in terms of Category A would likely just be to transfer some workers from OPT to H-1B status. Employers who want a worker for more than 2.5 years would likely take advantage of the expanded H-1B program. This would be especially true of the employers who wish to sponsor their foreign workers for green cards, which as I've explained is typically not the benevolent act it would seem to be; many employers, including big, household name firms, find this to be highlight attractive as it renders the worker immobile, unable to jump ship to another firm. By the way, I mentioned yesterday that more than wages are at stake, in the sense that many American new grads are shunted into the "talking jobs," doing only semi-technical work such as customer support, software testing and the like, while the H-1Bs are assigned the technical work. Sure enough, another reader responded to my posting yesterday that students at the well-known university he is affiliated with had complained that the jobs they were being offered didn't make use of their training. To those of you out there who are economists, a job is a job (at least at the same wages), but some students are deeply disappointed if they don't get to do development work. Ironically, being a outgoing and articulate is a negative here, as you are flagged as being right for the talking jobs. But in terms of wages and employability, while I shouldn't have given the impression yesterday that the Senate bill would have no impact at all on the new American grads (I didn't actually say this, but it probably sounded that way), I do believe the impact would be primarily on the older Americans, age 35 and above. If we didn't have OPT, things might be quite different, with more direct impact on new American grads as well. One side remark about OPT: The extension to 29 months was made during the George W. Bush administration, and defended by the Obama administration when it was challenged in court. Bush instituted this 29-month policy by executive fiat, rather than by going through Congress, an action of highly questionable legality, as the statute had specified 12 months. (The challengers, the Programmers Guild, lost in court anyway.) So I find it ironic that the Republicans are currently saying that Obama is operating under questionable legality in allowing, by executive fiat, health insurers to extend existing, substandard policies for a year. A bit hypocritical, I'd say. Norm Archived at http://heather.cs.ucdavis.edu/Archive/NoOneGoesHome.txt