It's Not As Bad As It Sounds 

Newsweek has an interesting article on how other countries are out pacing the U.S. in graduating people in technical fields. However, I find their analysis to be pretty flawed given my personal experience. First a few choice tidbits from the article:

As late as 1975, the United States graduated more engineering and scientific PhDs than Europe and more than three times as many as all of Asia, reports Harvard University economist Richard Freeman in a recent paper. No more. The European Union now graduates about 50 percent more, and Asia is slightly ahead of us. By Freeman's estimates, China has reached almost half the U.S. total and will easily overtake us by 2010. Among engineers with bachelor's degrees, the gaps are already huge. In 2001 China graduated 220,000 engineers, against about 60,000 for the United States, the National Science Foundation reports.

Not to belittle the accomplishment and importance of PhDs and Engineers (of which I'm one), they're hardly the only technical brains out there, especially in the software industry. In my relatively short career in fact, as an engineer I've been in the minority of those writing software. Most people have had computer science degrees (which is not counted as an engineering degree by most), or they've had a bachelor's degree in a non-technical field and picked up a programming book.

If you get into the mainframe world, it's even more diverse. Many mainframers have business or simply math degrees, not any real technical degree. Some don't even have a bachelor's degree. This is not meant as an insult. I find my engineering degree to have provided me with a lot of knowledge and best practices that many without that degree are lacking. But many skills can be learned outside of school.

I've also met a lot of people who come out of some Asian engineering programs and don't find myself to impressed. To a certain extent, many of these schools (especially in India) are more like factories trying to spit out as many as possible. Looking purely at the numbers of people with pieces of paper that say "Engineer" or "Doctorate", without looking at the quality of the corresponding coursework is pretty silly.

Freeman also documents a second worrisome reality: U.S. scientists and engineers aren't well paid, considering their skills and - especially for PhDs - the required time for a degree. This means, Freeman says, that "the job market... is too weak to attract increasing numbers of U.S. students." Consider some pay comparisons. From 1990 to 2000, average incomes for engineering PhDs increased from $65,000 to $91,000, up 41 percent; PhDs in natural sciences (physics, chemistry) rose from $56,000 to $73,000, up 30 percent. Meanwhile, average doctors' incomes increased from $99,000 to $156,000, up 58 percent; and lawyers went from $77,000 to $115,000, up 49 percent.

The author's of the article seemed to have missed the obvious connection between the first quote I have, and this one further down. For a long time the United States was cranking out a lot of people with various skill sets (whether degreed or not)... and now there's a huge number fighting for a smaller number of positions. When that happens, the salary goes down. That's basic economics... supply and demand. If one looks at the numbers especially in light of the dot com bubble bursting, then it should be no surprise. The lower salary then acts as a regulator making sure that the influx doesn't grow even further.

If one examines these numbers in light of the fact that one need not have a PhD or Engineering degree in order to do a technical job, then the numbers make even more sense. While a lot of company's specifically look for the degrees during hiring, may simply don't care, and look a lot more at experience, especially if it means forking out less money. It's also bad practice to compare engineers with medical doctors or lawyers, because you have to have a medical degree to practice medicine, and a law degree to pass the bar. No such constraint exists for most technical positions.

All in all, it would take better numbers an analysis to convince me of a real problem.