Titus digs up some tasty quotes from this discussion over at (ug) Megan McArdle's. Variously relevant to a number of threads that run through this blog. Each paragraph is a distinct excerpt, if that's not clear from the formatting:
I have a PhD from MIT in Chemistry, over half of my friends in a variety of disciplines left science for patent law, consulting, or finance because of the money. In my experience a rough average for my friends who stayed in their field was ~90-100K after 5-6 years of grad school. Patent Law (150K) and consulting (130K) start considerably higher. A career in medicine will generally have far better lifetime earnings, better geographic flexibility, and better job security as you age.
It's possible to get a half-decent-paying job out of college after spending 4 years of high school staring at the wall and 4 years of college drinking bong water and shotgunning six packs. Who would go through the relative hell of a science major when there are plenty of jobs for the reasonably intelligent yet unskilled?
A-grade engineers are unfortunately similar to Welsh longbowmen: devastatingly potent compared to their peers, but you have to start their training at age 10 or so. Simply upping the salaries of A-grade engineers won't magically create more of them. We know this, as we tried exactly that experiment in the boom.
In software development, the usual results of C-grade engineers is work that is of low quality, poorly tested, doesn't meet specifications, and is so poorly estimated as to take many times longer than expected. Sometimes it's so bad that it actually makes the code around it worse, through mere proximity (no, that's not an exaggeration). It takes very little of that in a typical software project to drive total costs above expected value. Because of that, it is not uncommon for C-grade software developers never to have actually shipped a system, because their projects were all shut down as business failures long before they were ready for production, or (much worse) yanked shortly after shipping to production. Hence, negative business value, as their salaries, benefits, and hiring costs have to be written off, as well as a good portion of the time of their co-workers to fix their messes.
The topic of "why H1Bs" has been done to death numerous times in forums beyond counting. I haven't yet seen a better explanation as that of ROI: becoming an A-type engineer as well as a hard scientist is a lifetime commitment. I like the Welsh longbowman analogy too; throw in Shao-Lin monks here as well. And what do you get out of this hard work, on average? A comfortable salary and engaging work, if you're lucky; engineers come out better on the salary front and scientists get more interesting work. Compare that with a career in sales or business management. Or, if you're not outgoing enough, in law. I'm sure the averages there look better and also there's a much likelier shot at real wealth. So why would someone -- who grew in this country, has a reasonably wide social network, speaks English natively -- invest more to get less?