Tom Friedman, in his Friday's NY Times column (subscription required) comments on Walter Isaacson's new biography of Einstein and asks:

"If Einstein were alive today and learned science the boring way it is taught in so many U.S. schools, wouldn’t he have ended up at a Wall Street hedge fund rather than developing theories of relativity for a Nobel Prize?"

While our schools can undoubtedly use improvement in their science curricula, there is a more important reason we're not training as many scientists in this country as we should - science just doesn't pay off, at least not until the scientist has reached an age when most other professionals in the US are thinking about how much to put in their 401Ks instead of worrying about where to find their first job. If Einstein were alive today, he'd have good reason for choosing to work at that hedge fund over the spartan life of a young scientist struggling to get a tenure-track job.

By the time you start making a reasonable salary as a scientist, you've watched for years as all your friends who are MBA graduates, computer engineers, physicians, lawyers, or even moderately paid teachers buy houses, take family vacations, buy their kids music, sports, and dance lessons, and start a substantial retirement savings - all while you were just barely getting by from paycheck to paycheck. Scientists begin saving for retirement and their kids' education at an age which financial self-help books tend describe, with half-hearted optimism, in phrases like 'not completely hopeless.'

For a science career, you live in poverty for 5-7 years of grad school, with no retirement benefits and poor health insurance. That's somewhat understandable - law students and med students also live in poverty during their graduate training (although a PhD program is longer, and senior science grad students are already working at a professional level that lawyers and physicians don't reach until after they graduate).

After grad school, a postdoctoral fellowship is absolutely necessary for any kind of academic career at a research university or in a senior position as an industrial scientist - in other words, the kind of permanent job that induces people to seek a PhD in the first place. Postdoctoral fellowships usually run at least three years now, with a salary in the 32-45k range (you maybe reach 45k after 2-3 years, if you have an outside source of funding). As in grad school, there are no 401Ks or other retirement benefits, and even when you do get decent health insurance, all the money for the premiums (including the 'employer' contribution) is considered taxable income through some perverted loophole in the tax code.

Scientists tend to be in their mid- to late 30's by the time they get a permanent job, and on average they don't get their first research grant (critical for establishing an independent research program) until age 42. This is extremely late for people who want at least some financial security before they start a family. Women scientists are disproportionately affected, because the brunt of the work of having children, biologically and usually socially, falls on women. Those who hope to have a family life are faced with the unpleasant choice of waiting a long time or risking their job and financial security to have children at a more normal age.

The whole process can be extremely discouraging and demoralizing. I have a PhD, more than 8 years of experience in the lab, several published papers, my own funding based on a successful, peer-reviewed fellowship proposal, and I earn less than my younger sister, a clothing store manager who successfully worked up from an entry level job in the two years she's been out of college. I work nights and weekends with no overtime pay. Instead of recording overtime, I fill out a time sheet at the end of the month stating my "hours absent" from the university. I have school debt that's accumulating interest, while my kids rely on grandparents to buy them shoes and swimming lessons.

The National Postdoctoral Association
puts it this way:

"Postdoctoral scientists have slipped between the cracks of the scientific workforce as a heterogeneous group of 'apprentice' scientists. They generally do not have well-defined expectations of employment, appropriate employment rights and responsibilities, commensurate or even normalized pay scales, performance evaluations, consistent employment benefits such as proper health care, pensions, occupational health insurance, or procedures for resolving conflict."

We could have the best public school science curricula in the world, but that would not be enough to overcome the large disincentives students face when considering a career in science.