The science policy-makers are restless. They want more American scientists and engineers.

A recent Op Ed piece from the National Academies called for more. Presenters at the recent Government-University-Industry Research Roundtable (GUIRR) Pan-Organizational Summit on the U.S. Science and Engineering Workforce, representing many organizations, called for more. The National Science Board (NSB) task force on national workforce policy (NWP) seems poised to do the same.

Yet, every faculty position at a research university or well-regarded liberal arts college draws hundreds of applications, and many young scientists are stuck in adjunct positions or seemingly endless postdocs. Can these national organizations really be claiming that we need to produce more scientists when the ones we already have are underemployed? What can they mean by "shortage"?

In their report presented to the GUIRR meeting, William P. Butz and his colleagues at the RAND Science and Technology Institute ask, "What would a 'shortage' of scientists and engineers look like?" and provide five possible answers. Here are the first four:

  • A historical decline in production,
  • A reduction in "market share" relative to "competitors" (i.e., other countries),
  • A failure of producers to meet their own expectations,
  • A failure of producers to meet the national need.

"Each of these concepts of shortage has a place," the report asserts, but each notion of "shortage" has a different solution.

So which kind of shortage do policy-makers claim we have? Arguments vary, but the most commonly invoked rationale for increasing the supply of U.S. scientists is number four: a failure to meet the national need--as perceived, that is, by the policy-makers themselves.

At the core of the policy-makers' argument is that the U.S. science and engineering (S&E) labor force is too dependent on foreign scientists. They have a point. By conservative estimates, 10% of U.S. residents holding S&E baccalaureate degrees were born abroad. Twenty percent of resident master's degree recipients and more than a quarter of S&E Ph.D.s were born abroad. And the production of domestic S&E Ph.D.s would have declined precipitously over the past few decades were it not for the ramping up of female and minority Ph.D. production.

Why is this a problem? Because it takes a long time to train an S&E labor force, and our dependence on foreign S&E workers makes us economically vulnerable to changes that we as a nation can't control. The foreign S&E labor force is unstable, the argument goes, so we need to reduce our dependence.

"If other countries and regions build up their indigenous S&E capabilities," reads the latest version of the National Science Board's Science and Technology Indicators, "they may diminish the relative attractiveness of the United States as a destination country." Though difficult to quantify, the report notes, "anecdotes suggest that experienced scientists and engineers, particularly those originally from Asia, are even now returning to their native countries." The U.S. is also experiencing increased competition for international scientists from other nations. Furthermore, "traditional donor countries may be able to moderate the outflow of their scientists and engineers," thereby decreasing the influx of scientists into this country. The report also notes that the aging of the workforce in certain countries (Germany and Japan are mentioned) will enhance demand for S&E workers in those countries in the coming years. And if the supply of foreign scientists and engineers dries up, the U.S. is likely to find itself with an acute shortage.

Another argument often made is that much of the research that will be done in the coming years will be "sensitive"--and that sensitive research is best done by U.S. citizens. So something must be done, the argument goes, to enhance the U.S. supply.

According to some, there is another potential advantage in seeking to increase the supply of scientists and engineers. Many organizations, including the American Association for the Advancement of Science (publisher of Science's Next Wave), advocate domestic S&E labor expansion as a way of increasing the diversity of the S&E labor force. A serious effort to increase the pool of U.S. scientists and engineers, these folks say, presents a great opportunity to draw new and different faces into the sciences and engineering.

These goals--increased U.S. economic security and a more diverse S&E workforce--are hard to fault. But policy-makers need to think hard about whether an expansionist S&E labor policy is, in isolation, the best means to achieve those worthy ends.

Of the definitions of "shortage" presented in the RAND report, the fifth is the one most often associated with labor markets. There is a shortage, says this definition, "if production is not meeting market demand, as indicated by rising prices." When there is a real economic shortage, prices--wages, in this instance--rise. But in recent years, wages have actually declined in real terms, and unemployment rates, though still low, have increased. Although we may have an S&E labor shortage according to RAND's definitions 1 through 4, we do not have a shortage in the usual economic sense. The policy-makers I've spoken to acknowledge as much. To quote the RAND report, "Neither earnings patterns nor unemployment patterns give any indication of an S&E shortage."

In a climate where wages aren't rising, what are the likely consequences of an increase in the S&E labor supply? If we focus on increasing supply without somehow generating a comparable increase in demand (e.g., by increasing investment in scientific research positions), we will create more scientists than we can employ. Salaries will fall. Working conditions will deteriorate. And young scientists--those who have not yet managed to entrench themselves in the workforce--will be hurt the most, as Michael S. Teitelbaum of the Alfred P. Sloan Foundation noted in his report to the GUIRR Summit.

What will this do for the cause of diversity? Many of the policy-makers who are calling for workforce expansion admit that a key reason that certain groups are underrepresented in science is that the "effort-to-reward ratio" for entering science is so poor. The best and brightest women and African Americans and Native Americans (not to mention many of the best and brightest white men) can make more money with less uncertainty and aggravation by working in other fields. Increasing the supply of scientists will only make this situation worse, making it harder, not easier, to increase the diversity of the scientific workforce.

So what should we do? If we are determined to increase the domestic supply of scientists while at the same time improving domestic diversity, one answer is obvious but unpalatable: Restrict the number of foreign S&T workers who can enter the U.S. workforce.

But no one--including me--wants to do this. The NSB's NWP task force isn't even considering it. "That wasn't even on the table," said one committee member in a recent telephone interview. Imported scientists make valuable contributions to American science and to other aspects of American life. Indeed, presentations at the October and November NSB meeting indicate that the NWP report will recommend that securing well-trained, foreign-born S&E workers should remain a major U.S. policy objective.

If import quotas and restrictions are off the table, then what can we do?

We can, and should, focus on the demand side instead of--or in addition to--the supply side. If we want to produce more American scientists, we need to create more demand for them--by investing more in science and by providing incentives for the private sector to do the same. Simultaneously, we should work to create more faculty and permanent staff positions in university science departments and to disassemble the mechanisms--such as extended postdocs and other "training" phases--that artificially depress wages for professional scientists. If we can do this, wages will rise, employment conditions will improve, and we'll draw more Americans, including underrepresented Americans, into science.

There is, of course, a place for supply-side fixes. Particularly dynamic fields--for example, computer science and information technology--might benefit from efforts to recruit and train more (especially minority) scientists. But a supply-side policy that fails to distinguish between vibrant and moribund fields will channel gifted underrepresented minorities into other professions and further alienate the underemployed young scientists whose numbers these policies aim to increase.

On its face, it's an odd proposal: to address a labor shortage by increasing demand. But it has one potent advantage over the other proposals that seem to be on the table: It links "shortage" definition four--a failure to meet national need--and definition five--a failure to meet market demand. If we are to meet these two national needs simultaneously--an increase in the production of domestic S&E labor and an increase in the diversity of the domestic S&E labor pool--then we will have to harness market forces. As the RAND report notes, "As young people observe this tightening labor market and consider lifetime employment prospects ... some of them will opt for S&E, rather than for clinical medicine, law, business, or another profession. As these people complete their education and join the workforce, total production of scientists and engineers will accelerate. The shortage will diminish."

Jim Austin is the editor of Science Careers. @SciCareerEditor on Twitter