As we’ve discussed on many occasions, the primary driver of economic well being in the rest of the 21st century will be innovation. For more than a century, the United States has been the world’s most vibrant economy — largely because it has been the consistent leader in scientific research and innovation. Regardless of whether you measured this innovation in terms of scientific papers published, or the number of patents issued, the U.S. always left the competition in the dust.

However, our lopsided advantage seems to be giving way to foreign competition. Americans are still getting plenty of patents, but their percentage of the total patents issued each year is dropping. Foreign applicants, particularly those from Asian countries, are starting to take over the lead in innovation in some fields. According to the New York Times, the United States’ share of its own industrial patents has declined to 52 percent.

The data on research papers show an even more startling shift. In 1983, Americans wrote 61 percent of the research papers that were published in scientific journals, according to a study by Physical Review. In just two decades, the percentage fell to 29 percent. That conclusion is supported in a study by the European Commission, which announced in 2003 that Europe took over the lead from the U.S. in producing scientific papers.

The statistics from the world of higher education are just as bleak. The number of freshly minted doctorates in the sciences fell 5 percent in just a single year, from 1998 to 1999, according to the National Science Foundation. That translates to 1,300 fewer American scientists working in U.S. corporations’ research labs or starting their own businesses.

The numbers tell us what is happening. But why does the United States appear to be losing its undisputed dominance in the sciences? Ever since the number of home-grown scientists and engineering graduates peaked in 1970, we’ve seen a decline in the number of Americans entering these fields. Some attribute it to poor preparation before college. However, a more likely driver seems to be relatively poor economic return on a doctorate in the sciences.

After spending four years in undergraduate school, four years in a graduate program, and another two or three years in a post-doctoral program, the recipient of a...