Not only is the U.S. losing ground in high technology exports, but its very capacity to develop new technologies is declining rapidly with respect to the rest of the world. According to Richard Freeman, the paper's author, the sheer population of Asian countries may allow them to train more scientists and engineers than the U.S. while devoting a smaller share of their economy to science and technology.
For a recipe on how not to gain a technological edge over your competitors, read this post: "Lessons from Soviet Science," which talks about how the Soviets failed to properly prioritize their research dollars, resulting in a massive number of institutes devoted to power engineering: A field considered vital in the 1930s when the Soviets drew up their strategy for developing science and technology.
Since results are so temporally divorced from effort (Fourier again), it is extremely hard for bureaucracies to determine which basic science projects deserve funding. So resources generally go either to projects with political significance (such as power engineering) or to large projects. It’s the “we have big money, let’s do something big with it” syndrome. The Dynamist has a nice reprint from Freeman Dyson that describes this latter phenomenon. Let’s take a look at the example of the Zelenchukskaya Observatory he picked from the USSR:One of the factors which the committee planning the observatory did not worry about was the Zelenchukskaya weather. I was on the mountain for three nights and did not see the sky. Even at Mount Palomar one may be unlucky and run into a string of cloudy nights. But at Zelenchukskaya the weather is consistently bad for the greater part of each year.