Yu Xie and Alexandra Killewald have good news for Americans. They have concluded that American science is NOT in decline. Their glass is more than half full, even brimming. There’s just one hitch. They really only looked at a small part of American science: mainly its human resources.
Graduates can be considered to be one output of America’s science establishment, but there are actually a few others that might even be more important–like discoveries from research and innovations from development. One can measure papers, patents, citations, prototypes and pilot plants, high-technology exports, and the investments that make these outputs possible. Those indicators of R&D are not nearly so favorable to the U.S.
The authors’ try to broaden their scope by citing a 2008 RAND report that largely based its findings on indicators from many years earlier. which I pointed out at the time. When your competitors’ indicators are increasing exponentially, it isn’t wise to use old data. Most notably, China has come out of nowhere with a skyrocketing challenge to the U.S. in many indicators of science and technology, as well as in business. This competition for market share extends to the placement of scientific papers in a fairly fixed number of slots in journals, explaining why American growth rates in publications tanked in recent years as they report in Chapter 2.
Revealingly, the authors divide those writing in this domain into two camps. They use the pejorative term “alarmist” to characterize those who think that American science is in decline, while they have no comparable term for the critics of the alarmists, like themselves. My thesaurus draws a blank for an antonym, but I might suggest “pollyannas.”
To be fair, the authors have provided a competent analysis of the limited domain of science education and jobs for graduates, and they do also cover some surveys of Americans’ attitude toward science. The NSF survey always reports that the public loves what it is doing. I wonder about that since so many of the American public seem to have swallowed a lot of denial propaganda about climate change, evolution, vaccines, and the age of the Earth.
I agree with the authors that there is no great shortage of American scientists–rather the opposite. Pay for scientists in the U.S. has not risen, as it would if there was really a shortage. Producing more would simply result in more underemployed post-docs. Systems engineers would recognize this as a classic problem. You have to find the bottleneck resource that is limiting overall performance of a system, since efforts to improve other resources will be wasted. While I understand that the authors want to look at the U.S. alone, learning from our competitors’ alternate universes can help with this identification. My stats show that China is surging ahead of us because it has been increasing real R&D investment by over 15% per year compared to our 3% or less. It takes money to do science today, and lots of it. Thus the bottleneck resource in the American science enterprise is R&D funding, not human resources. A book that is largely based on human resources can be a misleading guide to the question in its title.