Funding for scientific research has become risk-averse, favoring established researchers pursuing incremental discoveries, critics say. From a report: In August President Biden signed into law a massive boost to scientific research: roughly $200 billion, when fully funded, over the next several years. The bipartisan law’s premise is straightforward. In the long run economic growth depends on innovation, which in turn flows from how much we spend on research and development. The problem with that premise is the U.S. already spends plenty on R&D, about 3% of gross domestic product in recent years. That exceeds the peaks of the early 1960s during the space race. But the fruits of that effort have been elusive. Total factor productivity, the best metric for innovation’s contribution to growth, grew just 0.5% a year over the past decade, half its rate of the 1960s. Other evidence suggests breakthroughs, such as new drug discoveries, are becoming more expensive and time-consuming, and researchers spend ever more years acquiring training and experience before achieving novel discoveries. In other words, the U.S. is getting progressively less bang for its scientific buck.
Patrick Collison, the Irish-born co-founder of payments technology company Stripe, has spent a lot of the past five years pondering the problem of declining scientific productivity. One popular theory he rejects: the most important discoveries have been made and new insights are more incremental and difficult. That might be true of individual fields, such as physics, but can’t be true generally, he argues. “Every discovery opens up new frontiers and new questions and new possible avenues of investigation,” he said in an interview. “The surface area of our knowledge is continually expanding. Just like we are not exhausting the creative possibilities of the English language, because words can be combined in an infinite number of ways, it does not seem to me we are running out of possibilities of discovery in computer science.”
Research inputs — dollars and hours spent — are a lousy measure of innovation, he added. It doesn’t require more resources to make a hit movie or bestselling novel than a dud. The most innovative companies aren’t necessarily the biggest spenders. Tesla’s R&D spending in the past three years is 13% of that of General Motors and Ford Motor combined. Yet Tesla’s market value is more than five times its two rivals.’ Clearly, scientific productivity has something to do with how research is done, not how much. One culprit, in the view of Mr. Collison and many others, is that the institutions that fund science have become process-oriented, narrow-minded and risk-averse. Wary of failure, they favor established researchers pursuing narrowly focused, incremental ideas over younger scientists with more heterodox agendas.