To sustain economic growth, the United States must double its overall research effort every 13 years. That’s because it is taking an increasing amount of effort to generate enough ideas to power the economy, according to research by Stanford’s Nicholas Bloom and Charles I. Jones, Stanford PhD candidate Michael Webb, and MIT’s John Van Reenen.
Current research efforts in the US are 23 times greater than they were in the 1930s, having increased 4 percent a year in the decades since, the researchers write. However, research productivity—defined as the average amount of new research produced per individual researcher—has declined by 5 percent a year in the same time frame—halving every 13 years, according to Bloom, Van Reenen, Jones, and Webb.
The findings challenge economists’ widely held assumption that a constant number of researchers can generate enough new ideas to ensure constant growth. Many US companies have increased the number of researchers they employ to compensate for falling research productivity, and if current trends continue, companies and organizations will need to hire even more researchers to keep pace going into the future.
While Bloom, Van Reenen, Jones, and Webb find an economy-wide decline in research productivity, they focus on computing, agriculture, and health care to highlight what’s happening on a more granular level.
It would require 15 times more researchers today than 30 years ago to produce the same growth, according to the study.
Take computer chips: according to Moore’s law, the density of computer chips doubles every two years. But keeping up this pace requires more than 18 times the number of researchers today as in the early 1970s, reflecting declines in productivity, according to the study.
In the agricultural field, since 1969, aggregate yields of corn, soybeans, cotton, and wheat have doubled. And yet the study finds that research productivity in agriculture has declined about 5 percent a year. To offset the drop in research productivity, the number of researchers has had to expand by almost 24-fold for corn and soybeans and by six- to 10-fold for cotton and wheat to keep apace.
In health care, average life expectancy has climbed by about two years each decade since 1950. As research productivity declined about 5 percent each year, though, the effective number of researchers had to increase almost 16-fold to produce these gains in health outcomes, the researchers find.
At 15,000 companies across the US economy, the number of researchers has risen between 2 percent and 5 percent a year, the authors find. At the same time, research productivity has declined by 9 percent annually. Thus it would require 15 times more researchers today than 30 years ago to produce the same growth, according to the study. Bloom, Van Reenen, Jones, and Webb conclude that “declines in research productivity must be offset by increased research effort.”