Productivity isn’t just a function of innovation. New ideas and technology have the capacity to greatly boost the productive capacity of an economy, but those forces are for naught if new ways of doing things are not spread throughout the economy or if resources don’t flow to more productive companies. The benefits of such “allocative efficiency” is one reason why some researchers and policymakers are concerned about a less dynamic U.S. economy. If fewer companies are being started and workers are less likely to switch jobs, then productivity might take a hit. New research suggests that this is happening.

The paper is from Ryan Decker of the Board of Governors of the Federal Reserve System, John Halitwanger of the University of Maryland, and Ron Jarmin and Javier Miranda, both of the U.S. Census Bureau. The four economists use firm-level data on productivity to decompose (breakout into constituent parts) several trends in U.S. labor productivity growth. The decomposition roughly divides productivity growth into three buckets. The “within-firm” bucket captures productivity growth for existing firms. The second bucket indicates the change in productivity growth due to reallocation between existing firms. And the third bucket shows how much of productivity growth is due to the gains from firms entering the economy.

What the four economists find that is that productivity growth among existing firms isn’t the primary driver of the drop-off in overall productivity growth that we’ve seen since the turn of the century. The biggest decline among the three factors is in the second one, indicating a decline in reallocation between existing firms. In other words, employment and economic output are not flowing to the more productive firms. The third bucket also contributed to the post-2000 decline in productivity growth, as the net entry of new businesses also was on the decline, indicative of the potential impact of the declining start-up rate in the United States. The authors do note that the first factor—within-firm productivity—may also be contributing to some extent as productivity growth among high-productivity firms has declined.

As the authors note, many explanations of declining productivity growth popular among policymakers, the media, and some economists focus on slowing technological advancement or the mismeasurement of new technologies. This new decomposition points instead at declines in business dynamism as a possibly more fruitful avenue for getting a grip on why U.S. productivity growth has been slowing. The research by these four economists provides far from a definitive answer, yet the paper suggests a refocusing on research and conversation in this area that might be, well, very productive.