Conference on Entrepreneurship and Economic Growth

Supported by the Ewing Marion Kauffman Foundation
Manuel Adelino and David T. Robinson, Organizers
October 14-15, 2016

The Startup Deficit and Aggregate Productivity Growth

Aggregate productivity growth has slowed dramatically in the United States since 2005. The tight link between innovation and entrepreneurship in many models of economic growth suggests that frictions that impede entry may have long lasting effects on productivity growth. Recent work documenting declines in measures of business dynamism, the gross entry rate among them, is consistent with this view.

Titan Alon, David Berger, Robert Dent and Benjamin Pugsley examine whether, and to what extent, the relatively recent decline in productivity growth is a consequence of a longer-standing decline in the gross business entry rate since the 1980s. They analyze data spanning periods of both high productivity growth (1996-2004) and low productivity growth (2005-2012).

The analysis yields several significant findings. First, they document that the lifecycle profile of labor productivity growth is indeed downward sloping and convex. The figure presents their estimated results and illustrates that young firms do indeed register productivity growth contributions substantially faster than older incumbents. Furthermore, the differentials in growth rates are substantial but converge quickly; while the youngest firms grow very quickly relative to older incumbents, nearly two-thirds of the effect is gone after five years and the effect is nearly gone after 10 years.

They document that selection and allocation effects operating on young firms, rather than within firm productivity improvements, primarily drive these lifecycle effects. Mean productivity growth of young firms is negligibly different, and perhaps even smaller, than older firms. In contrast, the contributions by reallocation and selection are about equally important and constitute the bulk of the contribution of younger firms to aggregate productivity growth beyond that of older firms. The analysis confirms that these lifecycle contributions for non-censored firms ages 1-19 are robust to several controls and remain stable over the researchers' estimation period and in the low-growth (1996-2004) and high-growth (2005-2012) sub-periods they study.

These results suggest that the lasting effect of the startup deficit has contributed a meaningful drag on aggregate productivity growth of around 0.50 percent. The drag has come primarily from decreases in the forces of selection and reallocation that act most strongly on younger firms, which now account for a smaller share of economic activity. Furthermore, the fact that most of these lifecycle contributions come from between-firm market forces (selection, allocation) rather than within-firm innovations suggests also that the broader decline in dynamism observed in the aggregate economy may itself be a lasting consequence of the startup deficit that began over three decades ago.

The estimates also suggest one more major source of productivity drag: a declining rate of allocative efficiency amongst the economies largest and oldest (over 20-year-old firms). In comparing the high-growth and low-growth sub-periods, the only part of the age profile contributions that changed was that made by the 20-plus firms. Further breaking down the effect shows that the decline was driven exclusively by a decline in the allocative efficiency of these older, larger firms and more than offset modest within-firm productivity gains by this group. In other words, the most productive older firms have not gained market share at the expense of less productive firms. Given the increasing share of economic activity of this set of firms, the researchers find that this in fact was the largest drag on productivity growth registered around the productivity slowdown.

NBER Videos

National Bureau of Economic Research, 1050 Massachusetts Ave., Cambridge, MA 02138; 617-868-3900; email:

Contact Us