The consequences of unskewed U.S. business growth

Photo by iofoto, veer.com

Over the past couple of years, economists, analysts, and policymakers have started paying more attention to declining business dynamism—the rate at which new American businesses are started—in the United States. This decline has come among a variety of other downturns in the creation, destruction, and growth of American businesses.

The consequences of these trends can be pretty unsettling for anyone concerned about the growth of the U.S. economy—and we now have new concerns to add to that list. Recent research shows that the fastest-growing firms that were at the root of U.S. job creation in recent decades have slowed down since 2000. And this trend is very much present in the high-tech, “Silicon Valley” industries that are often heralded as the future of the U.S. economy.

This new research comes from a National Bureau of Economics Research working paper released yesterday by economists Ryan A. Decker of the Federal Reserve Board, John Haltiwanger of the University of Maryland, and Ron S. Jarmin and Javier Miranda of the U.S. Census Bureau. The paper covers a wide ground, but its main contribution is focusing on the changing distribution of growth among U.S. businesses. Decker and his co-authors examine how the growth rate of the firm at the 90^th percentile of job growth has changed over time compared to the growth rate of the median firm (in the exact middle of the distribution) and the firm at the 10^th percentile.

For several decades, the distribution of firm growth was positively skewed—the gap between the growth rates for the 90^th percentile and the median was larger than the gap between the median and the 10^th percentile. In other words, the firms leading in job growth were far ahead of the pack. But according to Decker and his team, in 1999, the 90^th percentile to 50^th percentile gap was 16 percent larger than the 50^th to 10^th gap. By 2007, the last year before the Great Recession, this gap was only 4 percent. And by 2011, the last year under study, essentially all the skewness was gone.

Furthermore, the decline in the gap between the 90^th percentile and the median was due to a large decline in the speed at which firms at the 90^th percentile grew. The “gazelle” firms don’t run as quickly as they once did. The decline is partly due to a declining number of startups, but it’s also because the young firms that do exist aren’t growing as fast as previous startups did. We have fewer startups, and those that exist grow slower. The result is that, since 2000, the U.S. economy is getting fewer jobs from startups and fast-growing new firms.

It’s possible that this trend could exist only in one sector or one group of them. But Decker and his co-authors show that the trend is prevalent in many firms, including the kind of high-tech firms that exist in the vaunted Silicon Valley. At the same time, the rate of growth for leading firms among those that are publicly listed has also been on the decline. This has happened as the number of public firms has been on the decline in the United States.

The ramifications of these trends are fairly significant. Young, fast-growing firms have long been a source of job growth and productivity growth in the United States—and slower job growth is clearly something the United States could do without. The declining growth rate of leading firms might also explain, in part, the collapse of the job ladder. If there are fewer jobs at faster-growing firms that need to hire workers, that would reduce the amount of job-to-job transitions. And other research has shown how a greater share of older firms in the economy could help explain the jobless recoveries of the 21^st century.

The declining dynamism of U.S. firms was apparent before the release of this new paper, but now its severity is even starker. The root of this trend and others related to it, however, is not well understood. Getting to the core of this slowdown will take time, effort, and research.