June 18, 2008 — Creative Destruction, Documented

A new study summarizes the churning of the U.S. economy and reveals its importance to productivity growth
By Doug Campbell

In North Carolina, the unemployment rate rose 0.2 percentage points between March and April to 5.4 percent. The number of unemployed workers rose by 7,882 and the total number of jobs in the state fell by 14,700, according to the Employment Security Commission of North Carolina.

But at the same time, the number of employed workers in North Carolina rose by 3,306. As ever, new jobs are being created even as old jobs are going away; in North Carolina's case, the old jobs tend to be in manufacturing and the new ones in services.

This is small solace to displaced workers, but the process of "creative destruction" is capitalism's dominant feature — one that can cause wrenching change while, over the long run, promoting breathtaking innovation. In fact, churning has long been thought of as a key to improving productivity, which in turn is essential to economic growth.

Joseph Schumpeter used the term "creative destruction" in his 1942 book, Capitalism, Socialism and Democracy, arguing that the capitalistic process "incessantly revolutionizes from within, incessantly destroying the old one."

Now comes a new study that documents the benefits of creative destruction. In "Turmoil and Growth: Young Businesses, Economic Churning, and Productivity Gains," economists with the University of Maryland, the Census Bureau, and the University of Chicago summarize their years of research on job churning. The authors built a longitudinal database — one that follows firms over time — to investigate the depth and breadth of the creative destruction process as few economists have done before. The Ewing Marion Kauffman Foundation funded development of the database as well as many of the authors' other papers.

Their key findings: Between 1977 and 2005 — a period that encompasses four recessions and several booms — the average annual number of jobs added to the economy as a percentage of total employment was almost 18 percent. Meanwhile, the job destruction rate was about 16 percent. In other words, the job creation rate was, on average, about 2 percentage points higher than the job destruction rate. Another way to put it is that net growth averaged 2 percent a year — but not because the "average" firm grew 2 percent; it was the result of the creative destruction process.

"The sheer magnitude of the restructuring and reallocation in the U.S. economy is quite remarkable," says John Haltiwanger, a University of Maryland economist and one of the "Turmoil and Growth" authors. "As soon as you see that you ask yourself — why does an economy have such a remarkable pace of churning? Because it's obviously costly to have some businesses in each and every period rapidly expanding while others are rapidly contracting."

The answer is that the creative destruction process is itself an engine of productivity growth. As Haltiwanger and his co-authors document, exiting businesses are less productive than surviving ones. Almost all of the productivity growth in the retail industry, for example, is attributable to net entry of new establishments. Likewise, the authors show that young firms (five years or less) experience productivity growth at a 3 percent higher clip than mature surviving establishments.

In basic economic terms, the process helps speed the reallocation of resources — such as workers and capital — to businesses with the greatest upside opportunities, and away from those that are fading. If that process is curbed — as in a heavily regulated economy that didn't allow certain industries to decline — the entry of new business would be stifled because the resources that would otherwise flow to them are being kept within the old industries. "If you tweak the destruction process, inadvertently you stifle the creation process," Haltiwanger says.

While mature, large businesses (in existence for 20 years and with 10,000 or more employees) tend to have very high probabilities of survival, they, too, encompass their own creative destruction processes. Within firms, offices and factories are opened or closed, swaths of workers are hired and fired. In this sense, it is not just the "young guns" that figure out what works and contribute to productivity gains. It is an economy-wide feature.

Economic churning may be painful for those in declining industries. But observing the process may help policymakers better understand how to respond to trends like the growth in offshored jobs. It seems clear, for example, that protectionist policies like trade barriers would be something that stifles the process, but perhaps the funding of worker-retraining programs would be useful to cushion the individual hardship the creative destruction process might create.

"We don't want to come off as saying this is costless," Haltiwanger says. "We recognize that it's very costly, both to businesses and to workers. Our counterpoint is that we understand that, and indeed we as economists or policymakers are trying to find ways to make the system work that much better."

Doug Campbell is an economics writer in the Richmond Fed's Research Department.