The Atlanta Fed's macroblog provides commentary and analysis on economic topics including monetary policy, macroeconomic developments, inflation, labor economics, and financial issues.

Authors for macroblog are Dave Altig, John Robertson, and other Atlanta Fed economists and researchers.

« Another step to enhance Fed transparency--and no better time than now | Main | Monetary policy exit: Is a bad bank the solution? »

March 30, 2011

A disturbing trend: No growth in total business establishments in U.S.

The last Atlanta Fed poll of small businesses in the Southeast suggested an uptick in confidence late last year. A similar upturn has been noted in the National Federation of Independent Business's survey of its members conducted in February of this year and released in March. This upturn is good news for the U.S. economic outlook, as small firms are one group that has lagged the economic recovery.

It's also good news, given the continuation of unimpressive readings from last week's release of the Quarterly Census of Employment and Wages (QCEW) for the second quarter of 2010. As we have noted previously and highlighted in this recent Wall Street Journal blog post,

"The recession caused a sharp decline in new business start-ups, intensifying job market losses and potentially putting future economic growth at risk."

The QCEW data also showed that the number of business establishments with payrolls in the United States has remained stuck at around 9 million since late 2007. By comparison, in the early 1990s there were about 6.5 million establishments, a number that rose to close to 8 million in 2000 before peaking at 9 million 2007.

The net creation of business establishments—that is, physical locations for conducting business such as manufacturing plants, retail stores and business offices—has in the past been a key ingredient in job growth in the United States. This growth is driven partly by demand from newly created businesses and by mature firms expanding their footprint by opening additional locations. The demand for physical space is also clearly important to the commercial real estate industry, which has been burdened by elevated vacancy rates in many markets and generally low demand for new space.

Another trend from the QCEW data is striking—the number of employees per establishment is much lower than it used to be. The average size of U.S. establishments was relatively stable during the 1990s, at around 16.5 employees per physical location. The 2001 recession was associated with a decline in the average size to about 16 workers per establishment, and the average size continued to track lower during the last decade, moving down to about 15 employees per establishment in 2007. The latest reading for the second quarter of 2010 was 14.3 workers per establishment, up from 14 workers in the first quarter.

Several possible explanations exist for these declines in average establishment size. First, there is a cyclical response to weak demand as firms cut their payrolls. Second, productivity gains over time allow a plant, store, or office to support the demand for its goods or services with fewer workers. Third, there is a secular trend away from industries that have a large average establishment size, such as manufacturing.

If one digs into the data, only one major sector has experienced a rise in average employment per location over time—health care. This growth is likely a result of increased demand for health care services, and those services are primarily embodied in the staff at doctors' offices and hospitals. Manufacturing, on the other hand, has witnessed dramatic declines in average plant size. During the 1990s, average plant size was relatively stable at around 43 workers. The average size then declined to about 38 workers following the 2001 recession and remained around that level through 2007 before declining again and reaching an average size of 33 workers per plant in the second quarter of 2010. This trend appears to primarily reflect a combination of secular shifts away from labor-intensive types of manufacturing where productivity gains have already played out—some apparel manufacturers, for instance—and sharp cyclical downturns.

Of course, something will have to give if there is employment growth—primarily more and/or larger establishments. Consider the following thought experiment: if the average size of establishments returns to the prerecession level of 15 workers per location, and private-sector jobs are added around a pace of 2.4 million a year pace (or 200,000 jobs a month), then we would see establishment growth return to the 1992–2007 average of about 160,000 per year. Clearly, we are currently far below that trend.

Photo of John Robertson By John Robertson
Vice president and senior economist in the Atlanta Fed's research department

March 30, 2011 in Employment , Real Estate , Small Business | Permalink


TrackBack URL for this entry:

Listed below are links to blogs that reference A disturbing trend: No growth in total business establishments in U.S. :


I'm not sure we'll ever go back to what we were before. Our productivity has to be exploding, so fast now that it can't be captured for some time. When MS windows came out, it took many years to transform how we do business. But, transform it did. well, that same process is happening again but this time the number of years to transformation should be much shorter.

Posted by: FormerSSresident | April 01, 2011 at 12:54 AM

Hi John,

You can really can see the significance of the decline due to recession. At least it has bottomed out.

Posted by: Intrinsic Value | April 01, 2011 at 06:25 AM

"If one digs into the data, only one major sector has experienced a rise in average employment per location over time—health care. This growth is likely a result of increased demand for health care services"

Or, actually, the much more likely (and obvious - if not politically permissible) result of the vast, superficial "subsidizing" of health care - one that has resulted in a grotesque misallocation of capital, the warping of the macroeconomy, and a decades-long medical uber-inflation which (along with academic tuition inflation) is treated by government insiders as though it is some sort of inexplicable "mystery".

As opposed the political payola and bribery it so clearly is.

Posted by: csa721 | April 03, 2011 at 11:58 AM

Looking at that chart of business establishments makes me wonder even more about the point of the Birth Death model. If it's supposed to take into account the birth and death of new businesses, then why did it ever add all those jobs on a monthly basis. The data is clear that new business formation simply hasn't been happening for years.

Posted by: apj | April 05, 2011 at 05:43 AM

The culture of fear and uncertainty, higher energy costs and open political/media hostility towards business doesn't help.

If you are a venture capitalist, angel investor or entrepreneur is this likely to cause you to think startup?

Why should you?

Posted by: Army of Davids | April 23, 2011 at 02:54 PM

I find it funny how these results contradict each other, companies are just finding new ways to cut costs, and that normally means cutting employees

Posted by: Best virtual office | November 02, 2011 at 01:50 AM

I'm with Best, these types of results are always contradictory. Businesses are boosting profits by cutting employees. That is not sustainable.

Posted by: Penny Stocks | November 03, 2011 at 06:02 PM

Post a comment

Comments are moderated and will not appear until the moderator has approved them.

If you have a TypeKey or TypePad account, please Sign in

Google Search

Recent Posts



Powered by TypePad