John Robertson, over at Macroblog, has an interesting post about what he calls a “disturbing trend.” According to data from the Quarterly Census of Employment and Wages (QCEW), the US has seen no growth in total business establishments. He says that the QCEW data shows “that the number of business establishments with payrolls in the United States has remained stuck at around 9 million since late 2007.”
Just to clarify, an establishment is a single physical location of some type of economic activity. In other words, a business. A single company may have multiple establishments. If establishments are not growing, something’s wrong.
Now of course, we know something’s wrong. If we just look at the straight job data, we see that from 2007 to 2011 our national job count has declined by 2%. You might have heard about this recession we had. But Robertson’s point stands: not adding new places of business is not nearly as good as adding new places of business.
His further point is that the size of establishments has reduced as well. In the 90’s, Robertson says, the average size of establishments stayed pretty stable at 16.5 workers, but then fell steadily through the last decade. The data from the second quarter of 2010 indicates that the current average establishment size is 14.3. Before the recession the average was 15.
These are both good observations, and helpful, as far as they go.
We thought it would be interesting to take a look at a different source. In Analyst we use the same data set for establishments that Robertson uses here, QCEW. However, County Business Patterns (CBP) is a Census dataset, and based on a sample, so it relies on very different methodology, and the numbers look quite a lot different. We also only have the data through 2008. However, it also allows us to split the establishment data up and see it in terms of number of employees.
|Number of Employees||2001||2007||2008||2001-2008 Change||2007-2008 Change|
First, note the way we’ve arranged the categories for this table. The vast majority of establishments employ two to four workers. To make the categories worth comparing, we’ve lumped establishments of five to 99 workers into one group, and then establishments of 100 to over 5,000 into another group.
This table shows us a definitive drop-off from 2007 to 2008 for establishments overall, and particularly for the two to 99 workers group. But within that group the two to four category is significantly harder hit. They’ve lost 2.1% as compared to the five to 99 group losing 0.6%, and the 100 to 5,000+ group gaining 1.3%. This all creates a picture of workers tending away from startups and toward larger companies in order to weather the recession.
The last thing to keep in mind is that this data is just looking at two points in time and telling us what the net change is. Saying that from 2007 to 2008 the nation lost 100,000 establishments doesn’t indicate that businesses generated no new establishments. It just tells us that more establishments failed. It’s not good news. But it doesn’t mean that nothing is happening.
Illustration by Mark Beauchamp