I'm always amazed how far partisan think tanks will spin their "research" to support political positions. All sides in the political debates do this. And demand for partisan studies has turned politically slanted research into a growth industry.
A recent - and impressive - example of partisan research spin comes from the left leaning Center for Economic and Policy Research. Their report, An International Comparison of Small Business Employment, states:
"we review internationally comparable data on the size of the small business sector in 22 rich countries. Across the full set of countries, for every measure we examine - including self-employment rates and the share of total employment in small enterprises - we find that the United States consistently has the lowest or among the lowest proportions of employment in small businesses."
The key phrase is "for every measure we examine."
The authors have selected four data sets to attempt to prove their point. Three of the data sets are, at best, odd choices: (1) the percentage of total manufacturing employment in manufacturing firms with fewer than 20 employees in 2006; (2) the percentage of computer related services employment in firms with less than 100 employees in 2001; and (3) research and development employment in enterprises with less than 100 employees in 2001.
Why these data sets - individually or together - are supposed to be good proxies for "the share of total employment in small enterprises" is beyond me. I could spend quite a bit of time on why they aren't, but won't bother. These data sets simply are not relevant and much better data exists (see below).
The fourth data set is self-employment. They use OECD numbers to claim the U.S. has very low self-employment relative to other industrialized nations.
I've pointed out in the past that OECD self-employment numbers for the U.S. only include unincorporated self-employment. Including incorporated self-employment increases the U.S. self-employment rate by about 50% - and moves the U.S close to the European average.
In addition to choosing unrepresentative or incomplete data sets, the report ignores readily available -and more timely - European Commission statistics on small business employment. This data shows about 53% of Europeans work for small and medium sized firms. Given the different definitions of small and mid-sized businesses, this is roughly the same as the 51% of Americans who work for small and mid-sized firms.
In other words, relevant and timely data clearly shows that the U.S. small business sector employment share is not significantly smaller than comparable countries.
The point of this study is to argue that our current health insurance system hurts small businesses and the self-employed. I don't disagree. And I think a good case can be made to support this position. Too bad they didn't make it.
One of the report authors has a Huffington Post article on the study. In his defense, he makes it clear that the study is partisan.
Comments