Data: Explore > Jobs > Self-Employment

[shawnbot]

This data comes from BEA, so we can script accessing the API and getting it in a form that's useful for us.

[shawnbot]

@Isabelle1512, can you explain this note in the data catalog?

It [...] must be calculated, by subtracting SA27N Full- and Part-Time Wage and Salary Employment by NAICS Industry from SA25N Full- and Part-Time Employment by NAICS Industry

I'm just not sure what the difference is between the SA25N and SA27N numbers, and why we would subtract one from the other.

[shawnbot]

@Isabelle1512 So, when I access the SA27N numbers from the BEA API, they're described as "Private nonfarm wage and salary employment: Mining". That actually sounds like the number we'd want for wage & salary in #926, no? If so, subtracting makes sense: the difference between the total and that number is the self-employment number. Can we use this wage and salary number instead of the BLS one?

[shawnbot]

I might be answering my own question here, but I compared the BLS and BEA numbers for wage and salary, and there is a difference. For California in 2013 I get 27,851 jobs from BEA and 24,726 from BLS, for an 11% difference.

[shawnbot]

@Isabelle1512 I'm going to assign this to you for review for now. We'll have numbers up on the site for you to check shortly.

[shawnbot]

@Isabelle1512 you can review the numbers on the site or in the raw data files (TSV, should open in Excel).

[Isabelle1512]

@shawnbot : you are right that BEA SA27N is very close to BLS QCEW. In fact, the BEA uses the QCEW as the data source for SA27N. What I learned this year is that, like the EIA, the BEA takes data collected by other agencies and performs additional analysis on it for the public to produce other data sets. I think depending on the various stages of release the BEA SA27N data may almost be identical to the BLS QCEW. The reason we can't use BEA SA27N for Wage and Salary employment is that it doesn't break down as narrowly by commodity as the BLS QCEW. CSO really wanted to talk about the differences between iron, copper, gold, coal, and oil and gas employment. Impossible to do with the BEA data set.

[Isabelle1512]

@shawnbot : to answer your other question, yes, subtracting the two tables gets us self employment. This particular route to self employment (there are others that are more direct!) enables us to include sole proprietors and partnerships under the self-employed total. I guess in the oil and gas industry, there are quite a lot of sole proprietors operating in the "oil patch," so Industry thought it was important to use this measure to capture those people.

[Isabelle1512]

@shawnbot : numbers look great.

[shawnbot]

Hooray, thanks @Isabelle1512!