Now, I'm Really Smelling a Rat!

Following the announcement of last month's unemployment rate of 8.5 percent , I questioned the accuracy of that number in my blog entry titled "Some Fuzzy Math On Unemployment". The Obama agency that released that number was the Bureau of Labor and Statistics (BLS). The announced 8.5 percent conveniently "matched" what Obama's economic team was predicting as a high for 2009. How intuitive!

Shortly after the release of the national number, it is normal for the BLS to release the state information for the same period. This data is now available at Wikipedia (Click To See Data).

Conveniently, that report doesn't provide any foot and crossfooting computations of the state numbers that one would expect from an agency that has the word "statistics" in their name. So I did just that; and I came up with this report that includes some extended data that could either disprove or prove the accuracy of their national report (Click to View My Revised Report).

In each state's case, I plugged in the population of that state* and, then, based on the unemployment rate, I extracted the proportional number* of the unemployed for each state. Then, I footed all that data by totaling the population all the unemployed; and, then, calculated the overall rate of unemployment.

So, guess what? When I did all that, I got an unemployment rate of 8.8% which is closer to the 8.7 percent that I thought was the real rate was and a number that is three tenths of a percent higher than the 8.5 percent the BLS announced.

While I might be wrong due to some rounding errors, I don't think I can be off by more than 1/10th of a percent. I thought this was interesting.

*Note: Actual unemployment rates are calculated on the basis of "available workforce" and not pure population numbers. So, using my form of calculation, the overall unemployment number of 26 million who are out of work is too high of a number because it includes portions of the population who are not part of the workforce; such as children, retirees, handicapped, etc. Furthermore, there are individual states who have higher or lower workforces based on the amount of retirees and children in their population's mix. For sure, a retirement state like Florida has an "unemployed" number by me that is greatly overstated. Offsetting that would be a state like Michigan were retirees have a tendency to move away in their later years. However, the bottom line of 8.8 percent should be very accurate because it was extracted on a "proportional basis" and it smooths out the workforce variances by state.