Which causes the unemployment rate to skew too low as the labor force is the denominator in the unemployment rate.
Lets solve this problem shall we?
The chart below shows the labor participation rate and the rolling five year average of that rate. The key takeaway is that over the past 60 years there has been a huge secular shift as woman entered the workforce (more on that from Calculated Risk here), which effectively ended the "norm" of a single worker household.
As is typical with long term trends, this datapoint appears to have overshot to the upside and is now rolling over (individuals may have entered the workforce at some point that never planned to work because the economy was so good). As such, a more "fair" methodology (in my opinion) is to use the five year average participation rate to account for these ebbs and flows. This also alleviates the issue of believing that 1.8 million people "chose" to leave the work force over the past 6 months (yes, that is what the BLS methodology indicates).
EconomPic Unemployment Rate
Denominator: The methodology for the denominator (based on my logic detailed above) is to use the five year rolling participation rate multiplied by the civilian institutional population (i.e. those that can work) to determine the labor force.
Numerator: The EconomPic methodology is to use the number of employed rather than unemployed (some of those not in the labor force are not countered as unemployed). As a result, the initial equation = the "employed rate". Taking 1 less this employed rate gets us to the unemployed rate.
The result:
Unemployment is reduced at each "low" prior to the low of the late 90's and most recent 2007 period (both of these closely tie) and roughly matches each "peak". That is until the most recent period, indicating that something is just not right.
Whereas the "official" unemployment rate is 10% flat, the EconomPic unemployment rate is 11.7%.
Source: BLS
Data looks good to me, good representation of the real world unemployment situation.
ReplyDelete