The blue line is the ratio of initial claims to the labor force. In the past, with claims levels this low, unemployment would have been down to 4% or 5%.
Continued claims (red line) aren't quite as low, but even they would point to 5-6% unemployment (the green line +2).
The orange line is the ratio of continuing claims to initial claims. This ratio appears to have a significant upward slope over the past 50 years. This is the result of an economy with less employment churn and/or longer average duration of unemployment. This trend is over a long enough time period that age demographics is probably not the explanation. There must be some political or cultural trend that is creating a less volatile labor market.
Extended unemployment insurance is not included in the continued claims measure. There are currently about 2.8 million unemployed workers with continued claims, or about 1.8% of the labor force. There are an additional 1.35 million workers on extended UI benefits, or about .9% of the labor force.
Adding the extended unemployment insurance numbers to the normal continued claims would move the continued claims series up to about 2.7% of the labor force, tracking right alongside the unemployment rate series.
Continuing claims are declining, but are starting to level out. On the other hand, extended UI rolls are shedding about 800,000 workers per year with a pretty linear trend. This, combined with the probability that Congress will shut down the program as the economy recovers, means that we should see the unemployment rate tracking with normal continued claims by around the end of 2014 or beginning of 2015, which would put UE at about 5% or 6% soon after, if continued claims are still declining slowly.
This suggests that there are some tailwinds for the unemployment rate, and we could see it continue to fall at a rate at least as fast as the recent trends.
Recent work by Farber and Valletta suggests (pdf), only pegged about .4% of the peak unemployment on EUI. But, they found that most of the added unemployment duration associated with EUI was the product of delayed exits from the labor force, not from delayed re-entries into the labor force. So, a winding down of EUI could lead to a reduction of a few tenths of a percent in the unemployment rate, even without any new resurgence in employment.
Normally, I have been arguing that the recent declines in the Labor Force Participation Rate are all demographic. This is a slight counter to that argument. EUI is probably causing LFP to be overstated by a few tenths of a percent. Without the EUI effect, LFP would probably be showing some additional cyclical weakness, even when demographics are properly accounted for.
I have found that rates of exit from unemployment at both long and short durations have been high during this recession. If a portion of even initial claims and exit rates of recent unemployed workers is also related to EUI, the effect of EUI on unemployment would have to be revised upward.
The good news is that the continued declines in EUI claims and the recent sharp improvements in regular unemployment insurance claims both presage a continued decline in the unemployment rate.