The other day Spencer posted a chart that Cactus had sent and then Cactus posted the link to the data. After looking at the data, and considering the question being asked: Are we seeing a recession? I thought, maybe looking at just median weeks of unemployment or number of unemployed is not going to capture the true picture.
During a recession or in determining a recession would not the amount of idle labor be more of a factor? If we have 5 of 10 people unemployed for 5 weeks, is that worse than 6 of 10 unemployed for 4 weeks?
For this chart I used average weeks unemployed times the number of unemployed looking at the first day of the first month and the first day of the seventh month to come up with a factor we can call Person Weeks Unemployed: PWU.
As you can see, our economy has been performing worse as time goes on. The Y axis number need to by X1000.
We are leaving a lot of work on the table and the swings have gotten larger. Larger swings in labor sitting idle is the opposite of what we are praising in the GDP swings. And, note when the big swings started, right when we de-coupled the rise in productivity from the rise in wages: 1974 Up until that time, the lost labor fluctuates around the 50,000 mark. During the first oil shock, we hit a new high of 125,000 person weeks. No wonder the mood sucked during Carter’s years. At least it came back down to the upper end of the range we had been in since 1958. Unfortunately, since we decided to focus on money (yes Reagan) we seem to waffle around 125,000. A full 75,000 person weeks of lost work.
However, what is more interesting is the peaks of PWU in relation to the official recession dates.
Recession 11/48 to 10/49 but the PWU peaks 1/1950 3 month delay
Recession 7/53 to 5/54, PWU peaks 7/1954, 2 month delay
Recession 8/57 to 4/58, PWU peaks 7/1958, 3 month delay
Recession 4/60 to 2/61, PWU peaks 7/1961, 5 month delay
Recession 12/69 to 11/70, PWU peaks 1/1972, 14 month delay
Recession 11/73 to 3/75, PWU peaks 1/1976, 10 month delay
Recession 1/80 to 7/80, PWU peaks 1/1981, 6 month delay
Recession 7/81 to 11/82, PWU peaks 1/1983, 2 month delay
Recession 7/90 to 3/91, PWU peaks 7/1992, 16 month delay
Recession 3/01 to 11/01, PWU peaks 7/2003, 20 month delay (updated to correct math error)
Some may say: Hey, Reagan did good! But, if we ignore the blip of economic reprieve it is 30 months from the end of the first recession to the peak of PWU after the second. And he set it good, going from a previous high of 125,000 to 223,000!
So what do you think? I think that what matters to people regarding a recession is whether they are working and not when the government states the turn-around began. Thus, the peak of a recession is in the eye of the beholder. If you’re a person earning money from labor, a recession these days can last a very long time. This data would suggest that what we are seeing in the Spencer post is not a decreased risk but a lull before the storm. One other thing. It appears the Republicans fail again. As a group they have the longest turn-around to seeing a reduction in lost labor. In fact, the recent Bush years could be considered the longest lasting recession ever based on my PWU metic.