Saturday, October 6, 2012

Updating a Regular Chart

Since I don't care much about baseball, I'll stick with my forte, inciteful comments about economics.  I've presented this chart from Calculated Risk Blog (post with graph) from time to time and it's worth checking again. 
It's easy to see that the recovery to pre-recession levels is poking along, but still (rough guess by extending that curve) another 3 to 4 years away, at the current slope.  Several of us have pointed out that there are plenty of reasons that the economy was going to slow down now, anyway, and that curve may truly never return to the employment numbers we had in 2005.  As I said last time I posted this, (just in July), I see something other people don't comment on.
note the curves where they return to 0% job loss compared to the peak employment before the recession.  Note that from right to left, the longest recessions are the current one, 2001, 1990, and 1981 in chronological order.  The spoiler is that the next longest one, the recession of 1957, seemed particularly deep and long.  I find the stretching out to the right of the '81 to 2001 recessions interesting, because it appears that recessions are getting both longer and deeper. 
At least, the 1990, 2001 and this 2007 recession follow the deeper trend; lasting longer holds back to the 1981 recession as well.  I suppose they separate out the 1980 and 1981 recessions because of some technical detail; I remember it as an extended period of general malaise. 
If you visit here regularly, I don't have to tell you I believe this is the result of the steadily increasing involvement of the government in "fixing" things.  The more they try to fix, the longer recessions take to heal.  Didn't their moms ever tell them, "stop picking at it, or it will never get better"?  Same basic idea.


  1. It would very interesting to me to see a similar chart that showed the labor participation rate change. Particularly for the 1980 through present incidents in the graph above. Do you know someplace that has some hard numbers I could use to put something like that together?

    1. Yeah, that would be an interesting thing to see.

      I took a look over at Shadowstats and didn't see a series like that. ( ) He might have it in there somewhere, but much of the site is subscriber-only. Likewise, I looked over at Calculated Risk, where this data is from, and didn't see it. (link to them in the article)

      Not sure who else to go to. The BLS might have a way to find it, but you'd ideally like to see it measured the same way over time, and I'm afraid BLS would mess with it like they do everything else. Shadowstats is good for that with the inflation and unemployment series he keeps by the older methods.

      Wish I could help more. If you find it, would you let me know?

    2. I've got something, not a pretty graph though. I took the BLS U6 from 1994, which is as far back as that goes; and the labor participation rate across the same dates. Crunched them together to see what percentage of the population the government thinks was actually working.

      I've got a google docs spread sheet I can share, but it's not real pretty.

    3. Cool. Thankfully, it's not the kind of data that needs a pretty graph. And with the free Open Office suite, it's just one operation away from a graph.

      But like you said originally, it sure would be nice to see that data extended back farther.

    4. BC - check this data out:

      it's sorted by state, but is one enormous text file - which imports nicely into your favorite spreadsheet.

      Hours of fun for the entire family. Assuming yours is as tragically geeky as mine!

  2. Ooh, thats gonna take some time to crunch through