I think this is the longest I've gone without blogging since I started my blog! It's due to a confluence of factors — writing my second book (now about 1/2 written, and I'm learning so much finally looking at all these dynamic equilibrium models together from the 30,000 foot level), and I've become ridiculously busy at my real job. It means I missed the 2nd anniversary of the dynamic information equilibrium model. The model came out as a series of posts, first for JOLTS data (10 Jan 2017), then for the unemployment rate (11 Jan 2017), and finally a forecast of the unemployment rate (18 Jan 2017). That last forecast was compared with a forecast from the FRBSF that went until the end of 2018 — it was flat at about 4.7% the entire period. As I said in that last post:
The interesting thing is that in either [dynamic equilibrium] model we shouldn't expect the flattening out over the two year period [Dec 2016 to Dec 2018] we see in the FRBSF model. We should expect either 1) a recession to start raising unemployment, or 2) a continuing decline (albeit at a slower rate). A constant unemployment rate won't happen, and in fact generally doesn't happen [1]. We might be able to test the various models here [2].
And we could!
The uptick in the latest data to 4% is being blamed on the US government shutdown in the business news, but it's really well within the normal fluctuations of the unemployment rate so any speculation as to its cause is pretty much unfounded without detailed micro data. As a side note, there are various reasons why people think the Fed paused its rate increases at it's meeting this past week — discussed thoroughly by Frances Coppola here. However, I think Tim Duy has the best theory of the Fed's reaction function: as long as inflation isn't a problem, the Fed doesn't hike rates if unemployment is flat.
The most recent data shows a relatively flat unemployment rate for the past few months, much like in 2016. Basically, the FOMC has some model like the FRBSF model in its collective head of a constant equilibrium unemployment rate. And it's precisely that model which led the FRBSF forecast astray.
Glad to see you back at it. FWIW I have to say I didn't understand your dynamic equilibrium stuff at all until fairly recently. Now that I am getting a bit of your arguments, it seems to me that your dynamic equilibrium concepts (if they hold up) are quite revolutionary in their implications for economic policy.
ReplyDeleteThanks Todd.
DeleteIn a sense, it's "punctuated information equilibrium" where periods of information equilibrium are interrupted by disequilibrium shocks. I'd personally just hope that maybe some people would stop coming up with random models using "logic" or things that "seem right", and instead base their models on empirical data.
Glad to hear from you again. A couple weeks ago I eyeballed 4.2% as the unemployment rate from recent levels that would have me thinking recession, and it looks like that might be just outside your... what is the accepted term for them? Uncertainty bands? Variance bands? Confidence interval? Unfortunately, I don't know what to make of shutdown effects, if any, on the data or the statistic. Of course, if I understand some of your recent work correctly, you would put more stock in your JOLTS models as recession indicators than the unemployment rate it seems.
ReplyDeleteAnd I agree with Todd, and think your predictions concerning the unemployment rate seem like they have to be a good example of empirical success. Again, good to know you're still working at the project.
Thanks!
DeleteThey're called "single prediction error bands" in Mathematica (as in each new point should be within the band 90% of the time (the confidence level). But confidence bands or error bands are fine.
Yes, JOLTS showed signs earlier in 2008, and the extended series also shows they came first in 2001. For the two recessions, hires is about 5 months ahead. But this sample size is small (only a two recessions) so it's possible unrate could come first.
Hey Jason, I'm looking at the applications of entropy to economics as well, but from a Boltzmann-Gibbs perspective rather than an information theoretic perspective. I couldn't find your email on this website, but I'd like to discuss it with you sometime! I think your work is directly related to mine.
ReplyDeleteCool! You can DM me on Twitter @infotranecon or email me through
Deletecontact (at) arandomphysicist.com