Good morning.
- Cities and global warming.
- Global warming and the denier.
- The UN and money … starve the beast?
- Of social contract and criminal law. I guess as a recent un-subscriber to social contract, I need to consider criminal law myself.
- Becker and Posner consider employment and the economy.
- On the whole, “not a news channel” thing. Another view on that here.
- Americans aren’t the only ones bad at history.
- Considering loving God.
- Well, that prediction went well.
- Some logic.
- Honduras.
- Science fiction and a rock star.
- Hmmm, perhaps because it was noticed?
- You go girl!
Well, that prediction went well.
Indeed, predictions rarely ‘go well’ in the sense that its almost impossible to predict a complex system with accuracy.
But the blogger errs when he asserts Keynesian thinking is in error. If the weatherman predicts a blizzard with 6 inches of snow and 12 inches shows up you haven’t demonstrated any serious problem with meterology. If a sunny 90 degree day shows up then yea there’s a problem.
So no unemployment benefits do not make unemployment.
Boonton,
You make a strong case against stimulus and related spending. After all, would you bet your retirement that the weatherman will be right?
And if a weatherman when it is starting to snow, predicts just a few inches and we get dumped on, then … we say “yea there’s a problem” and lo’ this is what we are telling these economists that, “yea there’s a problem.”
Let’s skip the analogy and cut to the chase. Suppose the actual unemployment figures were coming in right along the predicted ‘with stimulus’ curve….or for that matter if stimulus didn’t pass and the actual data points were coming in along the ‘without stimulus curve’. Would Obama’s critics accept that as ‘proof’ of Keynesian economics? No they wouldn’t. Why should the opposite hold?
Now let’s return to the analogy:
And if a weatherman when it is starting to snow, predicts just a few inches and we get dumped on, then … we say “yea there’s a problem” and lo’ this is what we are telling these economists that, “yea there’s a problem.”
In both cases we have a theory being applied to make a prediction. A prediction can be off for two main reasons in this case. The first is that the theory is entirely wrong. The second is that even with the right theory certain areas remain highly unpredictable. Long story short, there’s no coherent theory that would demonstrate stimulus making things worse so the more likely explanation is that the prediction was off and both the ‘with’ and ‘without’ curve should have been shifted up a few points.
Boonton,
You know that’s not true.
Feel free to present one. Keep in mind a coherent theory that predicts stimulus making things worse would also make other predictions which would have to be confirmed against the facts as they are now.
Let me use the weather example again:
The weatherman has a theory: When moist air hits a mass of cold air, the result is snow. How much snow requires knowing various variables like the mass of the different systems, how long they are in contact and so on.
Let’s say there’s an alternative theory: When moist air hits cold air heat is generated through the friction of the two masses rushing into one another.
If the weather man predicted 6 inches of snow and 12 inches showed up, I wouldn’t say the theory is bad. He could have been off in applying the theory or there could have been uncertainity in the measurements. While advocates of the friction theory might get a great chuckle at their rival being off, it really isn’t that impressive as an argument for their theory….nor is it much of an argument in terms of policy recommendations. It still makes sense for the mayor to order the plows to gear up and get ready to cancel school for the day.
I think the most obvious explanation for the graph is that both curves were several points too low. There are some other plausible possibilities, at the end of the day its not just how high the curves are but how much area they cover. 10% unemployment for two months is not nearly as bad as 8% unemployment for two years. It still may be that while the shape of the curves was off their total areas were pretty right.
In terms of coherent anti-stimulus theories, most of the ones I’m aware of center around ‘crowd out’ and cosist of perfectly rational actors either reducing spending to perfectly offset gov’t stimulus and/or gov’t borrowing leaving less room for private investment. Even then most of the anti-stimulus theories would probably predict a short term improvement that is offset in the longer run by the total cost of the stimulus. None would assert that in the very short term (and keep in mind we are not even talking 6 months here) stimulus causing unemployment to worsen.
If you want to take this issue seriously, then I suggest looking at the article below.
http://norris.blogs.nytimes.com/2009/11/09/did-unemployment-really-rise/
In reality, the government report says unemployment rates remained steady at 9.5 percent. And the number of jobs actually rose, by 80,000. And the number of jobs for college-educated Americans rose more than in any month in the last six years.
…
Studying the unadjusted numbers provides some indication that the hiring is starting to improve for better jobs. The number of jobs for college graduates, according to the household survey, rose 755,000 in October, before seasonal adjustments. That is the third-largest increase since the government started counting those figures, in 1992. (It trails increases of 895,000 in February 2002 and 755,000 in October 2003.)
No doubt if this was the Bush or McCain administration, you’d be hyping the above facts as examples of media bias. I’ve played with seasonal adjustments when I tutored people in stats classes and they are rather strange. On the one hand they are very real effects that you’d like to ‘filter’ out to get at what’s really happening. On the other hand they feel like a virtual world where you overwrite ‘real’ numbers with ‘fake’ ones.
What seasonal numbers do depend on, though, is a regular recurring seasonal effect. The economy, though, is not normal but is going through another Depression (sorry, 1982 is not a good model for this recession). It’s not clear whether the normal course of business when it comes to seasonal adjustments still hold water or not.