Friday, March 6, 2009

One Key Prediction Shown Wrong

Last fall I predicted that nonfarm payroll employment would not drop below 134 million.

As you can see below, the employment prediction was wrong.

Another key prediction -- that real GDP would stay above $11 trillion -- is still correct by a wide margin. This prediction will probably hold true, but the bad productivity data lately make me worry that I might someday be wrong on this one.

If and when GDP is below $11 trillion and employment is below 134 million, I call that a "severe" recession -- at least as bad as those in the 1970s and 1980s.


Unknown said...


I think your only mistake so far has been to overestimate the new administration's capacity for positive policy responses. If/when we continue to see the price of unemployment go down, your prediction is only bound to get worse.

Am I right?

Uncivilly Yours said...


You're late to the party. Everyone knows this is a terrible recession and that things are continuously getting worse NOT because there are "labour distortions" but because we suffered a huge shock to aggregate demand and that supply is adjusting accordingly. But then again, if everything is going to be seen from the "supply first" point of view it's easy to see how you're missing the plot.

Tino said...

“Everyone knows this is a terrible recession … because we suffered a huge shock to aggregate demand”
I once heard a wise central banker point out “I never knew as much macroeconomics as I after my fist semester in undergraduate”.

This is a unique and complicated event. Probably no one in the planet knows exactly what is going on, why the crisis happened, how best to deal with it. If your textbook IS-LM curve was enough the world would hardly be in panic. It’s foolish and arrogant to make certain claims. Mulligan went on a limb and put forward testable empirical predictions, because that’s how you do scientific economics the Chicago way, not hiding in a crowd.
The point of relying on a models and formulating empirical predictions to be tested (therefore risk failing) is not than you are never wrong, it’s that you are doing things in a scientific way, so that events can be used to evaluate models.

It’s easy to babble on without ever formulate your model, stating your assumptions and making testable prediction. That’s what most people do. But it’s usually not serious science. If you understood scientific methodology you would realize that Ex ante the value of a hypothesis that turns out wrong is exactly the same as one that turns out correct.

RPB said...

The real question is will the Fed be able to insert enough liquidity to replace the declining amount of credit. Or will the velocity just explode?

What happens, if in fact, several trillion in credit is revoked?