Menzie Chinn takes on stimulus and forecasting

From my former Econ Prof's very good blog:

There's been a lot of breast beating over the fact that the Administration underestimated the severity of the downturn.... I'll dispense with the clearly economically illogical arguments and try to tease out what is the "surprise" element in the 2009Q1 figures, and from that infer how much worse the economy was relative to what private sector forecasters predicted, conditional upon the passage of the ARRA.

First, define "surprise" as the deviation of the ex post value of X from the ex ante value:

X_t - ε(X_t | Ω _t-1)

Notice the information set Ω_t-1 does not include information that arrives between date t-1 and date t. Further note, I've used ε to denote subjective rather than mathematical expectations. If subjective expectations equal mathematical expectations, then one is using the rational expectations hypothesis, so that forecast errors are i.i.d. The rational expectations hypothesis is not an uncontroversial assumption (for instance, the efficient markets hypothesis is a joint hypothesis combining a particular asset pricing model and rational expectations), but for the sake of exposition, I'll use it here.

X_t - E(X_t | Ω _t-1)

Now, let X be GDP....

In order to get a feeling for the information and views roughly contemporaneous with the analysis of ARRA's effects, let's examine measured GDP as of 30 January, and the mean forecast from the WSJ survey of forecasters for early February. The early-February forecast incorporates the data and information used in the Administration's forecast -- actually a little more, since their data set ended in January (see this discussion of the Troika forecasting process here). Crucially, by this time, it was pretty well acknowledged that some sort of stimulus bill in the $800 billion range would be passed. Now, compare against actual GDP recorded as of the end-June 2009 in the 2009Q1 final release. The difference between the February forecast and the actual reported is the output surprise (relative to early February 2009 data).

The gap between actual and expected was 0.85 ppts (calculated as a log difference). In other words, GDP turned out to be lower by nearly one percentage point than expectations conditioning on the passage of the stimulus bill. That is a quantification of how much worse the economy was (in GDP terms) than anticipated, according to private sector forecasters. (This means that individuals who ascribe the worse-than-expected performance of the economy to the stimulus package cannot look to the ex post GDP realizations for support for their arguments.)

Very interesting stuff, but I'm not sure if I agree with the last parenthetical sentence. If GDP is a percentage point lower than what was expected conditioning on the expected passage of the stimulus, why does that mean that "individuals who ascribe the worse-than-expected performance of the economy to the stimulus package cannot look to the ex post GDP realizations for support for their arguments"? In fact, I don't see how it supports either a pro or anti-stimulus argument. GDP could be worse than expected because the economy was worse than the consensus, the stimulus did not live up to the Obama administration's expectations, or the very, very small chance the stimulus had a negative effect (none of these are mutually exclusive). Menzie makes a good point when he states that very little of the stimulus was spent in Q1, and thus should have little effect on Q1 GDP. I can see how that means it's not right to look at Q1 numbers and conclude the stimulus was ineffective, but not the logical conclusion Menzie makes.

Following the portion of the post I quote above, he goes into a rough econometric analysis of forecasting and what that means for the stimulus and direction of the economy. If you have basic knowledge of econometrics and a decent understanding of macroeconomics, I highly suggest you check it out.

In related news, Eric Cantor's shameless and wrong:

“I do think it is fair to day [sic] that the stimulus is a flop,” said House Minority Whip Eric Cantor , R-Va. “The goal that was set when we passed it was unemployment wouldn’t rise past 8.5 percent, and what we see now is businesses just aren’t hiring. Even the best projections have us losing 750,000 more jobs this year.”

I think it is fair to say that the stimulus might not have worked or that it might not have worked, but to say its a flop? There's no way to tell right now, and maybe ever. I think the effects of the stimulus will be debated in economic literature for a long time, and I'm not sure we will ever know what the effects were with any certainty. If you buy the Keynesian theoretical basics that Menzie taught me as a student of his (and I do), then I think you'll largely see the stimulus as helping. If the economy recovers or stays flat, then you will point to the stimulus as a major factor. If the economy gets worse, you'll probably claim it should have been larger. If you didn't agree with the rational for the stimulus in the first place then what happens to the economy probably won't change your mind. If the economy gets better, you might claim it would have anyways or the situation would have been even better without the spending extra money. If it gets worse, then flop it is.

Update 12:27pm 7/14/09: Menzie replies, "As of early Feb., one has an information set incorporating passage of the stimulus bill, if not with 100% certainty, then pretty high. Then you have the actual realization of GDP. You can decompose the "surprise" into a part that it "new" information regarding the state of the economy, and "new" information regarding stimulus composition and effectiveness. I set the new information regarding the latter at near zero. You could try your own decomposition; I'd welcome finding out your conclusions."

I'll have a reply soon but wanted to post his reply.