Monday, May 19, 2008

Ponderings on the Latest Forecast

According to CNN (or with a more exaggerated headline at yahoo), a survey conducted by NABE of economists say 56% expect we are starting or will have experienced a recession by the end of the year, though it will be mild and short as the credit market regains function. Thoughts:
  1. There were 52 forecasters surveyed, so it must be that 29 of the 52 agreed with this rather vague sentence about recession (55.7% rounds up to 56%).
  2. Intrade has responded, but is not yet as pessimistic as economists, as it is trading around 39.
  3. The survey date was April 17-May 1. The BEA released higher than expected numbers on April 30th, causing the Intrade shares to drop considerably that day. This would not be reflected well in the survey unless they were surveyed disproportionately on May 1. Even if they were, it is unlikely forecasters would have been able to include it in their model in such a short amount of time.
  4. Forecasters are the appropriate group to survey, better than the general economics population, but they have their own incentives via clients. Some have a incentive to be more optimistic, some more pessimistic, and since this is not a written or admitted bias I don't know how it weighs into the survey.
  5. Has someone changed the definition of "recession" on me? I understand it to be 2 consecutive quarters of negative growth. The survey estimates 1.4% of economic growth this year, sluggish but not a recession. The advance estimate for the first quarter is 0.6%. How do we have 2 negative quarters, 0.6%, and then average 1.4% on the year? Big about-face quarters of growth do not appear in professional forecasts, it is too risky for the forecaster.
  6. They discuss the Fed manipulation of the federal funds rate to help spur growth. Coincidentally, Meltzer is the guest on EconTalk this week discussing what the Fed is really doing.
  7. There is considerable discussion of how much of the tax rebates will be spent and when as well as the implications for economic growth. First of all, they were spent when the President and Congress made it clear they would be passed. Secondly, why debate whether the tax rebate will work on the logic it is supposed to when the logic itself is flawed. Assuming gravity is not a constraint, can a golf ball be hit farther if it has dimples or if it is smooth? Don't bother answering because it is functionally pointless, just like debating the tax rebates effect on the economy.


Anonymous said...

A "official" recession (i.e. an NBER one) is not defined by 2 consecutive quarters of negative real GDP growth. Rather it is defined by the NBER, specifically see below:

"The NBER does not define a recession in terms of two consecutive quarters of decline in real GDP. Rather, a recession is a significant decline in economic activity spread across the economy, lasting more than a few months, normally visible in real GDP, real income, employment, industrial production, and wholesale-retail sales."--NBER

I believe the recession of 2001 fits this description. So, it is possible that we are (or were) in a recession, but the GDP numbers don't indicate this, but all of the other statistics together do.

Also, surveys that ask for discrete outcomes may not get it right, since economic growth has a continuous set of outcomes. That is, little economic growth would not be viewed as a recession, but its consequences are more similar to those of a mild recession than those from strong economic growth. So, the surveys may suggest a recession and be wrong, but if they are wrong, we are probably witnessing low economic growth, which we are (or have). Personally, I believe the recovery is underway and began sometime in late March, but the data is reported with a delay and will not reflect this for a while.

Anonymous said...

I forgot to post this earlier, but here is more detailed information on the determination of an NBER recession.

Justin M Ross said...

Good point, NBER recession though vague is better than the textbook definition (at least the 3 macro texts on my desk). FWIW, Intrade goes by the textbook rather than the NBER definition, which is probably the better choice for gambling purposes.