Thursday, May 27, 2010

The 10% GDP output gap

With today's release of the second estimate of GDP numbers for Q1/10—which resulted in a very minor downward revision of annualized real growth from 3.2% to 3.0%—I thought I would revisit this chart, which compares the path of real GDP to a 3.1% annual growth path. My choice of a 3.1% growth rate harkens back to Milton Friedman's Plucking Model of growth, which I discussed in a post almost one year ago. In essence, his theory is that the U.S. economy has a built-in ability and/or desire to grow by a certain amount every year, and when it fails to achieve that, because of a recession-provoking disturbance of some sort, then it has a strong tendency to snap back to that long-term trend line once the economy has adjusted to the disturbance. This behavior has been documented by the Atlanta Fed: the sharper the recession, the stronger the recovery.

If this theory holds true, then currently the economy is about 10% below where it really wants to be. This would ordinarily lead to an explosive recovery. I have been arguing for over a year now that we won't get the explosive recovery (in which the economy would grow by 6-8% for a few years), primarily because of the monumental amount of fiscal "stimulus" this time around that is holding back growth by making the economy less efficient. Instead, I've been looking for 3-4% growth, and that's what we've been seeing so far. I think a cessation of fiscal "stimulus" spending would give the economy a huge boost. Note that this goes directly counter to what conventional wisdom is saying; everywhere you look these days you see people worried that the fourth quarter of this year is going to be weak because stimulus spending is scheduled to drop.

The real problem with the "output gap" we have today is twofold: on the one hand it encourages the Fed to remain hyper-easy, out of fear that the gap exerts strong deflationary pressure on the economy; and on the other, it encourages Washington to "do something," like extend unemployment benefits (which only reduces the incentives of the unemployed to seek work) and otherwise spend money (which takes money from the private sector that could otherwise be put to better use). To the extent we can cut spending, I think the economy will be better off. And if the November elections are going to be as transformative as I think they will be, then the prospect of major cutbacks in the size and role of government in coming years should be a cause for celebration, because then the economy will have a much better chance of closing the output gap rapidly, instead of over the course of many years. And the sooner the economy starts perking up, the sooner the Fed is going to have to normalize (i.e., raise) interest rates. This won't be a problem either, because current interest rates reflect the market's pessimistic view of future growth. Stronger growth and higher rates should go hand in hand.


seekingtraceevidence said...

The proper method is to run a regression on Real GDP from 1930 data. Today that value is 3.2%. Then you should add back in an estimate of core inflation. The 12mo Trimmed Mean PCE(Dallas Fed) is ~1% so the actual trend rests at ~4.2%.
Your chart method includes inflation which distortes your result.

Scott Grannis said...

seeking: No, my method does not include inflation. I'm only using real GDP. The growth rates I'm citing are also expressed in real terms.