"Kee" Points with Jim Kee, Ph.D.

Second quarter US GDP growth came in at a 1.7 percent annualized rate, up from 1.1 percent in the first quarter. That makes more sense to me, as the first quarter is typically the weakest (while thethird quarter is typically the strongest). There is some truth to the adage that, rather than looking at GDP to see where stocks are going, it makes more sense to look at the stock market to see where GDP is headed. That's because prices lead quantities, and stock price data is forward looking while GDP numbers are backward looking, reported with a time lag, and subject to frequently large revisions.

Indeed, economist Edward Lazear made that exact case in Friday's Wall Street Journal in an article titled "The Stock Market Beats GDP as an Economic Bellwether." Any economist who has been involved in building economic forecasting models, including this economist, has tried to use the stock market as an input for predicting the economy. My experience has been that it (the stock market) is much too volatile to be really useful. But Lazear's article made some interesting points, the most noteworthy being that the advance estimate from the Bureau of Economic Analysis (BEA) has missed the actual average growth rate, updated later when more complete data is available, by half. That's pretty striking! He makes the point that there is actually no predictive power in quarterly GDP figures, a fact that I have mentioned before based upon detailed research by noted UCLA econometrician Edward Leamer.

Updates in the calculated GDP numbers, something the BEA does about every five years, also added about 3 percent to the size of US GDP, roughly like adding a country the size of Belgium (Financial Times). The biggest change was the treatment of spending on research and development (R&D) and copyrights as investments. I wouldn't make too much of this, nor would I let business journalists persuade you into making too much of it. Economists have always felt that R&D is an investment with multi-year benefits, as is much of advertising but R&D is less controversial. Some of that spending is of course wasted, but some is truly an investment. When a steel company reinvests, it builds more plants; when a retailer does so, it builds new stores or leases them, which are called operating leases. And when a technology company or biotech company reinvests, for example, they spend on R&D. So that's an update of GDP methodology that I’m fine with.

Anyway, the increased US growth rate from the first quarter to the second reflects an acceleration in growth, albeit modest, which is consistent with recent manufacturing and non-manufacturing US PMIs (purchasing managers indices). Both showed the US economy gaining momentum last month. Elsewhere in the world, there are signals from global PMIs and other leading indicators that global growth might be picking up. Capital Economics points out that, while there is a lot of fanfare regarding the "great deceleration" in emerging markets, the crucial fact is that the "BRICs" (Brazil, Russian, India, and China) have been slowing for years and will probably level off in the future rather than fall further. I think that's the right view.

Finally, I'll take a stab at where interest rates should be. Currently, the 10-year US Treasury yield is 2.65 percent. That seems really low, but with low inflation expectations and low growth, how low should it be? One way to look at it would be to compare rates today with those of a similar period in which inflation expectations were well anchored, like the Bretton Woods era from 1944 to 1971. During that period, the dollar was linked to gold, and the world’s major currencies were fixed (rather than floating) to the dollar. The median 10-year yield was 3.56 percent, about 90 basis points higher than it is today. During the first 20 years of Bretton Woods, the median 10 year yield was 2.75 percent, or about 25 basis points higher than it is today. I’ll ball park it by saying that if inflation expectations remain low, the 10-year yield should rise 50 to 100 basis points. There are of course major shortcomings to this kind of analysis. For example, there should be some sort of “fiat currency premium” built into interest rates today because the dollar is not linked to anything. That should make interest rates (government yields) higher. Working against that is the fact that today we have major global buyers of Treasuries like China, Japan, and Korea that weren’t as significant back then. That would put downward pressure on yields. But overall, I think 50 to a 100 basis points is a good starting point.