Tuesday, November 14, 2006

Buying When The Yield Curve Is Inverted: Good Bet or Regret?



Those masters of the anti-motivational posters at Despair.com have it right: the worst mistakes are often the swiftest. So, it's with that in mind that I decided to take a look at the inverted yield curve in hopes of avoiding future regrets. After all, we always hear that inverted yield curves are bad for stocks, but here we are now--the curve having been inverted for a while--and we still stand at the summit above Dow 12,000.

I went back to January, 1970 (N = 1872 weekly periods) and identified all occasions in which the yield on the 3 month t-bill exceeded the rate on the long bond. This is the case at present. It turns out that we've only had 277 such inverted periods over the past 36 years.

When we look at the Dow Jones Industrial Average 40 weeks later, we see that the average price change following an inverted yield curve is a loss of -1.06% (105 up, 172 down). This is very much weaker than the average 40-week gain of 5.12% (1273 up, 599 down) for the entire sample. In short, buying the market during an inverted yield curve has indeed been a losing strategy.

Consider some of the historical periods with inverted curves: January-May, 1970; March 1973 - December 1974; September 1978 - April 1980; October 1980 - October 1981; February 1982 - April 1982; August, 1989; July - December 2000. All in all, pretty bad times to have been in stocks if your holding times were within a two-year horizon.

Note also that inverted periods can persist for quite a while before resolving in a decline and eventual Fed easing.

This time, of course, *could* be different. No doubt that's what our friend in the poster thought as he passed the summit.

Dr. Brett's Blog Links and Market Stats for Tuesday

No comments: