Does the yield curve mean anything?
When you’re zooming into a curvy section of the road, you need to slow down. It’s a simple matter of physics and engineering. A few years ago I was biking in the early morning and came into a curve too fast. I had good tires and a bike that was great around the corners. But there were wet leaves near the curb and my bicycle skidded out from under me. I experienced a very nasty fall.
But that’s not what’s supposed to happen in economics. In economics, people often assert that when there isn’t a curve in the financial indicators – when the yield curve is flat, or inverted – it’s the economy that’s supposed to slow down. It’s been observed that before each of the past seven recessions the yield of the 2-year US Treasury Note has been higher than that of the 10-year Note, going back 50 years. Seven-for-seven seems a pretty good record. Since the curve recently inverted, all we have to do now is wait, right?
But it’s not so simple. In the past, an inverted yield curve would cause a recession. The mechanics were fairly straightforward. Banks had a statutory maximum interest rate they could pay to customers. It was called “Regulation Q.” (Don’t ask why, but the Federal Reserve likes letter-based names.) From 1933 until 1986, Reg Q imposed maximum interest rates on all bank deposits. It had been thought that part of the cause of the Great Depression was excessive bank competition.
During this time, if the Fed raised interest rates above the Reg Q maximum, deposits would leave the banking system and banks would have to cut back on their lending. This would cause a credit crunch, the money supply would shrink, and a recession would inevitably follow.
But after Reg Q was repealed, we really had a different system. The yield curve couldn’t cause a recession any more. But could it be an indicator? There have been three recessions since 1986, and an inverted yield curve has preceded each of them. Does this mean anything?
The problem with this analysis is that three is a very small number. Any statistical work that uses a data set with only three elements won’t be significant. I decided to try to improve the picture. I examined the yield curve’s effectiveness as an economic indicator, using the 2s-10s spread and GDP. If the yield curve could predict a recession, it ought to be able to predict the opposite, an economic boom. I looked at the yield curve and GDP in a graph and in a regression analysis. For the regression, I used discreet 18-month periods of economic growth since 1985. That way, at least I had 21 data points to examine.
Source: Bloomberg, Federal Reserve, Charter Trust.
My results aren’t very convincing. The slope of the yield curve doesn’t appear to predict anything. Only once – out of three inversions – did the economy contract following an inversion. And the worst economic decline during this period was preceded by a steepening yield curve. My conclusion is that there is essentially no correlation between the yield curve and economic growth.
That doesn’t mean the yield curve is meaningless. When considered in the larger context of economic and financial indicators, it can provide insight into financial conditions and investor behavior. But there’s nothing mystical or magical about it.
The laws of physics govern the coefficient of friction between a bicycle tire and the road, and sharp curves require a slower approach. But there are no laws of physics governing financial indicators. They’re a matter of judgement. Amid all the breathless claims about crashes and booms and prophecies of doom, it’s important to keep this in mind.
Douglas R. Tengdin, CFA
Charter Trust Company
“The Best Trust Company in New England”