Friday, October 21, 2011
Further to my post of yesterday, here are some more measures of the shape of the yield curve. We know the Fed is trying to manipulate the bond market, first by QE, and second by Operation Twist. So it's possible the yield curve could be sending false signals. But I am of the belief that while the Fed can absolutely control short-term interest rates (it's policy target is the overnight rate), it has very little control over the yield curve from 5-10 years and out. That's because longer term interest rates are determined by inflation fundamentals and market expectations of what the Fed will do in the future.
Today, we know the Fed can't stay at zero forever; at some point the economy will pick up and the Fed will begin increasing short-term rates. Looking at the forward Treasury curve, we see that the market expects 3-mo. T-bill rates to rise from zero today to 0.71% in two years. 2-yr Treasury yields are expected to rise from 0.3% today to 1.3% in two years. 5-yr Treasury yields are expected to rise from 1.1% today to to 2.2%. In short, a positively-sloped yield curve is driven by the expectation that short-term rates will rise in the future. That is always the case when the yield curve is positively-sloped.
As the top two charts show, despite the Fed's attempts to depress long-term interest rates, the curve out to 30 years remains very steep by historical standards. That is fully consistent with the expectation that the economy will improve. It is not at all consistent with the view that we are on the cusp of another recession. If the latter were the case, the yield curve would be flat, especially in the 2-5 area. Yet even that part of the curve, where Fed expectations are very strong drivers of yields, the curve still has a definite positive slope. No matter how you look at it, the market fully expects things to improve in the future, not get worse.
Posted by Scott Grannis at 8:52 AM