Early July, I wrote an article about the steady rise in interest rates (GOVT, TLT and TRSY). My conclusion was that, although the speed at which the interest rate went up was pretty impressive, the current 10-year yield at that time (2.50%) was probably too low, instead of too high. Now, less than two months later, the 10-year yield has gone up even further and reached an intraday high of 2.93% on August 22. That is roughly 1.20% above the level at the start of the year. And asks for a fresh answer on the question; "Have bond yields risen too much?".
Bond yield model
My starting point is, again, a simple model to estimate the current 10-year bond yield. The model incorporates three factors that are related to economic growth, inflation and monetary policy. In general, higher economic growth, higher inflation and higher short-term interest rates are accompanied by higher bond yields, and vice versa. The ISM New Orders Index is included to represent economic conditions, the Core CPI is used as inflation gauge and the 3M-Libor rate is used as a proxy for the monetary policy.
The result is shown below. Based on the current ISM New Orders Index, Core CPI and the 3M-Libor the 10-year bond yield should be somewhere around 3.50%, still well above the current yield of 2.90%. As can be derived from the graph, the model estimate has spiked recently. This is mainly because of the ISM New Orders Index. The index improved firmly in July, rising from 51.9 to 58.3. This latest number on the New Orders is, however, high in historical context. Over the last 20 years the ISM New Orders Index has averaged just below 55. But, even when I use this historical average, the model estimates a bond-yield of 3.25%. So if anything, the model indicates interest rates are still too low.
The view that the current bond yield has not risen too much is not jeopardized by the other two factors in the model. Bloomberg expectations show that inflation is expected to rise this year and next. Higher realized inflation numbers will lead to a higher bond yield estimate. And with most Fed-members now 'comfortable' with the tapering suggested by Bernanke, the probability of a higher short term interest rates has increased somewhat. The graph below shows the implied probabilities from the Fed futures for different short-term interest rates for the last FOMC meeting of 2014. I used this meeting as the discussion of the first rate hike pivots around this FOMC meeting. The graph shows two sets of probabilities, one from June 29th and one from August 22 th. The changes are small but, overall, the chance of at least one rate hike has in increased by about 6%.
Then, off course, there is the Quantitative Easing factor. This is not incorporated into the model, which explains the discrepancy that has occurred between the actual bond yield and the model estimate in recent years. Estimates are that it takes about $1 trillion of bond purchases to lower the long-term interest rate by 0.40%. Data from the website of the Federal Reserve Bank of St. Louis show that the Fed has just reached a new 'milestone'. It now has a little over $2 trillion of bonds on its balance sheet. That's roughly $ 1.5 trillion more than at the start of QE, resulting in an estimated downward pressure of approximately 60 basis points (1.5*0.40%). Taking this into account would lead to a bond yield estimate of 2.90% (3.50%-0.60%), freakishly close to the actual yield.
There is one final point to be made concerning the effect of the Fed bond buying on the bond yield. As the minutes from the latest FOMC meeting show, the Fed members are feeling comfortable enough to support some kind of bond tapering. This will be a slow process, in which the amount of bond purchases will come down gradually. But given the fact that some of the bonds on the Fed balance sheet will mature, the downward pressure will also slowly diminish. This in turn will allow the bond yield to close the gap with the model estimate once more. As of today that would imply even higher rates, not lower ones.