I will be the first to say that rules, if not meant to be broken, were made to be bent. The grand central bank policy experiment of ZIRP and NIRP is effectively a bending of economic theory that, other than supporting asset prices, hasn’t provided the inflation boost they were projected to have. For a year and a half, the Federal Reserve has been threatening to raise interest rates and have not been able to do so. Still the question remains, what happens to long bond yields if rates begin to creep up.
Theoretically, long duration bonds should lose value much more so than short dated bonds. While this is more readily apparent with lower rated corporate bonds, it isn’t necessarily true of sovereign bonds such as the US 30-year Treasury bond. What people in my business tend to forget is that the yield curve is dynamic, yields along the curve do not go up in proportion to a rise in central bank interest rates. Sometimes, the short end rises faster and the long end remains static, resulting in a flatter yield curve, or in some cases an inverted yield curve, where short end rates are higher than the long end.
It is for this reason that I keep a close eye on the spread between the 2-year and 30-year Treasury. In other words, the difference in yields of the two points. As we speak the spread is clearly declining, as you can see in the following graph.
Still, asset managers to whom I have presented the model insist on arguing that a cycle of rising interest rates will have a negative impact on longer duration bonds. All I do is remind them of historical performance and tell them they are wrong to blindly trust theory. In fact, of the 6 interest raising cycles (green data line) we have seen in the US since the 1970’s, 4 have resulted in stable or increasing prices for 30-year US Treasury bonds (red data line).
The drop in long dated bonds in late 1970’s to early 1980’s was mostly due to global shock after the US dropped the gold standard. The period between 1994 and 1995 can be attributed to a rebalancing of the sovereign investment grade bond market after the founding of the European Union.
Other than that, every rise in interest rates (green data line) has been met by stable or increasing long bonds prices (red data line), including a flattening of the yield curve (pink data line).
Remember, the only two assets used in my model are the QQQ and TLT ETF’s. They are not held simultaneously. It is simple a rotation between the top 100 stocks in the Nasdaq and a portfolio of US Treasuries with 20+ years to maturity, utilizing passively managed etf’s in an actively managed portfolio. This simple algorithm has provided more than adequate annualized returns, before fees of 12.06% since July 2003. If you would have utilized the underlying assets, which allows for longer back-testing of the algorithm, you would have enjoyed 12.25% annualized returns since November of 1985. This is much better than the 9.79% annualized return in the Nasdaq or the 8.2% annualized return in the S&P 500.