Even in the current low interest rate environment when the Fed funds target rate is at 0.25% and the10 year yield is sub 3%, US treasuries remain an integral part in most portfolios. With never-ending speculation when the Fed is going to start tapering its asset purchase program, it is not an easy task for investors to decide upon the duration of the bond component in their portfolios. In his recent article, the "Bond King" Bill Gross argues that investors should focus on front end maturity positions and avoid anything longer than 7-10 years. Having read this suggestion, I decided to check what impact on portfolio risk parameters such a change would have, assuming that a starting point is a standard 60/40 portfolio that is invested in stocks (NYSEARCA:SPY) and long-dated treasuries (NYSEARCA:TLT).
To begin with, let's take look at the risk parameters for the SPY/TLT version of the 60/40 portfolio. A quick check on a freely available investor tool InvestSpy reveals that over the last 5 years, such a portfolio had an annualized volatility of 10.3% and a beta coefficient of 0.43:
Now let's see what happens if the maturity of the bond component is shortened to 7-10 years (NYSEARCA:IEF) or even to 1-3 years (NYSEARCA:SHY). In the case of IEF, the volatility inches up to 10.7%, whilst beta increases by 0.10 to 0.53. The same risk metrics for the SPY/SHY portfolio are 11.6% and 0.59, respectively.
The first observation is the substantial increase in beta, which means that the portfolio gets more reliant on equity market performance. This is due to the fact that longer dated treasuries are more negatively correlated with stocks, and this benefit reduces when TLT is replaced with IEF or SHY. Another point of note is that the total portfolio risk edges higher as the portfolio becomes less efficiently diversified, pushing the volatility estimate 1.3% higher in annualized terms. This may not sound like a substantial change, but to maintain the Sharpe ratio unchanged, depending on assumptions, the lower yielding SHY should deliver roughly 1% better total return than TLT. This is not to say that Bill Gross' suggestion is wrong, but more to remind investors that long-dated treasuries still act as an efficient diversification tool in equities-loaded portfolios.