# Head-to-head: A comparative analysis of bond ETFs

By Alex ChamberlinFeb. 24 2014, Published 8:52 a.m. ET

**TLT underperforms SPY**

TLT is the 20+ year Treasury Bond ETF issued by iShares and tracks Barclays U.S. 20+ year Treasury Bond Index. The underlying securities of the index are long-term U.S. Treasuries bonds. Here, we will see how the ETFs have performed through the returns they have yielded for the investors in the past one year. It will also give us an idea of how risky the ETF was, by looking at the movement of its price, that is, tracking its volatility. For the past 52-weeks (one year), the price of the ETF has fluctuated between a maximum of $121.64 to a minimum of $100.68—a change of $20.96. The current price is $106.58. The comparison of returns from TLT—change in its price for different time periods—is given below along with the returns from SPY for three, five, and ten years for comparison.

The fund has a beta of 0.73, that is, if the Barclays U.S. 20+ year Treasury Bond Index moves by 1%, TLT moves by 0.73% in the same direction. It implies that the fund is less volatile than the index it tracks, and so, provides some cushion when the market goes down. However, when the market goes up, it limits the ETF’s rewards as well.

We measure the riskiness of holding the ETF through a measure of volatility called standard deviation. The standard deviation shows how much variation or dispersion from the average exists. Given below is a comparison of volatility measured in terms of SD of TLT with SPY:

Time Period |
TLT standard deviation |
SPY standard deviation |
Average Treasury Rate |
TLT Sharpe Ratio |
SPY Sharpe Ratio |

3 years | 14.45 | 12.37 | 0.54 | 0.61 | 1.06 |

5 years | 13.95 | 15.28 | 1.51 | 0.21 | 1.20 |

10 years | 13.61 | 14.64 | 3.49 | 0.20 | 0.23 |

So, we can conclude that TLT performed well in the medium term (three years), at a marginally higher risk (that is, with high standard deviation), when compared to SPY. For a longer term of five and ten years, TLT underperformed SPY while trading at a slightly lower risk level.

**Sharpe ratio**

The risk–reward ratio can be gauged more accurately through a measure called the Sharpe ratio. It’s a reward-to-risk ratio that measures the excess returns per unit of volatility for holding the investment.

S(X) = (R – R(f)) / SD

Where R is the average return from investment X; R(f) is the best available rate of return of risk-free security (T-bills); SD is the Standard deviation of the return from X.

The higher the Sharpe ratio, the better it is for the investors. Putting it more simply, it’s the excess return % per unit of standard deviation. In our analysis, we have used the average ten-year Treasury yield to find out the Sharpe ratio for ten years for both TLT and SPY. Similarly, for a five-year comparison we used the average five-year Treasury yield; and for three-year term, the three-year Treasury has been considered as risk-free rates. Using the Sharpe ratio formula above (that is, TLT return minus treasury yield and divided by the standard deviation) and applying for each of three, five, and ten-year periods, we find that Sharpe ratio was higher for SPY for all the periods under consideration. In other words, SPY outperformed TLT. Among the TLT returns across the periods, the past three years produced maximum return for each unit of risk.