I got burned (sort of)
My current retirement portfolio is simple: roughly one-third ProShares UltraPro S&P 500 (NYSEARCA:UPRO), a 3x daily S&P 500 ETF, and two-thirds Vanguard Long-Term Bond Index Fund (MUTF:VBLTX). VBLTX is there to generate positive alpha and help the portfolio beat the S&P 500 by a few percentage points a year on average.
Unfortunately, VBLTX fell approximately 12% over the past five months, as U.S. Treasury yields rose. It's still up 5.1% year to date, but a 12% drawdown is enough to make me consider other options.
The purpose of this article is to look at metrics of performance for Vanguard's lineup of bond mutual funds going back to the beginning of 2010.
Pool of bond funds
Vanguard lists a total of 37 bond mutual funds on its website. You can find basic information on each fund there. Briefly, expense ratios are all very low, ranging from 0.06% to 0.25%; minimum initial investments are all $3,000 or $10,000; and two of the funds have purchase fees (0.25% for VICSX, 1% for VLTCX). Four of the funds were introduced after the start of 2010 and were therefore excluded.
CAGR vs. MDD
Perhaps the most important performance metrics are compound annual growth rate and maximum drawdown. Figure 1 shows CAGR vs. MDD for the 33 Vanguard bond funds, using data from January 4, 2010, to December 2, 2016.
As is often the case, funds that generated more growth tended to be susceptible to large drawdowns. As a holder of VBLTX, it strikes me that the junk-bond fund VWEHX (orange dot) had the same CAGR as VBLTX (both 7.1%), but a much smaller MDD (8.6% vs. 14.5%).
Mean vs. SD
Figure 2 shows mean vs. standard deviation of daily gains for the various funds.
Again, VWEHX jumps out at me as having a particularly favorable position on the graph. I drew in a gray dotted line to show that it dominates all funds except VSCSX and VWAHX. By that I mean you could combine VWEHX with cash to move on the dotted line towards (0, 0), and along the way, you would beat out every fund except these two in terms of maximizing expected returns for a given level of tolerated volatility. Equivalently, VWEHX has a higher Sharpe ratio than all funds except VSCSX and VWAHX.
The cluster of five long-term bond funds at the top-right is also noteworthy, in that their extra expected returns are probably not worth the inflated volatility. For example, VWESX's mean was 15.2% higher than VWEHX's, while its standard deviation was 256.4% higher.
Alpha vs. beta
For portfolio building, it's essential to have a feel for how much alpha a fund generates, and whether its beta is positive, negative, or neutral. Figure 3 shows alpha vs. beta for the Vanguard funds, using daily gains and Vanguard's S&P 500 index fund, VFINX, as the benchmark.
VWEHX differentiates itself here as the only fund with a positive beta. It generated less alpha than eight of the other funds.
The long-term treasury fund VUSTX was notable in that it had the highest alpha and most negative beta of all the funds.
The cluster of five funds at the top-right of Figure 2, which I gauged as having too much volatility for their level of expected returns, appear in a cluster again on this graph, at the top-left. High alpha and very negative beta bode well for these funds in terms of how they likely perform in conjunction with equities.
Looking at all 33 Vanguard bond mutual funds operating since the beginning of 2010, I would say that the junk-bond fund VWEHX is most impressive. It's hard to beat 7.1% annualized growth with a maximum drawdown of just 8.6%. It also had the third-highest Sharpe ratio, behind two much more conservative funds (VSCSX with 3.0% CAGR, VWAHX with 4.8% CAGR).
VWEHX's performance going back to January 2, 1980, is even more impressive: 9.1% CAGR (VFINX: 10.4%), Sharpe ratio of 0.091 (VFINX: 0.041), alpha of 0.031% and a beta of 0.039%. It experienced a 30.2% drawdown during the financial crisis (VFINX: 55.3%).
Overall, I have to say I favor VWEHX, with its long track record of excellent performance. My only concern is its positive beta, as negative-beta funds generally pair better with equities.
As for my current fund, VBLTX was outperformed by several other funds. VWESX had a meaningfully better CAGR (7.7% vs. 7.1%) and MDD (13.5% vs. 14.5%), and VWEHX had essentially the same CAGR but drastically lower MDD (8.6% vs. 14.5%).
Despite this apparent outperformance, I do wish to pair my bond fund with an equities fund, so I will stick with one of the five funds clustered at the top-left of the alpha vs. beta graph. I will likely either keep VBLTX or switch to the long-term investment-grade fund VWESX.
Disclosure: I am/we are long VBLTX.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Additional disclosure: The author used Yahoo! Finance to obtain historical stock prices and used R to analyze the data and generate figures. Any opinion, findings, and conclusions or recommendations expressed in this material are those of the author and do not necessarily reflect the views of the National Science Foundation.