Warning: Bond Funds That Act Like Stocks
Question: When is a bond fund not a bond fund? Answer: When it behaves more like a stock fund.
A variety of bond funds are promising flexibility, a multisector or alternative approach, or the latitude of an unconstrained strategy. There have been some successes in this realm. But is it still a bond fund if it behaves more like a stock fund?
Of course, I’m assuming bond and stock funds are combined in a diversified portfolio. From that perspective, I analyzed bond funds from three categories to assess how well they interacted with the stock portion of the portfolio. I wasn’t really interested in how well the bond fund performed on its own, but rather in how well it contributed to the overall success of a 60% stock/40% bond portfolio.
An important element of this analysis was examining the correlation of the various bond funds with the stock fund used in this study — SPDR S&P 500 (SPY) — an ETF that mimics the S&P 500. As a benchmark bond fund, I used Vanguard Total Bond Market ETF (BND).
The correlation of annual returns between SPY and BND from Jan. 1, 2008, to Dec. 31, 2014, was -0.38 (using NAV-based returns). This negative correlation is precisely what we anticipate between stock and bond funds: When stocks have negative returns, bonds tend to have steady positive returns.
When the correlation between stock and bond funds is positive, it indicates that the bond funds are behaving more like stock funds. If diversification of return patterns is what is wanted, using a bond fund that actually behaves like a stock fund will likely not be helpful in the long run — or even the short term.
Before getting into the results, we must establish a benchmark of performance. As already noted, in this analysis we used a 60/40 portfolio with SPY as the stock component and BND as the bond component. The annual returns of the 60/40 mix (assuming annual rebalancing) are shown in the table below.
It’s worth noting the performance of SPY in 2008 was -36.97%; BND saw a 5.18% return that year, and the 60/40 combination was at -20.11% for the year. Over the seven-year period from 2008 to 2014, the annualized return for the 60/40 combination was 7.01%, with a standard deviation of 13.21%.
The first category of bond funds I examined is described by Lipper as alternative credit, but, as you can see in the table, the term unconstrained is also used in the name of several funds. These funds tend to invest tactically in a wide variety of bond sectors that may include high-yield or non-U.S. bonds. Duration will also be tactically managed.
Each of these 10 bond funds was paired with SPY to determine if the bond fund added or subtracted value to the benchmark 60/40 performance.
The most striking observation was the correlation between the alternative bond funds and SPY. In every case but one, the correlation was high and positive — indicating the alternative bond fund was behaving more like SPY than BND. Only FPA New Income (FPNIX) had a negative correlation with SPY, and, not surprisingly, the 2008 performance of the SPY/FPA combo was very similar to the benchmark performance.
The other alternative bond funds, when teamed with SPY, actually made things worse in 2008. Indeed, all but two of the alternative bond funds by themselves actually had losses that year, the worst being
AllianceBernstein Unconstrained Bond Fund (AGSAX), at -16.52%.
Of the 10 alternative bond funds, only four of them — Iron Strategic Income Fund (IFUNX), Deutsche Unconstrained Income Fund (KSTAX), MainStay Unconstrained Bond Fund (MASAX) and Dreyfus Opportunistic Fixed Income Fund (DSTAX) — led to a higher seven-year annualized return when used in a 60/40 portfolio. But the slight performance advantage came at a cost: worse 60/40 performance in 2008 and a higher standard deviation (or volatility) of returns compared with the benchmark 60/40 portfolio.
The next category of bond funds is referred to by Lipper as flexible income — a group that also invests in a wide variety of bonds, from Treasuries to high yield to corporates.
As you can see in the table, these types of bond funds also behave a lot more like stock funds. Note the exceptionally high positive correlations of annual returns between each respective fund and SPY, with the exception of Cutler Fixed Income (CALFX) — which had a lower but still positive correlation.
This high correlation means there was no downside protection in 2008. Indeed, Northeast Investors (NTHEX) itself had a NAV return of -37.29% in 2008 — slightly worse than the -36.97% return of SPY. When it was teamed with SPY in a 60/40 mix the result was a loss of 37.10% in 2008. The seven-year performance using this category of funds was better than that of the 60/40 benchmark in all but two cases over this particular period — but that could quickly change if equities nosedive.
This group of flexible bond funds, by themselves, had an average return of 36.2% in 2009, whereas benchmark bond component BND had a return of 6.03%.
Flexible income bond funds are a different animal — use at your own risk.
The last category of avant-garde bond funds is multisector funds, which like the other categories can hold a wide variety of bonds. These funds will often hold a sizable percentage of lower-rated bonds.
The analysis showed a similar story to what we saw in flexible income funds: high correlation with stock funds, which led to large losses in 2008 compared with the benchmark 60/40 portfolio.
The lower section of the table shows the largest 10 out of 36 multisector bond funds, based on assets. Eight of the 60/40 SPY/multisector bond fund combinations had a higher seven-year performance than the benchmark 60/40 portfolio, but in all but one case they experienced larger losses in 2008 and higher volatility.
The only one of these funds that actually improved performance for the 60/40 portfolio both in 2008 and over the seven-year period was PIMCO Fixed Income Shares Series C (FXICX).
The odd fund in the group was Free Market Fixed Income (FMFIX): It had a negative correlation with SPY, but that negative correlation came at the cost of materially lower performance for the 60/40 mix over the seven-year period.
The real takeaway for advisors is this: When it comes to these bond fund categories, investor beware.
If you want a bond fund that largely acts like a stock fund, you have choices. But don’t expect that hybrid bond fund to behave like a traditional bond fund when equity markets get rocky. Labels such as unconstrained, flexible or multisector may simply mean these funds may turn on you when a traditional bond fund doesn’t.
Craig L. Israelsen, a Financial Planning contributing writer in Springville, Utah, is an executive in residence in the personal financial planning program in the Woodbury School of Business at Utah Valley University. He is also the developer of the 7Twelve Portfolio.