Swedroe: Remember To Look Under The Hood

We recently took a look under the hood of three similar, but different, passively managed U.S. small value funds—Vanguard’s Small-Cap Value ETF (VBR | A-100) and two structured portfolios: one from Dimensional Fund Advisors (DFSVX) and one from Bridgeway (BOSVX). (Full disclosure: My firm, Buckingham, recommends Dimensional and Bridgeway funds in constructing client portfolios.)

In that comparison, we noted one of the differences between VBR and the two other funds is that VBR—an index fund—includes REITs in its holdings, while DFSVX and BOSVX don’t. We discussed how the inclusion, or exclusion, of REITs significantly impacted the relative performance of the three funds in 2013 (when REITS dramatically underperformed small value stocks) and in 2014 (when REITS dramatically outperformed small value stocks).

Finally, we addressed some of the reasons a fund would consider excluding REITs: they’re tax inefficient; they have lower expected returns than small value stocks; and investors may already have exposure to REITs in other holdings. What’s more, if an investor wants exposure to REITs, they can own them directly in a REIT index fund and determine the amount of exposure they want.

The REIT Effect

Today we’ll see how the inclusion or exclusion of REITs impacted longer-term results. To begin, we’ll examine the performance of VBR and the Dimensional Fund Advisors (DFA) U.S. Small Cap Value Fund, DFSVX, over the 15-year period from 2000 through 2014.

During this period, VBR returned 10.70 percent per year, while DFSVX returned 11.62 percent. Note that over this period, the annualized return of the size factor was 3.6 percent and the annualized return of the value factor was 4.9 percent. Thus, you would expect DFSVX to have outperformed, because it has higher loadings on the size and value factors.

I would further note that during the most recent 15-year period, the annualized factors delivered somewhat higher returns than their long-term record indicates. From 1927 through 2014, for example, the annualized size factor was 2.2 percent and the annualized value factor was 4.0 percent.

VBR currently has about a 10 percent allocation to REITs. To determine how DFSVX’s exclusion of REITs impacted the relative performance of the funds during this period, we’ll now analyze the results from a portfolio allocated 90 percent to DFSVX and 10 percent to DFA’s REIT fund (DFREX). For investors who wanted exposure to REITs, this would be the way to replicate the holdings in VBR. It also makes the comparison of the two funds’ performance more of an apples-to-apples one.

From 2000 through 2014, DFREX returned 12.60 percent, outperforming DFSVX by almost a full percentage point. The annually rebalanced portfolio containing the two DFA funds would have returned 11.81 percent—1.11 percentage points more than the return of VBR.

Longer-Term Performance

We’ll now look at the funds’ performance over the last 10 calendar years, 2005 through 2014. During this time frame, the annualized return of the size factor was just 1.1 percent and the annualized return of the value factor was just 0.3 percent.

VBR returned 8.50 percent, DFSVX returned 7.93 percent and DFREX returned 8.14 percent. The DFA portfolio allocated 90 percent to small value and 10 percent to REITs, and rebalanced annually, would have returned 8.05 percent.

VBR outperformed the DFA portfolio by 0.45 percentage points, virtually matching the difference between the current expense ratios of the two portfolios (0.41 percent). The current expense ratios are: VBR, 0.09 percent; DFSVX, 0.52 percent; DFREX, 0.18 percent; and the 90/10 DFA portfolio, 0.49 percent. Also during this period, the size and value premiums were much smaller than their long-term averages.

Our final look will be at the most recent five calendar years: 2010 through 2014. In this period, the annualized size factor return was 1.2 percent and the annualized value factor return was -1.1 percent. During this period, VBR returned 16.57 percent, DFSVX returned 16.77 percent and DFREX returned 16.97 percent. The DFA portfolio allocated 90 percent to small value and 10 percent to REITs, and rebalanced annually, returned 16.99 percent.

In this case, the returns of the Vanguard fund, the DFA funds and the DFA 90/10 portfolio were very similar. This is about what we would expect, because the size and value premiums canceled each other out, and the return on REITs was only slightly higher than the return on small value stocks.

Conclusions

What conclusions can we draw? The first is that asset pricing models, while not perfect replications of the way the world works, do a very good job of explaining returns.

Second, if you expect that, over the long term, the size and value premiums will be about their historical averages and REITs will underperform small value stocks (from 1978 through 2014 the Dow Jones Select REIT Index returned 12.7 percent versus the 15.2 percent return for the Fama-French U.S. Small Value [ex-utilities] Index), then you should also expect that DFSVX will outperform VBR, due to its higher loadings on these factors.

On the other hand, we should also expect that there will be periods, possibly even very long ones, when the small and value factors will provide negative returns and REITs will outperform. During such periods, you would expect DFSVX to underperform. What you want to avoid is the mistake of chasing performance. Investment decisions should be based on long-term evidence, not recent performance.

The bottom line is that to avoid making mistakes in analyzing the performance of similar funds, even two passively managed funds in the same asset class, it’s important to ensure you’ve looked under the hood and understand each fund’s sources of returns.

Larry Swedroe is the director of research for the BAM Alliance, a community of more than 150 independent registered investment advisors throughout the country.