Comparing DFA- vs. Vanguard-Oriented Portfolios

Transcription

1 INVESTMENT POLICY RESEARCH: Comparing DFA- vs. Vanguard-Oriented Portfolios By Jared Kizer, CFA, Chief Investment Officer August 2018 Our Investment Policy Research articles provide an in-depth look at due diligence that has been completed by the BAM Advisor Services investment policy committee (IPC). We hope you find these articles to be educational and that they offer insight into the policies formed by our IPC. Many investors are interested in how similar funds compare to each other and why one might be preferred over another. This question often arises when the expense ratio (the fund company s compensation for managing the fund) is higher for a fund we recommend than a similar one. The temptation is to believe that all that matters is expense ratio, but that s just a starting point. The BAM Advisor Services IPC has been analyzing long-run fund performance and the academic research on fund performance long enough to know that expense ratio is not the only determinant of long-term performance, even when comparing funds that are playing in roughly the same sandbox.

2 In this piece, we address one of the most common questions we receive: What are the differences between portfolios of Dimensional Fund Advisors (DFA) funds compared to Vanguardoriented portfolios and why would an investor prefer one to the other? Historically, we have primarily recommended DFA s equity strategies, although we also recommend equity funds from Bridgeway Capital Management (U.S. small-cap value) and AQR Capital Management (for U.S. large-cap value). 1 Vanguard is the prominent provider of low-cost index fund strategies and, therefore, a fund family we frequently get asked to compare to the funds we recommend. Structural Differences Between DFA and Vanguard Funds We will tackle specific portfolio performance comparisons, but that exercise is not worth much without understanding the structural differences between the two fund companies. We cannot repeat this statement enough. Many investors start first with relatively short horizon performance comparisons, which aren t worth much either, without understanding how the funds are constructed. In general, when compared to Vanguard, DFA s equity funds tend to: 1. Own smaller stocks 2. Own stocks with lower price/earnings ratios, which we and others refer to as value stocks To be clear, this is not true of every single comparison you could make between DFA and Vanguard s funds but it is certainly true as a general statement. The reason DFA includes these types of stocks is because the academic evidence shows that small-cap stocks, particularly small-cap value stocks, have tended to generate higher returns than large-cap stocks, and value stocks have tended to generate higher returns than growth stocks in U.S., international and emerging markets. DFA has designed many of its strategies to try to capture these long-run phenomena, while this is less true for Vanguard s funds. As a general rule, this means that DFA s funds will tend to outperform Vanguard s during periods where small and value do well and vice versa. There can certainly be other reasons for differences in short-term performance for example, differences in sector allocations and possibly even differences in individual stock allocations but over longer periods, the differences in size of companies held and the valuation of asked to compare to the funds we recommend. Examining Long-Run Performance We tackle longer-run performance comparisons of DFA and Vanguard portfolios using the following methodology: 1. Separately comparing the performance of U.S., international and emerging markets equity fund portfolios. 2. Comparing the performance of DFA portfolios to Vanguard portfolios that are essentially marketcapitalization weighted. Market-capitalization weighted portfolios will have larger weights in the largest stocks. 3. Comparing the performance of a DFA U.S. equity portfolio to a Vanguard U.S. equity portfolio that tries to more closely match the small-cap and value characteristics of the DFA portfolio. Note: It s not possible to build an international or emerging markets Vanguard portfolio with this methodology, which is one of the main differences between the two fund companies. Vanguard does not have international or emerging markets equity index funds that allow for a substantial tilt toward small and value companies. Each of the three comparisons helps surface important aspects that can help investors understand the differences between the companies funds. For example, if an investor is perfectly comfortable owning a market-capitalization weighted global equity portfolio and expects it can achieve his risk and return goals, there s probably no reason to own any of DFA s equity strategies. Vanguard s market capitalization weighted funds are extremely low cost and do a great job of capturing of global equity market returns. If, however, an investor is open to tilting her portfolio toward small-cap and value stocks to attempt to capture higher long-run expected returns, it s helpful to know how such an approach has compared to a Vanguard marketcapitalization weighted portfolio. Further, as point three notes above, there s no way to build a globally diversified small-cap and value tilted Vanguard portfolio of index funds. It s possible to do this in U.S. equities but not in international and emerging market equities. Long-Term Performance Comparisons To keep things simple and to avoid being accused of data mining, all DFA portfolios are equally weighted and use highly diversified funds covering the major asset classes. The U.S. equity DFA portfolio is composed as follows: 25% DFA U.S. Large Company fund (DFUSX) 25% DFA U.S. Large-Cap Value III fund (DFUVX) 25% DFA U.S. Micro-Cap fund (DFSCX) 25% DFA U.S. Small-Cap Value fund (DFSVX) 1 BAM also manages most fixed income holdings in-house and recommends AQR and Stone Ridge for alternative investment strategies. 2

3 The international portfolio is composed as follows: 25% DFA International Large-Cap fund (DFALX) 25% DFA International Large-Cap Value III fund (DFVIX) 25% DFA International Small Company fund (DFISX) 25% DFA International Small-Cap Value fund (DISVX) The emerging markets portfolio is allocated as follows: 33% DFA Emerging Markets fund (DFEMX) 33% DFA Emerging Markets Small-Cap fund (DEMSX) 33% DFA Emerging Markets Value fund (DFEVX) For the Vanguard market-cap weighted portfolio, I use Vanguard Total Stock Market Index fund (VTSMX and VTSAX) for U.S. equity, Vanguard Developed Markets Index fund (VTMGX) for international equity and Vanguard Emerging Markets Index fund (VEIEX and VEMAX) for emerging markets equity. DFUSX s first full month of returns history was October 1999, so that drives the starting month in this particular analysis. In this comparison, we see that the DFA portfolio did generate higher compound returns than the Vanguard market-capitalization weighted approach. Volatility was right at 2 percent per year higher with risk-adjusted returns higher for the portfolio of DFA funds compared to Vanguard. Let s now look graphically at the year-by-year differences in performance between the two portfolios from 2000 through Figure 1: Year-by-Year Differences in U.S. Equity Portfolio Performance ( ) Note that in two of the cases, we list two tickers for Vanguard, one for the Investor share class of each fund and one for the Admiral share class of each fund. The Investor share class typically has a longer performance history, allowing us to extend the performance analysis but we switch over to the lower expense Admiral share class when it becomes available. All portfolios are rebalanced monthly, and each performance comparison goes back in time as far as possible. DFA Portfolios vs. Market-Capitalization Weighted Vanguard Portfolios U.S. Equity To make the data portion of this article more digestible, we ll cover the analytical results for U.S., international and emerging markets equity in turn. Table 1 relates the compound returns, volatility and Sharpe ratios for the U.S. equity comparison. Table 1: U.S. Equity Performance Comparison (10/1999 6/2018) DFA U.S. Vanguard U.S. Compound Return Volatility Sharpe Ratio In the U.S. equity comparison, this period was characterized by significant outperformance of the DFA portfolio compared to Vanguard in the early 2000s, followed by very similar performance starting in 2007 and after even though year-byyear differences were sometimes large. The average difference in performance from was +9.0 percent per year while it was 0.2 percent per year from Not surprisingly, these two sub-periods line up with periods that saw, respectively, very good performance of small-cap and value stocks compared to large-cap and growth, and similar performance of small-cap and value relative to large-cap and growth. The latter portion of the period also illustrates the well-known fact that strategies that tend to work over the very long term can (and will) go through long periods of flat or even negative performance without that being an indication the strategies no longer work. International Equity Table 2 replicates Table 1 except now comparing international equity fund portfolios instead of U.S. equity. 3

4 Table 2: International Equity Performance Comparison (9/1999 6/2018) DFA Int'l Vanguard Int'l Compound Return Volatility Sharpe Ratio Similar to the U.S. equity portfolio comparison, we see substantial outperformance of the DFA portfolio compared to Vanguard. However, here we also see similar volatility, indicating that unlike the U.S. equity market, a substantial tilt toward size and value did not increase portfolio volatility. Consequently, the percentage increase in Sharpe ratio is larger here compared to Table 1. Figure 2 presents the year-by-year differences in performance over the same period. Figure 2: Year-by-Year Differences in International Equity Portfolio Performance ( ) Table 3: Emerging Markets Equity Performance Comparison (5/1998 6/2018) DFA EM Vanguard EM Compound Return Volatility Sharpe Ratio Surprisingly, of the three equity markets, we are able to take this comparison back the longest, slightly longer than the 1999 start years for the U.S. and international equity comparisons. The results are most similar to the international equity result, higher realized return with similar volatility. Figure 3 plots the year-by-year performance differences. Figure 3: Year-by-Year Differences in Emerging Markets Equity Portfolio Performance ( ) Figure 2 shows a much more stable outperformance comparison compared to Figure 1. The DFA portfolio outperformed the Vanguard portfolio in all but four years. This shows that perhaps the cross-country diversification of size and value in tandem with possibly stronger size and value premia outside the U.S. tends to lead to a more stable performance pattern. It also illustrates one of the key differences that remains between DFA and Vanguard: It s impossible to get the depth of size and value tilts with Vanguard international equity index funds that are possible with DFA s funds. Emerging Markets Equity Table 3 relays the comparison for emerging markets equity portfolios Similar to the international equity comparison, we see a bit more stable performance pattern when compared to the U.S. result. The U.S. result was highly concentrated in the early period, while we do not see that to the same degree with either the emerging markets equity result in Figure 3 or the international equity result in Figure 2. This could again be due to the depth of cross-country diversification of the size and value premia and/or the fact that the size and value premia naturally tend to be stronger in international and emerging markets equity. The overall conclusions seem to be: Over the longest histories we have to work with, size- and value-tilted DFA portfolios have generated higher compound returns and risk-adjusted returns compared to a market-capitalization weighted Vanguard approach

5 2. The consistency of these results has been stronger in international and emerging markets equity compared to U.S. equity. The U.S. result is driven strongly by performance in the early 2000s, while outperformance has been more stable over time in the case of the international and emerging markets equity comparisons. 3. There are still, however, numerous years where the size- and value-tilted portfolios underperform, which is to be expected from any portfolio strategy (i.e., there are not strategies that always work even over relatively long periods of time). 4. Year-by-year differences to the positive or the negative can be large. The magnitude of this tracking error can be reduced by decreasing the depth of the size and value tilt, although this likely reduces long-term expected return as well. Comparing Similar U.S. Equity Portfolios In U.S. equities, it is possible to build Vanguard portfolios that have roughly similar size and value tilts compared to most DFA-oriented portfolios. In this section, we make this comparison but again emphasize that this is not possible to do in international and emerging markets equity. For this analysis, the DFA U.S. equity portfolio retains the same construction; therefore, the performance period remains the same (10/1999 6/2018): 25% DFA U.S. Large Company fund (DFUSX) 25% DFA U.S. Large-Cap Value III fund (DFUVX) 25% DFA U.S. Micro-Cap fund (DFSCX) 25% DFA U.S. Small-Cap Value fund (DFSVX) We use an optimization routine to identify the allocation between Vanguard Total Stock Market and Vanguard Small- Cap Value that allows the Vanguard portfolio s historical returns to most closely match the DFA portfolio. The result of this historical optimization analysis is a Vanguard portfolio that allocates 25 percent to Vanguard Total Stock Market and 75 percent to Vanguard Small-Cap Value. Table 4 relays the performance information for these two portfolios. Table 4: U.S. Equity Performance Comparison (10/1999 6/2018) DFA U.S. Vanguard U.S. Compound Return Volatility Sharpe Ratio This analysis identified a Vanguard portfolio that experienced very similar performance and had similar size and value tilts compared to the DFA portfolio. However, it required an allocation that may look odd to many investors, with threequarters of the portfolio to a single fund. Nevertheless, this is what s required to closely match the size and value tilts of the DFA portfolio. However, the two portfolios still do not experience identical performance on a year-to-year basis. Figure 4 makes this point Figure 4: Year-by-Year Differences in U.S. Equity Portfolio Performance ( ) To get a more accurate comparison, the Vanguard portfolio uses two different funds: Vanguard Total Stock Market Index fund (VTSMX and VTSAX) Vanguard Small-Cap Value Index fund (VISVX and VSIAX) We took this approach because the Vanguard U.S. equity portfolio requires a very large allocation to Vanguard Small- Cap Value to come close to the size and value tilts in the DFA portfolio. There are a number of reasons for this. First, the DFA portfolio includes two funds, DFA Micro-Cap and DFA Small-Cap Value, that have historically had exposure to very small companies and, in the case of DFA Small-Cap Value, a very deep tilt toward value. Second, Vanguard Small-Cap Value has historically not had the depth of tilt toward smallcap or value that DFA Small-Cap Value has. Consequently, to best match the characteristics of the DFA portfolio, the Vanguard portfolio requires a very large percentage allocation to Vanguard Small-Cap Value The annual return differences are obviously smaller than those in Figure 1, but we still see years where the return differences are large. In 2000, the DFA portfolio underperformed by 10.2 percent, while it outperformed by 9.5 percent in

6 The overall takeaway from this section is that it is possible to build a U.S. equity Vanguard portfolio that will come close to replicating the size and value tilts in DFA-oriented portfolios but at allocation percentages that seem askew for some investors. Summary Historically, DFA equity portfolios have generated higher returns and better risk-adjusted returns than Vanguard market-capitalization weighted portfolios. This performance has been more consistent in international and emerging markets equities compared to U.S. equities, but there are stretches in each of the three markets when the Vanguard strategy outperforms, corresponding to periods where largecap and growth are doing well relative to small-cap and value. Further, performance differences in any particular period can be large. In international and emerging markets equities, it s not possible to build Vanguard portfolios that match the size and value characteristics of DFA portfolios due to Vanguard s lack of small-cap and value-oriented index funds in those markets. In U.S. equities, one can build Vanguard portfolios that approximate the size and value tilts, but this requires a very large percentage allocation to Vanguard s Small-Cap Value fund, and performance of the Vanguard portfolio may still be significantly different from the DFA portfolio over shorter periods of time. Copyright 2018 The BAM ALLIANCE. This document and all attachments are provided to serve as a resource for clients within the BAM ALLIANCE. We hope you will use it to keep abreast of emerging research and investment trends, take advantage of upto-date advice, and revisit the principles on which the relationship with your advisor was forged. The opinions expressed by featured authors are their own and may not accurately reflect those of the BAM ALLIANCE. This article is for general information only and is not intended to serve as specific financial, accounting or tax advice. Past performance is not a guarantee of future results. Information regarding references to third-party sites: Referenced third-party sites are not under our control, and we are not responsible for the contents of any linked site or any link contained in a linked site, or any changes or updates to such sites. Any link provided to you is only as a convenience, and the inclusion of any link does not imply our endorsement of the site. BAM makes no warranties, expressed or implied, as to accuracy, completeness, or results obtained from any information on this report. 6