|
About IndexInvestor.com |
Privacy Policy |
Transaction Policy |
Legal Disclaimers |
Contact Us |
My Account |
Home
|
|||||
|
|
|
|
|
||
Life is filled with great dilemmas: boxers or briefs? Short skirt or long? Habs or Leafs? Man-U or Arsenal? Tokyo or Osaka? Aussie Rules, League, or Union? French or Italian food? And, of course, index mutual funds from Vanguard or Dimensional Fund Advisors?
Along with State Street Global Advisors and Barclays Global Investors, Vanguard and DFA are two of the world's leading managers of index investment products. Compared to the other three firms, DFA is in some ways unique. It has the strongest emphasis on indexed products, and perhaps the strongest association with very well-known academics, including Rex Sinquefield, Gene Fama, and Ken French. Moreover, there is a certain mystique about its retail mutual funds, which are only available through a select group of financial advisers. But is there anything to this, besides great marketing (which, of course, is nothing to sneeze at)? We've lost count of the number of times we've been asked this question. And that's why we've done the analysis in this article -- to see if we can settle the question once and for all. Our main approach will be an asset class by asset class comparison of the mutual fund products offered by DFA and Vanguard. We have deliberately left ETFs (and their main sponsors, BGI and SSGA) out of this analysis, because we wanted to do a mutual fund to mutual fund, apples to apples comparison.
We should also say up front that what we are doing, in essence, is comparing one all-star to another. Both DFA and Vanguard have well earned reputations for holding down their fund expenses, and for using their size and skill to limit their trading costs (in a recent survey of asset management firms with the lowest average transaction costs, both DFA and Vanguard ranked in the top ten). Moreover, at the margin, both firms also take actions to slightly enhance returns, including lending shares and departing occasionally from the underlying index weightings. As DFA notes in its prospectus, "rather than replicate an index in mechanical fashion, we allow slight variations from precise market weightings. This flexibility allows us to take advantage of favorable trading costs."On the other hand, there are also some important differences between the two firms. Compared to Vanguard, DFA is a much stronger advocate of the wisdom of using small cap and value tilts within different equity asset classes (e.g., domestic, foreign, and emerging market). We have written before about the wisdom of taking these tilts (our articles on these subjects can be easily accessed via the home page of our website). To briefly sum them up, there are three issues: (1) Does the small size and/or value premium exist? (2) If it does, what has caused it in the past? And (3) will that cause persist in the future? With respect to the size premium, we have noted our doubts about its existence, except in the case of microcap stocks (generally, stocks included in the bottom 2% or so of total market capitalization). With respect to the value premium, while the evidence for its existence seems compelling, its underlying cause remains unclear.
One school of thought (and DFA is in this camp), believes that the value premium reflects an efficient market delivering higher returns for bearing higher risk than is found in the broad equity market index. Unfortunately, different academics have yet to reach agreement on the nature of this additional risk. In contrast, the other school of thought believes that the value premium is a behavioral phenomenon that results from defects in the way investors process information. As such, they believe that by taking a value tilt it may be possible to earn higher returns than the broad market index, while taking on less risk. However, the validity of this argument necessarily depends on the existence of what are called "barriers to arbitrage." Theoretically, not all investors in the market should act irrationally. Hence, some smart investors should recognize the mistake that the irrational ones are making, and bid up the price of value stocks to the point that the expected additional return premium disappears. If you believe that the value premium is likely to persist into the future, you also have to believe in the continued existence of some very powerful barriers to arbitrage. Unfortunately, advocates of the "behavioral explanation" for the value premium have yet to make a convincing case to support this second argument.
It is interesting to note that DFA states that while its definition of value stocks is primarily based on the book/market ratio (consistent with Fama and French's research), it also notes that it may use other screening criteria, including price/cash flow and price/earnings, "as well as economic conditions and developments in the issuer's industry." Moreover, DFA's "criteria for assessing value are subject to change from time to time." In comparison, Vanguard uses indexes from Morgan Stanley Capital International (MSCI) in many of its funds. MSCI uses three criteria (book/market, price/earnings, and dividend/price to identify value stocks).
Over the long-term, we come down on the efficient market side of the argument, while recognizing that some investors can and do occasionally act irrationally. However, we find it hard to believe in a free lunch that lasts forever. In short, while taking a value tilt will, over the long-term, probably produce higher returns than the broad market index, it will also expose an investor to more risk, of one kind or another.
In talking about DFA's domestic equity funds, one of the terms you occasionally hear is the "CRSP Index. Before getting into our fund comparison, it will help to explain this index a bit more. As we have noted in other articles, when it comes to constructing an equity index, there are two basic approaches one can take. Either one include a fixed number of companies in the index, and vary the percentage of total market capitalization it covers, or one can take the opposite approach, targeting coverage of a fixed percentage of market cap, and letting the number of companies vary to achieve it. Indexes that start with a fixed number of companies (ranked by market capitalization) include those from Russell (e.g., the Russell 3000 Index), Standard and Poor's (e.g., the Standard and Poor's 500 Index), and Morgan Stanley Capital International (e.g., the MSCI Prime Market 750 Index). Indexes that start with a fixed percentage of market capitalization include those from Wilshire (e.g., the Wilshire 5,000, which covers 100% of market capitalization), Dow Jones (e.g., the Dow Jones Total Market Index covers 95% of market capitalization) and Morningstar (whose broad index covers 97% of market capitalization).
To put it charitably, the CRSP (which stands for the Center for Research in Securities Prices) takes a hybrid approach. It starts with the companies listed on the New York Stock Exchange, ranked by market capitalization, and divides them into ten equal groups (e.g., 178 companies in each group). Next it determines the market cap "breakpoints" for each group (that is, the high and low market capitalizations that define each group's boundaries). Using these breakpoints, it then assigns companies from the American Stock Exchange and National Association of Securities Dealers Automated Quote System (the NASDAQ) to different groups, which it calls "deciles." Stocks in deciles 1 and 2 are often called "large caps", those in deciles 3 to 5, "mid-caps", those in deciles "6 to 8, "small-caps", and those in declies 9 and 10, "micro-caps." Unfortunately, this can easily create confusion, because the "deciles" contain neither equal percentages of total market capitalization, nor equal numbers of companies. The 1996 example shown on the CRSP website shows that the top decile contained 203 companies that accounted for 58.6% of total market capitalization, while the tenth decile contained 2,426 companies that accounted for 1.3% of total market capitalization. Confusing, no?
Also confusing (though "interesting" might be a better word) is DFA's description of its approach to market capitalization weighting in its small company, real estate, and international funds: "Market capitalization weighting means each security is generally purchased based on the issuer's relative market capitalization. Market capitalization will be adjusted by [DFA] for a variety of factors. [DFA] may consider factors such as free float, trading strategies, liquidity management and other factors determined to be appropriate by [DFA] given market conditions. [DFA] may exclude the stock of a company that meets applicable market capitalization criteria if [DFA] determines, in its best judgment, that the purchase of such stock is inappropriate in light of other conditions. These adjustments will result in a deviation from traditional market capitalization weighting." As a result, DFA notes that "the weightings of certain countries…may vary from their weightings in international indices, such as those published by …Morgan Stanley Capital International." In other words, it appears as though there might be a little bit of active management going on at DFA to improve some of its funds’ performance.
One other issue that we need to address before going to the fund comparisons is financial adviser fees. As we noted, the only way an individual can invest in DFA funds is through a financial adviser. Vanguard funds can be directly purchased by individuals without having to go through a financial adviser. A survey done for DFA (available on its website) showed that "91.2% of DFA advisers charge clients a 1% annual fee on accounts up to $1 million, as opposed to a flat fee." Given this, we have decided to present DFA fund results in three different ways: before fund expenses, after fund expenses only, and after fund expenses plus a 1% adviser fee.
Large Capitalization Equity
DFA and Vanguard both offer S&P 500 Index Funds. Apart from the DFA adviser fee, they are virtually identical. DFA also offers an "enhanced" S&P 500 index fund, which attempts to use a combination of equity futures and debt to deliver slightly more return than the S&P 500 with slightly less risk. It charges more than twice the expense load as DFA's basic S&P 500 fund, but delivers virtually the same performance as the Vanguard product. Note that all the data in the following table are in nominal (i.e., including inflation) terms:
| Fund Ticker | Average Return, 5 Years Ended Oct04 (higher and positive is better) | Average Standard Deviation, 5 years Ended Oct04 (lower is better) | Gross Return/Standard Deviation (higher is better) |
Annual Expense Charge (lower is better) | Average Return Before Expenses | Average Return After Adviser Fee (DFA Only) |
| DFLCX |
(2.37%)
|
16.20%
|
(.15)
|
0.15%
|
(2.22%)
|
(3.37%)
|
| VFINX |
(2.28%)
|
16.23%
|
(.14)
|
0.18%
|
(2.10%)
|
|
| DFELX (enhanced) |
(2.07%)
|
15.98%
|
(.13)
|
0.36%
|
(1.71%)
|
(3.07%)
|
Vanguard also offers another large capitalization equity index fund. This one tracks the MSCI Prime Market 750 Index, which covers about 86% of the total capitalization of the U.S. public equity market. Because this fund recently switched to the MSCI index, we have used the index itself to estimate comparable five-year performance, which appears quite impressive (note that we have adjusted returns for fund expenses, but not the standard deviation):
| Fund Ticker | Average Return, 5 Years Ended Oct04 (higher and positive is better) | Average Standard Deviation, 5 years Ended Oct04 (lower is better) | Gross Return/Standard Deviation (higher is better) |
Annual Expense Charge (lower is better) | Average Return Before Expenses | Average Return After Adviser Fee (DFA Only) |
| VLACX |
(0.71%)
|
17.08%
|
(.05)
|
0.20%
|
(0.91%)
|
|
Large Cap Value Equity
According to DFA, its Large Cap Value Portfolio "invests in companies that have a market capitalization in the largest 90% of the total market universe." Vanguard's offering is based on the MSCI Prime Market 750 Value Index, which covers about 86% of total market capitalization. In this category, the DFA product outperforms Vanguard, even after taking adviser fees into account (again, we have used the MSCI index to proxy the fund's five year performance) through the end of October, 2004:
| Fund Ticker | Average Return, 5 Years Ended Oct04 (higher and positive is better) | Average Standard Deviation, 5 years Ended Oct04 (lower is better) | Gross Return/Standard Deviation (higher is better) |
Annual Expense Charge (lower is better) | Average Return Before Expenses | Average Return After Adviser Fee (DFA Only) |
| DFLVX |
6.97%
|
19.14%
|
0.36
|
0.30%
|
7.27%
|
5.67%
|
| VIVAX |
4.68%
|
14.38%
|
0.34
|
0.23%
|
4.91%
|
Using the same historical annual return, standard deviation and fee assumptions, we also conducted a simulation analysis of the two funds' performance over time. We took the perspective of two different investors. The first is accumulating funds for his retirement, and needs to achieve his goal must realize a compound nominal rate of return of at least 7.5% over the next five years. We used this rate of return because it is about the required return on a portfolio composed of 60% U.S. equity and 40% U.S. government bonds. Based on initial savings of $10,000, our investor's accumulation goal after five years is $14,356. We used a five-year time horizon because it corresponds to the length of our historical data sample. This helps to minimize the impact of estimation error on our conclusions. Our accumulating investor is interested in two questions: whether a give fund has a higher expected compound return than 7.5%, and the probability that the value of the fund will be equal to or greater than $14,356 at the end of the five year period.
The second investor is already retired. She also starts with savings of $10,000, but for income must withdraw $400 each year (4% of her initial portfolio). For the sake of comparison, we assume that to meet her needs she can also invest in a portfolio of government bonds with a weighted yield of 5%. If she does this, at the end of five years her portfolio will be worth $10,329 (assuming no change in interest rates). Our retired investor is focused on two objectives: minimizing the rate at which her portfolio declines in value (or, ideally, maximizing the growth in its value), and maximizing the probability of having at least $10,329 at the end of the five year period.
Obviously, both of these situations are unrealistic, because no investor would hold just one fund in his or her portfolio. However, these examples enable us to illustrate two important concepts. The first is called either "variance drain" or "volatility drag." This is an important concept that too few investors clearly understand. The essence of it is that investors with multiyear objectives are ultimately interested in their compound rate of return, and the higher a fund's volatility (standard deviation), the smaller will be its compound return in comparison with its average annual return. Here's an example that should help make this clear. Consider an investment that over five years earns annual returns of 10%, 5%, (20%), (5%), and 25%. Over this five-year period, the arithmetic average return on this investment is 3.00%. The standard deviation of these returns is 16.81%. Because of this variability, the compound average annual return over the five year period is only 1.87%. Volatility drag accounts for the difference. In some cases, this drag can be so large that it causes a fund with a lower level of average annual return to have a higher probability of achieving a long-term goal.
The second important issue is opportunity cost. The key point here is that a fund with relatively higher expenses potentially causes you to lose twice: first in the year the fees are charged, and then over time due to the additional returns you could have earned had the "excess" fees been invested in the fund.
The next table shows the following information for the DFA and Vanguard large cap value funds: (1) the expected compound annual return over five years for our "accumulating" investor; (2) the probability that, after five years, he will have achieved his goal; (3) the expected compound portfolio return for our retired investor, and (4) the probability that after five years she will have achieved her goal. Again, we show the DFA fund's results both without and with the 1% adviser fee;
|
Fund (Ticker)
|
Compound Annual Return for Accumulating Investor (higher is better)
|
Probability of Achieving Accumulation Goal (higher is better)
|
Compound Annual Return for Retired Investor Making Withdrawals (higher is better)
|
Probability of Achieving Retired Investor's Goal (higher is better)
|
| DFLVX with adviser fee |
4.37%
|
36%
|
(4.10%)
|
48%
|
| DFLVX without adviser fee |
5.55%
|
41%
|
(2.77%)
|
53%
|
| VIVAX |
3.86%
|
29%
|
(4.53%)
|
44%
|
As you can see, all of our analyses point to the same conclusion: in the large cap value category, the DFA offering is preferred to the one from Vanguard.
Small Capitalization Equity
Before looking at the performance of the two funds in this category, we note that they target somewhat different market capitalization ranges: the DFA fund theoretically covers the bottom 8% of market capitalization, while the Vanguard fund (which is based on the MSCI Small Cap 1750 Index) leaves out the bottom 2%, and covers the next 12% (that is, the 3rd through 14th percentiles). However, to limit trading costs, both funds also employ "buffer zones" which enable a company to temporarily move outside these target capitalization ranges without being dropped from the fund. Hence, the DFA fund can have more than 10% of its capitalization in companies that are above the 8th percentile, while the Vanguard fund can contain companies slightly above and below its target range. Again, because of the change in the Vanguard fund's target index, we have used performance data for the index itself. Now let's move on to our two performance comparisons:
| Fund Ticker | Average Return, 5 Years Ended Oct04 (higher and positive is better) | Average Standard Deviation, 5 years Ended Oct04 (lower is better) | Gross Return/Standard Deviation (higher is better) |
Annual Expense Charge (lower is better) | Average Return Before Expenses | Average Return After Adviser Fee (DFA Only) |
| DFSTX |
12.12%
|
25.73%
|
0.47
|
0.42%
|
12.54%
|
11.12%
|
| NAESX |
12.79%
|
20.34%
|
0.64
|
0.27%
|
13.06%
|
|
Fund (Ticker)
|
Compound Annual Return for Accumulating Investor (higher is better)
|
Probability of Achieving Accumulation Goal (higher is better)
|
Compound Annual Return for Retired Investor Making Withdrawals (higher is better)
|
Probability of Achieving Retired Investor's Goal (higher is better)
|
| DFSTX with adviser fee |
8.57%
|
54%
|
0.39%
|
62%
|
| DFSTX without adviser fee |
9.53%
|
57%
|
1.46%
|
65%
|
| NAESX |
11.03%
|
66%
|
3.39%
|
75%
|
In the small cap category, based on our analysis we prefer the Vanguard offering.
Small Cap Value Equity
DFA offers two funds in this category. The small value fund (DFSVX) invests in stocks from within the lowest eight percentiles of market capitalization that meet DFA's value screening criteria. The small XM value fund (DFFVX) does not include microcap stocks (those in the lowest 2.5% of market capitalization), and instead invests in value stocks located in the next 10% of market capitalization (i.e., between 2.5% and 12.5%). The XM fund is therefore quite comparable to the Vanguard Small Value Index Fund, which tracks the MSCI Small Cap 1750 Value Index (again, because the fund switched target indexes, we have used the index returns in our analysis). On to our results, which in this case (because of the short data series for the XM fund), are based on the three years ended October 29, 2004:
| Fund Ticker | Average Return,3 Years Ended Oct04 (higher and positive is better) | Average Standard Deviation,3 years Ended Oct04 (lower is better) | Gross Return/Standard Deviation (higher is better) |
Annual Expense Charge (lower is better) | Average Return Before Expenses | Average Return After Adviser Fee (DFA Only) |
| DFSVX |
22.45%
|
24.27%
|
.93
|
0.56%
|
23.01%
|
21.45%
|
| DFFVX |
21.52%
|
24.77%
|
0.87
|
0.47%
|
21.99%
|
20.52%
|
| VISVX |
20.29%
|
15.63%
|
1.32
|
0.27%
|
20.56%
|
|
Fund (Ticker)
|
Compound Annual Return for Accumulating Investor (higher is better)
|
Probability of Achieving Accumulation Goal (higher is better)
|
Compound Annual Return for Retired Investor Making Withdrawals (higher is better)
|
Probability of Achieving Retired Investor's Goal (higher is better)
|
| DFSVX with adviser fee |
18.59%
|
85%
|
11.57%
|
90%
|
| DFSVX without adviser fee |
19.89%
|
86%
|
12.97%
|
91%
|
| VISVX |
19.37%
|
96%
|
12.58%
|
98%
|
| DFFVX with adviser fee |
18.22%
|
83%
|
11.15%
|
88%
|
| DFFVX without adviser fee |
19.16%
|
85%
|
12.15%
|
89%
|
While the comparison is quite close in this category, on balance we prefer the Vanguard offering. As you can see, its substantially lower volatility more than makes up for its lower returns, enabling it to realize the highest probabilities of achieving our two investors' multi-year goals.
Microcap Equity
Theoretically, the benchmark for DFA's microcap fund is the CRSP 9-10 Index, which contains its bottom two "deciles" of companies. In recent years, about two thirds of the DFA fund's portfolio has been invested in companies in the lowest 2.5% of market capitalization, and one third in larger companies (i.e., those in the next 2.5% of market capitalization, up to the lowest 5%). As Vanguard does not yet offer a fund in this category (though MSCI has just launched its own microcap index), we have compared the DFA fund to the Bridgeway Ultra Small Company Fund, which aims to track the performance of the CRSP Decile 10 Index. While this fund is closed to new investors, it provides a good comparable for DFA's performance in this segment.
| Fund Ticker | Average Return, 5 Years Ended Oct04 (higher and positive is better) | Average Standard Deviation, 5 years Ended Oct04 (lower is better) | Gross Return/Standard Deviation (higher is better) |
Annual Expense Charge (lower is better) | Average Return Before Expenses | Average Return After Adviser Fee (DFA Only) |
| DFSCX |
15.93%
|
30.29%
|
0.53
|
0.56%
|
16.49%
|
14.93%
|
| BRSIX |
24.85%
|
24.48%
|
1.02
|
0.67%
|
25.52%
|
|
By default we prefer DFA in the microcap category because its fund is still open. However, the consequence of this is some performance dilution as the fund's growing size forces it to invest more of its capital in the stocks of slightly larger companies than those targeted by the Bridgeway microcap fund.
U.S. Commercial Property
Both Vanguard and DFA offer funds that track Real Estate Investment Trust Indexes. The former tracks the Morgan Stanley equity REIT Index, while the latter invests in shares of both equity and so-called "hybrid" REITs that invest in a mix of real estate equity and mortgages. As you can see in the following table, this slightly different focus gives the DFA product an edge, at least before adviser fees are charged.
| Fund Ticker | Average Return, 5 Years Ended Oct04 (higher and positive is better) | Average Standard Deviation, 5 years Ended Oct04 (lower is better) | Gross Return/Standard Deviation (higher is better) |
Annual Expense Charge (lower is better) | Average Return Before Expenses | Average Return After Adviser Fee (DFA Only) |
| DFREX |
20.32%
|
15.97%
|
1.27
|
0.41%
|
20.73%
|
19.32%
|
| VGSIX |
19.43%
|
16.18%
|
1.20
|
0.24%
|
19.67%
|
We call this category a toss-up, with no clear winner. On a before-adviser fees basis, the DFA product has the edge. It loses it, however, once adviser fees are taken into account.
Large Cap International Equity
In this category, one can see the impact of DFA's slightly more "active" approach to indexing. The Vanguard fund tracks the MSCI Europe, Asia and Far East (EAFE) Index.
| Fund Ticker | Average Return, 3 Years Ended Oct04 (higher and positive is better) | Average Standard Deviation, 3 years Ended Oct04 (lower is better) | Gross Return/Standard Deviation (higher is better) |
Annual Expense Charge (lower is better) | Average Return Before Expenses | Average Return After Adviser Fee (DFA Only) |
| DFALX |
9.00%
|
16.02%
|
0.56
|
0.43%
|
9.43%
|
8.00%
|
| VDMIX |
9.30%
|
16.65%
|
0.56
|
0.34%
|
9.64%
|
On balance, however, we prefer the Vanguard offering in this category.
Large Cap International Value Equity
Vanguard does not have an index product in this category. It does, however, offer an actively managed one. The comparison with the DFA product is not a pretty one.
| Fund Ticker | Average Return, 5 Years Ended Oct04 (higher and positive is better) | Average Standard Deviation, 5 years Ended Oct04 (lower is better) | Gross Return/Standard Deviation (higher is better) |
Annual Expense Charge (lower is better) | Average Return Before Expenses | Average Return After Adviser Fee (DFA Only) |
| DFIVX |
6.76%
|
17.38%
|
0.39
|
0.52%
|
7.28%
|
5.76%
|
| VTRIX |
2.25%
|
17.62%
|
0.13
|
0.62%
|
2.87%
|
No question about it. In international large cap value, we prefer DFA, hands down.
International Small Cap and Small Cap Value
Once again, Vanguard does not offer index products in these categories. It does, however, offer an actively managed fund that invests in international small cap companies. Once again, the clear winner is DFA.
| Fund Ticker | Average Return, 5 Years Ended Oct04 (higher and positive is better) | Average Standard Deviation, 5 years Ended Oct04 (lower is better) | Gross Return/Standard Deviation (higher is better) |
Annual Expense Charge (lower is better) | Average Return Before Expenses | Average Return After Adviser Fee (DFA Only) |
|
DFISX - Int'l Small Cap |
10.21%
|
16.07%
|
0.64
|
0.71%
|
10.92%
|
9.21%
|
| DISVX -- Int'l Small Cap Value |
13.58%
|
16.50%
|
0.82
|
0.81%
|
14.39%
|
12.58%
|
| VINEX |
7.69%
|
22.38%
|
0.34
|
0.73%
|
8.42%
|
On balance, we prefer DFA's international small cap value fund in this category.
Emerging Markets Equity
It is important to note that, unlike the EEM Exchange Traded Fund, neither the DFA nor the Vanguard emerging markets equity product exactly tracks the MSCI Emerging Markets Index. DFA's emerging markets universe includes 16 countries. In comparison, Vanguard's includes 18. The key difference between them -- and it could be an important one -- is that DFA includes Malaysia but excludes China, India and Peru, while Vanguard includes these three but not Malaysia.
We are also including emerging markets small cap and emerging markets value funds in this category, which DFA offers but which Vanguard does not. Here is how they compare:
| Fund Ticker | Average Return, 5 Years Ended Oct04 (higher and positive is better) | Average Standard Deviation, 5 years Ended Oct04 (lower is better) | Gross Return/Standard Deviation (higher is better) |
Annual Expense Charge (lower is better) | Average Return Before Expenses | Average Return After Adviser Fee (DFA Only) |
| DFEMX -- Emerging Mkts |
6.22%
|
23.21%
|
0.27
|
0.78%
|
7.00%
|
5.22%
|
| VEIEX |
6.64%
|
24.54%
|
0.27
|
0.53%
|
7.17%
|
|
| DFEVX -- Emerging Mkts Value |
9.77%
|
24.24%
|
0.40
|
0.86%
|
10.63%
|
8.77%
|
| DEMSX -- Emerging Mkts Small |
9.22%
|
22.26%
|
0.41
|
1.12%
|
10.34%
|
8.22%
|
If we were looking only at broad Emerging Markets funds, we would prefer the Vanguard offering. However, if we include Emerging Markets value and small cap funds, we prefer DFA's emerging markets value (DFEVX) offering.
Fixed Income Offerings
We use up to three different fixed income asset classes in our model target return portfolios: real return bonds, unhedged foreign currency bonds, and domestic investment grade bonds. Vanguard offers a real return bond fund (VIPSX), while DFA does not. Unfortunately, neither company offers an unhedged foreign currency bond fund. As a result, we recommend either the T. Rowe Price International Bond Fund (RPIBX), or the PIMCO Unhedged Foreign Bond Fund (PFBDX). That leaves us with a comparison between their domestic fixed income offerings.
Before moving on to fund comparisons, however, it is important to clearly understand DFA's approach to fixed income management. First, DFA is quite dubious about the long-term likelihood of earning higher risk adjusted returns by taking on more duration risk (that is, by investing in bonds with longer average maturities) beyond an intermediate point on the yield curve. They also believe that the bond markets are quite efficient, and that forecasting future interest rates (and hence consistently successful active bond management) is basically impossible. That being said, they also believe that , even in the absence of forecasting, "investors may be able to increase their risk adjusted returns [by employing]…a variable maturity strategy that shifts the maturities of the portfolio as the yield curve changes…[The approach] does not anticipate changes in the yield curve, rather it seeks to maximize risk-adjusted returns present in the [current] curve…In broad terms, this means shortening maturities in inverted curves [i.e., when short-term rates are higher than long-term rates], and extending them in [normal] upwardly sloped curves."
DFA also believes that while introducing foreign currency bonds into a fixed income portfolio can improve its risk/return trade-off, all foreign currency bond positions should be fully hedged against changes in exchange rates. They argue that "in our view, global bonds do not represent a separate and distinct asset class from domestic fixed income." Obviously, we disagree with this logic. However, for the sake of our comparison, we will accept it, and include currency-hedged foreign bond funds in our comparison.
We therefore evaluate the following DFA funds: Two-Year Global Fixed Income (DFGFX), Five-Year Global Fixed Income (DFGBX), Five-Year U.S. Government (DEFGX) which invests in securities that mature within five years, and Intermediate Term Government (DFIGX). This latter fund invests in dollar-denominated debt issued by both the U.S. Government, and in AAA rated dollar denominated debt issued by foreign governments and supranational agencies (e.g., the World Bank). This fund's average weighted maturity is between seven to ten years.
We compare these funds to two from Vanguard: VBMFX tracks the Lehman Brothers Aggregate U.S. Bond Market Index, which contains three types of investment grade bonds, of short, intermediate and long maturity: those issued by the U.S. government, those issued by corporations, and those backed by mortgages and other assets. We also include VBIIX, which tracks on index composed the same types of bonds, but with only intermediate maturities.
| Fund Ticker | Average Return, 5 Years Ended Oct04 (higher and positive is better) | Average Standard Deviation, 5 years Ended Oct04 (lower is better) | Gross Return/Standard Deviation (higher is better) |
Annual Expense Charge (lower is better) | Average Return Before Expenses | Average Return After Adviser Fee (DFA Only) |
| DFGFX |
4.28%
|
1.42%
|
3.01
|
0.25%
|
4.53%
|
3.28%
|
| DFGBX |
5.94%
|
3.77%
|
1.58
|
0.34%
|
6.28%
|
4.94%
|
| DFFGX |
6.34%
|
4.32%
|
1.47
|
0.27%
|
6.61%
|
5.34%
|
| DFIGX |
8.51%
|
6.74%
|
1.26
|
0.17%
|
8.68%
|
7.51%
|
| VBMFX |
7.10%
|
4.11%
|
1.73
|
0.22%
|
7.32%
|
|
| VBIIX |
8.55%
|
6.02%
|
1.42
|
0.20%
|
8.75%
|
|
Though it is a close call (at least before adviser fees), if our criterion is maximizing return, on balance we prefer the Vanguard offering (VBIIX) to DFIGX from DFA. On the other hand, if we were looking strictly at return per unit of risk, then we would prefer DFGFX.
Conclusion
As is so often the case when all-stars are compared, there is no clear winner when it comes to DFA versus Vanguard. Based on the performance data we have used in our analysis, we prefer the DFA offerings in Domestic Large Cap Value, Commercial Property, Large Cap International Value, International Small Cap, International Small Value, Emerging Markets Value and Emerging Markets Small Cap. We also prefer DFA for Microcap equity, where Vanguard lacks a comparable offering.
On the other hand, we prefer Vanguard's products for Domestic Large Cap Equity, Domestic Small Cap and Small Cap Value, Large International, Broad Emerging Markets Equity and Fixed Income. We also prefer Vanguard for real return bonds, where DFA lacks a comparable offering. We should also add two qualifications to these conclusions. First, due to short data series and changing underlying indexes, we have not done analyses to see if the differences in returns and volatility between funds that we observe are statistically significant. In some cases, they probably are, but in others they probably are not. More importantly, we should always keep in mind that most investors would be better off with indexed products from either DFA or Vanguard than they would be with actively managed funds. Finally, we wish that either Vanguard or DFA (or both of them) would introduce products in the commodities and unhedged foreign currency bonds asset classes.