Portfolios Using DFA Mutual Funds

A number of our U.S. subscribers have asked us how we would implement our model target return portfolio solutions using just DFA mutual funds. This section responds to that request.

The following table shows the DFA Funds that correspond to the asset classes we use in our model portfolio solutions. Please note that we have tried to select the broadest based funds we could find. Where no index funds are offered, we have used actively managed funds. As we have discussed in our writing, in order to take "tilts" within these asset classes, investors may use other index or actively managed funds.

As you can see, there are a number of asset classes where DFA doesn't offer a product (yet).

The first of these is real return bonds. In this case, we have substituted the Two-Year Global Fixed Income Fund (DFGFX).

The second asset class where DFA lacks a product is foreign currency bonds. Currently, DFA does not offer a fund that invests in unhedged non-dollar denominated investment grade bonds issues by foreign developed country governments. As we have discussed in our writing, our view of foreign currency bonds is that they have a valuable role to play in many portfolios, and that, for long term investors, the balance of research concludes that they should not have their exchange rate risk hedged. For investors with shorter-term horizons, however, this isn't the case, and there is a stronger argument to be made in favor of currency hedging. This latter position is the one DFA adopts. Its Global 5 year bond fund (DFGBX) fully (as near as we can tell) hedges its currency risk back into dollars. The result is, in effect, a broad U.S. dollar fixed income asset class, which has more attractive risk/return properties than many domestic-only bond indexes. As such, we regard DFA's Global 5 and Global 2 year bond funds as more substitutes (and very good ones at that) for other domestic bond funds rather than foreign currency bond funds as we use them in our asset allocation models. We therefore use the T. Rowe Price International Bond Fund (RPIBX), which invests in non-dollar bonds on an unhedged basis. Alternatively, one could use PIMCO's unhedged foreign bond fund (PFBDX) to track this asset class.

The third asset class where DFA lacks a product used in our model portfolios is foreign commercial property. In this case, we use either the Fidelity International Real Estate Fund (FIREX) or the Cohen and Steers International Realty Fund (IRFAX).

The fourth asset class missing a DFA product is commodities. As we understand it, there is still a fair amount of internal debate about this asset class within DFA. Apparently some see the current commodity indexes as little more than a bet on oil prices, while others believe they have broader diversification benefits. Given this, we use the PIMCO Commodities Real Return Fund (PCRDX) in listing of DFA funds. An alternative to this is the Oppenheimer Real Asset Fund (QRAAX); this fund places relatively more weight on energy commodities than the PIMCO fund (for more on this, see the section on commodities on our website).

Another asset class which lacks a DFA fund is timber. To track the performance of our model portfolios, we use a market cap weighted mix of timber real estate investment trusts: 70% Plum Creek Timber (PCL) and 30% Rayonier (RYN).

Equity market neutral (our proxy for uncorrelated alpha strategies) is another area in which DFA thus far lacks a product offering. In its place, we are using an equally weighted mix of five mutual funds that use “hedge fund-like” strategies. These include Hussman Strategic Growth (HSGFX), Analytic Global Long/Short (ANGLX), James Market Neutral (JAMNX), Rydex Absolute Return Strategies (RYMXQ), and JP Morgan Market Neutral (JMNAX).

The last asset class that is included in some of our model portfolios is U.S. equity market volatility, as tracked by the VIX index. Thus far, the only way to invest in this asset class is via futures contracts that track the VIX. Unfortunately, these are too complicated for most investors. We hope that at some point over the next two years, a futures-based index fund (similar to commodity index funds) becomes available in this asset class.

Finally, it is important to point out that many DFA funds (including some of the ones we use to implement our model portfolios) take what we term "tilts" within different equity asset classes, particularly toward value and small cap companies. Whether these tilts make sense comes down to your view on the relative efficiency of markets.

The people at DFA believe that markets are reasonably efficient (in the practical sense that the opportunities they offer for improving one's risk/return trade-off are quickly arbitraged away). Tilts therefore represent differential exposure to market risk factors, compared to the broad market index. Taking them should result in either more return with more risk (e.g., as is often claimed for small cap or value tilts), or less return with less risk, relative to simply investing in a broad asset class index. On the other hand, some people believe that markets are relatively inefficient (due to some combination of cognitively limited investors, differential access to information, and/or barriers to arbitrage). In their view, tilts can provide more return with less risk than the broad index (or, alternatively, less return with more risk, as the former also implies someone has blundered into the latter). In short, when taking tilts, we believe that one should be very clear about what one hopes to achieve, vis-a-vis investing in the broad market index. Like DFA, we generally come down on the "markets are practically efficient" side of the argument, while acknowledging that there are occasional exceptions to this rule (e.g., bubbles and crashes can and do happen, though not very frequently).

*Note that these are funds available to average investors. Those with more to invest may qualify for other share classes.

Update: January, 2006