Selecting the right funds in a portfolio strategy is critical for both self-directed investors and administrators of retirement plans such as 401(k) and IRAs. Working with too few funds limits diversification benefits and limits opportunities. On the other hand, too many funds can be confusing and misleading. Getting the right mix is especially important in a retirement plan as participants have a wide range of differing requirements and demands that will drive their financial planning. For ETF investors, the ever increasing number of ETFs (the latest tally is $782 billion asset in 810 funds) only exacerbates this problem.
We have developed guidelines for providing the optimal investment choices for an ETF based plan or a portfolio. The following are the key criteria to consider when making such choices.
Candidate funds should cover major asset classes at least. We consider the following the major classes.
- US Equity
- Developed Country Equity (ex. US)
- Emerging Market Equity
- Real Estate Investment Trusts (REITs, most US)
- Fixed Income
For global investors, we add Foreign Fixed Income as another major asset. Minor assets and styles are also possible candidates if strategies employed handle them effectively. We do not advocate going down to sector or industry levels since their volatile and concentrated nature defeats the diversification purpose
Since most ETFs are index funds, the following should be used to select an ETF.
- Liquidity: The minimum requirement should be at least 250K daily average volume in last 3 months – exclude ETFs with low trading volume. The other major factor to consider is the total net asset an ETF is held. In general, we would prefer at least $1 billion asset for a major ETF.
- Tracking error: how closely it tracks its stated index benchmark. Price/Nav discount/premium and bid-ask spreads are two important factors to consider here. In general, for two ETFs that have the same stated index benchmark, the one has tracks the benchmark more closely should be picked.
- Fee: all things equal, low expense fee is a key advantage for a long term portfolio. Vanguard ETFs and index funds have set high standard/bars for other ETFs to measure up to.
- Return: again, all things equal, return is another factor to use to distinguish ETFs.
In addition, it is critical to see how these selected funds work with each other. For example, consider the following two choices: US Equity, Emerging Market Equity and Fixed Income vs. US Equity, Developed Country Equity and Fixed Income. How would these asset classes work together? How would the actual funds behave when putting them together? MyPlanIQ uses a proprietary method to perform historical simulation to evaluate portfolio effectiveness.
Let’s try to select candidate ETFs for a simple core portfolio that consists of 6 key major assets mentioned above. We will build two funds – the one we think is the best and compare it with the most popular in each asset class.
US Equity: there are plenty choices here. The first comes to mind is the Wilshire 5000 Total Return index. Claymore launched an ETF (WFVK) to track this index in March 2010. However, since this is a fairly new ETF, its trading volume is light and thus is excluded. The other venerable index would be Vanguard Total Stock Market Index ETF (VTI). Its average daily trading volume in the past 3 months is more than 2 million shares. Another choice is the oldest and most popular S&P 500 ETF: SPDR SP 500 (SPY). This is the most liquid ETF and it has the least bid-ask spread. It has more than 200 million shares change hands daily. SPY, however, is a large cap index, not a broad base index. We prefer VTI over SPY.
Developed Market Equity: MSCI EAFE index (EFA) and Vanguard FTSE All-World ex-US ETF (VEU) are the two candidates. Vanguard Total World Stock (VT) is a one for all equity ETF that covers all of three equity asset classes (U.S. Developed and Emerging Markets). We prefer VEU over EFA, considering the low expense Vanguard charges for VEU.
Emerging Market Equity: iShares MSCI Emerging Market Index (EEM) and Vanguard Emerging Market Equity (VWO) are the two candidates. Matt Hougan from IndexUniverse and others have written several articles discussing both ETFs. Again, Vanguard VWO is clearly a winner after taking cost into account.
Real Estate Investment Truse (REITs): Dow Jones U.S. Real Estate Investment Trusts (IYR) is the most popular while iShares Cohen & Steer Realty Majors (ICF) is more concentrated on major REITs. Vanguard REIT Index ETF (VNQ) has the lowest expense, wide diversification and high trading volume. VNQ is clearly the choice here.
Commodities: Powershares DB Commodity Index ETF (DBC) and iShares S&P GSCI Commodity Index (GSG) are the two main choices. It is interesting to see that DBC has higher expense (1.3%) than GSG (0.75%) while GSG is more concentrated on energy commodities (74% vs. 55%). In general, DBC is more balanced though GSG has wider exposure among many commodities. This article has some more detailed writeup on the comparison between the two ETFs. DBC, though not as diversified as GSG, is more balanced in major commodities. We prefer DBC.
Fixed Income: For a broad base fixed income exposure, iShares Barclays (Lehman) Aggregate Bond (AGG) and Vanguard Total Bond Market (BND) are the clear two candidates. With its lower expense cost (0.11%), BND is better than AGG (0.2% expense). This article compares the two ETFs in more detail.
We have created two plans for the core portfolios. The candidate funds for Six Core Asset ETFs Most Popular are chosen based on their popularity (history and liquidity). The candidate funds for the other plan Six Core Asset ETFs are chosen based on more detailed criteria outlined above. The following table shows the selections of the two plans:
Developed Market Equity
Emerging Market Equity
The following table shows the performance of model portfolios: one Strategic Asset Allocation (SAA) Moderate and one Tactical Asset Allocation (TAA) Moderate portfolio are chosen for each plan.
1 Year Annualized Return
1 Year Sharpe Ratio
3 Year Annualized Return
3 Year Sharpe Ratio
5 Year Annualized Return
5 Year Sharpe Ratio
From the above table, one could clearly see that the selection of candidate funds for constructing a portfolio plays an important role in portfolio returns. The 2-3% difference of annualized returns in the past 5 years between the two corresponding portfolios is striking. It clearly shows that it pays to put more efforts in selecting funds.
The impressive performance of Six Core Asset ETFs supports the maxim ‘simpler is better’.
Disclosure: Long IYR, SPY, LQD