[revised on September 3, 2013 and February 3, 2014. The latest revision is an updated ETFportfolioDESIGNER2 that includes Vanguard’s Total World Stock ETF (VT). It was necessary to reset the DESIGNER’s time period to 2009 through 2013. Appropriate changes were made in the following discussion.]
I ‘like’ a portfolio of exchange traded funds (ETFs) that are listed in the New York Stock Exchange by trading symbols AGG, GLD, VNQ, and VWO [VT is an alternative to VWO]. Equal portions of capital are allocated to each fund. The best way to manage the portfolio is by rebalancing the ETFs. Ordinary investors can be cautiously optimistic about this high risk portfolio after reviewing the following information:
Among previously tested portfolios, a 4-sector model held hypothetical investments in market indices for emerging markets stocks, U.S. bonds, U.S. REITs, and global precious metals. The model was not an investable portfolio.
The 4-sector model was converted to an investable portfolio by substituting index ETFs for the market indices and assigning equal weightings (ref 1) to the ETFs. The big advantage of choosing index ETFs is their ease of trading in the stock market (ref 2). The objective of the investable portfolio, hereafter called the SmallTradesETFportfolio, is to outperform the U.S. stock market in the long term.
The conversion began by screening funds according to geography and asset class (ref 3). Selected ETFs were compared to their financial markets, appraised for risk, and tested for optimal portfolio mechanics.
Comparisons. Chart 1 shows the time course of total returns from ETFs and financial markets. The total returns are charted as monthly percentages of change in market value based on price changes and cash distributions (see endnote) .
Chart 1. Time course of total returns: All panels show the time course of monthly total returns from a market sector (black dashed line), older ETF (solid blue line), and newer ETF (solid orange line). ETFs are identified by 3-letter trading symbols. Total returns were set to 0% at inception of the newer ETF.
In chart 1, any spread between an ETF and its market sector illustrates the difference in investment performance. Smaller spreads, and therefore closer matches to sector performance, occurred among ETFs aligned to the precious metals and emerging markets stock sectors. Wider spreads appeared when AGG outperformed the U.S. Bond sector by 3-15 percentage points during the 2008 Credit crisis; also when VNQ and REZ consistently underperformed the Equity REIT sector by an average 8 percentage points.
Correlations. Chart 2 shows the correlation of returns between an ETF and its market index.
Chart 2, Correlation of total returns: ETFs are identified by 3-letter trading symbols. All panels compare the monthly returns of ETFs to their corresponding market sector. Monthly returns were calculated as a percentage change in market value that includes the accumulation of cash distributions from underlying holdings. The identity line shows hypothetically equal returns. “r” values are correlation coefficients for the relationship between ETF and sector returns. “r” values were not calculated for newest funds, CORP and GLTR, due to the scarcity of monthly returns.
In chart 2, the correlation coefficient (“r”) signifies the degree of alignment between concurrent monthly returns. If r were 1.0, the data would lie on an identity line where fund returns match the market returns. All r values in chart 2 were exceptionally high, which supports the visual impression from chart 1 that ETFs traced the time course of their market sectors. All of chart 2’s data except those for AGG were closely aligned to an identity line. About 12 of AGG’s monthly returns outperformed the U.S. Bond market during the 2008 Credit crisis and appear as outliers to their line of identity. The few outliers exerted no practical effect on AGG’s correlation coefficient.
ETF appraisals. The SmallTradesETFportfolio contains healthy, wealthy ETFs with proven ease of trading. Every ETF is eligible for holding in a tax-deferred brokerage account and otherwise offers a low tax burden when held in a taxable account (exception: returns from the precious metal held by GLD are taxed at the higher rate for ordinary income rather than lower rate for long-term capital gains). Here are the investment strategies and risks of the funds, with a link to their scorecards (ref 4):
AGG (iShares Core Total U.S. Bond Market ETF
) invests 95% of its capital in a basket of U.S. investment grade bonds that reflect the Barclays Capital U.S. Aggregate Bond Index
. AGG receives fixed income and capital gains from the bonds. The fixed income is derived from payments of interest and returns of principal. Investment grade bonds are less likely to default on payments of fixed income than non-investment grade (‘junk’) bonds. Investment grade bonds generally pay lower interest than junk bonds and are low-risk investments (ref 5).
VWO (Vanguard FTSE Emerging Markets ETF
) invests 95% of its capital in a representative sample of stocks listed in the FTSE Emerging Markets Index
. Emerging markets stocks are more volatile and less liquid than U.S. stocks. Stocks are generally high-risk investments that reward investors with payments of dividends and capital gains. VWO invests in emerging market stocks which have characteristically higher risk and higher returns than developed market stocks (ref 5).
- VT (Vanguard Total World Stock ETF) invests in stocks issued in the emerging and developed markets. VT is an alternative to VWO.
VNQ (Vanguard REIT ETF
) invests in real estate properties by purchasing shares of real estate investment trusts (REITs). Equity REITs are companies who invest pooled money into the ownership of real estate and distribute at least 90% of their taxable income to shareholders. Equity REITs are considered low-risk, high-return investments (ref 5). Because VNQ concentrates on real estate, its primary risk is a decline in the real estate market.
GLD (SPDR Gold Trust
) is a physical commodity fund that invests all capital in gold bullion. Gold only provides income when sold in the market for a profit. Investors typically buy gold bullion as an insurance policy against the devaluation of currency (ref 5). GLD shareholders risk losses from declining prices and damage or theft of the bullion.
Market forces continually change the value of a portfolio either to the benefit or detriment of the investor. The investor’s choices are to make no adjustments (“buy-and-hold”), sell rising assets to buy declining assets (“rebalance”), sell declining assets to buy rising assets (“reallocation”), or revise the investment strategy (ref 6).
The choices to buy-and-hold or rebalance the SmallTradesETFportfolio were examined by using a computer-assisted program to test a set of 5-year historical returns from the ETFs [the computer-assisted program is outdated and therefore replaced by ETFportfolioDESIGNER2. The discussions of Tables 1 and 2 are outdated, but they remain instructive]. The buy-and-hold strategy was to purchase each ETF with 25% of the total invested money and automatically reinvest the cash distributions. The rebalance strategy was divided into 2 plans for correcting the buy-and-hold portfolio. The scheduled plan made regular corrections during the life of the portfolio. The signaled plan made irregular corrections depending on when the holdings drifted from the allocation plan by an assigned error signal (ref 7).
The data in table 1 show that all choices outperformed the benchmark investment in U.S. stocks. Both rebalancing plans enhanced the portfolio CAGR of the buy-and-hold strategy, namely by 1.3 percentage points when rebalanced yearly and by 2 percentage points when rebalancing a 32% allocation error.
COLUMN HEADINGS: The investment strategies for the ETF Portfolio are “Buy-and-hold”, “Rebalance #1”, and “Rebalance #2”. The “Benchmark” portfolio is an index of the U.S. Stock Market.
ROW HEADINGS: “Rebalancing plan” shows the best schedule for “Rebalance #1” and best error signal for “Rebalance #2”. “Rebalance episodes” are the total number of rebalances. “Final market value” is computed from the weighted market returns. “Portfolio CAGR” is the compound annual growth rate; higher CAGRs reflect higher returns. “Sharpe ratio” is an adjusted annual rate of return; higher ratios reflect higher returns.
ASSUMPTIONS: Every portfolio is funded with $10,000 and 25% of the $10,000 is allocated to each holding. There are no fees for financial services, all cash distributions are automatically reinvested, and the portfolio holdings are never withdrawn.
Efficient investment. The total returns in table 1 were earned under the best of circumstances because there was no payment of fees for financial services and plenty of money was used for making the initial investment. Without fees, all levels of initial investment are equally efficient in yielding a return. But additional fees create a penalty margin that reduces returns by as much as 25 percentage points with $1,000 investments and 2 percentage points with $10,000 investments (ref 8).
So, how might trading fees and initial investment affect the SmallTradesETFportfolio? Table 2 demonstrates the efficiency of earning returns based on the initial investment and a $10 trading fee.
The data are total returns as measured by CAGR; higher CAGRs reflect higher returns. All trading fees are $10/trade.
ROW HEADINGS: “Initial investment” is the total amount of money spent on creating the portfolio, including trading fees. “Buy-and-hold” means that no trading occurred during the 5-year holding period, 2008-2012. “Rebalance #1” is a plan to rebalance the holdings on an annual basis. “Rebalance #2” is a plan to rebalance the portfolio when a holding drifts from the allocation plan by a 32% difference.
In table 2, the maximal returns of the buy-and-hold and rebalanced portfolios were achieved when the initial investment reached $15,000. There was no advantage to rebalancing the portfolio with only $2,000 of invested capital.
Financial services fees. ETF managers charge an annual expense ratio to specialists in the primary market, not to ordinary investors in the secondary market. The expense ratio reduces the net asset value of ETF shares in the primary market and the net asset value determines ETF share prices in the stock market (ref 2). Ordinary investors may pay an advisor’s fee when they are clients of a financial institution (do-it-yourself investors take pride in avoiding this fee). The typical advisory fee, about 1% of the annual portfolio value, reduces the total return of the portfolio by an amount that can be estimated in the computer-assisted program.
The SmallTradesETF Portfolio is designed for risk-tolerant investors who seek to outperform the U.S. Stock Market over a time period of many years. Each ETF tracks a unique sector of the financial markets with proven transparency, durability, and liquidity of fund operations. In terms of risk management, rebalancing the holdings helps protect from losses incurred during market declines. An initial investment of $4,000 is needed to gain the advantage of rebalancing the portfolio. Better results are obtained by investing at least $10,000. I prefer using a
30% 28% rebalancing signal (rather than a schedule) to rebalance the SmallTradesETFportfolio. The signaled rebalancing method is easy to obtain and use by clicking on this link, Rebalancing an Investment Portfolio.
Similar portfolios are advocated in newspapers (ref 9) and books (refs 10-12). The unique features of the SmallTradesETF portfolio are its simple allocation plan and tested rebalancing strategy. A major disadvantage is the uncertainty about future market returns. Fifteen years of historical data for a model portfolio and 5 years of historical data for an ETF portfolio are unreliably predictive of future returns (ref 10). Consider that ¼th of the Portfolio is invested in emerging markets stocks and that today’s emerging markets are in decline after attaining a historical peak (refs 13,14). How long will these markets decline? The optimist could argue that over 75% of the world’s population lives in the emerging economies where there’s tremendous capacity for growth. Today’s emerging economies have nearly half of the world’s GDP and their share of the global GDP seems to be growing despite blips in the trajectory (ref 14).
Endnote: Cash distributions are made by fund managers to fund shareholders. The typical sources of cash distributions are dividends and capital gains earned from the fund’s underlying assets.
Copyright © 2014 Douglas R. Knight
1. Beginners’ Guide to Asset Allocation, Diversification, and Rebalancing, U.S. Securities and Exchange Commission, SEC.gov/investor/pubs/assetallocation, Modified: 08/28/2009.
2. ETF structure, Small Trades Journal, a blog at WordPress.com.
3. XTF, ETF experts. XTF.com/Research/.
4. #ETF-scorecard, Small Trades Journal, a blog at WordPress.com.
5. Asset classes, Small Trades Journal, a blog at WordPress.com.
6. Jason Van Bergen, 6 Asset Allocation Strategies That Work. ©2013, Investopedia US, A Division of ValueClick, Inc., October 16, 2009.
7. #SmallTradesPortfolioREBALANCER, Small Trades Journal, a blog at WordPress.com.
8. Beware of trading fees, Small Trades Journal, a blog at WordPress.com.
9. Anna Prior. A Portfolio That’s as 2-sector as One, Two, Three. The Wall Street Journal, July 7, 2013.
10. William Bernstein. The Four Pillars of Investing: Lessons for Building a Winning Portfolio, McGraw-Hill, 2002.
11. Mebane T. Faber and Eric W. Richardson. Top of the Class: A review of The Ivy Portfolio. 4/6/2009, Advisor Perspectives, DShort.com.
12. John C. Bogle, The Little Book of Common Sense Investing. John Wiley & Sons, Inc. Hoboken, 2007.
13. Emerging economies: When giants slow down. Jul 27th 2013. The Economist.
14. Emerging vs developed economies: Power shift. Aug 4th 2011, 17:34 by The Economist online.