Asset allocation: the classic risk:return tradeoff
Three simple, but crucial, questions
Given the macroeconomic backdrop, we sought to address three simple yet fundamentally important questions for strategic asset allocation.
1. What is a reasonable range of expected returns (both nominal and real) for a balanced portfolio of stocks and bonds based on present market conditions?
2. How does this range of returns compare with both long-term historical averages and the more recent past in the US, as well as with the last decade in Japan?
3. Should this range of expected returns alter an investor’s approach to asset allocation in light of the prevailing concerns over future US growth, infl ation, and interest rates?
Simulating future portfolio returns
To examine potential implications for asset allocation, we used the Vanguard Capital Markets Model (VCMM) to generate 10,000 simulations of potential 10-year stock and bond return paths based on market conditions as of September 30, 2011, and various scenarios for future interest rates, inflation, and other risk factors.
Figure 1 presents the results. It shows the simulated return distributions for three hypothetical portfolios ranging from more conservative to more aggressive:
1. 20 percent equities/80 percent bonds;
2. 50 percent equities/50 percent bonds; and
3. 80 percent equities/20 percent bonds.
The chart on the left shows the distributions for nominal returns; the chart on the right displays the distributions of returns adjusted for the rate of expected consumer price index inflation. For reference, both charts also show how the hypothetical portfolios would have performed in 1926–2010 and 2000–2010.
Three implications for strategic allocation
Figure 1’s two charts have at least three key implications for strategic asset allocation. The fi rst is that balanced portfolio returns over the next decade are likely to be moderately below long-term historical averages (indicated by the red dots). Put another way, the meanvariance frontier of expected returns may now be somewhat lower for all but the most aggressive portfolios than has been realised, on average, over the past 85 years.
As an example, our VCMM simulations indicate that the average annualised returns on a 50 percent equity/50 percent bond portfolio are expected to centre in the 4.5 to 6.5 percent range in nominal terms, and in the 3.5 to 4.5 percent range in real terms over the next decade. This central tendency is below the actual average returns for such a portfolio from 1926 through 2010: 8.2 percent in nominal terms and 5.1 percent in real terms.
Viewed from another angle, the likelihood that our 50 percent/ 50 percent portfolio would achieve the 1926–2010 average nominal return is estimated at somewhat less than 25 percent; in terms of real return, the odds are higher, but still below 50 percent.
A second implication of the figure is that balanced portfolio returns over the next decade are likely to be moderately higher than those for 2000–2010 in both the US (4.5 percent, yellow dots) and Japan (1.6 percent), at least for portfolios with higher equity allocations. A primary reason is the outlook for the equity risk premium.
Specifically, our simulations suggest that the average return on a broad stock portfolio is likely to be higher than that for a broad bond portfolio given current equity valuations and as compensation for investors bearing greater equity-market risk. This may surprise some readers, considering the anticipated headwinds to long-term economic growth.
However, as discussed in previous Vanguard research papers (including Vanguard’s Economic and Capital Markets Outlook), it is market valuations—not consensus growth forecasts—that generally correlate with future stock returns. Recent stock valuation metrics, such as those for the S&P 500 Index, do not appear at extreme historical levels, whereas bond yields signal the possibility of below-average nominal bond returns in the near future. In other words, the expected equity risk premium is high.
The third, and most important, implication of the figure is that the simulated ranges of expected returns are upward-sloping. Simply put, higher risk accompanies higher (expected) return; more aggressive allocations have a higher—and wider—range of expected returns, with greater downside risk in the event that the equity risk premium is not realised over the next decade. Indeed, these expected risk:return tradeoffs among stocks and bonds show why the principles of portfolio construction remain, in our view, unchanged.
In fact, this upward-sloping, wider-tail pattern in Figure 1 reaffirms the beneficial role that bonds should be expected to play in a broadly diversified portfolio, despite their presently low yields and regardless of the future direction of interest rates. Although our scenarios generate below-average nominal returns for a broad taxable bond index over the next decade—a central tendency of 2 to 3 percent annually on average—bonds should be expected to moderate the volatility in equity portfolios in the years ahead. In addition to offering this diversification benefit, bonds would be expected to provide a higher nominal income stream in the event that interest rates rise.
Timeless asset allocation endures
Our VCMM simulations indicate that balanced portfolio returns over the next decade are likely to be moderately below long-term historical averages, especially for more conservative portfolios. Yet, while the mean-variance frontier of expected returns for such portfolios may now be modestly lower and steeper than the average over the past 75 years, we have shown that the basic principles of portfolio construction remain unchanged, given the expected risk:return tradeoff among stocks and bonds.
In our view, all investors should appreciate that market conditions today are less favourable for the future than they were in, say, 1980, when high bond yields and depressed price:earnings ratios provided tailwinds towards higher-than-average stock and bond returns during the 1980s and 1990s.
But the future need not be dark, either. Indeed, the present levels of interest rates and stock market valuations are, arguably, closer to the levels of the 1950s and 1960s, environments that over time produced respectable balanced portfolio returns.
Overall, we believe that realistically recalibrating one’s return expectations for a balanced portfolio is more prudent than making a drastic shift in allocation in an attempt either to defend against elevated market volatility or to pursue higher returns under the allure of higher yields, higher economic growth, or alternative investments. Investors who are unwilling, or unable, to lower their targeted rates of return or spending requirements may need to increase their savings rates—an approach that Vanguard research has shown can be quite effective in raising the odds of investment success.
An alternative tactic, for investors who feel locked to their return or spending targets, would be to adopt a somewhat more aggressive strategic asset allocation approach by increasing their equities holdings. Of course, a direct result of this approach would be for the investor to bear higher portfolio volatility and greater downside risk.
Joseph Davis is chief economist at Vanguard, Roger Aliaga-Díaz is a senior economist at Vanguard and Andrew J. Patterson is an investment analyst at Vanguard.
For further information contact: email@example.com
Total returns, income, and the search for yield
Vanguard believes it is important for investors to view their portfolios from a total return perspective, rather than simply in terms of potential income (in other words, the yield to maturity on a bond fund, or the dividend yield on an equity fund). An investor looking at total return will be concerned with capital gains and losses as well as with the income derived from bond interest and stock dividends. Past Vanguard research has highlighted the importance of understanding total return and the risks that can accompany a narrow focus on income.
Investors who increase their allocation to higher-yielding bonds or dividend-paying stocks in an attempt to meet spending needs based on income alone should be aware that their portfolio volatility is likely to increase as a result. (Such a change in strategic asset allocation is a ‘move to the right’ along the expected return frontiers in Figure 1.)
Figure 1 can serve as a reminder that, while an allocation to bonds may provide below-average total returns in the next decade, the volatility-dampening properties of bonds should be carefully taken into consideration when developing a sound investment strategy.