Global tactical asset allocation (GTAA) is confused by many investors with market timing. Market timing, shunned by nearly all individual investors and many institutions, involves picking market tops and bottoms and, based on that information, making dramatic short-term shifts in asset-class exposures. Needless to say, the successful market timer must have remarkable clairvoyance. The variables that can heavily influence market returns over the short term—that is, a day, a week, a month, even a year—are practically limitless. And when the market timer is wrong, any performance advantage, especially in relation to peers and benchmarks, quickly evaporates.
GTAA is different: It involves seeking to bear risk when risk is likely to be rewarded but moving to a conservative, diversified posture when it is not. Today, despite a sharp sell-off in many risky assets during the past year, the pendulum in the market remains on the risk side, not on the return side. In the last bear market, many investors thought they saw wonderful opportunities a year too early, in 2001, but significant return opportunities did not abound until late 2002. Today, patience is the key to eventually reaping meaningful profits.
The Research Affiliates GTAA process begins with building a long-term forecast that anchors portfolio allocations amid the short-term noise of the markets. These projections start with the simple premise that asset-class returns come from three distinct sources: income, growth in income, and changes in valuations.
The first source, income, is obviously the largest driver of bond returns. But many investors are surprised to learn that since 1926, income from dividends has also been the largest contributor to stock returns—and the dominant source of real equity returns net of inflation.1
For bonds, with income “fixed,” growth in income is, by definition, zero. Indeed, it is actually negative for most bond categories because occasional defaults inevitably take a bite out of the higher coupons. Equities, however, have positive expected income growth; that is, earnings that fund dividends grow along with the real economy. The rate of income growth, positive or negative, depends greatly on the economic and market environment, but the growth rate tends to revert back to a long-term average over time.
Finally, changes in valuations occur because investors may pay more or less for an asset class in 5 or 10 years. Changes in valuations over the short-term (up to a few years) are the most volatile and difficult to predict, and they can be the dominant source of positive or negative returns. But for spans of a decade or more, this component of returns diminishes in significance.
Because income and dividends comprise the lion’s share of long-term asset-class returns, let’s examine these elements in today’s market. Table 1 provides yields for eight asset classes as of July 2008 and March 2002.2 Across the board, bond and alternative categories offered much more competitive yields in 2002 than they do today. Only U.S. equities offer a yield premium (2.3% versus 1.4%), but one could argue that this premium is hardly a bullish figure because today’s current yield is virtually half of the long-term 4.4% contribution of dividends to stock market returns.3
We have combined these asset classes into efficient frontiers of portfolio choices in Figure 1 to reflect the overall risk/reward opportunities in the two bear markets.4 In this analysis, income yields are combined with the other two sources of portfolio returns—namely, projected growth of income and changes in valuations.
The results shown in Figure 1 are striking in two ways. First, the current bear market, with lower yields than the previous bear market in every asset class but equities (which are still low by historical standards), offers little in the way of compelling long-term opportunities. This circumstance contrasts with the earlier bear market, during which attractive assets besides U.S. equities were plentiful.
An even more interesting aspect is the steepness of the two efficient frontiers. The difference in steepness implies that investors were significantly more motivated to bear risk in the earlier bear market.
In 2002, a shift from the portfolio of 4% standard deviation to one of 8% standard deviation was projected to earn a return premium of 2.0 percentage points for bearing the incremental risk. Today, such a move would net only a projected 0.7 percentage point per year. This flat efficient frontier of today illustrates the relatively meager environment for upping one’s risk exposure.
Successful investing is about compounding incremental positive returns, not a cocktail of huge gains mixed with large drawdowns. We are not uneasy about bearing risk; we just want to be paid properly for doing so! By showing patience in today’s world of low returns, Research Affiliates will have the resources to pounce on opportunities when a crescendo of fear drives some of our markets to far lower prices. Such circumstances will occur because far lower prices reflect demand for a far higher risk premium.
Have patience, practice risk management, and stay in the game are the rules for winning in today’s GTAA environment. Concentrating on these three tenets now and taking calculated risks when they present themselves will ultimately earn us meaningful returns.
Learn More About the Author
1. Ibbotson, Roger G., and Peng Chen. 2003. “Long-Run Stock Returns: Participating in the Real Economy.” Financial Analysts Journal, vol. 59, no. 1 (January/February):88–98.
2. We selected March 2002 because the S&P 500 Stock Index declined cumulatively by 21.5% from its March 2000 peak through March 2002, which is virtually identical to the 21.9% sell-off the market experienced from November 2007 through July 2008. Of course, the S&P 500 continued to fall in 2002 and into early 2003, so much so that March 2002 served only as the halfway point of that bear market.
3. Indeed, today’s dividend figure may be artificially high because of the contribution of financial institutions over the last year. On a going-forward basis, this significant sector is likely to see dividend cuts. Similarly, the 2002 figure was biased down by the still hefty presence of the technology sector, with its essentially zero dividend yield.
4. These figures are for illustrative purposes only and are not actual projections used in any Research Affiliates portfolio.