GMI's power as benchmark inspires calculating its expected return for additional insight. Ideally, we're looking for an unbiased estimate of future performance. That's a high standard and difficult (impossible, really) to achieve. One way to get close to this ideal is by calculating GMI's implied risk premium (total return less an estimate for the "risk free" rate).
In contrast to the standard methodology of forecasting returns directly, this technique reverses engineers expected return based on several assumptions about risk. In particular, I'm making assumptions about 1) correlations for the major asset classes relative to GMI; 2) volatility for each asset class; and 3) GMI's ex ante Sharpe ratio. After calculating expected risk premiums for each asset class, I sum the weighted estimates (weighted by current market values). For some background on this technique, which I've modified slightly, see the "Reverse Optimization" section in Thomas Idzorek's monograph or Chapter 3 in The Portable MBA in Investment
The results are illustrated in the graph below, which compares GMI's historical risk premium on a rolling three-year annualized basis (red line) with two estimates of the benchmark's expected premium.
The green line is the estimated "tactical" premium for the near term—roughly the next three to five years. This estimate is based on predicting volatilities for each of the major asset classes using a simple a simple GARCH (1,1) model. In turn, I make some basic assumptions about correlations with GMI from these estimates on vol. I also calculate what's known as a modified Sharpe ratio for GMI. All of these estimates are computed monthly.
The "strategic" estimate of GMI's future risk premium, represented by the blue line in the chart above, assumes a stable Sharpe ratio of 0.2 for the portfolio. Why a 0.2 Sharpe ratio? The short answer: history suggests that this is a reasonable long-term guesstimate for a multi-asset class portfolio. Otherwise, the methodology for the "strategic" forecast of GMI's risk premium is identical to the tactical calculation.
Note that both the "tactical" projection and the actual real world record of GMI's risk premium bounce around the "strategic" estimate. That's what we would expect to see since the "strategic" estimate reflects a long-term outlook under the key assumption that the markets clear in the long run. Accordingly, the short-term realized and "tactical" estimate of GMI's premium fluctuate quite a bit. We can think of the wider fluctuations as a type of error relative to the trend. In that case, the deviations from the trend provide clues about the magnitude of GMI's premium in the near-term future.
Not surprisingly, GMI's risk premium was relatively high after the financial crisis of 2008 and Great Recession. Meantime, it's no shock to see that the implied risk premium for the near term has fallen recently.
In the short term, my estimates say that GMI's risk premium will be roughly 4.7% a year (plus whatever you expect for a risk-free rate to calculate the total return). But that's double the expected risk premium for GMI using the "strategic" methodology, which suggests that performance surprises are for multi-asset classes are likely to be negative for the foreseeable future.
Yes, a risk premium under 5%--perhaps well under 5%--is quite low. How could you boost expected return? You could start by considering an asset allocation that deviates significantly relative to GMI's passive mix. Or you could engage in a relatively aggressive strategy of tactical asset allocation. Or both. But keep in mind that relatively few professionals end up beating GMI over the medium- and long-term horizons. Those that do usually embrace a fair amount of risk, one way or another. There are still no free lunches in the money game. The good news is that it's relatively easy to enhance the odds of ending up in the upper half of the performance rankings. How? By keeping radical bets on asset allocation to a minimum.