The long-run performance outlook for the Global Market Index (GMI) ticked higher in January, reaching a 6% annualized forecast based on the average estimate for three models (defined below). Today’s estimate remains in line with the trailing 10-year return for GMI, an unmanaged, market-value-weighted portfolio that holds all the major asset classes (except cash).
Today’s revised long-run forecast for GMI is slightly above last month’s 5.9% projection. This estimate has been relatively steady at or near the 6% mark in recent months.
The underlying components of GMI continue to post relatively strong forecasts — defined as forecasts above the current trailing 10-year return. Estimates in this category are indicated by green boxes in the far-right column of the table below.
A glaring exception is the US stock market, which is projected to earn a substantially lower return than its performance for the past decade. GMI’s forecast is also below its 10-year performance, albeit by a small margin.
GMI represents a theoretical benchmark of the optimal portfolio for the average investor with an infinite time horizon. On that basis, GMI is useful as a starting point for asset allocation and portfolio design research.
GMI’s history suggests that this passive benchmark’s performance is competitive with the most active asset allocation strategies overall, especially after adjusting for risk, trading costs, and taxes.
Remember that all the forecasts above will likely be incorrect to some degree. However, GMI’s projections are expected to be more reliable vs. the estimates for the individual asset classes.
By contrast, predictions for the specific market components (US stocks, commodities, etc.) are subject to greater volatility and tracking error than aggregating forecasts into the GMI estimate. This process may reduce some of the errors through time.
Considering how GMO’s realized total return has evolved through time, consider the benchmark’s track record on a rolling 10-year annualized basis. The chart below compares GMC’s performance vs. the equivalent for US stocks and US bonds through last month.
GMI’s current 10-year return (red line) is a solid 6.0%. That’s fallen substantially from recent levels, but it’s been relatively steady lately.
Here’s a brief summary of how the forecasts are generated:
BB: The Building Block model uses historical returns as a proxy for estimating the future. The sample period used starts in January 1998 (the earliest available date for all the asset classes listed above).
The procedure is to calculate the risk premium for each asset class, compute the annualized return and then add an expected risk-free rate to generate a total return forecast.
We’re using the latest yield on the 10-year Treasury Inflation-Protected Security (TIPS) for the expected risk-free rate. This yield is considered a market estimate of a risk-free, real (inflation-adjusted) return for a “safe” asset.
This “risk-free” rate is also used for all the models outlined below. Note that the BB model used here is (loosely) based on a methodology originally outlined by Ibbotson Associates (a division of Morningstar).
EQ: The Equilibrium model reverse engineers expected return by way of risk. Rather than trying to predict return directly, this model relies on the somewhat more reliable framework of using risk metrics to estimate future performance.
The process is relatively robust because forecasting risk is slightly easier than projecting return. The three inputs:
* An estimate of the overall portfolio’s expected market price of risk, defined as the Sharpe ratio, which is the ratio of risk premia to volatility (standard deviation). Note: the “portfolio” here and throughout is defined as GMI
* The expected volatility (standard deviation) of each asset (GMI’s market components)
* The expected correlation for each asset relative to the portfolio (GMI)
This model for estimating equilibrium returns was initially outlined in a 1974 paper by Professor Bill Sharpe. For a summary, see Gary Brinson’s explanation in Chapter 3 of The Portable MBA in Investment.
I also review the model in my book Dynamic Asset Allocation. Note that this methodology initially estimates a risk premium and then adds an expected risk-free rate to arrive at total return forecasts. The expected risk-free rate is outlined in BB above.
ADJ: This methodology is identical to the Equilibrium model (EQ) outlined above with one exception: the forecasts are adjusted based on short-term momentum and longer-term mean reversion factors. Momentum is defined as the current price relative to the trailing 12-month moving average.
The mean reversion factor is estimated as the current price relative to the trailing 60-month (5-year) moving average. The equilibrium forecasts are adjusted based on current prices relative to the 12-month and 60-month moving averages. If current prices are above (below) the moving averages, the unadjusted risk premia estimates are decreased (increased).
The adjustment formula is simply taking the inverse of the average of the current price to the two moving averages. For example: if an asset class’s current price is 10% above its 12-month moving average and 20% over its 60-month moving average, the unadjusted forecast is reduced by 15% (the average of 10% and 20%).
The logic here is that when prices are relatively high vs. recent history, the equilibrium forecasts are reduced. On the flip side, when prices are relatively low vs. recent history, the equilibrium forecasts are increased.
Avg: This column is a simple average of the three forecasts for each row (asset class)
10yr Ret: For perspective on actual returns, this column shows the trailing 10-year annualized total return for the asset classes through the current target month.
Spread: Average-model forecast less trailing 10-year return.