At Butler|Philbrick & Associates, we don’t take anything on faith. Nor do we take expert opinions to heart, as we have shown time and again that experts make poor oracles. Instead, we believe in crunching the numbers ourselves to discover meaningful relationships in data. Where meaningful relationships exist, we apply statistical models to improve our chances of success.
Now that we have developed and deployed Version 2 of our systematic investment model, we decided to shift our attention to the development of a more robust model for forecasting long-term stock market returns. Traditional Advisors assume that the best estimate of future market returns in all market environments is the simple long-term average return on stocks: about 6.5% per year after inflation.
We hypothesized that it is possible to construct a statistical model using long-term market data which will allow us to make much more accurate predictions about long-term returns. It turns out that we were right. Those who are interested in the process we used, and the specifications of our model, are encouraged to read our full report.
There are several reasons why it may be useful to have a more robust estimate of future expected returns on stocks:
- People who are approaching retirement need to estimate probable returns in order to budget how much they need to save.
- A retiree’s level of sustainable income is largely dictated by expected returns over the early years of retirement.
- Investors of all types must make an informed decision about how best to allocate their capital among various investment opportunities.
Many investors do not know that traditional wealth advice is rooted in the assumption that the best estimate of future returns is always the average long-term return to stocks. No matter where markets are on the continuum from very cheap to very expensive, traditional Advisors will make recommendations on the assumption that investors should expect 6.5% inflation adjusted returns on stocks over all investment horizons.
To illustrate, imagine a retiree who visited a traditional Investment Advisor at the peak of the technology bubble in early 2000, when markets were more expensive than at any other time in the prior 130 years. This investor would have been advised to expect returns on his stock portfolio of 6.5% per year over his or her investment horizon, based on very long-term averages.
Our models suggest that this retiree should have expected inflation-adjusted returns to his portfolio of negative 2% per year over the subsequent 15 years, a difference in returns of 8.5% per year versus the long-term average. In fact, this investor would have experienced returns of negative 1.55% per year through December 31, 2010, and would need a return of almost 22% per year through 2015 to realize the traditional advisor’s year 2000 projections.
Table 1. applies this same analysis to other important periods over the past 100 years. We contrasted the return forecasts from a traditional long-term average approach with the forecasts from our valuation-based model, at a variety of transitional dates in stock markets, to demonstrate the improved accuracy of our valuation-based approach.
You can see that forecasts derived from long-term average returns yield over 400% more error than estimations from our valuation-based model over these 15-year forecast horizons (1.24% annualized return error from our model versus 5.24% using the long-term average). Clearly our model offers substantially more insight into future return expectations than simple long-term averages, especially near valuation extremes.
Chart 1. shows how closely our valuation-based model forecasted actual market returns over subsequent 15-year periods. The blue series is our model forecast, and the red series tracks actual market returns. The red line near the middle of the chart reflects a 6.5% annualized return.
So what does our model suggest about current market valuations and future expected returns? You can see from the chart’s blue line that expected returns from current levels are well below average. Even at the market’s lows in March of 2009, expected returns to stocks over the subsequent 15 years was just average, suggesting that markets simply achieved long-term average valuations at the market’s low. We were, and are, a far cry from the generational low valuations achieved around 1920, 1950, and 1980. If history is any guide, we may achieve those generational-low valuations – which represent once-in-a-lifetime opportunities to buy stocks – at some point in the next 5 to 10 years. Of course, we must endure another period of very low returns to achieve such low valuations.
So what level of annualized returns should we expect from stocks, after adjusting for inflation, over the next 5, 10, 15 and 20 years based on current valuations? Table 2. summarizes our model’s forecasts over these horizons. Only time will tell how accurate they might be, but history clearly proves that future returns will be much closer to these values than the long-term average of 6.5%.
The investment industry has a large vested interest in convincing you that the same approach that delivered poor returns over the prior decade will deliver much more robust results over the next few years. That way, you will be convinced to hold your money in the same traditional, high margin products that made banks so much money, and lost investors so much money, over the last 10 years.
In periods of low returns, investors must have the courage to adopt a different approach if they hope to achieve better-than-average results. Our Gestalt Architecture was engineered to deliver strong returns in all markets, including markets which drop in value over several years or months. How does your Advisor plan to deliver robust results in the likely event that future returns are well below average?