Why You Should Ignore 2021 Bond Return Forecasts
If you are wondering how to make use of the forecasts of 2021 corporate bond returns currently being published by various investment firms, the best advice: “ignore them.”
The story to keep in mind when you see one of those forecasts involves the late Nobel economics laureate Kenneth Arrow. During World War II he was part of a team of statisticians assigned to make weather forecasts. After a while, the statisticians concluded that their forecasts were not much of an improvement over pulling predictions out of a hat.
Arrow and his colleagues requested reassignment to more worthwhile duties, only to receive the following reply: “The Commanding General is well aware that the forecasts are no good. However, he needs them for planning purposes.”
For some reason, financial firms believe that at the end of December—but at no other time of the year—their clients need a 12-month total return forecast. Never mind how badly the strategists have failed in their previous attempts to predict returns over such a short a horizon. (You will notice that the forecasts are not accompanied by audited records of past accuracy.)
If these year-end forecasting rituals encourage income-oriented investors to attempt to time the market, they cannot just be dismissed as harmless fun. Jumping in and out of the market in anticipation of short-run movements in interest rates is a losing proposition. That is not to say, however, that it never pays to be forward-looking. A long-term investment plan that takes expected future returns into account can be beneficial.
Happily, long-run bond returns are inherently more predictable than short-run returns. This is because the forecasts are based on both a known factor, the present yield, and many unknowns. Prominent among the unknowns are what the state of the economy and the level of interest rates will be at the end of the forecast period. The difficulty with one-year forecasts is that the unknown part is a much bigger piece of the short-run than of the long-term picture. This is demonstrable using data from 1985 onward for the investment grade ICE BofA U.S. Corporate Index.
MORE FOR YOU
Short-Run versus Long-Run Forecasting
In statistical terms, the year-end yield-to-maturity (YTM) has a 92% correlation with the annualized total return over the succeeding five years. The comparable figure for one-year returns is 55%. This means that forecasters have a lot more room to get tripped up by unforeseeable events when they forecast for periods as short as 12 months than they do when taking the long view.
On average, the difference—higher or lower—between the end-of-year YTM and the one-year return is 5.10 percentage points. Based on the corporate index’s 1.80% YTM on December 29, it is reasonable to expect the 2021 return to come in somewhere between -3.30% and 6.90%. Forecasters face a daunting task in trying to narrow it down from there.
By contrast, the five-year annualized return is, on average, 0.97 percentage points higher or lower than the beginning YTM. By implication, there is a good chance that the surprises that will undoubtedly occur between now and the end of 2025 will not push the five-year annualized return outside the comparatively narrow range of 1.80% plus/minus 0.97%, or 0.83% to 2.77%. It is consistent with a responsible approach to investing to proceed on that premise and consider anew your appropriate percentage allocation to investment grade corporate bonds, given your age, financial circumstances, investment objectives, and other available choices.
In summary, one-year forecasts of bond returns have little value, other than entertainment value. It can be fun to see which prognosticator comes closest to the mark. Do not infer, however, that the winner triumphed through skill rather than luck. Above all, do not shoot yourself in the foot by reacting to predictions for periods as short as 12 months. Over that timeframe, the outcome is too heavily dominated by unknowable factors.
 Interestingly, the statistical team’s sense of futility involved long-range forecasts. Weather is more predictable in the short run than in the long run, the opposite of the case with bond returns.