Forecasting futility

Behavioral New World

June 1, 2024

Forecasting for personal finance

“Prediction is very difficult, especially if it's about the future.” — Niels Bohr

Personal financial planning involves forecasting—implicitly or explicitly—many variables, e.g., how will AI affect my career path? (I will use “prediction” and “forecasting” as synonyms.) Here I focus on two factors critical to successful financial planning: the performance of the stock market (in which many of us invest) and interest rates (which affect the interest we receive on savings as well as the interest we pay on consumer debt).

The potential value of accurate forecasting is obvious. For example, if you could have predicted the 10 worst days in each decade starting in 1930 (and withdrew from the market on those days), the return on your stock market investments would be more than 200 times greater than staying invested all the time.(link) Of course, the question is: Can we forecast the 10 worst days? More generally, to what extent can we forecast future stock market returns?

The key to profitable prediction is the ability to identify changes in a stock market index or changes in interest rates. I can predict the level of the stock market tomorrow fairly well—I’ll just guess today’s level and I’ll be within one percent almost all the time. But investment profits come from changes in the level of the stock market—as in, “buy low, sell high.”

The same is true for interest rates. It would be valuable to predict changes in interest rates, not their levels. Tomorrow’s interest rates will be very close to today’s, but we’re much more interested in changes in interest rates.

Here’s the challenge we all face: stock returns are more or less random. And random things cannot be predicted. Making it worse, we are neurologically wired to see patterns and do so even when there are no patterns (see my October 2023 newsletter). Nassim Nicholas Taleb discusses this bias in his excellent book, Fooled by Randomness.

What about interest rates? Consider this quote from Paul Krugman, Nobel prize winner in Economics, on Tuesday, May 21, 2024: “On interest rates, I am fanatically confused.” He goes on to say, “Anyone who claims to know for sure what the answer is to that [how interest rates will change], is deluding themselves.” If a Nobel prize winner can’t see the future of interest rates, why do we think we can?

Although the terms “risk” and “uncertainty” are often used interchangeably, the economist Frank Knight drew a helpful distinction between the two. Risk refers to a situation in which we know all the possible outcomes and their probabilities. Examples: flipping a fair coin or spinning a fair roulette wheel. But that’s not the environment in which we make financial decisions.

So how do we go about personal financial planning under conditions of uncertainty? First, don’t make decisions based on a forecast; the forecast is likely to be wrong! Yes, there are  “talking heads” out there telling you where the stock market is headed. But their forecasts are not much better than a random guess. Worse, some evidence suggests that the more confident the forecaster, the less accurate their predictions.[1]

Second, the suggestions made in my book, The Foolish Corner (link), apply here: hold a well-diversified portfolio with a low expense ratio, don’t trade actively, tailor the risk of your portfolio to your risk tolerance, and don’t look at your portfolio often.

Third, don’t waste your time reading pronouncements about what the Federal Reserve is going to do about interest rates. The time saved will have a higher return if you spend it investing in yourself (a topic for a future newsletter).

In short, “Those who have knowledge don’t predict. Those who predict don’t have knowledge.”— Lao Tzu

---

[1] See the book Expert Political Judgment by Philip E. Tetlock.