Forecasting Failures

By: Andrew J. Willms

President and CEO of The Milwaukee Company

"Forecasting is the art of saying what will happen, and then explaining why it didn't."


Irving Fisher was an influential and highly respected economist and statistician in the early 20th century. Considered by some scholars (including Nobel laureate Milton Freidman) to be “the greatest economist the United States ever produced.” [1] Fisher was of the opinion that the stock market was inherently rational and efficient. It was this belief that led to his prediction in 1929 that "stocks have reached what looks like a permanently high plateau." Three days later, the market plunged, suffering a historic collapse that presaged the Great Depression.

Fisher is far from alone in making market predictions that miss the mark. Since 2010, only 32% of forecasts for the year ahead that were classified by the forecaster as “sure things” were correct; by contrast, 64% were wrong. [2]

Notwithstanding these failures, the number of brokerage firms, financial planners, and portfolio managers who have or will be issuing market forecasts for the upcoming year are almost too numerous to count. Two recent studies have once again raised serious questions about the reliability of such predictions.

The first evaluated 3.8 million forecasts from 1994 to 2019 by stock market analysts in 45 countries. [3] The study considered whether following those forecasts would have led to higher returns. The study compared the top and bottom 20% of equities by consensus analyst recommendations. The study found that on an equal-weighted basis, U.S. analysts’ predictions would have under-performed.

Another recent study considered the accuracy of macro-economic forecasts. [4] This study found that, on average, independently-issued economic forecasts over-reacted to changes in the economic data on which the forecasts were based. At the same time, the researchers also found that, on average, forecasts based on a combination of independent economic forecasts under-reacted to changing economic conditions. The cause, it seems, can be attributed to a flaw in the methodology forecasters most commonly use when preparing consensus forecasts.

Whatever the cause, the lesson to learn from these studies, and many that came before them, is clear: Investors ought to take forecasts by market analysts and economists with a large grain of salt.

One might argue that predictions for a bull market are correct more often than not. But such an argument ignores that the stock market is higher year-end approximately seven times out of ten. As a result, predicating the stock market will go up in the future is about as helpful to investors as a prediction that it will be hot in Arizona next summer is to Arizonans.

Moreover, it’s when forecasters try to provide potentially more meaningful predictions -- such as what market sectors will outperform or what will happen to a specific stock -- that they run into the most trouble. It’s these types of predictions that are notoriously inaccurate.

For example, according to a New York Times report, in 2008, the median forecast called for the stock market to rise by 11.1%. In actuality the stock market fell by 38.5% that year. The foregoing is an extreme example, to be sure. But it illustrates the risk of placing too much faith in attempts to divine the future.

While I don’t put a great deal of faith in forecasts, I do think it’s helpful to prepare for what might happen. As my grandmother was fond of saying: An ounce of prevention is worth a pound of cure. In that vein, next week I will share with you my thoughts about some possible developments that might happen next year, and offer some ideas on how to prepare for these possibilities.

Thank you for reading. Have a great weekend.


[1] Harvard University professor Joseph Alois Schumpeter. [2] Swedroe, Larry, A Second Quarter Review of 2019 ‘Sure Things’ July,2019. Available at https://thebamalliance.com/blog/a-second-quarter-review-of-2019-sure-things/ [3] Azevedo, Vitor and Müller, Sebastian, Analyst Recommendations and Anomalies Across the Globe (October 5, 2020). Available at SSRN: https://ssrn.com/abstract=3705141 or http://dx.doi.org/10.2139/ssrn.3705141. [4] Bordalo, Pedro and Gennaioli, Nicola and Ma, Yueran and Shleifer, Andrei, Over-Reaction in Macroeconomic Expectations (August 2018). NBER Working Paper No. w24932, Available at SSRN: https://ssrn.com/abstract=3239268.

22 views0 comments

Recent Posts

See All