Pre-Tax Corporate Profits Forecasts for the US Economy: An unresolved Challenge
This study analyzes US pre-tax corporate profit forecasts from 1990 to 2024 and reveals the surprising anomaly that forecast accuracy does not improve with shorter horizons, alongside evidence of herding behavior, underestimation of variability, and a lack of excessive optimism among forecasters.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
In the world of investing, the most fundamental question is simple: will a company make more money next year than it did this year? For decades, financial experts have built their careers on answering this. They study business reports, talk to managers, and crunch numbers to predict the future profits of corporations. These predictions are the compass for investors deciding where to put their money. If the experts believe a country's companies will thrive, investors pour cash into that nation's stock market. If the experts see trouble, they pull back. The entire system of active investing relies on the assumption that these experts can see the future better than the past, and that their guesses get sharper the closer they get to the actual event. It is a reasonable belief: just as a weather forecast for tomorrow is usually more accurate than one for next month, a profit prediction made in December should be more reliable than one made in January.
A team of researchers from Germany decided to test this assumption with a massive, long-term look at the United States economy. They gathered every monthly prediction made by major banks and research institutes from 1990 to 2024 regarding the change in pre-tax corporate profits. They tracked these forecasts over a period of two years, watching how the predictions evolved month by month as the actual outcome approached. Their goal was to see if the experts truly learned from new information, or if they were simply guessing in the dark. What they found was a startling contradiction to everything the financial world believes about how forecasting works.
The study examined a vast collection of data, looking at how well these experts predicted the percentage change in US corporate profits year over year. The researchers measured the distance between what the experts predicted and what actually happened. They expected to see a clear pattern: as the months passed and the end of the year drew near, the errors in the predictions should shrink. The logic is that more information becomes available over time. Companies release quarterly reports, economic data is finalized, and the fog of uncertainty should lift. However, the data told a completely different story. The average error in these predictions was enormous, missing the mark by nearly 95 percent on average. More shockingly, this massive error rate did not improve at all as the deadline approached. A prediction made twenty-four months in advance was just as inaccurate as a prediction made just one month before the result was known. The experts were not getting better; they were staying exactly as wrong as they were at the start.
This result stands in stark contrast to how experts handle other economic numbers. When the same researchers looked at forecasts for gross domestic product, inflation, or housing starts, they saw the expected pattern. For those topics, the predictions became significantly more accurate as the forecast horizon shortened. The experts clearly knew how to use new information for those subjects. But for corporate profits, the improvement simply did not happen. The researchers ruled out the idea that the experts were just slow to react. The data showed that the experts did update their numbers frequently, often swinging wildly from one month to the next. They were not ignoring the news; they were reacting to it, yet their final guesses remained just as far off the mark as their initial guesses.
The researchers then looked at what the experts were doing with each other. They noticed that while the accuracy did not improve, the experts did start to agree with one another. As the deadline for the actual profit numbers approached, the spread of their predictions narrowed. At first, the experts had widely different opinions. By the end, they were all saying almost the same thing. This convergence suggests a behavior known as herding. Instead of each expert independently analyzing the new data to find the true answer, they seemed to be following the crowd. They were aligning their views with the majority opinion, perhaps to avoid the risk of being the only one wrong, but in doing so, they stopped trying to find the actual truth. They became a unified group of people who were all equally mistaken.
Another surprising discovery was that the experts were not overly optimistic, a common criticism in finance. In fact, the data showed they were too cautious. When the actual profits grew by a certain amount, the experts consistently predicted a smaller growth rate. They underestimated the potential of the economy. They also failed to imagine the extremes. The real world saw huge swings in profits, both massive gains and severe losses, but the experts' predictions were always clustered in the middle. They underestimated the volatility of reality, imagining a smoother, calmer world than the one that actually existed.
The implications of these findings are profound for anyone managing money. If the experts who are paid to predict the future cannot do so, and if their predictions do not get better with time, then the strategy of actively picking stocks based on these forecasts is fundamentally flawed. The study suggests that the inability to forecast corporate profits accurately is a major reason why trying to beat the market is so difficult. It implies that the complex, active strategies used by many investment firms may be chasing a ghost. The data indicates that for the US economy, the best approach for an investor might be to accept that the future is unpredictable and to stick with passive strategies that do not rely on these broken predictions. The experts are not seeing the future; they are simply following each other into a fog they cannot clear.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.