Every so often, the same question resurfaces: Are we facing a bubble? This usually happens when a market, an industry, or a stock rises sharply and rapidly. The common intuition is almost automatic: if something has risen too much, sooner or later it will have to crash. To analyze this situation, we’ll draw on a study by William Goetzmann, Otto Manninen, and James Tyler, which examines the history of the U.S. stock market between 1792 and 2024, challenging this deeply ingrained notion. Their central conclusion is as simple as it is provocative: real bubbles do exist, but they are far rarer than we tend to believe. 

The authors use an empirical definition that is accessible to any reader: a bubble is a sharp rise in prices followed by a decline severe enough to wipe out those previous gains. In other words, it is not enough for a market to rise sharply; to properly speak of a bubble, that rise must eventually collapse. This definition allows us to examine more than two centuries of data and count how many times something like this actually occurred. 

The aggregate market results are striking. When the U.S. stock market rose by at least 50% over three years, that episode was more often followed by further gains than by a total collapse. Over a five-year period, there were 144 instances of equivalent further gains compared to 50 complete reversals. Even over twelve-month periods, large gains did not consistently foreshadow a subsequent catastrophe. In other words, history does not support the idea that a boom is, in and of itself, a reliable signal of an impending crash. Moreover, when the authors look at major declines, they find something equally counterintuitive: severe crashes have tended to be followed more often by recoveries than by further declines of the same magnitude. 

But the study does not stop at the overall index. It also delves into the industry level, allowing researchers to observe many more instances and test whether the phenomenon repeats itself in specific sectors. To do so, it compares two major historical datasets: Cowles Commission industries from 1871 to 1938, and Fama-French industries from 1926 to the present. The result once again defies common sense. Sectors that had experienced extraordinary booms—with average gains of 300% over two years under one of their definitions—did not, on average, follow a subsequent path of collapse. Rather, after the boom, they tended to stabilize, with relatively flat subsequent trajectories and, in many cases, performance close to that of the general market. 

So where does the perception that booms are dangerous come from? This is where perhaps the most useful idea in the paper comes in. The authors show that booms are indeed associated with a higher probability of a crash, but not because they necessarily signal a negative average return. What they actually signal is greater volatility. After an extreme boom, the probability of a major crash does increase, yes, but so does the probability of new extraordinary gains. The boom does not select sectors “doomed” to fall; it selects sectors that have entered a regime of wide and uncertain swings. That is a crucial distinction, especially for non-specialists: it is one thing to foresee certain losses and quite another to warn that the path ahead is becoming more unstable. 

The study itself is also careful to acknowledge its limitations. The authors point out that many extreme observations are rare and occur at the same time, so it is unwise to draw exaggerated conclusions based on a few historical episodes. They also note that the United States has been a particularly successful market throughout history, so not everything learned there can be mechanically extrapolated to any other country. Even so, the overall message is powerful: stock market history does not suggest that every major rally inevitably ends in disaster. Rather, it shows that markets—and especially innovative industries—go through intense cycles in which euphoria, risk, and continuity coexist. In times when the word “bubble” is used lightly, this paper offers a lesson in prudence: rising sharply is not the same as being doomed to burst. 

Gabriel Haensgen
Portfolio Manager, Financial Funds, Fynsa AGF