Something no investor wants around.

Source: Wild Horizon/UIG/Getty Images

The Fear Factor Behind Stock Prices

Noah Smith is a Bloomberg View columnist. He was an assistant professor of finance at Stony Brook University, and he blogs at Noahpinion.
Read More.
a | A

Why do stocks go up so much? Some financial economists think they are converging on a tentative answer. 

Why stocks go up is one of the fundamental questions of finance. If it seems like a silly question, it’s only because the amazing performance of stocks relative to safe bonds -- known as the equity premium -- has been a fundamental, unchanging fact throughout our lifetimes. The Financial Times sums it up with the simple declaration: “In the long term, nothing beats stocks.” 

Why should this be true? If stocks are so awesome, you would think people would buy them until their price was very high, after which returns wouldn't be so good. The typical answer, of course, is that stocks carry a lot of risk, which investors don’t like. At some point you have to sell your stocks in order to use the money to consume things. Since stocks go up and down a lot, there’s a good chance that when you’re ready to sell, stocks will be at a low point. That’s consumption risk, and it will lead people to avoid stocks to some extent. That in turn pushes the price of stocks down today, making their expected returns higher. In other words, we usually teach people that high long-term stock returns are a compensation for risk. 

In the 1980s, some top economists challenged this view. Using a very simple model of how much people care about consumption risk, they found that people would have to be incredibly risk averse to generate the equity premium that we see. So the equity premium got relabeled the “equity premium puzzle” and spawned tens of thousands of econ papers. 

But maybe the puzzle is no puzzle at all. Only three years after the famous guys came out with the “equity premium puzzle,” an economist named Thomas Rietz quietly proposed a potential solution. You see, the famous guys had assumed that the consumption that people get from stocks is fundamentally derived from a normal distribution. 

Anyone who has worked in the finance industry immediately sees the words “normal distribution” as a red flag. We all know that stock returns are negatively skewed and fat-tailed -- in other words, that there are way too many huge, abrupt stock market crashes to be explained by a standard bell curve. (You would think the creators of the equity premium puzzle would have realized this, but hey, it was 1985.) 

About two decades later -- yes, economics moves at the speed of a glacier sometimes -- a very famous economist brought Rietz’s solution into the limelight. Robert Barro of Harvard University dug into the historical data and found that the number of wars and depressions in history could easily justify the equity premium. 

Since Barro’s paper, the idea of “rare disasters” -- what Nassim Taleb called black swans -- has become one of the main contenders for the solution to the equity premium puzzle. Economists Jerry Tsai and Jessica Wachter show this in a new survey paper. One of the helpful things about the rare disasters framework is that it also helps explain another long-standing puzzle: why stocks are so volatile in the first place. There is rarely an obvious explanation for why stock prices jump around, but it’s possible that these ups and downs represent people’s rising and falling fear that one of the rare disasters is about to strike. In fact, if you make certain assumptions about how the risk of rare disasters changes, you can solve just about any puzzle in the entire macro-finance literature. 

So the creators of the equity premium puzzle just got it wrong -- rare disasters exist, and that’s why stocks are scary. 

But some stubborn questions remain. First, what are these rare disasters? Barro’s 2006 sample, which looks at a lot of different countries, is heavily driven by the world wars and the Great Depression -- which may all be part of one single, huge disaster. Do people really think we are likely to go through another period like the early 20th century? Possibly, but it’s hard to measure. Some economists try to explain disasters as sudden huge drops in productivity, but these aren’t found in the data and that doesn’t really happen. 

Also, the disasters in Barro’s sample tend to involve high levels of inflation, while more recent stock market crashes tend to be very deflationary -- so it seems likely that the type of disasters that threaten us now are very different from those of the past. Will the equity premium remain the same even if the type of risk changes? 

In addition, there’s the question of correlations. In a lot of disasters, asset correlations spike to a very high level: Most stocks crash together. But that means that if people’s fear of stocks -- and therefore, stocks’ high expected return -- is driven by fear of disasters, then shouldn’t most stocks have about the same level of outperformance over time? But as Anisha Ghosh of Carnegie Mellon pointed out in 2012, this isn't what we see. 

So the equity premium puzzle might instead be the rare disaster puzzle. What huge, horrible thing are stock investors afraid of? What is the monster that is going bump in the night?

This column does not necessarily reflect the opinion of Bloomberg View's editorial board or Bloomberg LP, its owners and investors.

To contact the author on this story:
Noah Smith at

To contact the editor on this story:
James Greiff at