It's art, you know.

Photographer: Joe Klamar/afp/getty images

A Better Theory to Explain Financial Bubbles

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

What causes asset bubbles? This question is the great white whale of finance theory. We know that asset prices are given to spectacular rises and falls over short periods of time. Answering this question is hugely important, not just for people’s pensions and retirement, but for the whole economy, since crashes in asset prices can leave growth in the doldrums for years.

QuickTake Watching for Bubbles

But despite decades of research, finance academics have been unable to agree on a cause for this phenomenon. Do investors suddenly become optimistic about asset fundamentals, only to realize a couple of years later that it was all a mirage? Do they buy at prices they know are inflated, hoping to find a greater fool to sell to before the crash? Or do they simply follow the herd?

One possibility is that investors simply make mistakes when projecting future asset returns. If stocks have had an outstanding run for the past five years, or if earnings growth seems to have shifted to a faster trend, people might decide that this is the new normal, and pay prices that later turn out to be ludicrous.

In much of the economics profession, it’s still almost taboo to even consider this kind of human mistake. Most econ models are still based on rational expectations, the idea that people don’t systematically make errors when forecasting the future. This idea was advanced by many star economists of the 1970s and '80s, including the highly influential macroeconomist Robert Lucas. But in finance theory, economists have had more freedom to experiment. So a small but increasing number of papers are asking how markets would behave if investors improperly extrapolate recent trends into the future.

Back in 2005, Kevin Lansing of the Federal Reserve Bank of San Francisco showed that it doesn’t take very much human error to generate extrapolative expectations. If people start believing, even for a short time, that recent trends are the new normal, they will start paying higher prices, which locks the trend in place and lends credence to their belief. Eventually things spiral out of control before they come to their senses. Other economists have shown that even if just a fraction of investors think this way, it can cause repeated bubbles.

A recent paper by Edward Glaeser and Charles Nathanson applies the idea to housing markets. In their model, people decide how much a house is worth by making a guess about how much people will want to live in the area in the future. If buyers expect local demand to increase, it makes sense to pay more for your house, since the influx of other buyers will then drive up prices.

But how do you know whether demand will increase? One obvious way is to look at recent trends. If prices have been going up, it’s a signal that the area is hot. Glaeser and Nathanson note that in surveys, about 30 percent of homebuyers say outright that they use recent price trends to determine how much a home is worth; since others probably do this unconsciously, the true percentage is even larger.

Glaeser and Nathanson’s model makes one crucial assumption -- that investors rely on past prices to make guesses about demand, but fail to realize that other buyers are doing the exact same thing. When everyone is making guesses about price trends based on other people’s guesses about price trends, the whole thing can become a house of cards. The economists show how if homebuyers think this way, the result -- predictably -- is repeated housing bubbles and crashes. 

Extrapolative expectations can be used to account for a number of the phenomena that have puzzled and fascinated finance theorists for years. A 2013 paper by James Choi and Thomas Mertens showed that this kind of expectation formation can explain why U.S. stocks have enjoyed much higher returns than bonds for so many decades. And a masterful 2015 paper by David Hirshliefer, Jun Li and Jianfung Yu showed that extrapolative expectations about productivity levels can cause a lot of the things we see in the real economy -- volatile business investment, relatively stable consumption -- while also leading to a lot of the things we observe in financial markets, such as momentum, bubbles and high stock returns.

So more and more theories are showing that extrapolative expectations might be the answer to many of our financial and economic puzzles. They’re also consistent with what economists find when they conduct surveys that ask people about their predictions of the future. Extrapolative expectations could be the grand unified theory that allows economists to finally understand bubbles, crashes and even the business cycle itself.

But to gain wide acceptance, extrapolative expectations will have to overcome years of entrenched convention in the economics profession. The near-ban on using anything other than rational expectations is still very strong. In the hunt for truth, sociology is often the greatest barrier.

This column does not necessarily reflect the opinion of the editorial board or Bloomberg LP and its owners.

To contact the author of this story:
Noah Smith at nsmith150@bloomberg.net

To contact the editor responsible for this story:
James Greiff at jgreiff@bloomberg.net