The Bank of Canada Has Been Studying If Telling People A Bubble Exists Makes It Worse

Canada’s central bank has had researchers quietly studying bubbles overseas. The study, Bubbles, Crashes and Information Contagion In Large-Group Asset Market Experiments, appeared in the June issue of Experimental Economics. It was the centerfold issue, so obviously, I had to pick it up.

In the study, Bank of Canada (BoC) researchers looked at the role of news in bubble cycles, and its impact on consumers. They found people coordinate their expectations of price growth. When people are told, and believe, an asset is overvalued, they lower future price growth. This helps to stabilize the market for the most part, but large groups were a little more complicated. While they lowered their expectations, it was more difficult to break coordination. This resulted in larger bubbles that were difficult to stop. Though all bubbles pop, it just depends on how far they’ll deviate from fundamentals.

About The Bubble Experiment

The researchers conducted multiple experiments, where people forecast the price of assets. They did so for 50 periods, and the average forecast produced the price. Subjects were paid according to performance, giving it a real-world incentive structure. People were grouped into both small (6 people) and large (92 to 104 people) groups.

When the valuation of the asset became overvalued, random participants received a message. The asset had to reach 3x the fundamental value, then each person had a 25% chance of receiving a headline. The headline was simple, like “experts say the stock market is overvalued.”

The researchers had four major takeaways:

1. Large asset bubbles are “robust” in larger groups. Basically, they were more difficult to break. 
2. Information contagion occurs. When participants were told an asset was more expensive, they forecast smaller gains. Other participants, regardless of seeing the news, followed in line. 
3. Time-varying heterogeneity provides an explanation of bubble formation and crashes. That is, when expectations amongst the group become uniform, bubbles form. When expectations vary, some think it’s a bubble and others think it’s fairly valued, a market crashes.  
4. Bubbles are strongly amplified by coordination on trend extrapolation. If people see prices rise, they’re more likely to expect future price growth. This tends to accelerate price growth even faster, leading to bigger bubbles.

Here are some of the more interesting points, you no doubt will find relatable these days. 

Larger Groups of Investors Make Bigger Bubbles, And They Take Longer To Pop

Conventional wisdom holds that larger markets are more stable than small ones. If a market only has 5 people, one irrational person makes it 20 percent irrational. When that irrational person is in a group of 100, they’re only one percent. The research generally showed this as well, but with an important twist — coordination.

Once a large group entered a bubble, it was more difficult to break the “coordination” of investors. They based their expectations of future prices on what their peers were thinking. It produced groupthink, which made an even larger bubble. “Strong coordination of expectations amplifies the bubble,” wrote the researchers. 

It wasn’t until prices became extreme that some people started to think there may be a problem. Once the group became less uniform about the expectation, prices would crash. Though sometimes they pushed the valuation to the extreme.

News Produces Information Contagion And Helps Break Bubbles

Participants that read news on overvaluation made smaller future forecasts, relative to peers. Since humans coordinate expectations, other people generally adjusted their forecasts lower as well. This helped to create a market crash earlier in some cases.

The market crashing sounds like a bad thing, but crashing sooner than later is better. It prevented the absolute maximum losses that could possibly occur. Not all people saw the newsflash, but enough people coordinated to lower expectations. Though in some cases, the participants kept pushing prices to the absolute limit. Overall though, the researchers conclude the bubble news was stabilizing.

People Get Really Comfortable In Asset Bubbles

People get more comfortable as they experience more bubbles, even if they shouldn’t be. After gaining experience with the first bubble, crash and news included, expectations increased. By the second bubble, they began forecasting larger gains, deviating further from fundamentals. They lost more money, but they were willing to push it to the extreme, while thinking about it less. Inoculation against the fear of losing money, of sorts.

Investors Coordinate On Expectations

The researchers found people coordinate with others naturally, adopting similar expectations. It didn’t matter if they saw the overvaluation news or not. Most would coordinate their expectations of future price growth with those that did.

When prices climbed due to higher expectations amongst the cohort, so did expectations. This occurred even with knowledge of an overvaluation. The reality was less interesting than doing what their friends were doing. Real chimps with pants stuff.

In some groups lowered expectations weren’t enough to stop extreme bubbles. In these cohorts, the market reached the absolute upper bound of overvaluation. Due to coordination, they were willing to risk the maximum they could. Risk was less important than doing what others did.

Circling back to the point, the focus of the study was to determine if news impacts behavior. It not only does, but can play a crucial role in helping to deflate bubbles faster. As people see warnings, they adjust their expectations of price growth lower. Due to market coordination, this can help to stabilize the market over the long-term. Though sometimes it’s not enough to trigger a crash.

For a bubble to pop, participants need to have a varied opinion on where the valuation should be. Either information changes enough opinions and there’s no consensus, or prices hit an extreme. The latter, aka a hard limit, is when prices can no longer rise. It has the worst outcome possible, with maximum losses. What a fun topic for the Bank of Canada to explore now for no particular reason.

Like this post? Like us on Facebook for the next one in your feed.