Go to the Globe and Mail homepage

Jump to main navigationJump to main content

The pre-market price for Facebook stock is shown, Wednesday, May 23, 2012 at the Nasdaq in New York. Facebook stock rose in early trading, although still far below the $38 it was priced at before its initial public offering. (Mark Lennihan/Mark Lennihan/AP)
The pre-market price for Facebook stock is shown, Wednesday, May 23, 2012 at the Nasdaq in New York. Facebook stock rose in early trading, although still far below the $38 it was priced at before its initial public offering. (Mark Lennihan/Mark Lennihan/AP)

Why do we keep falling for economic bubbles - and will we ever learn? Add to ...

The guessing game of how much Facebook would ultimately be worth started long before the social-networking behemoth officially announced that it would become a public company. In early 2011, when Goldman Sachs bought a small stake, there was word that the site could be worth $50-billion (U.S.). Then last June, less than six months later, news network CNBC upped the ante to 12 figures on air, dubbing the company America’s “$100-billion baby.”

More related to this story

The general public devoured the speculation – investors believed that because they and everyone they know use Facebook incessantly, it must be worth an exorbitant amount. When it made its initial public stock offering a week ago, its valuation lived up to the hype at $104-billion – only for the stock price to drop nearly 20 per cent after three days of trading. On Wednesday, angry investors launched a class-action lawsuit charging that Facebook and its underwriters made “false and misleading representations” before the initial offering.

Was this a bubble bursting, or at least showing signs of strain? Companies in the social-media business have been getting scooped up for jaw-dropping amounts, despite often-remote potentials for profit. (Facebook itself recently bought Instagram, a smartphone photo app with zero revenue, for $1-billion.)

For those who lived through the fallout of the last tech bubble, in the late 1990s and early 2000s, this looks very familiar: Barely a decade ago, irrational expectations of Silicon Valley ended up sending the U.S. economy into recession; investors got hosed when the Nasdaq, the main market for tech listings, lost more than three-quarters of its total value. It is mind-boggling that we are back here so soon.

Such behaviour seems all the more absurd when you look back at the plethora of bubbles in history: People have fallen for everything from Holland’s “tulip mania” in the 1600s to the vicious Japanese housing bubble in the 1980s and the North American one that burst in 2008. We’ve even succumbed to mini-bubbles of frenzied speculation around trivial things, such as the Beanie Baby toy bonanza of the late 1990s. If humans are supposed to be the most intelligent species on Earth, shouldn’t we have learned our lessons by now?

Apparently not. University of Virginia economist Charles Holt ran an experiment that asked a group of subjects to buy and sell a fake asset, attempting to make the most profit. Previous runs of the test had often revealed bubble-like behaviour.

This time, though, smack in the middle of the simulation, one of the participants unexpectedly calculated the asset’s intrinsic value, based on its documented profits, and shouted it out. Mr. Holt assumed the experiment was doomed – surely everyone now would just buy and sell the asset at its true price.

But soon he saw something he could barely believe: Students started to buy at higher prices again; not long after, the person who had run the numbers started doing the same. Ultimately, they all traded based on what everyone else was doing, rather than what the numbers showed.

Experiments of this type are common in the school of behavioural economics. Butting heads with classical theory, these economists’ work is predicated on psychology and biology, not computer models.

While the field has been around for decades, its work is only now going mainstream, buoyed by technological advances that have allowed neurologists to study what happens to human brains in a bubble and adding scientific research to back up the behavioural economists’ theories.

Their message is simple: If we’re ever to avoid the “irrational exuberance” (in former U.S. Federal Reserve chairman Alan Greenspan’s famous phrase) that accompanies economic bubbles, we’ll have to get a better handle not on market forces, but on our brains.

A fatal formula: envy, overconfidence and habit

At 85 years old, Vernon Smith is one of the grandfathers of behavioural economics. A Nobel Prize winner, he is renowned for toying with theories of irrationality as early as the 1960s, when it was practically unheard of to question the crux of classical economics – that the economy is composed of rational actors following their self-interest.

One of his experiments asked subjects to bid on an asset whose value dropped by the same amount in every round of the simulation – a scenario similar to what a mining firm experiences, because its value drops every time it digs another ounce of its resource out of the ground.

Because the simulation’s linear trend was so easy to decipher, the subjects should have picked up on it quickly. But they didn’t. Comically, they made themselves believe that they could unearth some hidden value, and continued to bid up.

For decades, experiments of this sort have clearly shown that the subjects do not act rationally, but it has always been hard to pinpoint why. Study after study has highlighted a set of dominating characteristics that drive our behaviour in bubbles: envy, overconfidence and a reliance on past performance as a predictor of future gains.

On a suburban street, for example, homeowners who see their neighbours sell their houses for big profits are bound to want the same thing – that’s envy. The neighbours who have cashed in suddenly feel invincible, making them think they can correctly time the market again – overconfidence.

And both groups, in awe of a seemingly never-ending rising market, develop a myopic focus on the recent gains, rather than looking at historical averages – reliance on past performance.

Other studies have pointed to the phenomenon of comfort in numbers: Walking down a street, a passerby is more likely to choose a restaurant that has people sitting in the window than one that is empty, even though the quieter one could have better food. In uncertain situations, humans follow the lead of others, and at the very core, this is what happens in bubbles: People react to what everyone else is doing.

Another factor is the nature of regret: When we cash out an investment, our minds automatically calculate how much we could have made, theoretically, if we had kept riding the bubble.

Neuroeconomic experiments have showed that in bull markets, this fear of regret convinces people that they should invest more and more.

As the neuroscience guru Jonah Lehrer puts it, when we look at our handsome profits from investing 50 per cent of our portfolio, we quickly start to imagine how much money we would have made if we’d invested everything.

Then there is the danger that our brains, which are typically quite good at pattern recognition, sometimes overreach and see patterns that are not really there. This may be part of what is happening during a bubble.

The more complex the idea, however, the harder it is to prove with absolutely certainty. Even the most sophisticated behavioural economists will admit that it is almost impossible to be sure something like the restaurant effect is at play in a bull market.

Still, there is one theory almost no one disputes – the difference between thinking fast and thinking slow.

These terms were coined by Daniel Kahneman, a renowned behavioural psychologist and Nobel laureate, and explored in his recent bestselling book, Thinking, Fast and Slow: The “fast” system for decision-making relies on instincts and emotions; the “slow” one is more deliberate and analytic, but requires conscious, taxing effort.

Because humans are constantly inundated with overwhelming amounts of information, our fast systems have learned to make split-second decisions based on similar situations in the past. For the most part, this works out fine. But the human brain comes complete with some glitches.

“That’s sort of a necessary consequence of having a system that can do all kinds of other smart things,” says Scott Huettel, co-director of the Center for Neuroeconomic Studies at Duke University.

For example, the moon always looks bigger when it’s closer to the horizon than it does when it’s high in the sky – we are fooled by the context, although the moon itself hasn’t changed size.

But investors are loath to admit that there are limits to their rational abilities. Like a person in an emotionally charged bad relationship, humans are not willing to let go, because we too often believe that we are the exception – that we can make it work.

Mr. Kahneman came up with a name for this phenomenon: inside and outside views. With an outside view, a bunch of different items can be categorized based on their shared properties. With an inside view, certain items within the category appear to have unique traits.

An example is offered by Colin Camerer, a California Institute of Technology professor often regarded as the most important behavioural economist to emerge since Mr. Kahneman: While any marriage has about a 50-per-cent chance of surviving, try telling that to a newly married couple – they’re sure to believe their union is the one that will last.

Research in this area has discovered that is it particularly difficulty to adopt an outside view when ego is involved, when a lot of time has passed or when the issue is a personal matter.

All those characteristics are at play in bubbles: Ego is almost always a factor; investing your own cash inherently makes a bubble a personal problem; and the further back the last bubble was, the harder it is to see similarities.

There is an old saying in psychology departments: “People often question their eyesight or vision, but not their judgment.”

It’s different every time

The fact that people get caught up in economic bubbles over and over again may seem like a critical flaw in human evolution. But the real problem is that no two bubbles are precisely the same.

“What people are really good at is learning about events that they experienced personally, over and over again,” says Dr. Huettel, the Duke neuroscientist. “You aren’t going to get snookered by the same person in poker many times in a row because you pick up on irregularities.”

While bubbles seem to share general characteristics, we convince ourselves of even the slightest differences. For instance, the current tech bubble can be rationalized because it is predicated on the proliferation of social media, not the growth of the Internet.

There is also a lot of evidence that the surge of testosterone traders experience during a speculative rush can cloud decision-making.

John Coates, a Canadian-born research fellow in neuroscience and finance at the University of Cambridge and a former trader at Goldman Sachs and Deutsche Bank, has run experiments on trading floors in the City of London.

“Once you start making above-average profits, as most people do during a bull market, you start getting this high,” he says. “I think it’s enough to pretty much squash memory” of previous bubbles.

His new book, The Hour Between Dog and Wolf, details these findings and ties them back to what the behavioural economists started studying years ago.

Prof. Coates admits that even he, someone equipped with a PhD in economics from Cambridge, has fallen victim to the testosterone highs. “I don’t think I ever would have hit on this if I hadn’t experienced it myself,” he says. “We have an unstable biology, and it’s very powerful.”

Yet there are people – even whole firms – who appear to effectively game these bubbles. At a recent luncheon, Prof. Holt, the University of Virginia behavioural economist, had a conversation with a successful hedge- fund manager who confided that he did not trade on companies’ long-term fundamentals.

Instead, he looked at the previous seven days of trading and read newspaper headlines and television talking points, going by the theory that economist Burton Malkiel espoused – that “a blindfolded chimp throwing darts at the Wall Street Journal” had a 50-per-cent chance of beating the market, because humans are so subject to their irrational psychology.

The question, then, is whether humans can control their urges. Prof. Holt himself admits that even after studying bubbles for decades, he ultimately bid on a $1.2-million house at the height of the U.S. housing bubble.

Luckily for him, he backed out at the last minute because his father, an engineer, caught some structural problems during an inspection.

Human biases are “so ingrained that just knowledge isn’t enough to overcome them,” Dr. Huettel says.

“From an evolutionary point of view,” says Caltech’s Prof. Camerer, controlling urges “basically isn’t something that any other species needed the capacity to do.” Add in the financial incentives on Bay Street and Wall Street, and we practically jump at bubbles.

For traders, “there’s no downside to rolling the dice,” Prof. Coates says. “Bubbles occur once every five years, but in the meantime you’ve pocketed four bonuses, and you don’t give them back.”

Can we game the system?

There certainly are individual exceptions. Prof. Holt chuckles when he recalls that behavioural-economics pioneer Vernon Smith had bought a beautiful penthouse apartment near George Mason University in Virginia. A few years after Prof. Holt visited it, he asked some George Mason researchers what happened to Prof. Smith’s apartment. They said he had sold it because he realized the market was getting too heated.

Where Prof. Holt had almost bought a million-dollar home at the height of the bubble, Prof. Smith had smartly sold into the rally.

Collectively, though, behavioural economists and their neuroeconomist colleagues have had trouble coming up with theories of how to curb speculation. Prof. Coates, for example, advocates government intervention in the aftermath of a bubble, but he doesn’t see how the government can step in when things are too hot.

In his research, he found that in a bear market, when asset prices are falling, cortisol – testosterone’s hormonal opposite – floods traders’ bodies and inhibits them from taking on any type of risk.

Under those conditions, he says, governments must step in and facilitate trading for a short period to stabilize the markets.

But when the market is too frothy, he says, it would be very hard for regulators to tell investors the fun was over: “It’s very difficult to take away the punch bowl.”

In absence of any surefire cure, however, the underlying consensus is that acknowledging the pervasive influence of irrationality would be a start.

Glancing at the bestseller lists, it would appear that this message has caught hold – books such as Thinking, Fast and Slow, Nudge and even Freakonomics have sold scads of copies. Yet the people who read them often assume that it is everyone else who is irrational.

Instead, like an alcoholic trying to get clean, the first step is for each of us natural bubble addicts to tame our egos and admit that we have a problem.

Single page

Follow on Twitter: @timkiladze

In the know

Most popular video »

Highlights

More from The Globe and Mail

Most Popular Stories