Experimental Economics and Market Bubbles in the Lab
Can you grow a stock market bubble inside a classroom? Not a metaphor. An actual bubble where prices rise above what everyone in the room knows the thing is worth?
Turns out, yes. You can. And people have been doing it for decades.
Chapter 21 of Burton and Shah’s book covers experimental economics. It is a field that took ideas from physics labs and brought them into economics. And some of the results are wild.
Wait, Economics Has Labs?
For most of its history, economics was an armchair discipline. Theorize, build models, argue. But no experiments. In physics, you have a hypothesis, you test it. In economics, you just… debated.
That changed in the second half of the twentieth century. Researchers started building controlled market games. Real people. Real money on the table. Not hypothetical. Actual cash payments depending on how you play.
Here’s the thing. If you run an experiment without real money, it is just a classroom exercise. Students learn something, sure. But economists don’t trust it as real data. When real money is involved, participants are motivated the same way they would be in actual markets. That is what makes it legitimate research.
The setup usually looks like a game. There are buyers and sellers. Each person’s outcome depends on their own decisions and what others do. It is game theory in action. And most of these games are not zero-sum. One person winning doesn’t automatically mean another loses.
Bubbles in a Box
This is the part that blew my mind when I first read about it.
A bubble is when an asset’s price rises above its real value and keeps climbing. But in real markets, we have a problem. Nobody actually knows the fundamental value of a stock. Is Tesla overvalued? Is Apple undervalued? People argue about this endlessly, and nobody can prove anything because fundamental value is not directly observable.
But in a lab, you can fix that. You can create an asset with a known value. You tell participants exactly what dividends the asset pays, what the probabilities are, and what the terminal value is. Everyone has the same information. Everyone can calculate the expected value on their own. Some experiments even hand participants the expected value directly, just to be safe.
So, rational people with perfect information should price the asset at its fundamental value, right?
Wrong.
Vernon Smith and the Classic Bubble Experiment
Vernon Smith, who later won a Nobel Prize for this work, ran some of the earliest experiments on asset bubbles. The landmark study was by Smith, Suchanek, and Williams in 1988.
Here is how it worked. Participants were given some money and some units of an asset. They could trade over 15 periods. Everyone knew the dividend structure. Everyone could calculate what the asset was worth.
And bubbles formed anyway.
Out of 22 clean experiments (no disruptions or researcher intervention), 14 produced bubbles. That is almost two-thirds. Prices went above fundamental value, stayed there for a while, and then crashed. Even when people literally had all the information they needed to avoid overpaying.
The researchers were investigating two big questions. First, does giving everyone identical information prevent bubbles? Prior work suggested that bubbles need information asymmetry, that one group knows something another doesn’t. But SSW showed that is not true. Bubbles happen even with perfect, identical information.
Second question: does experience matter?
Experience Helps, But New Traders Keep Showing Up
This part is interesting and very relevant to real markets.
Yes, experienced traders produce fewer bubbles. When people who have been through the experiment before play again, they are less likely to bid prices above fundamental value. The market calms down with experience.
Later research by Dufwenberg, Lindquist, and Moore dug deeper into this. They found that you need at least one-third of traders in the market to be experienced before bubbles stop forming. Just a few experienced traders mixed in with rookies is not enough. The rookies drive the prices up anyway.
And this explains something important about real markets. Why do bubbles keep happening if experience prevents them?
Because new traders keep entering. Markets that are going up attract fresh participants. People who have never traded before see prices rising and want in. During the dot-com boom, your neighbor who never owned a stock suddenly had a day-trading account. During the crypto run, your cousin who couldn’t explain what a blockchain was became a Bitcoin evangelist.
The experienced traders know better. But when enough new, inexperienced people flood in, they overwhelm the experienced ones and push prices into bubble territory.
Can Anything Prevent Lab Bubbles?
Researchers tried a few things.
Short selling. The results are mixed. Some studies found that allowing short selling reduced bubbles. Makes sense. If you think something is overpriced, you can bet against it, and that should push prices down. But other researchers found the opposite. Haruvy and Noussair concluded that “simply adding short-selling capability does not appear to eliminate the trader behavior that underlies bubble formation.” So short selling alone is not a reliable bubble-prevention tool.
Futures markets. Porter and Smith found that adding a futures market (where people can bet on future prices directly) made bubbles smaller. Not eliminated, but smaller. The futures market gives people another way to express their view on the asset’s value, and that extra channel of information seems to take some air out of the bubble.
But nothing they tried killed bubbles completely.
The Endowment Effect Under the Microscope
Chapter 21 also touches on how experimental methods have been used to test the endowment effect. This is the idea that people value things more just because they own them. We covered it in earlier chapters.
Richard Thaler connected this to status quo bias. People tend to stick with defaults. Whatever option they start with, they keep. This has huge practical implications.
Think about retirement savings. If a company automatically enrolls employees in a 401(k) plan, most people stay enrolled. If employees have to opt in, most don’t bother. The default matters enormously because people are biased toward the status quo.
This is the basis for “nudge” theory. If you want people to make better choices, don’t force them. Just set better defaults. Make the good option the default, and human laziness (or inertia, if you want to be polite about it) does the rest.
Calendar Effects in the Lab
Here is a quirky one. Stock markets show calendar patterns. Stocks tend to do better in January than other months. Prices sometimes behave differently around holidays and weekends.
Researchers tested these effects in the lab. Coursey and Dyl created trading interruptions in their experiments (like weekends) and found price patterns similar to what we see in real markets around holidays.
Anderson, Gerlach, and DiTraglia ran identical experiments at two different times. One set in December, one in January. Same experiment, same structure, same everything. But prices in the January experiments were significantly higher.
Why? Nobody knows for sure. But the fact that calendar effects show up even in controlled lab experiments means they are probably not just statistical noise. Something real is going on with how people behave at different times of the year. Maybe optimism at the start of a new year? Fresh budgets? New Year’s resolution energy applied to investing?
This is an area where more research could really add to our understanding.
So What Does This Mean for You?
The big takeaway from experimental economics is uncomfortable but important.
Bubbles are not caused by ignorance. They are not caused by lack of information. They happen even when every single participant knows exactly what an asset is worth. Something deeper in human behavior drives prices above value and keeps them there until the crash.
Experience helps. If you have been through a bubble and crash before, you are less likely to get caught up in the next one. But there will always be new participants entering the market who haven’t learned that lesson yet. And if enough of them show up at the same time, the bubble inflates regardless of what the experienced folks do.
The practical lesson? When you see an asset going up and up and up, and everyone around you is excited, and new people who never invested before are suddenly pouring in, that is exactly the pattern that creates bubbles in the lab. Every single time.
Recognizing the pattern won’t make you immune to it. But it might make you a little more careful about chasing prices that have already run far above any reasonable value.
Previous: Neuroeconomics - Your Brain on Money