Experimental economics tests economic theories the way an experimental scientist tests a hypothesis, by running controlled experiments with real, meaningful financial stakes, rather than relying purely on observing existing, messy real-world markets or building purely theoretical models and reasoning about them abstractly. Because a real market bundles together dozens of interacting factors at once, an experimental economist can instead design a controlled setting that isolates one specific variable at a time, in a way genuinely uncontrolled real-world market data almost never allows.
A controlled experiment can isolate one variable in a way real markets simply can't
Real-world market data reflects the combined effect of countless factors happening simultaneously, prices, information, social context, individual psychology, all tangled together in ways that make it genuinely difficult to isolate exactly what's driving a specific observed behaviour. An economic experiment instead sets up a controlled environment where participants make real decisions, often for genuine financial payoffs, under conditions the researcher has deliberately designed so that only one specific variable of interest actually changes between different groups, letting the resulting behavioural difference be attributed to that one variable with real confidence.
Real financial stakes matter because unpaid hypothetical choices don't reliably predict real ones
Experimental economics insists on paying participants real money based on their actual decisions specifically because research has shown that people's choices in hypothetical, no-stakes scenarios don't reliably predict how they'd actually behave when genuine money is on the line. That insistence on real stakes is exactly what distinguishes rigorous experimental economics from simply surveying people about what they say they'd do, and it's precisely this combination, genuine financial incentives inside a carefully controlled setting, that's let experimental economists uncover real, systematic and reproducible patterns in how people actually make economic decisions.
What we're still unsure about
That controlled experiments with real financial stakes can isolate variables and reveal genuine, reproducible behavioural patterns is well established, confirmed by decades of experimental economics research that's reshaped how economists understand real decision-making. What's more genuinely an ongoing methodological question is exactly how well behaviour observed in a controlled laboratory setting, often involving relatively small stakes and university students as participants, actually generalises to real-world economic decisions made by a broader population facing genuinely larger stakes, and experimental economists continue to actively study and debate exactly how far laboratory findings can be trusted to carry over into real markets outside the experimental setting.
This sits inside Experimental Economics Methods, one of seven topics in Behavioral Economics, one of five domains in Economics, one of seventeen subjects the app can quiz you on.