A topic in the Open Knowledge Graph — a free, open map of 15,290 topics and the order to learn them in.
Experimental economics tests economic theories through controlled laboratory experiments with real monetary incentives. Its methodological principles — first articulated by Vernon Smith — include induced value theory (monetary payments control the utility function), saliency (payoffs must be large enough to matter), dominance (reward structure dominates other motivations), and no deception (unlike psychology experiments, economists typically never deceive participants, preserving the credibility of the experimental environment). These methods enable causal identification of how institutions, incentives, and information structures affect behavior. The field has produced fundamental insights about market efficiency, auction design, public goods provision, and bargaining, and earned Smith the 2002 Nobel Prize alongside Kahneman.
Experimental economics is the methodological backbone of behavioral economics — it provides the controlled environments in which theoretical predictions of both standard and behavioral models are tested against actual human behavior. Without the ability to create real economic situations in the laboratory, behavioral economics would be limited to observational data and thought experiments. Understanding the methodological principles is essential because they determine what can and cannot be concluded from experimental evidence.
The foundational methodological framework was articulated by Vernon Smith in the 1960s-80s. Induced value theory states that by paying participants real money based on their decisions, the experimenter can control the relevant utility function. If the experimenter pays you more when you choose A over B, and the payments are large enough to matter (saliency) and dominate other considerations (dominance), then your choice between A and B reflects the experimental incentive structure rather than unobserved personal preferences. This principle allows economists to test predictions about how rational agents should behave under specific incentive structures, because the experimenter knows the incentive structure with precision.
The no-deception norm is one of the sharpest methodological differences between economics and psychology. In a typical psychology experiment, participants might be given false feedback about their performance, told a cover story about the experiment's purpose, or presented with a confederate posing as another participant. Economics experiments reject all of these practices. The reasoning is both ethical and practical: if participants believe deception is possible, they will not take the experimental instructions at face value, and the experimenter loses control over the perceived incentive structure. An experimental economics lab builds its reputation on the commitment that "what we tell you is true," and this reputation is a shared resource that no individual experimenter should deplete.
Experimental designs in economics range from simple individual decision tasks (lottery choices for testing risk preferences, intertemporal choices for testing discounting) to complex market institutions (double auctions, posted-offer markets, sealed-bid auctions). Smith's early market experiments produced one of the field's most striking findings: simple double-auction markets converge to competitive equilibrium rapidly, even with very few traders who have no knowledge of economic theory. This result supported the efficiency of market institutions while simultaneously showing that individual rationality is not required for market-level efficiency — a finding that complicates the behavioral economics narrative.
The limitations of laboratory experiments have driven the field toward complementary methods. Field experiments — randomized controlled trials conducted in real economic settings — address external validity concerns by testing behavioral interventions in their intended context. Natural experiments exploit random or quasi-random variation in real-world policies to estimate causal effects. Online experiments expand participant diversity beyond the university student populations that dominate lab research. The most convincing evidence in behavioral economics comes from triangulation across methods: a phenomenon documented in the lab, replicated in the field, and consistent with natural experimental evidence carries much more weight than any single study.