ECON6037 Experimental Economics
Introduction to Experiments
February 19, 2024
Human Decision Makers
- Economics is built around assumptions about human behavior.
- These assumptions allow the construction of a well-defined utility function.
- For choice under risk, a key aspect of applied economics, there’s an additional assumption on cardinal utility.
- Further structure is added to model behaviour over time.
We need experiments to * Explore the distribution of behaviors and traits within a population. * Test the validity of assumptions about human behavior.
Economists Conception of Human Behavior
Following Marshall lecture by Rabin (2002):
\[\max_{x_i^t\in X_i} \sum_{t=0}^{\infty}\delta^t\sum_{s_t\in S_t}p(s_t)U(x_i^t|s_t)\]
- \(X_i\) is set of lifetime strategies, \(S_t\) set of state spaces.
- \(p(s_t)\) are rational beliefs, \(\delta \in (0,1)\) time-consistent discount factor.
- \(u(\cdot,s,t)\) true utility at time \(t\) is state \(s\).
Behavioral Economics in Historical Perspective
“We suffer more… when we fall from a better to a worse situation, than when we ever enjoy when we rise from a worse to a better.” - Smith (1790)
- Bentham, founder of modern utility theory, wrote extensively on the psychological foundations of utility.
- Jevons, Walras, and Menger all wrote about the psychological foundations of utility.
- Edgeworth (of box’s fame): Theory of Mathematical Psychics included a simple model utility of a person depending on another person’s payoff.
- Neo-classical revolution ended these approaches.
How Theory Can Fail
Non-standard preferences.
Non-standard beliefs.
Non-standard decision making.
Non-standard Preferences
Present-biased preferences: time inconsistency (\(\beta-\delta\) discounting).
Reference dependence: \(U(x_i|r,s)\) with \(r\) as a reference point.
Social preferences \(U(x_i,x_{-i}|s)\) where \(x_{-i}\) is the allocation of others.
Non-standard Beliefs
\[\tilde{p}(s) \neq p(s)\]
Overconfidence: wrong \(E(p)\) or wrong \(Var (p)\).
Law of small numbers: Wrong forecast of \(p(s_{t+1}|s_t)\).
Projection bias: wrong forecast utility \(\hat{u}(\cdot,s)\).
Experience effects: excessive updating of \(p(s_t|s_{t-1})\).
Non-standard Decision Making
Limited attention: consideration set a proper subset of the choice set.
Mental accounting.
Persuasion.
Emotions and happiness.
Psychology and economics
Uniting Psychology and Economics
- Integration of psychological concepts and theories into economic studies.
- Application of incentives in economic models, incorporating psychological insights.
- This interdisciplinary approach allows for diverse methodologies.
Areas of Psychology in Economics
- Social Psychology: Emotions, groups, discrimination.
- Cognitive Psychology: Kahneman & Tversky’s program.
- Personality Psychology: The BIG 5 model.
- Developmental Psychology: Child development.
- Comparative Psychology: Cultural differences and their economic impacts.
Approaching Psychology for Economists
Recommended Readings: - Social Psychology: “The Person and the Situations” by Ross and Nisbett, 2011. - Cognitive Psychology: “Judgement Under Uncertainty: Heuristics and Biases” by Kahnemann, Slovic, Tversky, 1982. - Key Psychology Journals: - Journal of Personality and Social Psychology. - Psychological Science. - Psychological Bulletin. - Psychologists seek new effects and frameworks for understanding.
Methodologies in Economic Psychology
- Experiments: Both in laboratory and field settings.
- Natural Experiments: Observational studies in real-world settings.
- Correlation Studies: Identifying relationships between variables.
- Structural Identification: Understanding underlying mechanisms.
Fields of Application in Economics and Psychology
- Public Finance: Present-bias, limited attention, social preferences.
- Development Economics: Present-bias, social preferences, risk preferences.
- Asset Pricing: Overconfidence, limited attention, market reactions.
- Corporate Finance: Overconfidence, reference dependence, media influence.
- Labour Economics: Reference dependence, social preferences, money illusion.
- Health Economics: Present-bias, menu choices, insurance decisions.
- Industrial Organisation: Present-bias, pricing strategies.
- Marketing: Menu effects, product placement.
- Environmental Economics: Social comparison, framing effects.
- Law and Economics: Behavioural biases, emotional factors in litigation.
Objectives of Experiments
Why Experiment in Economics?
Establish empirical regularities for new theories.
Test institutions and environments.
Provide policy advice and conduct wind-tunnel experiments.
Elicit preferences regarding goods, risk, fairness, and time.
Use in teaching to illustrate economic concepts.
Testing Theories in Economics
- Economic theory forms the basis for experimental design.
- Conditions of the theory are implemented (preference assumptions, technology, institutions).
- Predictions are compared with experimental outcomes.
- Analysis of why a theory succeeds or fails.
- Designing control treatments to allow causal inferences.
Joint Testing of Assumptions: - An experiment tests all underlying assumptions, including induced value theory.
Basis for New Theories
- Empirical regularities guide theory development (e.g., theories of fairness).
- Experiments can help select relevant equilibria in cases of multiple equilibria.
- Opportunity to advance beyond current theoretical limits (e.g., continuous double auction).
- Response to the failure of the homo oeconomicus concept: Development of new theories on bounded rationality, learning, fairness, etc.
Policy and wind-tunnel experiments
Evaluate Policy Proposals
- Does the reduction of entry barriers increase aggregate welfare?
- Which auctions generate the higher revenue? (e.g., in arts auctions or UMTS license auctions)
- Do emission permits allow efficient pollution control?
- How should airport slots be allocated?
- Workfare and welfare incentives: What are the distributional and welfare consequences of incentive compatible mechanisms?
- How does a privatized electricity industry with small numbers of suppliers and intermediate traders work?
The elicitation of preferences
Preference Elicitation
- How much should the government spend on avoiding traffic injuries?
- How much should be spend on the conservation of nature?
- In general: How should nonmarketable commodities be produced? How should they be distributed?
- Does relative comparison affect people’s preferences, i.e. are there important consumption externalities? Answer is important for tax policy, growth policy, etc.
A nonarbitrary and nonpaternalistic answer to these questions depends crucially on one’s view how much people value the above goods. Yet, measuring people’s values requires a theory of individual preferences and knowledge about the strength of particular “motives (preferences). This requires the testing of individual choice theories and instruments for the elicitation of preferences.
Teaching experiments
Better understanding of economic phenomena
- Markets
- Bargaining
- Social dilemma
- Own experience important
- “Failing” to behave rationally or optimally: Anomalous behavior
- Starting to think about economic questions differently
Origins of Experimental Science
Experiments in science…
- The old Greeks/Chinese: Observation of naturally occurring processes not experiments
- Experiments regarded as disreputable
- Arab ixtibar translated to Latin experior in 12th century Al Hassen and Ibn al Hadam Arab physicist
- First experimental tradition established in Physics (around 1600)
- i.e. Galileo inspired by observation varied pendula and ramps systematically
- Chemistry around 1800: Lab techniques to test molecular theory
- Biology: Mendel and Pasteur in 19th century
- Psychology: Middle of 19th century Wundt, Fechner, Helmholtz
… and Economics?
- Small theory on the emergence of experiments in disciplines
- Theory has to be sufficiently mature to develop testable predictions
- Need to emerge pioneers that develop laboratory techniques
Economics as Experimental Science
- Way of generating controlled data as opposed to using happenstance data as other empirical projects
- Economics is becoming an experimental science like physics and biology.
- Like theory, running experiments is an established method to explain and or describe economic activity.
Duhem-Quine Problem
- All tests of a theory require various auxiliary hypotheses that are necessary in order to interpret the observation as a test of the theory.
- Consequently, all tests of a theory are actually joint tests of the theory conditional on the auxiliary hypotheses. (Duhem and Quine, 1998)
The Duhem-Quine thesis: One can always rescue a theory from an anomalous observation by ex post hoc recourse to imaginative and persuasive auxiliary hypotheses. Conversely, every observational victory for a theory can be questioned by a suitable revision of the background knowledge in which the theory is embedded.
What should experimental economists do?
- … when theory works well,
- construct “stress tests”
- try to uncover edges of validity
- … when theory works poorly
- re-examine instructions
- increase the experience level of subjects
- try increased payoffs
- explore all sources of error in the attempt to find the limits of the falsifying conditions
These two approaches complement each other in helping to better understand to what actually determines human behavior and helps to improve our theories about it
Bed-time Reading
- Thinking fast and slow Kahneman 2011
- Good summary of Kahnemann and Tversky’s research program over last 40 years
- Few economic examples
- Scarcity: Why Having Too Little Means So Much Mullainathan and Shafir 2013
- Good summary of behavioral economics
- Lots of economic examples