"Baby, I Could Walk You All The Way Home"*

Before I review Engelmann and Strobel's “Inequality Aversion, Efficiency, and Maximin Preferences in Simple Distribution Experiment,” I think Bolton and Ockenfels bring up the essential question, “...what it is that experimental economics can hope to accomplish.(?)”

Bolton and Ockenfels like most economists consider themselves scientists. Experiments must be explained. Researchers must find causations. Economists must write theories. They want to “present a broader, more valuable map of economic behavior.”

What for? Why? Do economists really think they can predict behavior? Do they secretly hope that with these new theories they will be able to usher in a new wave of social planning with themselves doing the planning? What exactly is “economic behavior?” Until we attempt to answer these questions, economists cannot progress. I am not saying Bolton and Ockenfels are wrong. But they do not try and make their case. They do not tell the reader what a “broader, more valuable map of economic behavior” should be or what it will do. We are ignoring the fundamental philosophical question that forms any discipline. What is an economist's purpose?

I cannot answer this question, but I do know economic experiments yield individual decision information from a controlled setting that suggest how people will make other decisions. The ultimatum, bargaining, and distribution games are not parallel with real life. Subjects will never be told to split $10 in their life like they are told in the experiment. They will make similar decisions with their spouses, friends, and children but never will it be as blatant as in the experiment.

But just because we cannot ex ante predict behavior does not mean the
experiment has no value. That logic is a stupid way to examine the problem. The experimenter created this experimental institution, and certain results occurred. If we changed the experimental institution, the results would change. The main lesson from experimental economics is institutions matter. You cannot separate the decision from its context.

Experimental economics and economics will not progress until researchers decide what economic theory should do. Classical theorists ignore reality, but their theories still hold didactic and normative value. Nash equilibrium makes a young student look at a problem in a different way. As Engelmann and Strobel suggest new theories based on experimental results do not predict any better than older theories. But the new theories and the experiments that birthed them challenge students to think.

Economics has to decide what it wants to do. This competition between different economists pursuing different objectives can lead to progress. But researchers trying to match experiments and theories will impede economics. Experimentalists need to design new games and institutions that allow new behavioral hypotheses to be tested. Conducting the same simple experiments in an attempt to test new theory is a lazy exercise that will not push the boundaries of economics but instead stifle thought. We need new experiments and different contexts to see how people make decisions given different institutions. We need our experiments to be more parallel.

Economics must recognize its normative basis. Positive economists cannot refute this normative basis. No matter how many free riding experiments one does, Milton Friedman (or myself) cannot be convinced that taxes are a good thing. Economics is not mature enough for experiments to prove the validity or invalidity of economic theory, because our theory is still normative. We do not know enough about the brain, genetics, and environmental effects to “make a map of economic behavior.”

And it is foolish to pretend that economists can or should do this. Since Adam Smith the best economists have mixed philosophy, a loose interpretation of the scientific method, and a general desire to question their surroundings to create important works. Modern economics has forgotten philosophy, and it rarely questions reality but emphasizes what other economists have done. Equity, efficiency, and selfishness are normative issues that require debate in a philosophical forum. Economists cannot continue to argue over these issues in a positive forum without stagnating the profession.

Engelmann and Strobel's “Inequality Aversion, and Maxmin Preferences in Simple Distribution Experiments” has many interesting results. Their main conclusion is people's decisions cannot be predicted by a number of new and old theories. The game matters.

Distribution games entail an individual deciding how income is distributed amongst two other players. The decision maker's payoff stays the same in all three scenarios, but his fellow players' values change. The decision maker basically decides how equal all three players' payoffs are. He is implicitly trading off between equality and aggregate wealth for all three players.

Engelmann and Strobel vary the values to form the taxation, envy, and rich and poor games. Their results are mixed. The classic economic arguments of efficency, maxmin, and selfishness, probably does the best job of predicting decisions. People want to get as much money out of the researcher as possible, but it clearly does not explain all of the decisions. Other theories do better in certain games, but there are no clear predictors even for these simple games.

And their conclusions are what experimental economics can hope to accomplish. People differ. People care about different things. People are absolutely crazy. For some reason and because of market experiments, I find comfort in their craziness.

*The Boss