Experimental Economics

Amy Nagler

At 3 o’clock on a Wednesday afternoon, eight student participants take their seats in an economics laboratory on the second floor of the College of Agriculture and Natural Resources building. The Department of Agricultural and Applied Economics may seem like an unlikely host for an experimental economics facility, although the data generated here are used in research covering a surprising diversity of topics.

In order to motivate participants to reveal realistic market behavior as they trade an imaginary commodity over the computer network, they will be paid in cash at the end of the session based on their market earnings. Earnings accumulate during a sequence of trading periods and payouts can range from around $20 to over $100, depending on the experimental treatment, the assigned role, and the bargaining skill of the participant.

The current session will collect data from a series of treatments designed to investigate impacts of agricultural subsidy payments on market behavior. Specifically, we are interested in how subsidies are transferred between buyers and sellers via price negotiations as they trade in markets.

Other research conducted in this laboratory has been applied to better understand the role of alternative trading institutions in price discovery: How are prices influenced by whether trades are conducted in an auction rather than through direct negotiation between two individuals? Family economic behavior including links between childhood obesity and parenting, allocations of land for habitat conservation, and market concentration in cattle markets are just a few examples of the diverse subject matter explored here. The lab is also used as a teaching tool, giving students a hands-on understanding of economic theory and behavior.

A Do-it-yourself Behavioral Economics Experiment

Try this experiment with two people you know. Give ten one dollar bills to the first person. Instruct them that they may give a portion of this money to the second person–with one stipulation: If the second person is not happy with the amount passed on to them, neither player gets to keep any of the money and the entire $10 must be handed back to you.

What you will find is that neither player is apt to model classic economic theory. If both players were purely rational and self-interested, the smallest portion possible would always be handed over and this tiniest amount would always be accepted. What actually happens? The person you just endowed with ten free dollars is likely to hand over around $4 which the recipient will happily keep. We humans are self-interested, yes, but we are also generous.

Now watch what happens if the first person is a bit less generous and only hands over a dollar. The recipient will likely reject a perfectly free dollar to leave both players with nothing. Why? As economic agents humans act in ways that are self-interested, rational, generous, and we have expectations of fairness and punishment that also drive how we behave in different situations.

The Experimental Economics Toolbox

In our experimental market, a subsidy payment given to the buyer changes expectations and impacts how prices are negotiated. Behavioral experiments such as this “ultimatum game” are another tool economists at UW use to further understand how and why social and cultural expectations influence economic decision making.