Professor: Bill Harbaugh
538 PLC, 346-1244
Office hours: 2-4 Wed, or by appointment.
link to course notes and assignments.
Important:
This course is a mix of behavioral economics and experimental economics. I’ll start by describing behavioral economics.
By this point in your academic career you know that economics is the study of how people make choices in a world with constraints. Many of the economics courses you have taken probably emphasized how economists believe people should behave in such a setting. For example, you learn things like how to use indifference curves to find the allocation that maximizes utility, why we should ignore sunk costs, why we shouldn't cooperate in prisoner's dilemmas, and so on. Many of these "normative" statements about what people should do are based on some very specific assumptions about behavior, most notably that people make choices with the goal of maximizing their own utility.
Behavioral economics is different, in that is studies how people actually behave in economic settings, not how we think they should behave. Behavioral economists are particularly interested in situations where what people actually do differs from what the rational choice model says they should do. This is mostly because behavioral economists believe that by studying these anomalies we will ultimately be able to build better models of economic behavior. On the other, some of them also just seem to enjoy ripping into the established wisdom.
For example, behavioral economists have developed “hyperbolic discounting” models of choice over time. They argue that these models are better at describing actual behavior in areas such as retirement savings accumulation and addiction than the standard constant discount rate models are. Another example would be fairness. People seem to want to be nice to people who are nice to them, and mean to people who are mean. Behavioral models that incorporate these preferences lead to some interesting predictions about behavior and equilibria. They also have important implications for how we should think about subjects that are analyzed in traditional economics courses, such as wages and unemployment.
The other part of this course is experimental economics. Experimental economics is a methodology. Instead of testing theories using the sort of happenstance data that most empirical economists use, experimentalists get their data from experiments under controlled conditions. We use cash payoff so that people’s decisions have real consequences, just like decisions outside the lab do. I like to say that experimental economists are producers of data, not just consumers of it.
Note that experimental economics is not the same as behavioral economics. Many experimentalists are not that interested in overturning the theoretical foundations of rational choice theory. Instead, they may find that the theory doesn't make specific predictions about the situation they are trying to deal with, and they are trying to figure out which equibrium is most likely to happen. Some economists use experiments to predict how people will behave in radio spectrum auctions, for example. Economists should have used experiments to study the electricity auction mechanism before it was set up in California, for another. Other experimental economists believe that the anomalies behavioral economists claim to have identified are not general enough to bother with, perhaps because they believe they tend to go away with learning, or in situations where the stakes are higher.
This course is designed to give you a background in both behavioral and experimental economics. We will study behavior in about 10 different experiments. The experiments will be conducted on you and they will have real money payoffs. While this course is mainly about what experiments reveal about human behavior, you will also learn something about the design and analysis of economic experiments.
Bill's Economic Incentive Scheme:
Economics is about actual, not hypothetical behavior and for this reason economic experiments involve real, not hypothetical consequences. Economists make sure people make careful decisions in experiments by making sure they have incentives to do so. In this class we will do this with the usual incentive - money. The money comes from you. There is no textbook for this class – all readings are available on the internet, or through handouts I will give you. Instead, there is a $75 lab fee, payable via cash, a check made out to me, Bill Harbaugh, or paypal sent to my account, wtharbaugh@gmail.com. The payment is due Wednesday of the second week of class. This lab fee will be used to provide the payments for the experiments that we will do.
I will deposit this money in a savings account, adding (at least) $100 of my own funds to “sweeten the pot.” I will keep a running balance of your earnings in the experiments. I will pay you your earnings out of this account on or before the final exam. Any amount of the account balance that is not paid out in the experiments will be divided up among students evenly. If you drop the course, I will prorate and refund your lab fee based on the proportion of class meetings that you were officially enrolled for.
Depending on your decisions, and the decisions of others in the experiments, it will be possible for you to earn substantially more than the $75 lab fee in these experiments. Of course, it is also possible for you to earn nothing – particularly if you miss some classes.
I will give you a receipt for your payment, and a copy of these rules to sign on Wed of the second week of class.
Homework:
There will be 4 homework assignments. They will be assigned about a week before they are due. The point of homework is to give you incentives to learn the material and some feedback about your level of understanding as we cover new material. Given this, it can be a very good thing to do homework in a group. You can learn a lot from explaining things to other people, and from hearing other perspectives on a problem, challenging them, and having your own answers challenged. I still learn this way, from the questions I get asked while teaching. However, if you are just relying on someone else to figure out the problems, and then copying out their answers, you are not engaging in this give and take process, and you're probably not learning much. I am not particularly worried about this copying from the point of view of evaluating your knowledge. This will show up on the exams. However, I do want to make sure that the homework serves the purpose of helping you learn the material. When I receive copies of the homework from different people, but with identical answers, I worry that people are relying too much on others, and not working through the problems themselves. Therefore, if you work in groups, once you have worked out an answer stop working together on the problem. Write your interpretation of the answer down on your homework. At that point, don't compare answers anymore. This will make sure that the answer is your own.
Exams:
There will be a midterm on Mon, October 23, in class. The final is 10:15 Monday of finals Week.
Grading:
Homework (4)
25 points each
Midterm
100 points
Final
100 points
Total
300 points
Masters Degree Students:
If you are enrolled in the 528 version of this class you will need to write a 15 page research paper to get credit for this class. You are responsible for meeting with me within the first two weeks of class to go over the requirements for this paper. This paper will count for 30% of your grade, and the weight on homework and tests will be reduced accordingly.
Course Content:
This course is built around the subjects listed below. We will spend about 2 lectures on each. Check this online syllabus frequently, for updates!
Professors usually want you to do the readings before class to avoid blank stares from their students. Students usually prefer to wait until after class, when they know which parts to skip. I’d prefer that you wait, so your decisions in the experiments won’t be influenced by what you read. But you might prefer to read ahead – maybe it will help you earn more money! These papers range from easy to very, very difficult. If you find parts tough going, you are not alone. Do your best and I will give more explanation in class. All of the articles are available on-line. You should be able to link to these directly, so long as you are on campus. If you are off campus, you may need a VPN.