V55.0602.01 Yaw Nyarko
TR 09:55-11:10 New York University
Location: Main 806 269 Mercer St. #711B
Office hrs: Thursday 3:30-4:30PM (212)-998-8928
(or by appointment) yaw.nyarko@nyu.edu
Societies and the Social Sciences: An Economic Perspective
Summary of Class
Societies are made up of individuals. Social outcomes are the result of individual decisions. To determine social outcomes therefore, it is important to study how these individual decisions are made. This class will study social science through the prism of individual decision making. The class will take a decidedly neo-classical economics perspective on this. In particular we shall model individuals as making rational decisions given their understanding of the world around them and given the constraints, financial and informational, that they face.
Increasingly, this economic perspective is being used in popular discussions on social questions. This includes the current topical issues on welfare payments and welfare dependency; race, intelligence and education policy (as popularized by Hernstein and Murray's "The Bell Curve"); minimum wage laws; social security; crime and punishment. All these issues are now being addressed using the tools of neo-classical economics and individual decision making. Although these topics will not be directly addressed in the class, techniques will be studied which can be used in understanding the issues involved.
This is therefore a class in decision theory, that is, how individuals make decisions under various situations. In each case we shall suppose that the individual chooses actions to maximize some criterion or function. Ronald Reagan once said "an economist is the only professional who sees something working in practice and then seriously wonders if it works in theory(1)." (I am no Republican, but Reagan did have some good lines!) The philosopher, Sir Karl Popper once said that the aim of the social sciences should be to understand "the logic of the situation(2)." The goal of this class will be to study the "theory" and the "logic of the situation" in many social science problems.
I have divided the class into four parts. The first two parts will cover single agent decision theory. These two parts involve, respectively, decision-making under uncertainty and decision making across time. The last two parts will be on multi-agent decision, where there is a non-trivial interaction between many individuals. The material on multi-agent decision theory will be divided into what is called normal form games where individuals move simultaneously and extensive form games where agents move sequentially.
As mentioned above, the first part of the class is decision-making under uncertainty using the expected utility criterion. This concept will be illustrated in many examples. We shall use this criterion to study what is called risk aversion, which is the claim that individuals do not like risk. It is this risk aversion which causes individuals to buy commercial insurance to insure against house fires, earthquakes, theft, medical accidents, etc. We shall study examples of such insurance. We shall also use the expected utility criterion to illustrate the "optimal" allocation of one's savings or wealth among a number of assets or securities. We shall "solve" examples of such portfolio allocation decisions.
The section on decision making across time will begin by a study of discounting. This is based on the simple premise that a dollar today is worth more than a dollar next year, since by obtaining the dollar today and putting this in the bank one would obtain a dollar next year in addition to the interest on that dollar. This concept is used in "pricing" returns which will be obtained over a number of years. This "price" will be called the net present value. In this section we shall also study problems where a decision has to be made in each of several periods but where a decision in one period constrains the set of decisions that can be made in subsequent periods. The technique of solving such problems will be something called backward induction. This technique will be used in many examples.
The third part of the class is on Normal form games. Normal form games are those where there are many agents each choosing actions simultaneously. We shall use such models to study the models of firm's pricing behavior in economic markets. In such models, one paradoxical thing that often happens is that in the course of each agent trying to maximize their individual profits, they choose actions which are bad for them collectively. In particular if each chooses an different action it is possible for each and every one of them to be better off. The study of such issues falls under the topic of "Pareto Optimality." The fourth section studies extensive form games which are those where there are many agents but where the agents do not necessarily move at the same time, but may move sequentially, one after the other. These problems involve each individual trying to guess what the others will do when it is their turn to choose an action. Such models also involve the use of threats of the form "if you don't choose action x which I like, I will choose action y which I know you hate." We shall study threats and classify them into "credible" and "incredible" threats.
The best way to study for this class is to work through as many of the exercises as possible. Reading the material of the class will provide one with the basic principles. However, it is only by doing the exercises that one really obtains a deep understanding of these issues. You do not have to purchase any of the recommended books. Many of you will find that coming to class and taking notes will be enough (in addition to the problem sets), and may find the textbooks not too helpful.
V55.0602.01 Yaw Nyarko
TR 09:55-11:10 New York University
Location: Main 806 269 Mercer St. #711B
Office hrs: Thursday 3:30-4:30PM (212)-998-8928
(or by appointment) yaw.nyarko@nyu.edu
Societies and the Social Sciences: An Economic Perspective
Course Outline
I. Decision Making Under Uncertainty
1. Probability and Expectation
a. Frequentist and Subjectivist Definitions of a Probability; Lotteries
b. The Expected Payoff Maximization Principle. Some simple examples
2. Expected Utility
a. Expected Utility Versus Expected Value (or Expectation).
b. Some Examples
c. Risk Aversion
d. The St. Petersberg Paradox.
Part B: Some Applications
3. Commercial Insurance Markets
4. Stock Markets and Portfolio balance.
5. State Lotteries and Gambling Institutions
6. The Dutch Book
7. Informational Issues
8. Violations of Expected Utility Maximization?
II. Inter-temporal Decision Making
1. Discounting and Net Present Values
a. Discounting future incomes; Net Present Value (NPV).
b. Pricing Some Simple Financial Assets
c. Infinite Horizon discounting
d. Intertemporal Decision Making Via NPV
2. Ponzi Schemes, Pyramids and speculative bubbles
III. Multi-Agent Decision-Making
1. Normal Form Games
a. Definition of a normal form game
b. Examples of 2x2 Matrix Games
c. Best Responses to beliefs/actions
d. Dominant Strategies
2. Nash Equilibrium
a. Definition of a Nash Equilibrium
b. Nash Equilibria in the 2x2 Matrix Games
c. Other Examples: cake-sharing and Hotelling
d. The Nash Equilibria of Some nxm Matrix Games
3. Auctions
4. VOTING
5. Mixed or Randomized Strategies
Definition of mixed Strategies.
Examples
Monitoring and Auditing
6. Pareto Optimality
a. The Definition
b. Some Examples
c. The Tragedy of the Commons
7. The LEMONS problem and the used car market.
8. The Free-Rider Problem and the Groves-Clarke et. al. mechanism
IV. Sequential Decision-Making
1. Definition of an Extensive Form Game
2. Sub-game Perfection, Incredible Threats, Backward Induction
a. The Definitions
b. Examples: Battle of the Sexes. IN-CLASS EXPTS:
c. Other Examples?
3. Bargaining
a. Finitely Repeated Sequential-Move Bargaining Problem; examples.
b. Infinitely Repeated Sequential-Move Bargaining Problem; examples.
4. Repeated Games
a. The Main Concepts - supergames; examples of strategies; tit-for-tat etc.
b. Finitely Repeated Games
c. Nash Equilibria vrs Sub-game Perfect Nash Equilibria
d. Trigger Strategies and the role of the discount factor.
Class Textbooks (on reserve at Bobst Library)
There is no formal Class textbook. Some students, however, may find the following books useful (most current edition of each book): Copies are available on reserve at Bobst Library.
1. Coleman, A.:' Game Theory and its Applications," Butterworth-Heineman Ltd, Oxford.
2. Dixit, A. And B. Nalebuff, "Thinking Strategically," W.W. Norton, New York.
3. Gardner, R. "Games for Business and Economics," John Wiley and Sons.
4. Ken Binmore (1992): "Fun and Games: A Text on Game Theory," D.C. Heath, Lexington, Mass.
5. H. Scott Bierman and Luis Fernandez (1993):"Game Theory with Economic Applications" Addison-Wesley, Inc.
6. Baird, D. R. Gertner and R. Picker (1994):"Game Theory and the Law," Harvard University Press, Cambridge, MA.
Grades
The requirements for a grade in the class are as follows: There will be one mid-term exam on October 22nd, 1998. This will be worth approximately 40% of the class grade. There will also be final exam worth approximately 60% of the class grade. There are no make-up exams given. The mid-term exam will be on Thursday October 22nd, 1998 at the usual place and time of the class. The final exam will be in the middle of December - the exact time and place will be announced by the university. There will be no other exams given; there will be no exceptions. There will also be problem sets given out in class. Students will also be required to take part in experiments which will either be in class or in the Economics Department Computer Laboratory at 269 Mercer St. Rm 713. Performance in the problem sets and computer experiments will be used in determining your letter grade for the class in addition to the mid-terms and final, especially when your grade falls in the boundary of two letter grades. This of course means that the problem sets may be very important in determining your final class grade.
Class and Recitation Sections
Lecture V55.0602.001 Tue & Thr 09:55-11:10 Main 806
V55.0602.002 Tue: 2:50-4:05 pm Wave 470
V55.0602.002 Tue: 4:20-5:35pm Main 412
V55.0602.002 Wed: 8:30-9:45am Main 801
V55.0602.002 Wed: 2:50-4:05 pm Shim 125
V55.0602.002 Wed: 4:20-5:35 pm Wave 668
Morse Academic Plan Tel #: (212)-998-8119
1. Quoted by McMillan (1992) p. 9.
2. Quoted by McMillan (1992) p. 9.