Past Talks s4: Fall 2008A Field Experiment on Bargaining
Marco Castillo
of Georgia Tech University
and visiting scholar at GMU
Friday, December 5th from 4:00 to 5:30
ABSTRACT
We present field experimental evidence of gender discrimination in a competitive environment. We exploit the existence of a large and widely used competitive market where prices are settled by face-to-face bargaining to test for the existence of differential treatment. To do this, we trained confederate buyers to implement a simple bargaining strategy that consisted of repeating a maximum acceptable price until the offer is accepted or rejected. This protocol forces sellers to reveal their strategies and their willingness to compromise with different buyers. While we find that initial p ent with attempts to screen by the seller, negotiations are more likely to continue with women than men. Also, women obtain better prices and are rejected less frequently. These results are puzzling because sellers are likely to face similar buyers whether they reject a man or a woman. The data reveal no evidence that these differences are due to selection issues or sellers obtaining more enjoyment out of providing services to one gender over the other.
Lost in the Mail: A Field Experiment on Corruption
Ragan Petrie
of Georgia State University
and visiting scholar at GMU
Friday, November 7th from 4:00 to 5:30
Crime, corruption, privatization and development are all hot topics here at Mason. This week, Professor Petrie speaks to all four with an innovative field experiment examining the postal system of Peru. She and her co-authors sent letters from America to Peru, and found that 21% of mail containing cash never reached it's destination, nor did 15% of mail without money. As they write in their introduction:
In many developing countries, recent research shows that corruption in the provision of public services is not only widespread but can create important inefficiencies and inequities (Bertrand, Djankov, Hanna, Mullainathan, 2007; Reinikka and Svensson, 2004). In an environment where the provision of public services is privatized, it is not clear if corruption is lessened or not. We show that in the presence of moral hazard, even private firms providing public services, such as mail delivery, can face a significant level of corruption. Importantly, we show that this has equity consequences. The cost of crime is not equally shared by all citizens — the middle class suffers from corruption the most.
ABSTRACT
Efficient markets require the free and safe movement of goods and services. Corruption in the mail sector, therefore, can hamper the development of commerce and electronic markets in particular. We present the results of a field experiment designed to detect corruption in the privately-run Peruvian mail sector and to measure its differential impact across populations. We subtly, and realistically, manipulate the content and information
available in several pieces of mail sent to households across Lima and detect not only high levels of shirking but of corruption as well. Twenty-one percent of the mail containing money and 15% percent of the mail not containing money never arrived at its destination. Interestingly, we find that middle-income neighborhoods, rather than the rich or poor neighborhoods, are the most likely to suffer from crime. Our study demonstrates two important issues. Privatization has been unable to extricate moral hazard, which
then leads to corruption and results in barriers to market expansion. And, those who are the targets of corruption are not equally distributed across the population.
After the War: An experimental study of post-conflict behavior
Frans van Winden
of University of Amsterdam and CREED
Friday, October 24th from 4:00 to 5:30
reception following
Truland, room 335, Arlington Campus
ABSTRACT
Most models of conflict concentrate on how players allocate resources between productive and fighting efforts. After a conflict, the winner is assumed to take control of all the resources of the loser. In this paper we show experimentally that this simplification misses an important component of a conflict, namely the reaction of the defeated player. We find that, if given the choice to destroy some of their resources, many defeated players prefer to do so rather than let the winner take any of it. Given this behavior, incentives to invest in weapons as opposed to production change and in some cases lower levels of conflict are achieved. Furthermore in settings with repeated interaction, the behavior of players in post-conflict stages can serve as a form of costly communication which enables players to reach a peaceful outcome in the future.
"When the Principal Demands a Corrupt Agent: Neural Capital, Rank Fetishism and Productivity"
Elias L. Khalil
of Monash University, Australia
Friday, October 17th from 4:00 to 5:30
reception following
Truland, room 335, Arlington Campus
Abstract: The dominant view of corruption is based on the principal-agent framework: corruption undermines the interest of the principal. This view cannot explain why corruption, in many cases, is accepted and even demanded by the public, the principal. This paper provides a general theory that provides an answer. It redefines corruption as privileges enjoyed by people of high rank, what is called "rank fetishism." The principal demands people in authority to indulge in privileges to enhance, via heightened neurotransmitters, their own neural capital.
"Trust and Reciprocity in 2-node and 3-node Networks"
Mary Rigdon
University of Michigan
Friday, September 26th from 4:00 to 5:30
reception following
Truland, room 335, Arlington Campus
ABSTRACT
In this paper we focus on the interaction between exogenous network structure and bargaining behavior in a laboratory experiment. Our main question is how competition and cooperation interact in bargaining environments based on networked versions of the investment game. We focus on 3-node networked markets and vary the network structure to model competition upstream -- multiple sellers paired with a monopsonistic buyer -- and competition downstream -- a monopolistic seller paired with multiple buyers. We describe two kinds of models of trust for such networked environments, absolute and relativized models, and use this structure to generate a general hypothesis about these environments: that information crowds in cooperation on the competitive side of the market. The experimental results support this hypothesis.
Lecture Attachment
Wired for Survival: Human Nature, Lawgivers, and other Wifemen
Margaret Polski
Center for the Study of Neuroeconomics,
George Mason University
Friday, September 12th from 4:00 to 5:30
reception following
ABSTRACT
In Wired for Survival: Rational (And Irrational) Choices from the Gas Pump to Terrorism (September, 2008) I argue that we are biologically wired for survival. Our most important political and economic institutions are codified not in law, treaty, government, or culture, but in the neural networks embedded in our bodies and our brains. For centuries we have debated whether growth and prosperity depends upon better people or better institutions. The debate raged anew among the 18th century Anglo-Scots political economists, whose work provides a foundation for the U.S. political economy and scholarship in institutional economics. Yet today when we think about economic organization and regulation, we typically focus on discovering regulatory arrangements that produce technically efficient outcomes without considering the real objective functions of the people who are the object of our efforts. Then we attempt to select the "best" fit, create artificial enhancements or reengineer ourselves, institutions, organizations, and networks to achieve efficiency ideals. North (2005) argues that to create adaptive institutions, our minds must evolve. But the evidence suggests that can't change our minds or our institutions without changing our brains. Successful adaptation requires remapping ourselves, the way that we interact with others, and consequently, the way that we think and make choices. This essay summarizes what we know about the biological basis of choice and develops implications for economic organization and regulation.
Lecture Attachment
Guilt aversion vs. intentional reciprocity in a trust game
Omar Al-Ubaydli
George Mason University
Friday, September 5th
Lecture from 4:00 to 5:30
Party starts at 6:00
Lecture in Truland, room 335, Arlington Campus
Party at Quincy Plaza
3900 N Fairfax Dr, Arlington, VA 22203 US
Please send RSVP for the party to Yoanna Ganeva <yganeva@gmu.edu>
ABSTRACT
If you think that others are expecting you to behave in a certain way, does that make you more or less likely to behave in that way? We estimate this causal effect in a trust game with punishment. We measure what a subject thinks that other subjects expect him to do in three different ways.First, by directly eliciting what a subject thinks that others think that he will do. Second, by first eliciting what his partner in the trust game thinks that he will do and then reporting it to him without the partner's explicit knowledge. Third, by eliciting what neutral observers think that he will do, telling him this, and then asking him what he thinks others think that he will do, using the observers' expectations as an instrument. Comparing the first and third allows us to check for endogeneity of elicited expectations. Comparing the second and the third allows us to check for a deception effect. We find that in the context of the trust game with punishment, what you think others expect you to do has no effect on how you behave. We find an endogeneity effect but no deception effect. We interpret the predictions in light of the guilt-aversion and intentional-reciprocity models.
 |
|
