"Field Experiments in Economics" – 24th Middlebury Economics Conference April 26-27, 2003
Participants: (in bold) / Abstracts
Anabela Botelho,Glenn W. Harrison (U of South Carolina), Marc A. HirschandElisabet E. Rutstrom (U of South Carolina)-
"Bargaining Behavior, Demographics and Nationality: What Can the Experimental Evidence Show?"
Bargaining Behavior appears to vary across nations. What drives these apparent differences? We reconsider the evidence provided by previous experiments, and undertake some new experiments that expand the controls for demographics. We show that inferences about country effects are sensitive to the way in which the data are analyzed and the controls that are incorporated. From a methodological perspective, inferences about country effects cannot be made independently of inferences about demographic effects unless one assumes a priori that there is no interaction between the two. Given the surge of interest in cross-country comparisons of behavior in economics experiments, the methodological subtleties of indentifying country effects need to be considered.
Glenn Harrison (U of South Carolina) andJohn List (U of Maryland) -
"What Constitutes a Field Experiment in Economics?"
Experimental economists are leaving the reservation. They are recruiting subjects in the field rather than in the classroom, using field goods rather than induced valuations, and using field context rather than abstract terminology in instructions. We argue that there is something methodological fundamental behind this trend. Field experiments differ from laboratory experiments in many ways. Although it is tempting to view field experiments as simply less controlled variants of laboratory experiments, we argue that this would be a serious mis-characterization. What passes for "control" in laboratory experiments might in fact be precisely the opposite if it is artificial to the subject or context of the task. We propose five factors that can be used to determine the field context of an experiment: the nature of the subject pool, the nature of the commodity, the nature of the task or trading rules applied, the nature of the stakes, and the nature of the information that the subjects bring to the task.
Ernst Fehr (University of Zurich and CEPR) and Lorenz Goette(University of Zurich and University of California at Berkeley) -
"Do Workers Work More if Wages Are High? Evidence from a Randomized Field Experiment"
The canonical model of life-cycle labor supply predicts a positive response of labor supplied to transitory wage changes. We tested this prediction by conducting a randomized field experiment with bicycle messengers. In contrast to previous studies we can observe in which way working hours as well as effort respond to a wage increase and we have full control regarding the workers' anticipation of the wage increase.
We find that workers increase monthly working time and decrease their daily effort but since the working time effect dominates the effort effect overall labor supply increases. This provides evidence in favor of intertemporal substitution with regard to total labor supply. However, the decrease in daily effort contradicts the canonical model of intertemporal labor supply with time separable preferences, since the wage in our experiment directly rewarded effort. We show that a simple model of loss averse, reference dependent, preferences can account for both the increase in working time and the decrease in daily effort. Moreover, we elicit independent individual measures of loss aversion and show that workers who are more prone to loss aversion are more likely to reduce effort in response to higher wages. Our model and our results also reconcile the seemingly contradictory evidence reported in previous studies (Camerer et al. 1997, Oettinger 1999) of high frequency labor supply.
Eric Bettinger (Case Western Reserve University and NBER) and Robert Slonim(Case Western Reserve University) -
"The Effect of Educational Vouchers on Academic and Non-Academic Outcomes: Experimental Evidence from a Randomized Natural Experiment"
The effectiveness of school vouchers to improve opportunities for disadvantaged students is hotly contested. We present evidence form a privately-run, educational voucher program in Toledo, Ohio. The program focused on low-income families with children in grades K-8 and used a lottery to randomly determine voucher recipients. Like other studies, we measure the effects of the voucher on traditional measures of student learning (e.g. test scores and grade repetition). We also measure the effects of the voucher on other outcomes that schools likely effect. Specifically, we use economic experiments to measure voucher effects on altruism, patience, and overconfidence, for both successful and unsuccessful voucher applicants, and their parents. If vouchers affect these important economic behaviors, then researchers and policymakers may have to reevaluate the cost/benefit analysis of vouchers. This study is the first to use economic decision-making laboratory experiments to directly assess major policy debate. Our results provide mixed support for the effectiveness of the voucher on academic and non-academic outcomes. While we find that voucher winners scored higher on achievement tests than unsuccessful applicants, the difference is not robust to the inclusion of covariates or across multiple test score measures. We do find that voucher winners were more altruistic towards charitable organizations but were not more patient or less overconfident than unsuccessful applicants. We also document sereral results that add robustness to the emerging economics literature examining children's behavior, and show that several behavioral dispositions previously only observed with college students and adults (e.g., hyperbolic discounting and gender and ethnic differences) are present in children's behavior.
Dean S. Karlan(Princetron University) -
"Using Experimental Economics to Measure Social Capital and Predict Financial Decisions"
Questions remain as to whether results from experimental economics games are generalizable to real decisions in non-laboratory settings. I conduct a survey and two experiments, a Trust game and a Public Goods game, to measure social capital. Social capital purports to provide incentives to individuals to abide by otherwise difficult to enforce contracts. I then examine whether behavior in the games predicts repayment of loans to a Peruvian group lending microfinance program. I find that individuals identified as "trustworthy" by the Trust game are more likely to repay their loans. Individuals identified as "trusting" default more and save less, suggesting that those who "trust" more in the game are prone to taking bad risks. Behavior in a public goods game does not predict any future decisions in this context. Those with more positive attitudes towards society, as measured by three questions identical to those used in the General Social Survey, are more likely to repay their loans.
Abigail Barr(University of Oxford) -
"Cooperation and Shame"
Using two economic experiments I investigate how a sample of rural communities in Zimbabwe approach social dilemmas. When provided with an opportunity to impose sanctions in the context of a public goods game, fourteen out of eighteen communities achieved higher levels of cooperation. In thirteen communities the imposition of shame-based sanctions in the form of light-hearted criticism was observed. The resulting data revealed that: both non-cooperators and cooperators were criticised; community members cared about what their neighbours thought of them and made adjustments to their behaviour accordingly; the overall pattern rather than individual experiences of criticism affected subsequent behaviour; those who made low contributions and witnessed the criticism of others who made similar contributions, made higher contributions subsequently; while those who experienced such criticism first-hand made significantly smaller adjustments to their behaviour; those who made high contributions and witnessed the criticism of others who made similar contributions, made lower contributions subsequently; and to the extent that an opportunity to criticise passed by unexploited subsequent levels of cooperation were reduced.
Jeffrey Carpenter(Middlebury College), Stephen Burks(University of Minnesota-Morris) and Eric Verhoogen (University of California-Berkeley)
"Comparing Students To Workers: Ther Effects of Stakes, Social Framing, and Demographics on Bargaining Outcomes"
We compare the behavior of three subject groups in one-shot Ultimatum Game and Dictator Game bargaining experiments: young students at Middlebury College, older students at Kansas City Kansas Community College (KCKCC), and employees at two warehouses near Kansas City, Kansas. Most participants play for stakes of US$100. In the UG the student and employee groups from Kansas were similar to each other, and made slightly higher offers than the Middlebury students. In the DG the employees were markedly and significantly more fair-minded than the students. We find, however, that an analysis of subject group differences across demographically dissimilar groups, that does not control for demographic differences and also test for interaction effects, is likely to be misleading.
Donna L. Bahry (Penn State University) and Rick K. Wilson(Rice University)
"Confusion or Fairness in the Field? Rejections in the Ultimatum Game under the Strategy Method."
Field experiments conducted in the Russian Republics of Tatarstan and Sakha showed that proposers in the "ultimatum game" almost invariable made offers that split a day's wage at 50/50. These experiments were conducted in a variety of settings, from small isolated villages to large urban areas. While the consistency in proposer behavior seemed odd, it was not given the expectations of the responders. Using the strategy method to elicit a profile of rejections, we find that responders invoke an exceedingly strong norm of a 50/50 split or nothing at all. This paper focuses on the profile of rejections and combines the experimental data with data drawn from an extensive face-to-face attitudinal survey conducted on the same subjects six weeks prior to participating in the experiment.
Abigail Barr(University of Oxford) and Bill Kinsey (Vrije Universiteit Amsterdam) -
"Do Men Really have no Shame?"
Microfinance is one of the most commonly applied development interventions of our time. It is also one of the most gender-biased. In part, this is due to targeting. However, it might also relate to the emphasis placed by microfinance providers on group-loans. If women have a comparative advantage when it comes to functioning in groups, they might self-select into microfinance provided as group loans, while men seek alternative sources of credit. This paper explores the possibility that such a comparative advantage exists and that it relates to women's greater propensity to feel shame and/or induce feelings of shame in others. It uses data derived from an economic experiment conducted in 12 Zimbabwean villages to test a series of hypotheses. The findings suggest that men regard others less than women when deciding how to behave; that, even after controlling for this, they are more likely to attract criticism; and that they are no less responsive than women to such shame-inducing, social sanctioning. Finally, while men are no more inclined to sanction others they are less effective than women at effecting a resultant improvement in behaviour.'
Justine Burns(University of Cape Town) -
"Insider-Outsider Distinctions in South Africa: The Impact of race on the behaviour of High School students"
South African society has experienced numerous changes since the advent of democracy in 1994. Yet, it is unclear whether these changes have been successful in undermining race-based distinctions and sterotypes propagated and enforced under the apartheid regime. This paper reports the results of a trust game and dictator game played by high school students in the Western Cape, South Africa, and finds evidence that race-based distinctions persist in behaviour. Black students, while favoured in the dictator game, are not trusted in the trust game. This distrust arises both from an expectation that black students would return less in the trust game, as well as prejudice towards blacks in a strategic environment. Moreover, preliminary evidence suggests that these behavioural biases may be particularly important for students in racially homogenous schools and students whose close friends are all from the same race group.
Simon Gachter(University of St. Gallen and CESifo) and Benedikt Herrmann (University of Gottingen) -
"Norms of cooperation among urban and rural dwellers: Experimental evidence from Russia"
We investigate norms of cooperation by conducting public goods experiments with and without punishment among student and non-student urban and rural dwellers in Russia. There are strong differences between the generations: our non-students were socialized in the collectivist former Soviet Union; and our student subjects experienced their formative years in the transformation period of the post-Soviet era. A second dimension is the still-substantial gap between urban and rural areas in Russia. In the absence of punishment, we find strong differences between subject pools in cooperation rates, but in the presence of punishment differences between subject pools vanish. Punishment does not significantly affect cooperation rates - a finding which is in stark contrast to most of the existing evidence. Punishment behavior is strongly different between subject pools and reflects a couple of interesting social interaction effects.