| JUN/2013 |
| JEEA |
 |
 |
| Journal of the European Economic Association (JEEA) Volume 11 Issue 3 |
 |
| Social Norms: Theory and Evidence from Laboratory and Field |
| The June 2013 issue of JEEA contains eight papers on the topic of social norms. While JEEA has published special issues on an invited topic, this issue has an altogether different origin. A number of ordinary submissions in the years 2010-2012 had social norms as a broad topic. These submissions differed by methods -- applied theory, laboratory experiments, field experiments -- but they had a clear unifying theme. The board of editors decided that publishing these papers together would highlight the common themes and the interest in the topic -- hence, this June 2013 issue. The first set of papers are investigations of social norms in the laboratory. Krupka and Weber introduce a novel experimental technique to elicit social norms on acceptable behavior: subjects play coordination games with a monetary incentive to guess what other players find to be acceptable play. Using the elicited norms from this simple technique, the authors help make sense of behavior in variants of the dictator game which standard social preference models have a hard time explaining. Indeed, in a horse race with other social preference models, the norm-based elicitations add substantial predictive value. A related paper in this June 2013 issue is Gächter, Nosenzo, and Sifton which analyzes how social norms elicited with the Krupka-Weber procedure explain behavior in a three-person gift exchange game. In these other games, the horse race between models of social preferences and the social norm measures favors the more standard social preference models. While additional evidence is needed to ultimately assess how much experimental measures of social norms can add to social preference models, the Krupka-Weber method is an innovative and influential contribution in the literature. A second experimental paper which introduces a methodological innovation is Föllmi-Heusi and Fischbacher. This paper proposes a simple design to measure honesty in the laboratory. Subjects are given a die which they roll privately. They then report the roll of the die to an experimenter. Since the payout is increasing in the outcome of the roll, there is an incentive for the subjects to cheat by providing higher numbers. While at the individual level it is impossible to tell whether a draw of, say, 6 is an outcome of chance or of cheating, it is possible at the population level to make inferences about the occurrence of dishonesty. The paper finds that there is a significant degree of dishonesty at the margin, but few people who cheat maximally by choosing a 6. The paper also explores the comparative statics of dishonesty with respect to stakes, privacy, and other determinants. This design has already been used in a number of other studies on the topic, and is related to designs that Dan Ariely, Francesco Gino, and coauthors have used to also study honesty. The final laboratory experiment in the special issue is Capellen, Moene, Sørensen, and Tungodden. This paper builds on earlier work by the authors which introduces a novel experimental methodology to separate the intensity of fairness preferences from differences in fairness norms. This methodology is used in an interesting design using subjects from different countries to elicit preference for same-country subjects and generosity towards subjects from a poorer country, among other things. This special issue also contains an ambitious applied theory paper, Dal Bó and Terviö which provides a model of social norms. Such models are notoriously hard to write, as the concept itself of social norms is hard to pin down. The authors specify a model with self-signalling aspects and, after characterizing some general properties, they apply it to understand moral behavior in a variety of economics settings. The third set of papers, in addition to the laboratory experiments and the applied theory, consists of field experiments. Fellner, Sausgruber, and Traxler and Pruckner and Sausgruber are conceptual related field experiments, if in very different settings. Fellner, Sausgruber, and Traxler considers a sample of Austrian tax-payers who were not complying the the compulsory tax associated with the ownership of a television set. Using a large-scale field experiment, they investigate the impact of sanctions versus the impact of social norms, implemented both as moral reminders as well as Cialdini-type reminders that the large majority of taxpayers pay such tax. They find precisely estimated sizeable effects of the sanctions, which they compare to the moral reminders. This experiment is a precursor to experiments on tax evasion which the Nudge Unit in the UK government is implementing. Pruckner and Sausgruber instead use the settings of newspapers sold in the streets in a box to measure honesty, since they can observe how many people pick a newspaper without paying the required fee. Finally, Costa and Kahn analyze the power of social norms in the now-famous O-Power implementation, based again on Cialdini's pioneering work on social norms. Households receive comparisons of they energy consumption to neighboring homes. These comparisons lead to an estimated decrease of energy consumption of about 2 percent, a reliable effect estimated in other papers. The novel finding in Costa and Kahn is the interaction of this effect with political identity. While households registered as Democrat indeed record a significant decrease in energy consumption in response to the social comparison intervention, there is no effect in households registered as Republican. This suggests that in the implementation of nudges of this type one ought to pay careful attention to the heterogeneity in demographics in the population LIST OF ABSTRACTS |
|
 |
* If your EEA membership has expired and you no longer wish to renew it or receive EEA Newsletters, please reply to this message with the Subject as "Cancel".
=============================================
The European Economic Association
http://www.eeassoc.org
=============================================