2013 Regional ESA Conference

October 24-26, 2013

Hotel Paradox

Santa Cruz, California

Last Revised 08:21:56 PDT, 2013.10.22


October 24, 5:30 - 8:00 pm


Friday, October 25, 8:00 am -9:00 am

Plenary Session: Rachel Croson, University of Texas Arlington, "Experimental Economics Imperialism"
Chair: Tim Cason

Friday, October 25, 7:30 am - 6:30 pm

Registration Desk Open, Hotel Paradox
Paradox Breakfast Buffet, Solaire Restaurant, 6:30 - 8:30 am
Paradox Lunch Buffet, Solaire Restaurant, 12:00 - 1:20 pm

Friday, October 25, 9am - 6:00 pm

Experimental Software and Hardware Demos, Hotel Paradox


Friday, October 25, 9:20am - 10:20am

Session 1, Sequoia A: Auctions: Bidding Behavior

Session 2, Sequoia B: Conflict and Contests

Session 3, Sequoia C: Games: Sophistication and Expertise

Session 4, Sequoia D: Market Design 1

Session 5, Grove: Risk Preference

Session 6, Cypress: Interpersonal Influence

Session 7, Fitness Center: Cooperation and Institutions 2

Friday, October 25, 10:40am - 12:00pm

Session 1, Sequoia A: Decision Making Under Ambiguity

Session 2, Sequoia B: Repeated Games 1

Session 3, Sequoia C: Finance 1

Session 4, Sequoia D: Belief Formation

Session 5, Grove: Environmental 1

Session 6, Cypress: Bargaining (Behavioral)

Session 7, Fitness Center: Cooperation and Institutions 1

Friday, October 25, 1:20pm - 2:40pm

Session 1, Sequoia A: Networks

Session 2, Sequoia B: Repeated Games 2

Session 3, Sequoia C: Finance 2

Session 4, Sequoia D: Special Session: Pay Protocols

Session 5, Grove: Environmental 2

Session 6, Cypress: Labor 1

Session 7, Fitness Center: Beliefs and Others Behavior

Friday, October 25, 3:00pm - 4:20pm

Session 1, Sequoia A: Cooperation and Institutions 3

Session 2, Sequoia B: Repeated Games 3

Session 3, Sequoia C: Industrial Organization 1

Session 4, Sequoia D: Biased Beliefs

Session 5, Grove: Behavioral Finance

Session 6, Cypress: Inequality and Social Preferences

Session 7, Fitness Center: Legislative Bargaining and Committtee Behavior

Friday, October 25, 4:40pm - 6:00pm

Session 1, Sequoia A: Cooperation and Institutions 4

Session 2, Sequoia B: Risk Preference: Social Effects

Session 3, Sequoia C: Industrial Organization 2

Session 4, Sequoia D: Norms

Session 5, Grove: Labor 2

Session 6, Cypress: Charitable Contributions 1

Session 7, Fitness Center: Voting Behavior


Saturday, October 26, 8:00am - 9:00am

Plenary Session:
Welcome: Sheldon Kamieniecki, Dean of Social Sciences, UCSC
Speaker: Al Roth, Stanford University, "Market Design" [tentative]
Chair: Dan Friedman

Saturday, October 26, 8:00 am - 10:00 am

Registration Desk Open, Hotel Paradox
Paradox Breakfast Buffet, Solaire Restaurant, 6:30 - 8:30 am
Paradox Lunch Buffet, Solaire Restaurant, 12:00 - 1:20 pm

Saturday, October 26, 9am - 6:00 pm

Experimental Software and Hardware Demos, Hotel Paradox


Saturday, October 26, 9:20am - 10:20am

Session 1, Sequoia A: Learning

Session 2, Sequoia B: Social Information 1

Session 3, Sequoia C: Survey Methods: Characterizing Behavior

Session 4, Sequoia D: Gender 1

Session 5, Grove: Market Design 2

Session 6, Cypress: Charity 2

Session 7, Fitness Center: Dynamic Choice

Saturday, October 26, 10:40am - 12:00pm

Session 1, Sequoia A: Spatial Interactions

Session 2, Sequoia B: Social Information 2

Session 3, Sequoia C: Lab to Field 1

Session 4, Sequoia D: Gender 2

Session 5, Grove: Firm Behavior 1

Session 6, Cypress: Lies and Deception

Session 7, Fitness Center: Tournaments and Contests 1

Saturday, October 26, 1:20pm - 2:40pm

Session 1, Sequoia A: Cooperation and Institutions 5

Session 2, Sequoia B: Experimental Methodology

Session 3, Sequoia C: Social Preferences 1

Session 4, Sequoia D: Time Preferences

Session 5, Grove: Strategic Sophistication

Session 6, Cypress: Gender and Stereotypes

Session 7, Fitness Center: Tournaments and Contests 2

Session 8, Poolside: Preferences and Behavior

Saturday, October 26, 3:00pm - 4:20pm

Session 1, Sequoia A: Coordination

Session 2, Sequoia B: Lab to Field 2

Session 3, Sequoia C: Social Preferences 2

Session 4, Sequoia D: Charitable Contributions 2

Session 5, Grove: Experiments in Development

Session 6, Cypress: Macroeconomics

Session 7, Fitness Center: Auctions: Institutions

Saturday, October 26, 4:40pm - 6:00pm

Session 1, Sequoia A: Communication and Cooperation

Session 2, Sequoia B: Lab to Field 3

Session 3, Sequoia C: Social Preferences 3

Session 4, Sequoia D: New Data: Inference from Decision Times

Session 5, Grove: Reference Dependent Preferences

Session 6, Cypress: Trust and Trustworthiness

Session 7, Fitness Center: Auctions: Incentives

Abstracts

Does anticipated loss-aversion explain overbidding in first-price auctions?

Anmol Ratan
Monash University

ABSTRACT:

A large body of experiments claims that individuals often interpret outcomes as gains and losses. Bidders may apply similar interpretations to auction outcomes once they develop expectations regarding winning or losing in auctions. Using this approach, overbidding with respect to risk-neutral Bayesian Nash equilibrium prediction in first-price auctions can be rationalized. In this paper, I provide a test of anticipated loss aversion in induced value first-price auctions. In my experiment individuals compete against pre-programmed computer opponents. In one treatment, computers submit risk-neutral Nash bids (NE auctions) and in another treatment computers submit zero-profit bids (ZP auctions). Payoffs are manipulated to control for expected utility risk aversion or other alternative explanations for treatment differences. A unique treatment effect is obtained for loss averse bidders: In ZP auctions, anticipated loss-aversion predicts underbidding (overbidding) as compared to NE auctions for levels of loss aversion that lie below (above) a unique threshold. The experimental data supports the unique treatment effect for levels of loss aversion that lie below (above) this unique threshold. The paper contributes to emerging literature which suggests that expectations can drive references to explain behavior.

Author(s): Anmol Ratan
Topic: Markets: Auctions, Decision Theory: Bounded Rationality
Back to session: Auctions: Bidding Behavior


The Willingness-to-Pay Discrepancy between Incentive Compatible Auctions and Posted Price Offerings

Shang Wu
University of Delaware

ABSTRACT:

Auction experiments are commonly used to elicit participant's values for a wide range of items and services. These auctions are theoretically incentive compatible, so are believed to give an unbiased estimate of consumers' willingness to pay (WTP). This study tests this by comparing WTP from a second price Vickrey auction to a posted price offering in a within subjects, homegrown value setting. Jars of honey are offered to 115 adult participants. Consumers' WTP are found to be about 50% lower in the auction compared to the Posted Price. We consider several possible explanations for this difference in behavior, and eliminate anchoring and yea-saying effects as likely factors. Unfamiliarity with the mechanisms appears to be a plausible explanation.

Author(s): Shang Wu, Jacob Fooks, Kent Messer and Deborah Delaney
Topic: Methodology: Lab, Field and External Validity, Methodology: Experimental Design, Tools and Practices
Back to session: Auctions: Bidding Behavior


Bidding behavior in affiliated-values first-price auctions

Theodore Turocy
University of East Anglia

ABSTRACT:

Bidding in excess of the (risk-neutral) Bayes-Nash prediction is a robust result in both private-values and common-values first-price auctions. These two environments are generally studied separately, and explanations for the phenomenon differ, with risk aversion, joy of winning, or regret being advanced most commonly for private values auctions, while explanations about failing to learn about the winner's curse receive attention in the common value setting. We conduct a series of experiments using a framework in which two bidders receive independent, uniformly distributed private signals. We vary the way these signals map into values, in such a way that the standard independent private values case as well as a rescaled version of the wallet game are special cases. For intermediate parameterisations, values are affiliated. Our setting has the property that the risk-neutral Bayes-Nash equilibrium is the same for all parameterisations, while varying the degree of commonality of values; whereas models (such as cursed equilibrium) which incorporate misperception of the winner's curse make qualitatively different predictions in the common-value setting. We therefore can evaluate the incremental effect of failing to correct for the winner's curse relative to the baseline of aggressive bidding in the private values setting.

Author(s): Timothy Cason, Theodore L. Turocy
Topic: Markets: Auctions, Games: Information
Back to session: Auctions: Bidding Behavior


Social Learning in Continuous-Time All-Pay Auctions

Daniel Stephenson
Texas A&M University

ABSTRACT:

This study investigates social learning in continuous-time all-pay auctions. Instead of conducting discrete rounds, our subjects earn payoffs continuously over time and can adjust their bids at will throughout the experiment. In combination with the all-pay auction's continuous strategy space, this continuous-time approach gives us a fine-grained picture of the empirical dynamics, allowing us to disentangle alternate sources of behavioral noise. To identify the role of social information in the learning process, we conduct two information treatments. In our local information treatment, subjects receive real-time feedback regarding their own payoff via an intuitive graphical interface. In our global information treatment, subjects also observe the the bids and payoffs of others. Subjects in both treatments exhibit persistent cyclical behavior. Moreover, we find that social information raises the frequency of cycles, aggravates overbidding, and decreases behavioral noise.

Author(s): Daniel Stephenson
Topic: Games: Contests, Decision Theory: Learning
Link: danielgstephenson.com
Back to session: Auctions: Incentives


Curved or Twisted: Incentive Effects of Proportional Grading, and Other Consequences of All-Pay Auctions

Andrew Brownback
University of California--San Diego

ABSTRACT:

Consider a contest with a fixed proportion prizes, such as a set fraction of A grades per student, or of open tenure slots per Assistant Professor. One could organize the contest as an all-pay auction, where higher effort can increase the likelihood of a prize. The policy question we pose is how effort varies if the contest is run as a single large auction or as a set of separate smaller auctions. What are the consequences for the allocation of effort by skill level, and on the aggregate effort? We solve the model theoretically and test the predictions in a laboratory with auctions featuring 2 players and 1 prize or featuring 20 players and 10 prizes. We found that subjects perform strikingly similarly to the risk neutral predictions, thus providing us with different policy prescriptions depending on the relative welfare weights on the extracting effort from low- or high-skilled contestants.

Author(s): James Andreoni, Andrew Brownback
Topic: Markets: Auctions, Games: Contests
Back to session: Auctions: Incentives


Modeling Unobserved Consumer Heterogeneity in Experimental Auctions: A Censored Random Parameters Approach

Marco Palma
Texas A&M University

ABSTRACT:

Data collected using experimental auctions often contain multiple bids from the same respondent and are typically analyzed using panel-data models. Econometric methods used have considered bid-censoring and random effects that capture heterogeneity in the intercepts, but have failed to account for heterogeneity in the coefficients. In this paper, we compare different estimation methods, and provide evidence that employing a Random Parameters Tobit model in the analysis of experimental auction data allows accounting for unobserved consumer heterogeneity in the treatments and handling bid-censoring. When the model is applied to data collected in a non-hypothetical sealed-bid second-price Vickrey auction conducted to elicit willingness-to-pay for food safety standards in specialty melons, findings indicate that valuations are censored and heterogeneous.

Author(s): Alba J. Collart, Marco A. Palma
Topic: Methodology: Experimental Design, Tools and Practices, Markets: Auctions
Back to session: Auctions: Incentives


An Experimental Investigation of Procurement Auctions with Asymmetric Sellers

Li Hao
University of Arkansas, Fayetteville

ABSTRACT:

Auctions are commonly used procurement mechanism. However, most of the research to date has focused on suppliers that are a priori symmetric in that their costs are drawn from a common distribution. While such an assumption may be applicable in some contexts, in many cases a supplier's cost has to do more with their own operations or economies of scale and scope. This paper reports the results of controlled laboratory experiment designed to compare prices between first price and second price procurement auctions when sellers are asymmetric. The results indicate that first price auctions generate lower prices regardless of market composition and that second prices are either equally or less efficient depending on the market composition.

Author(s): John Aloysius, Cary Deck, Li Hao
Topic: Markets: Auctions,
Back to session: Auctions: Incentives


Auctions with Tokens

Michael Caldara
University of California, Irvine

ABSTRACT:

Traditional solutions to the assignment problem are less than ideal in high contention environments. When preferences are similar but vary in magnitude, market clearing prices may be high and random serial dictatorship may be inefficient. In this paper, we explore the efficiency and welfare properties of several alternative mechanisms. Participants are endowed with an allotment of non-monetary tokens to bid with in place of money. We study both open outcry and sealed bid auctions when token budgets are fixed and when participants can buy and sell tokens in a pre-market.

Author(s): Michael Caldara; David Porter
Topic: Markets: Auctions, Markets: Market design and Matching
Back to session: Auctions: Institutions


Theory and Experiments on Bidding Behavior in the Bucket Auction

Joshua Foster
University of Arkansas

ABSTRACT:

Participation rates are directly relevant to the auctioneer's ability to generate revenue. A novel auction mechanism called the bucket auction has shown incredible potential for charity auctions as a way to generate revenue among bidders with impure social preferences. In this paper, a theoretical model is designed to predict bidder behavior in for-profit and charity bucket auctions. This theory addresses the bucket auction's participation rate by endogenizing how bidders choose endowments of tokens used for bidding.

Author(s): Joshua Foster
Topic: Markets: Auctions, Social Behavior: Other-regarding Preferences
Back to session: Auctions: Institutions


Auctions or Contests? Resource allocation under strategic uncertainty

Katerina Sherstyuk
University of Hawaii, Dept of Economics

ABSTRACT:

We compare methods to allocate resources or operator licenses in industries where the future value of the resource is unknown at the time of allocation as it depends on ex-post joint actions of operators. For example, the value of radio spectrum licenses may depend on ex-post investments into infrastructure development by the operators. Some countries consider infrastructure development among the top priories for telecommunication industry. In this project, we investigate the effects of operator license allocation methods on ex-post industry performance in environments with strategic uncertainty. We assume that the future profitability of holding a license is uncertain at the time of license allocations as it depends on the investment decisions of all operators in the industry. We use the experimental laboratory to compare three methods to allocate licenses: (1) auction; (2) pure investment commitment contest, where the only selection criterion is the investment commitment to the future infrastructure development; and (3) bureaucratic allocation procedure, modeled as lottery. After the licenses are allocated, the winners make investment decisions which we model as a minimal effort coordination game. We consider which allocation method is more successful in promoting higher investment levels and higher profitability for operators. We also investigate whether overbidding is common in auctions.

Author(s): katerina Sherstyuk, Nina Karmanskaya and Pavel Teslia
Topic: Games: Coordination, Markets: Auctions
Back to session: Auctions: Institutions


It's not all in the face - Misjudging your opponents in competitive bargaining.

Georg Primes
University of Vienna

ABSTRACT:

In social interactions the human face is an essential source of information that helps us to assess and understand our counterpart. Studies confirm its value as they show that participants willingly invest parts of their endowment to gain access to visual information on co-players. We report results from nine sessions of a round-robin ultimatum game tournament. Players were introduced to their co-players via standardized portrait photographs. We expected players to adjust their bargaining behavior according to their first impression of their partner. Hence, players may vary their strategy from subgame perfection to generous bargaining depending on the current counterpart. The strategy method, eliminating learning effects, allowed us to use three behavioral categories to analyze competitive bargaining: i) minimal acceptable offer (MAO), ii) offer size, and iii) variability in behavior. The competitive environment of the game did not seem to affect observable MAO levels. Proposers choosing offer levels closer to subgame perfection performed poorly in this competition. Surprisingly, strategically differentiating between co-players negatively affected the overall performance of a participant. Generally, winners constantly offered at a higher level while demanding less from their competitors. Hence, a competitive environment may produce the same bargaining strategies than a non-competitive framework. In contrast to our expectations, players who adjusted their strategy to different co-players failed to assess them correctly. Their offers were rejected more often than the offers of players who chose to 'ignore' the photographs and treat co-players equally. This is particularly surprising as humans' proneness and ability to assess a face are discussed in many studies from the fields of biology and psychology.

Author(s): Georg Primes, Johannes Jaschke
Topic: Psychology and Biology: Other, Games: Bargaining
Back to session: Bargaining (Behavioral)


I see you, you see me: how perceived facial dominance and trustworthiness influence bargaining behavior.

Johannes Jaschke
University of Vienna

ABSTRACT:

People rapidly make judgments from faces that influence decision processes and, thus, outcomes of social interactions. In particular, facial trustworthiness and dominance are regarded as good proxies of whether to approach or avoid a person that might cause harm. Deducing intentions and behavioral strategies from facial cues should thus have evolutionary significance, reducing the probability of conflict while increasing the probability of future interactions. Applying nine sessions of an ultimatum game (UG) we investigated whether dominance and trustworthiness perceptions impinge on behavior in support of recent evolutionary theories.

Author(s): Johannes Jaschke, Georg Primes
Topic: Psychology and Biology: Cogonition, Social Behavior: Other
Back to session: Bargaining (Behavioral)


The Demand for Expressing Emotions

Kristian Lopez Vargas
Economics - University of Maryland

ABSTRACT:

Abstract Economics has mainly studied communication for its instrumental purpose, that is, for how it affects play and outcomes in the material interaction. We research on the non-instrumental aspect of communication. In particular, we study the material value and purpose of ex-post verbal expression in a modified Power-to-Take game, where previous research suggests expression is likely to be driven by mood-emotional episodes. Our main findings show this type of expression has a non-negligible material value and, that purely expressive as well as reciprocal motives are both significant components of this valuation. We also find that the anticipation of expression induces a more pro-social taking behavior. Finally, we document that stronger mood and emotional reactions are associated to higher valuation for expression. Our research indicates that incorporating expression possibilities in economic interactions can have an important impact on well-being via other psychological mechanisms besides those instrumental and informational channels normally studied in cheap talk.

Author(s): Brit Grosskopf & Kristian Lopez-Vargas
Topic: Games: Bargaining, Psychology and Biology: Emotions
Link: https://docs.google.com/file/d/0BybNqMF3VOCYaW9XN19NUDlvVzQ
Back to session: Bargaining (Behavioral)


In Search of the Angry Button

Theo Offerman
University of Amsterdam

ABSTRACT:

Anger at another individual can be viewed as a means of creating credibility of a threat to punish, when other commitment devices are absent. In an ultimatum game, angrier responders are relatively likely to reject low offers, and thus can win better terms if they can communicate to proposers that they anger easily. Moreover, proposers can gain an advantage if they can distinguish whether a responder is likely to anger at a low offer. We report an experiment to explore the extent to which individuals have this potentially valuable skill of detecting an angry button. Participants view photographs and videos of other individuals in various emotional states, including anger and neutrality. They then must identify which individuals reject and which accept low offers when they subsequently play an ultimatum game.

Author(s): Boris van Leeuwen, Charles Noussair, Theo Offerman, Sigrid Suetens, Matthijs van Veelen and Jeroen van de Ven
Topic: Games: Bargaining, Psychology and Biology: Emotions
Back to session: Bargaining (Behavioral)


Belief in Mean Reversion and the Disposition Effect

Peiran Jiao
Claremont Graduate University

ABSTRACT:

The disposition effect is the investors' tendency to disproportionately sell more winners than losers. This paper experimentally evaluates the relative importance of its two competing behavioral explanations: belief in mean reversion and prospect theory. In the experiment, fully 61% percent of the participants exhibited significant disposition effect. I use participants' risk preference elicitations in the domains of gains and losses, and their belief elicitations about price movements, to predict their selling behavior of assets and the magnitudes of the disposition effect. I find belief in mean reversion has more prediction power than prospect theory. While both theories explain selling decisions, individual participants' belief in mean reversion alone explains 17% of the between-participant variations of the disposition effect, whereas prospect theory parameters and trading experience in real-world financial markets turn out to be insignificant conditional on beliefs.

Author(s): Peiran Jiao
Topic: Markets: Finance, Decision Theory: Beliefs
Back to session: Behavioral Finance


The Disposition Effect and Realization Preferences: a Direct Test

Jacopo Magnani
UCSC

ABSTRACT:

This paper provides a new experimental method to test for the hypothesis that individual investors have an irrational preference for selling winning stocks vis-a'-vis selling losing stocks. Using a real option framework, I design an experiment where optimal behavior is characterized by an upper and a lower selling thresholds in the asset price space, thus producing a clear rational benchmark. The design ensures that the disposition effect is qualitatively different from the behavior induced by risk aversion, an important confound in previous studies. I find that subjects tend to delay selling losers beyond the optimal point and anticipate selling winners before the optimal liquidation time.

Author(s): Jacopo Magnani
Topic: Decision Theory: Risk, Markets: Finance
Back to session: Behavioral Finance


Do Lottery Payments Induce Savings Behavior? Evidence from the Lab

Emel Filiz Ozbay
University of Maryland

ABSTRACT:

This paper presents the results of a laboratory experiment designed to investigate whether the option of a Prize Linked Savings (PLS) product alters the likelihood that subjects choose to delay payment. By comparing PLS and standard savings products in a controlled way, we find strong evidence that a PLS payment option leads to greater rates of payment deferral than does a straightforward interest payment option of the same expected value. The appeal of the PLS option is strongest among men, self-reported lottery players, and subjects with low bank account balances. We use the results of our experiment to structurally estimate the parameters of the decision problem governing time preference, risk aversion, and probability weighting. We employ the parameter estimates in a series of policy simulations that compare the relative effectiveness of PLS products as compared to standard savings products.

Author(s): Emel Filiz-Ozbay, Jonathan Guryan, Kyle Hyndman, Melissa Kearney, and Erkut Y. Ozbay
Topic: Decision Theory: Risk,
Link: http://econweb.umd.edu/~filizozbay/savings.pdf
Back to session: Behavioral Finance


Nominal Loss Aversion and Individual Investment Decisions

Thomas Stephens
WU Vienna

ABSTRACT:

We investigate the role of nominal loss aversion in individual investment decisions. As part of a large-scale internet experiment, we present a large, heterogeneous sample of subjects with a series of financial investment decisions. Each decision combines a real and a nominal environment. Subjects choose how much to invest, and how to allocate their investment between a relatively safe and a relatively risky option. The real scenarios include both gains and losses, with each real scenario presented as a nominal gain and a nominal loss. This design allows us to distinguish between aversion to real and aversion to nominal losses.

Author(s): Thomas A. Stephens, Jean-Robert Tyran
Topic: Decision Theory: Bounded Rationality,
Back to session: Behavioral Finance


The Impact of Variance and Skewness of Court Awards on Eminent Domain

Eric Cardella
Texas Tech University

ABSTRACT:

Legal disputes are often negotiated and settled prior to going to court. While pre-court settlements are common, they are not universal. Much of the prior literature on court settlements has focused on the role of information asymmetries leading to settlement failures. In this paper, we take a complementary approach by exploring how uncertainty of the underlying court award distribution impacts settlements. Specifically, we develop an experimental design, based on a stylized eminent domain bargaining game between a buyer and seller, which enables us to systematically test how increases in variance and skewness of the award distribution impact negotiation behavior and settlement rates. In the aggregate data, we find that increases in variance decrease settlement rates. We also find that increases in skewness have a non-monotonic effect on settlement rates; relatively small initial increases in skewness increase settlement rates, but settlement rates start to decrease as skewness increases further. As part of the experimental design, we also gather individual measures of risk aversion and prudence of using a lottery choice task, which enables us to incorporate these individual risk preference measures into the analysis of the results.

Author(s): Eric Cardella; Carl Kitchens
Topic: Decision Theory: Risk, Games: Bargaining
Back to session: Belief Formation


Performance Uncertainty and Optimistic Beliefs

Stephanie Heger
Washington University in Saint Louis

ABSTRACT:

In choosing among occupations, investment strategies or when considering the option to become self-employed, individuals may make poorly informed decisions since they over-estimate the probability of outcomes with high payoffs. An intervention aimed at providing information, such as a training program for future entrepreneurs, could therefore be very valuable. However, its value would rely on an understanding of what sort of information to provide---and to whom. We design an experiment that distinguishes between two patterns in beliefs that could explain why individuals over-estimate the likelihood of success. The first is performance over-estimation, whereby the individual believes his performance is better than it actually is. The second is optimism, whereby, irrespective of his performance, the individual overweights the probability of exogenously determined events that he prefers. Having isolated them in the laboratory, we show that these two beliefs patterns are positively correlated at the individual level. Next, we examine beliefs in a combined setting where individuals simultaneously face performance uncertainty and preferences over outcomes. We show to what extent beliefs in this combined setting can be attributed to performance overestimation versus optimism. This decomposition has an interesting gender component. Male over-estimation in the combined setting is mostly attributed to performance over-estimation. For females, over-estimation in the combined setting is mostly attributed to optimism.

Author(s): Stephanie Heger and Nicholas Papageorge
Topic: Psychology and Biology: Gender and Individual Differences, Psychology and Biology: Gender and Individual Differences
Back to session: Belief Formation


Filter Bubbles, Echo Chambers, and Ideological Segregation in Online News Consumption

Justin Rao
Microsoft Research

ABSTRACT:

We use detailed browsing behavior data from over one million US-located internet users to examine patterns of news consumption. We use the textual content of each story read to filter out stories related to sports, weather, entertainment, and ?nance using a machine learning approach. The remaining stories are categorized as either news or opinion. For news stories, we find moderate segregation across users. Segregation does not differ much by consumption channel---organic browsing and social referrals from a site such as Facebook show similar patterns. For opinion stories, users are far more segregated and this segregation increases dramatically when opinion is consumed from social channels. For opinion, users that tend to consume content from one side of the political spectrum almost never read articles from the other side. The social channel is by far more prevalent for opinion as compared to regular news.

Author(s): Seth Flaxman, Sharad Goel and Justin M. Rao
Topic: Applied Economics: Other, Social Behavior: Communication
Link: www.justinmrao.com/bubbles.pdf
Back to session: Belief Formation


Reducing within-group overconfidence through group identity and between-groups confidence judgements

Philip Brookins
Florida State University

ABSTRACT:

Overconfidence in one's relative performance is a wide-spread phenomenon that in many cases has negative consequences for the decision maker. Similar to individual (or within-group) overconfidence, it was also shown that individuals are, on average, overconfident about the performance of their group relative to other groups. We hypothesize, however, that this between-groups overconfidence should mitigate within-group individual overconfidence. Simply put, if the decision maker believes that her group's ranking is high it must be because her beliefs about the performance of other members of her group are inflated, which we expect will lead to a reduction in her individual within-group overconfidence. We test this hypothesis in a laboratory experiment. Using a two-by-two between-subject design, we manipulate subjects' group identity and between-groups confidence judgments, while within-group confidence judgements are present in all four treatments. We find, as hypothesized, that the salience of between-groups confidence judgments significantly reduces within-group individual overconfidence. We also find an equally strong reduction of within-group overconfidence from the group identity manipulation. For managers whose goal is to improve calibration of their employees, our results imply that activities promoting between-groups (e.g., between-firms) comparisons and strengthening organizational identity should be helpful.

Author(s): Philip Brookins, Adriana Lucas, Dmitry Ryvkin
Topic: Psychology and Biology: Cogonition, Social Behavior: Group Behavior
Back to session: Belief Formation


An Experimental Investigation on Belief and Higher-Order Belief in the Centipede Games

Yun Wang
University of Pittsburgh

ABSTRACT:

This paper explores people's beliefs behind non-backward induction behavior in laboratory centipede games. We elicit players' beliefs about opponents' strategies and 1st-order beliefs. The measured beliefs help us to infer the conditional probability systems of both players. The inferred CPS's and players' actual strategy choices identify why they fail to reach the BI outcomes. We find that subjects maximize their monetary payoffs according to their stated beliefs less frequently in the Baseline Centipede treatment where an efficient non-equilibrium outcome exists; they do so more frequently in the Constant Sum treatment where the efficiency property is removed. Moreover, subjects believe their opponents' maximizing behavior and expect their opponents to hold the same belief less frequently in the Baseline Centipede treatment and more frequently in the Constant Sum treatment.

Author(s): Yun Wang
Topic: Decision Theory: Beliefs, Games: Other
Back to session: Beliefs and Others Behavior


An Experimental Analysis of the Cognitive Processes Underlying Beliefs and Perception Manipulation

Alison Sanchez
Univ. of Calif., San Diego

ABSTRACT:

The main focus of studies on beliefs has thus far been on subjects' abilities to form accurate beliefs about others actions. In this paper we focus on whether and how individuals model and manipulate the cognitive processes responsible for belief formation in others. The manipulation strategies subjects utilize reveal not only how they think, but also how they model the way others think. We gather data on choices and beliefs in a unique one-shot trust game paradigm in order to investigate these cognitive mechanisms. We introduce a new way to capture the difference between players' stated beliefs and their true beliefs. We find that as selfish behavior increased, sophisticated subjects lied about what they believed about their opponent's strategy, often blaming their victim, in order to preserve their own social image. Our results demonstrate that individuals have sophisticated awareness of the cognitive mechanisms behind decisions and use this knowledge to strategically manipulate information to influence the perceptions of others so that they may behave selfishly, but maintain a positive social image while doing so.

Author(s): James Andreoni and Alison Sanchez
Topic: Decision Theory: Beliefs, Psychology and Biology: Cogonition
Back to session: Beliefs and Others Behavior


Inequality and Relative Ability Beliefs

Jeffrey Butler
Einaudi Institute for Economics and Finance

ABSTRACT:

In this study I present experimental evidence of a novel channel yielding inequality persistence: inequality itself may be self reinforcing. In an initial experiment, results suggest that individuals respond to salient inequality by adjusting their performance beliefs to justify the inequality. Subsequent experiments reveal that i) it is beliefs about relative ability an ostensibly stable trait - rather than effort provision - that respond to inequality; and that ii) unequal pay in an initial task affects willingness to compete on a subsequent task for male participants. I conclude by discussing some implications of these findings.

Author(s): Jeffrey V. Butler
Topic: Decision Theory: Beliefs, Psychology and Biology: Gender and Individual Differences
Link: http://eiefexperiments.org/papers/IneqRelAbilBeliefs_PaperPlusTables.pdf
Back to session: Beliefs and Others Behavior


Theory of Mind: Theory and Experiment

Erik Kimbrough
Simon Fraser University

ABSTRACT:

This paper provides an evolutionary foundation for our capacity to attribute preferences to others. This ability is intrinsic to game theory, and is a key component of Theory of Mind, perhaps the capstone of social cognition. We argue here that this component of theory of mind allows organisms to efficiently modify their behavior in strategic environments with a persistent element of novelty. Such environments are represented here by multistage games of perfect information with randomly chosen outcomes. ??Theory of Mind?? then yields a sharp, unambiguous advantage over less sophisticated, behavioral approaches to strategic interaction. In related experiments, we show the subscale for social skills in standard tests for autism is a highly significant determinant of the speed of learning in such games.

Author(s): Erik O. Kimbrough, Nikolaus Robalino, Arthur J. Robson
Topic: Games: Information, Games: Other
Back to session: Beliefs and Others Behavior


A Cold Shower for the Hot Hand Fallacy

Adam Sanjurjo
UCSD

ABSTRACT:

In their seminal 1985 paper, Gilovich, Vallone and Tversky (GVT) show that the belief in the hot hand by players, coaches and observers of basketball is a fallacy. This was an important early result for the behavioral economics program because it was one of the first examples of people systematically committing a cognitive fallacy under high stakes in their domain of expertise. The original results have been subsequently reinforced by the combination of the repeated failures to detect a hot hand effect and the continued insistence on its validity by professionals and lay persons alike. The hot hand fallacy has become an exhibit of the power of cognitive illusions more generally; not only are they economically meaningful, but they can be robust to learning and advice. We demonstrate that the hot hand belief is not a fallacy. The conclusion of the original GVT study relies principally on a controlled shooting experiment because of confounds present in game data. We design a shooting task and develop statistical measures that together have superior identifying power. We find overwhelming evidence of the hot hand both in our study and in earlier controlled shooting studies. Additionally, in a betting task using videos of our shooters, we find that participants can detect streakiness, performing significantly different than chance levels.

Author(s): Adam Sanjurjo (with Joshua B. Miller)
Topic: Decision Theory: Beliefs, Decision Theory: Bounded Rationality
Back to session: Biased Beliefs


Consistent Individual Deviations from Bayesian Belief Updating and the Demand for Information

Sandro Ambuehl
Stanford University

ABSTRACT:

How do individuals value noisy information that guides economic decisions, and how does this relate to belief updating? Such interactions are all but inescapable in most economic contexts, but empirical evidence is scarce. Our experiment uses modifications of familiar updating tasks to measure both individuals valuation of information, and the change in beliefs that such information causes. Our design enables us to fully explore preferences over the space of information structures in a two-states two-signals setting. We test for key behavioral patterns in willingness-to-pay for information, and relate this to belief updating. The aggregate data are not far from the predictions of the standard model, but this masks substantial individual heterogeneity. Subjects differ consistently in their responsiveness to information the extent that their beliefs move upon observing signals. This underlying property links individuals behavior across different choice problems: more responsive subjects value information more highly, and are quicker to attain extreme beliefs in a qualitatively distinct choice setting.

Author(s): Sandro Ambuehl and Shengwu Li
Topic: Decision Theory: Beliefs, Decision Theory: Learning
Back to session: Biased Beliefs


Attribution Bias in Strategic Environments

Dominik Duell
New York University

ABSTRACT:

In a laboratory experiment based on a strategic career-concerns setting, we analyze how sharing a membership in a social group affects principals' assessments of the causes of agent performance. The treatments vary whether (artificially induced) group identification is specifically primed as well as the in-group vs. out-group matching of the principals and agents. We elicit from the principals their beliefs about agents' type and effort and analyze how principals' explanations of agents' performance vary depending on whether they share the social identity with the agents. We find that when principals use the sanctioning tools at their disposal to incentivize agents, principals' and agents' choices sustain what can be described as a ``strategic attribution bias'': upon observing good outcomes, principals attribute them more readily to their agents' effort and reward their agents more frequently when they share a social identity; and in turn, agents who share social identity with their principals tend to invest more into effort in expectation of principals' attribution. However, when principals do not use the sanctioning tools to induce better outcomes, there is no evidence of attribution bias.

Author(s): Dominik Duell and Dimitri Landa
Topic: Decision Theory: Beliefs, Social Behavior: Other
Back to session: Biased Beliefs


Luck, entitlement, and redistribution: experimental evidence

Olga Bogach
Brigham Young University

ABSTRACT:

Recent evidence suggests that the United States has both a larger variance of contemporary income and a far higher correlation between the earnings of fathers and sons than several OECD countries. For Americans, who have always believed that their country is a land of opportunity, this has led to a widespread debate about the meaning of economic opportunity in general, and the particular nature of the American Dream. On the other hand, researchers have shown that the countries with more income mobility across generations demonstrate lower returns to education and fewer work hours (Blanden 2009; Rogerson 2006). Thus, opportunity, or the ability to be rewarded for hard work or intelligence, may actually be lower in these countries.

Author(s): Olga Bogach, Lars Lefgren, David Sims
Topic: Applied Economics: Labor Market,
Back to session: Biased Beliefs


The Price of Warm Glow

Robert Slonim
University of Sydney

ABSTRACT:

This paper presents a theoretical model and experimental evidence to explain the volunteering puzzle where agents prefer volunteering time to donating money when monetary donations are, ceteris paribus, more efficient for providing resources to charity. In the model agents receive heterogeneous utility from pure and impure altruism (Andreoni 1989) that permits warm glow to vary between monetary donations and volunteering, thus allowing preferences for impure altruism to rationalize inefficient allocation decisions. We define a measure of the price of impure altruism as the additional proportion of income sacrificed by a donor to give in the dimension that maximizes her utility, holding the overall charitable contribution constant. To test the predictions of the model we ran an experiment in which we varied within-subjects the costs and benefits of monetary and volunteer donations. We also varied between-subjects the emphasis on the donation value to the charity (pure altruism) or the sacrifice to the donor (impure warm-glow altruism). Consistent with the model's predictions, the experiment shows that the pure altruism frame increases the efficiency of donation choices, substitutability of donations between money and time and crowding out. Nonetheless, while the greater impurity framing results in a more inefficient allocation of resources, empirically we find it increases overall charitable donations. We discuss the implications of our experimental results for both theory and policy.

Author(s): Andrew Lilley and Robert Slonim
Topic: Applied Economics: Charitable Giving, Decision Theory: Other
Back to session: Charitable Contributions 1


Taking, Giving, and Impure Altruism in Dictator Games

Edward Millner
Virginia Commonwealth University

ABSTRACT:

We show that, if giving is equivalent to not taking, impure altruism could account for List's (2007) finding that the payoff to recipients in a dictator game decreases when the dictator has the option to take. We examine behavior in dictator games with different taking options but equivalent final payoff possibilities. We find that recipients tend to earn more as the amount the dictator must take to achieve a given final payoff increases, a result consistent with the hypothesis that the cold prickle of taking is stronger than the warm glow of giving. We conclude that not taking is not equivalent to giving and agree with List (2007) that the current social preference models fail to rationalize the observed data.

Author(s): Oleg Korenok, Edward L. Millner, and Laura Razzolini
Topic: Applied Economics: Charitable Giving, Social Behavior: Other-regarding Preferences
Link: https://docs.google.com/file/d/0BzCJ3n5I9KJJa25lU0N5aVNKNEE/edit?usp=sharing
Back to session: Charitable Contributions 1


Transaction Costs and Inertia in Charitable Giving

Maros Servatka
Department of Economics, University of Canterbury

ABSTRACT:

This paper uses a laboratory experiment to analyse the effect that transactions costs and inertia have on charitable giving. We conjecture that transaction costs are likely to have a greater effect on donations if the solicitations are received when potential donors are busy (when the opportunity cost of time is high) as opposed to when they have time on their hands to donate (when the opportunity cost of time is low). If donations do not have to be made immediately, inertia could also become a factor as people might intend to give, but then postpone making the payment until they have more time, and having postponed making the donation once, keep doing so until the opportunity to donate has passed. We find that introducing a transaction cost to a standard Dictator Game with charity as a recipient, and manipulating whether the donation can be made when we know subjects have time on their hands, reduces donations providing evidence of a transaction cost effect. Some weak evidence is found that giving people more time to donate reduces donations, which is consistent with an inertia effect.

Author(s): Stephen Knowles and Maros Servatka
Topic: Applied Economics: Charitable Giving, Social Behavior: Other-regarding Preferences
Back to session: Charitable Contributions 1


The Role of Volunteer Leaders in Fundraising

Wei Zhan
TAMU

ABSTRACT:

Instead of making direct appeal to potential donors, charities often utilize volunteer leaders to make an appeal on their behalf. While these fundraising techniques are commonly used, there is little research on the mechanisms underlying their effectiveness. We design a series of lab experiment to study the role of leadership in fundraising, focusing on three dimensions: 1) how potential donors react to changes in information; 2) how potential donors react to the status of leaders; and 3) how potential donors react to the changes in social distance of the leaders. The study addresses both the aggregate effect of leadership and the variability among donors in responses to the variation in leader's role. Of particular interest is the interaction of the effectiveness of leaders and donors' motives for giving.

Author(s): Wei Zhan, Catherine Eckel
Topic: Applied Economics: Charitable Giving,
Back to session: Charitable Contributions 1


A Finite Mixture of Behaviors in the Care Game

Ericka Scherenberg Farret
University of Texas at Dallas

ABSTRACT:

We analyze the complexity of the interactions between employer and employee in a care work environment using data from a three person sequential game. We assume that individuals are motivated by different reasons that lead them to different responses in this setting and propose the estimation of a Bayesian finite mixture model to uncover the possible existence of sub-populations within the experimental sample and the parameters of interest. We found evidence of two sub-populations for the players. This technique allowed us to classify each observation into each sub-component and observe their patterns of behavior.

Author(s): Ericka Scherenberg-Farret, Catherine Eckel, Patrick Brandt
Topic: Applied Economics: Other, Applied Economics: Labor Market
Back to session: Charitable Contributions 2


The effect of feedback on performance of charity auctions.

Svetlana Pevnitskaya
Florida State University

ABSTRACT:

Auctions are commonly used to raise money for charities. It has been shown that in auctions, and other mechanisms, the types of feedback provided to participants can significantly affect behavior and therefore the outcome. We explore the effect of different types of feedback on the performance of charity auctions. We employ a lab-field experimental design where a portion of revenue raised in laboratory auctions is donated to actual charities chosen by participants. Subjects also have an option to make a donation to a charity of their choice. We find that the magnitude of funds raised, and the selection of charities, are affected by the type of feedback provided to subjects.

Author(s): Svetlana Pevnitskaya and Robert White
Topic: Applied Economics: Charitable Giving, Markets: Auctions
Back to session: Charitable Contributions 2


You Think You Are Bad? That's (Also) An Asset

Adrian Stoian
California State University East Bay

ABSTRACT:

Could the order of moral decisions influence those decisions? We look in particular at the case when a positive moral decision such as donating to Red Cross is interacted with a negative one such as cheating. In a simple model the order should not influence those decisions but in the model we discuss it could however play a role if people make inferences about themselves that in turn affect the latter decisions. The experimental evidence confirms this second line of thought: subjects cheat much more if they are first presented with the opportunity to cheat and donate less comparing with the subjects who are first asked to donate and then are given the opportunity to cheat. A negative self image pays off as it translates into higher monetary rewards.

Author(s): Doru Cojoc and Adrian Stoian
Topic: Applied Economics: Charitable Giving, Social Behavior: Lying and Cheating
Back to session: Charitable Contributions 2


Follow the leader with a 2-stage decision model

David Goldbaum
University of Technology Sydney

ABSTRACT:

A key feature of the follow-the-leader setting is that the universal benefit gained through cooperatively coordinating is undermined by competition-inducing payoff differentials. The conflicting incentives arise from the reward for selecting an option that is popular among the cohort of players with a premium for adopting it early. In order to generate a high payoff for everyone, a leader has to emerge to guild the choice of the remainder of the population. The competition arises from the premium paid to the leader, who adopts a choice in advance of the population of followers willingly waiting to learn of the leader's decision. Success requires that players take part in and respond to the endogenous evolution of the social network structure to bring about the equilibrium.

Author(s): David Goldbaum, AJ Bostian
Topic: Social Behavior: Group Behavior, Games: Networks
Back to session: Communication and Cooperation


The ABC of Cooperation and Framing Effects in Social Dilemmas

Simon Gaechter
University of Nottingham

ABSTRACT:

We study the role of heterogeneous social preferences in explaining framing effects in social dilemmas. We show that framing effects in give-some versus take-some public goods games can be decomposed into two components: (i) the effect of frames on beliefs and (ii) the effect on cooperation attitudes as elicited by the strategy method. We find that framing has a fairly strong effect on both components. We argue that the likelihood of observing significant framing effects in direct response contributions in either direction depends on the random sampling of the two components. Using a simulation approach we show that likely framing effects sizes are most of the times negligible and roughly normally distributed around zero. This explains the mixed results found in previous literature.

Author(s): Simon Gaechter, Felix Koelle, Simone Quercia
Topic: ,
Back to session: Communication and Cooperation


Spokesperson - to be or not to be?

Orsola Garofalo
Universitat Autonoma de Barcelona

ABSTRACT:

We study how the separation between making and communicating a choice affects the decision-maker's choice and the reaction of the individuals affected by that choice. In a modified dictator game, a decision-maker allocates a fair or unfair amount of money, and then decides whether to delegate or not to a spokesman the communication of the allocation chosen. Receivers can then decide to punish or not. First, we find that decision-makers choose unfair allocations more frequently when receivers can only punish by the same amount both the decision-maker and spokesman. Second, we find that decision-makers delegate the communication more frequently when receivers are free to assign whom to punish and the amount of punishment. Third, we find that, while receivers often punish unfair allocations, such behavior is unaffected by delegated communication.

Author(s): Orsola Garofalo and Christina Rott
Topic: Social Behavior: Communication,
Back to session: Communication and Cooperation


Participation and Spatial Coordination in Conservation Incentive Schemes: The Role of Transaction Costs and Communication

Tim Cason
Purdue University

ABSTRACT:

An Agglomeration Bonus (AB) is used in some environmental conservation incentive schemes to reward private landowners to spatially coordinate their land use decisions and enhance the supply of environmental benefits--ranging from habitat and species protection, water pollution control, and pollination services. The AB creates a coordination game with multiple Pareto ranked Nash equilibria corresponding to different spatially coordinated land use patterns. We experimentally analyze the levels of participation and performance when landowners have to incur a transaction cost prior to participating in the scheme. The experimental design varies (within-session) the transaction costs at two levels, which affects the risks and payoffs of coordinating on the different equilibria. We also implement a treatment involving non-binding pre-play communication between subjects. To study spatial coordination, all subjects are arranged on a circular local network with every individual having two strategic neighbors. The results indicate that communication substantially improves participation rates and coordination on the Pareto efficient Nash equilibrium under both transaction costs conditions. Without communication, participation in the scheme is greater with lower transaction costs, but conditional on participation subjects more frequently choose the efficient strategy that can lead to the AB with higher transaction costs. Subjects overall succeed in coordinating on the efficient outcome more often in the low transactions costs treatment, and a history of successful coordination raises subsequent coordination in either treatment.

Author(s): Simanti Banerjee, Timothy Cason, Frans de Vries and Nick Hanley
Topic: Games: Coordination, Games: Networks
Back to session: Communication and Cooperation


Commitment Problems in Conflict Resolution

Roman Sheremeta
Case Western Reserve University

ABSTRACT:

Commitment problems are inherent to non-binding conflict resolution mechanisms, since an unsatisfied party can ignore the resolution and initiate conflict. We provide experimental evidence suggesting that even in the absence of binding contractual agreements individuals often avoid conflict by committing to the outcome of a conflict resolution mechanism. Commitment problems are mitigated to a greater extent for groups who opt-in to the conflict resolution mechanism, but only when opting-in is costly. Although conflict rates are higher when opting-in is costly than when it is free or exogenously imposed, commitment problems are greatly reduced amongst those groups who choose to opt-in.

Author(s): Roman M. Sheremeta
Topic: Games: Contests,
Back to session: Conflict and Contests


Agent Bargaining

William Ingersoll
University of Arizona

ABSTRACT:

Negotiations in the real world can rarely be represented by a simple bargaining session between two parties. Agent bargaining, when one player represents another party in a bargaining situation for some form of compensation, is one such complicating circumstance from the real world. We are interested in exploring the effects that this third entity has on the outcome of the negotiations. We construct a laboratory experiment emulating a simple example of agent bargaining. We test a hypothesis formulated using sequential-Nash-bargaining and also propose behavioral explanations for the observed behavior. We find that sequential-Nash-bargaining does a poor job explaining our observations, and that using a weighted minimization of the differences between each of the three parties as a focal point fits the observations better.

Author(s): William Robert Ingersoll and Alex Roomets
Topic: Games: Bargaining,
Back to session: Conflict and Contests


An Experimental Analysis of Asymmetric Power in Conflict Bargaining

Charles Holt
University of Virginia

ABSTRACT:

Abstract: Demands and concessions in a multi-stage bargaining process are shaped by the probabilities that each side will prevail in an impasse. Standard game-theoretic predictions are quite sharp: demands are pushed to the precipice with nothing left on the table, but there is no conflict regardless of the degree of power asymmetry. Indeed, there is no delay in reaching an agreement that incorporates the (unrealized) costs of delay and conflict. A laboratory experiment is used to investigate the effects of power asymmetries on conflict rates in a two-stage bargaining game that is (if necessary) followed by conflict with a random outcome. Observed demands at each stage are significantly correlated with power, as measured by the probability of winning in the event of disagreement. Demand patterns, however, are flatter than theoretical predictions, and conflict occurs in a significant proportion of the interactions, regardless of the degree of the power asymmetry. To address these deviations from the standard game-theoretic predictions, we also estimate a logit quantal response model, which generates the qualitative patterns that are observed in the data. This one-parameter generalization of the Nash equilibrium permits a deconstruction of the strategic incentives that cause demands to be less responsive to power asymmetries than Nash predictions.

Author(s): Charles Holt, Katri Sieberg, David Clark, Tim Nordstrom, and William Reed
Topic: Games: Bargaining,
Link: http://www.mdpi.com/2073-4336/4/3/375
Back to session: Conflict and Contests


Effects of Endogenous Selection of Nested PD games on Pro-social behavior and outcomes

Malcolm Kass
University of Texas - Dallas

ABSTRACT:

A small but growing strand of research investigates the pro-social effects of democracy. In these studies, subjects select the institution in which they interact using a collective choice mechanism such as voting. Typically subjects choose among punishment or communication institutions in a baseline social dilemma structure, and the results point to greater cooperation when institutions are endogenously selected. However, most of the literature addresses institutions that alter the equilibrium predictions, leading to the idea that this social welfare-enhancing effect of democracy will only be effective when the subject's choice set includes institutions that serve to align personal and group incentives.

Author(s): Malcolm Kass
Topic: Social Behavior: Other, Public Choice: Voting and Rent Seeking
Back to session: Cooperation and Institutions 5


Legitimate Authority, Procedural Fairness, and Communication: A Public Goods Experiment

Eric Dickson
New York University

ABSTRACT:

Enforcing the law is a primary function of governments, but regimes vary markedly in the extent to which their activities are viewed as legitimate by the general population. Using a series of public goods experiments, we assess the dynamics of legitimacy and citizen-regime cooperation and conflict. In one study, we explore the effects of an authority's investment in accurate information on her legitimacy, exploring the roles of both procedural and outcome-based fairness. In a second study, we allow group members to rat out their counterparts for failing to contribute to a public good, and measure variation across institutional treatments in the extent to which group members are willing to tattle to the authority.

Author(s): Eric Dickson, Sanford Gordon, and Gregory Huber
Topic: Public Choice: Public Goods and Common Pool Resource, Social Behavior: Norms and Morals
Back to session: Cooperation and Institutions 1


Boycotting and Buycotting: Third-Party Punishment and Reward in Groups

Dimitry Mezhvinsky
The Ohio State University

ABSTRACT:

This study experimentally demonstrates that buyers engage in boycotting and buycotting at a cost to themselves, in order to respectively punish and reward a seller for policies of which the buyers are neither the victims nor the beneficiaries. Notably, the experimental results show that third-party punishment (see Fehr and Fischbacher, 2004) and reward occurs even when the mechanism for executing such punishment and reward is not straightforward and when abstaining from punishment and reward may be viewed as the neutral or default choice by the third party, rather than the extreme choice as in other studies of third-party norm enforcement. This study finds that the average buyer does not condition her boycotting and buycotting decision on the actions of other buyers. This finding suggests that information revelation, rather than conformity preferences, may be the primary driving force behind the conditional cooperation observed in other studies of pro-social behavior. Those buyers who condition their buycotting decisions on others buyers' actions do so in one of two ways, each of which is predictive of behavior in other buycotting settings. These patterns of behavior may be explained by differences in buyers' aversion to different types of guilt; that is, differences in the manner in which buyers wish to avoid disappointing others.

Author(s): Dimitry Mezhvinsky
Topic: Social Behavior: Norms and Morals, Social Behavior: Other-regarding Preferences
Back to session: Cooperation and Institutions 1


Cooperation under Third-Party Rent Extraction

Tobias Cagala
University of Erlangen-Nuremberg

ABSTRACT:

Despite its prevalence, the influence of third parties on public good contributions has attracted little attention in the experimental literature. We study the effect of a third party administrator who can extract rents from the pool of public good contributions. Our design identifies the total effect of positive and negative rent extraction and disentangles the channels through which contributions are affected. We find that, in line with theory, positive rent extraction increases cooperation while negative rent extraction does not affect cooperation. The main channels through which rent extraction affects cooperation are inequality aversion and reciprocity.

Author(s): Tobias Cagala, Ulrich Glogowsky, Veronika Grimm, Johannes Rincke
Topic: Public Choice: Public Goods and Common Pool Resource, Social Behavior: Other-regarding Preferences
Back to session: Cooperation and Institutions 1


To punish or withhold? Implicit versus explicit sanctions in a risky social dilemma: evidence from Kamchatka, Russia.

E Lance Howe
University of Alaska, Anchorage

ABSTRACT:

In many small-scale societies, households cooperate across a range of activities, from hunting and herding to food sharing and other activities that pool risk. Incentives to cooperate vary across activity type and local institutions often arise that reinforce pro-social behavior in critical domains. In some remote rural regions, particularly in Western Alaska and Far East Russia, local institutions rarely include any explicit sanctions for non-cooperative or antisocial behavior.Rather, implicit punishment is more commonly observed. To better understand why these norms have developed we design an experiment to test the effect of implicit versus explicit sanctions on different types of cooperation. In July 2013 we conducted a modified public goods game in two remote rural communities of Kamchatka Russia where households are known to cooperate in different domains (e.g.fishing, hunting, and food sharing). The experiment builds on a standard public goods game but introduces idiosyncratic risk and an opportunity to share. Subjects therefore have an opportunity to cooperate in a team production domain and a risk pooling domain. We look at the use and effects of explicit versus implicit sanctions on cooperation in this environment. Consistent with the ethnography of the region, we find subjects are reluctant to use explicit punishment. The related effects on cooperation in the social dilemma vis-a-vis cooperation in risk pooling will discussed.

Author(s): Lance Howe, James Murphy, and Olga Bogach
Topic: Public Choice: Public Goods and Common Pool Resource, Applied Economics: Economic Development
Back to session: Cooperation and Institutions 1


Endowment Heterogeneity and Informal Sanctions in a Public Goods Experiment

David Kingsley
University of Massachusetts Lowell

ABSTRACT:

This paper reports on a public goods experiments which interacts endowment heterogeneity and informal sanctions. Using a model of inequality aversion the hypothesis tested is whether the disadvantaged, low endowment, group members will successfully enforce higher contributions to the public good from the advantaged, high endowment, group members in order to lessen inequality. Results suggest that despite heavy use of the informal sanction high endowment group members do not contribute more to the public good. This is in contrast to the increased contributions and welfare of group members in a homogenous endowment informal sanction treatment. As a result the welfare of all group members in the heterogeneous endowment informal sanction treatment is substantially lower than baseline measures. This result suggests that informal sanctions, which are effective when endowments are homogenous and social norms clear, are ineffective and actually worsen welfare when endowments are heterogeneous and social norms less clear.

Author(s): David C. Kingsley
Topic: Social Behavior: Other-regarding Preferences, Public Choice: Public Goods and Common Pool Resource
Back to session: Cooperation and Institutions 2


Investing in Institutions for Cooperation

Alexander Smith
Worcester Polytechnic Institute

ABSTRACT:

In conventional repeated public good games, the productivity of the public good (the MPCR) is chosen by the experimenter. In this paper, I report the results of an experiment in which subjects choose how much to invest in the productivity of the public good, and then how much to contribute to the provision of the public good. Preliminary results suggest that subjects are initially willing to make significant investments in the productivity of the public good in spite of the fact that some subjects free ride in the contribution stage. Over time, the free riding behavior erodes investment in productivity, suggesting that one consequence of poor cooperation is decreased investment in the institutions that facilitate cooperation among individuals.

Author(s): Alexander Smith
Topic: Public Choice: Public Goods and Common Pool Resource, Public Choice: Voting and Rent Seeking
Back to session: Cooperation and Institutions 3


Does external regulation crowd out cooperation?

Anna Lou Abatayo
University of Hawaii at Manoa / East-West Center

ABSTRACT:

It is widely believed that there is strong experimental evidence to support the idea that externally imposed regulations crowd out the intrinsic motivations of common pool resource users to cooperate. We identify an important confound in previous experimental work on crowding out effects and design a series of experiments to disentangle what we believe is a communication effect from the effect of whether a regulation is endogenously created or exogenously imposed. When we impose the exact same regulations on subjects either endogenously or exogenously, we observe no differences in cooperation. However, when we add communication to our endogenous regulation treatment, we observe significant behavioral differences between the externally imposed and endogenously chosen regulations treatments. Our results strongly suggest that prior experimental evidence for regulations crowding out cooperation is actually driven by communication crowding in cooperation.

Author(s): Anna Lou Abatayo and John Lynham
Topic: Public Choice: Public Goods and Common Pool Resource, Social Behavior: Communication
Back to session: Cooperation and Institutions 5


Uncertainty in Payoff-Equivalent Appropriation and Provision Games

Brock Stoddard
University of South Dakota

ABSTRACT:

It remains an open empirical question whether social dilemmas are more serious in appropriation games (such as common-pool-resource extraction games) or provision games (such as public-goods games). This study examines this question in a laboratory experiment that compares cooperative behavior within six pairs of payoff-equivalent appropriation and provision games. Using a menu design, subjects make decisions in multiple one-shot settings, without receiving feedback on others' actions until all decisions are finalized. Probabilistic uncertainty is also examined across game types. In some game pairs, the individual return or the group return is drawn from a discrete uniform probability distribution, after all decisions are finalized. In other game pairs, the individual and group returns are deterministic. Finally, in some game pairs the pecuniary incentives are high while in other game pairs the pecuniary incentives are low. The experimental data reveal that average behavior is not significantly different across each pair of appropriation and provision games. However, compared to the respective certain conditions, an uncertain individual return has a stronger effect on raising average contribution decisions in provision games than in appropriation games.

Author(s): Brock Stoddard
Topic: Public Choice: Public Goods and Common Pool Resource, Social Behavior: Group Behavior
Back to session: Cooperation and Institutions 3


Setting the Bar - An Experimental Investigation of Immigration Requirements

Menusch Khadjavi
University of Hamburg

ABSTRACT:

In the face of poverty-motivated migration streams from developing countries as well as demand of and competition for highly qualified talents, migration is one of the pressing issues worldwide. It however remains unclear what perceptions and procedures shape efficient migration policy. We use a public good game experiment with endogenous groups to investigate how perceived out-group potential (and associated bargaining power) and debate among in-group members shape immigration requirements. We employ the minimal group paradigm and immigration requirements are set by in-group voting. Our results show that debate appears to foster fair and efficient requirements if perceived migrant potential is high. Conversely, debate in the perceived low-potential case leads to the tightest requirements, so that fewer out-group subjects immigrate. We discuss associated migration policy implications.

Author(s): Menusch Khadjavi, Jasper D. Tjaden
Topic: Public Choice: Public Goods and Common Pool Resource, Social Behavior: Group Behavior
Back to session: Cooperation and Institutions 3


Punishment with Uncertainty in Public Goods Experiment

Qilin Zhang
The Hong Kong Polytechnic University

ABSTRACT:

In stranger-matching public good experiments with punishment that has 50% chance to fail, we study how uncertainty affects both peer and third- party first-round monetary punishment, which in turn affect cooperation, contribution and overall welfare. Data displays that punishment frequency under uncertainty changes little, indicating uncertainty dose not affect punisher's willing to punish; robust test suggests this is because uncertainty doesn't change people's preferences that trigger punishment. However, data shows punishment severity under uncertainty decreases by about 25%, indicating that risk-averse attitude plays a negative role in people's decision of how severely they punish. This is also predicted by our social-preference-based prospect model. Further, our study shows that although uncertainty decreases pro-social punishment, it also decreases anti-social punishment. Overall, because of the decrease in punishment that reduces the cost of punishing and because of the unchanged-expected-punishment-severity technique employed in our experiment which prevents cooperation and contribution from decaying, social welfare increases. This suggests that we can reach same social welfare simply by increasing punishment severity yet at a lower level of cost of punishment enforcement resource.

Author(s): Qilin Zhang, Yan Zhou, Peiran Jiao
Topic: Public Choice: Public Goods and Common Pool Resource,
Back to session: Cooperation and Institutions 3


Debt and (Future) Taxes: Financing Intergenerational Public Goods

J. Forrest Williams
Texas A&M University

ABSTRACT:

Many public goods inherently involve an intergenerational component. For example, groups fund projects to enhance environmental quality that help them; those projects also impact the group's children's livelihood. Additionally, most public projects have two funding sources: current and future incomes (e.g., taxes and debt). I construct a public goods experiment in which an agent's contribution to the public good not only impacts her welfare, but also impacts future agents who play the game. In one treatment she may borrow from this future player. I find that, adding debt-financing as a means of funding the public good increases contributions to the project, yielding higher endowments for future agents. These gains are, however, offset by debt repayment so that economic growth is not possible for future generations without additional debt.

Author(s): J. Forrest Williams
Topic: Public Choice: Public Goods and Common Pool Resource, Public Choice: Other
Back to session: Cooperation and Institutions 4


Verbal Feedback and Punishment in Public Good Games.

Rulliere Jean-Louis
University of Lyon

ABSTRACT:

This paper investigates the potential use of verbal communication as a feedback tool to sustain cooperation in public good games. If our daily lives tell us that it is natural to reward or punish others using communication, very few researchers have actually tackled this problem. Even less studied are both the impact of verbal feedback on the dynamics of decision making and the emotional mechanisms involved in these dynamics. Because of its high emotional load and low cost compared to monetary punishment, we saw verbal feedback as a potentially successful mechanism in order to raise prosocial behavior.

Author(s): Benjamin PELLOUX, Jean-Louis RULLIERE and Frans van WINDEN
Topic: Public Choice: Public Goods and Common Pool Resource, Social Behavior: Communication
Back to session: Cooperation and Institutions 4


Voluntary and Mandatory provision of Common Pool Resources with Heterogeneous appropriators

Fatemeh Momeni
Purdue

ABSTRACT:

This paper studies the extraction behavior of heterogeneous common pool resource users under different provision mechanisms in a laboratory setting. In the first two treatments, the provision rule is exogenously determined: In regressive treatment poor and rich individuals pay equal amount of their wealth to the provision of the resource while in progressive treatment the entire provision cost is paid by rich members. Finally, in the third -endogenous- treatment players voluntarily contribute to cover the provision costs through a Provision Point Mechanism.

Author(s): Fatemeh Momeni
Topic: Public Choice: Public Goods and Common Pool Resource, Social Behavior: Other-regarding Preferences
Back to session: Cooperation and Institutions 4


Effects of Threshold Uncertainty on Common Pool resource Depletion

Sara Adler
UCSB

ABSTRACT:

Many natural and common-pool resources have intrinsic thresholds for environmental consequences. In addition, governments, in order to achieve conservation of these common pools, create their own targets, which may be sporadically enforced or readjusted over the life of the resource. This paper examines both group and individual behavior in an experiment designed to model common pools in which there exist such thresholds for consequences by using three different games. Results demonstrate that since the true threshold is often unknown to policy makers, and guesses are not permanently and irrevocably established, it was best to allow the participants to develop their own beliefs regarding the value of the threshold in order to maximize the life of the resource. Revealing the true threshold provided no significant gains in terms of the life of the resource. Once individuals passed the threshold, greediness ensued, displayed by participants withdrawing the maximum number of tokens. The results of this study will have implications for evaluating and establishing governmental resource policies for water and other common pools and will anchor future applications of common pool resource depletion experiments which examine the behavioral decision making process.

Author(s): Sara E Adler
Topic: Public Choice: Public Goods and Common Pool Resource,
Back to session: Cooperation and Institutions 4


Cooperation among heterogeneous agents

Claudia Keser
The University of Gottingen

ABSTRACT:

I will present a series of experiments, in which we investigate whether and how unequal endowment distributions affect cooperation in repeated linear VCM experiments. In keeping with Hofmeyer, Burns and Visser (HBV, 2007), we observe that moderate asymmetry does not change the overall contribution level. Players tend to provide the same fraction of their endowments. HBV call this the fair-share rule. However, when we create an asymmetry important enough to render the social optimum unattractive to the super rich, we observe a decrease in the overall contribution level. The rich then do not play according to the fair-share rule, but make contributions at the same level as the poor. They appear to seek a constraint social optimum, in which everybody makes the same contribution. I will then demonstrate how, in situations with moderate asymmetry, we can use minimum-contribution requirements to change the cooperation norm and achieve higher group contribution levels than under symmetry.

Author(s): Andreas Markstadter, Martin Schmidt
Topic: ,
Back to session: Cooperation and Institutions 2


Justification and Legitimate Punishment

Fangfang Tan
Max Planck Institute for Tax Law and Public Finance

ABSTRACT:

Punishment can lose its legitimacy if the enforcer can profit from delivering punishment. We use a controlled laboratory experiment to examine how justification can combat profit-seeking punishment and promote the legitimacy of punishment. In a one-shot sender-receiver game, an independent third party can punish the sender upon seeing whether the sender has told the truth. Most third parties who can profit from punishment punish the senders regardless of how the senders. However, when they have to provide explanations for their punishment decisions, majority third parties punish the sender if and only if the sender lies. Our data also suggest that senders are more likely to perceive punishment as legitimate and behave honestly when they know the enforcer has to justify the punishment decision. Our findings suggest that the justification requirement plays an important role in building effective punishment institutions.

Author(s): Erte Xiao (Carnegie Mellon University) and Fangfang Tan (Max Planck Institute for Tax Law and Public Finance)
Topic: Social Behavior: Norms and Morals, Social Behavior: Lying and Cheating
Link: http://mpra.ub.uni-muenchen.de/47154/
Back to session: Cooperation and Institutions 5


Cooperation and punishment among heterogeneous agents

Martin Schmidt
University of Gottingen

ABSTRACT:

We investigate the effect of the possibility to punish individual others in an asymmetric public-good experiment. One player with a low endowment and two players with a high endowment make voluntary contributions over 25 periods. We focus on contributions as well as punishment behavior by both player types. In a baseline treatment without punishment, we observe that players tend to contribute the same percentage of their endowments. This implies that the players with a high endowment contribute significantly more than those with a low endowment. Furthermore, we observe the typical downward trend in group contributions.

Author(s): Claudia Keser and Martin Schmidt
Topic: Public Choice: Public Goods and Common Pool Resource,
Back to session: Cooperation and Institutions 2


Second Thought: Theory and Experiment in Social Dilemma

Tatsuyoshi Saijo
ISER, Osaka University

ABSTRACT:

Why have we been using second thought? This paper shows that second thought is not an innocent device in our daily life, but is human wisdom that plays an important role to resolve problems such as social dilemma. We design a simplest possible mechanism to achieve Pareto efficiency, and then compare the performance of the mechanism with and without second thought. The mechanism with second thought performs very well in theory and experiment. Moreover, the mechanism with second thought is robust against deviation from rational choices.

Author(s): Tatsuyoshi Saijo and Yoshitaka Okano
Topic: Public Choice: Public Goods and Common Pool Resource,
Back to session: Cooperation and Institutions 5


Noisy best response in coordination games: strategic vs. payoff uncertainty

Daniel Saunders
University of California, Santa Barbara

ABSTRACT:

In an environment with strategic interaction, multiple equilibria yields a coordination problem with strategic uncertainty, even in the idealized setting of complete information. This uncertainty has proved substantive in practice, not a mere artifact of theory, as demonstrated by several laboratory experiments in which it leads to coordination failure. This paper examines two such environments: the threshold game with a unitary public good and the market entry game with a finite capacity. I employ Logit Quantal Response Equilibrium to explicitly model strategic uncertainty, and use the limiting properties of LQRE to resolve the equilibrium selection problem. The model predicts that the strategic uncertainty generated by noisy best responders has greater influence on group-level outcomes than payoff uncertainty when agents are highly boundedly rational. I conduct a two-by-two experiment for each game and estimate the logit parameter, varying group size and information. The dominance of strategic uncertainty bears out empirically, and LQRE is able to capture certain group size and player type treatment effects, upon which standard theory is silent.

Author(s): Daniel Saunders
Topic: ,
Back to session: Coordination


Cooperation and Trustworthiness in a repeated Interaction

Ulrich Glogowsky
University of Nuremberg?Erlangen

ABSTRACT:

Cooperation in public good provision often depends on the trustworthiness of an administrator who may reduce contributors' returns from cooperation to her own benefit, and who herself may respond to the degree of cooperation among contributors. This paper analyses the interdependencies between cooperation and trustworthiness in a repeated game that replaces the mechanical public good administration in the Public Good Game with a human administrator. We present a new approach for visualising conditional behaviour in repeated sequential games. Our approach identifies the causal effects of cooperation on trustworthiness and vice versa by combining standard methods from time-series analysis with a design based identification strategy. We find that contributors and the administrator strongly respond to each other, resulting in cooperation and trustworthiness being strongly mutually interdependent. Furthermore, cooperation and trustworthiness are rather driven by changes in cooperation than by changes in trustworthiness.

Author(s): Tobias Cagala, Ulrich Glogowsky, Veronika Grimm, Johannes Rincke
Topic: Public Choice: Public Goods and Common Pool Resource, Methodology: Experimental Design, Tools and Practices
Back to session: Coordination


Prominence and Leadership in a Minimum Effort Game

Doru Cojoc
University of Iowa

ABSTRACT:

We examine in the laboratory whether a prominent group member can help groups overcome their coordination difficulties in a repeated minimum effort game. Subjects are paired in groups of 6 and play multiple rounds of the minimum effort game with fixed matching. In one treatment, groups have access to ex-post monitoring. In the other, individuals can also observe in real-time the choices of one prominent member, who in turn can observe the choices of all group members.

Author(s): Doru Cojoc
Topic: Games: Coordination, Social Behavior: Group Behavior
Link: https://dl.dropboxusercontent.com/u/11212329/Coordination%20with%20Prominent%20Members.pdf
Back to session: Coordination


Competition, cooperation, and collective choice

Ernesto Reuben
Columbia University

ABSTRACT:

The ability of groups to implement efficiency-enhancing institutions is emerging as a central theme of research in economics. This paper explores voting on a scheme of intergroup competition which facilitates cooperation in a social dilemma situation. Experimental results show that the competitive scheme fosters cooperation. Competition is popular but also that the electoral outcome depends strongly on specific voting rules of institutional choice. If the majority decides, competition is almost always adopted. If likely losers from competition have veto power, it is often not, and substantial gains in efficiency are foregone.

Author(s): Thomas Markussen, Ernesto Reuben, and Jean-Robert Tyran
Topic: Public Choice: Public Goods and Common Pool Resource, Public Choice: Voting and Rent Seeking
Back to session: Coordination


On Measuring Ambiguity Aversion

Charles Sprenger
UCSD

ABSTRACT:

Our objective in the present study is to return to the consistency requirements Subjective Expected Utility (SEU) in order to develop deeper insights into individual decisionmaking and potentially provide identification of competing non-SEU models. In addition to consistency a l'Ellsberg between two subjective events, Subjective-Subjective Consistency, we consider Subjective-Objective Consistency. When considering mixtures between subjective and objective event bets a decisionmaker must also behave as if a single probability measure governed her choice. The distinction between Subjective-Subjective Consistency and Subjective-Objective Consistency is important for two reasons. First, of the two popularized non-SEU models we discuss, both deliver Subjective-Subjective inconsistencies while only one maintains Subjective-Objective Consistency. Hence investigating Subjective-Objective Consistency allows for separation that a standard Ellsberg experiment cannot deliver. Second, though Ellsberg style experiments are frequent in the literature, questions testing Subjective-Objective Consistency have yet to be implemented experimentally. Particular forms of violations may lead to interesting insights on the structure of decision making when subjective ambiguity can be replaced with objective uncertainty.

Author(s): James Andreoni, Tobias Schmidt, Charles Sprenger
Topic: Decision Theory: Ambiguity, Decision Theory: Risk
Back to session: Decision Making Under Ambiguity


The effect of ambiguity on preferences over gains and losses: Experimental evidence

Oscar Zapata
University of Calgary, Canada

ABSTRACT:

Many of the choices individuals make are best characterized as decision making under (Knightian) uncertainty or ambiguity as opposed decisions made under risk. For example, choices about mitigation and adaptation to climate change are made under ambiguity due to the limited and incomplete knowledge about climate patterns and the likelihood of extreme events. Similarly, choices about personal protection are made without knowing the true probabilities surrounding personal risks from criminal activity. In this paper, we explore how individuals make use of the available information in risky and ambiguous environments over gains and losses. In decision environments with potential gains, we find that individuals display ambiguity aversion and are less willing to bear risk in environments where probabilities are unknown. However, we find that in decision environments involving losses, individuals display loss aversion that outweighs the effects of ambiguity versus risk. Moreover, we identify errors in probability estimates in ambiguous situations in which errors increase when individuals face losses.

Author(s): Robert Oxoby and Oscar Zapata
Topic: Decision Theory: Ambiguity, Decision Theory: Risk
Back to session: Decision Making Under Ambiguity


Unintended Hedging in Ambiguity Experiments

Alex Roomets
Heidelberg University

ABSTRACT:

We point out a potential hedging problem in standard Ellsberg-type experiments. This hedging problem may yield an incorrect classification of ambiguity averse subjects as expected utility-maximizers (or ambiguity loving)or vice versa. We demonstrate that subjects in these experiments can choose combinations of ambiguous options in a way that eliminates all ambiguity in expected payoffs, and we propose a new classification strategy for asymmetric Ellsberg urn that allows the identification of ambiguity averse subjects who employ hedging

Author(s): Joerg Oechssler and Alex Roomets
Topic: Decision Theory: Ambiguity, Methodology: Experimental Design, Tools and Practices
Back to session: Decision Making Under Ambiguity


No Two Experiments are Identical

Yoram Halevy
University of British Columbia

ABSTRACT:

We study choice between bets on the colors of two balls, where one ball is drawn from each of two urns. Though you are told the same about each urn, you are told very little, so that you are not given any reason to be certain that the compositions are identical. We identify choices that reveal an aversion to ambiguity about possible differences between urns, thus identifying a source of uncertainty different from the usual Knightian distinction between risk and ambiguity. Choice behavior is studied in a controlled high-stakes laboratory experiment, and the ability of new and existing models to rationalize the experimental findings is examined.

Author(s): Larry G. Epstein and Yoram Halevy
Topic: Decision Theory: Ambiguity,
Back to session: Decision Making Under Ambiguity


Do Cab Drivers Charge for Congestion? A Traffic Field Experiment in Lima, Peru

Hernan Bejarano
Chapman, Economic Science Institute

ABSTRACT:

Little is know about how accurately individuals perceive congestion and how their perceptions affect their routing and commuting decisions. However, unregulated taxi systems where taxi drivers and customers bargain over trip fares make it possible to study how individuals perceive and bargain over splitting uncertain congestion costs. The assumption is that negotiated fares are a function of expected travel times and thus provide relevant information for choosing routes and departing times. This paper uses a naturally occurring field experiment in Lima, Peru to analyze if taxi drivers attempt to transfer congestion costs to customers. The experiment involved collaborators hailing taxis and negotiating fares at different times of day between specific locations with alternative route choices as treatments. A total of 1100 trips were conducted following the same bargaining protocol and the randomly assigned route treatments, from which taxi fares, travel times and routing decisions were collected. Results show that taxi drivers request fares that discriminate between routes that take longer travel times, but do not request enough money to prevent earnings per minute from decreasing during congested periods. Fares for congested trips are one third lower than metered fares constructed estimates. Insufficiency of fares increment during peak collaborators hours could be generated by either highly competitive market conditions, failure to perceive longer travel times or a combination of both. Customers waiting time and proportions of taxis over total traffic support the idea that drivers might perceive peak traffic hours as a competitive market therefore transferring only partially cost of congestion into fares. These results highlight that unregulated taxi systems are successful at generating incentives for taxi drivers to make optimal routing decisions. However, they fail to improve traffic miscoordination because taxi driver misperceptions are left uncorrected. Real time routing and taxi market information technology could be used to improve the overall use of traffic network capacity.

Author(s): Hernan Bejarano
Topic: Field Experiments: General, Applied Economics: Other
Back to session: Environmental 1


An Experimental Study of a Common Property Renewable Resource Game in Continuous Time

Dina Tasneem
McGill University

ABSTRACT:

Fisheries, forests, and water all share the common characteristics of valuable common-pool renewable resources, which are integral to any economy.I experimentally study a differential game, in which a stock of resource replenishes itself at a constant proportional rate. Two players simultaneously extract the resource as a private good in real time. In this differential game, there are multiple equilibria, where the rate of extraction is either a linear or a non-linear function of the stock level. The theory thus predicts multiple Pareto ranked non-cooperative Markovian equilibria that can arise from different levels of aggressiveness of extraction, and that result in different levels of steady state resource stock and long-run welfare. My main contribution is to provide an empirical basis for equilibrium selection in this game which I experimentally run in real time. Of particular interest is the linear strategy, which is most frequently studied and cited in the literature: it is the most efficient non-cooperative equilibrium strategy in this model. I nd that a substantial proportion of pairs of subjects exhibit cooperative behaviour in the sense that they reach the best steady state in a completely non-cooperative setting. I also nd evidence that subjects condition their strategies on the state variable. However, a detailed examination of the extraction strategies does not provide evidence for the linear Markovian strategy. In fact, strategies exhibit a wide range of heterogeneity, with many di erent steady states observed in the data. The steady-state stock level is positively a ected by the initial stock level, a fact upon which the theory is silent. My experiment is the first to empirically test this important class of model in real time in the experimental laboratory.

Author(s): Hassan Benchekroun, Jim Engle-Warnick and Dina Tasneem
Topic: Games: Other, Public Choice: Public Goods and Common Pool Resource
Back to session: Environmental 1


Driving Responses and Opinions: How Do Drivers React to Road Pricing?

Rami Harb
Atkins NA

ABSTRACT:

Congestion pricing is a candidate method for managing congestion and for compensating for scarce transportation funding. However, lack of public acceptance and political support of road pricing policies are still major hurdles to their implementation. This study is the first of its kind to collect data on actual route choices in two traffic regions (Atlanta and Orlando) based on numerous, controlled and randomly assigned, relative route prices. Based on analysis of this data it is clear that road pricing works in the sense that drivers react to the price changes and choose the relatively less costly route. We further collect survey data from the same respondents on opinions towards various pricing policies such as taxes, fixed tolls, dynamic tolls, as well as conversion to carpool lanes with no additional taxes or tolls. We explore the relationship between individuals' route choices and their opinions towards congestion pricing policies. Finally, this paper evaluates the effect on participants' opinions of taking part in this study. We find that as the relative cost of using the expressway decreases, more and more low income drivers select the expressway, approaching the propensity of higher income drivers. We also find that the inclination to express support for traffic congestion solutions is correlated with the propensity to use Expressways. We do not find any evidence that the particular preferences expressed over various proposed solutions is related to driving behavior, however. On the other hand, respondents change their opinions as they participate in our study, and in response to major events in their local communities.

Author(s): Rami Harb PhD, Elisabeth Rutstrom PhD
Topic: Field Experiments: General,
Back to session: Environmental 1


Experiments in Environmental Service Provisioning Mechanisms with Spatially Explicit Externalities: An Application in Coastal Infrastructure

Jacob Fooks
University of Delaware

ABSTRACT:

The design and implementation of public programs that provide compensation for the voluntary provision of public goods by landowners has a long and rich literature. Of recent interest has been the design of mechanisms that specifically address non-linearities in underlying natural system dynamics. Sea level rise, coastal inundation, and protective infrastructure development is a setting that has received considerable recent attention where these issues are of particular interest. In the aftermath of Hurricane Sandy there have been a rash of policy debates and legal suits related to investment in sea walls, dikes, and dune restoration. Addressing these issues calls for new policy tools to align incentives of landowners with heterogeneous costs and incentives. To this end, this research testbeds several public good provisioning mechanisms, including two variations reverse Vickrey-Clarke-Groves mechanisms and a machine learning based Adaptive Mechanism which tries to infer and react optimally to subjects' decision heuristics. Experiments are conducted in a spatially explicit public good setting, with participants framed as coastal property managers with the opportunity to invest in protective infrastructure. Externalities from this protection propagate spatially according to shallow wave kinematic routing equations commonly used by government agencies in flood modeling.

Author(s): Jacob R Fooks
Topic: Public Choice: Public Goods and Common Pool Resource, Markets: Auctions
Back to session: Environmental 2


Evaluating Common Pools to Mitigate Shutdown and Market Risk in Multi-output Quota Markets with Stochastic Production Ratios

Christopher Anderson
University of Washington

ABSTRACT:

Tradable output permits are being introduced into increasingly complex regulatory environments, including pollution production and fisheries where multiple regulated products are produced. In some cases, the production ratios among products are highly stochastic-due to emission equipment failure or fish species intermixing-making it hard for users to predict their quota needs. In our motivating field case, the West Coast US groundfish fishery, some rockfish species have total quotas that amount to less than one fish per vessel, and are encountered randomly (without little ability to avoid) on 2-5% of fishing trips; many harvesters will not encounter one at all during a year, but a harvester who encounters one early faces significant shutdown risk or quota cost. Harvesters in this fishery have, paradoxically, chosen to create a common pool of their individually allocated quota for these limiting species, rather than face market risk for buying quota if they need it. We explore conditions under which quota holders might prefer this ??risk pool?? arrangement using a novel quasi-continuous experimental common pool environment. Subjects play the role of fishermen in a two-species fishery, choosing whether to fish in each week of a 52-week fishing season with a contemporaneous price externality and a time-varying probability of encountering a limiting species. We find risk pools can improve outcomes over quota markets for infrequently encountered and severely limiting species, particularly if harvesters are uncertainty averse, or able to even slightly improve on Nash outcomes for pooled quota.

Author(s): Christopher M. Anderson, Thomas W. Sproul, Hirotsugu Uchida
Topic: Public Choice: Public Goods and Common Pool Resource, Markets: Market design and Matching
Back to session: Environmental 2


Price Controls and Banking in Emissions Trading Programs

James Murphy
University of Alaska Anchorage

ABSTRACT:

We present results from laboratory emissions markets designed to investigate the effects of price controls and permit banking on limiting permit price risk. While both instruments reduce between-period price volatility and within-period price dispersion, combining price controls and permit banking yields important benefits even though the price controls are not activated very often. The combined policy produces lower price volatility than either policy alone and lower price levels than banking alone. However, banking, price controls, and the combination all produce higher between-period emissions volatility. Hence, for emissions markets that seek to control flow pollutants with strictly convex damages, efforts to limit permit price risk can result in higher expected damages.

Author(s): John K Stranlund, James J Murphy & John M Spraggon
Topic: Applied Economics: Other, Markets: Other
Back to session: Environmental 2


Game Form Representation and Knowledge Spillovers

Kory Garner
Purdue University

ABSTRACT:

This paper looks at knowledge spillovers in a simple game. The game is a competition between two players and has multiple ways in which it may be presented. An experiment is conducted where some subjects are given an easier game form while others are given a more difficult form. We find that when subjects are given the more difficult game form representation they make fewer mistakes in later rounds if they have been playing against an individual with the easier representation.

Author(s): Karthik Kannan, Sandra Maximiano and Kory Garner
Topic: Games: Other, Games: Information
Back to session: Experimental Methodology


Capture vs. Recruitment of Subjects in Experiments Measuring Social Preferences

Linda Kamas
Dept. Economics, Santa Clara University

ABSTRACT:

This paper examines the effects of measuring pro-social preferences using subjects who have been recruited with the promise of pecuniary reward compared to students captured by an unknown experimentalist conducting experiments during a class period. Using data from five experiments and more than 1,000 subjects, we show that conclusions about social preferences differ significantly, with recruited students far more likely to be self-interested and efficiency maximizing and less likely to be inequity averse. This finding may help explain the different results of researchers trying to determine whether inequity aversion or surplus maximization better characterizes other-regarding behavior. We argue that capturing subjects in classrooms more likely to leads to accurate conclusions about social preferences than recruiting subjects, but we also acknowledge that the preferred recruitment method is sensitive to the types of behaviors being explored in the experiment.

Author(s): Linda Kamas and Anne Preston
Topic: Social Behavior: Other-regarding Preferences, Methodology: Experimental Design, Tools and Practices
Back to session: Experimental Methodology


A Comprehensive Comparison of Risk Elicitation Methods

Tamas Csermely
Vienna University of Economics and Business

ABSTRACT:

The question of how to measure and classify people's risk preferences is of substantial importance in thefield of Economics. Not surprisingly, many different ways have emerged on how to elicit risk preferences most accurately, with the Multiple Price List (MPL) method being the most prevalent. We conduct a holistic investigation of different MPL methods, inspired by different risk elicitation methods in the literature. Our findings suggest substantial differences in elicited risk preferences depending on the version of the MPL used. In addition, elicitation methods vary to a great extent in terms of within-method consistency as well. Based on two games we use as benchmarks, we determine the best version of MPL, also using other criteria, namely two measures of within-method consistency and self-reported perception of complexity.

Author(s): Tamas Csermely, Alexander Rabas
Topic: Decision Theory: Risk, Methodology: Experimental Design, Tools and Practices
Link: "(The paper has not been uploaded to any sites, but the working paper is available upon request)"
Back to session: Experimental Methodology


Measuring Risk Aversion to Guide Policy Naturalistic Tasks and Respondents

Elisabet Rutstrom
Georgia State University

ABSTRACT:

When measuring risk attitudes for use in policy analysis it is important to recognize that the context may matter. We propose a method that goes beyond simple context framing using text and pictures and instead relies on Virtual Reality simulations. Such simulations have been shown to improve the accuracy of perceptions in experimental context. We ask whether elicited risk preferences differ when using simulations or stylized lottery tasks, and compare students (a low cost convenient subject pool) to field subjects. While we see initial differences across our 2x2 treatment design, we find that they converge with repetition.

Author(s): Vinayak V. Dixit, Rami C. Harb, Jimmy Martinez-Correa, Elisabet E. Rutstrom
Topic: Decision Theory: Risk, Methodology: Lab, Field and External Validity
Back to session: Experimental Methodology


Dictates, Ultimatums and Spousal Empowerment in Rural India

Urmimala Sen
Georgia State University

ABSTRACT:

We conducted artefactual field experiments in rural India with simple bargaining games in which two players bargain over division of an endowment. These games are ultimatum, dictator and two variations of bargaining games - a dictator game with role choice game (where the FM can choose to give up their role of dictator to the SM) and an empowerment game (where the FM can empower the SM to accept or reject their proposed offer). Eight treatments are played: in four we provide information that the other player is the spouse and in the remaining four variations that information is not provided. Both new variations are strategically equivalent to the dictator game since one player has the unilateral ability to determine the allocation yet the FM may instead choose to share decision-making power with the other. We find that when subjects are unaware that they are playing with their spouse, they choose to keep the dictator role for themselves or not to empower the other player. We also find that the divisions in these games (with no spouse information) are actually far less equitable than traditional dictator games. Therefore we find support for subjects' concern for distributive fairness but not procedural fairness.

Author(s): James C. Cox and Urmimala Sen
Topic: Games: Bargaining, Field Experiments: General
Back to session: Experiments in Development


Insurance structure, risk-sharing and investment choices: an empirical investigation

Laura Munro
London School of Economics and Political Science

ABSTRACT:

I conduct an artefactual field experiment in Gujarat, India to disentangle the effect of variations in access to information and weather insurance designs on the investment decisions of insured farmers. I focus on one particularly policy-relevant - and to-date unexplored - variation: whether weather insurance is sold to groups or individuals. Prior studies have found that farmers make riskier investment choices - and thus can achieve higher average agricultural productivity - when reducing their risk exposure through agricultural insurance. However, these articles have focused only on the effect of individual insurance. I examine whether this effect differs across individual and group products, as - analogous to joint liability microcredit products with asymmetric information - a group insurance structure could be subject to moral hazard. Farmers able to hide investment choices and attribute crop losses to rainfall deficiencies could foreseeably make riskier investments and seek higher proportions of group insurance payouts. On the other hand, access to information about peer decisions may lead to group pressure and constrain risk levels undertaken in ex-post investment choices. To test these predictions, I randomly endow groups of smallholder farmers with different forms of insurance, i. group insurance with hidden information, ii. group insurance with access to information, and iii. individual insurance with access to information, and then ask farmers to choose between three fertilizer products with varying levels of risk and return. Experimental results confirm my predictions: farmers covered by group policies pursue more conservative investments when groups are able to monitor investment decisions than when these investments are hidden from the group. Farmers with group insurance and access to information also make more conservative decisions than farmers with individual insurance and access to information.

Author(s): Laura Munro
Topic: Games: Information, Social Behavior: Group Behavior
Link: https://www.dropbox.com/sh/2dewhbssvpejdha/HTYUkaQulc
Back to session: Experiments in Development


Creating Bubbles Out of Thin Air: The Effect of Biased Reward Functions on Price Convergence and Volatility

Jens Schubert
Virginia Commonwealth University

ABSTRACT:

Using a set of laboratory experiments, we study the impact of financial analysts' incentives (reward functions) on price convergence and price volatility in markets with positive and negative feedback loops (reflexivity). Positive (negative) feedback plays an important role in demand-driven (supply-driven) markets, respectively. Our experimental findings indicate that in markets with positive feedback, biased reward functions can result in market prices being far removed from fundamental values (bubble formation). In markets with negative feedback loops, biased incentives can result in market price deviations from fundamental values; however, they do not lead to bubble formation. Our study adds to the asset price bubble literature by showing how biased economic incentives can contribute to the formation of economic bubbles.

Author(s): Jens Schubert, Luiz R. Lima, Michael K. Price
Topic: Markets: Finance,
Back to session: Finance 1


Bank Run Psychology

Oege Dijk
Gothenburg University

ABSTRACT:

There are two views on the mechanism for bank run contagions. One view states they are based on fundamentals and economic linkages between banks. The other that they are based on psychology: the panic hypothesis. However there is no good existing theory on how bank run psychology would affect decision-making. In this paper we test a theory that the presence of background fear changes people's decision-making and thus leads to a higher likelihood of bank runs. We test this through a bank run experiment where we induce background fear through an autobiographical reflection task. We find that in the fear treatment subjects are twice as likely to withdraw their deposits from a shaky bank than in the baseline treatment. Subjects in the fear treatment are also more likely to withdraw after observing previous withdrawals, but unlike in the baseline treatment are not more likely to keep their deposits in the bank after observing others do the same. Panic seems to be a one way street.

Author(s): Oege Dijk, Radboud University Nijmegen
Topic: Psychology and Biology: Emotions, Decision Theory: Bounded Rationality
Back to session: Finance 1


Effect of Visual Representation and Experience on Asset Market Bubbles

Anya (Savikhin) Samak
University of Wisconsin-Madison

ABSTRACT:

We conduct a laboratory experiment to investigate whether observational learning and visual information display can decrease bubble formation in asset market bubbles with up to three experience levels. We first vary whether transaction prices are displayed in a column of text or in a graphical display (with time on the X-axis and price on the Y-axis). Second, we conduct pre-market sessions in which each participant is ??matched' with a trader from a prior market and observes, but does not participate in, the market as it unfolds. We find that among inexperienced and once-experienced subjects, markets with the graphical display result in bubbles that are lower in amplitude relative to markets with the tabular display. Observational learning, like experience, significantly reduces the amplitude of bubbles in a subsequent market. Moreover, there is some evidence of greater learning in markets with graphical display as compared to text display.

Author(s): Tim Cason and Anya (Savikhin) Samak
Topic: Markets: Auctions, Decision Theory: Learning
Back to session: Finance 1


Marked to Market Accounting and Financial Market Instability: Experimental Evidence

Shengle Lin
San Francisco State University

ABSTRACT:

We evaluate the performances of marked-to-market accounting standards in laboratory asset market against two alternative accounting methods: historic cost and fundamental value accounting. We find that the pricing effect is trivial when accounting standards do not inject cash into the market. When loan injection is determined by one of the three valuation methods, price bubbles start to appear. We find that the markets with fundamental value accounting standards are immune to defaults in the presence of price bubbles. Markets with both historic cost and marked-to-market accounting standards experienced defaults, with the latter having even larger sizes of liquidations. We attribute the differences in default likelihood to the differential capacity of de-leveraging in the face of a liquidity shock associated with the three accounting standards.

Author(s): Shengle Lin (presenter); David Porter; Glenn Pfeiffer
Topic: Markets: Finance, Decision Theory: Risk
Back to session: Finance 1


An Experimental Analysis of Bubble: Private Information and Trading Behavior

Tomoki Kitamura
Aoyama Gakuin University and NLI Research Institute

ABSTRACT:

Why a bubble occurs is partly explained by asymmetric information or the interaction between rational and behavioral investors. In this paper, we experimentally investigate the relationship between investor's perceptions about the existence of asymmetric information and trading behavior. When it is common knowledge that private information may exist but no such information exists in actuality, investors tend to make perceptions about unknown information related to stock prices. A bubble occurs after investors believe that stock prices could accurately reflect private information. This mechanism of bubble emergence is different from that discussed in past studies on asymmetric information or interaction. In this study, we prepared two treatments??treatment A: no investor has actual private information and treatment B: some investors have private information. We employed a within subject design wherein the subjects traded stocks first under treatment A, next under treatment B, and finally, under treatment A again. In the first treatment A, we found that the stock price was almost fair considering the expected value of dividends. In treatment B, the stock price converged to the price that fully reflects private information. In the second treatme

Author(s): Munenori Nakasato (Aoyama Gakuin University) and Tomoki Kitamura (NLI-Research Institute)
Topic: Markets: Finance,
Back to session: Finance 2


Limelight on Dark Markets: An Experimental Study of Liquidity and Information

Michael McBride
UC Irvine

ABSTRACT:

The goal of this paper is to study how informational frictions affect asset liquidity in OTC markets in a laboratory setting. The experiments replicate an OTC market similar to the one used in monetary and financial economics (Shi, 1995; Trejos and Wright, 1995; Duffie, Garleanu, and Pedersen, 2005): individuals are matched bilaterally and at random, there are gains from trades due to differences in technologies and endowments, and the terms of trade are determined through a simple bargaining protocol. Subjects buy commodities that have different private values with assets that have common values and can be subject to a private information problem. The asset plays the role of a medium of exchange, but this role can be affected by its lack of recognizability. We study a benchmark experiment where the OTC bargaining game takes place under complete information, a set of experiments with adverse selection where the terminal value of notes are determined exogenously, and a set of experiments with hidden actions where subjects can produce fraudulent notes at some cost.

Author(s): Aleksander Berentsen, Michael McBride, Guillaume Rocheteau
Topic: Markets: Macroeconomics, Games: Bargaining
Back to session: Finance 2


Understanding the channels of financial contagion: Theory and experiments

Isabel Trevino
new york university

ABSTRACT:

The sudden and unanticipated spread of financial crises across countries without a clearly understood transmission pattern has been an ongoing topic of study for both academics and policy makers. Two important mechanisms of financial contagion are a fundamental channel, based on financial or real links, and a social learning channel derived from noisy observations about the behavior of agents in foreign markets that move first. I model this environment using global games and test the model experimentally in an effort to distinguish the relative strength of these two channels. While the theory makes clear statements about which channel should be relevant in the different treatments of the experiment, we observe systematic deviations in the way subjects use the information at their disposal. Two main biases arise in the data: a base rate neglect bias, by which agents underweight their prior, and thus put less weight on the fundamental channel, and an overreaction bias where subjects put too much weight on the behavior they observe in a foreign country, i.e. on the social learning channel. The impact of these biases depends on the environment of the economy, which is defined by the experimental treatment. These results have important consequences in terms of welfare and provide a characterization of the conditions in the economy that can potentially increase or reduce the strength of each of these channels in spreading contagion.

Author(s): Isabel Trevino
Topic: Games: Coordination, Games: Information
Back to session: Finance 2


Trustees and agents in corporate governance

Giuseppe Danese
Simon Fraser University

ABSTRACT:

In this paper I discuss two approaches to the role of managers of business enterprises. The first approach views managers as agents of a principal, the second as trustees of the enterprise. Both approaches recognize that it is desirable to have institutional structures that restrain managerial opportunism: agency focuses on monitoring (high-powered incentives in Williamson's terminology), trusteeship on expectations of behaviour (low-powered incentives). I present the results of an experiment designed to investigate the relative effectiveness of high- and low-powered incentives in restraining managerial opportunism and attracting investments by a group of workers exerting real effort in a virtual private organisation.

Author(s): Giuseppe Danese
Topic: Social Behavior: Norms and Morals, Applied Economics: Labor Market
Back to session: Firm Behavior 1


Money Illusion and Price Competition

Enrique Fatas
University of East Anglia

ABSTRACT:

We experimentally analyze the effect of money illusion in price competition. Experimental firms decide first a market and then repeatedly compete in prices with another firm. Markets are equivalent in real terms and only differ in the currency the price competition is run under. We find that prices increases with the nominal exchange rate, being the effect persistent over time. Pricing decisions are consistent with a behavioral equilibrium model based on coarse grid pricing. The model captures average treatment effects, market price dispersion and accounts for more than 90% of all symmetric price profiles observed in the laboratory.

Author(s): Enrique Fatas, Guillermo Mateu and Antonio J Morales
Topic: Decision Theory: Bounded Rationality, Markets: Industrial Organization
Link: --
Back to session: Firm Behavior 1


Optimal Organizational Response: An Experimental Study of Firm Structure

Miguel Martinez
Universitat Pompeu Fabra

ABSTRACT:

We experimentally test a model of how uncertainty affects the degree of decentralization of decision rights in a firm. In particular, we examine the tradeoff between coordination and adaptation to tasks faced by many firms. We consider a task reallocation framework where the manager makes two decisions: what team to use and who will reallocate tasks; i.e. herself (centralization) or her team (decentralization). In the latter case, a coordination conflict may arise. In our model, team composition endogenously affects the tension between members while the manager faces uncertainty over the nature of the tasks. The model predicts a threshold on the level of uncertainty such that for low uncertainty the manager chooses a centralized organization, and for high uncertainty, a decentralized one. Moreover, in a centralized organization, reducing uncertainty should make team composition more heterogeneous, while a decentralized organization should be unaffected by the manager's uncertainty and always more heterogeneous than under centralization. We experimentally test the model and find that managers choose centralized organizations more often than predicted, although managers correctly opt for more decentralization as uncertainty rises. Interestingly, we find an unexpected Loss Aversion Effect that appears over time. When managers observe a worker's decision against their interest on previous rounds, they overreact, choosing a disproportionally homogeneous team. As uncertainty is reduced, this effect becomes more evident.

Author(s): Martinez Carrasco, Miguel Angel and Hamman, John
Topic: Applied Economics: Labor Market, Social Behavior: Group Behavior
Back to session: Firm Behavior 1


Estimating Dynamic Games of Oligopolistic Competition: An Experimental Investigation

Tobias Salz
New York University

ABSTRACT:

Dynamic oligopoly estimators have become a workhorse for industry studies in empirical industrial organization. In this paper we implement a dynamic oligopoly model in the laboratory. We recover the parameters of this game by applying a structural dynamic oligopoly estimator to the laboratory-generated data. This allows us to compare these recovered estimates to the true parameters of the economic structure that we implemented. We also compute a counterfactual using the recovered parameters and compare the predicted play to the actual play coming from an experimental treatment that implements that counterfactual. Counterfactual calculations are typically performed under the assumption that the same type of equilibrium is played under the new parameter constellation. Implementing the counterfactual as a treatment allows us to check in the actual data how appropriate this assumption is. Lastly, we also assess to which extent laboratory subjects can be described by markov perfect behavior, which is an important assumption for the estimation of dynamic oligopoly estimators.

Author(s): Tobias Salz and Emanuel Vespa
Topic: Games: Repeated Games, Markets: Industrial Organization
Back to session: Games: Sophistication and Expertise


COORDINATION, COMMON KNOWLEDGE AND AN H1N1 OUTBREAK

Natalia Mishagina
CIRANO

ABSTRACT:

We experimentally measure the ability of subjects to coordinate on issues related to the H1N1 influenza virus before and after an outbreak. Our instrument consists of an H1N1 survey converted into a coordination game. We show that the degree of coordination on beliefs about (1) H1N1 prevention, (2) H1N1 vaccine side effects, and (3) actions to take if sick all increase after the outbreak. More generally, we provide the first test of coordination on meaningful strategy labels in response to a naturally occurring field coordination device. Our results provide a unique insight into the responses of beliefs to a major health event.

Author(s): Jim Engle-Warnick, Sonia Laszlo, Natalia Mishagina, Erin Strumpf
Topic: Social Behavior: Other, Social Behavior: Other
Link: http://www.cirano.qc.ca/pdf/publication/2013s-10.pdf
Back to session: Games: Sophistication and Expertise


Blind Stealing: Experience and Expertise in a Mixed-Strategy Poker Experiment

John Wooders
UTS

ABSTRACT:

We explore the role of experience in mixed-strategy games by comparing, for a stylized version of Texas Hold-em, the behavior of experts, who have extensive experience playing poker online, to the behavior of novices. We find significant differences. The initial frequencies with which players bet and call are closer to equilibrium for experts than novices. And, while the betting and calling frequencies of both types of subjects exhibit too much heterogeneity to be consistent with equilibrium play, the frequencies of experts exhibit less heterogeneity. We find evidence that the style of online play transfers from the field to the lab.

Author(s): John Wooders & Matt Van Essen
Topic: Games: Contests,
Back to session: Games: Sophistication and Expertise


When do Women Compete with Men? Gender Differences in Competitive Behaviour in an Investment Decision Experiment

Qiyan Ong
Nanyang Technological University

ABSTRACT:

The gender wage gap has been partially attributed to differences in competitive behavior: The evidence suggests women are less willing to compete, and less effective at competing, than men. However, the appropriate policy - if any - depends crucially on the extent to which gender differences in competitive behavior are caused by the gender-stereotype of the tasks involved, or by the gender composition of competitor groups. Our study introduces experimental variation in the gender stereotype of an investment task while holding other all other features of the task fixed. Rather than use priming or suggestion for manipulating gender stereotypes, we alter the task itself directly in a naturalistic manner. Consistent with previous studies, we show that the gender gap in tournament entry exists in stereotypical male tasks. However, there is no gender gap in the same task, when modified to appeal to stereotypical female abilities. Nor is there a gender gap in entry in tasks which contain both stereotypical male and female elements, conditional on competitors being restricted to the same gender. This suggests that the documented gender gap in tournament entry in stereotypical male tasks is mainly caused by women's unwillingness to compete with men and women's unwillingness to compete in tasks devoid of any stereotypically female elements.

Author(s): Qiyan Ong; Walter Theseira
Topic: Applied Economics: Labor Market, Psychology and Biology: Gender and Individual Differences
Back to session: Gender 2


Young Consumer's Demand for Natural Sweeteners

Mariah Ehmke
University of Wyoming

ABSTRACT:

Health conscious consumers are increasingly concerned about the caloric content and glycemic index of sweeteners added to food. Currently, the average American consumes 22 teaspoons of added sugar in processed foods per day. Young people typically consume higher amounts of sweeteners via candy, sports drinks, and soda (Smed, Jensen et al. 2007). Recently, the American Heart Association issued a statement recommending no more than six teaspoons for women and nine teaspoons for men of added sugar in processed foods per day (Winslow and Wang 2009). Of particular concern is the glycemic index of sweeteners-or how quickly sweeteners raise one's blood sugar level after consumption. While much of recent concern about added sugar focuses on high fructose corn syrup, other industries, including the honey and beet sugar industries, are likely to be affected by these new recommendations and consumer sentiment.

Author(s): Ehmke, Mariah; Katrina Krutkramele, Chris Bastian, Linda Thunstrom, Enette Larson-Meyer, and Matt Andersen
Topic: Decision Theory: Preferences, Decision Theory: Risk
Back to session: Gender 2


Do Differences in Willingness to Compete in Laboratory Experiments Explain Gender Differences in Labor Market Outcomes

Anne Preston
Santa Clara University

ABSTRACT:

We present the results from the first three years of an ongoing project (8 years in total) which tests the hypothesis that competitiveness and other personality characteristics affect labor market outcomes and contribute to the gender gap in labor market performance. We compare individuals' behavior in the lab during their senior years in college to their subsequent career performances by following the laboratory subjects for five years after graduation. This presentation links measures of willingness to compete and confidence to early career wages for the first two waves of students at Haverford College and Santa Clara University.

Author(s): Linda Kamas and Anne Preston
Topic: Psychology and Biology: Gender and Individual Differences, Applied Economics: Labor Market
Back to session: Gender 2


Exploring gender differences in preferences over competition using switch points and relative payoffs

John Ifcher
Santa Clara University

ABSTRACT:

Niederle and Vesterlund (2007) find gender differences in preferences for competition: in a binary choice, 73% (35%) of men (women) choose a $2.00 Winner-Take-All (WTA) payment over a $0.50 Piece-Rate (PR) payment for solving addition (math) problems. This gender difference is not explained by variant performance or risk and feedback aversion; rather, the authors attribute it to men being more overconfident than women, and a greater preference for competing among men than women. To more fully understand gender differences in preferences for WTA and PR payments, we offer subjects a series of choices between PR payments, ranging from $0.00 to $2.00, and a $2.00 WTA payment. We identify the strength of each subject's preference for a PR payment relative to a WTA payment by observing his or her switch point: the minimum PR payment the subject preferred to a $2.00 WTA payment. The mean and median switch points are substantially greater for men than for women, $0.94 versus $0.66, and $0.85 versus $0.60, respectively. Further, switch points enable us to examine how much greater a WTA payment has to be relative to a PR payment so that a subject chooses the WTA payment. Women, on average, choose the WTA payment as long as the relative payoff is greater than 3.0 (= $2.00 WTA payment ?? $0.66 mean switch point). In contrast, men, on average, choose the WTA payment as long as the relative payoff is great than 2.1 (= $2.00 WTA payment ?? $0.94 mean switch point). Thus, compared to men, women require a 43 percent relative payoff premium, on average, to choose the WTA payment. Lastly, we test whether the preference for competition is related to participation in college or high school varsity sports, choice of major, enrollment in college-level STEM (science, technology, engineering, and math) courses, GPA, and a measure of narcissism.

Author(s): John Ifcher and Homa Zarghamee
Topic: Psychology and Biology: Gender and Individual Differences, Methodology: Experimental Design, Tools and Practices
Back to session: Gender 2


Parental socialisation effort and the intergenerational transmission of risk preferences

Teodora Boneva
University of Cambridge

ABSTRACT:

We study the transmission of risk attitudes in a unique survey of mothers and children in which both participated in an incentivized risk preference elicitation task. We document that risk preferences are correlated between mothers and children when the children are just 7 to 8 years old. This correlation is only present for daughters. We show that a measure of parental involvement is a strong moderator of the association between mothers' and daughters' risk tolerance. These findings support a role for socialization in the intergenerational transmission of preferences that predict economic behaviour.

Author(s): Sule Alan, Nazli Baydar, Teodora Boneva, Thomas Crossley, Seda Ertac
Topic: Psychology and Biology: Gender and Individual Differences,
Link: http://www.ifs.org.uk/publications/6752
Back to session: Gender and Stereotypes


Affirmative Action and Stereotype Threat

Anat Bracha
Research Department, the Federal Reserve Bank of Boston

ABSTRACT:

Stereotype threat is when a reminder of a negative stereotype leads individuals to conform to the stereotype. This effect is often demonstrated following a simple prime such as reminding one of her race or gender. Affirmative action, a policy that recognizes a group in a need of a support, may therefore serve unintentionally as a prime leading to stereotype threat. This paper experimentally investigates whether gender-based affirmative action, in the form of a quota, has such a negative unintended consequence. We use quantitative GRE questions, and find that while affirmative action has a positive effect on the performance of low-ability women, it negatively affects high-ability women. Since we find no evidence that high ability women exert less effort under affirmative action or single-sex competition, our findings suggest there are two opposing factors associated with affirmative action, consistent with the hypothesis that affirmative action may have an unintended negative consequences due to stereotype threat.

Author(s): Anat Bracha, Alma Cohen and Lynn Conell-Price
Topic: ,
Back to session: Gender and Stereotypes


Income attraction: An online dating field experiment

Jue Wang
Peking University, HSBC Business School

ABSTRACT:

Marriage rates have been decreasing in the West contemporaneously as women's relative wages have been increasing. We found the opposite pattern in China. Prior empirical studies with US marriage data indicate that women marry up (and men marry down) economically. Furthermore, if the wife earns more, less happiness and greater strife are reported, the gender gap in housework increases, and they are more likely to divorce. However, these observational studies cannot identify whether these consequences were due to men's preference for lower income women, or women's preference for higher income men, or to other factors. We complement this literature by randomly assigning income levels to 360 unique artificial profiles on a major online dating website and recorded the incomes of nearly 4000 visits. We found that men of all income levels visited women's profiles with different income levels at roughly equal rates. In contrast, women at all income levels visited men with higher income at higher rates, and surprisingly, these higher rates increased with the women's own income. Men with the highest level of income got ten times more visits than the lowest. As well as suggesting that women's preference for men's income may drive income patterns in marriage, the correlation between labor market conditions and rates of marriage, the wage premium for married men, and men in general, our findings also suggests an explanation for why the gender gap in housework is higher for women who earn more than their husbands. To our knowledge, this is the first field experimental study of gender differences in preferences for mate income.

Author(s): David Ong, Jue (Julie) Wang
Topic: Field Experiments: General, Markets: Market design and Matching
Link: https://dl.dropboxusercontent.com/u/21811713/Income%20attraction.pdf
Back to session: Gender and Stereotypes


Gender, Age and Norm Enforcement: Evidence From a Matrilineal and a Patriarchal Society

Edward Asiedu
University of G?ttingen

ABSTRACT:

This paper uses an experiment to study the impact of womens norm enforcement on collective action. Research suggests that women are less influential when using dominant forms of communications. Two explanations have been given to this phenomenon. The first is that, lower status individuals exhibit a much lower willingness to sanction norm violation. The second assumes the cultural environment. From the experimental data collected from the patriarchal Ewe and the matrilineal Ashanti cities in Ghana, we find that, women leaders in the matrilineal city are as effective as the men in enforcing group norms whilst women leaders in the patriarchal city are less effective in enforcing group norms. The data also shows that women and men in the patriarchal city show similar sanctioning behaviors. We argue that most of the negative impact of women's leadership on collective action may be driven more by factors that are external to the woman, like the culture or general social perception. In terms of the impact of age differences of leaders on collective action and norm enforcement, we do not find such evidence.

Author(s): Edward Asiedu, Marcela Ibanez
Topic: Field Experiments: General, Psychology and Biology: Gender and Individual Differences
Back to session: Gender and Stereotypes


Demanding or Deferring : Cross-cultural Gender Differences in Communication

Siyu Wang
George Mason University

ABSTRACT:

Substantial attention has been paid to the impact of gender differences on various economic outcomes and behaviors (e.g., wage gaps, willingness to compete and risk attitudes). It is well accepted that males and females communicate in different ways. Little effort has been made to understand the impact of gender differences in communication styles on economic outcomes. We investigate consequences of gender differences in communication in the context of a coordination game in different cultures. We use a coordination game because it is economically relevant and communication is able to substantially impact play in this environment. It is also known that the role of males and females in society is culturally dependent. Little is known about how gender differences in communication vary with culturally determined gender roles. Our investigation also informs this issue. We developed a game that enables one to draw inferences regarding a specific gender role in a certain culture. In particular, are men expected to defer to women, or vice-versa? We measured gender differences in communication styles in three different well controlled environments and investigated how communication impacts economic coordination behaviors of males and females in different cultural contexts.

Author(s): Siyu Wang and Daniel Houser
Topic: Social Behavior: Communication, Games: Coordination
Back to session: Gender 1


Fact or Fiction? Gender and the Truthfulness of Campaign Messages

Jonathan Woon
University of Pittsburgh

ABSTRACT:

We investigate whether men and women candidates craft different kinds of campaign messages prior to an election and whether such differences are related to subjects' willingness to become political candidates. In our experiment, messages prior to the election are cheap talk and subjects' rewards are based on a real-effort (anagram) task that we use to represent objective ability. In the control condition, group members have common incentives to elect the best member of their group as a representative. In our treatment conditions, we add a private incentive (bonus) for subjects who are elected as the group representative and vary whether the bonus is for individuals or also applies to other members of the group (the latter should activate prosocial preferences). We find that men are more likely than women to use precise (rather than vague) messages and to overstate their ability when creating campaign messages. These results may help to explain why fewer women choose to run for political office, and may provide insights into the types of campaign environments that would encourage the most qualified candidates to run.

Author(s): Kristin Kanthak and Jonathan Woon
Topic: Psychology and Biology: Gender and Individual Differences, Social Behavior: Lying and Cheating
Back to session: Gender 1


Gender differences in negotiation

Muriel Niederle
Stanford University

ABSTRACT:

There is a lot of scattered evidence on gender differences in Negotiation. In this paper we try to assess the extent to which gender differences are on the intensive margin - how well men and women negotiate - and the extensive margin - do they enter negotiations. We also assess the role of uncertainty in exacerbating gender differences.

Author(s): Christine Exley, Muriel Niederle and Lise Vesterlund
Topic: Psychology and Biology: Gender and Individual Differences, Games: Bargaining
Back to session: Gender 1


Higher Quality and Lower Cost from Improving Hospital Discharge Decision Making

Vjollca Sadiraj
Georgia State University/ AYSPS

ABSTRACT:

Excess, wasted costs in healthcare are estimated to exceed the size of the entire defense budget. Effectively addressing this problem requires decreasing the cost of healthcare while increasing its quality by improving healthcare decision making. This paper reports research on improving decisions about hospital discharges, decisions that are now made by physicians and based on mainly subjective evaluations of patients' health status. The research has three components: (a) econometric identification of patient medical and demographic variables that discriminate between likelihood of successful discharge and likelihood of subsequent readmission within 30 days (the Medicare threshold); (b) development of decision support software that can be used in more effective, evidence-based discharge decision making; and (c) empirical evaluation of effectiveness of the decision support software. Experimental data reported herein indicate that utilizing the decision support software with change in the default option reduces (i) patient length of stay in the hospital and (ii) the probability of readmission for higher risk patients.

Author(s): James C. Cox, Vjollca Sadiraj, Kurt E. Schnier and John F. Sweeney
Topic: Applied Economics: Other,
Back to session: Industrial Organization 1


Experiments as Insturments: Understanding Consumer Behavior in Sponsored Search

Matt Goldman
UCSD

ABSTRACT:

In this paper we show how one can re-purpose business experimentation to estimate causal effects unrelated to the intent of the original experiment. Using a large dataset from a major search engine, we apply this technique to examine the impact of ad position in sponsored search on click and purchasing patterns. Our findings reveal substantial heterogeneity in click position effects across advertisers and an overall slightly positive impact of position on conversions, with limited heterogeneity. Using experiments in this way to estimate auction ranking parameters has the potential to shape markets using GSP-like mechanisms to allocate and price.

Author(s): Matt Goldman, Justin Rao
Topic: Methodology: Experimental Design, Tools and Practices, Markets: Market design and Matching
Link: http://econ.ucsd.edu/~mrgoldman/exp_instruments1.pdf
Back to session: Industrial Organization 1


Shrouding add-on information: an experimental study

Tobias Wenzel
University of Duesseldorf

ABSTRACT:

We explore how competition affects firms' obfuscation strategies in laboratory experiments. Firms sell a base good and an add-on product. The price of the add-on may be shrouded and, if so, myopic consumers pay too much (Gabaix and Laibson, 2006). Shrouding is an equilibrium but an unshrouding equilibrium coexists. In our experiments, competition matters in that only duopolistic markets are frequently shrouded whereas four-firm markets are not. With repeated interaction, shrouding rates do not increase. However, for duopolies only, the opportunities to shroud facilitate tacit collusion on the base good price: the unshrouding equilibrium serves as a credible punishment if deviations occur.

Author(s): Hans-Theo Normann and Tobias Wenzel
Topic: Markets: Industrial Organization,
Back to session: Industrial Organization 2


Buy-sell clause with an investigation cost

Xiangdong Qin
Shanghai Jiao Tong University

ABSTRACT:

This paper studies how a closely-held partnership is dissolved with a buy-sell clause between two partners asymmetrically informed about the entity's business. We introduce a theoretical model showing that the managerial partner, who has better knowledge about the partnership, can get an upper hand in the dissolution by triggering the buy-sell clause at a preemptive price if his private value is sufficiently high or low. The limited partner, who must pay an investigation cost to learn his private value, will be deterred from investigation and motivated to choose the option preferred by the managerial partner. Laboratory experiments are used to test how buy-sell clause affect partnership dissolution in the presence of such information asymmetry and investigation cost. We find evidences that managerial partners propose preemptive prices when their values are relatively low, but fail to take full advantage when their values are high, that is, they trigger the clause at a price too high and grant extra profits to their partners. Our theory explains several phenomena commonly observed in legal practices that a proposer often regrets his early decision when his partner's choice is opposite to what he intended. Meanwhile, a responder often requires more information disclosure about the partnership.

Author(s): Xiangdong Qin and Fangzhou Zhang
Topic: Markets: Industrial Organization, Games: Bargaining
Back to session: Industrial Organization 2


Talking about others: Gossip as a means to increase trust and cooperation

Dietmar Fehr
Social Science Research Center (WZB) Berlin

ABSTRACT:

Exchanging social information about absent parties (gossip) appears to be an important aspect in daily-life conversations. While individuals gossip for various reasons, gossip may also function as a social control mechanism, for example, through the manipulation of individuals' reputation. In this study we investigate whether gossip promotes cooperation and what motivates subjects to gossip. We conduct trust games where we introduce a third party that can gossip about a trustee's behavior. A third-party's material payoff is not affected by the decisions made by the trustor or trustee, but it is costly for the third party to gossip. We find that third parties are willing to sacrifice a substantial amount of their endowment in order to gossip. More specifically, we observe that third parties pay more for communicating deviant behavior than for communicating norm compliant behavior of trustees. The possibility to gossip leads to both higher cooperation and higher trust even though all interactions take place anonymously. The results suggest that gossip allows trustors to selectively interact with trustees who have gained a high reputation.

Author(s): Dietmar Fehr and Lorenz Kurrek
Topic: Social Behavior: Communication, Social Behavior: Norms and Morals
Back to session: Interpersonal Influence


Fairness and Persuasion - How Stakeholders' Statements influence Third-Party Distribution

Marco Kleine
Max Planck Institute for Research on Collective Goods

ABSTRACT:

We study whether and to what extent impartial third parties are influenced by opinions of stakeholders when deciding about a fair allocation in a third-party distribution decision. A production stage prior to the distribution allows for different focal fairness rules to be considered. We compare two Communication treatments that allow one of the two stakeholders to state his or her opinion to a Baseline treatment without communication opportunities. We find that statements of stakeholders indeed affect the decision making. Yet, third parties deviate from their initial fairness judgment only if stakeholders' requests are moderate and largely ignore high monetary claims. Due to this asymmetric influence with regards to moderate and non-moderate statements, stakeholders in the Communication treatments who state their opinion are allocated significantly less money than their counterparts in the Baseline treatment.

Author(s): Marco Kleine, Pascal Langenbach, Lilia Zhurakhovska
Topic: Social Behavior: Communication, Social Behavior: Norms and Morals
Back to session: Interpersonal Influence


Interpersonal Influence

Lucas Coffman
Ohio State University

ABSTRACT:

How does one influence another? We conduct an experimental analysis of communication and decision-making to better understand the nature and allocation of influential ability. We consider settings in which a buyer must decide how much he is willing to pay for a good, a seller's payoff increases in willingness to pay, and the seller can communicate with the buyer. We find that (1) communication increases the likelihood of valuations increasing, despite the conflict of interest; (2) influence works primarily through changes in buyers' perceived self-interest, and yet (3) a minority of sellers intentionally manipulate buyers other-regard and outperform as a result. Surprisingly, (4) influence depends less on the identity of the seller than on the identity of the buyer, as well as (5) on seller-buyer homophily.

Author(s): Lucas C Coffman; Paul Niehaus
Topic: Social Behavior: Communication, Social Behavior: Other-regarding Preferences
Back to session: Interpersonal Influence


On Measuring the Preferences of the Poor

Chenna Reddy Cotla
George Mason University

ABSTRACT:

We conduct economic experiments to measure the risk attitudes of a random sample of over 9,650 adults from rural poor households in Peru. Experimental treatments were randomly assigned to households as well as whether the lottery was paid or hypothetical. Each subject was offered two multiple price list lotteries that differed on whether they included positive only or positive and negative pay-offs and whether pay-offs or probabilities were changed. Using multiple price lists allow us to directly test whether decisions are reasoned or due to error. We find that the unexplained variance of individual preferences over gains is large, and average behaviour is close to risk neutral. In contrast, the reaction to losses is uniform and very negative. Women are no more or less risk averse than men, and choices are less risk taking in paid lotteries. Also, despite the large stakes used in the lotteries used, about 40% of subjects can be classified as choosing inconsistently, with women being more likely to be inconsistent. Inconsistent behaviour is more common in instruments that vary probabilities rather than pay-offs as well as when lotteries are paid. The results suggest that an important difference between the poor and other populations is in their ability/desire/understanding to make deliberate decisions and their marked reaction to losses.

Author(s): Marco Castillo, Chenna Reddy Cotla, Ragan Petrie and Maximo Torero
Topic: Decision Theory: Risk, Decision Theory: Preferences
Back to session: Lab to Field 1


Counterfactual Simulations in Experimental Economics: A Study of Blended Physician Payment Structures

Ellen Green
University of Delaware

ABSTRACT:

I use counterfactual simulations to predict the impact of blended payment structures between capitation, fee-for-service, salary, and pay-for-performance on agent behavior. I elicit individuals' responses to these popular physician payment structures from my laboratory experiment, Green 2013, to explain their responses to the blended payment structures. Within the simulation model, I maximize the healthcare provider's payoff with other-regarding behavior towards the patient to find the optimal payment structure. The simulations indicate that the optimal blend of these payment structures is a combination of the capitation and fee-for-service, rather than the popular single incentive scheme used in the healthcare industry.

Author(s): Ellen Green
Topic: ,
Back to session: Lab to Field 1


Caste, Efficiency and Fairness with Public Goods and Common Pool Resources

Urmimala Sen
Georgia State University

ABSTRACT:

The central question in this essay is how the Indians from different castes (a social hierarchical system in India - officially abolished in 1950) perform in social dilemma situations, created by voluntary contributions to public goods and appropriation from common pool resources. Theories of “economic man” and “unconditional social preferences” predict that behavior is identical whether we have private or common property while reciprocal preferences theory predicts that behavior will not be the same. We also examine whether belonging to the same or different caste matters to the subjects. “Artefactual” field experiments were conducted in West Bengal, India with villagers who differ by caste to ascertain whether caste differences exacerbate social dilemmas and which, if any, theory predicts the outcomes. Results indicate a significant lack of trust and trustworthiness among the lower caste individuals, implying a major policy failure. These results are important for the socio-economic development of India - India has made substantial progress in the past few decades, yet there appears to be significant gaps in this progress.

Author(s): James C. Cox, Vjollca Sadiraj & Urmimala Sen
Topic: Field Experiments: General, Public Choice: Public Goods and Common Pool Resource
Back to session: Lab to Field 1


Charitable Dictators? Elite Attitudes in a Developing Country Context

Catherine Porter
Heriot-Watt University, Edinburgh

ABSTRACT:

This paper investigates the attitude of elites in a developing country to local charities. We play a modified dictator game (DG) with students who act as a control group, and also with civil servants and well-off private sector employees. The sample includes both Ugandans and expatriates. In the DG, participants are invited to choose from a list of 20 local charities, and then divide a sum of (real) Ugandan Shillings (equivalent to around $10) between themselves and the charity. Finally, participants were asked whether they would like to add their own money to the amount already placed in the envelope for the charity. The average donation was high relative to other studies (50%). Contrary to many DG findings, non-students (civil servants) were not significantly more generous than students- this could be related to very poor levels of salary in the Ugandan civil service. Only very few individuals donated their own money to the charity, despite 30% of participants donating their whole endowment. The post-game questionnaire elicited attitudes to charity/non-governmental organisation and international aid performance in Uganda, and a surprising negative correlation is found with donations. Preliminary analysis shows that the donation is affected more strongly by attitudes to poverty and human nature (e.g. agreeing strongly with the statement ??some people are poor through no fault of their own) than by the belief that money will be well spent.

Author(s): Catherine Porter, Kim Lehrer
Topic: Applied Economics: Charitable Giving, Applied Economics: Economic Development
Back to session: Lab to Field 2


Social norms about bribe-giving and bribe-taking in India

Tushi Baul
New York University Abu Dhabi

ABSTRACT:

We used a novel experimental methodology introduced by Krupka and Weber (2008) to measure social norms regarding social acceptability of bribes in India. We used a stylized situation obtaining a drivers license in which the possibility to engage in unethical behavior is common. We asked subjects to rate the social appropriateness of different situations depicting bribe giving and bribe taking at a drivers license testing facility. Using a coordination game approach, we provided incentives to the subjects to truthfully reveal their beliefs about common views on the appropriateness of engaging in this type of unethical behavior. We found that the social appropriateness rating varied with the bribe amount. For smaller bribes, there was lack of coordination on the modal social acceptability rating, whereas larger bribes were considered inappropriate by the majority of participants. We also varied the information regarding common behaviors at the drivers license testing facility by telling subjects in some treatments that bribe taking by public officials was prevalent. When bribe giving and bribe taking were socially acceptable, participants perceived bribes to be less socially inappropriate. Also, bribe giving was more socially acceptable than bribe taking in these socially acceptable bribing situations.

Author(s): Ritwik Banerjee, Tushi Baul and Tanya Rosenblat
Topic: Social Behavior: Other, Applied Economics: Economic Development
Back to session: Lab to Field 2


Coordination games with farmer groups: Experimental evidence from Senegal

Angelino Viceisza
Spelman College

ABSTRACT:

Most decisions are made in group contexts where one person’s behavior is impacted by others. As such, it is important to understand conditions in which individuals are able to coordinate. Prime examples are farmer, microfinance or community groups. We conduct lablike (artefactual or extra-lab) field experiments (that is, coordination games) with members of naturally occurring farmer groups in Senegal to assess some determinants of coordination. Preliminary evidence suggests that there is variation in whether or not subgroups of farmers created for purposes of the experiment achieve coordination in the lab. We also find that (1) the potential benefits to coordination (the so-called premium), (2) the presence of external uncertainty (in addition to strategic uncertainty) with regard to the premium, and (3) cheap talk significantly impact coordination. These findings seem to be consistent with previous evidence found with student samples. To the best of our knowledge, we are one of the few studies that explore coordination at the lablike level with members of pre-existing groups. Given this unique opportunity, we use our findings to inform the design of subsequent randomized field experiments with these and other farmer groups.

Author(s): Kodjo Aflagah, Tanguy Bernard, Angelino Viceisza
Topic: Field Experiments: General, Games: Coordination
Link: https://sites.google.com/site/viceisza/research
Back to session: Lab to Field 2


Advice and Herding: An Artefactual Field Experiment with Current and Potential Immigrants

Natalia Candelo Londono
CUNY Queens College

ABSTRACT:

Previous research shows that social networks are an important factor in the emigration decision. However, the causal mechanisms through which social networks have this impact have not been tested. We suggest that the effect is due, in part, to the sharing of information within the social network. We measure this information sharing in a field experiment. We find that more information is transmitted to individuals closer in the social network. Advices are transmitted the most, and in average, reflect participants' own decisions. These findings have important implications for understanding how herds are formed, and what information is likely to be impactful, in the immigration decision.

Author(s): Natalia Candelo, Rachel T. A. Croson, Catherine Eckel
Topic: Field Experiments: General, Social Behavior: Communication
Back to session: Lab to Field 2


Dictator Game Generosity and Real World Outcomes

Catherine Weinberger
Univ of Calif Santa Barbara

ABSTRACT:

A longitudinal research project begun more than 10 years ago is beginning to demonstrate that generous behaviors observed in a well-known laboratory experiment are correlated with later outcomes in the real world. The dictator game offers each player a choice between maximizing earnings and donating money to others. In this longitudinal study, more than 1800 students played a dictator game with charity recipients and completed a survey about career choices. More than 1000 were resurveyed 2-3 years later to determine labor market and other outcomes. More than 675 individuals participated in the most recent wave of the survey, conducted 7-9 years after subjects played the dictator game. Data analysis is at a very early stage. Preliminary analysis is based on samples of men and women employed full-time seven years after the base year dictator game. Within these samples, those who were generous when they played the dictator game seven years previously are less likely to be in the upper tail of the earnings distribution. These differences are statistically significant in quantile regressions at the upper quartile, but not at the median. Other preliminary results include some sorting to occupations and job tasks. For example, generous women are more likely to work with children and are also more likely to be in a supervisory or managerial position.

Author(s): Catherine J. Weinberger
Topic: Applied Economics: Labor Market, Field Experiments: General
Back to session: Lab to Field 3


Does competition among public officials reduce corruption? An experiment

Danila Serra
Southern Methodist University

ABSTRACT:

We investigate experimentally whether an increase in the number of officials providing the same service reduces extortionary corruption, i.e., bribe demands for services that clients are entitled to receive. In the experiment, citizens apply for a license at one of many offices; public officials can demand bribes and citizens can visit multiple offices at a cost. We find that increasing competition between officials has either no effect or a positive effect on bribes, depending on the size of search costs. We compare our results with those obtained in a market setting with multiple sellers of a homogeneous good.

Author(s): Dmitry Ryvkin, Danila Serra
Topic: Applied Economics: Other, Markets: Industrial Organization
Back to session: Lab to Field 3


Intrinsic Motivation, Effort, and the Call to Public Service

Sheheryar Banuri
The World Bank

ABSTRACT:

Pay schemes in the public sector aim to attract applicants that are both high ability and highly motivated. In an environment with weak incentives, little monitoring, and low sanctions for underperformance, attracting motivated staff is critical. Raising public sector pay runs the risk of attracting high ability, yet unmotivated staff, yielding a net loss in output. We test the effectiveness of raising public sector pay in attracting a motivated, high ability workforce. In addition, we ask whether public servants are more motivated than private sector subjects, and finally, whether higher motivation yields higher levels of effort. Using a sample of students destined for the private and public sectors in Indonesia, we use dictator games and real effort tasks to show that (1) public sector workers are more motivated (to provide public goods) than private sector workers, (2) highly motivated workers are more likely to provide higher effort, and (3) high public sector pay is more likely to attract under-motivated workers. Implications for the reform of public sector pay is discussed.

Author(s): Sheheryar Banuri and Philip Keefer
Topic: Applied Economics: Labor Market, Public Choice: Other
Back to session: Lab to Field 3


Monetary Incentives and Support Groups Help People Help Themselves: A Field Experiment

Alessandra Cassar
U San Francisco

ABSTRACT:

Oakland based Family Independence Initiative (FII) claims to be a successful model for poverty alleviation. The FII program is structured around mutual support groups that, through a bottom-up approach of self-selection of goals and small monetary incentives for achieved results, encourage participants to lift themselves from poverty. As the popularity of this program continues to gain momentum in the US, we designed a field experiment to measure its impact. We enrolled close to 200 individuals in four experimental treatments (group, no-group, incentives, no-incentives) and a fifth external control group. Our experimental data show that the full FII program of mutual help groups and incentives is indeed delivering on its promise to improve people's economic life: more goals are achieved and monthly sales values are significantly higher than in the other treatment groups. In addition to investigating the mechanisms behind its effectiveness, our research finds that the FII model is replicable in a developing country setting, promising a new potentially successful yet inexpensive way to help people help themselves out of poverty.

Author(s): Paulina Aguinaga, Alessandra Cassar, Jennifer Graham and Lauren Skora
Topic: ,
Back to session: Lab to Field 3


Dependency Status and Preferences for Social Insurance

John Hamman
Florida State University

ABSTRACT:

Current thinking on the origins and size of the welfare state in developed economies places individuals human capital assets and labor market risk front-and-center. Yet this literature largely ignores the fact that people live in households in which they can depend on others but also have others depending on them. The influence of dependency status' on individual preferences for social insurance programs is largely unknown. We report preliminary results from an ongoing laboratory experiment designed to estimate the effect of mutual dependency on preferences for policies that insure against labor market risk. Early results seem to indicate that 1) willingness to vote in favor of a social insurance policy is responsive to unemployment risk, 2) in a low risk situation, having one's welfare tied to the actions of another increases support for social insurance policies, and 3) task effort varies in conjunction with a variety of factors including risk level, partner's unemployment status, and an individual's vote in favor of insurance.

Author(s): John Ahlquist and John Hamman
Topic: Applied Economics: Labor Market, Decision Theory: Risk
Back to session: Labor 1


The More You Know: How Information affects job search in a large field experiment

Laura Gee
Tufts University

ABSTRACT:

Job search is costly, so job seekers only invest the time in an application if they first believe that the job is a good fit and second believe they have a chance of obtaining an employment offer. Showing the number of previous applicants for a position allows a job seeker to update his priors about both job quality and potential competition. When the number of previous applicants is low this signals low quality, and low competition. On the other hand, when the number of previous applicants is high this signals the opposite: high quality and increased competition. Using a large scale field experiment with 2 million job seekers applying to over 100,000 positions on an online job networking website, I test how previous applicant count affects the likelihood of application, makeup of the applicant pool, and job taking behavior. I find that increasing the amount of information increases the likelihood of application, but does so differentially depending on the current applicant count.

Author(s): Laura K. Gee
Topic: Applied Economics: Labor Market, Field Experiments: General
Back to session: Labor 1


Right Contract for Right Workers? Survey and Experimental Evidence

Tracy Liu
University of Michigan

ABSTRACT:

It is well observed that companies design different contracts for regular staff and temporary workers. Different contract choices include: an explicit incentive contract, e.g. piece rate; an implicit incentive contract, e.g. fixed wage+ bonus; and finally fixed wage only contract. Results from a Word Bank survey with 3493 Chinese companies show that firms are more likely to choose a fixed+bonus contract when facing long-term workers while they always choose piece-rate for short-term workers. Furthermore, we conduct a real-effort principal-agent game experiment to study the effect of long-term relationship as well as the role of bonus in fixed+bonus contract. Specifically, we find that 40% principals choose the fixed wage+bonus contract for long-term workers and this implicit contract is as effective as piece-rate. In contrast, piece-rate is dominantly chosen for short-term workers and it is the only effective mechanism for workers productivity. Additionally, we find that the attractiveness of fixed wage+bonus disappears when bonus is not available, suggesting that fixed payment alone can not be an effective mechanism to improve workers' performance.

Author(s): Wei Chi, Tracy Xiao Liu, Xiaoye Qian, Qing Ye
Topic: Applied Economics: Labor Market, Games: Repeated Games
Back to session: Labor 1


Trust, Trustworthiness, and Success in Business

Mongoljin Batsaikhan
Georgetown University

ABSTRACT:

Do successful entrepreneurs trust others more? Are they more trustworthy than less successful ones? The answers depend on how profitable trust is; one might choose not to trust if that is more profitable than trusting. Although recent literature tends to argue that social capital improves business outcomes, there has not been enough evidence on this at the micro level. It is also not clear whether trusting others pays off when doing business in developing countries, as trusting others too much can be costly under insufficient contract enforcement, thus not profitable for business. This study has implemented a trust game with small business owners in Mongolia but links the trust measure to sales data from real life in order to show a robust correlation between the two. The results show that trusting more (or investing more in trust) is associated with higher earnings in real business. This raises the very interesting question whether successful entrepreneurs who demonstrate more trust do so deliberately because they anticipate reciprocity or whether it is simply an inherent trait which they possess. As documented in the past (Karlan 2005), trustworthiness is associated with loans from bank. I also show that successful entrepreneurs invest in an amount in the trust game that is close to the optimal level of trust that maximizes the earning at individual level as well.

Author(s): Mongoljin Batsaikhan
Topic: Field Experiments: General, Social Behavior: Norms and Morals
Back to session: Labor 1


Overreaction or Under-reaction: How Do People Respond to Wage Changes?

Tai-Sen He
Nanyang Technological University

ABSTRACT:

This paper experimentally examines individuals' work effort choices in response to wage changes. I designed a simple, real-effort experiment with four treatments. Subjects experience a wage increase in one treatment and a wage decrease in another treatment; the other two treatments are used as control treatments in comparison with the former two treatments. I find that subjects overreact to the wage increase but under-react to the wage decrease. Male subjects' average effort is higher than female subjects. However, interestingly, the wage increase and the wage decrease both have a more pronounced effect on female subjects than their male counterparts. The results have implications for research on the wage elasticities of labor supply.

Author(s): Tai-Sen He
Topic: Applied Economics: Labor Market, Psychology and Biology: Emotions
Back to session: Labor 2


Job Tenure as a Signal of Effort in Labor Markets with Incomplete Contracts

Frederic Schneider
University of Zurich

ABSTRACT:

A common assumption in labor economics is that effort is neither publicly observable nor enforceable. However, one widely available and utilized kind of information in labor markets is a worker's employment history. We hypothesize that work history may indirectly convey information about a worker's past effort. In an experimental labor market, we vary public knowledge about workers' previous employment history. We then introduce a shock that requires all market participants to switch partners. We observe that workers who have a history with longer tenure spells are more likely to get a job offer, and that they can command higher wages after the shock. In the periods before the shock, we find that employer switches become less frequent compared to the control condition without information about employment history. This effect also diminishes the chances of unemployed workers to find a job, thus creating longer unemployment spells. We conclude that information about tenure can serve as a substitute for information about effort, but that this can exacerbate labor market rigidity as workers stick to their jobs in an effort to obtain longer tenure with their current firm and previously-employed workers are more desirable to firms.

Author(s): Frederic Schneider and Roberto Weber
Topic: Applied Economics: Labor Market, Social Behavior: Other
Back to session: Labor 2


Cycle frequency in standard Rock-Paper-Scissors games: Evidence from experimental economics

Zhijian Wang
Experimental Social Science Laboratory, Zhejiang University

ABSTRACT:

The Rock-Paper-Scissors (RPS) game is a widely used model system in game theory. Evolutionary game theory predicts the existence of persistent cycles in the evolutionary trajectories of the RPS game, but experimental evidence has remained to be rather weak. In this work, we performed laboratory experiments on the RPS game and analyzed the social-state evolutionary trajectories of twelve populations of N=6 players. We found strong evidence supporting the existence of persistent cycles. The mean cycling frequency was measured to be 0.029+/-0.009 period per experimental round. Our experimental observations can be quantitatively explained by a simple non-equilibrium model, namely the discrete-time logit dynamical process with a noise parameter. Our work therefore favors the evolutionary game theory over the classical game theory for describing the dynamical behavior of the RPS game. (Physica A: Statistical Mechanics and its Applications Volume 392, Issue 20, 15 October 2013, Pages 4997-5005)

Author(s): Bin Xu, Hai-Jun Zhou, Zhijian Wang
Topic: Decision Theory: Learning, Games: Repeated Games
Link: http://dx.doi.org/10.1016/j.physa.2013.06.039
Back to session: Learning


Heterogeneity in Experienced-Weighted Attraction Learning and Its Relation to Cognitive Ability

Shu Heng Chen
national chengchi university

ABSTRACT:

In this article, we apply the experience-weighted attraction (EWA) learning model to examine the learning behavior of a set of 108 subjects, who participated in one of the six series of 15- to 20-person Keynes' beauty contest experiments, held at National Chengchi University in year 2010. Our earlier study (Chen, Du and Yang, 2013) has found a positive relation between cognitive ability, measured by a working memory test (Lewandowsky et al. 2010), and cognitive hierarchy, in the sense of level-k reasoning. Through the analysis of the estimated Markov transition matrix, we further found that the subjects with a higher cognitive ability have a di erent dynamic behavioral pattern from those with a lower cognitive ability, which indicates the possible e ect of cognitive ability on learning. Hence, as a subsequent study, we examine this possibility by directly applying the EWA learning model, which was rst applied to the beauty contest experiment data by Camerer and Ho (1999). We consider two di erent action spaces in terms of granulation: one which has a ner division (Camerer and Ho, 1999) and hence a large number of choices, and one which has a rather coarse division directly corresponding to the Nagel's classi cation of reasoning levels (Nagel, 1995). We found that the high cognitive able subjects di er from the low cognitive subjects in the estimated parameters of the EWA models, when one uses the coarse division. Among the ve parameters of the model, we are particularly interested in the one which is normally understood as the capability to do counterfactual reasoning or imagination (the parameter ). We found that of the more cognitive able subjects is signi cantly larger than that of the less cognitive able subjects (0.588 vs. 0.489), and if we further restrict the sample to the tailed 25% subjects, is even low to 0.277. However, this relation between cognitive ability and disappears when the EWA model is applied to version with ner division. This result may lead to a fundamental question concerning the applicability of (generalized) reinforcement learning to the situation when a large number of many possible choices are presented.

Author(s): Shu-Heng Chen, Ye-Rong Du, Lee-Xieng Yang
Topic: Decision Theory: Learning, Psychology and Biology: Cogonition
Back to session: Learning


Lone Wolf or Herd Animal? An Experiment on Choice of Information and Social Learning

John Duffy
University of Pittsburgh

ABSTRACT:

Why do individuals consider information about the choices made by others? One explanation is that that they have intrinsic social preferences for such information –they are innately conformist, or “herd animals”. A second, more instrumental explanation is that the choices made by others may objectively convey more accurate information, e.g., about the true state of the world. We design and report on an experiment that tests these competing explanations. Subjects in our experiment face the problem of correctly guessing the unknown, binary state of the world in each of two periods. The unknown state of the world is known to persist from the first to the second period with probability 0<p<1. Each subject in a group of size n first receives a private, independent, noisy but informative signal regarding the true state of the world in period 1 that is accurate with probability .5<q<1, and they are then incentivized to correctly guess the state of the world in period 1. The rational choice prediction is that they should guess according to their private signal. Prior to period 2 and without knowledge of the true state of the world in period 1, subjects must decide whether to draw a new signal about the true state of the world in period 2, again with accuracy q, (i.e., to act as `lone wolves’) or to observe the guesses made by the other n-1 subjects about the true state of the world in period 1 (i.e., to act as herd animals). After observing this information they are again incentivized to correctly guess the state of the world in period 2. We fixed n and q and studied variations in the persistence parameter, p. When p is sufficiently high in our setting, social information is more accurate than private information so that rational agents should choose to observe the guesses made by the n-1 others and to guess according to the majority opinion. For intermediate values of p, rational agents should chose private information and follow it while for low values of p, rational agents should again choose social information about the guesses made by others but because of the low persistence of the state, they should guess the opposite of the majority opinion about the true state of the world. We find strong support for all three of these predictions in our experimental data. Most of our subjects look to the decisions of others only when it is rational to do so – that is, not because they are conformists or have intrinsic social preferences for such information but because such information may convey useful information about the true state of the world. We also explore individual differences in the choice of information using a personality survey and cognitive tests.

Author(s): John Duffy, Ed Hopkins, Tatiana Kornienko
Topic: Decision Theory: Learning, Decision Theory: Preferences
Back to session: Learning


Legislative Bargaining with Teams

Anthony Bradfield
The Ohio State University

ABSTRACT:

We experimentally investigate the Baron-Ferejohn (1989) model of legislative bargaining, with the addition of two player teams. The main prediction is that teams will outperform individuals according to the demanding truth-wins norm. Further, we use chat between team members to determine whether other regarding preferences are common, in particular looking for confirmation of a type of conditional fairness. Results show that teams meet the truth-wins norm for forming coalitions of minimal size, but there is little evidence for fairness concerns. Other tests show that the failure to play the equilibrium refinement may be due to a low continuation value for the proposer if their proposal is rejected.

Author(s): Anthony Bradfield and John Kagel
Topic: Games: Bargaining,
Back to session: Legislative Bargaining and Committtee Behavior


Cheap-Talk, Back Room Deals and Multilateral Bargaining

Chloe Tergiman
NYU

ABSTRACT:

We design an experiment to show how and why presence and type of communication matters. We use a multilateral bargaining setting, and focus on the specific strategic setup of the divide-a-dollar game between more than two players under different voting rules (majority versus unanimity). Relative to a situation where no communication is allowed, an unrestricted communication protocol leads to more inequality and closer-to-equilibrium outcomes under the majority rule, while it removes all inequality and leads to further-from-equilibrium outcomes under the unanimity rule. Under the majority rule, subjects use the communication tool to engage in back-room deals, while under the unanimity rule, subjects communicate publicly. Private communication channels are used by non-proposers to compete for favors from proposers while public communication channels are used to express preferences for fairness and equality. We further show that in order to reach close-to-equilibrium outcomes in the majority setting, private communication channels are a necessary component of communication protocols.

Author(s): Marina Agranov and Chloe Tergiman
Topic: Games: Bargaining,
Link: http://strategy.sauder.ubc.ca/tergiman/home/Research.html
Back to session: Legislative Bargaining and Committtee Behavior


How Transparency Kills Information Aggregation (And Why That May Be a Good Thing)

Sebastian Fehrler
University of Zurich

ABSTRACT:

We show theoretical and experimental results on the role of transparency in committee decision making and deliberation. We present a model in which committee members have career concerns and unanimity is needed to change the status quo. Transparency leads to a break down of information aggregation, causing more in-correct group decisions. However, if the cost of wrongly changing the status quo is high enough, the principal will be better off in expectation under transparency than under secrecy - he is helped by the failure of information aggregation. We test the model in a laboratory experiment with two member committees playing under three levels of transparency. We observe strong effects of transparency on committee error rates and information aggregation that are largely consistent with the model's pre-dictions. On the individual level, we observe strong effects on deliberative behaviour which go in the predicted direction but are less pronounced than in theory.

Author(s): Sebastian Fehrler and Niall Hughes
Topic: Public Choice: Voting and Rent Seeking, Social Behavior: Communication
Back to session: Legislative Bargaining and Committtee Behavior


Communication in Legislative Bargaining

Andrzej Baranski
Ohio State University

ABSTRACT:

Previous experimental tests of the Baron Ferejohn (BF) model of legislative bargaining show that the qualitative predictions of the stationary subgame perfect equilibrium (SSPE) hold. However the model fails quantitatively regarding the measure of proposer power. In the standard BF setting there is no room for meaningful communication because every allocation is subgame perfect (SPE), but in our design we actually obtain a reduced set of SPE that allows us to test whether preplay communication is credible or not. We show that the set of credible messages coincides with the SSPE predicted allocation. In the presence of communication the proposer's share gradually increases to match very closely the SSPE prediction and chat screens reveal that players are credibly communicating. We provide a quantitative measure of the impact of communication on the share approved by voters which are willing to settle for less in the presence of communication. Fairness concerns vanish quickly and voters compete for a position in the minimum winning coalition by lowering their reservation shares over time and explicitly asking the proposer to exclude the other member. We are able to record the first amount requested by a voter and the final agreed share which allows us to determine the effect of communication in the negotiation process. We reject the hypothesis that communication enhances altruism in our setting. Selfish voting behavior is further confirmed by the non-significance of the coefficient that measures how the other member's share affects ones probability of voting in favor of a given proposal. We provide a treatment with both public chat and private chat, in the sense that voters can communicate with each other without the proposers intervention. We find that only a 20% of voter attempts to collude are successful.

Author(s): Andrzej Baranski and John H. Kagel
Topic: Games: Bargaining, Social Behavior: Other-regarding Preferences
Back to session: Legislative Bargaining and Committtee Behavior


Truth or Guilt: Theory and Experiments on Lying in Games

Lilia Zhurakhovska
University of Erlangen-Nuremberg

ABSTRACT:

Why do people prefer to tell the truth? Guilt aversion- the desire to not let down the other person's expected payoff- is a plausible explanation for why individuals might prefer to tell the truth in settings in which lying is payoff-maximizing. Yet recent studies have found that individuals prefer to tell the truth even when lying is Pareto optimal or helps the other player. This paper is the first to experimentally test whether guilt aversion explains why individuals do not lie. We propose a theory of 'psychological guilt aversion' which depends on lying per se and not on monetary outcomes. Using a novel experimental test, we find that so-called 'lying aversion' is partly explained by guilt aversion, but more robustly explained by our theory of psychological guilt aversion. Gender differences in sensitivity to both types of guilt are discussed.

Author(s): Adam Eric Greenberg, Paul Smeets, and Lilia Zhurakhovska
Topic: Social Behavior: Lying and Cheating,
Link: http://www.economic-theory.rw.uni-erlangen.de/team/academic-staff/lilia-zhurakhovska.shtml
Back to session: Lies and Deception


Broken Promises and Hidden Partnership: Theory and Experiment

Jingnan Chen
ICES, GMU

ABSTRACT:

Previous research shows unenforceable promises can promote both trust and reciprocity. While this can benefit existing exchange, in dynamic environments promises might hinder one's willingness to form new trust relationships, even when doing so would be efficient. This arises in both business and social contexts, including industry non-compete agreements and personal relationship commitment decisions. Despite its practical importance, we are aware of no previous experimental research that informs behavior when facing tension between keeping a previous promise and taking advantage of a new trust opportunity. Here we take a step in that direction. We report data from an experiment using three person trust games with a hidden trust action. In our game, the second mover reciprocates the first mover's trust by not trusting the third mover, and players can make non-binding promises to each other prior to making decisions. We find people are willing to break promises to pursue new trust relationships, but only when the party involved in the new trust makes a promise to cooperate. When promises encouraging new trust are not made, we find that people are overwhelmingly likely to keep old promises and avoid entering new and potentially beneficial hidden trust relationships. Our data provide novel evidence on the role of promises in affecting economic outcomes, and shed new light on the contexts under which promises are likely to be broken.

Author(s): Jingnan (Cecilia) Chen; Daniel Houser
Topic: Games: Other, Social Behavior: Communication
Back to session: Lies and Deception


An experimental study on dishonesty: The effect of ambiguity and framing

Merve Akbas
Duke University

ABSTRACT:

We experimentally examine whether ambiguity and framing of decisions influence the frequency of lying. We use a task in which we vary the difficulty of perceiving the truthful answer and we manipulate the framing of the decision as lying to gain versus lying to avoid a loss. We find that (1) when the decision is ambiguous, loss frame increases the frequency of lying and (2) when the decision is simpler, the level of lying decreases for both frames and the difference between them becomes insignificant. Our results suggest that ambiguous decisions are more likely to involve dishonesty and they are more likely to be influenced by the environment than simple decisions.

Author(s): Merve Akbas and Dan Ariely
Topic: Social Behavior: Lying and Cheating,
Back to session: Lies and Deception


Lying to catch up, or lying to keep up

Quoc Tran
Bridgewater State University

ABSTRACT:

This paper examines the effect of information on misreporting behavior. Subjects participate in a misreporting game: each participant is provided with a high or low initial endowment, and then with two twenty-sided die, and is asked to roll the die, sum the two numbers on the die, and to report the outcome. Payoffs linearly increase in the outcome of the die, and all reporting in anonymous. The subjects are asked to roll the die thirty times (although only a single roll is paid). Expectations are also collected. We conduct four treatments, three of which provide subjects with information about the average endowment level of the group. These treatments vary the average endowment (low, medium, and high). A baseline is also conducted where no information on group average is provided. Using student subjects at the University of Indonesia, we found that poor subjects (with low initial endowment levels) engage in greater misreporting as compared to rich subjects (with high endowment levels), particularly when average endowment information is revealed. Implications for policy are discussed.

Author(s): Ritwik Banerjee, Sheheryar Banuri, Ha Nguyen, and Quoc Tran
Topic: Social Behavior: Lying and Cheating, Social Behavior: Other
Back to session: Lies and Deception


Banking Networks and Information Processing

Ciril Bosch-Rosa
Technical University Berlin

ABSTRACT:

Based on Allen et al. (2011) we run an experiment which investigates how banking networks, information, and short-term credit all interact in financial markets. Using a simple setup with six banks, we investigate whether or not clustered networks are more prone to systemic credit dry-ups even if they are less susceptible to individual bank failures.

Author(s): Ciril Bosch-Rosa, Max Caldwell
Topic: Markets: Finance, Decision Theory: Risk
Back to session: Macroeconomics


Informational Asymmetries in Laboratory Asset Markets with State Dependent Fundamentals

Andreas Markstaedter
Georg-August-Universitat Gottingen

ABSTRACT:

Healthy financial markets rely on asset prices that are consistent with rational expectations and correctly follow the underlying economic growth path. In this paper we use the experimental economics method to investigate the formation of market prices in asset markets with state-dependent fundamentals. We are particularly interested in two informational aspects, (1) the role of traders who are informed about the true state (i.e., insiders) and (2) the provision of Bayesian updates of the assets' state-dependent fundamental value to all traders.

Author(s): Claudia Keser, Andreas Markstaedter
Topic: Games: Information, Markets: Finance
Back to session: Macroeconomics


How Does Transparency Affect Emergent Equilibria in Large Online Macroeconomic Environments?

Jack Rogers
University of Exeter

ABSTRACT:

We present the results from an online experimental environment used for teaching 260 first year undergraduate students during 10-hour sessions across 8 days. Each student interacted with others by logging in any time and place of their choosing, and playing a dual role as both a producer selling a unique (monopolistic) good, and a consumer buying posted offers. The ultimate objective was to accumulate as much utility as possible, which then counted towards their final grade, and overall winners were also incentivised with cash prizes. We characterise the nature of emergent equilibrium behaviour and show that transparency reduces overall average prices and increases aggregate welfare. We argue that this can be explained by a desire to avoid the appearance of selfishness in the transparent environment.

Author(s): Jack Rogers
Topic: Markets: Macroeconomics, Methodology: Experimental Design, Tools and Practices
Back to session: Macroeconomics


Asset Trading and Monetary Policy in Production Economies

Luba Petersen
UBC

ABSTRACT:

In this paper we demonstrate how an experimental general equilibrium economy can be implemented in a laboratory setting in a simplified, time- and cost-efficient manner. We then introduce an asset market to study speculative behavior within a general equilibrium setting where subjects' objective is to maximize their utility from consumption and leisure. Subjects are endowed with units of an asset but are not obliged to participate in the asset market. Asset prices consistently grow above fundamental value and do not decline significantly with learning. Finally, we introduce a policy preventing subjects from borrowing to buy assets. The borrowing constraint does not have any effect on asset prices. Asset market activity has no significant effect on the real economy in either treatment.

Author(s): Guidon Fenig, Mariya Mileva and Luba Petersen
Topic: Markets: Macroeconomics,
Link: http://www.sfu.ca/~lubap/Site/Luba_Petersen_files/Asset%20Trading%20and%20Monetary%20Policy%20in%20Production%20Economies.pdf
Back to session: Macroeconomics


Experiments on Strategic Behavior and Equilibrium Selection in Two-Sided Matching Markets

Ahrash Dianat
George Mason University

ABSTRACT:

We investigate strategic behavior and equilibrium selection in a dynamic version of the Gale-Shapley deferred acceptance algorithm. Previous experimental work on two-sided matching has largely studied the properties of static matching mechanisms, in which the strategy choice faced by laboratory subjects is what preference list to submit to the experimenter. In this context, the evidence is mixed on whether subjects behave according to theoretical predictions and whether subject behavior leads to outcomes that are ex-post stable. In particular, we find that agents react to information in ways that contradict theory. Proposers react to information that is irrelevant and responders ignore information that is potentially valuable. This leads to equilibrium selection effects. Two further experiments, one that further manipulates information and a second that manipulates the size of the market, are implemented to test the reasons behind these findings.

Author(s): Marco Castillo and Ahrash Dianat
Topic: Markets: Market design and Matching,
Back to session: Market Design 1


An Experimental Study of Chinese College Admissions

Ming Jiang
University of Michigan

ABSTRACT:

We experimentally study a student placement problem modeled after the college admissions in China, where students have correlated preferences over colleges and students' admission priorities are determined by test scores. We evaluate in this context the performance of three matching mechanisms: the Boston mechanism, the Deferred Acceptance mechanism, and the Chinese Parallel mechanism. We find that participants are most likely to reveal their true preferences under the DA mechanism, followed by the Chinese Parallel mechanism and then the Boston mechanism. Specifically, we find that under the Chinese Parallel mechanism and the Boston mechanism, participants manipulate in a way that reflects their rankings in test scores. Furthermore, we do not observe significant difference in stability and efficiency across mechanisms when first choices are manipulated in the above way.

Author(s): Yan Chen, Ming Jiang, Onur Kesten
Topic: Markets: Market design and Matching,
Back to session: Market Design 1


An Experimental Test of a Complex Market Design: Changing the Course Allocation System at Wharton

Judd Kessler
The Wharton School

ABSTRACT:

This paper reports on an experiment conducted at the Wharton School of the University of Pennsylvania, testing the efficacy of a new mechanism for assigning students to schedules of courses. The experiment compared the new mechanism, Budish's (2011) approximate competitive equilibrium from equal incomes (CEEI), to Wharton's current mechanism, a fake-money auction that Wharton introduced in 1996, and which was subsequently adopted by numerous other professional schools. In the experiment, CEEI outperforms Wharton's course auction on quantitative measures of efficiency and fairness, as well as qualitative measures of perceived strategic simplicity and student satisfaction. The experiment was also successful in the Roth (1986) sense of ??whispering in the ears of princes??, both in that it persuaded the Wharton administration to adopt CEEI beginning in academic year 2013-2014, and in that results from the experiment have guided aspects of the real-world implementation.

Author(s): Eric Budish and Judd Kessler
Topic: Markets: Market design and Matching, Methodology: Experimental Design, Tools and Practices
Back to session: Market Design 1


Behavioral Market Design: Improving Welfare on Focal--Rather than Equilibrium--Paths

Daniel Fragiadakis
Stanford University

ABSTRACT:

Well-documented in mechanism design are the theoretical tradeoffs between welfare and incentive compatibility. In many environments, the ``cost of strategyproofness'' is high, motivating the use of manipulable mechanisms, many of which have good theoretical properties when subjects play equilibria. Recent behavioral game theory experiments, however, cast doubt on subjects' abilities to find equilibria in the lab. In this paper, we relax the assumption that agents always play equilibria, taking a new behavioral approach to market design. We develop an algorithm that (1) illustrates the costs and benefits of making different non-truthful reports, (2) has focal strategies that balance the tradeoffs, allowing us to make behavioral predictions, and (3) yields levels of welfare near that of a social planner under the predicted strategies. Over half of our subjects conform to prediction, leading to subjects earn more than under strategyproof algorithms. We compare our algorithm to two field mechanisms as well. Field algorithm 1 yields only slightly lower welfare than our mechanism, but submitted preferences are much further from the truth under field algorithm 1 compared to ours, giving the market designer a less accurate evaluation of the algorithm's performance. Field algorithm 2's submitted preferences represent true preferences roughly as well as ours, but has significantly lower welfare.

Author(s): Daniel E. Fragiadakis, Peter Troyan
Topic: Markets: Market design and Matching,
Back to session: Market Design 2


Matching and Chatting: An Experimental Study of the Impact of Network Communication on School-Matching Mechanisms

Tingting Ding
New York University

ABSTRACT:

While, in theory, the school matching problem is a static non-cooperative one-shot game, in reality the matching game is played by parents who choose their strategies after consulting or chatting with other parents in their social networks. In this paper we compare the performance of the Boston and the Gale-Shapley mechanisms in the presence of chatting through social networks. Our results indicate that allowing subjects to chat has an important impact on both the strategies that our subjects choose and their welfare. This implies that encouraging parents to talk or offering parental education may be welfare enhancing.

Author(s): Tingting Ding and Andrew Schotter
Topic: Markets: Market design and Matching,
Back to session: Market Design 2


Lying through Their Teeth: Third Party Advice and Truth Telling in a Strategy Proof Mechanism

Pablo Guillen
The University of Sydney

ABSTRACT:

We test the effect of advice on the well known top trading cycles (TTC) matching algorithm in a school choice frame work. We compare three treatments involving third party advice [right advice (R), wrong advice (W), and both right and wrong advice (RW)] to a no-advice baseline (B). In line with previous literature the truth telling rate is higher than 80% in the baseline, but it becomes as low as 35% in the W treatment. Truth telling rates are also significantly lower in R than in B, and much lower in RW than in B. This evidence suggests that a vast majority of participants in our experiment were confused. Truth telling seems to work only as a default strategy, and participants can be heavily influenced by advice. The real life implementation of matching mechanisms may have been misguided by some laboratory experimentation.

Author(s): Pablo Guillen, Alexander Hing
Topic: Markets: Market design and Matching, Decision Theory: Bounded Rationality
Link: http://hdl.handle.net/2123/9255
Back to session: Market Design 2


The Strategic Formation of Networks: Experimental Evidence

Arya Gaduh
University of Arkansas

ABSTRACT:

We use a laboratory experiment to explore dynamic network formation in a six-player game where link creation requires mutual consent. The analysis of network outcomes suggests that the process converges to the pairwise-stable (PWS) network when a PWS network exists and not to converge to any network when a PWS network does not exist. However, convergence is far more frequent for strongly stable networks. The analysis at the single choice level indicates that the percentage of myopically rational behavior is high. Deviations from myopic rationality are more prevalent in early turns, when marginal payoff losses are small and when deviations involve excessive links that may be removed unilaterally later on.

Author(s): Juan Carrillo, Arya Gaduh
Topic: Games: Networks,
Back to session: Networks


Centrality, Control and Exclusion in Network Social Dilemmas

David Rojo Arjona
Chapman University

ABSTRACT:

We often observe individuals in central positions who control the flow of benefits within a group or the distribution of final output. We experimentally test the effects of such control on cooperation within a group when players can exercise the disciplining power of exclusion from the group. First we examine a public goods game where a central player connects two sub-groups thus increasing the returns for both sub-groups. Second, we investigate a group trust game where the central player additionally decides how much of the group output to distribute to the rest of the group. Generally, central and peripheral players present asymmetric contributions and this asymmetry depends on the game. In particular, the central player free-rides more than the peripheral players in the public goods game but this is reversed in the trust game. And although, in the trust game, central players return similar amounts to the public good game, result show that periphery players in trust games exclude central player disproportionally more often. This suggests that peripheral subjects judge central control perverse.

Author(s): David Rojo Arjona, Arthur Schram, Abhijit Ramalingam and Boris van Leeuwen
Topic: Games: Networks, Public Choice: Public Goods and Common Pool Resource
Back to session: Networks


Hysteresis of Network Formation

Julian Romero
Purdue University

ABSTRACT:

We implement a network formation experiment using the set-up of Bala and Goyal (2000) where the cost of link formation changes continuously over time. For a large range of cost values, there are multiple equilibrium networks, some of which have large variations in overall efficiency. Our aim is to study the equilibrium selection process and determine whether there is hysteresis. The presence of hysteresis has important implications on equilibrium selections, and therefore overall efficiency in the network. In particular, a temporary change in the cost can cause a significant increase in overall efficiency.

Author(s): Kory M. Garner and Julian Romero
Topic: Games: Networks, Games: Coordination
Back to session: Networks


Growing the Right Stars: A Laboratory Study on Network Formation with Heterogeneous Agents

Rong Rong
Weber State University

ABSTRACT:

In a star network, small numbers of agents gather information and distribute it to a large group. Theoretical studies suggest that stars are the most efficient network structure to disperse information (Galeotti and Goyal, 2010). However, previous experimental evidence shows that very few stars emerge in the laboratory due to severe coordination failure (Goeree et al, 2008). To facilitate coordination, an earlier study of ours looked into the effect of institutions on star network formation. We successfully identified two institutions that increase the frequency of stars: the one that is characterized by budget constraint and intellectual property right protection (Rong and Houser, 2013). In this paper, we further investigate the effect of these two institutions in an environment with heterogeneous agents and imperfect information. The results show that these two institutions not only increase the frequency of stars but also allow people to coordinate on the right star (the most efficient star).

Author(s): Rong Rong, Daniel Houser
Topic: Games: Networks, Games: Coordination
Back to session: Networks


The impact of glucose administration on Bayesian v. heuristic based choice

David Dickinson
Appalachian State University

ABSTRACT:

We examine the impact of glucose on a Bayesian choice task that creates a separating equilibrium between high-level Bayesian choice and lower-level reinforcement heuristic choice. Consistent with a dual systems framework, we hypothesize that glucose administration will both increase reaction times and improve Bayesian accuracy because it should shift decision making towards the more deliberate system 2 and away from the more automatic system 1 decision process. We study 113 subjects randomly assigned to either a glucose or placebo drink condition, who make choices over several incentivized easy and difficult choices of the Bayesian task. Our results indicate a significant glucose effect on reaction times. In this case, glucose administration has a main effect of increasing reaction time, as predicted, but glucose also improves the marginal decrease in reaction times experienced across trials. We find that glucose administration significantly increases the likelihood of Bayesian choice over reinforcement heuristic-based choice for those subjects indicating an above-average comprehension of the task structure. Additionally, we find that Bayesian choice likelihood increases across trials on easy choices when administered glucose. We interpret this, as well as the reaction time result, as evidence that glucose may improve learning, particularly on easy tasks. Together, these results suggest that there is a beneficial impact of glucose on deliberative decision making, though some of the results depends on the difficulty of the task and the comprehension of the decision environment being faced.

Author(s): David L Dickinson, Todd McElroy, Nathan Stroh
Topic: Psychology and Biology: Cogonition, Decision Theory: Bounded Rationality
Back to session: New Data: Inference from Decision Times


Intuitive generosity and error prone inference from decision time

Maria Recalde
University of Pittsburgh

ABSTRACT:

In an attempt to better understand what drives decisions in strategic environments researchers have begun to draw inference from the time it takes individuals to make decisions. Examining decision times in a series of games Rand, Greene, and Nowak (2012) find that subjects who reach their decisions relatively quickly are more generous than those who take time to decide. This negative correlation between decision time and giving causes them to infer that cooperation is intuitive, and that time is needed for individuals to act selfishly and reject the instinctive impulse to cooperate. We report data demonstrating that the inference on decision time is sensitive to the environment in which cooperation is measured, and argue that confusion is a strong confound when the equilibrium lies on the boundary of the strategy space. We examine behavior in two public goods games with interior Nash equilibria in dominant strategies, one with an equilibrium toward the bottom of the strategy space and one with an equilibrium toward the top of the strategy space. Looking at the game with a low interior Nash equilibrium, we replicate Rand et al.'s finding that participants who make decisions quickly tend to be more cooperative. However the correlation between time and contributions reverses when the equilibrium contribution is high. In this case we find that individuals who make decisions quickly are less generous than those who take time to make their choices. Our results suggest that the negative correlation between decision time and generosity documented in the literature may not reflect `spontaneous giving,' but rather the fact that confused participants quickly select a contribution that lies, on average, in the middle of the strategy space.

Author(s): Maria P. Recalde, Arno Riedl, Lise Vesterlund
Topic: Applied Economics: Charitable Giving, Methodology: Experimental Design, Tools and Practices
Back to session: New Data: Inference from Decision Times


Combining Response Times and Choice Data Using a Neuroeconomic Model of the Decision Process Improves Out-of-Sample Predictions

John Clithero
California Institute of Technology

ABSTRACT:

A basic problem in empirical economics involves using data from one domain to make out-of-sample predictions for a different, but related environment. When the choice data are binary, a canonical method for making these types of predictions is the logistic choice model. We investigate if it is possible to improve out-of-sample predictions by changing two aspects of the canonical approach: 1) Using response times in addition to the choice data, and 2) Combining them using a model from the neuroeconomics literature, called the Drift-Diffusion Model (DDM). We compare the out-of-sample prediction accuracies of both methods using both simulations and real experimental data. In both cases we find that the DDM method outperforms the Logit prediction method. Furthermore, although the improvement in prediction accuracy is small for the case in which items have very similar or very different values, the DDM method improves prediction accuracy substantially for intermediate cases.

Author(s): John A. Clithero and Antonio Rangel
Topic: Psychology and Biology: Neuroeconomics, Methodology: Other
Link: http://www.hss.caltech.edu/~clithero/Paper_ForESA.pdf
Back to session: New Data: Inference from Decision Times


Rationalizing Context-Dependent Preferences: Divisive Normalization and Neural Constraints on Decision-Making

Ryan Webb
New York University

ABSTRACT:

The physiology of neural systems places constraints on the form of neural computation. In particular, divisive normalization is a canonical feature of neural computation which allows the brain to efficiently code sensory information, and has now been observed in neural signals which encode valuation and implement decision-making. We demonstrate that a neuroeconomic choice model employing normalization provides an efficient means for the brain to represent valuations given the constraints of a neural system, yielding the fewest possible choice errors. However, normalization predicts the composition and size of the choice set will influence choice behaviour, including seemingly adverse violations of the axiom of Independence of Irrelevant Alternatives and increased choice inconsistency. We present evidence from two lab experiments which illustrate such behaviour, and demonstrate that divisive normalization must be explicitly included in the econometric specification of a behavioural model. This emphasizes the role neurophysiological constraints play in decision-making, and provides a normative explanation for some types of context-dependent preferences.

Author(s): Ryan Webb; Paul W. Glimcher; Kenway Louie
Topic: Psychology and Biology: Neuroeconomics, Decision Theory: Bounded Rationality
Back to session: New Data: Inference from Decision Times


Harnessing the Power of Norm

Carrie Wenjing Xu
University of Michigan

ABSTRACT:

The impact of norms has been widely studied in decision making processes. Drawing attention to norms can induce more socially appropriate behaviors. However, when norms are not explicitly defined, how people perceive norms may be biased and depend on identities and actions. Therefore, it is important to find the best way to draw attention to norms and nudge behaviors. We design an experiment to study how social norms can be cued and manipulated and the effect on behaviors. We use Dictator Game and vary when subjects are asked to think about the socially appropriate allocation decisions a dictator should make. Results suggest that asking for norms at the beginning of the game, before subjects are assigned as dictator or recipient, can greatly increase fair splits in the Dictator Game. The variance in norm perception is also the least among all the treatments. This is critical for policy interventions. We need to choose the right way to harness the power of norms and induce desirable behaviors. Unlike experiments that ask about beliefs, the experimental design also controls for experimental demand effect.

Author(s): Carrie Wenjing Xu
Topic: Social Behavior: Norms and Morals, Applied Economics: Charitable Giving
Back to session: Norms


Ignorance and Ethical Behavior: Lessons from the Lab

Albena Neschen
University of Cologne

ABSTRACT:

In thepaper we examine the nature of strategic ignorance in a simple dictator game task taking two perspectives: Firstly, from a business ethics view, we will argue how people should manage the trade-off between their and the interests of others in a situation where they can remain ignorant about the consequences of their actions. Secondly, by evaluating data from the laboratory, we will illustrate how people actually handle their responsibility for others dependent on their accountability for remaining ignorant. Our laboratory experiment is based on a series of simple one shot dictator games. In our experimental treatments, we systematically vary the opportunity for the dictator to be informed about the receiver's endowment. Our findings show that dictators who intentionally remain ignorant mainly do so in order to exploit the receiver.

Author(s): Albena Neschen, Bernd Irlenbusch, Gari Walkowitz, Julian Conrads
Topic: Social Behavior: Norms and Morals, Games: Information
Link: albena.neschen@uni-koeln.de
Back to session: Norms


Countering Manipulation in Prediction Markets? The Role of Incentives

Meng-Chien Su
National Taiwan University

ABSTRACT:

Manipulation is one of the biggest concerns for using prediction markets to predict future events such as elections and terrorist attacks, since various interest parties have the incentives to mislead forecaster. Unfortunately, legal restrictions prevent most prediction markets from providing incentives for traders to arbitrage and counter manipulation. We consider four different incentive schemes and evaluate their ability to aid prediction markets in resisting manipulation. In this laboratory prediction market where informative traders trade state-claims and observers of the market attempt to forecast the true state, we find that uninformative manipulators can successfully mislead both final market price and forecasters' predictions when the prediction markets uses only play money, though they fail in real money markets. Using lottery tickets provide similar resistance to manipulation as real money, while rank-order tournaments perform as bad as play money. This suggests organizing sweepstakes (but not performance contests) as a legal way to provide incentives for prediction markets to deter potential manipulation.

Author(s): Meng-Chien Su, Joseph Tao-yi Wang
Topic: Special Topic: Experimental Payments, Markets: Market design and Matching
Back to session: Norms


Follow the Crowd or Your Conscience: an Experimental Study of Norms and Social Preferences

Michael Kurschilgen
Max-Planck Institute for Research on Collective Goods

ABSTRACT:

In a modified dictator game, I study the effect of conformity and morality on other-regarding preferences. Conformity requires information about others' behavior. Morality requires self-reflection about one's own moral standards. I find that self-reflection makes people both less selfish and, even more interestingly, more heterogeneous. The latter results from people having competing normative concerns (efficiency vs. equality). Self-reflection makes them more extreme. In contrast, information has almost no effect on people's revealed preferences while having a strong effect on people's moral judgments. The experimental results can be well explained within the framework of identity utility (Akerlof & Kranton 2000).

Author(s): Michael Kurschilgen
Topic: Social Behavior: Norms and Morals, Social Behavior: Other-regarding Preferences
Back to session: Norms


Do people anticipate loss aversion

Sally Sadoff
UCSD

ABSTRACT:

Despite a large literature demonstrating loss aversion, there is little evidence on a central prediction of these models: people prefer gain frames to loss frames. Previous work has measured behavior of subjects assigned to either a loss or a gain frame, but has not measured preferences for the loss or gain frame given the option. We elicit willingness to pay to participate in an effort task in which the incentives are framed either as a gain or as a loss. We then measure effort in the task to test whether, as the model predicts, people prefer the gain frame but work harder in the loss frame. In addition to the theoretical literature on loss aversion, our study informs an empirical puzzle: despite evidence that loss frames are more effective, most workplace contracts are framed as bonuses. Observational studies have suggested that this finding may be due to employee preferences for gain frames but there is little experimental evidence identifying this potential mechanism.

Author(s): Alex Imas, Sally Sadoff, Anya Samak
Topic: Decision Theory: Preferences,
Back to session: Reference Dependent Preferences


Uncertainty and the WTA-WTP Disparity for Private and Public Goods

Jens Schubert
Virginia Commonwealth University

ABSTRACT:

Abstract We study the robustness of the willingness to accept - willingness to pay disparity for private and public good lotteries in the presence of uncertainty in settings where decision makers move from one state of uncertainty to another. The scenarios that we create in the laboratory mimic public policy scenarios where a proposed policy not only increases expected benefits but also affects the relative risk of outcomes. In particular, we vary the level of expected benefit increases and the level of relative risks. Our experiments explore the existence of the disparity and its interdependence with uncertainty and social preferences. In the public good lottery setting, the changes in relative expected benefits and risks differ across individuals in a group. To our knowledge, our study is the first experimental test that comprehensively addresses the impact of uncertainty on the WTP-WTA disparity in the context of public goods. Our findings suggest that the disparity persists for both private and public good lotteries. While the experiments provide significant evidence for social preferences in the public good setting, social preferences in the public good setting cannot explain the size of the disparity.

Author(s): Jens Schubert, Christian Vossler, Michael K. Price
Topic: Public Choice: Public Goods and Common Pool Resource, Decision Theory: Risk
Back to session: Reference Dependent Preferences


Deciding for Others Reduces Loss Aversion

Erik Wengstrom
Lund University

ABSTRACT:

We study risk taking on behalf of others, both with and without potential losses. A large-scale incentivized experiment is conducted with subjects randomly drawn from the Danish population. On average, decision makers take the same risks for other people as for themselves when losses are excluded. In contrast, when losses are possible, decisions on behalf of others are more risky. Using structural estimation, we show that this increase in risk stems from a decrease in loss aversion when others are affected by their choices.

Author(s): Ola Andersson, Hokan J. Holm, Jean-Robert Tyran and Erik Wengstrom
Topic: Decision Theory: Risk, Social Behavior: Other
Back to session: Reference Dependent Preferences


Examining when Reference Points Lead to Negative Wage Elasticities in Work and Volunteer Tasks

Christine Exley
Stanford University

ABSTRACT:

Theories of reference dependence have the perhaps surprising implication that labor supply can go down with increased wages, a negative wage elasticity that is not consistent with the neoclassical model. This study seeks to directly test theory's prediction of a negative wage elasticity in a laboratory experiment and to uncover the contexts under which reference points matter for labor supply. With a simple design closely following Abeler, et al (2011), participants complete an effort task before realizing whether they will be paid (1) a fixed payment of $8, or (2) a variable payment that is equal to their effort times their wage rate. There is an equal probability of either payment outcome occurring. Crucially, we vary the wage itself in our experiment rather than the reference point, which allows us to draw direct conclusions about wage elasticities in the presence of reference-dependence. We also vary whether individuals earn money for themselves or for a charity, since a growing body of work shows that the impact of incentives can be quite distinct in these two environments. Based on the preliminary results of this experiment, we expect to confirm two results. First, in both environments, there exists a range where participants complete less of the task on average when wages are increased, directly providing evidence of negative wage elasticities in a reference-dependent context. Second, this range is larger when individuals are earning money for a charity.

Author(s): Christine Exley and Stephen Terry
Topic: Applied Economics: Charitable Giving, Social Behavior: Other-regarding Preferences
Back to session: Reference Dependent Preferences


Play it Again: Partner Choice, Reputation Building and Learning in Restarting Finitely-Repeated Dilemma Games

Kenju Kamei
Bowling Green State University

ABSTRACT:

Previous research has shown that opportunities for 2-sided partner choice in finitely repeated social dilemma games can promote cooperation through a combination of sorting and opportunistic signaling, with late period defections by selfish players causing an end-game decline. In six treatments that vary the cooperation premium and the informational support for reputation formation, we let each subject play multiple voluntary contribution super-games to study the competing forces of (a) learning about the benefits of reputation, and (b) learning about backward unraveling. With a high cooperative premium and good information, investment in reputation grows across super-games, although end-game unraveling also begins earlier and becomes more pronounced.

Author(s): Kenju Kamei and Louis Putterman
Topic: Public Choice: Public Goods and Common Pool Resource, Games: Repeated Games
Back to session: Repeated Games 1


Voluntary Contribution Mechanism Played over an Infinite HorizonVoluntary Contribution Mechanism Played over an Infinite Horizon Voluntary Contribution Mechanism Played over an Infinite Horizon

Lijia Tan
Wang Yanan Institute for Studies in Economics

ABSTRACT:

Recent empirical evidence shows that cooperation improves when games are infinitely repeated. We experimentally investigate the evolution play in an infinitely repeated Voluntary Contribution Mechanism (VCM). We establish a baseline by also conducting finitely repeated VCM sessions. We find that in infinitely repeated VCM games (1) average contributions are similar to that of finitely repeated VCM games when subjects are inexperienced; (2) most groups have a non-monotonic trend of contribution with repetition; and (3) contributions remain at the same level after an unexpected restart in contrast to a significant increasing of contributions in finitely repeated VCM games. The data provides strong support for heterogeneous subjects, which may explain the non-monotonic trend of average contributions in infinitely repeated VCM games. This non-monotonic trend is caused by one category of subjects who expect others to contribute in period t as they did in period t-1.

Author(s): Lijia Tan, Lijia Wei
Topic: Public Choice: Public Goods and Common Pool Resource, Social Behavior: Group Behavior
Link: http://papers.ssrn.com/sol3/papers.cfm?abstract_id=2232925
Back to session: Repeated Games 1


Dump, Date, or Marry: Endogenous Group Formation with Varied Contract Length

Sean D'Evelyn
Loyola Marymount University

ABSTRACT:

This paper examines an experimental setting inspired by the marriage market. During the experiment, subjects are divided into pairs with the ability to revise pairings. Each period, subjects make two decisions. The first decision is whether or not to take a prosocial action that benefits their partner at a cost to themselves. The second decision determines their relationship status. In the Dump-Date-Marry treatment, subjects have the option of either being rematched (dumping), remaining with their current partner (dating), or making their current partnership permanent (marrying). The other two treatments are the Dump-Date treatment and the Dump-Marry treatment where one of the relationship options is not available. For my theoretical predictions, I look at the set of sequential equilibria in each treatment. As predicted by the set of sequential equilibria, prosocial actions are most frequent in the Dump-Marry treatment, but there is no significant difference in prosocial actions between the primary treatment and the Dump-Date treatment. Prosocial actions decline over time in all treatments. Contrary to theory, a majority of subjects choose to get married early in the Dump-Date-Marry treatment even though such an action is not part of a sequential equilibrium.

Author(s): Sean D'Evelyn
Topic: Social Behavior: Other, Social Behavior: Other-regarding Preferences
Back to session: Repeated Games 1


Flexibility and collusion with imperfect monitoring

Jan Potters
Tilburg University

ABSTRACT:

Flexibility - the ability to react swiftly to others' choices - facilitates collusion by reducing the gains from defection before opponents react. Under imperfect monitoring, however, flexibility may also hinder collusion by inducing punishment after too few noisy signals. The combination of these forces predicts a non-monotonic relationship between flexibility and collusion. We implement an indefinitely repeated game with noisy information about opponents' action. We vary how long players have to wait before they can change their output. The results lend at best limited support to the predicted non-monotonicity when collusion is tacit, and no support when collusion is facilitated by communication.

Author(s): Maria Bigoni, Jan Potters, Giancarlo Spagnolo
Topic: Markets: Industrial Organization, Social Behavior: Communication
Back to session: Repeated Games 1


The Impact of Monitoring in Infinitely Repeated Games: Perfect, Public, and Private

Guillaume Frechette
New York University

ABSTRACT:

In this paper we propose an experimental design to compare the impact of monitoring on cooperation in infinitely repeated games. More specifically, we study a prisoner's dilemma, with randomly termination, in the laboratory. Three monitoring structures are investigated, while keeping fixed the strategic form of the stage game: perfect monitoring, imperfect public monitoring, and imperfect private monitoring. Under perfect monitoring, payoffs are stochastic (as under imperfect monitoring), but actions are observed. Under imperfect monitoring, each agent observes a noisy signal of their opponent's action, but not the action itself. In the public monitoring case, the signals received are common knowledge; under private monitoring, a player does not know the signal observed by the opponent. We explore the choices and strategies used under each of these information structures and relate the observed behavior to the important forces present in theory. We find that subjects are able to sustain cooperation under imperfect monitoring, even when monitoring is private.

Author(s): Masaki Aoyagi, V. Bhaskar, and Guillaume Frechette
Topic: Games: Repeated Games,
Back to session: Repeated Games 2


A Comment on Cycles and Instability in a Rock-Paper-Scissors Population Game: A Continuous Time Experiment

Siqian Zhu
Zhejiang University

ABSTRACT:

In Timothy N. Cason, Daniel Friedman and Ed Hopkins's recent paper, Cycles and Instability in a Rock-Paper-Scissors Population Game: A Continuous Time Experiment (The Review of Economic Studies, 2013), the authors claim that control treatments (using simultaneous matching in discrete time) replicate previous results that exhibit weak or no cycles. In this comment, we point out that the computing method of counting CCT (counter-clockwise transits) and CT (clockwise transits) is erroneous, which leads to the incorrectness of a crucial index, the cycle rotation index. Additionally, after we rectify the computing method and replace the former null hypothesis of Wilcoxon Signed Ranks test with a more explicit but interchangeable one, the result shows that one of the two Discrete treatments for the Stable game S exhibits significant cyclical behavior, contradicting the authors' previous statement.

Author(s): Siqian Zhu
Topic: Games: Repeated Games, Decision Theory: Learning
Back to session: Repeated Games 2


Testing Global Games, Risk Dominance, and Payoff Dominance in Repeated Global Stag Hunt Games

Ajalavat Viriyavipart
Texas A&M University

ABSTRACT:

The theory of global games has shown that in any 2X2 game, the players should conform to Harsanyi and Selten's risk dominance criterion. We test this prediction in a global stag hunt game. The results show that different cohorts coordinate on different thresholds and we observe small differences between sessions with private and common information. In sessions with private information, a few cohorts coordinate on thresholds close to the global games prediction but many cohorts coordinate on thresholds close to the payoff-dominant threshold. We argue that historical accident and dynamic process play a key role in forming mutually consistent expectations in this game.

Author(s): John Van Huyck and Ajalavat Viriyavipart
Topic: Games: Coordination, Games: Repeated Games
Back to session: Repeated Games 2


Repeated Battle of Sexes: Experimental Evidence and Individual Evolutionary Learning

John Ledyard
Caltech

ABSTRACT:

We study experiments with human subjects in the repeated Battle of the Sexes game, under full and under partial information about other player's payoffs and with symmetric and asymmetric payoffs. We observe higher rates of alternation, where subjects are alternating between two pure strategy Nash equilibria, every two periods, in a full information, symmetric payoffs treatment (around 45%) and the lowest (around 5%) in the partial information (no information about the other player's payoffs), asymmetric payoffs treatment. A fraction of players exhibit patterns of alternation with longer horizon (around 15%) If pairs of subjects do not alternate every two periods, their behavior converges to one of the Nash equilibria. Finally, there is a fraction of pairs that exhibit neither alternating nor the Nash equilibrium behavior. We construct a more 'sophisticated' Individual Evolutionary Model (IEL) which has an ability to evolve more complex strategies that include alternating behavior. This turns out not to be enough to evolve alternating behavior. The model initialized with randomly generated strategies of various degree of complexity converges to one of the Nash equilibria. However, if we start our artificial agents out with 'some good' ideas, the model evolves to alternation with frequencies that match those of the experimental data.

Author(s): Jasmina Arifovic and John Ledyard
Topic: Games: Repeated Games,
Back to session: Repeated Games 2


Cooperation in Stochastically Evolving Environments: An Experimental Study of Public Information in Markov Games

Andrew Kloosterman
New York University

ABSTRACT:

A Markov game generalizes an infinitely repeated game by allowing the stage game played each period to differ. Instead of repeating the same stage game every period, the stage game played each period is determined by a Markov process. This paper presents a laboratory experiment that tests one result from a companion theory paper, Kloosterman 2013, which shows that less accurate public information about future stage games can lead to higher welfare achievable by players through play of an equilibrium. The experimental game is an asymmetric partnership game where two players must both exert costly effort to complete a project that is beneficial to only one of them. Subjects are indeed responsive to the signal, but less accurate information does not increase the frequency of the cooperative effort choice (and therefore welfare) as theoretically predicted. The paper also presents results from a preliminary experiment, with no information, regarding how different expectations about the future affect initial choices in a stochastic prisoner's dilemma environment. Consistent with theory, experienced subjects learn to cooperate in the initial period when they expect the future to most likely consist of dilemmas where a large punishment is possible.

Author(s): Andrew Kloosterman
Topic: Games: Repeated Games,
Back to session: Repeated Games 3


Endogenous Incomplete Contracts: Theory and Experiments

Steven Wu
Purdue University

ABSTRACT:

While it is well known that incomplete contracts can create inefficiencies in one-shot settings, economists know less about how the *degree* of contractual incompleteness affects trading outcomes when traders are able to repeat trade. In contrast to the received literature on incomplete contracts, we develop a model that endogenizes the degree of contractual incompleteness rather than assume that contractual form is exogenously imposed. While our model replicates the standard finding that increasing the degree of incompleteness leads to greater inefficiencies in one-shot trading, once repeat trading is allowed and if the discount factor is sufficiently high, then a high degree of contractual incompleteness becomes negatively associated with inefficiency thereby reversing the standard result. We also extend the model by introducing a monitoring technology that can partially verify performance. This enables partial enforcement of contracts (e.g. a third-party can determine whether the traded good meets some minimally acceptable standard, but fine measurement of quality beyond that is not possible). Partial variability is consistent with many monitoring technologies in practice. The model predicts that, in low-discount rate environments, partial enforcement would be useful in preventing the collapse of trade and ensuring at least minimally acceptable quality goods are traded. However, in high discount-rate environments, partial enforcement may crowd out relational contracts that facilitate the trade of high quality goods leading to an overall reduction in efficiency.

Author(s): Nisvan Erkal and Steven Y. Wu
Topic: Games: Repeated Games,
Back to session: Repeated Games 3


An Experimental Investigation into the Success or Failure of Joint Ventures

Kyle Hyndman
University of Texas at Dallas

ABSTRACT:

Joint ventures between firms are very popular, yet many studies suggest that the chances of success are, at best, 50-50. Prominent reasons for failure include unmet expectations, opportunistic behavior by partners and mistrust among others. We conduct a series of experiments to understand the relative importance of these factors. In our experiments, subjects are able to choose the length of their interaction (one period, a fixed, finite number of periods, or indefinitely repeated) and also whether to include an exit clause which allows subjects to terminate their relationship early. Results suggest that indefinitely binding agreements with the option to exit within the first two periods yields the best outcomes as those matches which are not profitable break-up,while those that are continue.

Author(s): Kyle Hyndman and Dorothee Honhon
Topic: Games: Repeated Games, Games: Coordination
Back to session: Repeated Games 3


The Role of Information in Explaining Decoy Effects

Salar Jahedi
University of Arkansas

ABSTRACT:

The inclusion of an irrelevant alternative into the choice set is known to influence behavior. Specifically, consumers are more likely to purchase an item when it is placed alongside worse-looking alternatives, or decoys, than when it is offered alone. I test the extent to which decoys matter in environments with varying degrees of uncertainty. With high uncertainty about product quality, decoy effects appear quite large and are consistent in size with many marketing studies. Once product quality is fully known, the decoy effect persists but is much smaller in size.

Author(s): Salar Jahedi
Topic: Decision Theory: Preferences, Decision Theory: Bounded Rationality
Link: http://comp.uark.edu/~sjahedi/bargains.pdf
Back to session: Risk Preference


Dictating the Risk

Marina Schroder
University of Magdeburg

ABSTRACT:

We introduce the risk dictator game that allows us to elicit an individual's social preferences for increasing or decreasing the risk that another individual bears. The cost of modifying the other individual's risk is an increased own risk. We modify the domain of payoffs using binary lotteries in gains or losses, but keeping dictators always at least as well-off as receivers. We find that about half of the dictators keep the own risk at a minimum, neither increasing nor decreasing the receivers risk. About one fourth of the dictators exhibit purely pro-social preferences for decreasing the receivers risk. About one fourth of the dictators exhibit both pro-social and anti-social preferences, sometimes increasing and sometimes decreasing the risk the receivers bear. We observe only very few purely anti-social dictators, who only increase the receivers risk. Surprisingly, the distribution of pro-social and anti-social preferences seems independent from the relative expected income situation. Dictators do not reduce the risk of receivers, who are threatened by a loss, more than those, expecting a positive outcome.

Author(s): Abdolkarim Sadrieh and Marina Schroder
Topic: ,
Back to session: Risk Preference


Unexpected Utility: Testing Expected Utility and Its Alternatives using Convex Risk Budgets

James Andreoni
University of California, San Diego

ABSTRACT:

We reconsider the experimental literature on risk preferences with a new elicitation device in which subjects choose a probability p of winning a prize x, where p and x lie on a linear budget constraint, r_1*p + r_2*x = m. Note that the choice space is smooth, continuous, and all options involve risk. With this framework we derive a number of strict tests of alternative theories. Our analysis assumes simply that preferences depend on the probability p and prize x; U = u(p; x), and employ mostly non-parametric tests that reveal important aspects of preferences. The data offer a clear rejection of Cumulative Prospect Theory, but is consistent with Disappointment Aversion. However, neoclassical Expected Utility theory, even when restricted to CRRA utility, performs unexpectedly well.

Author(s): James Andreoni and William Harbaugh
Topic: Decision Theory: Risk,
Link: http://econ.ucsd.edu/~jandreon/WorkingPapers/AndreoniHarbaughUU.pdf
Back to session: Risk Preference


Cultural Persistence: The Ancient Origins of Risk Preferences

Benjamin Enke
University of Bonn

ABSTRACT:

What precisely does culture mean and how persistent is it? Utilizing representative data on risk preferences from 73 countries, we study the origins of the large heterogeneity in attitudes towards risk both within and across societies. We show that the distribution of individual traits bears a significant association with human genetic variables, which have their origins thousands of years ago and constitute natural proxies for persistent cultural heterogeneity. Specifically, we establish that differences in risk attitudes between countries bear a strong relationship to their genetic distance. In addition, as a within-country analogue, the heterogeneity in risk attitudes within a population is related to the diversity of the respective genetic pool. These findings suggest that the distribution of one of the key economic traits is not random, but rather traces back to human migration patterns in ancient times. This implies that attitudes towards risk should be understood as highly persistent parts of a country's long-run culture.

Author(s): Anke Becker, Thomas Dohmen, Benjamin Enke, Armin Falk
Topic: Decision Theory: Preferences,
Back to session: Risk Preference: Social Effects


Other Regarding Preferences and Risk Taking

Rudy Santore
University of Tennessee

ABSTRACT:

We report the results of a group experiment designed to test whether a subject's incentive to take a monetary gamble depends on other subjects in the group taking an identical gamble. Under the standard assumption of self-interest, subjects' decisions should be independent of other subjects' incomes. However, if subjects have other-regarding preferences, then the incentive to take a gamble can be shown to depend on whether other subjects in the group have uncertain incomes. To see why, suppose that subjects A and B have identical initial incomes and that A is inequality averse. Any gamble taken by B induces ex post inequality and exerts a negative externality on A. But if A takes the same gamble as B, there will be no ex post inequality. Thus, inequality aversion should, conditional on B taking a gamble, increase the incentive for A to take the same gamble.

Author(s): Rudy Santore, Stephen Cotten, and Justin Roush
Topic: Social Behavior: Other-regarding Preferences, Decision Theory: Risk
Back to session: Risk Preference: Social Effects


Following the Followers: The Dynamics of Risky Preferences and the Decisions of Leaders

Jason Aimone
Baylor University

ABSTRACT:

While everyone experiences repeated choices regarding risk and uncertainty in their day-to-day lives, leaders are tasked with making such decisions that affect not only their own interests but those who have chosen to follow that leader's choices. Such political and economic leaders thus must then choose whether to continue with the risk choices their current followers chose to follow or change their choices to in order to either attract further followers or to satisfy their own personal risk preferences. Little research has attempted to elicit risk preferences in such repeated risk taking environments in either solo or leadership environments. Here we explore a dynamic risk environment where subjects make repeated Eckel Grossman (2008) style gamble decisions for themselves but know that others, who chose to follow that subject, will receive that same gamble. We find that when making repeated risk decisions, decision-makers demonstrate an increase in variance in those decisions when in a leadership context relative to when their risk decisions affect only themselves. However this increased variance is not associated with change in average risk level of those choices. Further we observe in our study that after observing an increase in the number of people ??following?? a leader, that leader's subsequent risk choice is more likely to shift to a less risky gamble. We find that individual characte

Author(s): Jason A. Aimone and Brooks King-Casas
Topic: Decision Theory: Risk, Psychology and Biology: Neuroeconomics
Back to session: Risk Preference: Social Effects


How does origin of leadership influence risk-taking behavior of representative decision makers?

Luke Jones
Valdosta State University

ABSTRACT:

Individuals in many different contexts are required to make decisions under uncertainty on the behalf of others. Recent experimental evidence in economics suggests that people behave differently in such circumstances than they do when faced with otherwise equivalent decisions made solely on their own behalves. The contexts in which representative decision making under uncertainty occur are numerous, and accordingly there are multiple ways in which one can assume the role of representative decision maker. However, current research is only beginning to examine the effects of different contextual factors in the representative decision making environment, and no previous studies have investigated the origin of leadership. In this paper, we address the role of the origin of leadership using an experiment in which a representative decision maker chooses between receiving a certain monetary payment and playing a lottery, and in which her choice affects not only her own payoff, but the payoff of a randomly assigned dependent. We find that the degree of risk aversion one exhibits in a role of leadership depends on how one obtains the leadership position. In particular, when individuals are randomly assigned to the role of representative decision maker, they are more risk averse than individuals operating solely on their own behalves. However, when individuals earn the position of representative decision maker, they behave in a way that is statistically identical to those making decisions without dependents.

Author(s): Luke R. Jones and Attila Cseh
Topic: Decision Theory: Risk, Social Behavior: Other-regarding Preferences
Back to session: Risk Preference: Social Effects


Social Information and Contribution Behavior: On the Malleability of Climate Change Mitigation Preferences

Sara Elisa Kettner
University of Heidelberg

ABSTRACT:

How malleable are individual preferences for climate change mitigation? We investigate this question in an artefactual field experiment by presenting different pieces of information to our participants. We show social information on past contribution behavior to increase individual giving and demonstrate that this crowding-in-effect stems from the social content of the information and not from psychological effects of simple anchoring. By eliciting subjects beliefs, we find a positive relationship with giving while anticipated beliefs change under social information as well. Lastly, we detect characteristics that stand in relation to contributions and sensitivity to social information. Our data reveals that females are more reactive to the presented information.

Author(s): Timo Goeschl; Sara Elisa Kettner; Johannes Lohse; Christiane Schwieren
Topic: Public Choice: Public Goods and Common Pool Resource,
Back to session: Social Information 2


Can Subtle Provision of Social Information Affect What Job You Choose (and Keep)? Experimental Evidence from Teach For America

Clayton Featherstone
Stanford University

ABSTRACT:

It has been well documented that information about the actions of others can affect small-stakes decisions. We show that a subtle provision of such social information can also influence a very high-stakes decision: whether to take (and keep) a job as a public school teacher. In an experiment involving thousands of admits to Teach For America (TFA), those provided with data about the high matriculation rate in the previous year are more likely to accept the job. Moreover, this effect persists into the second semester of teaching, even though one-sixth of those in the control group who initially accepted the job have left TFA by then. As expected, the effects are stronger for those more marginal in their decision to join TFA. Our results suggest that social information can have a powerful effect on high-stakes behavior and should be considered as a potential tool for policy.

Author(s): Lucas C. Coffman, Clayton R. Featherstone, Judd B. Kessler
Topic: Field Experiments: General, Applied Economics: Labor Market
Back to session: Social Information 2


Tax Evasion Spillovers

Oliver Himmler
Max Planck Institute Bonn

ABSTRACT:

Tax evasion creates tax evasion: lab experiments have established that the observation of others withholding their due tax payments is an important driver of an individual's own decision to engage in tax evasion. That those who are in a position to evade their taxes do take advantage of this opportunity has been shown by Kleven et al. (2011), yet in real life many people cannot evade taxes themselves, e.g. because their income is reported to the authorities by their employer. It is therefore an open question whether absent the 'direct response mechanism' of own tax evasion, individuals react to information about the evasive actions of others via adjustment behavior in domains other than taxation. Our field experiment tests for such spillovers. We confront library workers with facts about tax evasion by celebrities, and subsequently measure the spillover effects on work effort as well as illegal behavior at the workplace. While total work output did not differ significantly across treatments, individuals in our two tax evasion information treatments exhibited much higher rates of stealing (25% and 33%) when compared to the neutral sessions (0% and 4%). The surge in illegal behavior cannot be explained by compensating for higher work effort or by background characteristics. This provides first causal evidence that tax evasion not only creates more tax evasion, but fosters other economically important forms of adverse behavior. Both tax evasion and stealing at the workplace constitute illegal behaviors, and this tentatively suggests that spillovers are more likely to occur in domains that are closer in nature.

Author(s): Sebastian Goerg (FSU and MPI Bonn), Oliver Himmler (MPI Bonn), Tobias Koenig (WZ Berlin)
Topic: Social Behavior: Norms and Morals, Field Experiments: General
Back to session: Social Information 2


Social Coupons: Mechanism Design for Social Media

Tanya Rosenblat
Iowa State University

ABSTRACT:

We develop a simple model of referral mechanisms in social networks. Firms use referrals to exploit private information that consumers have about the preferences of their friends. Our model exhibits a trade-off between the virality of a referral mechanism and the quality of consumers that are recruited through referrals. We show that firms will generally want to limit the number of referrals per customer. We test our model through a field experiment on Twitter and Facebook.

Author(s): Tanya Rosenblat
Topic: Games: Networks, Field Experiments: General
Back to session: Social Information 2


Nudge nudge and say no more: How do reminders in the waiting room affect antibiotics use?

Linnea Wickstrom Ostervall
Stockholm University, Department of Economics

ABSTRACT:

It's not a matter of if; it's a matter of when bacteria become resistant to currently available antibiotics. As no major new contributions to the arsenal of antibiotics have been made the past decades, we are facing the threat of neither being able to cure common infections nor perform advanced surgery. We need to limit antibiotics use to delay resistance, with the hope of buying enough time to find replacements. However, guidelines on prudent use are not adhered to. Patients tend to over-demand antibiotics; doctors overprescribe them. The purpose of this study is to examine if an effective way to lower antibiotics use may be to remind out-patients of the association between antibiotics use and resistance, and whether phrasing of the reminder matters. The method employed is randomized field experiment. The level of randomization is the clinic for distributing reminders - framed as questionnaires - and the individual patient for type of reminder. Two different phrasings of the reminder are used, in addition to a placebo questionnaire. The intervention was carried out at eighteen clinics during the flu season in Stockholm, early March 2012. At treated clinics, prescription rates were about 15 percent lower than at control clinics. This is a substantial effect. There is, however, no difference in effect between the individual patients, regardless if they received any of the reminders or the placebo. This suggests doctors, not patients, were affected by the reminders. There are two possible channels: either the intervention lends support to doctors in arguing for prudent use of antibiotics, or doctors adhere more strictly to guidelines on prudent use when they know they are being monitored. A follow-up experiment was carried out fall 2012 to investigate this further. The results suggest monitoring is at least not the full story; interaction between doctor and patient seems highly important.

Author(s): Linnea Wickstrom Ostervall
Topic: Public Choice: Public Goods and Common Pool Resource,
Back to session: Social Information 1


Confucianism and Preferences: Evidence from Lab Experiments in Taiwan and China

Joseph Wang
National Taiwan University

ABSTRACT:

This paper investigates how Confucianism affects individual decision making in Taiwan and in China. We found that Chinese subjects in our experiments became less accepting of Confucian values, such that they became significantly more risk loving, less loss averse, and more impatient after being primed with Confucianism, whereas Taiwanese subjects became significantly less present-based and were inclined to be more trustworthy after being primed by Confucianism. Combining the evidence from the incentivized laboratory experiments and subjective survey measures, we found evidence that Chinese subjects and Taiwanese subjects reacted differently to Confucianism.

Author(s): Elaine M Liu, Juanjuan Meng, and Joseph Tao-yi Wang
Topic: Social Behavior: Communication, Social Behavior: Lying and Cheating
Back to session: Social Information 1


Blood donation registry one year on: Long term effects and the role of social information

Carmen Wang
Harvard Business School

ABSTRACT:

Blood shortages are common throughout the world. High seasonal variation in donations and massive spikes during disasters (e.g., 9/11, Hurricane Katrina and the Australian Bushfires) reveal enormous potential blood supply, suggesting that frequent shortages during normal times may be due to poor market signals (e.g., no price mechanism). We introduced a blood donation registry that will improve upon a national blood agency's central clearinghouse functions; the registry collects and aggregates information on potential donors' willingness-to-donate to more efficiently coordinate donor actions to meet excess demand. This study follows up on our registry performance analyses to report on the long-term effects one and half years after introduction. We will also discuss the role of social information in designing registries; the registry may not only efficiently disseminate blood demand information but can also provide information on other donors' behavior. We'll present an upcoming study on how to better disclose social information to uphold a norm that induces higher voluntary blood donations.

Author(s): Ellen Garbarino, Robert Slonim and Carmen Wang
Topic: Markets: Market design and Matching, Social Behavior: Norms and Morals
Back to session: Social Information 1


It's Not just the thought that counts --- An experimental study on hidden cost of giving

Xiaofei (Sophia) Pan
Harvard University

ABSTRACT:

Receiving a gift can create an impulse to reciprocate, even when doing so is inefficient and potentially harmful to a third party. For instance, people may exhibit such impulses after receiving bribes. This paper provides a theoretical framework and experiment evidence for a pure gift effect: people feel in debt to the gift they receive and a gift receiver will favor the gift giver from whom he/she receives the gift over a third party who has paid the same cost to send the same gift but the gift was not received by the receiver. We further show that the pure gift effect does not depend on whether the intention of giving is for future returns. Our findings suggest that when reciprocating to the gift become socially undesirable, we may need to guard against any gift an agent may receive rather than focusing on the intent of the underlying gift.

Author(s): Xiaofei (Sophia) Pan and Erte Xiao
Topic: Social Behavior: Other-regarding Preferences,
Back to session: Social Preferences 1


Inequality of Opportunities and Redistribution: an Experiment

Gustavo Caballero
University of Calgary

ABSTRACT:

Income inequality may arise due to differences in individual effort and differences in opportunities at earning the income. However, people tend to perceive as fair those outcomes arising from differences in effort but not those arising from inequity. In that manner, individual preferences towards redistribution depend on how much information one possesses about the sources of a given inequality and how the redistribution is allocated. Here I present an experiment where an individual divides $20 between herself and a participant who has zero. The role of sender or receiver depends on a probability that is higher if more effort is made on a multiplication question task but differs between two groups for each possible level of effort. This paper inquiries how individual preferences towards redistribution responds to having information about differences in effort and inequity. Additionally it looks into the effect that general presence of inequity has on the willingness to help.

Author(s): Gustavo Caballero & Robert Oxoby
Topic: Social Behavior: Other-regarding Preferences, Decision Theory: Preferences
Back to session: Social Preferences 1


Other-regarding Preferences in Experimental Dynamic Volunteer's Dilemmas

Cheng-Tse Lin
National Taiwan University

ABSTRACT:

Dynamic volunteer's dilemmas have been modeled as individuals deciding whether to volunteer or not based on cost-benefit analysis, as in the war of attrition game. However, this analysis is usually carried out assuming self-interest, without other-regarding preferences. In this paper, we investigate the role of other-regarding preferences in dynamic volunteer's dilemmas using lab experiments in which two players decide when to jump in and volunteer for the pair and contrast the results with a control treatment where subjects play against computers. We find that subjects are more likely to volunteer when dealing with the computer rather than with other participants. Our experimental data provide direct evidence that other-regarding preferences do play a critical factor in this problem, but through envy instead of altruism.

Author(s): Cheng-tse Lin and Joseph Tao-yi Wang
Topic: Social Behavior: Other-regarding Preferences, Public Choice: Public Goods and Common Pool Resource
Back to session: Social Preferences 1


Benefit of the doubt: the importance of intentions and control in cooperation

Garret Ridinger
University of California, Irvine

ABSTRACT:

Recent models of fairness can be separated into two types: outcome-based models and intention-based models. Outcome-based models suggest that people care about fair distributions while intention-based models suggest people care about how outcomes are achieved. Models of intentions typically assume individuals have full control over their choices while outcome based models suggest that intentions are not behaviorally relevant. However, it seems that people care about both fair outcomes and the intentions of others. In addition, individual control likely influences whether a choice is viewed as intentional. This paper develops a mixed motives model that combines concerns for outcomes, intentions, and individual control. In order to test the equilibrium predictions of the mixed motives model, I use a sequential prisoner's dilemma when the first mover's choice has a chance to be reversed by the computer. Experimental treatments varied the first mover's control over their choices (high versus low) and whether the second mover was told what the first mover chose (observable versus unobservable). In the high versus low treatment, both first and second movers are less likely to cooperate when control is high. In addition, second movers are more likely to cooperate when the computer cooperates in the low control condition. This suggests that when first movers have little control, the second movers evaluate fairness based on outcomes. In the observable versus unobservable treatment, second movers are significantly less likely to cooperate when they are told what the first mover chose. However, there is no significant difference in cooperation by first movers. Evidence suggests that second mover's give the first mover's the benefit of the doubt. Overall, the results cast doubt on fairness models that assume only outcomes or only intentions matter. Instead, the evidence is consistent with the mixed motives model and suggests that intentions, outcomes, and control are all important factors in understanding cooperation.

Author(s): Garret Ridinger
Topic: Social Behavior: Other-regarding Preferences, Social Behavior: Norms and Morals
Back to session: Social Preferences 1


Signaling Smarts: Revealed Preferences for Self and Social Perceptions of Intelligence

Clay McManus
University of Tennessee

ABSTRACT:

We design a laboratory experiment to test for self and social image concerns in a domain where decisions signal intelligence. Money-maximizing behavior in the experiment sorts subjects by academic ability, as measured by performance on verbal analogy questions. The design allows subjects to privately make costly changes in behavior to impact perceptions of their ability. In treatment, sorting behavior is publicly revealed, while decisions in control remain private. We find evidence that individuals are motivated to self-signal intelligence in the private setting. Social observation has the opposite effect, discouraging ability signaling. These effects are more pronounced in males. The results suggest that framing a decision as a social signal of intelligence can crowd out intrinsic motivations and instead demotivate individuals. Audience effects can have a counterintuitive impact on signaling behavior in the ability dimension, in contrast to the near universally positive findings on how social observation impacts generosity.

Author(s): T. Clay McManus and Justin M. Rao
Topic: Social Behavior: Norms and Morals, Methodology: Experimental Design, Tools and Practices
Link: https://docs.google.com/viewer?a=v&pid=sites&srcid=ZGVmYXVsdGRvbWFpbnx0Y2xheW1jbWFudXN8Z3g6MjQwY2Y5ZjRkNzFlOTQ5NA
Back to session: Social Preferences 2


Norms Make Preferences Social

Erik Kimbrough
Simon Fraser University

ABSTRACT:

We explore a unifying explanation for prosocial behavior in which people care not about others' payoffs per se, but whether their own behavior accords with social norms. Individuals who are sensitive to norms will adhere to them so long as they observe others doing the same. A model formalizing this generates both prosociality (without relying on explicit distributional preferences) and well-known context effects (for which simple distributional preferences cannot account). A simple experiment allows us to measure individual-level norm- sensitivity and to show that norm-sensitivity explains heterogeneity in prosociality in public goods, dictator, ultimatum, and trust games.

Author(s): Erik O. Kimbrough, Alexander Vostroknutov
Topic: Social Behavior: Norms and Morals, Social Behavior: Other-regarding Preferences
Link: http://ideas.repec.org/p/sfu/sfudps/dp13-01.html
Back to session: Social Preferences 2


Luck, choice and responsibility - an experimental investigation of distributive justice

Johanna Mollerstrom
Harvard University

ABSTRACT:

This paper uses a series of economic experiments to study fairness views in situations involving luck, focusing on the extent to which people are held responsible for outcomes influenced by luck. The design allows for two types of luck - option luck and brute luck - which differ with respect to whether the agents can influence the risk exposure. Option luck involves risk that can be insured against, while brute luck is risk beyond the participants' control. In line with several prominent theories of distributive justice, we find that agents are to a larger extent held responsible when an outcome is due to option luck rather than brute luck. Additionally, our results uncover a pattern in distributional behavior which is not well accounted for in the theories of justice: compensation in situations involving bad brute luck is not unconditional - it depends on the participants risk preferences, revealed by how they handle option luck. We find that more risk tolerant agents generally receive less compensation in cases of bad brute luck. We outline a theory of distributive justice with which this behavior is in line.

Author(s): Johanna Mollerstrom, Bjorn-Atle Reme
Topic: Social Behavior: Norms and Morals, Social Behavior: Other-regarding Preferences
Back to session: Social Preferences 2


Costly Signalling of Intentions in the Trust Game

Severine Toussaert
NYU

ABSTRACT:

The literature on experimental trust games has recently devoted special attention to the role of perceived intentions. More precisely, theories of reciprocity and guilt aversion suggest a causal link between the Receiver's belief about the Sender's intentions and his propensity to return. While this link has been extensively analyzed, the literature has stopped short of examining the consequences of the Receiver's belief-dependent motivations for the Sender's incentives to trust in the first place. If perceived intentions indeed matter for the Receiver, shouldn't the Sender internalize it when deciding whether to invest? The present study provides new evidence on the role of perceived intentions and, in particular, on the way they may feed back into the Sender's incentives. In order to tackle this issue, a modification of the standard binary trust game is studied under 3 treatment conditions. In all 3 treatments, both players are informed that the Sender's decision will be replaced with some probability by the choice of a random device. This probability is common knowledge. In Treatment 1, the Receiver makes a choice without knowing the Sender's decision while in Treatment 2, he makes a choice conditional on the Sender's decision. In Treatment 3, the Sender is offered the choice to inform the Receiver of his decision at some cost. If perceived intentions matter, the Receiver's propensity to return should be higher (lower) if the Sender invested (didn't invest) in Treatment 2 compared to Treatment 1. Senders who foresee this should be more likely to pay to signal their trust in Treatment 3.

Author(s): Severine Toussaert
Topic: Social Behavior: Other, Social Behavior: Other-regarding Preferences
Back to session: Social Preferences 2


Markets and the emergence of social relationships: A Laboratory Experiment

Seung Ginny Choi
Interdisciplinary Center for Economic Science

ABSTRACT:

The market is an area where buyers and sellers exchange goods and services. It is also a sphere where activity beyond competition and exchange takes place and where social connections develop. As the scope of a market expands, including an increasing number of people and touching more aspects of our lives, it is not surprising there has emerged a debate about the social consequences of markets. This paper studies connections between social ties and the socio-economic environment within which market exchange occurs. Our results provide evidence that markets promote not only efficient economic exchange but also play an important role in creating the fabric of social interaction.

Author(s): Ginny Seung Choi and Virgil Henry Storr
Topic: Markets: Other,
Back to session: Social Preferences 3


When is inequality Fair? An experiment on the effect of procedural justice and agency

Sevgi Yuksel
New York University

ABSTRACT:

We investigate how fairness views on the distribution of income depend on our beliefs about the process which generates the inequality. Specifically, we examine how two crucial features of this process affect fairness views: (1) Procedural justice - equal treatment of all; (2) Agency - one's ability to determine his/her income. We do this in a lab experiment by manipulating the income generating process, and eliciting ex-post redistribution decisions on total earnings. We vary equality of opportunity (procedural justice) and one's ability to make choices which consequently influence their income (agency). Our main findings can be summarized as follows: (1) Inequality resulting from risk is considered to be fair only when risk is chosen freely; (2) Introducing inequality of opportunity significantly increases redistribution, however the share of subjects redistributing none remain close to the share of subjects redistributing fully revealing an underlying heterogeneity in the population about how fairness views should account for inequality of opportunity; (3) Finally, we find evidence to suggest that redistribution decisions reward risk-takers, which can be interpreted as a preference for efficiency in our design.

Author(s): Merve Akbas, Dan Ariely, Sevgi Yuksel
Topic: Social Behavior: Norms and Morals, Social Behavior: Other-regarding Preferences
Back to session: Social Preferences 3


Social Preferences in Exploding Offers

Xiaoyuan Wang
Texas A&M University

ABSTRACT:

Anecdotal evidence suggests people respond negatively to take-it-or-leave-it (i.e., exploding) offers and this may make sellers more reluctant to use such offers than theory would dictate. This paper examines this conjecture in an experimental search market for goods with two sellers which buyers visit sequentially in a random order. Each seller can choose either an exploding offer (which is nullified once the buyer visits the other seller) or an offer that the buyer may revisit. In this environment, there is a unique Nash equilibrium where sellers should use an exploding offer and charge the highest possible price. Experimental sellers quickly converge to this equilibrium when matched against computer buyers programmed to play optimal strategy. However, when matched against human buyers, sellers do not follow equilibrium play, avoiding exploding offers and offering lower prices, and buyers reject exploding offers more frequently. These results cannot be explained by risk aversion nor are not consistent with inexperienced buyers in other experimental search games. They are reminiscent of the rich literature on ultimatum games, suggesting buyers may have a non-monetary incentive to reject exploding offers.

Author(s): Alexander L. Brown, Ajalavat Viriyavipart, Xiaoyuan Wang
Topic: Markets: Industrial Organization, Social Behavior: Other-regarding Preferences
Link: http://xwang.org/project-details-exploding-offers/
Back to session: Social Preferences 3


The Effects of Anger and Happiness on Pro-Social Behaviour

Brit Grosskopf
University of Birmingham

ABSTRACT:

This paper investigates the effects of induced positive and negative emotions on pro-social behaviour in a one shot voluntary contributions mechanism where personal gains and group benefits are mis-aligned. Our specific focus is on the impact of anger and happiness on contribution and punishment behaviour. Our findings suggest subjects, on average, in the Negative emotions treatment contribute less than those in the Positive emotions treatment and overall welfare is lower when subjects are in a negative mood. We also find that how punishment is used is affected by moods: negative mood subjects punish harsher than positive mood subjects.

Author(s): Michalis Drouvelis and Brit Grosskopf
Topic: Games: Other, Psychology and Biology: Cogonition
Back to session: Social Preferences 3


Paradoxes and Mechanisms for Choice under Risk

James C. Cox
Georgia State University

ABSTRACT:

Experiments on choice under risk typically involve multiple decisions by individual subjects. The choice of mechanism for selecting decision(s) for payoff is an essential design feature unless subjects isolate each one of the multiple decisions. We discuss theoretical properties of commonly used mechanisms and new mechanisms introduced herein. We report experiments that generate data that show systematic differences across mechanisms in subjects' revealed risk preferences. We illustrate the importance of these mechanism effects by identifying their implications for tests of classic properties of theories of decision under risk. We also identify behavioral properties of mechanisms that introduce bias in elicited risk preferences from cross-task contamination.

Author(s): James C. Cox, Vjollca Sadiraj, and Ulrich Schmidt
Topic: Special Topic: Experimental Payments,
Link: http://excen.gsu.edu/workingpapers/GSU_EXCEN_WP_2013-08.pdf
Back to session: Special Session: Pay Protocols


Elicitation of Risk and Uncertainty Preferences

Yoram Halevy
University of British Columbia

ABSTRACT:

We will discuss the possibility and mechanics of eliciting preferences over uncertain prospects using variations of the price list method.

Author(s): Yoram Halevy
Topic: Special Topic: Experimental Payments,
Back to session: Special Session: Pay Protocols


Incentives in Experiments: The RPS Mechanism

Paul Healy
Ohio State University

ABSTRACT:

I will describe the Azrieli et al. result that, under monotonicity, the Random Problem Selection (RPS) mechanism is essentially the only incentive compatible mechanism. Then I will discuss ways in which monotonicity may be violated, and when paying for every decision may be warranted. Finally, I will show new experimental results on monotonicity and the RPS mechanism, joint with Alex Brown.

Author(s): Paul J. Healy, Yaron Azrieli, Chris Chambers, Alex Brown
Topic: Methodology: Experimental Design, Tools and Practices, Decision Theory: Other
Link: http://healy.econ.ohio-state.edu/papers/Azrieli_Cambers_Healy-CombinedDecisions.pdf
Back to session: Special Session: Pay Protocols


An attention-based stationary concepts for Experimental 2x2-Games

Qiqi Cheng
Zhejiang University

ABSTRACT:

This paper introduces an attention-based view of payoffs and choices in the mixed strategy games and brings this view to the exist models, Nash equilibrium, Quantal response equilibrium and Impulse balance equilibrium. The view posits that individuals can't equally attend to the two choices if they are not equivalent in drawing the attention, leading to a weight-related effect on each choice, and finally having an effect on the stationary state. The extended models based on this view give a significant improvement in prediction the stationary state in zero-sum games, by examining data from published papers [Selten, R. and T. Chmura (2008), McKelvey, R. D., T. R. Palfrey and R. A. Weber (2000)]and predicts at lease as well as other three stationary conceptions in overall games.

Author(s): Qiqi Cheng
Topic: Games: Repeated Games, Decision Theory: Bounded Rationality
Back to session: Strategic Sophistication


Decomposing Bounded Rationality and Behavioral Models

Ryan Kendall
University of Southern California - LABEL

ABSTRACT:

This paper uses a new approach to model human decision making as being influenced by separately identified components within a game. New mathematical results allow for a unique decomposition of a game into two disjoint components; what we refer to as Strategic and Behavioral. The Strategic component is sufficient to determine the prediction for personal payoff maximizing equilibrium concepts such as Nash, Logit-Quantal Response (QRE), and Logit-Noisy Introspection (NI). A natural byproduct of exhaustively classifying the selfish incentive is a unique representation of the non-selfish, or cooperative, incentive. Human responsiveness to this previously unknown variable, labeled the Behavioral component, is evaluated using laboratory experiments which yield two main results. First, data from these experiments suggest that humans systematically respond to a game's Behavioral component which is ignored by Nash, QRE, and NI models. Second, while data from our experiments fit models of agents with secondary preferences for traits such as altruism or equity, our data also fit a model of agents with secondary preferences for a nonsensical trait (what we describe as a Corner Model). That is, we find a similar model-fit between dissimilar model-motivations which introduces concerns about the meaningfulness of some standard behavioral models.

Author(s): Ryan Kendall and Daniel Jessie
Topic: Social Behavior: Other-regarding Preferences, Decision Theory: Bounded Rationality
Back to session: Strategic Sophistication


Failures of Unraveling in Disclosure Experiments: A Level-K Analysis

Daniel Martin
Paris School of Economics

ABSTRACT:

Using laboratory experiments, we implement a two-player disclosure game in which senders have access to costless verifiable messages about their type in order to examine whether limited disclosure could be driven by lack of strategic sophistication. We find that senders converge to full disclosure when feedback about the behavior of other players is available, but when players have no such feedback, as in many applied settings, there is a consistent level of partial disclosure. Partial disclosure can be explained using a level-k model, and fitting this model to the experimental data reveals that most players in the no feedback treatment behave consistent with a single level of sophistication in each role, and that the level of sophistication is higher when in the role of sender.

Author(s): Ginger Jin, Michael Luca, and Daniel Martin
Topic: Markets: Industrial Organization, Psychology and Biology: Cogonition
Back to session: Strategic Sophistication


Characterizing Financial and Statistical Literacy

Morten Lau
Durham University

ABSTRACT:

We characterize the literacy of an individual in a domain by their elicited subjective belief distribution over the possible responses to a question posed in that domain. We consider literacy across several financial, economic and statistical domains. We find considerable demographic heterogeneity in the degree of literacy. We also characterize the degree of consistency within a sample about their knowledge, even when that knowledge is imperfect. We show how uncertainty aversion might be a normatively attractive behavior for individuals who have imperfect literacy. Finally, we discuss extensions of our approach to characterize financial capability, the consequences of non-literacy, social literacy, and the information content of hypothetical survey measures of literacy.

Author(s): Amalia Di Girolamo, Glenn W. Harrison, Morten I. Lau and J. Todd Swarthout
Topic: Decision Theory: Beliefs,
Back to session: Survey Methods: Characterizing Behavior


The Size of the LGBT Population and the Magnitude of Anti-Gay Sentiment are Substantially Underestimated

Katherine Baldiga
Ohio State University

ABSTRACT:

Measuring sexual orientation, behavior, and intolerance is difficult because responses are biased towards socially acceptable answers. We test whether social desirability bias exists even under very private and anonymous conditions. Identifying the direction of social desirability bias also allows us to shed light on norms surrounding sexual orientation and related views. We run an experiment on 2,516 participants drawn from the United States' general population. Participants were randomly assigned to either a best practices method that was computer-based and provides privacy and anonymity, or to a veiled elicitation method that further adds a veil to individual responses. Answers in the veiled method do not enable inference about any particular individual, but can be used to accurately estimate aggregate incidence rates. The veiled method shows many, though not all, questions receive biased responses even under extreme privacy, and, in some cases, the bias is substantial. The veiled method increased self-reports of non-heterosexual identity by 65% (p<0.05) and same-sex sexual experiences by 59% (p<0.01). The veiled method also increased the incidence rates of anti-gay sentiment. Respondents were 67% more likely to express disapproval of an openly gay manager at work (p<0.01) and 71% more likely to say it is okay to discriminate against homosexuals (p<0.01). The results show non-heterosexuality and anti-gay sentiment are substantially underestimated in existing surveys, and the privacy afforded by current best practices is not always sufficient to eliminate bias. Finally, our results indicate that, somewhat paradoxically, it may be socially undesirable to be gay, or unaccepting of gay individuals.

Author(s): Katherine Baldiga, Lucas Coffman, Keith Marzilli Ericson
Topic: Psychology and Biology: Other, Applied Economics: Other
Back to session: Survey Methods: Characterizing Behavior


Eliciting Private Information with Noise: The Case of Randomized Response

Andreas Blume
University of Arizona

ABSTRACT:

The paper formalizes Warner's (1965) randomized response technique (RRT) as a game and implements it experimentally, thus linking game theoretic approaches to randomness in communication with survey practice in the field and a novel implementation in the lab. As predicted by our model and in line with Warner, the frequency of truthful responses is signicantly higher with randomization than without. The model predicts that randomization weakly improves information elicitation, as measured in terms of mutual information, although, surprisingly, not always by RRT inducing truth-telling. Contrary to this prediction, randomization significantly reduces the elicited information in our experiment.

Author(s): Andreas Blume, Ernest K. Lai, Wooyoung Lim
Topic: Games: Information, Social Behavior: Lying and Cheating
Link: https://docs.google.com/viewer?a=v&pid=sites&srcid=ZGVmYXVsdGRvbWFpbnxhbmRyZWFzM2JsdW1lfGd4OjIzMDljMDlhNWQzZTk2Nzc
Back to session: Survey Methods: Characterizing Behavior


The Source of Inequality and Preferences for Wealth Redistribution

David Owens
Haverford College

ABSTRACT:

The redistribution of wealth is a major function of economic policy, as well as a common source of political discord. Proponents of redistributive policies often cite luck as the mechanism that determines wealth, while opponents often cite the crucial role of effort in its determination. In a hybrid within- and between- subjects experimental setting, this paper manipulates the mechanism that determines an unequal distribution of wealth, and measures subjects' preferences for redistribution. Each subject participates in settings where luck, intelligence and effort determine their wealth, and express their preferences for redistribution both before and after their wealth is known. According to preliminary results, the most redistribution by far is preferred when luck determines wealth, with intelligence a distant second. Self interest predictably drives changes in preferences, with the wealthy preferring less redistribution, and the poor more, after learning where they stand in the distribution.

Author(s): David Owens
Topic: Social Preferences, Redistribution
Back to session: Inequality and Social Preferences


Characterizing Preferences for Giving to Parents in an Experimental Setting: Reciprocity or Altruism

Maria Porter
Michigan State University

ABSTRACT:

This paper examines how adult children share with parents in an experimental setting. Respondents play a series of dictator games in which budget sets and relative prices of giving to parents or strangers are varied across games. All respondents play all variations of these games, separately with parents and strangers as recipients. Respondents generally satisfy the Generalized Axiom of Revealed Preference (GARP), that is, have a rational preference ordering for giving to each recipient. Yet, they treat gifts to parents and gifts to strangers as distinct goods. Subjects are more sensitive to the relative price of giving when parents are the recipients as opposed to strangers. We characterize four preference classes: selfish, selfless, utilitarian, and Rawlsian. We also estimate a CES utility function for preferences over giving to strangers and to parents. Finally, we randomize the amount of information revealed to parents, and find evidence of strategic motivations for giving. In comparison to those whose parents receive no information, players give more to parents who are more likely to reciprocate after the game, that is, those who receive full information regarding the games.

Author(s): Maria Porter and Abi Adams
Topic: Decision Theory: Preferences, Applied Economics: Charitable Giving
Back to session: Inequality and Social Preferences


Modelling Socially Responsible Investment Decisions

Paulo Peneda Saraiva
Nottingham University Business School

ABSTRACT:

I use experimental methods to model individual preferences for socially responsible investments (SRI) under risk. SRI are financial products that generate returns to the investor and yield benefits to the Environment, to Society, and/or to Corporate Governance. In the experiment, participants allocate money between two risky investments with financial, and in some cases, socially responsible attributes. I use standard economic axioms of revealed preferences (WARP/SARP/GARP) to investigate if investors follow the economic model of utility maximization when making these investment decisions. Despite the heterogeneity in investment behavior, I find that investors' choices do not severely violate these axioms. I use a two-parameter utility function to estimate individual investor's preferences for SRI which vary from perfect substitutes to Leontieff. In addition, I find that including a socially responsible attribute in only one investment leads to higher consistency violations in preferences, than when both investment products have, or have not, socially responsible attributes.

Author(s): Paulo Peneda Saraiva
Topic: Decision Theory: Preferences, Social Behavior: Other-regarding Preferences
Back to session: Inequality and Social Preferences


Behavioral Poverty Traps and Time Preferences in Liberia

Julian Jamison
Consumer Financial Protection Bureau, Washington DC

ABSTRACT:

We use a field experiment involving high-risk young men in urban Liberia, along with lab-in-the-field decision tasks, to answer a key question in the field of development and political economy: can preferences such as present bias and non-cognitive skills such as impulse control and conscientiousness be actively modified, thereby reducing poverty and social instability? Economists and psychologists often treat these attributes as fixed in adults, but in practice a large number of policy interventions are predicated on behavior change in these areas. Our approach pairs two complementary randomized interventions: a large unconditional cash grant and a psychosocial transformation program. Our total sample of 1000 youth is therefore split into four groups, randomly receiving one or the other or neither or both of the interventions. In the initial results we see some promising signs of sustained behavior change, indicating that adults may be more malleable than previously thought. There is an increase in the patience measure (using an incentivized game) from all interventions, including the cash grant. We also see a significant reduction in self-reported drug use, particularly from the treatment program, and in impulsive behavior (measured in part using novel inhibition-related psychometric tasks) from all interventions. Finally, we see some reduction in anti-social behaviors such as theft and aggression, particularly from those enrolled in the treatment program, although - as with the patience and some other measures - the effect is stronger in the short-term than the long-term.

Author(s): Chris Blattman, Julian Jamison, and Margaret Sheridan
Topic: Discounting: Executive Function, Endogenous Preferences: lab-in-the-field
Back to session: Inequality and Social Preferences


Testing time preference elicitation methods with multiple-price-list format

Younjun Kim
Iowa State University

ABSTRACT:

Individual time preference is usually measured by a sequence of questions in a table with Multiple Price List (MPL) format. The questions ask subjects to choose between a smaller and sooner payment and a larger and later payment. We investigate whether elicitation of a discount rate and time consistency (or time stationarity) is affected by the MPL format under two types of outcome framing; in a delay framing (speed-up framing), smaller and sooner payments are fixed (varied), and larger and later payments are varied (fixed) so that subjects elicit a larger and later (smaller and sooner) payment indifferent to a smaller and sooner (larger and later) payment. Under a within-subject design we ask subjects a question included in a MPL table and ask the question separately from the MPL table. If MPL format does not affect subjects' responses, the responses would be the same between the two identical questions. Also, under a between-subjects design we check whether elicited time consistency is the same regardless of the two types of framing. Our experimental data find that subjects are more likely to choose a larger and later payment in a question included in a MPL table compared to the question asked separately from the MPL table, which suggests that elicited discount rates from a single question are higher than those from a MPL table. Together with elicited loss aversion, we find that more loss-averse subjects are more likely to choose a larger and later payment in a MPL table relative to in a single question. These patterns hold in both types of framing. We also find that elicited time consistency is different between the two types of framing; subjects look more future-biased under a delay framing but present-biased under a speed-up framing. Our findings raise a future research question which time preference elicitation methods better predict actual behaviors.

Author(s): Younjun Kim and Tanya S. Rosenblat
Topic: Methodology: Experimental Design, Tools and Practices,
Back to session: Time Preferences


The Role of Attention in Intertemporal Choices

Geoffrey Fisher
Caltech

ABSTRACT:

We run an eye tracking experiment designed to understand how visual attention is allocated in a simple intertemporal choice environment. Subjects make decisions between receiving a payment today or a larger, delayed payment. We find that those who are more patient spend significantly more time looking at the monetary amounts, as opposed to the delays when those amounts will be implemented. Those who are impatient spend more time looking at the word today. Additionally, we run several experiments designed to exogenously vary how attention is deployed and report how it impacts behavior.

Author(s): Geoffrey Fisher and Antonio Rangel
Topic: Psychology and Biology: Neuroeconomics, Decision Theory: Preferences
Back to session: Time Preferences


Demand for Commitment: Time Inconsistency or Signaling?

Jeffrey Naecker
Stanford University

ABSTRACT:

Much previous research has looked at commitment demand as the smoking gun for sophisticated time-inconsistency. We propose an alternative explanation for the demand for these devices: commitment allows individuals to signal their intentions or goals to others. In this paper, we present the results from a field experiment where participants could choose a commitment device; their choice was either kept private or revealed to others, depending on the treatment. We find strong demand for the commitment technology in our private treatment, which is indicative of some agents being time-inconsistent sophisticates. However, we also find significantly higher demand for commitment in the public treatment. This phenomenon is consistent with some subjects desiring to signal to others that they are especially committed to or interested in the task at hand. This result reveals that time inconsistency alone may not explain all demand for commitment seen in the literature.

Author(s): Christine Exley and Jeffrey Naecker
Topic: Social Behavior: Other-regarding Preferences, Field Experiments: General
Back to session: Time Preferences


The Unforeseen Benefits of EBT: What Can They Tell Us About Dynamic Inconsistency?

Michael Kuhn
UCSD Economics

ABSTRACT:

Intra-month cycles in household consumption and expenditures are often considered hallmarks of dynamically-inconsistent planning. I find that food-stamp households with more children and gender-balanced adult populations experience this phenomenon more severely prior to the introduction of Electronic Benefit Transfer (EBT), and that the introduction of EBT eliminates this gap. I propose an explanation based on collective dynamic inconsistency --present bias generated by the aggregation of differing time preferences within households-- and that EBT affects this aggregation process by solidifying more dictatorial control over the food stamp resources. This implies that policies that improve the property rights of a single recipient over a transfer disbursement and shortening the frequency of disbursement would help improve budgeting, especially for families with young children.

Author(s): Michael A. Kuhn
Topic: Applied Economics: Other, Social Behavior: Group Behavior
Back to session: Time Preferences


Incentive Substitutability and Heterogenous Motives in an Innovation Contest - Field Experimental Evidence

Annika Mueller
The Wang Yanan Institute for Studies in Economics

ABSTRACT:

The offering of high-powered pecuniary incentives is the basis for eliciting effort in contests and tournaments. In many instances, winning a contest not only garners cash prizes for the winners but also generates a signal that may have its own intrinsic value and serve as an additional incentive for participation and effort exertion. In this paper we present the results of a field experiment where we systematically varied cash prizes, a community status signal and a job market signal for 1200 subjects solving a difficult NASA problem in a contest setting. We find that participants are motivated to exert effort for heterogeneous motives that are uncorrelated with skill which results in the treatments drawing on an equivalent skills distribution of participants. Hence while motivation to participate is conditional on treatment, the skills of the participants are not. We show that once activated, rates of participation, effort exertion and performance do not vary across treatments. Hence a 3 times boost in cash rewards is an equivalent to generating a community status leader-board or a job market signal in the form of a letter of recommendation. Our results highlight the need for both theory and practice to explicitly include the costless signal generation aspect of contests as an important criteria for tournament design.

Author(s): Kevin Boudreau (London Business School, Harvard-NASA Tournament Laboratory); Karim R. Lakhani, (Harvard Business School, Harvard-NASA Tournament Laboratory); Michael E. Menietti (Harvard-NASA Tournament Laboratory); Annika M. Mueller (WISE)
Topic: Games: Contests, Field Experiments: General
Back to session: Tournaments and Contests 1


Competing Openly or Blindly in Crowdsourcing Contests?

Lian Jian
University of Southern California

ABSTRACT:

Crowdsourcing contests --- contests in which participants compete to win prizes by solving contest holders' problems --- has become an important way of aggregating knowledge and expertise from the crowd on the Internet. In practice, two types of mechanisms are used to organize these contests: simultaneous and sequential. In a simultaneous contest, contestants submit their solutions independently without access to one another's solutions while in a sequential contest, contestants submit their solutions sequentially and each can view all prior submissions before making their decisions. Most prior research has focused on simultaneous contests, with only a handful that have studied sequential contests. In this paper, under the condition of incomplete information, we analytically characterize the expected highest quality in sequential contests, compare it with the same measure in simultaneous contests, and also conduct laboratory experiments to test our theoretical predictions. Comparative statics analysis reveals that while the expected highest quality monotonically increases in the talent level of the participant pool in simultaneous contests, it is not monotonic in sequential contests. Rather, in sequential contests, there is an optimal overall talent level in the participant pool that maximizes the expected highest quality. In addition, we also analytically show that for contests with a small number of participants, simultaneous contests produce better quality best solutions. This latter theoretical result is supported by our laboratory experiment. In the experimental data, we also observe a bifurcation effect, which is present in both types of contests but is much more intensified in sequential ones. That is, high-ability contestants work extremely hard and low-ability ones tend to drop out entirely.

Author(s): Lian Jian and Tracy Xiao Liu
Topic: Games: Contests, Markets: Market design and Matching
Back to session: Tournaments and Contests 1


Managing Sabotage in Promotion Tournaments

Tim Salmon
SMU

ABSTRACT:

Promotion tournaments are a common way to incentive employees to engage in effort by firms. While it is clear how this can work to motivate effort during the tournament phase, the effect on post tournament effort is less clear. Of clear concern is the degree to which the loser of the promotion accepts this outcome. If the loser does not accept the outcome, they may well attempt to sabotage the winner which in turn harms the company. We examine the degree to which the design of the tournament mechanism influences the degree to which sabotage behavior occurs. We also examine the degree to which workers self-selecting into tournaments may help counter any negative responses to tournament losses.

Author(s): Tim Salmon
Topic: Applied Economics: Labor Markets, Social Behavior: Group Behavior
Back to session: Tournaments and Contests 1


Externalities in the Colonel Blotto Game

Zhengzheng Wang
Texas A&M University

ABSTRACT:

This paper investigates a non-zero sum Colonel Blotto Game experiment where subjects are divided into two teams. Battlefields lost by one player on a team will impose negative externalities on the other players within the same team. This design captures a situation where an attack on one country imposes additional costs on other countries on the same side of the conflict. For example, a terrorist attack on a prominent US target increases costs to other countries, as they expend resources to harden their potential targets. Experimental results show that, when externalities are added into the game, participants tend to bid more on battlefields and try to win. We also find that either in the baseline (with no externality) or in the treatment (with externality on one of the two teams of players), participants bid less that the Nash Equilibrium of the game. Expenditures are highest in the team where externalities are present.

Author(s): Zhengzheng Wang, Catherine Eckel
Topic: Games: Contests,
Back to session: Tournaments and Contests 2


Do contests bring out the worst in us or the worst among us?

Glenn Dutcher
University of Central Missouri

ABSTRACT:

Given the importance of contests in society, the amount of deception present in these contests is problematic. However, in order for designers to implement more efficient contests, they need to know if the contest environment encourages deception or if deceivers self-select into contests. We develop a model for this environment and test the predictions with a real effort experiment where subjects have the opportunity to deceive in order to increase their expected payouts. Our two main treatments focused on if subjects were forced into a contest (treatment 1) or if they could choose between a contest or piece rate pay scheme (treatment 2). We compare the deception in these treatments against a baseline where subjects were paid piece-rate. As expected, we detect a significant increase in the amount of deception when subjects are forced into a contest. This result is driven by beliefs of others and ability - those who deceived more in treatment 1 were primarily of lower ability and/or believed others also deceived more. Concerning selection, ability and beliefs once again play a vital role. Subjects selection criteria for the contest pay scheme are based on the amount of deception they are comfortable with given their beliefs about their relative ability and the amount of deception by others. Specifically, we observe high ability subjects choosing to avoid the contest because they (correctly) believe the amount of deception in the contest will be quite high. Likewise, we observe low-ability subjects selecting the contest who are much more comfortable with deceiving than others. These findings have direct implications for contest designers and the welfare effects of contests.

Author(s): E. Glenn Dutcher, Daniela Glatzle-Rutzler, Dmitry Ryvkin
Topic: Games: Contests, Social Behavior: Lying and Cheating
Back to session: Tournaments and Contests 2


An Experimental Investigation of Simultaneous Multi-battle Contests with Complementarities

Cary Deck
University of Arkansas

ABSTRACT:

This paper reports the results of laboratory experiments that are designed to test theoretical predictions in a multi-battle contest with geographic complementarities. The specific setting is a game of Hex where control of each region is determined by a Tullock contest and the overall winner is determined by the combination of claimed regions. We find that in a game with only a few regions, aggregate behavior across regions is largely consistent with the theoretical predictions. However, examining individual level behavior suggests that bidders are not behaving in accordance with the model, but rather pursuing focused attacks. This intuitive behavioral approach is also found to occur in larger games where the theory is undeveloped.

Author(s): Cary Deck, Sudipta Sarangi, Matt Wiser
Topic: Games: Contests,
Back to session: Tournaments and Contests 2


Is a Picture Really Worth a Thousand Words? Investigating the Relative Value of Appearances in an Investment Experiment

Walter Theseira
Nanyang Technological University

ABSTRACT:

We investigate how subjects use both hard (quantitative data) and soft (photographs) information to decide who to lend money to in an investment decision experiment. Data for the underlying investment decisions are taken from a real-life microfinance market, allowing us to combine field data with a laboratory experiment approach. We find that subjects are able to use both soft and hard information to guide their investment decisions. Subjects are more likely to invest in a borrower, when the available information is associated with a higher likelihood of actual repayment. Subjects provided with both types of information are able to make better decisions, compared to subjects provided with only soft or hard information. We elicit and compare the value subjects place on information with the effectiveness of that information in improving decisions. We find that in general, subjects display tremendous overconfidence in their ability to use the information provided to make accurate decisions.

Author(s): Qiyan Ong and Walter Theseira
Topic: Markets: Finance, Applied Economics: Other
Back to session: Trust and Trustworthiness


Ageism & Cooperation

Eric Schniter
Economic Science Institute - Chapman University

ABSTRACT:

Age-based discrimination can be socially and economically undesirable yet we know little about the accuracy of age stereotypes and their effects on cooperation. To investigate the effects of age stereotypes on cooperation, we presented older adults (over age 50) and younger adults (under age 25) with incentivized belief elicitation tasks (about anticipated sender-receiver interaction behaviors) and then a series of same, different, and unknown-aged group interactions in a sender-receiver game. Younger and older adults shared beliefs about uncooperative younger adults and cooperative older adults. Consensual stereotypes about age groups demonstrated wisdom of crowds. Individuals consistently stereotyped their out-group as relatively different and more uncooperative than observed to be. While younger adults earned more by acting uncooperatively with older adults, older adults earned less by acting relatively cooperative with younger adults (despite beliefs). We discuss how costs and benefits of ageism are relevant to intergenerational cooperation in an aging society.

Author(s): Eric Schniter, Timothy Shields
Topic: Decision Theory: Beliefs, Social Behavior: Group Behavior
Link: http://papers.ssrn.com/sol3/Delivery.cfm/SSRN_ID2175778_code1711759.pdf?abstractid=2175778&mirid=1
Back to session: Trust and Trustworthiness


Financial Information and Advice

Markus Noth
University of Hamburg

ABSTRACT:

Consumer protection efforts in financial markets have recently been increased. More information on products must be provided and several restrictions on financial advisers have been implemented or are discussed. We investigate two open questions. First, we explore the assumption that more information actually leads to better decisions with and without additional advice. Second, knowing about commissions might either help to minimize them or shift the focus to one aspect of the overall fees. Our experimental results show that advice and information on commissions reduce the positive effect of financial product information especially for subjects with rather low financial literacy.

Author(s): Markus Noth
Topic: ,
Back to session: Trust and Trustworthiness


Trust and Contracts: An Experimental Study

Benjamin Ho
Cornell

ABSTRACT:

The relationship between trust and the legitimacy of institutions has been analyzed in fields ranging from institutional theory in sociology, political science, psychology and economics. Economic studies (Knack and Keefer, 1997; La Porta et al., 1997, 1998, 2008; Zack and Knack, 2001; Algan and Cahuc, 2010; Ho and Hoffman, 2013) have shown that understanding the relationship between trust and the rule of law has important implications for understanding development, investment decisions, and the distribution of wealth. Trust between investors and entrepreneurs is an important facilitator of investment and production. Campos-Ortiz et al (2012) found that experimental subjects coming from countries with higher levels of trust devote more resources to production and are more likely to refrain from theft. Despite the importance of the subject, there has been little study on how institutional trust affects contracts are written. Our paper analyzes the effect of institutional trust on contract structures, specifically on the way gains are shared, by developing an investment game in which subjects contract on how to share the surplus. We vary institutional trust by providing a dispute resolution system that can accommodate contract breach by the agent, assigning some of our subjects to a game in which judges can be bribed and others where this is not possible. We analyze the differences in contracts that arise as a result of the different environments, and look at the total investments in the two groups. We compare our experimental findings to the analytical predictions given by a behavioral game theory model that accounts for fairness, reciprocity and social norms.

Author(s): Benjamin Ho and Evsen Turkay
Topic: Social Behavior: Lying and Cheating, Social Behavior: Norms and Morals
Back to session: Trust and Trustworthiness


Are Political Statements Only Expressive? A Laboratory Experiment

Cortney Rodet
Florida State University

ABSTRACT:

Many forms of political activity, such as voting, are often thought to be expressive rather than instrumental, that is, the action taken is not expected to influence directly the political outcome. We present evidence suggesting the presence of an (perceived) instrumental benefit of political statements by individuals. Two hundred subjects have the opportunity to make a sometimes costly comment on another person's choice of a political donation between two rival political groups. In one treatment, the decision-maker receives the comments before the choice is made - when they may have some persuasive impact. In the other, the comments come after subjects know the result. Subjects are 100% more likely to comment in the pre rather than the post treatment; moreover, most (75 %) of the comments made prior to the decision suggest a plan of action to the decision maker, while most (93 %) of the comment made after affirm one or another of the political positions. That is, political expression is expressive, but is also instrumental. Finally, in both treatments, the rate of commenting is inversely related with the price, indicating that political expression--both instrumental and expressive--is rational.

Author(s): Cortney S. Rodet, Jared P. Barton
Topic: Social Behavior: Communication,
Back to session: Voting Behavior


Weather, mood, and voting: An experimental analysis of the effect of weather beyond turnout

Anna Bassi
UNC at Chapel Hill

ABSTRACT:

Theoretical and empirical studies show that inclement weather on an election day reduces turnout, potentially swinging the results of the election. Psychology studies, however, show that weather affects individual mood, which -- in turn -- affects individual decision-making activity potentially beyond the simple decision to turn out. This paper evaluates the effect of weather, through its effect on mood, on the way in which voters who do turn out decide to cast their votes. The paper provides experimental evidence of the effect of weather on voting when candidates are perceived as being more or less risky. Findings show that, after controlling for policy preferences, partisanship, and other background variables, bad weather depresses individual mood and risk tolerance, i.e., voters are more likely to vote for the candidate who is perceived to be less risky. This effect is present whether meteorological conditions are measured with objective or subjective measures.

Author(s): Anna Bassi
Topic: Public Choice: Voting and Rent Seeking, Decision Theory: Risk
Back to session: Voting Behavior


Getting Out the Vote: Information and Voting Behavior

Yi-Yi Chen
Washington University in St. Louis

ABSTRACT:

This paper uses laboratory experiments to study the impacts of information on voting behavior in the Palfrey-Rosenthal pivotal voter model (P-R model) and compares two information-revealing mechanisms: One through which active supporters show their support without paying costs (e.g., polls), which can be viewed as cheap talk; the other where active supporters have to pay their time (e.g., participating in campaign activities) or money (e.g., contributing to super PACs) to support their preferred candidates, thus providing more certainty in the form of actual levels of support for a party than the first mechanism. In experiments, active supporters, played by robots, are set to always vote, while human participants are expected to vote strategically. To capture the difference between the two mechanisms, the first mechanism reveals the distribution of active supporters of a party, while the second one provides the actual number of the active supporters of that party. Three findings emerged. (a) In general, participants followed the main ideas of the P-R model, with appropriately responding to the cost of voting and the belief of being pivotal. (b) However, when participants were informed of being in an advantageous position by the second mechanism, they became more willing to vote, which can be attributed to that leading in an interim stage has a positive psychological impact on performance in tournaments. (c) In the first mechanism, participants overestimated their probability of being pivotal when they were in an advantageous position but underestimated it when being in a disadvantageous position.

Author(s): Yi-Yi Chen
Topic: Public Choice: Voting and Rent Seeking, Games: Contests
Back to session: Voting Behavior


Sequential Search by Committees with Rank-dependent Payoffs: An Experimental Study

Darryl Seale
University of Nevada Las Vegas

ABSTRACT:

We extend the classical secretary problem by studying sequential search by committees. In our setup, the committee consists of two players who inspect multi-attribute alternatives (applicants) one at a time with no backward solicitation. The players have uncorrelated absolute ranks over the alternatives (rank 1 being the best), and each player's payoff decreases with his absolute rank of the accepted alternative. Absolute ranks are not revealed until the search terminates; players only learn the relative ranks of the alternatives as they inspect them sequentially. Results from a controlled experiment show that subjects over-searched relative to equilibrium, especially when each player was not informed about the other player's ranking of the presented alternatives. Decisions were affected by theoretically irrelevant cues including the relative rank of the previous alternative and the other player's relative rank of the current alternative.

Author(s): Vincent Mak, Amnon Rapoport, and Darryl A. Seale
Topic: Games: Coordination, Games: Repeated Games
Link: https://dl.dropboxusercontent.com/u/25430273/Sequential%20search-April%2011-2013.pdf
Back to session: Voting Behavior


Strategic Self-Ignorance

Linda Thunstrom
University of Wyoming

ABSTRACT:

We examine strategic self-ignorance-the use of ignorance as an excuse to over-indulge in pleasurable activities that may be harmful to one's future self. Our model shows that guilt aversion provides a behavioral rationale for present-biased agents to avoid information about negative future impacts of such activities. We then confront our model with data from an experiment using prepared, restaurant-style meals-a good that is transparent in immediate pleasure (taste) but non-transparent in future harm (calories). Our results support the notion that strategic self-ignorance matters: nearly three of five subjects (58 percent) chose to ignore free information on calorie content, leading at-risk subjects to consume significantly more calories. We also find evidence consistent with our model on the determinants of strategic self-ignorance.

Author(s): Linda Thunstrom, Jonas Nordstrom, Jason Shogren, Mariah Ehmke, Klaas van 't Veld
Topic: Decision Theory: Risk, Decision Theory: Ambiguity
Back to session: Dynamic Choice


Time and State Dependence in an Ss Decision Experiment

Ryan Oprea
UCSB

ABSTRACT:

Flow earnings in a laboratory experiment decline the further a Brownian state variable, z, evolves from its optimal level z*. Optimal state dependent models predict subjects will pay a fixed cost to return z to z? only when z strays outside a critical inaction region around the optimum. Subjects average adjustment points are remarkably close to optimal levels, but as in the field they do not establish true state dependent inaction regions, suggesting significant time dependent components in adjustment rules. Structural estimates of the parameters of a bounded rationality model suggest subjects experience substantial cognitive costs from responding to the state, accounting for these patterns. Cross treatment results suggest that these costs - and the resulting degree of state dependence in adjustment - are powerfully influenced by the volatility of the stochastic process, a finding with potentially important policy implications.

Author(s): Ryan Oprea
Topic: ,
Back to session: Dynamic Choice


Markov Strategies: An Experimental Test

Emanuel Vespa
UCSB

ABSTRACT:

In infinite time horizon Dynamic Games, subgame-perfect equilibria in which agents use Markov strategies is the most popular solution concept. In this paper we provide an experimental test of the restriction to Markov strategies. We study two simply dynamic games. A dynamic version of the prisoners' dilemma and a dynamic version of a common pool problem. The key difference between these games is that in the former an agent's action affects the other's payoffs in the current period, while in the latter it affects payoffs only in the future. Our experimental design allows us to recover an estimate of the strategies that our subjects use and we can assess the popularity of Markov behavior.?

Author(s): Emanuel Vespa (UCSB) and Alistair Wilson (U Pittsburgh)
Topic: ,
Back to session: Dynamic Choice


Hotelling in Continuous Time and Action Spaces

Curtis Kephart
UCSC

ABSTRACT:

We investigate experimentally the importance of continuous time on the four-player Hotelling location game with a uniform customer density and inelastic demand. The static pure strategy Nash equilibrium consists of two firms paired-up at the first and third quartiles of the linear city. In repeated simultaneous move games (discrete time grid), we largely replicate previous findings, subjects fail to converge to the static equilibrium. In continuous time treatments where subjects are able to change actions asymmetrically, the static game pure strategy equilibrium more clearly organizes the data.

Author(s): Curtis Kephart
Topic: ,
Back to session: Spatial Interactions


Why Central Paris is Rich and Downtown Detroit is Poor: A Laboratory Experiment

Laurent Denant-Boemont
University of Rennes 1 & CREM-CNRS

ABSTRACT:

In major urban areas, income heterogeneity for households may produce differ- ent patterns of spatial segregation. In Paris for instance (e.g., see Brueckner et al, 1999), richer households tend to be located in the city centre whereas poorer ones are more frequently located in the suburbs. In other cities, like Detroit or Sydney, a reverse location pattern is observed, since richer households are to be often found in the suburbs, a signification fraction of poorer households residing in CBD (Central Business District, see, e.g. Glaeser et al., 2008). Such stylized facts may be explained by urban economic theory, in particular by refinements of the canonical monocentric model of Alonso (1964). More precisely, it can be shown that location pattern depends, assuming income heterogeneity for house- holds, upon the relative value of income elasticities viewing two consumption goods, housing demand and transport demand. In particular, if consumption of housing is more (less) income elastic than commuting cost (viewed as the opportunity cost of time, see Becker, 1965), then the rich will live in the suburbs, as they will bid higher (lower) than poorest households for lands located there. As a consequence, various location patterns across metropoles may be explained by such comparative statics regarding bid-rent equilibrium.

Author(s): Henri Busson and Laurent Denant-Boemont
Topic: Applied Economics: Other, Markets: Industrial Organization
Back to session: Spatial Interactions


Urban Costs and the Spatial Structure of Cities: A Laboratory Experiment

Sabrina Hammiche
University of Rennes 1 & CREM-CNRS

ABSTRACT:

This paper presents a laboratory experiment to investigate how urban costs might determine the internal economic structure of urban areas (either monocentric or polycentric) by influencing location choices of firms and households. Worplaces can be mostly located in the CBD, giving a monocentric city, or decentralized within the suburbs, taking the shape of a polycentric city. Our experimental design is in part guided by the theoretical model of Cavailhes et al. (2007) that shows how the trade-off between communication costs faced by firms in order to access essential facilities located in the CBD, and commuting costs borne by workers, determine the degree of cities' polycentricism by influencing the distribution of the workplaces within it. In the experiment, groups of 16 subjects participated in a two-step auction game, where firms and workers first negotiated to find a job contract and second where workers bid for land in order to find a home. The game is repeated for four rounds and different experimental treatments are implemented, each defining a given scenario for communication costs and commuting costs. The chosen benchmark is a polycentric city treatment, where no communication cost exists for firms, giving them no incentive to locate in the CBD. In two other treatments, the communication cost is positive. In the Monocentric City treatment, commuting cost for workers is low, giving a clear incentive for firms to locate in the CBD, as workers do not suffer from potential high commuting costs, giving an outcome where all firms should locate in the CBD. In the Hierarchical City treatment, the commuting cost is very high for workers, giving workers a strong incentive to live close to their workplace and accepting lower wages as a result. Experimental results are in line with theoretical predictions: firms tend to locate in the CBD under the monocentric treatment, whereas more distant locations are accepted leading to a hierarchical outcome as workers propose lower wages for not working in the CBD so as to escape high commuting costs.

Author(s): Michiel Bliemer, Laurent Denant-Boemont, Sabrina Hammiche, David Hensher and Corinne Mulley
Topic: Applied Economics: Other, Markets: Industrial Organization
Link: https://docs.google.com/viewer?a=v&pid=sites&srcid=ZGVmYXVsdGRvbWFpbnxsYXVyZW50ZGVuYW50Ym9lbW9udHxneDo0MDY2ZmFhZjhjMjBlMzE0
Back to session: Spatial Interactions


Good News, Bad News, and Social Image: The Market for Charitable Giving

Luigi Butera
GMU

ABSTRACT:

This paper experimentally investigates how donors respond to news about the efficiency of their charities, that is, to real prices of giving greater than 1, and how the response depends on that information being public or not. We find that as long as charity efficiency remains private information, individuals reward better-than-expected charities (good news) by increasing their donations. On the other hand, donors largely ignore bad news when giving happens under full anonymity. However, when charity efficiency is revealed to others, some donors decrease their contribution in response to good news, and they increase it when news is bad. This emergent behavior accounts for 34% of subjects that do respond to new information. We show that the latter behavior is driven by image-motivated donors, who treat the size of their gift and the efficiency of their recipients as substitutes in terms of social image payoffs.

Author(s): Luigi Butera
Topic: ,
Back to session: Charity 2


Dual-Process Reasoning and Giving Behavior

Zach Grossman
UCSB

ABSTRACT:

Recent experiments on whether cognitive load will increase or decrease altruistic behavior have generated conflicting results. In a giving a experiment, we combine a cognitive load manipulation with an affective system manipulation designed to induce sympathy for the recipients, to test the hypothesis that depleting the resources of the cognitive system will lead to a greater increase in giving when the affective system is already activated.

Author(s): Zachary Grossman (UCSB) and Joel van der Weele (Amsterdam)
Topic: Applied Economics: Charitable Giving, Decision Theory: Bounded Rationality
Back to session: Charity 2


Substitutes vs. Compliments in Charitable Donations

Alexander Brown
Texas A&M

ABSTRACT:

In a previous study, Brown Meer and Williams (2013), we develop a platform where experimental subjects earn money by performing an unpleasant task at a fixed piece rate and have an option to donate to charity. In a variety of sessions, we found subjects would unambiguously give more to charity when they worked directly for charity than when they work for themselves and could donate their earnings to charity. We test the robustness of this result by varying the fixed piece rate in which subjects work for themselves or charity. Subjects may either work for themselves and give their earnings to charity or work directly for charity. Some subjects do substitute to the higher earning activity. However, there is a general preference for working for charity even when the piece rate is greater when working for oneself than charity. That is, individuals still chose to work directly for charity even though they could benefit charity more by working for themselves at a higher rate and donating their earnings to charity. We parameterize a charitable giving utility model and find our results are consistent with a differential warm glow of charitable donations of time and money.

Author(s): Brown, Alexander L., Jonathan Meer, and J. Forest Williams, Texas A&M University
Topic: ,
Back to session: Charity 2


The Impact of Overconfidence in Acquiring Information

Klajdi Bregu
University of Arkansas

ABSTRACT:

The goal of this research project is to quantify overconfidence in undergraduate students and determine whether this overconfidence leads to a preferred type of feedback. This is accomplished by using a lab experiment composed of two parts. The first part establishes the measures of overconfidence, while the second part looks at how students acquire helpful information. An exit survey was given in order to relate these findings to major types and other demographics. The results are useful in assessing whether students have accurate perceptions of their performance and the performance of others. Various factors are measured to determine their effect on willingness to accept help. We find that students who are overconfident tend to ask for much less help than those who are not overconfident.

Author(s): Klajdi Bregu
Topic: Decision Theory, Games: Information
Back to session: Preferences and Behavior


Small Stakes Risk Aversion and the Endowment Effect

Erin Fairweather
Mount Saint Mary's College

ABSTRACT:

We study the interaction of the endowment effect and risk aversion under alternative preference elicitation mechanisms. We induced an endowment effect in both the paired-choice Multiple Price List design and a novel paired- choice Becker-DeGroot-Marschak design by giving subjects scratch-off lottery tickets at the beginning of the experiment; some subjects were given a high-variance ticket and some a low-variance ticket, and all were offered the opportunity to trade their endowed ticket for the alternative. We find evidence of an endowment effect for both mechanisms, but subjects in the Multiple Price List mechanism appear to be significantly more risk averse than their counterparts in the Becker-DeGroot-Marschak mechanism, a result consistent with simpler non-paired-choice lotteries in Sprenger (2011) and James (2007, 2011).

Author(s): Erin Fairweather and Sean Crockett
Topic: Risk Aversion, Endowment Effect
Back to session: Preferences and Behavior


Eliciting Preferences with Rank-ordered? Choices

Sean Crockett
Baruch College

ABSTRACT:

In two separate papers, we introduce a novel mechanism that permits preference elicitation over many dimensions. Subjects are asked to rank a set of consumption bundles from favorite to least favorite. The probability that a given bundle is awarded to the subject is increasing in its rank, thus making the mechanism incentive compatible. With Ryan Oprea, I use the mechanism to study the endowment effect. Subjects are endowed with one of the ranked bundles, making the mechanism incentive compatible down to the rank of the endowment, and participate in the mechanism twice, permitting study of how elicited preferences change after the acquisition of new endowments. With Craig Brown, I use the mechanism to study social preferences in a dictator game, where the two dimensions used in the experiment are money to self and money to an anonymous other.

Author(s): Sean Crockett, Craig Brown and Ryan Oprea
Topic: Preference elicitation, Endowment effect
Back to session: Preferences and Behavior


How individual preferences get aggregated in groups - An experimental study

Ben Greiner
University of New South Wales

ABSTRACT:

This paper experimentally investigates how individual preferences, through unrestricted deliberation, get aggregated into a group decision in two contexts: reciprocating gifts, and choosing between lotteries. In both contexts we find that median group members have a significant impact on the group decision, but particular other members also have some influence. Non-median members closer to the median tend to have more influence than extreme members. We do not find evidence that group choice exhibits a shift in a particular direction that is independent of member preferences and caused by the group decision context itself. We find that group deliberation not only involves bargaining and compromise, but it also involves persuasion: preferences tend to shift towards the choice of the individual's previous group, especially for those with extreme initial preferences.

Author(s): Attila Ambrus, Ben Greiner, Parag Pathak
Topic: ,
Back to session: Preferences and Behavior