Archives du séminaire Behavior seminar

Behavior seminar

Le 25/04/2024 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Using consumer surveys and barcode data, this study reveals a persistent 20% gap in red meat consumption between single men and women in the US. Investigating whether this disparity stems from gender stereotypes, we collect survey data to assess the influence of attitudes, beliefs, and implicit biases on meat consumption. Findings suggest that the gap is largely due to preferences and perceived needs rather than differences in beliefs about environmental, health, or ethical impacts. Moreover, an Implicit Association Test uncovers a strong bias linking meat with masculinity. The study further examines the stickiness of these consumption stereotypes through an experiment with identity priming and a de-biasing treatment, analyzing their effects on meat consumption expectations and conjoint analysis outcomes.

Colson-Sihra Eve () How Sticky are Consumption Stereotypes? Evidence from the Meat Gender Gap

Behavior seminar

Le 04/04/2024 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


People sometimes avoid information about the impact of their actions as an excuse to be selfish. Such “willful ignorance” reduces altruistic behavior and has detrimental effects in many consumer and organizational contexts. We report the first meta-analysis on willful ignorance, testing the robustness of its impact on altruistic behavior and examining its underlying motives. We analyze 33,603 decisions made by 6,531 participants in 56 different treatment effects, all employing variations of an experimental paradigm assessing willful ignorance. Meta-analytic results reveal that 40% of participants avoid easily obtainable information about the consequences of their actions on others, leading to a 15.6-percentage-point decrease in altruistic behavior compared to when information is provided. We discuss the motives behind willful ignorance and provide evidence consistent with excuse-seeking behaviors to maintain a positive self-image. We investigated the moderators of willful ignorance and address the theoretical, methodological, and practical implications of our findings on who engages in willful ignorance, as well as when and why

Shalvi Shaul () Ignorance by Choice: A Meta-Analytic Review of the Underlying Motives of Willful Ignorance and Its Consequences

Behavior seminar

Le 28/03/2024 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


In a world where the majority and the minority group have equal distributions of talent, where candidates are objectively and accurately evaluated, and no discrimination occurs, the underrepresentation of the minority group in selective positions is nonetheless highly sticky. If the sample of candidates from the minority group is numerically smaller, at equal distribution of talent, the most qualified candidate is more likely to belong to the majority sample, mirroring its larger numerical size. If future samples of candidates respond to the realized selection in the expected direction–increasing if the selection came from the sample, decreasing or increasing less if it did not–the higher probability of success in the majority sample will persist. We capture this process with a well-known statistical model: the Polya urn. The richness of existing results and the streamlined model allow us to study and compare different policy interventions. A simple app (https://caron.shinyapps.io/Women-Men-Polya-Urns/) allows readers to run their own experiments. Two robust results are that temporary affirmative actions interventions have long -term equalizing effects, and that any decline in the quality of selected candidates is self-correcting, even while the intervention lasts.

Casella Alessandra () Women, Men, and Polya Urns. On the Persistence of Underrepresentation

Laura Caron, Alessandra Casella and Victoria Mooers

Behavior seminar

Le 21/03/2024 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We propose a method to identify individuals' marriage markets under the assumption that the observed marriage allocations are stable. Our specific aim is to learn about (the relative importance of) the individual's observable characteristics that define these markets. In a first step, we use a nonparametric revealed preference approach to construct inner and outer bound approximations of individuals' marriage markets from the observed marital matchings. Then, we use the machine learning method called Support Vector Machine (SVM) to estimate a robust boundary between these inner and outer bound approximations. The method estimates the threshold (as a linear function of individual characteristics) that defines whether two potential partners operate in the same marriage market. We demonstrate the practical usefulness of our method through an application to Dutch household consumption data

MERLINO Luca Paolo () Identifying Marriage Markets”, which is a joint work with Bolletta (Saclay), Cherchye (KUL), De Rock (ULB) and Demuynck (ULB).

Behavior seminar

Le 14/03/2024 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We use data from the SHARE and HRS surveys (2010-2020) to estimate the relationship between socio-economic status, mortality, risk and duration of dependency for people aged 50 and over. The results show that wealthier individuals have a higher probability of survival, regardless of the estimation method used (OLS or IV). The strength of this relationship varies from country to country, and in Europe we observe that in the more Bismarckian countries wealth explains survival more than in the Beveridgian countries. Finally, on the basis of a Cox survival model, we also show that the least wealthy individuals are those who have a greater probability of becoming dependent, but also those who experience the longest periods of loss of autonomy in old age. Our results therefore identify a triple penalty linked to socio-economic status, summarised by the finding that the poorest people live for a shorter time, are more likely to become dependent and do so for a longer period.

Schoenmaeckers Jérôme () We are not all equal! Impact of socio-economic status on loss of autonomy in the old age

Mathieu Lefebvre, Université de Strasbourg; Jérôme Schoenmaeckers, HEC-Liège

Behavior seminar

Le 07/03/2024 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


The paper reports from a large-scale study of people’s fairness preferences and beliefs, where 65 000 individuals from 60 countries make real distributive choices. We establish causal evidence on the role of the source of inequality and efficiency considerations for inequality acceptance, and we provide a rich description of people’s beliefs about the main sources of inequality and the cost of redistribution. We find large heterogeneities in both preferences and beliefs and show that they are strongly associated with people’s policy views on redistribution. The paper also studies how people’s fairness views relate to various country characteristics. In particular, we show that there are striking differences between the developed and developing countries in both fairness preferences and beliefs.

Almås Ingvild () Fairness Across the World

Behavior seminar

Le 29/02/2024 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We investigate fairness preferences regarding matching mechanisms by employing a spectator design. Participants choose whether they want to make a costly voting decision for the Boston mechanism or the serial dictatorship mechanism (SD) played by other participants. The Boston mechanism generates justified envy since some participants are forced to submit their true preferences, while the strategy-proof SD satisfies envy-freeness. A high share of individuals vote for the Boston mechanism when priorities are based on merit, and this share further increases when priorities are determined by luck. At the same time, fairness as envy-freeness and strategyproofness plays a role in people's revealed preferences when the priorities are based on merit. The results suggest that fairness as envy-freeness and strategy-proofness plays a role for spectators' voting decisions but many people believe that clever strategic choices create entitlements on their own.

MECHTENBERG Lydia () Fairness in Matching Markets: Experimental Evidence

Dorothea Kübler, Tobias König, Lydia Mechtenberg, and Renke Schmacker

Behavior seminar

Le 14/12/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


A prevailing narrative suggests that citizens who live under democratic rule take it for granted, potentially leading to democratic backsliding. This paper advances a more optimistic perspective: electoral turnovers—elections displacing the head of government—can rejuvenate democratic support among citizens, as they signal political elites’ commitment to democratic ideals and their willingness to relinquish power upon defeat. To test this argument, I aggregated data from multiple international surveys, covering approximately 500,000 respondents whom I then matched with a century’s worth of electoral data. This dataset allows me to identify the electoral turnovers to which each respondent has been exposed throughout their lives. I show that past exposures to turnovers, especially the first one, bolster support for democracy for about 15 years. In line with the proposed mechanism, the effect is stronger in new and imperfect democracies where the commitment of political elites to democratic ideals is uncertain.

BOL DAMIEN (KCL(London)) Renewing Democracy: How Past Exposures to Electoral Turnovers Reinforce Citizen Support

Behavior seminar

Le 07/12/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4

SHOGREN Jason (KCL(London)) *; () ;

La séance est annulée

Behavior seminar

Le 30/11/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Economic theory provides us with models of network formation. The multiplicity of equilibrium and the possibility of large payoff inequality in these models motivate an experimental study of their validity. This paper presents an experiment based on the unilateral link formation model by Bala and Goyal (2000). In this model, an individual can choose to form links with others in order to access benefits. The value flow can be either one-way (benefits only from forming a link) or two-way (benefits from forming or receiving a link) in the game. We consider two group sizes (n = 10 and n = 50) for each value flow model (4 treatments in total). We find that the subjects’ performance is relatively close to the theoretical prediction in the two-way case for all groups. That is, agents create sparse, unequal, and small average distance networks and achieve high aggregate efficiency. In the one-way flow case, consistent with the theoretical prediction, the networks are sparse and have a larger average distance and smaller degree inequality, but the deviation from the theoretical prediction is larger as compared to the two-way model, especially when the group size is large. The poorer performance in the one-way case is due to the fact that the efficient network structure —the cycle network— is more difficult to form as compared to that of the two-way model — the star network, holding individual decision accuracy constant.

Moisan Frédéric (KCL(London)) An Experiment on a Linking Game

Behavior seminar

Le 23/11/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Recent work on finite decision trees has characterized pre-rational preference relations over the space of Anscombe/Aumann consequence lotteries as those that are complete and transitive, as well as satisfying the independence axiom for behaviour under risk, and the sure-thing principle for behaviour under uncertainty. In case the preferences are continuous over Marschak triangles, these properties imply that pre-rational preferences have a subjective expected utility representation. This past work considered trees containing four kinds of node: (i) decision nodes; (ii) terminal nodes with consequences; (iii) chance nodes at which a roulette lottery is resolved; (iv) event nodes at which a horse lottery is resolved. This talk will recognize the relevance of three further kinds of node: (v) quantum nodes, which can be replaced by a combination of decision, event and chance nodes; (vi) timed consequence nodes, which need not be terminal nodes; (vii) enlivenment nodes, which can be initial nodes of an enlivened continuation subtree about which nothing can be predicted before at least one earlier decision has to be made. The main focus will be on trees with timed consequence nodes in which, in the absence of menu consequences that include non-trivial continuation subtrees, the previous results on pre-rational preferences apply for an extended domain of elongated consequences taking the form of intertemporal consequence streams. The last part of the talk will consider whether preferences without an expected utility representation, like those in the potential addict example or in the Allais and Ellsberg paradoxes, can be made pre-rational by extending them to give weight to menu consequences

HAMMOND Peter (KCL(London)) An Arboretum of Decision Trees


Texte intégral

Behavior seminar

Le 16/11/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We investigate how inequality affects what is considered most necessary to purchase. Using consumption surveys of frequent purchases from poor households in India, we estimate a structural model of demand to isolate the impact of inequality on needs from supply-side effects. We rely on plausibly exogenous inequality shocks following the 1991 liberalization reforms. We find that inequality increases the need of the poor for "little luxuries" (sweets, processed food and drinks, intoxicants), which generates significant expenditure reallocation at the expense of calorie intake. This accounts for three fourth of the decline in calorie consumption of the poor over the period.

BELLET Clément (KCL(London)) Does Inequality Affect the Needs of the Poor?"

Behavior seminar

Le 09/11/2023 de 16:00:00 à 17:00:00

Bâtiment A, Rez de chaussée, Salle 4


The “social welfare function” (SWF) framework for policy evaluation grows out of theoretical welfare economics and is widely used in some areas of economics, such as optimal tax theory and climate economics. The framework has three key components: an interpersonally comparable well-being measure, which converts the possible social outcomes of policy choice into vectors (lists) of interpersonally comparable well-being numbers; the SWF itself, a rule for ranking well-being vectors, such as a utilitarian SWF or, instead, a “prioritarian” SWF that gives extra priority to the worse off; and an “uncertainty module,” for ranking policies understood as probability distributions across outcomes.

ADLER Matthew (KCL(London)) Social Welfare Functions and Health Policy

Behavior seminar

Le 19/10/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


A multidimensional well-being measure that respects individual preference heterogeneity, even when these preferences are incomplete, is presented in this paper. We discuss how the proposed well-being measure can be implemented in a non-parametric way, using the Adaptive Bisectional Dichotomous Choice (ABDC) method (Decancq and Nys, 2021). The ABDC method consists of two steps. In the first step, respondents are presented with a series of dichotomous choices between pairs of life situations that consist of their actual and hypothetical life situations. Considering three dimensions of well-being (income, health, and social relationships) we measure well-being for a final sample of 2288 Dutch citizens. About 27 percent of respondents indicate to have incomplete preferences over these three dimensions, by selecting the response category ”I don’t know” in the algorithm. Results indicate that individuals with better health are less likely to answer “I don’t know”, suggesting that individuals have more complete preferences when assessing situations similar to their current lives. On average, respondents are willing to trade off approximately 24% of their income for perfect health and social relationships (although the variability is large). The identification of the worst-off individuals is dependent on the well-being measure used, underscoring the importance of careful consideration when selecting appropriate measures of well-being. Furthermore, our data enable us to compare three methods for eliciting preferences to estimate our multidimensional measure of well-being: subjective well-being, contingent valuation, and ABDC. The evidence suggests that results are more consistent for individuals with complete preferences.

BURONE SCHAFFNER Santiago German () Measuring multidimensional well-being when preferences differ: a non-parametric approach

Behavior seminar

Le 28/09/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


An occupation is usually characterized as a combination of what people do and how much they are paid for it, with little attention to the fact that work arrangements also define when people are paid for their labor. This paper contributes to this discussion by investigating how much value people assign to having a short delay between their tasks and the associated compensation. Using a national experimental survey with ridesharing drivers in Brazil, I document that this population is willing to forgo about 40 percent of their earnings, on average, to receive on the same day of their rides, compared to the alternative of being paid with a month's delay. Text analysis methods provide evidence that short delay-to-pay tends to be more critical if working more hours is one's primary adjustment margin in response to financial emergencies. Finally, I provide experimental evidence that increased attention to their domestic budget can make drivers marginally more likely to take up delayed compensation under large multipliers. Those three results suggest that the payment schedule can be a crucial labor market feature for workers under constrained liquidity.

Scarelli Thiago () Worker’s Preference over Payment Schedules: Evidence from Ride-Hailing Drivers

Behavior seminar

Le 21/09/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Despite the well-known health consequences of lead exposure, an estimated 6 to 10 million lead service lines still deliver drinking water to homes throughout the US. Disadvantaged communities are disproportionately exposed to lead service lines, contributing to health and human capital disparities. This paper studies the effects of public lead service line replacements using children's blood lead test data with confidential address information, home sales data, and geocoded public service line installation data from Rhode Island. Replacing public lead service lines significantly reduces child blood lead levels by about 0.4 ug/dL, or 13 percent, and increases the price of home sales by 7-8 percent, indicating that homeowners value these replacements

Marcus Michelle () Burying the Lead: Effects of Public Lead Service Line Replacements on Blood Lead Levels and Property Values

Behavior seminar

Le 07/09/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


The present paper analyzes optimal redistributive income taxation in a Mirrleesian framework extended with other-regarding preferences at the individual level. We start by examining a general model where the other-regarding preference component of the individual utility functions is formulated to encompass almost any form of preferences for other people’s consumption, and then continue with four prominent special cases. Two of these reflect self-centered inequality aversion, based on Fehr and Schmidt (1999) and Bolton and Ockenfels (2000), whereas the other two reflect non-self-centered inequality aversion, where people have preferences for a low Gini coefficient and a high minimum consumption level in the society, respectively. We find that other-regarding preferences may substantially increase marginal tax rates, including the top rates, and that different types of other-regarding preferences have very different implications for optimal taxation

JOHANSSON STENMAN Olof () Optimal Taxation and Other-Regarding Preferences, by Thomas Aronsson and Olof Johansson-Stenman.

Behavior seminar

Le 29/06/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


A significant proportion of primary care expenditures arise from the prescription of drugs, with physicians having some degree of autonomy in deciding whether and which drug to choose for a particular patient. While medical reasons and guidelines primarily influence these decisions, economic factors may also play a role. Given the ongoing trend of practice mergers or closures in the English NHS over several decades, we investigate the per-patient prescribing volume and costs in increasingly concentrated markets. We employ quarterly data from 2015 to 2019 for 7,300 practices in the English national health system. Concentration is evaluated using a population-weighted Herfindahl-Hirschman Index that is based on predicted patient flows rather than observed ones to address endogeneity concerns. Our findings suggest that greater market concentration leads to a reduction in total drug prescriptions and their associated costs, particularly for generics and over-the-counter drugs. However, the concentration effect does not appear to influence primary care prescriptions for insulin. Therefore, the impact is more significant for drugs that allow GPs greater discretion in their prescribing behavior.

HERR Annika () Market Concentration and Prescribing Behaviour in Primary Care

David Probst and Nils Gutacker

Behavior seminar

Le 08/06/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


This paper explores the relationships between overall diet quality and attitudes toward risk and time in a general population survey. Our survey combines (i) a stateof- the-art food frequency questionnaire with (ii) a choice-based preference module to elicit individual risk and time preferences. We conduct this survey on a representative sample of the French population. Using a Hierarchical Bayes framework, we jointly estimate individual risk aversion and impatience parameters. We show that risk and time preferences signicantly explain the individual heterogeneity in key aspects of diet quality, even after controlling for socio-demographic characteristics. We nd that more impatient and more risk-seeking individuals have a poorer overall diet quality, that more impatient individuals have a higher daily energy intake, and that more risk-seeking individuals consume more alcohol.

NEBOUT Antoine () What You Eat is What You Are: Risk Attitude, Time Preference, and Diet Quality

Antoine Nebout ,Emmanuel Kemel, Florent Vieux , Sandrine Péneau, Nicole Darmon, Noemi Berlin , Emmanuel Paroissien

Behavior seminar

Le 01/06/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Are the United States still a land of opportunity? We provide new insights on this question by invoking a novel approach that allows us to measure unfair inequality in the joint distribution of income and wealth. We show that inequality of opportunity has increased by 56% over the time period 1983-2016. Increases are driven by two distinct forces: (i) a less opportunity-egalitarian distribution of income until 2000, and (ii) a less opportunity-egalitarian distribution of wealth after the financial crisis in 2008. In sum, our findings suggest that the US have consistently moved further away from a level playing field in recent decades.

HUFE PAUL () Multidimensional Equality of Opportunity in the United States

Behavior seminar

Le 24/05/2023 de 14:00:00 à 15:00:00

Bâtiment A, Rez de chaussée, Salle 4


In addition to offering many insights concerning public policy, Behavioral Economics challenges the premises of standard Welfare Economics. Behavioral Welfare Economics seeks to either modify the standard choice-based methods for measuring economic well-being so that they accommodate the issues that arise in Behavioral Economics, or to replace them with something else, such as hedonic measurement. This talk will provide a broad conceptual overview of the challenges that arise in Behavioral Welfare Economics, as well as the potential solutions that have emerged over the past twenty years, along with new developments.

BERNHEIM Douglas () The Challenges of Behavioral Welfare Economics

Behavior seminar

Le 17/05/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


This presentation will summarize ongoing work examining impacts of early life environments in the early 20th Century in the US and associated effects on longevity for these cohorts using newly released linked data between the 1940 Complete Count Census and Death Records. Specific examples include in utero exposure to pollution, natural disasters, and clean water.

FLETCHER Jason () Early Life Conditions and Old Age Mortality

Behavior seminar

Le 20/04/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


According to Epictetus, mental freedom and happiness can be achieved by distinguishing between, on the one hand, things that are upon our con- trol (our acts, opinions and desires), and, on the other hand, things that are not upon our control (our body, property, offices and reputation), and by wishing for nothing that is outside our control. This article proposes two accounts of Epictetus’s precept: the I account of Epictetus’s precept requires indifference between outcomes differing only on circumstances, whereas the IB account requires indifference between outcomes involving the best replies to circumstances. We study the implications of these precepts on the preference relation and on the existence of Epictetusian rationality. The I account implies that the preference relation satisfies in- dependence of circumstances, whereas the IB account implies robustness to dominated alternatives. Unlike the IB account, the I account rules out (counter)adaptive preferences. Finally, when examining game-theoretical implications of Epictetusian rationality, we show that the two accounts of Epictetus’s precept exclude the existence of prisoner’s dilemmas.

PONTHIERE Gregory () Epictetusian Rationality

Behavior seminar

Le 13/04/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Going to court is always a tough decision to take. Indeed, you never know your probabilities to win, the time it will require and how much you will win or spend. Predictive justice is an AI tool that uses big data to bring information to lawyers and litigants about their probabilities to win and the potential rewards. This objective information should reduce the difficulty to make the ambiguous decision of going to court or not. Seeing judicial decision as a decision under natural ambiguity, we test the effect of three types of information on attitudes:partial ambiguity, risk and similarity. In a complement study, we elicit the willingness to pay to access this information. Results first show that attitudes toward ambiguity differ under natural and artificial settings. Then, under both settings, the information provided to a complete ambiguous situation changes the behaviors but with no major difference according to the content of information.The valuations of the types of information are relatively in line with these attitudes.

MASSONI Sébastien () Judicial Decision under Ambiguity and Predictive Justice

V. Teixeira

Behavior seminar

Le 06/04/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We investigate impacts of antidepressant treatment on academic achievement among Danish children who are referred to a child psychiatrist. Leveraging quasi-random assignment of patients to psychiatrists with different prescribing tendencies, we find that treatment significantly increases test scores, especially in math and among girls. Treatment effects are larger among children of less educated mothers who, in general, are less likely to be treated, indicating negative selection on observables. However, a marginal treatment effects approach reveals that the effects are larger among children most likely to benefit from treatment, suggesting positive selection on unobservables. The marginal treatment effects are almost always positive, suggestive of under-prescribing of SSRI for children. This conjecture is confirmed by policy experiments indicating that expanding treatment among children of less educated mothers yields large test score increases, but restricting access to treatment among children of highly educated mothers harms their performance

DAYSAL Meltem () Antidepressant Use and School Performance: Evidence from Danish Administrative Data

Behavior seminar

Le 23/03/2023 de 16:30:00 à 17:30:00

Bâtiment A, Rez de chaussée, Salle 4


Analyses of self-reported-well-being (SWB) survey data may be confounded if people use response scales differently. We use calibration questions, designed to have the same objective answer across respondents, to measure general (i.e., common across questions) scale-use heterogeneity. In a sample of ~3,350 MTurkers, we find substantial such heterogeneity that is correlated with demographics. We develop a theoretical framework and econometric approaches to adjust for this heterogeneity. A key simplifying assumption, motivated by our evidence, is that different respondents’ scale-use functions are linear transformations of each other. We apply our new estimators in several standard SWB applications. Adjusting for general-scale-use heterogeneity changes results in some cases, and our framework predicts when adjustment will matter.

BENJAMIN Daniel () Adjusting for Scale-Use Heterogeneity in Self-Reported Well-Being

Behavior seminar

Le 09/03/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Statistical analysis of life satisfaction data relies on three fundamental assumptions about the properties of such data: (1) Individuals perceive the scale as being linear (“cardinality”). (2) All individuals use the scale in the same way (“interpersonal comparability”). (3) The way that individuals use the scale does not change over time (“intertemporal comparability”). Sceptics of life satisfaction scales question the credibility of these assumptions. Advocates respond with evidence of psychometrics validity. For example, that life satisfaction declined markedly during the outbreak of the COVID 19 pandemic. Yet while this attests to the useability of life satisfaction scale data, it does not clarify the precision of the associated metrics and thus the extent of that useability. Some applications, such as cost-effectiveness analysis using life satisfaction data, may be compromised by especially severe violations of the three assumptions above. We need to better understand these issues of degrees if we are to responsibly apply life satisfaction scale data in policy. Assessing the credibility of these assumptions requires understanding the life satisfaction ‘reporting function’. This is an affective, cognitive, and linguistic process that subjectively assesses life satisfaction and then maps that assessment to a response category on a life satisfaction scale question. This study explores the reporting function using cognitive interviews: essentially asking respondents to ‘think out loud’ while answering life satisfaction scale and follow up questions. The resulting qualitative data is analysed to explicate the reporting function and evaluate the three assumptions above.

FABIAN Mark () What Do Responses to Life Satisfaction Scale Questions Mean? Evidence from Cognitive Interviewing

Behavior seminar

Le 16/02/2023 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Policy makers put great emphasis on the role of information about carbon emissions in achieving sustainable decisions by consumers. We conduct two studies to understand the optimal targeting of such information and its effects. First, we conduct an incentivized and representative survey among US consumers (N = 1, 022) to investigate awareness of climate impact and willingness to mitigate it. We find a large variation in the perceptions of the carbon emissions of different consumption behaviors, with an overall tendency to underestimate these emissions. We also find a positive but highly concave willingness to mitigate climate impact. We combine elicited misperceptions and willingness to mitigate in a structural model that delivers sharp predictions about where to best target information campaigns. In an experiment with actual consumption decisions (N = 2, 081), we then test for the effect of CO2 information on the demand for beef, a product predicted to be a productive target for information. Correcting misperceptions has no effect on the demand for beef, both in absolute terms and compared to a predictably less productive target of information, i.e. the demand for poultry. Our dataset allows us to hone in on the underlying reason for this null effect

IMAI Taisuke () Correcting Consumer Misperceptions about CO2 Emissions

Davide Pace, Peter Schwardmann, Joël van der Weele

Texte intégral

Behavior seminar

Le 15/12/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


During the COVID-19 pandemic, UK residents could choose to opt into the vaccination program, but not which vaccine they received. Using this natural experiment, we investigate individuals' beliefs about COVID-19 vaccines before and after vaccination. Although the majority of people were expecting a type of vaccine (Pfizer), about 20% of them were assigned to another one (Moderna). We study how people who received Moderna (treatment group) updated their beliefs differently from those who received Pfizer (control group). We document three results. First, individuals overly optimistically updated their beliefs about the safety and effectiveness of the vaccine they received, so that the average beliefs of the control and treatment groups diverge. Second, we show that this divergence is driven by those who were particularly skeptical about a vaccine but then were assigned to it, and those who were particularly keen on a vaccine but then did not receive it. Third, we find that people adapt not only their beliefs, but also their preferences: the desirability of a vaccine adjusted according to which vaccine people received. These findings shed light on the predictable nature of individuals' beliefs in the field, and are in line with a dynamic of wishful beliefs updating. They suggest that wishful thinking can cause beliefs divergence, even when similar individuals are exposed to similar information environments, hold similar priors, and have no room for ex-post rationalization of previous decisions.

Prati Alberto () *Why do beliefs diverge? Evidence from a natural experiment on COVID-19 vaccines

Behavior seminar

Le 01/12/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We discuss the behavioral game theory of tennis, with particular focus on a literature testing minimax predictions. We look at some novel data for insights.

Nax Heinrich () *Tennis

Behavior seminar

Le 24/11/2022 de 14:30:00 à 15:30:00

Bâtiment A, Rez de chaussée, Salle 4


This study explores the effect on mental health and life satisfaction of working in an automatable job. We utilise an Australian panel dataset (HILDA), and estimate models that include individual fixed effects, to estimate the association between automatable work and proxies of wellbeing. Overall, we find evidence that automatable work has a small, detrimental impact on the mental health and life satisfaction of workers within some industries, particularly those with higher levels of job automation risk, such as manufacturing. Furthermore, we find no strong trends to suggest that any particular demographic group is disproportionately impacted across industries. These findings are robust to a variety of specifications. We also find evidence of adaptation to these effects after one-year tenure on the job, indicating a limited role for firm policy.

STRINGER Eliza-Jane (London School of Economics) *People Versus Machines: The Impact of Being in an Automatable Job on Australian Worker's Mental Health and Life Satisfaction

Behavior seminar

Le 10/11/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We document new facts on the distribution of male fertility and its relationship with men's labor market outcomes. Using Norwegian registry data, we uncover a "retreat from fertility": the gap in male childlessness between low and high earners has widened by almost 20 percentage points over the last thirty years, resulting in a remarkable compression of the fertility distribution. Using firm bankruptcies, we show that men experiencing negative labor market shocks are persistently less likely to become fathers and be partnered for at least 15 years after the event. We conclude by documenting that men's fertility penalty to job loss has increased markedly over the last three decades. A challenging labor market fails to shield low income workers with serious implications for family formation.

Walther Selma (London School of Economics) *Male Fertility: Facts, Distribution and Drivers of Inequality co-authored with Bernt Bratsberg and Andreas Kotsadam

Behavior seminar

Le 20/10/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We experimentally investigate the impact of information disclosure on managing collective harms that are caused jointly by a group of liable agents. Subjects interact in a public bad setting and must choose ex ante how much to contribute in order to reduce the probability of causing a common damage. If a damage occurs, subjects bear a part of the loss according to the liability-sharing rule in force. We consider two existing rules: a per capita rule and a proportional rule. Our aim is to analyze the relative impact of information disclosure under each rule. We show that information disclosure increases contributions only under a per capita rule. This result challenges the classical results regarding the positive effects of information disclosure, since we show that this impact may depend upon the legal context. We also show that while a proportional rule leads to higher contributions than a per capita one, the positive effect of disclosure on a per capita rule makes it as efficient as a proportional rule without information disclosure.

LEFEBVRE Mathieu (London School of Economics) *Information Disclosure under liability

Behavior seminar

Le 13/10/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


This study reports the results of the first artefactual field experiment designed to measure the prevalence of aversion toward social risks for health and wealth in a large and demographically representative sample. We identify social risk preferences for wealth and health for losses and gains, and decompose these attitudes into four different dimensions: viz., an individual risk dimension, a collective risk dimension, an ex-post inequality dimension and an ex-ante inequality dimension. The results of a non-parametric analysis suggest that aversion to risk and inequality is the modal preference for outcomes in health and wealth in the domain of gains and losses. Furthermore, we find stronger ex-ante than ex-post inequality aversion for health gains, and the opposite for health losses. These differences are of the same sign but not significant for wealth gains and losses. A parametric decomposition reveals that attitudes toward individual risk are moderated by the domain of the outcomes; respondents are risk neutral for individual members of society in the loss domain for health and wealth. These results highlight the importance of considering different components of social risk preferences when managing social risk.

ATTEMA Arthur (London School of Economics) *Decomposing Social Risk Preferences for Health and Wealth

Behavior seminar

Le 29/09/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


This talk reports on recent work showing how to use response time data and choice frequencies to elicit preferences nonparametrically when choices are stochastic. It discusses theoretical results and their empirical implementation. The talk also discusses two recent applications. The first shows how we can separate non-transitive preferences from noise-generated transitivity violations, a problem so far unresolved. The second shows how to apply the techniques to improve the analysis of survey data and determine whether a group (defined, e.g., by gender, age cohort, socioeconomic status, etc.) prefers an option over its alternative more than another group, even when individual choices are noisy, and without any assumptions on the noise

ALOS-FERRER Carlos (London School of Economics) Learning Preferences from Response Times

Behavior seminar

Le 15/09/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


*Identity motives—strivings to view oneself in certain ways—affect people’s willingness to protect their health, buy consumer products, vote for politicians, or die for their country. Yet, research into identity motives has focused mainly on a small fraction of humanity who inhabit “Western” societies. I will present findings from two large studies in which we measured identity motives among >12,000 members of cultural groups spanning 35 nations on all inhabited continents. Across highly diverse cultural, socioeconomic, political, and environmental contexts, people structured their identities to view themselves as: accepted by others (belonging), positively valued (positive self-regard), distinguished from others (distinctiveness), persisting through time (continuity), competent and capable (efficacy), and having a life that matters (meaning). These common motives underlie the superficially divergent expressions of identity observed across human cultures.

BECKER Maja (London School of Economics) *Common motives underlie identity construction across highly diverse cultural contexts

Behavior seminar

Le 08/09/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We study the effects of providing quantitative and qualitative information about top earners on people's attitudes and tax preferences towards the richest 1 percent. We conduct an online experiment with 2000 French respondents assigned randomly to quantitative information on the income levels of the top 1 percent and their respective income sources (capital vs. labor), or to quantitative information plus qualitative egalitarian descriptions on the consequences of inequality and high capital income shares. We find that: (i) respondents overestimate the income of the richest 1 percent and want them to pay a higher income tax rate than the current one. (ii) Quantitative information shifts attitudes about top earners towards the unfavorable spectrum, but it has no effect on the desired income tax rate at the top. The effect on attitudes comes mainly from information on income sources. (iii) Quantitative plus qualitative information leads respondents to choose a higher income tax rate for the richest 1 percent. The effect comes mainly from the qualitative information on the consequences of inequality.

Barrera Oscar (London School of Economics) Quantitative and Qualitative Economics on Attitudes and Tax Preferences Towards the Top 1 Percent. (Avec Emmanuel Chávez)

Behavior seminar

Le 30/06/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Many countries suffer from severe droughts, and it is predicted that they'll become more frequent with forecasted changes in climatic conditions. In agricultural regions, drought is often anecdotally associated with considerable economic stress and worsening mental health. However, robust evidence on the overall microeconomic effects of drought, particularly from high-income countries, is limited. In this study, we use panel data from large administrative databases to estimate the economic and mental health impacts of Australian droughts on farmers, other agricultural workers, and their broader communities. Results indicate that droughts in agricultural areas have large negative effects on the incomes of farmers, moderate negative effects on other agricultural workers, and small approximately zero effects on non-agricultural workers. Crucially, the negative economic shocks on agricultural workers are psychologically harmful, with estimates indicating large increases in suicides. In contrast, we find no changes in GP and allied health service use, and only small increases in the use of antidepressant and anxiolytic medication. The large increase in suicides and the small changes in healthcare use suggest that mentally unwell agricultural workers are not receiving the treatment they need.

JOHNSTON David (London School of Economics) Impacts of Droughts on Economic and Mental Wellbeing

Behavior seminar

Le 23/06/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Initial success is often thought to result in a run of future successes; a phenomenon coined the Winner Effect. One reason posited for this effect is the psychological and motivational impact of success on later performance. I will present three studies investigating this possibility and show that a positive effect of success on later performance can be identified in both field and laboratory data. Using experimental data and applied theory, I will discuss the likely mechanisms behind this effect. The articles linked to the presentation: Article 1 (field data): https://academic.oup.com/ej/article/129/624/3107/5536246 Article 2 (lab experiment): https://onlinelibrary.wiley.com/doi/full/10.3982/QE1679 Article 3 (theory): in preparation

PAGE Lionel (London School of Economics) How success breeds success

Behavior seminar

Le 09/06/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


The available evidence from anthropology, economics, and psychology suggests that sensitivity to the emotions of shame and guilt varies across cultures. So does (over)confidence in ability and skills. Is there a connection between these observations? We address this question theoretically and empirically. We find significant evidence, consistent with our model, of a negative relationship between the cultural importance of shame relative to guilt and individual confidence. The relationship holds across countries, and for US immigrants relative to the culture of origin countries.

DESSI Roberta (London School of Economics) Shame, Guilt, and Motivated Self-Confidence

Junjie Ren (National University of Singapore) and Xiaojian Zhao (Monash University)

Behavior seminar

Le 02/06/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We study how the presence of women in top academic positions affects future hiring decisions and the characteristics of PhD students. We exploit a natural experiment that generated exogenous variation across Spanish university departments in the probability that male and female faculty are promoted to Associate and Full Professor positions. From 2002 to 2006, the composition of review committees was randomized, and applicants were significantly more likely to be promoted if, by luck of the draw, they were assessed by a committee including a colleague, a co-author, or a former advisor. Our analysis shows that the gender of these “marginal” successful applicants has long-term consequences on the gender composition of the department, the number of male and female PhD students, and their research topics.

ZINOVYEVA Natalia (London School of Economics) Women in top academic positions: Is there a trickle-down effect?

Manuel Bagues and Giulia Vattuone

Behavior seminar

Le 19/05/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Understanding attitudes towards inequality among the “working rich” matters for any policy aimed at increasing the level of redistribution in society. We investigate this question using a unique sample of nearly 1,000 graduates from a highly ranked MBA program and a representative sample of Americans. We first show that high-earning MBAs are far more likely to know their rank in the income distribution. We then explore whether and how comparisons with peers or others (i.e. reference groups) shape their preferences for redistribution. Asking them to rank within their family, colleagues or classmates leads to an average 18% drop in the income share allocated to the richest 1% but has no discernible effect on their taxation preferences. We discuss the respective contribution of the comparative and normative functions of reference groups as potential mechanisms.

BELLET Clément (London School of Economics) Perceived Inequality and Preferences for Redistribution Among High Earners: Do Reference Groups Matter?

Behavior seminar

Le 12/05/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Ultra-processed food consumption has been recently associated with the prevalence of obesity and non-communicable chronic diseases such as some types of cancers, diabetes and heart diseases. However, ultra-processed food consumption is not so-well documented across individuals or across food product categories. In this paper, we construct a unique longitudinal dataset, between 2002 and 2015, of at-home food consumption of French households to which we merge information regarding nutritional values and degree of food processing. Over the period, we find increasing consumption patterns of ultra-processed products, yet still relatively low when compared to Anglo Saxon countries levels of consumption. French lower consumption levels are most likely a result of higher prices for ultra-processed products relatively to less-processed products, the reverse of Anglo saxon prices. We then develop a panel data econometric analysis to identify the determinants of ultra-processed food consumption exploiting the within- and between-household variations. We find that socio-economic demographics would explain more the between household variation that the within-household variation. In fact differences between households are due to obesity status, youth, poverty, rural, north and east location which are all associated with higher levels of ultra-processed consumption. Nonetheless, becoming old, obese, or moving does not have a strong impact in UP consumption. Some life events with impact in time constraint, such as having young children, return to labour activity, or becoming single, are associated higher UP consumption. Yet, the main driver for within-household UP consumption reduction are prices. These results suggest that policies aiming at decreasing UP consumption should , on the one hand, focus on specific population groups and, on the other hand, use price reductions as lever for within household change of consumption patterns.

Bonnet celine (Toulouse School of Economics, INRAE) Between- and within-household consumption of ultra-processed food in France

Behavior seminar

Le 21/04/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Identity motives—strivings to view oneself in certain ways—affect people’s willingness to protect their health, buy consumer products, vote for politicians, or die for their country. Yet, research into identity motives has focused mainly on a small fraction of humanity who inhabit “Western” societies. I will present findings from two large studies in which we measured identity motives among >12,000 members of cultural groups spanning 35 nations on all inhabited continents. Across highly diverse cultural, socioeconomic, political, and environmental contexts, people structured their identities to view themselves as: accepted by others (belonging), positively valued (positive self-regard), distinguished from others (distinctiveness), persisting through time (continuity), competent and capable (efficacy), and having a life that matters (meaning). These common motives underlie the superficially divergent expressions of identity observed across human cultures.

BECKER Maja (Toulouse School of Economics, INRAE) CANCELLED Common motives underlie identity construction across highly diverse cultural contexts

Behavior seminar

Le 14/04/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Social interactions determine many economic behaviors, but information on social ties does not exist in most publicly available and widely used datasets. We present results on the identification of social networks from observational panel data that contains no information on social ties between agents. In the context of a canonical social interactions model, we provide sufficient conditions under which the social interactions matrix, endogenous and exogenous social effect parameters are all globally identified. While this result is relevant across different estimation strategies, we then describe how high-dimensional estimation techniques can be used to estimate the interactions model based on the Adaptive Elastic Net GMM method. We employ the method to study tax competition across US states. We find the identified social interactions matrix implies tax competition differs markedly from the common assumption of competition between geographically neighboring states, providing further insights for the long-standing debate on the relative roles of factor mobility and yardstick competition in driving tax setting behavior across states. Most broadly, our identfication and application show the analysis of social interactions can be extended to economic realms where no network data exists.

DE PAULA Aureo (Toulouse School of Economics, INRAE) Identifying Network Ties from Panel Data: Theory and an Application to Tax Competition

Behavior seminar

Le 31/03/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Football is the national sport of most of the planet. This paper examines how happy the outcomes of football matches make us. We calibrate these results relative to other activities and estimate the dynamic effects these exogenous events have on our utility over time. We find that football – on average – makes us unhappier – so why would we go through the pain of following a football team? This behavioural choice paradox occupies much of the paper, so we investigate why we go on following our teams, even though matches make us more unhappy on average. We examine how much our story changes if we examine the dynamic effects of football matches over time in different hours before and after the game and the extent to which our happiness is influenced by what we would rationally expect the result to be beforehand – as based on the betting odds.

DOLTON Peter (Toulouse School of Economics, INRAE) Is Football a Matter of Life and Death – Or is it more Important than that?

George MacKerron (University of Sussex)

Behavior seminar

Le 24/03/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


The outcomes from strategic decision-making (such as market entry or technology adoption) depend on structural features of situations and types of players involved. Even in identical situations, players differ in their perceptions of situations, goals, and strategic sophistication. Informed by behavioral and experimental economics, we present and integrate two classes of heuristic player types under strategic uncertainty in new situations, exemplified within the simplest class of games, two players–two actions (2x2) games, e.g., Prisoner's dilemma or Entry games. One class anticipates others' behavior and (iteratedly) best-replies to beliefs (called iterated reasoning or level-k heuristic), while the other is guided by goals, e.g., equality or social optimum, ignoring procedural details, such as others'reasoning (called efficiency heuristic). To understand the implications of game structure and player types, we develop a behavioral system of 2x2 games. Due to the fundamental differences of the two classes, the large set of 2x2 games collapses to four distinct game classes for efficiency types and five, albeit different ones, for iterated reasoning, and to 14 in a joint system based on (behavioral) gametheoretic features. Thus, advanced knowledge of players' strategic capabilities or goals considerably simplifies the strategic analysis by inducing a categorization of games. Furthermore, we predict differences in the heterogeneity of behavior and outcomes, depending on the presence of different player types and game structure.

NAGEL Rosemarie (Toulouse School of Economics, INRAE) Integrating Level-k, Efficiency Heuristics, and a Behavioral Taxonomy for 2x2 Games

Behavior seminar

Le 17/03/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


This paper develops a willingness-to-pay elicitation method that combines the strengths of choice experiments and experimental auctions. While these tools are standard for estimating consumer values of goods and their attributes, they tend to produce dissimilar results since their methodological discrepancies may trigger distinct types of behaviour. The Global & Analytical willingness-to-pay elicitation method (G&A method) generates for each individual two measures of willingness-to-pay, one derived from an initial holistic approach and the other from a reflective assessment about the relative importance of the attributes composing the good. We applied this method to the wine market, and in particular to the question of how consumers value the Corporate Social Responsibility (CSR) attribute. We show that while analytical assessment leads to reduced willingness-to-pay for wine, the marginal willingness-to-pay for CSR increases relatively to the other attributes.

MULLER Laurent (Toulouse School of Economics, INRAE) A Global & Analytical Willingness-to-Pay Elicitation Method. The case of the Corporate Social Responsibility attribute for wine

Behavior seminar

Le 10/03/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


It is widely hypothesized that anxiety about adverse future outcomes motivates people to adopt comforting beliefs or to engage in wishful thinking. However, there is little direct causal evidence for this effect. In a first experiment, participants perform a visual pattern recognition task where some patterns may result in the delivery of an electric shock, a proven way of inducing anxiety. Participants engage in significant wishful thinking: they are less likely to correctly identify patterns that they know may lead to a shock. A second and third experiment establish that participants also engage in wishful thinking in anticipation of monetary losses and that the phenomenon is robust to another perceptual task, which draws on different cognitive processes. Across our three experiments, greater ambiguity of the visual evidence is associated with more wishful thinking. Our within-subject design allows us to detect wishful thinking at the individual level. We find that wishful thinking is heterogeneous across and stable within individuals.

VAN DE WEELE Joël (Toulouse School of Economics, INRAE) Anticipatory Anxiety and Wishful Thinking

with Jan Engelmann, Mael Lebreton, Nahuel Salem-Garcia and Peter Schwardmann

Behavior seminar

Le 24/02/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Gendered norms have major implications for women’s labor market outcomes. Notably, a recent literature finds that child-rearing norms and the prescription that the husband should be the main breadwinner lead to behavioral changes affecting women’s labor supply. Motherhood reduces participation and hours of market work, while women who earn more than their husbands have been shown to react in ways that reverse that gap. In this paper we use panel data for the US to examine to what extent these two different norms interact. We start by asking whether child-rearing norms affect women who are the main breadwinner and those who are not in the same way, and then turn to how mothers and childless women react when breaking the male-as-the-breadwinner norm. Our results show that the breadwinner norm has an effect only on mothers, suggesting that the salience of gender norms may depend on the household’s context. Concerning child-rearing, we find that although the labor supply of women who earn more than their husbands initially responds to motherhood less than that of secondary earners, the two groups converge after 10 years. Moreover, women in the former category exhibit a disproportionately large increases in the share of housework they perform after becoming mothers. These results indicate that norms still prevail over considerations of comparative advantage, and that the presence of children pushes women to seek to compensate breaking a norm by adhering to another one

Garcia-Penalosa Cecilia (AMU - AMSE MARSEILLES) Interactions amongst gender norms: Evidence from US couples

Behavior seminar

Le 17/02/2022 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We explore decision-making under uncertainty using a framework that decom- poses uncertainty into three distinct layers: (1) risk, which entails inherent ran- domness within a given probability model; (2) model uncertainty, which entails subjective uncertainty about the probability model to be used; and (3) model mis- specification, which entails uncertainty about the presence of the correct probability model among the set of models considered. Using a new experimental design, we measure individual attitudes towards these three layers of uncertainty and examine the role of each of them in characterizing attitudes towards ambiguity. In addition to providing new insights into the underlying processes behind ambiguity aversion, our study provides the first empirical evidence of the role of model misspecification in decision-making under uncertainty.

BERGER Loïc (CNRS) Three layers of uncertainty and the role of model misspecification

Behavior seminar

Le 16/12/2021 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4

, () *

Behavior seminar

Le 02/12/2021 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Transport has significant externalities including carbon emissions and air pollution. Public health research has identified additional social gains from active travel, due to health benefits of physical exercise. Per mile, these benefits greatly exceed the external costs from car use. We introduce active travel into an optimal fuel taxation model and analytically characterise the optimal second-best fuel tax. We find that accounting for active travel benefits increases the optimal fuel tax by 49% in the US and 36% in the UK. Fuel taxes should be implemented jointly with other policies aimed at increasing the uptake of active travel.

MATTAUCH Linus () Healthy Climate, Healthy Bodies: Optimal Fuel Taxation and Physical Activit

Behavior seminar

Le 25/11/2021 de 10:00:00 à 11:00:00

Bâtiment A, Rez de chaussée, Salle 4


We measure the effects of a network of heroes in legitimizing and diffusing extreme political behaviors. Exploiting newly-declassifed intelligence files, novel voting data and regimen- tal histories, we show the home municipalities of French line regiments arbitrarily rotated through Philippe Pétain's command during the heroic WWI battle of Verdun, though similar before WWI, increasingly espouse Pétain's authoritarian political views thereafter, raising 7% more active Nazi collaborators per capita during the Pétain-led Vichy regime (1940-44). The effects are similar across joining Fascist parties, German forces, paramilitaries hunting Jews and the Resistance, and collaborating economically.

Grosjean Pauline () Heroes and Villains: The Effects of Heroism on Autocratic Values and Nazi Collaboration in France


Texte intégral

Behavior seminar

Le 18/11/2021 de 16:00:00 à 17:00:00

Bâtiment A, Rez de chaussée, Salle 4

KIMBALL Miles () Adjusting for Scale-Use Heterogeneity in Self-Reported Well-being

Daniel J. Benjamin (UCLA) Kristen Cooper (Gordon) Ori Heffetz (Cornell & Hebrew U) Jiannan Zhou (Colorado)

Behavior seminar

Le 21/10/2021 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Job insecurity has consequences outside of the labour market. Using the 2012 Fornero reform as a natural experiment, a difference-in-differences framework based on a firm-size discontinuity and individual data coming from the Italian Survey on Household Income and Wealth, our results suggest that greater job insecurity reduces consumption and increases savings. We also show that the changes in consumption and savings are a function of the family structure and of the rank in the household income distribution. Last, greater job insecurity reduces all types of consumption except food expenditures and the extra-savings it caused are either invested in safe assets or kept on savings account.

Lepinteur Anthony () Job insecurity, savings and consumption: an Italian experiment

Behavior seminar

Le 14/10/2021 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We examine the relationship between time delay and generosity in a laboratory experi- ment using modified dictator games. In a between-subject design we vary the time of payout for the dictator, the receiver or both. In a within-subject design, we vary the endowment of the dictator as well as the price of giving. We derive and test the predictions of several models of social preferences and test if the giving decisions are representable by a utility function.

RIEDL Arno () Intertemporal Social Preferences

Behavior seminar

Le 23/09/2021 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


In our model, players contribute to two local public goods for which they have different tastes and sponsor costly links to free ride on others’ contributions. We characterize the Nash equilibria of the game. In these equilibria, either there are two large contributors who might free ride on each other, or several contributors whose neighborhood of free riders does not overlap. As linking costs increase, agents seek connections to others, whose type is closer to their own, i.e., society becomes more homophilous. Polarization increases if links are intrinsically cheap and decreases otherwise. Moreover, if moderate agents emerge as large contributors, welfare increases, while polarization decreases in societies with low extremism.

MERLINO Luca Paolo () SEMINAIRE ANNULE - Homophily and Polarization in Endogenous Networks

Behavior seminar

Le 07/09/2021 de 15:00:00 à 16:00:00

Bâtiment A, Rez de chaussée, Salle 4


Whether or not to vaccinate one's child is a decision that a parent may approach in several ways. The vaccination game, in which parents must choose whether or not to vaccinate a child against a disease, is one with positive alternalities (herd immunity). In some societies, not vaccinating is an increasingly prevalent behavior, due to deleterious side effects that parents believe may accompany vaccination. The standard game-theoretic approach is to assume that parents make decisions according to the Nash behavioral protocol, which is individualistic and non-cooperative. Because of the positive externality that each child’s vaccination creates for others, the Nash equilibrium suffers from the free-rider problem. However, in more solidaristic societies, parents may behave cooperatively -- they may optimize according to the Kantian protocol, in which the equilibrium is efficient. We test, on a sample of six countries, whether vaccination behavior conforms better to the individualistic or cooperative protocol. To do so, we conduct surveys of parents in these countries, to ascertain their beliefs about the subjective probability and severity of deleterious side effects of vaccination. Our analysis shows that in all the countries of our sample, the Kant model dominates the Nash model. We conjecture that, due to the free-rider problem inherent in the Nash equilibrium, social norms have developed, quite generally, inducing parents to vaccinate with higher probability than they would in the non-cooperative solution.

ROEMER John () Game-theoretic analyses of childhood vaccination behavior: Nash vs. Kant

Philippe De Donder, Toulouse School of Economics; Humberto Llavador, Universitat Pompeu Fabra; Stefan Penczynski, University of East Anglia; John E. Roemer, Yale University; Roberto Velez, Centro Estudios Espinosa Yglesias, Mexico City. Research Assistants: Collin Schumock, Austin Jang

Behavior seminar

Le 17/06/2021 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


The available evidence from anthropology, psychology, and economics shows that sensitivities to shame and guilt vary across cultures. So does the tendency to exhibit overconfidence. We develop a theoretical framework to analyze this variation, interpreted as reliance on different psychological incentive mechanisms. We find that overconfidence and guilt are typically complementary, whereas shame can be a substitute for overconfidence. Individual-level data on US immigrants appears consistent with the model's predictions

DESSI Roberta () Motivated Beliefs, Shame and Guilt across Cultures

Behavior seminar

Le 06/05/2021 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


On average, firstborns complete more education than their laterborn siblings. We study whether this effect is amplified by genetic endowments. Our family-fixed effects approach allows us to exploit exogenous variation in birth order and genetic endowments among 15,019 siblings in the UK Biobank. We find that those with higher genetic endowments benefit disproportionally more from being firstborn compared to those with lower genetic endowments, providing a clean example of how nature and nurture interact in producing skills. Moreover, since parental investments are a dominant channel driving birth order effects, our results are consistent with dynamic complementarity in skill formation.

von HINKE Stéphanie (University of Bristol) Dynamic complementarity in skill production: Evidence from genetic endowments and birth order

Dilnoza Muslimova, Hans van Kippersluis, Niels Rietveld, Fleur Meddens

Texte intégral

Behavior seminar

Le 15/04/2021 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Using the 2016 merger of French regions as a natural experiment, this paper adopts a difference-in-difference identification strategy to recover its causal impact on individual subjective well-being. No depressing effect is found despite increased centralization; life satisfaction has even increased in regions that were absorbed from economic and political viewpoints. The empirical evidence also suggests that local economic performance improved in the concerned regions, which includes a faster decline in the unemployment rate. In this setting, economic gains have likely outweighed cultural attachment to administrative regions.

WILNER Lionel (INSEE CREST) How do citizens perceive centralization reforms? Evidence from the merger of French regions

Behavior seminar

Le 01/04/2021 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We study whether bonus payments for information provision can improve the information flow between physicians. We consider the situation of a primary care physician (PCP) deciding on the provision of information of varying qualities to a specialist while referring a patient. Our theoretical model, which includes altruism and the aversion to lose profit relative to a reference profit, predicts that bonus payments increase the provision of both high- and low-quality information. Running a controlled laboratory experiment we find support for this prediction. Moreover, we observe that in case that the beneficiary of information provision receives a higher payoff than the PCP, PCPs more often pass on high-quality information when the beneficiary is a patient instead of a specialist. In case that the beneficiary receives a lower payoff than the PCP, the type of the beneficiary (specialist or patient) does not affect the provision of high-quality information. As a consequence, specialists more often receive high-quality information when they earn smaller instead of larger profits than PCPs, while there is no such difference for patients. As such our observations suggest that the level of high-quality information transferred by PCPs depends on an interaction of who benefits from information provision and the payoff relation between beneficiaries and PCPs.

KIFMANN Mathias (INSEE CREST) Rewards for information provision in patient referrals: a theoretical model and an experimental test

Behavior seminar

Le 18/03/2021 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Human groups can markedly differ in fairness and cooperation norms, and these differences can create intergroup misunderstandings and conflict. At the same time, humans also trade and travel across cultural divides, suggesting that they can learn and adapt to new culture-specific conventions and rules of engagement. While such adaptions avoid intergroup conflict and benefit intergroup exchange, how humans learn group-specific rules that are often implicit and distinct from already learned values and norms remains poorly understood. Here we examine this fundamental learning process underlying social rule acquisition. We created three populations with different yet unobservable rules of engagement and varied whether or not decisions affected interaction partner outcomes. Participants made bargaining offers to responders from these different populations and could observe whether their offer was accepted or rejected. Participants quickly adapted to group-specific rules in learning environments without social consequences, but were overly generous and ended up misrepresenting what would be acceptable when decisions affected their partner’s outcomes. We propose a computational model, combining Bayesian principles and social preferences, that mechanistically explains how generosity leads to biased sampling, impeded learning, and false beliefs about what offers are deemed acceptable. Using functional neuroimaging, we mapped key computational variables in two major brain networks, previously associated with value-based and social decision-making. Results suggest that generosity, related to brain regions associated with decision-conflict and perspective-taking, can induce self-fulfilling beliefs in pro-sociality norms that may help to increase cooperation and reduce conflict between distinct groups but also create inaccurate stereotypes and economic inefficiencies.

LEBRETON Mael (INSEE CREST) *Generosity Biases the Learning of Cultural Conventions

Behavior seminar

Le 04/03/2021 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


The milk addiction paradox refers to an empirical finding in which non-addictive commodities such as milk appear to be addictive. This paradoxical result seems more likely when consumption is persistent and with aggregate data. Using both simulated and real data, we show that the milk addiction paradox disappears when estimating the data using an AR(1) linear specification that describes the saddle-path solution of the rational addiction model, instead of the canonical AR(2) model. The AR(1) specification is able to correctly discriminate between rational addiction and simple persistence in the data, to test for the main features of rational addiction, and to produce unbiased estimates of the short and long-run elasticity of demand. These results hold both with individual and aggregated data, and they imply that the AR(1) model is a better empirical alternative for testing rational addiction than the canonical AR(2) model.

DRAGONE Davide (INSEE CREST) Solving the milk addiction puzzle

Co-author : Davide Raggi

Texte intégral

Behavior seminar

Le 17/12/2020 de 15:00:00 à 16:00:00

Bâtiment A, Rez de chaussée, Salle 4


The magnitude and nature of the COVID-19 pandemic prevents public health policies to rely on coercive enforcement. Practicing social distancing, wearing masks and staying at home becomes voluntary and conditional on the behavior of others. We present the results of a large scale survey experiment run in nine countries with representative samples of the population (by age and gender) and find that both empirical and normative expectations play a vast and significant role in compliance, beyond the effect of any other individual or group characteristic. In our survey experiment, when empirical and normative expectations of individuals are high, compliance goes up by 55% (relative to the low expectations condition). Similar results are obtained when we look at self-reported compliance among those with high expectations (37% higher). Our results are robust to different specifications and controls, and driven by an asymmetric interaction with individuals’ trust in government and trust in science. Holding expectations high, the effect of putting trust in science is substantial and significant in our vignette experiment (22% increase in compliance), and even larger in self-reported compliance (76% and 127% increase before and after the lockdown). By contrast, putting trust in government generates modest effects. At the macro level, the country level of trust in science, and not in government, becomes a strong predictor of compliance.

FATAS Enrique () In Science We (Should) Trust


Texte intégral

Behavior seminar

Le 10/12/2020 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


This article provides experimental evidence of the impact of a school-based intervention aimed at developing students’ growth mindset and internal locus of control. Using a randomized experiment conducted in 97 disadvantaged middle schools in France, we follow 23,000 students over four years to assess the impact of the intervention. We find a 0.07 percent of a standard deviation increase in GPA, associated with a change in students’ mindset, improved behavior as reported by teachers and school registers, and higher educational and professional aspirations. Given the low intensity of the intervention—12 one-hour in-class sessions over four years—and its small cost—65 euros per student—the intervention is particularly cost-effective in comparison with most education programs. Importantly, while all students eventually benefit from the program, better-behaved and higher-performing students benefit more.

ALGAN Yann () The Role of Mindset in Education: A Large-Scale Field Experiment in Disadvantaged School

With Coralie Chevallier5, Elise Huillery, Adrien Bouguen, Axelle Charpentier

Behavior seminar

Le 03/12/2020 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4

DUCHENE Sébastien () *

Behavior seminar

Le 19/11/2020 de 14:30:00 à 15:30:00

Bâtiment A, Rez de chaussée, Salle 4


We establish a theoretical link between three phenomena at the core of behavioral economics: the Endowment Effect, Loss Aversion, and violations of Expected Utility as in the Certainty Effect. In our model, all jointly stem from one single force: uncertainty about the utility function to use and caution. Behaviorally, we show that our model is derived from positing a form of the certainty effect, that we show implies both Loss Aversion and the Endowment Effect. We analyze further implications of our model and demonstrate how it can organize existing empirical evidence of the Endowment Effect, and how it is conceptually and behaviorally distinct from other popular approaches, e.g., Cumulative Prospect Theory.

ORTOLOVA Pietro () Caution and Reference Effects

Simone Cerreia-Vioglio and David Dillenberger

Behavior seminar

Le 05/11/2020 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


For prospect theory (PT), a truth-telling mechanism for assessing personal beliefs, the quadratic scoring rule, is extended to obtain a new measurement tool for loss aversion. Under the plausible assumption that utility is linear for small scale gains and similar for losses, the loss aversion index (short: lambda ) measures the size of the concave kink of the gain-loss utility function at the reference point. We control for the bias captured by decision weights in PT, and quantify efficiently with only three quadratic scores. The additional benefits of the new tool are inherited from characteristic properties of scoring rules: they are relatively simple to implement, easy to understand by subjects, incentive compatible, and they dispense of chained measurements. In an experiment, we demonstrate these features of the new tool for risk and extend it to ambiguity; we find a median value of lambda= 1 at the aggregate level for both sources of uncertainty. Probability and event weighting, while less pronounced, accord with earlier findings from the literature; specifically, we find that these weights depend on the sign of the corresponding outcomes, supporting reference-dependent preferences. Event weighting persist at the individual level; after controlling for these weights, we find very few subjects who are loss averse or gain seeking.

L’HARIDON Olivier (Université Rennes 1) An Effective and Simple Tool for Measuring Loss Aversion

Craig S. Webb (Department of Economics, University of Manchester) and Horst Zank (Department of Economics, University of Manchester)

Behavior seminar

Le 01/10/2020 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


The use of sanctions is often invoked to bolster international agreements. For instance, it has been proposed to link climate change agreements – where costs of compliance normally exceed the benefits for a country - with trade sanctions – where a country normally benefits from having sanctions removed. We investigate whether sanctions may improve international cooperation experimentally. Participants are involved in a “collective risk social dilemma” (CRSD), where groups of individuals are threatened by the possibility of a large loss in their savings. The loss probability is proportional to the amount of “insurance” bought by individuals. Crucially, insurance bought by an individual reduces the risk of loss for all individuals. Self-interested individuals will free ride on others’ without buying any insurance. This would lead to a “tragedy of the commons” because the collective loss would occur with certainty. The cooperative solution is instead for the group to reduce the probability of loss to 30%. We involve German and Russian participants because, according to previous research, the two populations belong to different cultural areas with respect to the effectiveness of sanctions in sustaining cooperation. German participants can typically rely on sanctions to significantly increase group cooperation, whereas sanctions are typically detrimental in Russia, because of the high levels of anti-social punishment – i.e. punishment of cooperators – and of vengeance against punishers. This is the first experiment where the effectiveness of punishment is tested internationally. The use of populations from different cultural areas makes the prospect of cooperation particularly demanding. We implemented eight different experimental conditions, in which the CRSD is played within countries or between countries, with or without sanctions, either revealing or not revealing the nationality of the counterparts. We find that: (1) German groups achieve higher cooperation - that is, probability of loss avoidance - than Russian groups in within-country interactions. (2) Introducing sanctions significantly increases the probability of loss avoidance in Germany, but not in Russia. (3) Cooperative behavior by Russian participants significantly increases and converges to cooperative behavior by German participants in international interactions when sanctions are available. The convergence is slower when sanctions are not available. (4) Results are the same regardless of whether national identities are revealed or not revealed. (5) Expected payoffs are always higher in the non-sanction conditions than in sanction conditions. Arguably, this is the case because people are risk averse and attach high value to a certainty premium.

GRIMALDA Gianluca (Université Rennes 1) Sanctions and International Interactions Improve Cooperation to Prevent Collective Losses

Alexis Belianin, Heike Hennig-Schmidt, Till Requate, Marina V. Ryzhkova

Behavior seminar

Le 24/09/2020 de 14:00:00 à 15:00:00

Bâtiment A, Rez de chaussée, Salle 4


What notion of individual welfare do happiness questions measure? We clarify the theoretical assumptions about time horizon and weight put on others that underlie existing economic applications of self-reported well-being (SWB) data. In a survey, we ask respondents what time horizons and other individuals they had in mind when answering SWB survey questions, including commonly used measures of life satisfaction and happiness. Respondents put most weight on the immediate present and on themselves—but not enough to interpret SWB data as measuring common notions of flow utility and self-centered utility. We find differences across SWB questions and across sociodemographic groups. We discuss implications for existing and future work.

HEFFETZ Ori (Université Rennes 1) What Do Happiness Data Mean? Evidence from a Survey of Happiness Respondents

Daniel J. Benjamin, Jakina Debnam, Marc Fleurbaey, Miles Kimball

Behavior seminar

Le 10/09/2020 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


This paper is a cross-national study of physical pain over the business cycle. It is, to our knowledge, the first inquiry of its type. Using a sample of 146 countries, and fixed-effects methods, the paper shows that physical pain falls in a boom and rises in a downturn. The increases in pain are borne almost exclusively by women. Estimated effect-sizes are substantial. The paper’s conclusions have paradoxical aspects. The counter-cyclicality of physical pain, for example, is not what would be predicted by conventional analysis. During an economic expansion, people typically work harder and longer, and many kinds of accidents and injuries increase. Nor are the paper’s results due to unemployed citizens experiencing more pain (although they do). At the conceptual level the study’s findings are consistent with an important hypothesis proposed recently, using different kinds of evidence, by brain and behavioural-science researchers such as Katja Wiech and Irene Tracey (2009) and Eileen Chou and colleagues (2016). That hypothesis states that economic worry itself can create physical pain.

OSWALD Andrew (Université Rennes 1) Physical Pain, Gender, and the Business Cycle in 146 Nations

Lucia Macchia

Behavior seminar

Le 25/06/2020 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Efficient learning and decision-making in uncertain environments constitutes an important challenge for human and machine intelligence. Even for simplest, binary decisions, it requires inferring properties of one’s environment on the basis of imperfect evidence. This cognitive inference process has been studied across a wide range of vastly different paradigms, from categorizing ambiguous stimuli to choosing among risky reward sources. A common, defining feature of human decisions made under uncertainty is their large trial-to-trial variability. Prominent theories have assigned the source of this decision variability to the peripheries of cognitive inference: sensory errors during perceptual decisions, and stochastic choices during economic decisions. During this talk, I will show how the ubiquitous variability of human decisions made under uncertainty is driven by neither of these two sources, but by computation noise during cognitive inference itself. During probabilistic reasoning, the majority of suboptimal errors arise from random noise in otherwise near-optimal computations. During reward-guided learning, the same type of computation noise drives the majority of non-greedy decisions - which existing theories usually describe as the result of a trade-off between choosing a known option and exploring uncertain alternatives. In the brain, computation noise correlates with fluctuations of activity in prefrontal regions implicated in decision-making and cognitive control, and increases during pharmacological manipulation of the locus coeruleus-norepinephrine (LC-NE) system - a neuromodulatory pathway involved in arousal and exploratory behavior. Together, these findings indicate most behavioral variability, rather than reflecting stochastic choices or active exploration, is due to the limited precision of mental computations.

WYART Valentin (Université Rennes 1) ANNULE - Variable by choice or by chance? The surprisingly large contribution of computation noise to human decision-making under uncertainty


Texte intégral

Behavior seminar

Le 18/06/2020 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We study the use of excuses to justify socially stigmatized actions, such as opposing minority groups. Rationales to oppose minorities change some people’s private opinions, leading them to take anti-minority actions even if they are not prejudiced against minorities. When these rationales become common knowledge, prejudiced people who are not persuaded by the rationale can pool with unprejudiced people who are persuaded. This decreases the stigma associated with anti-minority expression, increasing public opposition to minority groups. We examine this mechanism through several large-scale experiments in the context of anti-immigrant behavior in the United States. In the first main experiment, participants learn about a study claiming that immigrants increase crime rates and then choose whether to authorize a publicly observable donation to an anti-immigrant organization. Informing participants that others will know that they learned about the study substantially increases donation rates. In the second main experiment, participants learn that a previous respondent authorized a donation to an anti-immigrant organization and then make an inference about the respondent’s motivations. Participants who are informed that the respondent learned about the study prior to authorizing the donation see the respondent as less intolerant and more easily persuadable.

ROTH Jonathan (Université Rennes 1) I Have Nothing Against Them, But...

Leonardo Bursztyn, Ingar Haaland, Aakaash Rao

Texte intégral

Behavior seminar

Le 11/06/2020 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Does politics influence citizens’ willingness to comply with government guidelines that severely restrict their freedom to engage in habitual private behavior? We examine how government action and political attitudes combine to induce compliance during times of extreme crisis. Based on panel surveys collected before and during the height of the COVID-19 crisis in France in March 2020, we examine the links among the exercise of public authority, citizens’ political preferences, and their expressed willingness to adjust everyday travel, hygiene, and social behavior.

ANDERSON Chris (Université Rennes 1) The Long Arm of the State: Citizen Compliance in Times of Crisis

Behavior seminar

Le 04/06/2020 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4

ALGAN Yann (SciencesPo – Ecole d’Affaires Publiques) ANNULE

Behavior seminar

Le 07/05/2020 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


When economists aggregate people’s wellbeing to make judgement about the overall good of a society, they sometimes discount later wellbeing compared with earlier wellbeing. This makes good sense only if all wellbeing is dated, which implies that wellbeing is separable across times. But this sort of separability makes it hard to take proper account of the value of extending people’s lives. Any solution to this problem will depend on a theory about the value of population. The upshot is that any theory of discounting is committed to a particular ethics of population.

BROOME John (SciencesPo – Ecole d’Affaires Publiques) ZOOM - Population, separability and discounting

Behavior seminar

Le 23/04/2020 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Economic models typically assume agents to positively discount their future utility, that is, to enjoy sooner benefits more than later ones. A growing empirical literature has documented systematic violations of positive discounting for some non-monetary items. Yet, the reason for such violations is unclear. A key role might be played by the degree of temporal fungibility, intended as the extent to which a good can or cannot be saved for a later period. Herein, we report a longitudinal laboratory experiment where we study time preferences for a good with high market fungibility but no temporal fungibility: time spent in the laboratory. Subjects are asked to allocate money (control) and free time (treatment) over two future periods. We design a novel allocation environment which allows to disentangle discounting and convexity properties of the utility function. We find that, ceteris paribus, people prefer to allocate more money to the sooner than the later period, but more free time to the later than the sooner period. Personal timetables and heterogeneous preferences cannot solely explain this asymmetry. Our results suggest that positive discounting can lead to non-negligible predictive errors in many relevant domains, such as health, education and labor. They also invite to reconsider the interpretation of the discount factor in monetary decisions.

PRATI ALBERTO (Ecole d’Economie Aix-Marseille) ZOOM. Time preferences for duration

Maria Bigoni, Stéphane Luchini

Behavior seminar

Le 02/04/2020 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Using the Markets in Financial Instruments Directive questionnaire answers, we examine the gender gap in financial literacy of retail clients of a large European retail bank. Using a database of more than 80,000 retail clients, we identify 12,124 matched pairs of spouses (married or cohabiting). We find a gender gap in financial literacy for all individuals and show that the absolute average gap is reduced for spouses who answered the questionnaire at the same time. We also identify couples in which wives exhibit higher financial literacy than their husbands (inversed gender gap) and we determine the couple' characteristics in each spouses' financial literacy gap category. Our results are consistent with prior studies on the specialization of tasks between spouses and add new insights by showing how individual and couples' characteristics impact the gender gap in financial literacy.

BROIHANNE Marie-Hélène (EM Strasbourg) ZOOM. Spouses gender gap in subjective financial literacy under MiFID profiling

Behavior seminar

Le 05/03/2020 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Many political news stories reach citizens via a two-step process, transmitted to them indirectly via their social networks. Yet, why do some news stories “go viral” and become transmitted heavily in citizens’ social networks with strong impact on their political opinions while others go by almost unnoticed? In this talk, I integrate theories from cognitive and evolutionary psychology into classical political science research on the flow of political communication to argue that political news stories that resonate with deep-seated psychological biases will be transmitted more and have stronger impact on citizens’ political opinions when they transmitted in their social networks. Focusing on evolved biases for cheater detection, vivid social information, and negativity, I present experimental evidence from large-scale chain transmission studies collected in Denmark and the United States as well as observational analyses of Facebook data supporting this argument.

AAROE Lene (EM Strasbourg) Evolved psychological biases and dissemination of political information in social network


Texte intégral

Behavior seminar

Le 09/01/2020 de 10:30:00 à 11:30:00

Bâtiment A, Rez de chaussée, Salle 4


This paper presents the results of a field experiment designed to reduce gender discrimination in student evaluations of teaching (SET). In the first intervention, students receive a normative statement reminding them that they should not discriminate in SETs. In the second intervention, the normative statement includes precise information about how other students (especially male students) have discriminated against female teachers in previous years. The purely normative statement has no significant impact on SET overall satisfaction scores, suggesting that a blanket awareness-raising campaign may be inefficient to reduce discrimination. However, the informational statement appears to significantly reduce gender discrimination. The effect we find mainly comes from a change in male students’ evaluation of female teachers.

BORING Anne (EM Strasbourg) Reducing Discrimination in the Field: Evidence from an Awareness Raising Intervention Targeting Gender Biases in Student Evaluations of Teaching

Arnaud Philippe

Texte intégral

Behavior seminar

Le 21/11/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Fertility decisions have large long term effects on economic outcomes. This is true for both parents and children. Timing of selection into motherhood is especially important for young women as a birth can have large effects on human capital accumulation (HC). Individual fertility and HC investment decisions are however most likely endogenous and it is thus hard to establish causality in this relationship. This is also the case when considering the impact of parental selection on child outcomes. We therefore exploit the introduction of the contraceptive pill in the Netherlands as an exogenous shock to birth control technology that greatly enhanced women’s fertility decision making process. We first document that there was a massive drop in the proportion of women who had children after it became freely available to all women in 1970. We also investigate if the adoption of this new technology was linked its socially acceptability by looking at heterogeneity of the fertility response depending on an area’s religiosity. We proxy this by proportion votes for conservative religious parties and find much larger birth drops for more liberal municipalities relative to conservative ones. We then investigate how the pill impacted women's education and labour market outcomes and how this in turn influenced their children’s long-run outcomes with a focus on their offending behaviour.

Marie Olivier (EM Strasbourg) The Power of the Dutch Pill: Birth Control as Crime Control?

Esmée Zwiers (Princeton University)

Behavior seminar

Le 14/11/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


This paper shows that psychological attitude towards oneself is affected by recent and past random negative economic shocks. Combining individual panel data with rainfall variations during the growing season in Ethiopia, we find that experiencing a severe drought lowers individuals self-esteem by almost half a standard deviation. Furthermore, past similar droughts experienced during early childhood and adolescence have a persistent effect on adulthood self-esteem. We also find that children of low self-esteem individuals also report low level of self esteem. The economic implications are discussed.

DI FALCO Salvatore (EM Strasbourg) Economic Adversity and the Intergenerational Transmission of Personality

Behavior seminar

Le 10/10/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


This paper presents a model of discrimination in collective decisions. The focus is on the role of the institutional set-up for whether discrimination is mitigated or exacerbated. In particular, I consider the interaction of committee composition - homogeneous versus diverse with the decision making rule - unanimity versus majority voting. The analysis suggests that under uncertainty about the extent of the other decision makers' biases diversity in committees may help avoid own group favoritism. In homogeneous committees unanimity rule is more conducive to discrimination than majority rule.

DASKALOVA Vessela (EM Strasbourg) Discrimination in Collective Decisions

Behavior seminar

Le 19/09/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


In this article, we study voluntary contributions of heterogeneous groups to a public good in an experiment. Members of the same group have either low or high external marginal returns. We vary the level of information about the heterogeneity and about a contributor’s type between groups. While controlling for the net costs of contributions, we find that the level of information determines how types in heterogeneous groups vary in their contributions. When the type of a contributor can be identified, types with high returns contribute more, otherwise the effect disappears or even reverses with low types contributing more than high types. This result provides evidence for the so-called “poisoning-of-the-well” effect and demonstrates how it interacts with the information structure of the environment. Without any information about heterogeneity, there is no difference in contributions by types.

KROGER Sabine (EM Strasbourg) Public good production in heterogeneous groups: An experimental analysis on the relation between external return and information

Gerlinde Fellner-Röhling, Erika Seki

Texte intégral

Behavior seminar

Le 12/09/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


The emotion that someone expresses has consequences for how that person is treated. We study whether people display emotions strategically. In two laboratory experiments, participants play task delegation games in which managers assign a task to one of two workers. When assigning the task, managers see pictures of the workers and we vary whether getting the task is desirable or not. We find that workers strategically adapt their emotional expressions to the incentives they face, and that it indeed pays off to do so. Yet, workers do not exploit the full potential of the strategic display of emotions.

HOPFENSITZ Astrid (EM Strasbourg) The strategic display of emotions

Daniel Chen, Boris van Leeuwen, Jeroen van de Ven

Texte intégral

Behavior seminar

Le 27/06/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


In this paper we investigate the relationship between ethical choices and anti-social behaviours. To address this issue we ran a within-subjects laboratory experiment that included both a classic (hypothetical) moral dilemma (using the well-known Trolley problem) and a real-payoff money-burning experiment. A main contribution is that our Trolley dilemmas separate purely utilitarian from more clearly immoral choice options. Our results show that choices in both environments respond to incentives (i.e., the relative price of the ethical decision), and Trolley problem decisions are consistent with previously known results—individuals prefer no action over action, as well as indirect over direct responsibility, when negative consequences would be similar in either instance. In analyzing the determinants of anti-social money burning, our data identify money burning due to inequality aversion, but we also find some evidence of pure nastiness. Importantly, we find that utilitarian behaviour in the Trolley dilemma is not linked to antisocial money burning, which contrasts with previous conclusions in the literature. Nevertheless, we observe that the willingness to commit more clearly ethically dubious acts in the Trolley problem significantly predicts money burning and, more specifically, nastiness. We conclude that choices in hypothetical environments may be useful for predicting antisocial behaviours that have real payoff consequences and efficiency implications.

MASCLET David (Université de Rennes 1) Using ethical dilemmas to predict anti social choices with real payoff consequences : an experimental study

David L. Dickinson

Texte intégral

Behavior seminar

Le 20/06/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Previous work has shown that preferences are not always stable across time, but surprisingly little is known about the reasons for this instability. I examine whether variation in people’s emotions over time predicts changes in preferences. Using a large panel data set, I find that within-person changes in happiness, anger, and fear have substantial effects on risk attitudes and patience. Robustness checks indicate a limited role of alternative explanations. I further address potential endogeneity concerns by exploiting information about the death of a parent or child. This identification strategy confirms a large causal impact of emotions on preferences.

MEIER Armando (Université de Rennes 1) Emotions, Risk Attitudes, and Patience


Texte intégral

Behavior seminar

Le 18/06/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4

, (Université de Rennes 1) *

Behavior seminar

Le 06/06/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


What is rational to do for a decision-maker who is not able to identify states of the world or final consequences, and is not able to assign probabilities to all the relevant events? A set of dominance conditions deriving from stochastic dominance, formulated in a model in which the objects of preferences are act-event pairs, implies expected utility and, in a variant of Savage's theorem, the use of a well-defined probability measure compatible with the decision-maker's beliefs. As in Skiadas (1997b), the model is flexible enough to incorporate feelings such as ambiguity aversion in the evaluation of consequences. However, there is a sense in which ambiguity aversion is acceptable for individuals but appears unreasonable (which is not the same as irrational) for policy-makers.

Fleurbaey Marc (Princeton University) Rationality under risk and uncertainty

Behavior seminar

Le 23/05/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We study how participants' behavior in an experimental public good game is affected when they know that information about their choices and outcomes, together with different sets of information about their identity, will be transmitted the following year to a set of new, unknown participants - with no payoff linkages between the two sets of players. We explore two questions: first, does knowing that they will be acting as role models for future players influence individuals' choices? Second, how does behavior vary with the degree of future visibility of participants' identity? We find no evidence of a significant role model effect, but a surprising effect of visibility: when subjects know their photo will be transmitted to future, unknown and unrelated participants, they contribute significantly less. We consider different possible explanations, and argue that the most convincing is a sucker aversion explanation according to which subjects in the photo treatment become more sensitive to being perceived as exploited by their peers.

DESSI Roberta (Princeton University) Public goods and future audiences: acting as role models?

Giuseppe Attanasi, Frédéric Moisan, Donald Robertson

Behavior seminar

Le 16/05/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


A growing number of firms claim to place a strong emphasis on the “happiness” of their employees as a way to boost performance. Yet, despite this, there is a lack of causal field evidence on the link between mood and productivity. We measure employee mood at a large telecommunications firm using a novel weekly survey over a 6 month period, and link these reports with detailed individual-level administrative measures of workplace behaviors and performance. Being in a positive mood improves weekly sales by around 13%. Exploiting variation in local weather conditions, we show that na¨?ve OLS estimates are a lower bound on the causal effect of mood on productivity. We discuss various threats to the validity of our instrumented analysis, and consider a number of different mechanisms. The data suggest that the effect of mood on performance runs largely through workers converting more calls to sales, rather than working any faster or putting in more hours.

BELLET Clément (Princeton University) The Impact of Employee Mood on Productivity

Jan-Emmanuel De Neve et George Ward

Behavior seminar

Le 09/05/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We conduct a field experiment to evaluate the short and long term effects of two interventions targeting the dietary habits of low income families with young children. In one treatment, families received food groceries at home for free for twelve weeks and were asked to prepare five specific healthy meals per week. In the other treatment, families were simply asked to reduce snacking and eat at regular times, also for twelve weeks. We evaluate the impact of the interventions on diet and BMI over the course of three years. We find evidence that children's BMI distribution shifted significantly relative to the control group, i.e. they became relatively ``thinner''. This effect persists three years after the intervention for the first intervention, but fades away for the second. We find evidence that children reduced their sugar intake following both treatments. However, we find little evidence that their preferences changed in favor of healthier foods. A possible explanation is that children were restricted access to foods high in sugar in the treated groups. Parents, on the other hand, do not appear to have changed their diet as a result of the interventions, neither in the short run nor in the longer run.

BELOT Quentin (Princeton University) The Formation and Malleability of Dietary Habits: A Field Experiment with Low Income Families

Noemi Berlin, Jonathan James and Valeria Skafida

Texte intégral

Behavior seminar

Le 18/04/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


How does electoral rule disproportionality affect the structure of the party system (i.e. the number and the policy platforms of the competing parties)? By studying a model, where both party entry and platform choice are endogenous, we are able to provide a unified theory: An increasing electoral rule disproportionality exhibits: a) a first-order negative effect on platform polarization, b) a second-order negative effect on the number of parties (as polarization decreases, centrist parties are squeezed between other contenders and prefer not to enter), and c) an additional third-order negative effect on polarization via the reduction of the number of parties. We then conduct a laboratory experiment and strongly confirm the theoretical predictions of the model.

BOL Damien () Electoral Rules, Strategic Entry and Polarization

Konstantinos Matakos, Orestis Troumpounis, Dimitrios Xefteris

Behavior seminar

Le 11/04/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


During human evolution, individuals interacted in bands connected by limited migration and sometimes conflicts. If the spread of preferences, from one generation to the next, depends on their material success, which preferences will prevail ? Building on population biology models of spatially structured populations, and assuming preferences to be private information, we characterize which preferences, if any, cannot be displaced, once established. We find that such uninvadable preferences represent different motives when expressed in terms of fitness than when expressed in terms of material payoffs. At the fitness level, individuals appear be driven by a mix of self-interest and a Kantian motive, which involves evaluating one's behavior in light of what own fitness would be if others were to choose the same behavior. This Kantian motive is borne out from kin selection (be it genetic or cultural). At the material payoff level, individuals appear to be driven by these two motives, but also in part by an other-regarding motive (spite or altruism) towards individuals with whom they interact. We show how population structure---size of bands, migration rates, probability of conflicts between bands, cultural loyalty towards parents---shape the relative importance of these motives.

Alger Ingela () Evolution of Preferences in Structured Populations

Laurent Lehmann (Lausanne University) et Jörgen Weibull (Stockholm School of Economics)

Behavior seminar

Le 28/03/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Recent work in organizational economics stresses the importance of clarity in relational contracts as a reason behind the persistent performance differences that we observe between seemingly similar enterprises. Even though the conditions necessary to establish cooperation through relational contracts are straightforward in theory, in practice establishing a relational contract poses a number of difficult challenges. Specifically, trading parties face the clarity problem: they need to establish mutual understanding about the content of the relational contract and they need to succeed in adapting this mutual understanding to new situations when the environment changes. We hypothesize that building the relational contract on general principles, rather than relying on specific but narrow rules, helps achieve clarity and improves performance in repeated games with uncertainty about the future, and report results from an experiment assessing this hypothesis.

HERZ Holger () Clarity in Relational Contracts: Rules vs. Principles

Behavior seminar

Le 21/03/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Sleep deprivation is common around the world. While sleep medicine has established that inducing acute sleep deprivation substantially worsens cognition, we know little about the real-world impacts of improving sleep. We hire 450 individuals in urban India as data-entry workers and offer a random subset different interventions to increase their sleep: (i) devices to improve their home-sleep environment, (ii) additional financial incentives to increase sleep, and (iii) the opportunity to take a short nap in the afternoon. We present three sets of results. First, the interventions increase night sleep duration by 20 to 40 minutes (on a base of 5.5 hours per night in the control group) with no detectable changes in sleep efficiency. Individuals assigned to the nap treatment sleep on average about 12 minutes during their naps. Second, contrary to predictions by most sleep experts and economists, improved night sleep lowers labor supply slightly (6.5 minutes) and does not significantly improve productivity or earnings. In contrast, naps increase productivity by about 2-3 percent, although the increase is insufficient to fully counteract the associated reduction in labor supply relative to working through that time. Third, increased night sleep improves health as measured by a composite health index by 0.1 units, and naps improve an index of well-being by a similar amount. Taken together, we find little evidence of increased sleep causing impacts on short- and medium-run economic outcomes that could be easily discernible by individuals, thus providing a possible explanation for the persistence of widespread sleep deprivation found in many settings.

SCHOFIELD Heather () Sleepless in Chennai: The Consequences of Improving Sleep among the Urban Poor

Behavior seminar

Le 14/03/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We conduct a field experiment to test the demand for flexibility and for soft and hard commitment among clients of a microfinance institution. We offer a commitment contract inspired by the rotating structure of a ROSCA. We find substantial demand for both saving and credit contracts, with many respondents willing to take up either — suggesting that many microfinance clients borrow to save. Additional treatments test ex ante demand for soft commitment (e.g., reminders), hard commitment (e.g., penalty for missing an instalment), and flexibility (e.g., option to postpone an instalment). We find no demand for any of these features, in isolation or in combination: individuals appear to actively dislike them all. These findings complement a literature showing that commitment devices induce financial discipline. Our results suggest that many commitment devices used in practice may be seen as overly restrictive ex ante, even for a population with a demonstrated demand for commitment products.

QUINN Simon () When Nudge Comes to Shove: Demand for Commitment in Microfinance Contracts

Behavior seminar

Le 21/02/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We conduct an experiment to study how beliefs about behavioral responses to taxation and preferences over equality–efficiency trade-offs relate to the political disagreement on redistribution. We use a novel method to elicit incentivized beliefs from a sample of 13,900 Americans about how taxes affect people’s effort choices, and we elicit incentivized equality–efficiency preferences. We find that Democrats and Republicans have virtually identical beliefs about behavioral responses to taxation. Furthermore, we find that beliefs about behavioral responses to taxation fail to predict people’s support for equalization of incomes in society. Equality– efficiency preferences, by contrast, strongly predict both people’s political affiliation and their support for equalization of incomes in society. We also explore the role of motivated beliefs and identity politics by priming respondents about the political disagreement on redistribution. The treatments increase political polarization in preferences, but do not polarize beliefs. Our findings suggest that the political divide on redistribution relates more to people’s preferences than to their beliefs about the behavioral responses to taxation.

TUNGODDEN Bertil () Beliefs about Behavioral Responses to Taxation

Alexander W. Cappelen and Ingar K. Haaland

Texte intégral

Behavior seminar

Le 14/02/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4

, () *; () ;

La séance est annulée

Behavior seminar

Le 07/02/2019 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


International cooperation to supply a global public good like climate change mitigation is more vulnerable to free riding than international cooperation to liberalize trade. Can it help to link these different issues? Simple theory shows that, in the best possible case, linkage is a coordination game in which equilibrium selection is unreliable. Taking our game into the experimental lab, we find that the institution players use to decide about linkage and the relative gains of trade matter for the use and success of linkage.

DANNENBERG Astrid () Coercive Trade Agreements for Supplying Global Public Goods

Behavior seminar

Le 20/12/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


In the context of a weight loss challenge, I use the menu choice approach of Gul and Pesendorfer (2001) to provide new insights on the link between commitment and temptation. First, I study commitment to eating healthy by eliciting participants’ preferences over a set of lunch reimbursement options, which differed in their food coverage. Extracting information from the entire ordering, I develop measures of temptation to study its source, strength and structure, and validate those measures with survey data. Finally, I test whether temptation revealed through menu choice can predict other behaviors that could be symptomatic of self-control problems, such as take-up of, and performance on, a goal setting contract. In this rich environment, I find a tight link between commitment and temptation. First, nearly 50% of participants strictly preferred a coverage that excluded the foods they rated as most tempting and unhealthy. Second, those who revealed their temptation through menu choice were more likely to take up the contract and less likely to achieve their goals. The elicitation of menu preferences thus offers a promising venue for measuring self-control problems.

TOUSSAERT Séverine () Revealing temptation through menu choice: Evidence from a weight loss challenge

Behavior seminar

Le 13/12/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Decision theory, as used in economics, is based on axioms that are usually interpreted as principles of rationality. When decision theory is used to explain actual behaviour, the usual justification is that, in the situations to which the theory is applied, individuals can be expected to reason correctly. Thus, there is an implicit assumption that if an individual reasons correctly, his preferences will satisfy the rationality axioms. The same assumption often appears in behavioural economics, as the claim that observed deviations from standard decision theory are the result of ‘errors’ and that the satisfaction of individuals’ ‘true’ (i.e. error-free) preferences can be used as a normative criterion. But very little effort has been made to explain what correct reasoning is, or how it leads to the satisfaction of rationality axioms. A few writers have expressed scepticism about whether any process that might plausibly be called ‘correct reasoning’ can achieve this (e.g. Broome, Rationality through Reasoning, 2013; Cubitt and Sugden, Economics and Philosophy, 2014; Infante, Lecouteux and Sugden, Journal of Economic Methodology, 2016). Broome argues specifically that one cannot achieve rationality by ‘second-order’ reasoning from the premise that one’s preferences ought to satisfy rationality properties. In this paper, I re-examine the famous case of Savage’s (1954) response to the Allais Paradox. Savage is the creator of the canonical axiomatisation of rational choice theory, and is explicit that his axioms are axioms of rationality, analogous with principles of logic. But when confronted with Allais’s decision problems, he discovered that his preferences contravened his own axioms. His response was to persuade himself that his original preferences were erroneous. I analyse the process of reasoning by which he reached this conclusion. I argue that Savage’s reasoning is first-order in Broome’s sense, and so not vulnerable to Broome’s objection, but that it draws on his own mental states in a way that goes beyond formally ‘correct’ reasoning. My conclusion is that Savage’s own position is coherent and defensible, but on the issue of whether rationality can be achieved by reasoning, the sceptics are right.

SUGDEN Robert () Are violations of rational-choice theory errors that reasoning can correct? A re-examination of Savage’s response to the Allais Paradox

Franz Dietrich and Antonios Staras

Behavior seminar

Le 15/11/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Can participation in financial markets lead individuals to re-evaluate the costs of conflict, change their political attitudes and even their votes? Prior to the 2015 Israeli elections, we randomly assigned Palestinian and Israeli financial assets to likely voters, and incentivized them to actively trade for up to seven weeks. No political messages or non-financial information were included. The treatment systematically shifted vote choices towards parties more supportive of the peace process. This effect is not due to a direct material incentive to vote a particular way. Rather, the treatment reduces opposition to concessions for peace, and increases awareness of the broader economic risks of conflict. While participants assigned Palestinian assets are more likely to associate their assets' performance to peace, they are less engaged in the experiment. Combined with the superior performance of Israeli stocks during the study period, the ultimate effects of Israeli and Palestinian assets are similar. How does engagement with markets aect social-economic values and political pref- erences? A long line of thinkers have debated the nature and direction of such eects, but claims are dicult to assess empirically because market engagement is endoge- nous. We designed a large eld experiment to evaluate the impact of nancial markets, which have grown dramatically in recent decades. Participants from a national sample in England received substantial sums they could invest over a six-week period. We assigned them into several treatments designed to distinguish between dierent the- oretical channels of in uence. Investment in stocks led participants to adopt a more right-leaning outlook on issues such as merit and deservingness, personal responsibil- ity and equality. Subjects also shifted to the right on policy questions. These results appear to be driven by growing familiarity with, and decreasing distrust of markets. The spread of nancial markets thus has important and under-appreciated political ramications.

SHAYO Moses () How Do Markets Affect Values and Behavior? Emerging Patterns from Three Studies

Saumitra Jha, Yotam Margalit and Itay Sisso

Texte intégral

Behavior seminar

Le 08/11/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


It is well known that uncertainty is a key consideration in theoretical health economics analysis. The literature has shown that uncertainty is a multifaceted concept, with the individual’s optimal response depending on the formal nature of the uncertainty and the time horizon involved. This paper extends the literature by considering uncertainty with regards to the cumulative effect on health capital of on-going health behaviours. It uses techniques of stochastic optimal control to analyze uncertainty which can be represented as a Weiner process and shows how, in a Grossman health investment framework, the optimal lifetime health investment trajectory might be affected. JEL codes: I1, I12

LAPORTE Audrey () Making the Grossman Model Stochastic: Investment in Health as a Stochastic Control Problem

Behavior seminar

Le 11/10/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Is workers' health more sensitive to losses than gains in job security? While loss aversion, whereby losses loom larger than gains, is typically examined in relation to decisions about anticipated outcomes, I first show using a large sample of workers from the European Household Community Panel and value-added models that losses in job security have a larger eff ect on health than equivalent job security gains. Second, I address endogeneity issues using the 1999 rise in the French Delalande tax as a quasi-natural experiment. It allows evaluating separately the causal impact of exogenous gains and losses in job security on workers' health. Di fference-in-diff erences estimation results con rm that lower job security generates signifi cant and robust losses in self-assessed health. Meanwhile a greater feeling of job security does not translate into a higher level of self-assessed health. These results are in line with the predictions of the model linking job security to health under the hypothesis of loss aversion built in this paper. This article also demonstrates that losses in health induced by lower job security are not transitory.

Lepinteur Anthony () The Asymmetric Experience of Gains and Losses in Job Security on Health


Texte intégral

Behavior seminar

Le 27/09/2018 de 12:00:00 à 13:30:00

Bâtiment A, Rez de chaussée, Salle 4


The 20th century has witnessed a rapid pace of cultural change. This paper focuses on cultural change in one particular area -- attitudes towards gay people -- and argues that the AIDS crisis was an important propagator of change. We examine this hypothesis empirically in a variety of ways

Fernandez-Urbano Roger () Cultural change

Sahar Parsa and Martina Viarengo

Behavior seminar

Le 20/09/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


The identities (social categories) a person holds are important determinants of choice. However, questions remain about why people who have the same identity often behave differently, and how multiple identities interact with other aspects of the self-concept. To understand these questions, I first examine how people mentally represent or organize information about the self-concept. More specifically, I propose a novel theoretical approach to the self-concept that suggests that aspects of the self-concept are seen as defining of an individual to the extent that they are perceived as causally central, having an influence on many other aspects of the self-concept. I then examine the implications of this account of the self-concept for identity-consistent behaviors—do differences in subjective beliefs about how an identity is causally connected to other aspects of the self-concept predict differences in identity-consistent behaviors? I provide evidence that people who perceive an identity (e.g., identity as a Democrat or Republican) as more causally central are more likely to act in identity-consistent ways (e.g., vote for their party’s candidate) than people who possess the same identity but perceive it as more causally peripheral (i.e., connected to fewer other aspects of the self-concept).

CHEN Stéphanie () Representations of the self-concept and identity-based choice

Oleg Urminsky

Behavior seminar

Le 28/06/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4

Stutzer Alois () Women leaving the playpen: The emancipating role of female su

Michaela Slotwinski (University of Basel)

Behavior seminar

Le 21/06/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Corruption is a major threat to economic and social development. Democracy in itself is not necessarily conducive to less corruption. Voters may lack information on politicians' wrongdoings, and electoral institutions may make it hard for them to remove corrupt politicians from office. From these premises, one might expect that more information and more open electoral systems, that is, systems giving voters more freedom to express their preferences over individual candidates, should help remove corrupt politicians from office. We propose a simple theoretical model describing voters' behavior under closed list and open list proportional representation systems, and derive predictions regarding the impact of electoral rules and information on candidates' electoral prospects. We test these hypotheses in a survey experiment performed in Paraguay taking advantage of a rare social uprising following a corruption scandal. We find that under the more open system, supporters of the incumbent party -the most corrupt party- do actually exhibit a preference for corrupt politicians, and that this is not due to a lack of information. Besides, under the open system, vote shares for the big political parties increase, especially so for the incumbent. Based on this evidence, we challenge the conventional view that more open electoral systems are necessarily good at fighting corruption.

Van der Straeten Karine () Voting corrupt politicians out of office: Evidence from an Experiment in Paraguay

Rumilda Canete, Pepita Miquel (Toulouse School of Economics & IAST) and Stephane Straub (Toulouse School of Economics & IAST).

Behavior seminar

Le 07/06/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We find substantial effects of fathers' multiple-partner fertility (MPF) on children's long-term educational outcomes. We focus on the children in fathers’ “second families,” emphasizing the case in which the second families are nuclear families – households consisting of a man, a woman, their joint children, and no other children. We analyze outcomes for almost 80,000 children born in Norway in 1986-1988 who, until they were at least age 18, lived with both biological parents. This analysis cannot be done using existing US data sets. Children who spent their entire childhoods in nuclear families but whose fathers had children from another relationship living elsewhere were more likely to drop out of secondary school (24% vs 17%) and less likely to obtain a bachelor's degree (44% vs 51%) than children in nuclear families without MPF. Our probit estimates imply that the marginal effect of fathers' MPF is 4 percentage points for dropping out of secondary school and 5 percentage points for obtaining a bachelor's degree. Our analysis suggests that the effects of fathers' MPF are primarily due to selection.

POLLAK Catherine () Fathers' Multiple-Partner Fertility and Children’s Educational Outcomes

Behavior seminar

Le 31/05/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Abstract We study fairness in economies where humans consume one private good and one public good, the latter representing the welfare of other species. We show that it is impossible for a social evaluator to be egalitarian with respect to humans and be speciesist, in the sense of always respecting unanimous preferences among humans (even when they imply harming other species). One solution is to only impose a respect for unanimity among humans when it does not lead to a decrease in the welfare of other species. Characterizing social preferences satisfying these properties reveals a surprising link between concerns for other species, egalitarianism among humans, and unanimity: the latter two imply a form of dictatorship from humans with the strongest preferences for the welfare of other species. Keywords: Welfare Economics, Animal Ethics, Egalitarianism, Eciency, Fairness.

Fleurbaey Marc () If you're an egalitarian, how come you're not a vegan?

Martin Van der Linden (Utah State University)

Behavior seminar

Le 17/05/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4

ELLIS RANDALL P. (Department of Economics Boston University) Provider Agency, Health Care Innovation, Pricing, and Regulations: (or How to Reduce Health Care Spending by 50 Percent)

Behavior seminar

Le 03/05/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4

Prati Alberto () Retrospective self-serving believes on dating events: an empirical study using subjective well-being data

Behavior seminar

Le 12/04/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Recent business scandals, economic upheavals and claims of broad types of unethical conduct among professional economists (e.g., List et al., 2001) have contributed to a growing discourse on returning economics to its origins as a “moral science” and on strengthening the emphasis on ethics in economics teaching and research, e.g., see Atkinson (2011), Bruni and Sugden (2013), DeMartino (2011), Sandel (2013), and Schiller and Schiller (2011). A now voluminous literature that began with Marwell and Ames (1981) and Carter and Irons (1991) asks “does studying economics lead to more self-interested behaviour?” Here we turn this question on its head and ask “does studying ethics in an economic context affect moral attitudes or behavior?” This presentation reports the results of two studies. “Does Studying Ethics Affect Moral Views?: An Application to Economic Justice” considers whether the oldest and most traditional form of ethics instruction, namely philosophical ethics, can influence views of economic justice using the responses in surveys to contextually rich vignettes. “An Experimental Study of the Behavioral Effects of Ethics in the Economics Classroom” reports the results of economics experiments conducted with students in introductory economics classes following lectures on ethics. The results are sometimes enigmatic: non-effects obtain, even when the costs of reporting an effect are low in the survey study, whereas significant effects sometimes appear, when participants have material stakes in the economics experiments. These findings suggest that considerable subtlety is required in contemplating types of ethics instruction and in interpreting the effects.

Konow James () Can Economic Ethics Affect Attitudes and Behavior?

Behavior seminar

Le 05/04/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4

Palminteri Stefano () TBA

Behavior seminar

Le 29/03/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Abstract: We study how employees learn about the salaries of their peers and managers, and how those beliefs affect their own behavior. We conducted a field experiment with a sample of 2,000 employees from a multi-billion-dollar corporation. We combine rich data from surveys and administrative records with an experiment that provided some employees with accurate information about the salaries of others. First, we document large misperceptions about salaries and identify some of the sources of these misperceptions. Second, we find significant behavioral elasticities with respect to the perceived salaries of other employees. These effects are different for horizontal and vertical comparisons: while higher perceived peer salary decreases effort, output and retention, higher perceived manager salary has a positive effect on those same outcomes. We discuss evidence on the underlying mechanisms, and implications for pay inequality and pay transparency.

Cullen Zoe (HBS) How Much Does Your Boss Make? The Effects of Salary Comparisons

Behavior seminar

Le 15/03/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


In this paper we report the results of a repeated experiment in which a central bank buys bonds for cash in a quantitative easing (QE) operation in an otherwise standard asset market setting. The experiment is designed so that bonds have a constant fundamental value which is not affected by QE under rational expectations. By repeating the same experience three times, we investigate whether participants learn that prices should not rise above the fundamental price in the presence of QE. We find that some groups do learn this but most do not, instead becoming more convinced that QE boosts bond prices. These claims are based on significantly different behaviour of two treatment groups relative to a control group that doesn't have QE.

HANAKI Nobuyuki (HBS) An experimental analysis of the effect of Quantitative Easing

Behavior seminar

Le 08/03/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Since the Industrial Revolution, human societies have experienced a high and sustained rate of economic growth. Recent explanations of this sudden and massive change in economic history have held that modern growth results from an acceleration of innovation. But it is unclear why the rate of innovation drastically accelerated in England in the 18th century. An important factor might be the alteration of individual preferences with regard to innovation due to the unprecedented living standards of the English during that period, for two reasons. First, recent developments in economic history challenge the standard Malthusian view according to which living standards were stagnant until the Industrial Revolution. Pre-industrial England enjoyed a level of affluence that was unprecedented in history. Second, Life History Theory, a branch of evolutionary biology, has demonstrated that the human brain is designed to respond adaptively to variations in resources in the local environment. In particular, a more favorable environment (high resources, low mortality) triggers the expression of future-oriented preferences. In this paper, I argue that some of these preferences – a lower level of time discounting, a higher capacity to accept losses, a lower materialistic orientation and a higher tendency to explore – are likely to increase the rate of innovation. I review the evidence regarding the impact of affluence on preferences in contemporary we well as in past populations, and conclude that the impact of affluence on neuro-cognitive systems may partly explain the modern acceleration of technological innovations and the associated economic growth.

Baumard Nicolas (Département d'Etudes Cognitives de l'ENS) PSYCHOLOGICAL ORIGINS OF THE INDUSTRIAL REVOLUTION
Chevalier Coralie (Département d'Etudes Cognitives de l'ENS)

Behavior seminar

Le 15/02/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Poor households regularly borrow and lend to smooth consumption, yet we see much less borrowing for investment. This cannot be explained by a lack of investment opportunities, nor by a lack of resources available collectively for investment. This paper provides a novel explanation for this puzzle: investment reduces the investor's need for informal risk sharing, weakening risk-sharing ties, and so limiting the amount of borrowing that can be sustained. I formalise this intuition by extending the canonical model of limited commitment in risk-sharing networks to allow for lumpy investment. The key prediction of the model is a non-linear relationship between total income and investment at the network level -- namely there is a network-level poverty trap. I test this prediction using a randomised control trial in Bangladesh, that provided capital transfers to the poorest households. The data cover 27,000 households from 1,400 villages, and contain information on risk-sharing networks, income, and investment. I exploit variation in the number of program recipients in a network to identify the location of the poverty trap: the threshold level of capital provision needed at the network level for the program to generate further investment. My results highlight how capital transfer programs can be made more cost-effective by targeting communities at the threshold of the aggregate poverty trap.

Advani Arun (Département d'Etudes Cognitives de l'ENS) nsurance Networks and Poverty Traps

Behavior seminar

Le 25/01/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4

RAVESTEIJN BASTIAN () TBA

Behavior seminar

Le 18/01/2018 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


Abstract: Classic economic theory focuses on static and stable preferences. However, there is growing evidence that cognitive, emotional and visceral states (e.g. fatigue, stress or hunger) can mediate behavioral biases and shape preferences (DellaVigna, 2009). In particular, Symmonds & al. (2010) and Levy & al. (2013) provide evidence that risk attitudes fluctuate with metabolic states. In this paper, we follow this stream of research and propose an experimental design with a specific hunger manipulation mechanism using high-protein drink and an original risk attitude elicitation tool that allows parametric estimation of the components of Prospect Theory by using a convex budget line allocation methodology (Choi & al., 2007). The experiment (N=107) took place in the Social Sciences Experimental Lab (Xlab) at the University of California, Berkeley. Fasting requirement combined with the nutritional-shake tasting activity resulted in a successful manipulation of hunger. Our preliminary results suggest a limited impact of hunger on the utility function and loss aversion parameters. There could be a positive effect of hunger on the level of optimism (represented by the elevation of the probability weighting function) suggesting a lower risk aversion for hungry (fasting) individuals for small gain probabilities. These results have to be confirmed by further analysis.

NEBOUT Antoine () Hunger Games II: Do Hunger Affects Risk Preferences?

Lydia Ashton, Emmanuel Kemel

Behavior seminar

Le 21/12/2017 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


The Netherlands stands out for offering a generous public coverage of long-term care (LTC) services to its disabled elderly population. In our paper, we investigate whether the Dutch system ensures socio-economic horizontal equity in the use of LTC services, i-e whether individuals with similar needs for LTC receive the same amount of services, irrespective of their income. While most studies of horizontal equity in health care use typically rely on a statistically derived measure of needs, we use the eligibility assessment made by the Dutch independent central LTC assessment agency as an explicit norm of vertical equity. We exploit rich administrative data on the universe of the individuals aged 60 or more eligible for public LTC in 2012 (N=616,934). Our data allow us to construct a measure of LTC use (resp. needs) as the monetary value of all institutional care and home care services the individual used (resp. was entitled to) in 2012, while providing individual socio-economic and demographic information. We find substantial pro-poor concentration of LTC use, only partially offset by poorer individuals having higher needs for LTC. The differential gap between use and needs across the income distribution is especially marked among those eligible for home care. Income, age and household composition contribute to pro-poor income-related horizontal inequity, while regional differences in use, origin and wealth show a negligible contribution.

Tenand Marianne () Long--term care use in the Netherlands: equal treatment for equal needs? An assessment using administrative data.

Behavior seminar

Le 14/12/2017 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


This paper uses an equivalent income approach to quantify the domes- tic welfare loss due to the Syrian Civil War. Focusing on the (income, life expectancy) space, we show that the equivalent income has fallen by about 60 % in comparison to the pre-conáict level. We also Önd that the di§er- ential between the equivalent income and the standard income for 2016 lies between $75 and $144. Although this low willingness to pay for com- ing back to pre-conáict survival conditions can be explained by extreme poverty due to the War, the small gap between standard and equivalent incomes tends to question the extra value brought by the latter for the measurement of standards of living in situations of severe poverty. We examine some solutions to that puzzle, including a more general speciÖ- cation of the utility function, the shift from an ex ante approach (valuing changes in life expectancy) to an ex post approach (valuing changes in dis- tributions of realized longevities), as well as considering population ethical aspects. None of those solutions is fully successful in solving the puzzle.

PONTHIÈRE Grégory () The Domestic Welfare Loss of Syrian Civil War: An Equivalent Income Approach

Behavior seminar

Le 30/11/2017 de 12:30:00 à 13:30:00

Bâtiment A, Rez de chaussée, Salle 4

Van den Berg Gerard () Conditions Early in Life and the Ensuing Shape of the Age-Earnings Profile over the Full Working Life

(joint with public eco, see public eco)

Behavior seminar

Le 16/11/2017 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We study the tradeoffs between centralized and decentralized management using a new experimental game, the decentralization game. This game models an organization with two divisions and one central manager. Each division must choose or be assigned a product type. Both divisions benefit from coordinating their product types, but each prefers to coordinate on products that are close to its local tastes. The central manager aims to maximize the sum of division payoffs. Which product type achieves this goal varies with taste shocks that are known to the divisions but not the central manager. Under centralization, the central manager assigns products to divisions after receiving the divisions’ messages about the state of the world (i.e., the taste shock); under decentralization, the divisions choose their own products. Contrary to the theory, overall performance is higher under centralization than under decentralization. Communication between divisions and suggestions from central managers modestly improves performance under decentralization. Nonetheless, centralization remains the best-performing organizational form.

BRANDTS JORDI () Centralized vs. Decentralized Management: An Experimental Study

Behavior seminar

Le 09/11/2017 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We provide a model where hospitals compete on quality under a fixed price regime to investigate (i) whether hospital competition, as measured by an increase in fixed prices or increased patient choice, increases or reduces the gap in quality between high- and low-quality hospitals, and as a result, (ii) whether competition increases or reduces (pure) health inequalities. The answer to the first question is generally ambiguous, but we find that that the scope for competition to result in quality convergence across hospitals is larger when the marginal health gains from quality decrease at a faster rate. Whether competition increases inequalities depends on the type of inequality. If marginal health gains decrease at a sufficiently slow rate, health inequalities due to postcode lottery will increase (decrease) whenever competition induces quality dispersion (convergence). Competition reduces health inequalities between high- and low-severity patients if patient composition effects, due to high-severity patients being more likely to exercise choice, are small.

SICILIANI Luigi () Competition and Equity in Health Care Markets (with Odd Rune Straume, University of Minho)

Behavior seminar

Le 12/10/2017 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We study how cooperation-enforcing institutions dynamically affect values and behavior using a lab experiment designed to create individual specific histories of past institutional exposure. We show that the effect of past institutions is mostly due to “indirect” behavioral spillovers: facing penalties in the past increases partners’ cooperation in the past, which in turn positively affects ones’ own current behavior. We demonstrate that such indirect spillovers induce persistent effects of institutions. However, for interactions that occur early on, we find a negative effect of past enforcement due to differential learning under different enforcement institutions.

JACQUEMET Nicolas., Emeric Henry, GALBIATI Roberto () Learning, Spillovers and Persistence: Institutions and the Dynamics of Cooperation

Behavior seminar

Le 05/10/2017 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We present a multi-modal database for the analysis of human emotional states in the dynamic context of game. Peripheral physiological signals (ECG, EDA, respiration, EMG, temperature, pupil size), accelerometer signals, video were recorded for 58 participants who played 3 matches of a football simulation game (FIFA 2016) of different difficulty levels. After each match, participants were asked to annotate events which have triggered their emotions in terms of categorical emotions (boredom, frustration, anger, fear, happiness) and the arousal/valence score by reviewing the game recording. Global game experience in terms of the level of difficulty, implication and amusement evaluation and a ranking of the 3 matches were also collected from questionnaires. Combined with the global game experience evaluation, the database proposes a multi-scale view (local event scale and global match scale) for evaluating user emotions. Methods and results are presented from a local event view - classification of the emotional moment and no emotional moment, emotion recognition on the emotional moment, and from a global evaluation view - game experience evaluation. The database is made publicly available. The proposed database is the first available one which allows analysing the human emotional responses from physiological signals under dynamic interactive context of game.

YANG Wenlu (LIP6-UPMC) Automatic emotion recognition from physiological signals in a video game context

Behavior seminar

Le 28/09/2017 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


This paper studies the long- run impact of learning via coarse feedback when a similarly labeled action may have consequences that depend on the state of the economy. The decision maker is only informed about the payoff obtained in the past by fellow decision makers who had chosen this action, with no specification of the state of the world when the choice was made.This nature of feedback suggests that decision makers face an ambiguous environment even when a lot of data has been accumulated. We consider a lab experiment with two binary choice problems in which one action is subject to coarse feedback but not the other and we allow for the possibility that the action associated to the coarse feedback would be subject to an ambiguity discount. Estimating such a noisy best-response model, we find that there is no ambiguity discount, thereby suggesting that subjects behave in a noisy-best response way as in the valuation equilibrium introduced in Jehiel and Samet (2007). We suggest how our findings can be applied to technology adoption, discrimination and investment problems.

SINGH Juni () Understanding choices under coarse feedback: a lab experiment

Behavior seminar

Le 21/09/2017 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


One problem with the Becker-Murphy model of Rational Addiction, at least in the eyes of many public health specialists, is that it does not explain why so many rational, forward looking, smokers should apparently find it so hard to quit, especially since the terminal conditions are part of an intertemporal optimization problem.  In this paper we apply techniques of stochastic control theory to introduce uncertainty into the individual’s perception of how her stock of addiction will accumulate over time as a consequence of her time path of smoking.  We assume that addiction capital is basically unobservable, so she cannot adjust her smoking behaviour according to a feedback policy rule but instead builds uncertainty into her consumption plan from the beginning.  We discuss the differences between the equation explaining her lifetime smoking trajectory in the deterministic and stochastic cases, and find that the quadratic utility function which underlies the familiar lead-lag consumption form of rational addiction equation is not, in fact, capable of allowing for the type of uncertainty which we consider here.

LAPORTE Audrey () Why Should Rational Smokers Find it Difficult to Quit?  Introducing Uncertainty into the Rational Addiction Model

Behavior seminar

Le 07/09/2017 de 11:00:00 à 12:00:00

Bâtiment A, Rez de chaussée, Salle 4


We conduct an experiment where participants choose between actions that provide private benefits but may also impose losses on strangers. Three legal environments are compared: no law, strict liability for the harm caused to others, and an efficiently designed negligence rule where damages are paid only when the harmful action causes a net social loss. Legal obligations are either perfectly enforced (Severe Law) or only weakly so (Mild Law), i.e., material incentives are then nondeterrent. We investigate how legal obligations and social norms interact. Our results show that liability rules strengthen pro-social behavior and suggest that strict liability has a greater effect than the negligence rule.

ESPINOSA Romain () Laws and Norms: Experimental Evidence with Liability Rules?

Behavior seminar

Le 29/06/2017 de 12:00:00 à 13:00:00

Bâtiment A, Rez de chaussée, Salle 4

FIORINI Luciana () The Opportunity Criterion: An Axiomatic Approach

Behavior seminar

Le 22/06/2017 de 12:00:00 à 13:00:00

Bâtiment A, Rez de chaussée, Salle 4

DOHMEN Thomas () *

Behavior seminar

Le 08/06/2017 de 12:00:00 à 13:00:00

Bâtiment A, Rez de chaussée, Salle 4

JACQUEMET Nicolas () How do hospital-specialty characteristics influence health system responsiveness? An empirical evaluation of in-patient care in the Italian Region of Emilia-Romagna

Behavior seminar

Le 01/06/2017 de 12:00:00 à 13:00:00

Bâtiment A, Rez de chaussée, Salle 4

RENOU Ludovic () *; () ;

La séance est annulée

Behavior seminar

Le 18/05/2017 de 12:00:00 à 13:00:00

Bâtiment A, Rez de chaussée, Salle 4

Séance reportée (initialement N. prenom ... () *

Behavior seminar

Le 04/05/2017 de 12:00:00 à 13:00:00

Bâtiment A, Rez de chaussée, Salle 4

GALLO Edoardo (Université de Cambridge) Financial Contagion in Networks: A Market Experiment

Syngjoo Choi & Galloy Brian Wallace

Behavior seminar

Le 27/04/2017 de 12:00:00 à 13:00:00

Bâtiment A, Rez de chaussée, Salle 4


Abstract. This paper considers a stationary model of inventory management in a rich setting in which unsold units carry over, in contrast with the full depreciation of unsold units that is implemented in laboratory studies of the newsvendor problem. The model permits an array of costs associated with restocking, understocking, depreciation, financing, and holding inventories. The extra dimensions make it possible to hold the optimal inventory constant, while adjusting parameters that change the frequency of stockouts and the risk associated with storage and depreciation. This framework facilitates an investigation of factors that influence the nature and severity of behavioral biases observed in simpler newsvendor settings. Optimal inventory decisions are derived and tested with a laboratory experiment. We consider four main questions in the inventory literature: the “pull-to-center” effect, the “recency” effect, the effect of increased up-front costs, and the effect of risk aversion.

BOULU-RESHEF Béatrice (Université de Cambridge) Inventory Management with Carryover in a Laboratory Setting: Going Beyond the Newsvendor Paradigm

Charles A. Holt (University of Virginia)

Behavior seminar

Le 20/04/2017 de 12:00:00 à 13:00:00

Bâtiment A, Rez de chaussée, Salle 4


An often, but not widely appreciated assumption underlying theoretical predictions of optimal insurance is that marginal utility of consumption evaluated at any given income level remains constant whether the individual is well or ill. However, if utility of consumption is increased or decreased by ill health, this will have an effect on the theoretical predictions of optimal choice of health insurance. In addition to the question of optimal insurance, the issue of state dependence also relates to the validity of asking patients (and not citizens) for their valuations of health improvements in stated preference studies when these are to guide the resource allocation of communal funds. To the extent that patients are in poor health and there is significant negative or positive state dependence, the marginal utility of income will not be equivalent to that of the average taxpayer, and thus, derivation of societal net benefits will be biased. Moreover, negative state dependence could be used as a justification for transferring resources from non-healthy/disabled to health individuals. Hence, the presence of state dependence has methodological as well as policy relevance. Despite the importance of the subject, very few studies have been conducted. These studies show contradictory evidence. The approach presented in this paper mimics a simple insurance market, where individuals can self-insure across two periods of time (one in which the person is in good health and one in which the person is in poorer health). The health states are presented as certain events, thus eliminating any problems associated with cognition and probabilities. The focus of this study is not to derive an estimate of state dependence, but to determine whether or not individuals exhibit positive or negative state dependence when confronted with a short term reduction in health and to qualify whether the severity of a given health state affects the sign of state dependence.

GYRD-HANSEN Dorte () A stated preference approach to assess whether health status impacts on marginal utility of consumption


Texte intégral

Behavior seminar

Le 30/03/2017 de 12:00:00 à 13:00:00

Bâtiment A, Rez de chaussée, Salle 4


Cesarean section rates are historically high in Canada as they are across the OECD. We investigate the roles of physician incentives and spillover effects of a major research trial on breech deliveries. We find that doubling the relative compensation of C-sections over vaginal deliveries increases the probability of a C-section by at most four percentage points. We also show that the publication of the “Hannah Term Breech Trial” had spillover effects on the C-section rate for vertex births. The secular increase in the C-section rate is largely accounted for by observable characteristics of births, mothers and health care systems.

STABILE Mark () Explaining the Rise in C-sections: the contributions of physician incentives and research spillovers

Behavior seminar

Le 23/03/2017 de 12:00:00 à 13:00:00

Bâtiment A, Rez de chaussée, Salle 4


We conduct an “artefactual” field experiment to incorporate different risk taking measures within the Innovation Panel (IP) of the UK Household Longitudinal Survey (UKHLS). We randomly allocate to an experimental module a nationally representative sample of 661 adult respondents to the IP Wave 6 (IP6) who complete the incentive-compatible tasks by Holt and Laury (2002) and by Binswanger (1980) and Eckel and Grossman (2008). They also respond to the survey questions by Dohmen et al. (2011) for self-reported willingness to take risks in general, in finance, and in health. One year later (IP7) the same measures are repeated for 411 of these respondents. This design allows us to systematically test the validity of the measures along three dimensions. First, we look at cross-validity by investigating their associations at one point in time. Second, we look at temporal stability by comparing the responses between IP6 and IP7. Third, we look at external validity by considering a range of health and financial behaviors. Concerning cross validity, we find evidence that the different risk taking measures correlate and map into each other, although only imperfectly. When the analysis accounts for individual observed heterogeneity and for the appropriate degree of nonlinearity, the experimental measures of risk aversion and the self-reported willingness to take risks predict each other. About temporal stability, we find significant persistence of the measures over time. Finally, we find mixed evidence concerning external validity, especially on health and nutrition behaviors.

GALIZZI Matteo (London School of Economics) Temporal stability, cross-validity, and external validity of risk-taking measures: experimental evidence from a UK representative sample

Behavior seminar

Le 16/03/2017 de 12:00:00 à 13:00:00

Bâtiment A, Rez de chaussée, Salle 4

BOMMIER Antoine (London School of Economics) Empirical identification of time preferences: Theory and an illustration using convex time budgets

Bruno Lanz

Behavior seminar

Le 09/03/2017 de 12:00:00 à 13:00:00

Bâtiment A, Rez de chaussée, Salle 4


In many different contexts, connected candidates are more likely to be hired or promoted than unconnected ones. This may be due to favoritism or better information on candidates' abilities. Attempts at identifiying both effects have generally relied on productivity measures collected after hiring. In this paper, we develop a new method to identify favors and information from data on hiring. Under natural assumptions, we show that observable characteristics have a lower impact on the probability to be hired for connected candidates and that this reduction precisely captures the information effect. We then show how to recover biases due to favors from overall shifts in hiring probabilities. We apply this new method on data on academic promotions in Spain. We find weak evidence of information effects and strong evidence of favoritism. These results are consistent with those obtained from later productivities.

BRAMOULLÉ Yann () Hiring through Networks: Favors or Information

kenan Huremovic

Behavior seminar

Le 02/03/2017 de 12:00:00 à 13:00:00

Bâtiment A, Rez de chaussée, Salle 4


This study estimates the effect of a targeted early childhood intervention program on global and experienced measures of maternal well-being utilizing a randomized controlled trial design. The primary aim of the intervention is to improve children’s school readiness skills by working directly with parents to improve their knowledge of child development and parenting behavior. One potential externality of the program is well-being benefits for parents given its direct focus on improving parental coping, self-efficacy, and problem solving skills, as well as generating an indirect effect on parental well-being by targeting child developmental problems. Participants from a socio-economically disadvantaged community are randomly assigned during pregnancy to an intensive 5-year home visiting parenting program or a control group. We estimate and compare treatment effects on multiple measures of global and experienced well-being using permutation testing to account for small sample size and a stepdown procedure to account for multiple testing. The intervention has no impact on global well-being as measured by life satisfaction and parenting stress or experienced negative affect using episodic reports derived from the Day Reconstruction Method (DRM). Treatment effects are observed on measures of experienced positive affect derived from the DRM and a measure of mood yesterday. The limited treatment effects suggest that early intervention programs may produce some improvements in experienced positive well-being, but no effects on negative aspects of well-being. Different findings across measures may result as experienced measures of well-being avoid the cognitive biases that impinge upon global assessments.

Doyle Joseph () Can Early Intervention Improve Maternal Well-being? Evidence from a Randomized Controlled Trial

Behavior seminar

Le 02/02/2017 de 12:00:00 à 13:00:00

Bâtiment A, Rez de chaussée, Salle 4

ELSNER Benjamin () Simultaneous Treatment Bias in the Estimation of Peer Effects

Ingo ISPHORDING

Behavior seminar

Le 26/01/2017 de 12:00:00 à 13:00:00

CLARK Damon () What makes a successful life? The role of financial deprivation

Behavior seminar

Le 19/01/2017 de 12:00:00 à 13:00:00

Bâtiment A, Rez de chaussée, Salle 4

CALVO-PARDO Hector (University of Southampton) Informative social interactions

Luc Arrondel, Chryssi Giannitsarou and Michael Haliassos

Texte intégral

Behavior seminar

Le 12/01/2017 de 12:00:00 à 13:00:00

ERISEN Cengiz () Tolerance and perceived threat toward Muslim immigrants in Germany and the Netherlands

Cigdem Kentmen-Cin

Texte intégral

Behavior seminar

Le 05/01/2017 de 12:00:00 à 13:00:00

YIN Rémi () Impulsiveness, Situational Influences and Health Behaviors. Evidence from Evening and Night Work

Behavior seminar

Le 15/12/2016 de 12:00:00 à 13:00:00

WEBER Roberto (University of Zurich) Do gender preference gaps impact policy outcomes?

Eva Ranehill

Behavior seminar

Le 08/12/2016 de 12:00:00 à 13:00:00

Bâtiment A, Rez de chaussée, Salle 4

FOURNIER Patrick (University of Zurich) The Human Negativity Bias: A Comparative Experimental Study

Behavior seminar

Le 01/12/2016 de 12:00:00 à 13:00:00

LASLIER Jean-François (University of Zurich) Preferences for voting rules

Behavior seminar

Le 24/11/2016 de 12:00:00 à 13:00:00

PONTHIÈRE Grégory (University of Zurich) Missing poor, selection bias and counterfactuals

Mathieu Lefebvre et Pierre Pestieau

Behavior seminar

Le 17/11/2016 de 12:00:00 à 13:00:00

ETILÉ Fabrice () Childhood circumstances and the heterogeneity in the psychological response to major life events

Behavior seminar

Le 03/11/2016 de 12:00:00 à 13:00:00

LEPINTEUR Anthony () The Shorter Workweek and Worker Wellbeing: Evidence from Portugal and France


Texte intégral

Behavior seminar

Le 13/10/2016 de 12:00:00 à 13:00:00

FORTIN Bernard () Gender peer effects heterogeneity in obesity

Rokhaya Dieye

Texte intégral

Behavior seminar

Le 06/10/2016 de 12:00:00 à 13:00:00

SENIK Claudia () Gender and socialism

Behavior seminar

Le 29/09/2016 de 12:00:00 à 13:00:00

Campus jourdan,Bâtiment A, Rez de chaussée, Salle 4

ARRONDEL Luc () Pourquoi la demande d'actions baisse-t-elle pendant la crise : Préférences ou anticipations ?

Behavior seminar

Le 22/09/2016 de 12:00:00 à 13:00:00

Campus jourdan,Bâtiment A, Rez de chaussée, Salle 4

HERRERA ARAUJO Daniel () QALY-fying VSL: Finding a link between Value per Statistical Life and QALY

Behavior seminar

Le 00/00/0000 de 00:00:00 à 00:00:00

Bâtiment A, Rez de chaussée, Salle 4

STRINGER Eliza-Jane, NEBOUT Antoine () *

coordonnées

archives

agenda

site web