|
on Experimental Economics |
|
Issue of 2024‒07‒15
34 papers chosen by |
| By: | Detemple, Julian |
| Abstract: | Experiments are an important tool in economic research. However, it is unclear to which extent the control of experiments extends to the perceptions subjects form of such experimental decision situations. This paper is the first to explicitly elicit perceptions of the dictator and trust game and shows that there is substantial heterogeneity in how subjects perceive the same game. Moreover, game perceptions depend not only on the game itself but also on the order of games (i.e., the broader experimental context in which the game is embedded) and the subject herself. This highlights that the control of experiments does not necessarily extend to game perceptions. The paper also demonstrates that perceptions are correlated with game behavior and moderate the relationship between game behavior and field behavior, thereby underscoring the importance and relevance of game perceptions for economic research. |
| Keywords: | dictator game, trust game, game perceptions |
| JEL: | C90 D01 D91 |
| Date: | 2024 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:safewp:297999&r= |
| By: | Maxine Bonneau (Imperial College London); Tanya O'Garra (Imperial College London); Praveen Kujal (Middlesex University Business School and Chapman University) |
| Abstract: | We examine how the moral (or neutral) content of an issue influences the tendency to falsify attitudes, given varying social and monetary incentives to engage in such ‘preference falsification’. We conduct an incentivized two stage online experimental study where, in a prior first stage, attitude strength over moral and neutral issues is elicited. Then, in the second stage participants in groups of ten were asked to express their preferences regarding the moral or neutral issues for each possible combination of supporters and opposers in their group, each associated with varying monetary payoffs. More than half of the participants falsify their preferences between the two phases for both moral and neutral frames. The rate is significantly lower for the moral (vs neutral) issues. Participants’ average monetary cost to avoid falsifying preferences is higher in moral treatments, and increases with the level of attitude strength. |
| Keywords: | Morality, preference falsification, incentives |
| JEL: | C9 C90 C92 D02 D7 D74 |
| Date: | 2024 |
| URL: | https://d.repec.org/n?u=RePEc:chu:wpaper:24-11&r= |
| By: | Irene Maria Buso; Daniela Di Cagno; Werner Gueth; Lorenzo Spadoni |
| Abstract: | We investigate, both theoretically and experimentally, an institutional mechanism designed to enhance cooperation. In this mechanism, contributors have the option to voluntarily contribute to the public good and decide whether to join a (sub)group where partners equally share the contribution cost. Theoretically, stable cost-sharing partnerships enhance efficiency since their partners fully contribute, while outsiders would free-ride. Our data reveal that individual joining and contribution behaviors do not always align with benchmark predictions: partnerships are not always formed, and when they are, they are not always of the optimal size; partners often contribute less than maximally, and outsiders more than minimally. Nonetheless, we document systematic evidence of partnership formation and significantly improved provision of public goods across rounds. |
| Keywords: | Public Good, Group Formation, Group Size, Experiments |
| JEL: | C92 H41 D85 |
| Date: | 2024–06–01 |
| URL: | https://d.repec.org/n?u=RePEc:pie:dsedps:2024/309&r= |
| By: | Elisa Stumpf; Silke Uebelmesser |
| Abstract: | We study beliefs about wealth inequality and preferences for wealth redistribution. For this, we conduct a large-scale online survey in Germany. First, we analyze how well participants are informed about the German wealth distribution and their position in it. Second, we investigate how preferences for wealth redistribution are affected by an information experiment. One treatment group receives information about the shape of the German wealth distribution, while another treatment group receives information about their position in this distribution. Using a multidimensional approach to measure preferences for wealth redistribution, we find no significant average treatment effect for either treatment in the full sample, although those who overestimate their position reduce their aversion to inequality after learning their position, while those who underestimate their position are more likely to agree that anyone can become successful through hard work. We employ a data-driven approach to further investigate heterogeneity in treatment effects and present evidence that younger participants decrease their support for redistribution after learning about the shape of the wealth distribution. In contrast, older participants decrease their support after learning their position in the distribution. |
| Keywords: | wealth distribution, preferences for redistribution, inequality, survey experiment, information provision |
| JEL: | C90 D31 D63 D83 |
| Date: | 2024 |
| URL: | https://d.repec.org/n?u=RePEc:ces:ceswps:_11126&r= |
| By: | Yosuke Hashidate (Sophia University); Tetsuya Kawamura (Tezukayama University); Fabrice Le Lec (Univ. Lille, CNRS, IESEG School of Management, UMR 9221 – LEM – Lille Economie Management, F-59000 Lille, France); Yusuke Osaki (Waseda University); Benoît Tarroux (Université Lumière Lyon 2, CNRS, Université Jean-Monnet Saint-Etienne, emlyon business school, GATE, 69007, Lyon, France) |
| Abstract: | This article investigates impure motivations in social preferences through an experiment in which participants choose menus of social allocations (i.e. allocations of gains between themselves and another participant). Menu choices reveal the presence of impure motivations: according to a parsimonious theoretical model, negative motivations (e.g. guilt) will imply a preference for smaller menus, whereas positive ones (e.g. pride) a preference for larger sets. By varying the level of publicity of within-menu choice, we can also observe the importance of self- and social image. Data from France and Japan show unambiguously an important impact of impure motivations on preferences on menus, with the suggestion that negative ones exceed positive ones. |
| Keywords: | Social preference; menu preference; impure motivation; warm glow; guilt; shame; pride; self-esteem |
| JEL: | C91 D01 D64 D91 |
| Date: | 2024 |
| URL: | https://d.repec.org/n?u=RePEc:gat:wpaper:2406&r= |
| By: | Bellani, Luna; Berriochoa, Kattalina; Kapteina, Mark; Schwerdt, Guido |
| Abstract: | We study the effects of information on attitudes towards inheritance taxation using survey experiments fielded in Germany. We show that information about tax allowances increases demand for higher taxes and shifts public opinion from favoring abolition to supporting the tax. Effects are primarily due to a prevalent underestimation of tax allowances and the alteration of people's expectations of being affected by such taxes. In contrast, information highlighting the increasing proportion of inherited wealth only negligibly affects policy demand. Our results suggest that pocketbook motives and misinformation may contribute to explaining the paradox of limited demand for inheritance taxation despite growing inequality concerns. |
| Keywords: | capital taxation, equality of opportunity, inheritance tax, information, randomized experiment |
| JEL: | H52 I22 D72 D83 |
| Date: | 2024 |
| URL: | https://d.repec.org/n?u=RePEc:zbw:cexwps:299225&r= |
| By: | Horton, John J. (MIT); Johari, Ramesh (Stanford University); Kircher, Philipp (Cornell University) |
| Abstract: | In a labor market model with cheap talk, employers can send messages about their willingness to pay for higher-ability workers, which job-seekers can use to direct their search and tailor their wage bid. Introducing such messages leads – under certain conditions – to an informative separating equilibrium that affects the number of applications, types of applications, and wage bids across rms. This model is used to interpret an experiment conducted in a large online labor market: employers were given the opportunity to state their relative willingness to pay for more experienced workers, and workers can easily condition their search on this information. Preferences were collected for all employers but only treated employers had their signal revealed to job-seekers. In response to revelation of the cheap talk signal, job-seekers targeted their applications to employers of the right "type, " and they tailored their wage bids, affecting who was matched to whom and at what wage. The treatment increased measures of match quality through better sorting, illustrating the power of cheap talk for talent matching. |
| Keywords: | sorting, cheap-talk, gig-economy, freelancer, field-experiment, online job search platform |
| JEL: | J64 D83 C87 |
| Date: | 2024–06 |
| URL: | https://d.repec.org/n?u=RePEc:iza:izadps:dp17052&r= |
| By: | Bernhard Kasberger; Simon Martin; Hans-Theo Normann; Tobias Werner |
| Abstract: | Algorithms play an increasingly important role in economic situations. These situations are often strategic, where the artificial intelligence may or may not be cooperative. We study the deter-minants and forms of algorithmic cooperation in the infinitely repeated prisoner’s dilemma. We run a sequence of computational experiments, accompanied by additional repeated prisoner’s dilemma games played by humans in the lab. We find that the same factors that increase human cooperation largely also determine the cooperation rates of algorithms. However, algorithms tend to play different strategies than humans. Algorithms cooperate less than humans when cooperation is very risky or not incentive-compatible. |
| Keywords: | artificial intelligence, cooperation, large language models, Q-learning, repeated prisoner’s dilemma |
| JEL: | C72 C73 C92 D83 |
| Date: | 2024 |
| URL: | https://d.repec.org/n?u=RePEc:ces:ceswps:_11124&r= |
| By: | Marek Kapera |
| Abstract: | In this article I study whether the interim preferences of the consumer can be expected to converge to their real preferences in the process of preference discovery. I construct a subjective expected utility model of the consumer, where the uncertainty results from the imperfect knowledge of their own preferences. This uncertainty is partially resolved by experimental consumption. Under the assumption that the subjective probability of the consumer satisfies learning monotonicity, I identify the equivalent conditions for the consumer to experiment. My results show that the interim preferences never fully converge to the real preferences of the consumer. Instead, the preference discovery either terminates, meaning that the consumer ceases to experiment, or only experiments within some neighborhood of the best currently known alternative, and never sufficiently explores their preferences. |
| Keywords: | Taste uncertainty, Preference discovery, Learning through consumption, Conditional preferences, Experimental preferences |
| JEL: | D11 D83 D91 |
| Date: | 2024–03 |
| URL: | http://d.repec.org/n?u=RePEc:sgh:kaewps:2024098&r= |
| By: | Niklas Klarnskou; Philippos Louis; Wouter Passtoors |
| Abstract: | We use a lab experiment to examine the effect of feedback on bidder behavior in procurement e-auctions. We compare ‘Rank-only’ to ‘show lead bid’ feedback, two regimes applied frequently by procurement professionals. A choice among the two is often based on rules-of-thumb that rely on initial ‘bid compression’, i.e. the spread of the bids submitted pre-auction. The use of such criteria finds no support in existing economic theory. A common assumption in theoretical auction models is that bidders face no opportunity cost from participating in a dynamic auction. This may not hold in situations where the expected value of a contract does not justify a long-time commitment to the bidding process on the part of bidders. In our experiment participants face the choice of remaining active in an auction vs. exiting and being rewarded with a diminishing outside option. Showing the lead bid accelerated bidders’ learning. In the presence of opportunity costs, this can lead to substantially different outcomes conditional on the initial bid compression. With low bid compression, the bidder with the lowest cost wins more frequently, enhancing efficiency, but faces reduced competition by the others, which hurts the buyer’s potential outcome. The opposite is true when bid compression is high. Rank only feedback achieved similar overall levels of efficiency, with higher benefits for the buyer. Crucially, these outcomes are not as sensitive to initial bid compression as in the case of ‘show lead bid’ feedback. A discouragement effect emerging in the ‘Show-lead-bid’, but not in the ‘Rank-only’ regime can explain these results. |
| Keywords: | Procurement auctions, feedback, opportunity cost, competition, bid compression, discouragement effect, lab experiment |
| JEL: | C92 D44 D83 |
| Date: | 2024–06–14 |
| URL: | https://d.repec.org/n?u=RePEc:ucy:cypeua:04-2024&r= |
| By: | Marion Dupoux; Benjamin Ouvrard |
| Abstract: | Pigouvian taxes are often unpopular among the general public. We test the effectiveness of social information provision to improve support for such taxes. In a lab experiment that involves a market game with externalities, we provide subjects with information about other participants’ personal opinions about the “right thing to do” (voting, or not, for tax implementation). To gain insight into the causal mechanism by which social information impacts subjects’ votes, we also elicit personal, normative, and positive beliefs. Our findings demonstrate a causal effect of social information provision on subjects’ support for the tax, and that subjects’ changes in beliefs is a causal mechanism through which this increased support for the tax is made possible. We also show that subjects who experience the tax are more likely to support it, and that the tax significantly reduces externalities in the game. We therefore highlight the pivotal role of beliefs in voting behaviors and the acceptability of Pigouvian taxes. |
| Keywords: | Beliefs, Externality Game, Pigouvian Taxes, Social Information, Voting Behavior |
| JEL: | C92 D91 H23 |
| Date: | 2023–05 |
| URL: | https://d.repec.org/n?u=RePEc:gbl:wpaper:2024-05&r= |
| By: | Henning Hermes; Philipp Lergetporer; Fabian Mierisch; Frauke Peter; Simon Wiederhold |
| Abstract: | We provide the first causal evidence of discrimination against migrants seeking child care. We send emails from fictitious parents to > 18, 000 early child care centers across Germany, asking if there is a slot available and how to apply. Randomly varying names to signal migration background, we find that migrants receive 4.4 percentage points fewer responses. Responses to migrants also contain substantially fewer slot offers, are shorter, and less encouraging. Exploring channels, discrimination against migrants does not differ by the perceived educational background of the email sender. However, it does differ by regional characteristics, being stronger in areas with lower shares of migrants in child care, higher right-wing vote shares, and lower financial resources. Discrimination on the child care market likely perpetuates existing inequalities of opportunities for disadvantaged children. |
| Keywords: | child care, discrimination, information provision, inequality, field experiment |
| Date: | 2023–04 |
| URL: | https://d.repec.org/n?u=RePEc:bav:wpaper:225_discriminationchildcaremarketfieldexperiment&r= |
| By: | Fuhrmann-Riebel, Hanna (University of East Anglia); D'Exelle, Ben (University of East Anglia); López Vargas, Kristian (University of California, Santa Cruz); Tonke, Sebastian (Max Planck Institute for Research on Collective Goods); Verschoor, Arjan (University of East Anglia) |
| Abstract: | Tackling environmental pollution requires a permanent change in regular, repeated behavior of households. Bringing about change in such behavior may require interventions that are not limited to a single point in time, yet little evidence exists on how frequently we need to target households to initiate behavioral change and to form new habits in regular pro-environmental behavior. To fill this gap, we investigate the impact of mobile text reminders on households' recycling behavior in urban Peru, by randomly varying the frequency of reminders over a nine-week treatment period. We find that reminders increase both the likelihood that households start to recycle, and the frequency of recycling among households that already recycled before the intervention. The effects are stronger if reminders are repeated over a longer period. Our findings suggest that low-cost mobile text reminders can support repeated pro-environmental behavior, and that some repetition may be needed to maximize their effectiveness. |
| Keywords: | recycling, habit formation, limited attention, reminders, Peru |
| JEL: | C93 D83 D90 D91 Q53 Q58 |
| Date: | 2024–05 |
| URL: | https://d.repec.org/n?u=RePEc:iza:izadps:dp17013&r= |
| By: | Francesco Fabbri; Giulio Principi; Lorenzo Stanca |
| Abstract: | We represent preferences that exhibit absolute or relative attitudes towards ambiguity without assuming convexity of preferences. Our analysis is motivated by the recent experimental evidence by Baillon and Placido (2019) indicating that ambiguity becomes more tolerable as individuals are better off overall. Decreasing absolute ambiguity aversion is characterized by constant superadditive certainty equivalents and admits an act-dependent variational representation (Maccheroni et al., 2006). Decreasing relative ambiguity aversion relates to positive superhomogeneity and admits an act-dependent confidence preference representation (Chateauneuf and Faro, 2009). We apply our characterizations to retrieve a classic risk sharing result on the efficiency of trade and subjective beliefs of the individuals (Rigotti et al., 2008). |
| Date: | 2024–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2406.01343&r= |
| By: | Franziska Heinicke; Wladislaw Mill; Henrik Orzen |
| Abstract: | This paper addresses bargaining with a nuclear option. Players with access to such an option have the power to cause enormous damage to their negotiation partners. Figurative nuclear options are available in many important real-world settings and, being the ultimate threat, are often seen as effective in putting maximal pressure on the other party and as possibly efficiency- improving. On the other hand, since going nuclear is typically also very costly to the nuclear-option holder herself, the credibility of a nuclear threat may be questionable. We report the results from unstructured one-shot bargaining experiments and examine to what extent a nuclear option increases bargaining power, makes agreements more likely, and affects efficiency. We find that nuclear-option holders do not generally benefit while the other party is worse off compared to a baseline setting, particularly when the other party is intrinsically—i.e., save for the nuclear threat itself—in a strong position. Furthermore, the nuclear option increases the number of negotiations that end in agreements that are not efficiency-improving. Thus, the presence of a nuclear option in our bargaining setting is overall detrimental. |
| Keywords: | Nuclear option; Power asymmetry; Bargaining; Experiment |
| JEL: | C91 D91 D01 C78 J52 |
| Date: | 2024–06 |
| URL: | https://d.repec.org/n?u=RePEc:bon:boncrc:crctr224_2024_559&r= |
| By: | Md Tawhidul Islam (School of Economics and Management, Kochi University of Technology); Kenta Tanaka (Faculty of Economics, Musashi University, Tokyo, Japan); Koji Kotani (School of Economics and Management, Kochi University of Technology) |
| Abstract: | Public bads prevention problems, such as climate change, require people to cooperate above a certain threshold which is ambiguous and varies in many situations. In that case, people conjecture and share some information about the threshold. However, little is known about how sharing such information affects people to cooperate. We experimentally examine how people’s cooperative choices are influenced by ambiguity and sharing information about the conjectures in public bads prevention, hypothesizing that sharing the information does not necessarily contribute to cooperation. We conduct the laboratory experiment with 400 subjects under five treatments each of which differs in ambiguity as well as in presence or absence of sharing the information. We find that (i) the percentages of cooperative choices are nonmonotonic, decreasing and then increasing over ambiguity levels and (ii) sharing the information tends to uniformly discourage cooperation, and the negative impact becomes prominent as the ambiguity levels rise. The result demonstrates an adverse effect between sharing information and threshold ambiguity on cooperation, being in sharp contrast with the literature. Overall, this study suggests that how or what information is shared among people should be carefully reconsidered for resolving any public bads problem involving threshold ambiguity, as everybody is able to easily publicize her conjectures during an era of digital democracy. |
| Keywords: | Threshold ambiguity, sharing information, public bads, cooperation |
| Date: | 2024–06 |
| URL: | https://d.repec.org/n?u=RePEc:kch:wpaper:sdes-2024-3&r= |
| By: | Lebedinski, Lara (Institute of Economic Sciences, Belgrade); Carneiro, Pedro (University College London); Urzua, Tamara Arnold (World Bank); Perng, Julie (University of Göttingen); Boudet, Ana Maria Munoz (World Bank); Sosa, Katia Herrera (World Bank) |
| Abstract: | In this paper we evaluate experimentally the impacts of a parenting program delivered virtually to 1, 431 families with children ages 2-6 years old in Serbia. We compare two program modalities to a control group. In the first (standard) modality, only the main caregiver -mainly mothers- participates in the training, while in the second (plus), two caregivers -mothers and fathers- participate in the training. We find that the standard intervention has a positive effect of 0.28 standard deviations on parent-reported child development outcomes, but no such impact is found for the plus modality. We fail to observe statistically significant impacts of either treatment across most variables measuring parental behaviors and home environments, with two exceptions: parents in the standard treatment became more conscious about their child's learning, while parents in the plus modality became less likely to use physical punishment to discipline children. |
| Keywords: | parenting program, child development, virtual delivery |
| JEL: | I10 |
| Date: | 2024–05 |
| URL: | https://d.repec.org/n?u=RePEc:iza:izadps:dp17028&r= |
| By: | Motta Café, Renata; Yarygina, Anastasiya; Escalante, Lisseth |
| Abstract: | A large body of research has estimated the effects on tax collection of informing taxpayers of their obligations. This paper examines the effects on voluntary tax payments of providing taxpayers with information on their obligations that is collected through massive information cross-checking rather than traditional auditing. This information is inexpensive to collect, but may not be sufficiently accurate to promote taxpayer compliance. We conducted a randomized controlled trial to evaluate the effectiveness of self-regularization interventions on tax compliance among simplified tax regime firms. We examine the treatments that use this information in three different ways: messaging, providing tax compliance manuals, and assisted regularization, in which tax auditors assist taxpayers in the self-regularization process. We find that assisted regularization increased the reported monthly income by 20 percent (US$1, 160), which also nearly closed the tax evasion gap and reduced inconsistencies in tax declarations by 67 percent (37-point reduction). The manual and message interventions had positive, albeit smaller, effects. While the assisted regularization intervention had the largest effects, a cost-effectiveness analysis reveals that this intervention is not optimal for smaller firms because of the costs incurred by the tax authority. No effects were observed on firms' tax compliance in declarations filed in the post- intervention periods. |
| Keywords: | public finance;State capacity;Tax evasion;electronic tax audit;tax self-regularization;Simples Nacional |
| JEL: | H26 H32 H71 H83 |
| Date: | 2024–06 |
| URL: | https://d.repec.org/n?u=RePEc:idb:brikps:13603&r= |
| By: | Subhasish M. Chowdhury (Department of Economics, University of Sheffield, Sheffield S1 4DT, UK); Dan Kovenock (Economic Science Institute, Chapman University, Orange, CA 92866, USA); Anwesha Mukherjee (Department of Economics, University of Reading, Reading RG6 6AA, UK) |
| Abstract: | Contests are situations in which agents compete by irreversibly expending costly resources in an attempt to win a prize. Due to their applications in conflict, rent-seeking, organizational incentives, sports, litigation, and political campaigns, contests are widely applied in the social sciences. In this survey we summarize some main results and recent developments of experimental studies in contest theory. We also point out their broader applications in the social sciences. |
| Keywords: | Contest; Experiment; Conflict; Survey |
| JEL: | C91 C92 D72 D74 |
| Date: | 2024–06 |
| URL: | https://d.repec.org/n?u=RePEc:shf:wpaper:2024004&r= |
| By: | Lund, Crick (King's College London); Orkin, Kate (University of Oxford); Witte, Marc J. (Vrije Universiteit Amsterdam); Walker, John (University of Oxford); Davies, Thandi (University of Cape Town); Haushofer, Johannes (Stockholm University); Murray, Sarah (John Hopkins University, Baltimore); Bass, Judy (Johns Hopkins University); Murray, Laura (John Hopkins University, Baltimore); Tol, Wietse (University of Copenhagen); Patel, Vikram (Harvard University) |
| Abstract: | Mental health conditions are prevalent but rarely treated in low- and middle-income countries (LMICs). Little is known about how these conditions affect economic participation. This paper shows that treating mental health conditions substantially improves recipients' capacity to work in these contexts. First, we perform a systematic review and meta-analysis of all randomized controlled trials (RCTs) ever conducted that evaluate treatments for mental ill-health and measure economic outcomes in LMICs. On average, treating common mental disorders like depression with psychotherapy improves an aggregate of labor market outcomes made up of employment, time spent working, capacity to work and job search by 0.16 standard deviations. Treating severe mental disorders, like schizophrenia, improves the aggregate by 0.30 standard deviations, but effects are noisily estimated. Second, we build a new dataset, pooling all available microdata from RCTs using the most common trial design: studies of psychotherapy in LMICs that treated depression and measured days participants were unable to work in the past month. We observe comparable treatment effects on mental health and work outcomes in this sub-sample of highly similar studies. We also show evidence consistent with mental health being the mechanism through which psychotherapy improves work outcomes. |
| Keywords: | labor, development, human capital, mental health, psychotherapy |
| JEL: | D9 I14 O1 J24 |
| Date: | 2024–06 |
| URL: | https://d.repec.org/n?u=RePEc:iza:izadps:dp17071&r= |
| By: | David MASCLET (Univ Rennes, CNRS, CREM – UMR6211, F-35000 Rennes France and CIRANO, Montreal, Canada); David L. DICKINSON (Appalachian State University, IZA, ESI) |
| Abstract: | We present a theoretical framework of individual-decision making that incorporates both moral motivations and social influence into the utility function. The main idea of the paper is that individuals face a trade-off between their material individual interests and their desire to follow moral obligation. In our model, we assume that moral motivation is weak or conditional in the sense that it may be influenced by others’ actions. Specifically, in our framework one’s moral obligation is a combination of two main components: an autonomous component and a social component that captures the influence of others. Our theoretical framework is able to explain many stylized results commonly observed in the literature and suggests a different mechanism to explain economic behavior. |
| Keywords: | Behavioral Economics, Ethical Decision Making, Fairness |
| JEL: | C9 D9 D91 |
| Date: | 2024–06 |
| URL: | https://d.repec.org/n?u=RePEc:tut:cremwp:2024-04&r= |
| By: | Kene Boun My; Julien Jacob; Mathieu Lefebvre |
| Abstract: | We propose a new theoretical framework to analyze the incentives provided by different allocations of liability in the case of (semi)autonomous devices which are a source of risk of accident. We consider three key agents, an AI provider (scientist), a producer and a consumer, and look at the effect of different rules of sharing liability on the decision making of each type of agent. In addition we test the theoretical predictions in an original lab experiment. We show that liability on the scientist and the producer is efficient in reducing their misbehaviors. We also find that liability on the consumer increases her incentives to control the risk of an accident (in case of a semi-autonomous device). However, the absence of consumer’s control (full autonomous device) and liability decreases the consumer’s propensity to buy the good. We complete our study by making a social welfare analysis. It highlights the importance of letting the producer liable in order to provide the consumer with confidence in the technology, especially in the case of a full autonomy of the good. |
| Keywords: | AI, Liability Sharing Rules, asymmetric information, lab experiment. |
| JEL: | C91 D82 K13 K32 |
| Date: | 2024 |
| URL: | https://d.repec.org/n?u=RePEc:ulp:sbbeta:2024-24&r= |
| By: | Serge Garcia; Katrin Erdlenbruch; Boniface Derrick Mbarga |
| Abstract: | This article investigates residential choice in flood-prone areas with attractive natural amenities. In a discrete choice experiment involving 472 French homeowners, we analyse the effects of flood risk information disclosure. Respondents make trade-offs between house characteristics, amenities and location in flood-prone areas, with two information treatments about the consequences of flooding and protection measures. We also examine the influence of existing information tools. The econometric models reveal a general aversion to flood-prone areas and a negative effect of information about the consequences of flooding. Buyer-tenant information influences the decision to leave flood-prone areas, while zoning influences the decision to stay. |
| Keywords: | choice experiment; flood risk-amenity trade-off; information treatment; mixed logit; attribute willingness to pay; residential choice. |
| JEL: | Q54 Q51 C25 |
| Date: | 2024 |
| URL: | https://d.repec.org/n?u=RePEc:ulp:sbbeta:2024-22&r= |
| By: | Smith, E. Keith; Kolcava, Dennis |
| Abstract: | Public support is a crucial, often necessary, component of political feasibility for policy proposals. Conjoint experiments are commonly utilised to assess support for policies, particularly how public opinion varies by specific policy instruments, and packaged designs. Yet, while robust methodologies have been identified to compare relative levels of support between policy instruments, current strategies often limit substantive interpretations of the absolute level of support. Given the importance of majority support thresholds across democratic settings, identifying absolute levels of support can provide critical information when assessing the feasibility of policy proposals. Here, we empirically explore a simple methodological advancement -- how do estimations of relative and absolute levels of support vary by discrete choice, proposal vote and binary ratings response methods? Drawing upon evaluations of support for carbon taxation in the United States as a case study, we find that similar levels of relative support can be found across response methods, but that the absolute level varies substantially by method and whether abstention is allowed. We further evaluate response methods by efficiency and consistency, to develop a set of recommendations towards utilising multiple response items to simultaneously assess relative and absolute levels of support for policies in conjoint-experimental designs. |
| Date: | 2024–06–06 |
| URL: | https://d.repec.org/n?u=RePEc:osf:osfxxx:837ws&r= |
| By: | Jingru Jia; Zehua Yuan; Junhao Pan; Paul McNamara; Deming Chen |
| Abstract: | When making decisions under uncertainty, individuals often deviate from rational behavior, which can be evaluated across three dimensions: risk preference, probability weighting, and loss aversion. Given the widespread use of large language models (LLMs) in decision-making processes, it is crucial to assess whether their behavior aligns with human norms and ethical expectations or exhibits potential biases. Several empirical studies have investigated the rationality and social behavior performance of LLMs, yet their internal decision-making tendencies and capabilities remain inadequately understood. This paper proposes a framework, grounded in behavioral economics, to evaluate the decision-making behaviors of LLMs. Through a multiple-choice-list experiment, we estimate the degree of risk preference, probability weighting, and loss aversion in a context-free setting for three commercial LLMs: ChatGPT-4.0-Turbo, Claude-3-Opus, and Gemini-1.0-pro. Our results reveal that LLMs generally exhibit patterns similar to humans, such as risk aversion and loss aversion, with a tendency to overweight small probabilities. However, there are significant variations in the degree to which these behaviors are expressed across different LLMs. We also explore their behavior when embedded with socio-demographic features, uncovering significant disparities. For instance, when modeled with attributes of sexual minority groups or physical disabilities, Claude-3-Opus displays increased risk aversion, leading to more conservative choices. These findings underscore the need for careful consideration of the ethical implications and potential biases in deploying LLMs in decision-making scenarios. Therefore, this study advocates for developing standards and guidelines to ensure that LLMs operate within ethical boundaries while enhancing their utility in complex decision-making environments. |
| Date: | 2024–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2406.05972&r= |
| By: | Pierpaolo Battigalli; Giovanni Di Bartolomeo; Stefano Papa |
| Abstract: | We examine the interplay between guilt aversion and inequality in decision-making, speci…cally in the context of a dictator game. Considering different initial allocations of the surplus, we investigate how the material opportunity cost of reducing inequality in‡uences sharing behavior, revealing complex relationship among guilt-driven choices and opportunity costs. |
| Keywords: | guilt aversion; psychological games; dictator games; second-order beliefs |
| JEL: | A13 C91 D01 D64 |
| Date: | 2024–06 |
| URL: | https://d.repec.org/n?u=RePEc:sap:wpaper:wp252&r= |
| By: | Ruoxuan Xiong; Alex Chin; Sean J. Taylor |
| Abstract: | We study the design and analysis of switchback experiments conducted on a single aggregate unit. The design problem is to partition the continuous time space into intervals and switch treatments between intervals, in order to minimize the estimation error of the treatment effect. We show that the estimation error depends on four factors: carryover effects, periodicity, serially correlated outcomes, and impacts from simultaneous experiments. We derive a rigorous bias-variance decomposition and show the tradeoffs of the estimation error from these factors. The decomposition provides three new insights in choosing a design: First, balancing the periodicity between treated and control intervals reduces the variance; second, switching less frequently reduces the bias from carryover effects while increasing the variance from correlated outcomes, and vice versa; third, randomizing interval start and end points reduces both bias and variance from simultaneous experiments. Combining these insights, we propose a new empirical Bayes design approach. This approach uses prior data and experiments for designing future experiments. We illustrate this approach using real data from a ride-sharing platform, yielding a design that reduces MSE by 33% compared to the status quo design used on the platform. |
| Date: | 2024–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2406.06768&r= |
| By: | Pablo Guillen; Rami Tabri; Edward Wang |
| Abstract: | We propose a modification of the University Admissions Centre (UAC) mechanism to allow preference lists to be submitted in batches until the applicant is matched to a seat. Batching eliminates truncation and thus recovers strategy-proofness, allowing for the clearinghouse to provide simple advice. The current UAC mechanism uses a constrained list, giving incentives to students to strategize. We test the efficiency of our modification in an individual decision-making matching experiment in which we compare the batched mechanism with the current mechanism, with and without advice. Results show that while the batched mechanism exhibits greater efficiency for student welfare, better advice is required to improve truth-telling and thus avoid suboptimal matches. |
| Date: | 2024–06 |
| URL: | https://d.repec.org/n?u=RePEc:syd:wpaper:2024-13&r= |
| By: | Federico A. Bugni; Mengsi Gao; Filip Obradovic; Amilcar Velez |
| Abstract: | Randomized controlled trials (RCTs) frequently utilize covariate-adaptive randomization (CAR) (e.g., stratified block randomization) and commonly suffer from imperfect compliance. This paper studies the identification and inference for the average treatment effect (ATE) and the average treatment effect on the treated (ATT) in such RCTs with a binary treatment. We first develop characterizations of the identified sets for both estimands. Since data are generally not i.i.d. under CAR, these characterizations do not follow from existing results. We then provide consistent estimators of the identified sets and asymptotically valid confidence intervals for the parameters. Our asymptotic analysis leads to concrete practical recommendations regarding how to estimate the treatment assignment probabilities that enter in estimated bounds. In the case of the ATE, using sample analog assignment frequencies is more efficient than using the true assignment probabilities. On the contrary, using the true assignment probabilities is preferable for the ATT. |
| Date: | 2024–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2406.08419&r= |
| By: | Can Celebi (University of Mannheim); Stefan Penczynski (School of Economics and Centre for Behavioural and Experimental Social Science, University of East Anglia) |
| Abstract: | In our study, we compare the classification capabilities of GPT-3.5 and GPT-4 with human annotators using text data from economic experiments. We analysed four text corpora, focusing on two domains: promises and strategic reasoning. Starting with prompts close to those given to human annotators, we subsequently explored alternative prompts to investigate the effect of varying classification instructions and degrees of background information on the models’ classification performance. Additionally, we varied the number of examples in a prompt (few-shot vs zero-shot) and the use of the zero-shot “Chain of Thought†prompting technique. Our findings show that GPT-4’s performance is comparable to human annotators, achieving accuracy levels near or over 90% in three tasks, and in the most challenging task of classifying strategic thinking in asymmetric coordination games, it reaches an accuracy level above 70%. |
| Keywords: | Text Classification, GPT, Strategic Thinking, Promises |
| Date: | 2024–06 |
| URL: | https://d.repec.org/n?u=RePEc:uea:wcbess:24-01&r= |
| By: | Regula HAENGGLI FRICKER; Weil Leopold |
| Abstract: | Legitimacy is the bedrock of any democratic regime. Effective governance crucially relies on citizens’support of the political system and the decisions it produces. This study investigates how participatory budgeting as a novel form of political participation relates to legitimacy perceptions. In a conjoint vignette survey experiment, we asked respondents to rate the legitimacy of hypothetical participatory budget decision processes, relating to a Swiss local context and differing in three attributes: (1) the opportunity to introduce policy proposal (agenda-setting), (2) the opportunity to vote on a proposal according to different voting methods, (3) whether the decision is in line with the respondent’s policy preferences or not. We find that participation in agenda-setting, (preferential) voting, and outcome favorability significantly increase the perceived legitimacy of participatory udgeting processes. However, our findings suggest that novel participation opportunities do not alleviate the difference in perceived legitimacy between the winners and losers created by policy decisions. |
| Keywords: | Legitimacy, Democratic Innovations, Participatory Budgeting, Voting Methods, Decision-marking |
| Date: | 2023–09–01 |
| URL: | https://d.repec.org/n?u=RePEc:fri:fribow:fribow00532&r= |
| By: | Berger, Benedikt; Adam, Martin; Rühr, Alexander; Benlian, Alexander |
| Abstract: | Owing to advancements in artificial intelligence (AI) and specifically in machine learning, information technology (IT) systems can support humans in an increasing number of tasks. Yet, previous research indicates that people often prefer human support to support by an IT system, even if the latter provides superior performance – a phenomenon called algorithm aversion. A possible cause of algorithm aversion put forward in literature is that users lose trust in IT systems they become familiar with and perceive to err, for example, making forecasts that turn out to deviate from the actual value. Therefore, this paper evaluates the effectiveness of demonstrating an AI-based system’s ability to learn as a potential countermeasure against algorithm aversion in an incentive-compatible online experiment. The experiment reveals how the nature of an erring advisor (i.e., human vs. algorithmic), its familiarity to the user (i.e., unfamiliar vs. familiar), and its ability to learn (i.e., non-learning vs. learning) influence a decision maker’s reliance on the advisor’s judgement for an objective and non-personal decision task. The results reveal no difference in the reliance on unfamiliar human and algorithmic advisors, but differences in the reliance on familiar human and algorithmic advisors that err. Demonstrating an advisor’s ability to learn, however, offsets the effect of familiarity. Therefore, this study contributes to an enhanced understanding of algorithm aversion and is one of the first to examine how users perceive whether an IT system is able to learn. The findings provide theoretical and practical implications for the employment and design of AI-based systems. |
| Date: | 2024–06–18 |
| URL: | https://d.repec.org/n?u=RePEc:dar:wpaper:146095&r= |
| By: | Höfmann, Michelle; Pott, Christiane; Quick, Reiner |
| Abstract: | This study investigates the impact of two changes to the auditor's report — a separate section addressing going concern uncertainties (GCU section [GCUsec]) and information on management and auditor responsibilities — and the characteristics of the audit committee on bank directors' perceptions and decisions. In a 2 × 2 × 2 between‐subjects experimental design with 85 German bank directors, we observe that a GCUsec in the auditor's report leads to more unfavourable decisions. Contrarily, explanations of responsibilities and different characteristics of the audit committee do not significantly impact on bank directors' perceptions and decisions. We thus confirm the effectiveness of the International Auditing and Assurance Standards Board (IAASB)'s revision of the International Standard on Auditing (ISA) 570 to enhance the informational value and decision usefulness of the auditor's report. |
| Date: | 2024–06–17 |
| URL: | https://d.repec.org/n?u=RePEc:dar:wpaper:146062&r= |
| By: | Andersson, Henrik; Ouvrard, Benjamin |
| Abstract: | We implement an online survey on a sample of 1, 088 French respondents to assess their willingness-to-pay (WTP) for meat substitutes and to test the effectiveness of informational treatments aimed at encouraging a switch to these substitutes. Using insights from the mental accounting theory, our treatments inform respondents about the carbon content of the different alternatives. We show that there is no significant difference in the WTP between the veggie and meat-like alternatives, both exceeding the WTP for cultured meat. Second, we detect weak and heterogeneous effects of our informational treatments. Third, our study emphasizes the need for careful consideration in study design, as certain results appeared to challenge the independence of irrelevant alternatives principle. |
| Keywords: | Externalities; Meat substitutes; Mental accounting; Willingness-to-pay |
| JEL: | C99 Q18 Q51 |
| Date: | 2024–06 |
| URL: | https://d.repec.org/n?u=RePEc:tse:wpaper:129428&r= |