We study how individuals value reductions in passive and undesirable waiting time using a real-incentive laboratory experiment. Participants were required to endure enforced inactivity – no phone, no reading, no conversation – and could pay to reduce the waiting time. By varying both the size of the time reduction (within-subject) and the initial waiting duration (between-subjects), we elicit willingness to pay (WTP) for time savings and compute the implied value of time (VOT) in euros per hour. We find that WTP is positive on average but increases less than proportionally with the amount of time saved. As a result, the VOT decreases with both the size of the time reduction and the initial waiting time. These findings highlight reference dependence and diminishing marginal sensitivity to time savings, challenging the assumption of a constant VOT commonly used in transport appraisal. Our results support a more context-sensitive approach to time valuation.
This article documents a gender difference in how information about a reference risk level shapes subsequent risk choices and provides evidence on the mechanisms behind it. We study two complementary settings: professional divers who receive information about a reference risk level before deciding whether to choose it again, and a two-stage lottery laboratory experiment in which participants receive information about a first-stage reference gamble. Across both studies, the more positive the information, the more likely men are to change away from the reference risk level, whereas women become relatively more likely than men to stick with it. A mediation analysis suggests that part of this divergence can be explained by a mechanism consistent with an emotional channel: information changes the perceived attractiveness of available choices in a gender-specific way.
We test whether minimal, non-informative messages can nudge tax compliance beyond standard deterrence. In a within-subjects lab experiment, we randomize exposure to either a “reminder” that leaves audit probability unchanged or an informative “warning” tied to higher audit probability, and estimate effects on both the probability of evasion and the share of income evaded. A short non-informative reminder, holding incentives fixed, lowers the probability of evasion by about 16 percentage points, with no detectable effect on the evaded share among evaders; informative messages add at most marginal effects once audit probability is controlled for.
Non-cognitive skills, particularly personality measures, are increasingly utilized as predictors for various outcomes in economic literature. The prevailing assumption in this literature is the stability of non-cognitive skills. We scrutinize this assumption by combining population-based Northern Finland Birth Cohort 1966 data with highly reliable data from multiple Finnish registers and assessing the stability of Cloninger’s temperament traits (n = 2897) and active coping (n = 3006) between ages 31 and 46. We find mean-level changes in all traits to be marginal, except for a moderate decline in novelty seeking. Family, income- or employment-related life events do not generally correlate with trait changes. However, health-related changes are associated with altered harm avoidance. Some associations also exist between health changes and active coping alterations. Studies exploring these traits’ impact on life outcomes should consider their interplay with health to mitigate potential endogeneity bias. Given the similarity between the Big-Five trait neuroticism and harm avoidance, our results suggest future research to explore the connection between health changes and neuroticism.
Conflict, defined as a condition in which one individual's interest entails a loss for another, is a common feature of daily life. At the same time, humans rely on interaction in public goods to enhance and maintain their standard of living, particularly in matters involving public goods. This study examines the relationship between interaction in public goods provision and transparency, operationalized as transparency on contributions to a public good, and their effects on conflict behavior using an experimental approach. Interaction and transparency exposure introduced through a public goods game, while conflict was measured using a contest game. Our results indicate that interaction experience in public goods environments is associated with reduced conflict frequency and improved aggregate welfare, primarily through lower levels of conflict investment. In contrast, transparency of public goods contributions does not significantly influence the likelihood of conflict, but it promotes convergence in conflict investment behavior. This study concludes that interaction and contribution transparency in public goods settings may contribute to enhanced social stability.
We report the results from a controlled laboratory experiment using a variant of the one-shot Ultimatum Game in which the pie available for sharing increases with the proposer’s own demand, creating an explicit equity–efficiency trade-off for both roles. Using data from 148 participants across three conditions – a baseline with a fixed pie, and one of two efficiency-loss conditions, in which a fairer distribution incurs either a moderate or severe reduction in the amount available – we investigate whether and how participants’ decisions, in the role of proposers and responders, are affected by the equity–efficiency trade-off. Our data confirm that participants in both roles tend to frequently aim at (almost) equal-split of the available amount, and that there is nonetheless a sizeable heterogeneity in demands. Overall, the evidence suggests that efficiency losses have limited effects on participants’ demands, although proposers tend to increase their demands under severe efficiency losses, while responders’ demands remain comparatively stable.
The share of couples where the wife out-earns the husband is increasing globally. This shift raises questions about how gender norms influence household specialization when women have a comparative advantage in the labor market. This paper conducts a laboratory experiment to causally identify how gender norms shape the division of mixed-gender pairs when women are the more productive partners. In a 2 × 2 Battle of the Sexes game, we exogenously manipulate the salience of gender norms and the relative productivity of men and women. Our results show that men consistently adhere to traditional gender norms. They are more likely to opt for higher-paying “Career” roles when gender norms are salient, regardless of relative productivity. In contrast, women’s responses are heterogeneous. When both genders earn the same from the labor market, norm salience leads women to be more likely to choose the high-earning option, which departs from traditional norms. When women earn more, however, they adopt a more conservative strategy. These patterns appear to be influenced by expectations about partner behavior and norm-dependent inequality aversion. Our findings highlight the efficiency costs associated with persistent gender norms in economies where women are increasingly likely to match or exceed their male partners in earning potential.
Digital financial services may help informal savings groups use safer and more efficient financial tools, but take-up remains limited in many low-income settings. Using a clustered randomised controlled trial, we evaluate the effectiveness of an intervention that combines financial literacy training, mobile money instruction, and SMS reminders among Village Savings Loan and Association (VSLA) members. The findings show that the intervention led to higher reported knowledge of mobile money services and use of selected functions. It also improved self-reported financial inclusion indicators and perceptions of mobile money as a secure and efficient tool for VSLA savings and deposits. Treated participants were less likely to believe that losing a phone would imply losing mobile money balances. These findings suggest that training combined with SMS reminders can improve knowledge, self-reported use, and perceptions of mobile money among individuals already engaged in informal savings groups.
Green crowdfunding is an emerging field of social entrepreneurship supporting the development of eco-friendly projects. With rising competition, research focuses on factors driving campaign success, mostly relying on secondary database analyses. In a shift from previous research, this study focuses on individual-level analysis and, adopting a twofold perspective, investigates the interplay between entrepreneurs’ and backers’ psychological and demographic characteristics and their effect on funding decisions in donation-based crowdfunding. On the one side, we experimentally manipulated entrepreneurs’ age and gender to analyse their effects on participants’ funding decisions, measured as i) liking; ii) willingness to fund the campaign; iii) amount of money donated. On the other side, we examined how backers’ personality traits, stereotypes, and environmental values impact funding decisions. Results from an online study involving 954 participants showed that age and gender of the entrepreneur did not affect backers’ funding decisions. However, backers’ age, gender, and agreeableness influenced funding behaviour above and beyond individual differences in environmental values. Notably, the results suggested that psychological and environmental dispositions shape evaluative and intentional responses but fail to predict monetary behaviour. These findings offer insights into the green crowdfunding market and highlight the importance of targeting backer profiles to promote sustainable investment decisions
Probability discounting is a widely used framework for studying attitudes to risk in psychology, with applications spanning addiction, gambling and clinical research. Despite its prevalence, the model’s connection to established economic theories of choice under risk, such as expected utility theory and prospect theory, remains poorly understood. I critically examine the theoretical foundations of the probability discounting model, arguing that it rests on an outdated conception of reinforcement learning. I show that the model is formally isomorphic to the dual theory of choice under risk, but limited to binary prospects with a fixed parametric probability weighting function. I review the methodological and statistical approaches commonly used in probability discounting research and employ a structural econometric framework to test the validity of the model’s implied probability weighting function. Using data from a widely cited study, I demonstrate that the probability discounting model is overly restrictive because it cannot capture the simultaneous overweighting and underweighting of probabilities, a common finding in experimental studies of choice under risk. These results suggest that researchers in applied domains should transition to the more flexible, well-established models of choice under risk used in economics and decision theory.
Although standard models often suggest a link between Internet usage and life satisfaction, omitted variables can compromise the reliability of these estimated effects. Previous studies addressing this endogeneity have been limited to a single explanatory variable. However, Internet usage is typically recorded in survey data as ordinal categories, which may require more than one variable to capture its effect. This paper develops a control function method that tests and corrects for the endogeneity of a set of dummy variables representing Internet use frequency in a linear model for life satisfaction. Utilizing the proposed methodology and a novel instrumental variable, the study reveals that Internet use frequency is essentially unrelated to the life satisfaction of Europeans.
We study the North-South divide in Italy, a paradigmatic case of within-country differences, to investigate whether regional differences in cooperation and trust stem from stable generalized preferences or context-dependent expectations. We conducted a survey experiment with 2405 Italians under three conditions: (a) no information about the counterpart's region, (b) knowing they share region, (c) knowing they are from opposite area. No regional differences were found in the no-information condition, but significant decrease in trust beliefs emerged when Southerners interacted with fellow Southerners. Our results suggest that negative in-group expectations drive regional differences, though evidence is mixed across methods.
We model creativity as capital built by costly creative effort that complements social capital and is often accompanied by routines that economize attention and time. Higher effort costs are associated with lower entry into the creative state, while openness and trust are associated with higher productivity of creative effort mainly through creative capital. Using field survey data from an Italian music festival and a recursive bivariate probit, we find that costs are negatively associated with creativity, whereas creativity is strongly associated with festival collaboration, volunteering, and territorial cooperation. Consistent with a routinization perspective, the creativity–engagement link is stronger when participation occurs in more socially “structured” environments. To encourage creativity, policies should reduce cognitive frictions and improve the productivity of creative effort.
Large Language Models (LLMs) are increasingly embedded into Computer Control Agents (CCAs), enabling them to act autonomously on behalf of users. While prior research has examined behavioral biases in LLM-generated text, little is known about whether such biases and heuristics extend to automated actions. This study investigates whether CCAs exhibit the decoy effect — a well-documented bias in human decision making, where the presence of an inferior alternative shifts preferences toward a target option. Using the open-source framework browser-use, 3600 simulations across five canonical decision scenarios with six state-of-the-art LLMs are analyzed. Results show an aggregate decoy effect where CCAs select the target more often when a decoy is present (43.4%) than when it is absent (38.0%; p<.001), consistent with bounded rationality observed in humans. However, the magnitude and direction of the effect vary across models, with some displaying no significant bias. These findings extend evidence of heuristics in LLM behavior from text outputs to autonomous actions, highlighting both opportunities and risks for deploying CCAs in decision-making contexts. Implications for digital nudging, system design, and the use of LLMs as simulations in behavioral research are discussed.
We report an experiment that tests whether violations of the stationarity principle of the exponentially discounted utility model are mistakes. In normative decision theory, a mistake is defined as a choice that a decision-maker does not wish to maintain following an analysis of their decision. If observed violations of the stationarity principle are mistakes of that type, its normative credentials remain intact. The intertemporal decisions of 448 subjects were observed and categorized according to whether they were stationary or not. Subjects were then provided with a normative argument in support of a counterfactual behavior, alongside the opportunity to revise their decisions. The majority of subjects made stationary choices and did not revise them, despite the normative reason to do so. Those choices cannot be attributed to mistakes. However, final choices also exhibit systematic present-biased violations of the stationarity principle that cannot be attributed to errors. Hence, whilst the stationarity principle cannot be excluded as a requirement of rationality for the majority of our subjects, nor can it claim to be universal.
We evaluate a multi-faceted intervention aimed at improving social inclusion and reducing prejudice against individuals with visual impairment. The intervention, randomly assigned to upper-secondary school students, consists of an awareness-raising activity and a simulation-based inter-group contact activity. While we find positive effects on knowledge of visual impairment, perspective-taking and empathic concerns, and general societal attitudes towards persons with visual impairment, no improvements are observed in terms of implicit attitudes or multidimensional attitudes. Moreover, the intervention does not improve outcomes measured through incentivized choices, such as the willingness to pay for social interaction with persons with visual impairment, beliefs regarding their performance and outcomes in various domains, and altruism towards them. The evidence suggests that assessing impacts only on knowledge and general attitudes, as is commonly done in the literature, may not suffice to determine the extent to which such interventions are successful at improving social inclusion for persons with visual or other forms of impairment.
Communicating food safety and risk information to consumers is essential to reducing the incidence of foodborne illness. However, despite widespread public health campaigns, behaviour change remains limited. This study investigates whether framing individual responsibility in campaign messages influences consumers' stated preferences for food safety campaigns. Using a web-based discrete choice experiment with 2343 Scottish adults, participants were randomly assigned to one of three groups: a control group (no prompt), a treatment with a statement-based responsibility prompt or a treatment with a question-based prompt. Respondents completed a series of choice tasks evaluating hypothetical campaigns that varied by delivery channel, timing and message style, alongside an opt-out option ("no campaign"). Results show that question-framed prompts significantly increase the likelihood of choosing a campaign over the opt-out, relative to both the control and statement conditions. While the overall effects are modest in magnitude, they are consistent with prior evidence on the cognitive and motivational impact of question-based framing. These findings suggest that simple, lowcost message framing strategies, particularly those encouraging reflection and self-persuasion, can enhance the perceived effectiveness of public health campaigns and may support improved food safety communication design.
Despite widespread concern for animal welfare, a gap persists between consumers' attitudes and their purchasing behaviour. To address this discrepancy, our study examined various nudging strategies within a simulated, twodimensional virtual supermarket in Germany. Using a 2 & times; 2 between-subjects design, we assessed the effectiveness of a social norm nudge, an information nudge and a combination of both, while also exploring the mediating role of descriptive normative beliefs. Our results reveal significant main effects for both nudges, but no interaction effect. This suggests that each nudge independently increased the rate of purchasing products with higher animal welfare standards. Although the combined nudge condition resulted in the largest overall impact, the effects were additive rather than super-additive. Additionally, while participants generally held high levels of descriptive normative beliefs, the social norm nudge did not further elevate these beliefs. Nonetheless, participants with higher descriptive normative beliefs tended to purchase more products with higher animal welfare standards. The study findings emphasise the need for a nuanced understanding of how combined nudging strategies impact purchasing behaviour in supermarkets.
The widespread adoption of smart travel cards has substantially enhanced the flexibility of pricing mechanisms in public transport. This study examines the impact of a monthly voucher on increasing the utilization of public transport services. The analysis is based on a controlled field experiment conducted within Bogota's public transport system. We provide empirical evidence of the presence of Mental Accounting Bias (MAB) in the context of transport voucher use. Specifically, we demonstrate that the increased ridership observed among voucher recipients can be attributed not only to the voucher's monetary value but also to a behavioral response triggered by a perceived expansion of their transport budget. This finding highlights the role of MAB in explaining why this intervention was effective, whereas other demand-side subsidies, such as fare discounts, have proven less successful in this context. To establish the presence of MAB, we employ three distinct methodological approaches, each drawing on different sets of evidence collected during the experiment. They provide compelling evidence of MAB. To our knowledge, this study presents the first empirical investigation of MAB in public transport. The potential to induce MAB through transport vouchers offers valuable insights for the design of policies aimed at increasing public transport ridership.
