In this project, we manipulate the public observability of forecasts and outcomes of a physical task. We explore how these manipulations affect overconfidence (OC). Participants in the experiment are asked to hold a weight after predicting how long they think they could do it for. Comparing the prediction and outcome times (in seconds) yields a measure of OC. We independently vary two dimensions of public observability (of the outcome and of the prediction). Additionally, we manipulate incentives to come up with an accurate prediction. This design allows us to shed light on the mechanism behind male and female OC. Following the existing literature, we formulate several hypotheses regarding the differences in predictions and outcomes for males and females in the presence of the public observability of predictions and outcomes. Our experimental data do not provide support to most of the hypotheses: in particular, there is no evidence of a gender gap in overconfidence. The most robust finding that emerges from our results is that incentives on making correct predictions increase participants’ forecasts on their own performance (by about 24%) and their actual performance as well, but to a lower extent (by about 8%); in addition, incentives to predict correctly in fact increase error for females (by about 33%).

We conduct an experiment on an online game, exploring the effect on gameplay behavior of voluntary commitment devices that allow players to limit their gameplay. Approximately 25% of players use the devices. Median and 75th percentile device users use devices approximately 60 and 100% of the time, respectively. Players who chose to use the device were those who had previously played longer and more frequently than those who chose not to use the device. Offering the commitment devices decreased session length and session frequency by 2.8 and 6.1%, respectively, while increasing weeks of play by 5.5%. Our results are consistent with some players having self-identified self-control problems, leading to longer and more frequent play than they would prefer, and to demand for commitment, and also with commitment devices creating a more rewarding experience, leading to longer-lasting involvement with the game. Our results suggest incentivizing or requiring commitment devices in computer games.

Experimental work using real married couples has shown that efficiency in intra-household allocations is influenced by information asymmetry between spouses. We conduct a lab-in-the-field experiment in rural India to test the extent to which lack of complete information on spousal preferences related to a bundle of private goods can affect allocation dynamics as well as expectations about allocations. We first show that there exist information asymmetries in spousal preferences, and that our information intervention helps reduce gendered misperception in beliefs about allocations and actual allocations, especially for men. However, information on spousal preferences does not significantly affect the final allocation decision, suggesting that husbands and wives may be responding to existing gender norms. We outline implications for experimental work on intra-household bargaining, and for policy.

The expanding literature on lying has exclusively considered lying behavior within a one-dimensional context. While this has been an important first step, many real-world contexts involve the possibility of simultaneously lying in more than one dimension (e.g., reporting one’s income and expenses in a tax declaration). We experimentally investigate individual lying behavior in one- and two-dimensional contexts to understand how the multi-dimensionality of a decision affects lying behavior. Our paper provides the first evidence regarding the pure effect of dimensionality on lying behavior. Using a two-dimensional die-roll task, we show that participants distribute lies unevenly across dimensions, which results in greater over-reporting of the lower-outcome die.

The Internet presents today’s researchers with unprecedented opportunities to conduct field experiments. Using examples from Economics and Computer Science, we present an analysis of the design choices, with particular attention to the underlying technologies, in conducting online field experiments and report on lessons learned.

Misreporting—a form of lying—is common in online labor and remote work settings. We execute an experiment on Amazon MTurk to determine how ex-ante honesty oaths and worker beliefs impact lying behavior across a range of plausible and implausible lies. Using a novel quantile-style exposition of the types of lies reported, we find that oaths elicit more truthful behavior, reducing both small, plausible lies and large, implausible ones. Shirking is reduced under oath. Worker expectations of group reporting are positively related to individual reporting of plausible lies.

Governments across the world have implemented restrictive policies to slow the spread of COVID-19. Recommended face mask use has been a controversially discussed policy, among others, due to potential adverse effects on physical distancing. Using a randomized field experiment (N = 300), we show that individuals kept a significantly larger distance from someone wearing a face mask than from an unmasked person during the early days of the pandemic. According to an additional survey experiment (N = 456) conducted at the time, masked individuals were not perceived as being more infectious than unmasked ones, but they were believed to prefer more distancing. This result suggests that wearing a mask served as a social signal that led others to increase the distance they kept. Our findings provide evidence against the claim that mask use creates a false sense of security that would negatively affect physical distancing. Furthermore, our results suggest that behavior has informational content that may be affected by policies.

Loss framing and checklist formatting are two oft-cited tools for encouraging behavior change, but there is little causal evidence on their impact in field settings. We partnered with the City of Philadelphia to test the effectiveness of these tools to increase completion of the City’s wellness program. In our experiment, 5235 City employees and retirees were randomly assigned to receive one of four postcard versions (using a 2 × 2 design), whereby we varied both framing (gain or loss) and how instructions were provided (information only or information in checklist format). Our results suggest that neither loss framing nor the checklist formatting significantly influenced the likelihood that individuals would complete the wellness tasks, or how quickly they completed the tasks. We conclude that this specific form of employee behavior may be difficult to influence through the “passive” behavioral interventions we tested, and suggest that a more “active” approach may be required in such instances.

Access to information via social media is one of the biggest differentiators of public health crises today. During the early stages of the Covid-19 outbreak in January 2020, we conducted an experiment in Wuhan, China to assess the impact of viral social media content on pro-social and trust behaviours and preferences towards risk taking with known and unknown probabilities. Prior to the experiment, participants viewed one of two videos that had been widely and anonymously shared on Chinese social media: a central government leader visiting a local hospital and supermarket, or health care volunteers transiting to Wuhan. In a control condition, participants watched a Neutral video, unrelated to the crisis. Viewing one of the leadership or volunteer videos leads to higher levels of pro-sociality and lesser willingness to take risks in an ambiguous situation relative to the control condition. The leadership video, however, induces lower levels of trust. We provide evidence from two post-experiment surveys that the video’s impact on pro-sociality is modulated by influencing the viewer’s affective emotional state.

The evidence on the welfare effects of kinship is mixed, suggesting both positive and adverse effects of kinship. This study looks into the differential effects of kinship on trusting and trustworthy behaviour by investigating the subjects’ motives and drivers of differential behaviour towards kin and non-kin. We conducted an economic experiment with households of rural India. We found that kin are trusted more than non-kin and that differential trust towards kin and non-kin is mainly driven by higher other-regarding preferences towards kin rather than being due to differences in expected reciprocity between kin and non-kin. We observed a heterogeneous effect of kin on trustworthy behaviour: kin exhibit low trustworthiness when they are not close to other kin, while they exhibit higher trustworthiness when they have close kin in the network.

Recruitment of representative and generalizable adult samples is a major challenge for researchers conducting economic field experiments. Limited access to representative samples or the high cost of obtaining them often leads to the recruitment of non-representative convenience samples. This research compares the findings from two field experiments involving 860 adults: one from a non-representative in-person convenience sample and one from a representative online counterpart. We find no meaningful differences in the key behaviors of interest between the two samples. These findings contribute to a growing body of literature demonstrating that non-representative convenience samples can be sufficient in certain contexts.

We present evidence from a natural field experiment designed to shed light on whether individual behavior is consistent with a neoclassical model of utility maximization subject to budget constraints. We do this through the lens of a field experiment on charitable giving. We find that the behavior of at least 80% of individuals, on both the extensive and intensive margins, can be rationalized within a standard neoclassical choice model in which individuals have preferences, defined over own consumption and their contribution towards the charitable good, satisfying the axioms of revealed preference.

In this paper, we propose a network model to explain the implications of the pressure to share resources. Individuals use the network to establish social interactions that allow them to increase their income. They also use the network as a safety and to ask for assistance in case of need. The network is therefore a system characterized by social pressure to share and redistribute surplus of resources among members. The main result is that the potential redistributive pressure from other network members causes individuals to behave inefficiently. The number of social interactions used to employ workers displays a non-monotonic pattern with respect to the number of neighbors (degree): it increases for intermediate degree and decreases for high degree. Respect to a benchmark case without social pressure, individuals with few (many) network members interact more (less). Finally, we show that these predictions are consistent with the results obtained in a set of field experiments run in rural Tanzania.

This paper investigates risk preferences using an artefactual field experiment conducted with a non-standard subject pool of farmers in Ghana. I introduce an alternative methodology for studying preferences following replication of a seminal risk elicitation procedure by Binswanger (Am J Agric Econ 62(3):395407, 1980). An important feature of both approaches is that they are easy to understand and, hence, are particularly suitable for eliciting preferences among subjects with low levels of formal education. I successfully replicate Binswanger's study, documenting how his original result of the moderate level of risk aversion for an average farmer can be generalized to a different country. However, using my alternative approach, whereby lotteries are presented in the loss domain, I find that half of my experimental subjects violated expected utility theory. This approach is of relevance to the current literature on studying risk preferences among subjects with poor literacy skills.

The widespread Internet “piracy” continues to fuel the debate about business models impervious to copyright infringement. We studied the displacement effects of “piracy” on sales in the book industry. We conducted a year-long large-scale field experiment: in the treatment group, we removed unauthorised copies appearing on the Internet and observed the sales data, whereas in the control group, we simply observed sales. We were able to substantially curb the unauthorised distribution, which resulted in a small, positive effect on sales. While using classical analysis we found it not to be significantly different from zero, a Bayesian approach using previous “piracy” studies to generate a prior led to the conclusion that protecting from piracy resulted in a significant sales boost of about 9 per cent.

Many studies have investigated the role of socio-demographic factors (including gender, age, race), cognitive ability and cultural factors on time and risk preferences. Yet, research regarding the effect of mindfulness on risk and time preferences has been limited. This study investigates the association between mindfulness and time/risk preferences. We conducted a survey on a representative sample of the French adult population (N = 1154) in Spring 2020. We assessed individual mindfulness through the Mindful Attention Awareness Scale (MAAS), and measured time and risk preferences with incentive-compatible economic games as well as self-reported questionnaires. Our results suggest that a higher level of mindfulness is associated with higher risk aversion and patience for stated preferences, but we found no relationship for revealed ones. We also observe that a higher level of mindfulness is related to greater time consistency, as we found a negative and significant association between the MAAS and the present and future biases.

Government programs promoting locally produced foods have risen dramatically. But are these programs actually convincing consumers to pay more for locally produced food? Studies to date, which have mostly relied on hypothetical stated preference surveys, suggest that consumers will pay premiums for various local foods and that the premiums vary with the product and presence of any geographic identity. This study reports results from a large field experiment involving 1,050 adult consumers to reveal consumers’ willingness to pay (WTP) premiums for “locally produced” foods – mushrooms and oysters. Despite strong statistical power, this study reveals no positive effect of the locally produced label on consumer WTP. These null results are contrary to most of the existing literature on this topic. The finding that consumers are not willing to pay more for local foods has important implications for state and federal agencies that promote labeling campaigns that seek to increase demand and generate premiums for locally produced foods.

U.S. soybean farmers are currently grappling with dicamba herbicide drift. Using a network diffusion framework that accommodates key features of soybean farmer networks, we estimate the damages incurred from dicamba drift across different regions. Under our baseline assumptions, we estimate an average yield loss of 3% and predict sizable levels of forced switching to dicamba-resistant seed in response to drift. The relative importance of drift on damage and seed choice holds across a range of economic and network assumptions. In the absence of policy, this damage may cause regional adoption rates of dicamba-resistant soybean seed to increase.

This paper studies whether school-based financial education has spillover effects from children to parents. Leveraging data from a large-scale experiment with public high schools in Peru and credit bureau records on the parents of the youth targeted, this study measures the impact of providing personal finance lessons during secondary school on parental financial behavior. Financial education lessons in the school yield limited average spillover effects, but lead to sizable effects on parental financial behavior within disadvantaged households. Among parents from poorer households, the treatment reduces default probability by 26%, increases credit scores by 5%, and increases current debt levels by 40%. The treatment has stronger effects among the parents of daughters, who experience a significant 6.7% increase in their credit score and a 28% reduction in their loan portfolio in arrears. Among the parents of boys, most of the spillover effects are muted.

Higher education enrolment and graduation rates have increased rapidly inter-generationally across much of the world, offering employers the promise of more knowledgeable recruits and promising individuals new means of social advancement. In the case of Bolivia, the labour force is becoming more heterogeneous over time, which could imply positive effects induced by a closer match between labour supply and recruiters’ needs. However, we show that this is not the case. We revisit the transition mechanisms from college to the workplace, positing recruiters’ interpretations of educational credentials as a crucial determining factor for employability in the formal sector. In a two-branch correspondence study, 2848 fictitious CVs were sent to 1424 formal firms in the three main urban Bolivian areas. We find a large university reputation premium. Applicants from well-valued universities are around 40% more likely to receive a positive response – a 2.25 percentage point advantage from a 7.87% baseline likelihood. Thus, the increasingly heterogeneous labour force is generating additional informational frictions in the labour market, rather than promoting a more efficient matching process.