Job loss is a common shock in low-income countries, but formal insurance against it is typically limited. In the aftermath of a mass layoff in Ethiopia’s Hawassa Industrial Park, we quasi-experimentally study the impacts of job loss on worker welfare, and then experimentally document the impacts of expanding job-loss insurance payments. We find that job loss has persistent and substantial impacts on long-run employment and consumption expenditure; that job loss payments substantially moderate the fall in consumption expenditure, with large negative impacts on employment if payments are given as a lump sum; and that individuals have a high willingness to pay for additional job-loss insurance.

...

...

...

...

...

...

...

...

A central factor when choosing an action is its causal effect on the outcome of interest. Yet, causal information is often lacking. People instead observe correlational or historical data, along with causal interpretations and recommendations provided by experts who frequently disagree with each other. Our laboratory experiments study choice in such settings, where beliefs concern the structure of the data-generating process rather than merely magnitudes. Roughly half of our subjects attempt to determine the fit of the causal interpretations to past data, as the literature on model persuasion assumes. We characterize the limits to their ability to do so. Half the subjects’ choices are co-determined by the interpretations’ promises of future payouts, as the literature on narrative competition assumes, or by the downside these choices entail if they are mistaken. Subjects also commonly employ heuristics such as Occam’s razor. The fact that they typically prefer flexibility over parsimony insures them against bad choices in some settings but has the opposite effect in others. Our estimates predict well out of sample and closely agree across two different identification strategies in two different samples. We also study the extent to which behavior is robust to framing and the relation between subjects’ choices and their political attitudes and psychological characteristics. Our results characterize the cases in which subjects’ behavioral tendencies render them most receptive to misleading interpretations. They also inform the literatures on narrative competition and model persuasion.

Economic research on gender gaps in preferences and economic outcomes has identified gaps with respect to sex—a binary classification as either a “man” or “woman.” We investigate the value of incorporating a continuous measure of self-reported gender identity into economics by exploring whether gender identity correlates with variation in economic decisions and outcomes beyond the explanatory power of binary sex. We validate a novel and simple measure of continuous gender identity (CGI) that correlates with measures used in gender research outside economics. We employ this validated measure in five datasets (N=8,213) measuring several dimensions of economic preferences and educational and labor market outcomes. Our results yield mixed evidence that CGI correlates with economic behaviors and outcomes when accounting for binary sex: CGI has little relationship with incentivized preference measures, but is correlated with a number of unincentivized preference measures and with educational choices and labor market outcomes that are stereotypically associated with females. Our results suggest that incorporating self-reported measures of gender identity may have some value for understanding economic behaviors and for targeting policy.

Job seekers face uncertainty about their abilities, and whether these match with job requirements. Such uncertainty may result in sub-optimal job search outcomes and job matches. We conduct a field experiment among job seekers in Addis Ababa, Ethiopia. Participants underwent a skill assessment and were asked about their willingness to pay (WTP) for information about their relative performance on a test of general intelligence. WTP is positive for about 80 percent of the population, and is associated with gender and personality. Feedback provision leads individuals to update their beliefs which only persists for individuals with low WTP. We provide evidence that suggests imperfect recall as potential mechanism for the lack of persistence. Feedback increases job search intensity but relatively lower for initially overconfident individuals and those with negative or zero WTP, which results in lower realized wages for these groups. The heterogeneity in belief updating, recall and job search behavior is consistent with some overconfident job seekers being unable to forget information and, thus, to maintain motivated beliefs, but being sophisticated about this inability to forget.

The transition towards sustainable behaviors is a crucial step in mitigating climate change. However, individuals are often poorly informed about the consequences of their actions, especially in contexts in which it is not obvious what is the ‘right thing’ to do. This study reports evidence from a field experiment on Internet data consumption. Evidence from a survey we conduct shows that most of smartphone users lack knowledge about the environmental advantages of Wi-Fi networks relative to mobile ones (e.g., 4/5G). We design and test two behavioral interventions aiming at informing users about the consequences of their consumption choices and shifting Internet data consumed on mobile to Wi-Fi networks. Results show that information alone is not sufficient to produce a change in consumption levels. However, when combined with personal goals, we observe a significant and persistent effect. Such a cost-effective solution can potentially reduce Co2 emissions by 2.4 kg per year/person, remarking the advantages of behavioral tools for policymaking.

The development of formal models of decision making under risk has been shaped largely by decisions between options with monetary outcomes, with the most prominent model being cumulative prospect theory (CPT). Whereas CPT is good at describing choices between monetary lotteries, it shows poorer performance in the context of decisions between options with nonmonetary and nonnumerical outcomes (e.g., medications with possible side effects). We suggest that this may be due to affective processes and context-dependent evaluation---which are not considered in CPT---playing a larger role in nonmonetary than in monetary choices. We propose three psychologically motivated modifications to CPT's modeling framework to capture these differences: (a) representing the subjective value of a nonmonetary outcome by an affect rating (rather than a monetary equivalent); (b) determining the subjective affective value of an outcome relative to the value of the worst outcome in the choice problem; (c) assuming that the probability weighting for an outcome depends on the amount of affect triggered. We submit model variants of CPT implementing the proposed modifications to a model comparison in three empirical data sets. For choices between options with negative nonmonetary outcomes (medications with possible side effects), these modifications substantially improve CPT's performance relative to that of the original version of CPT. The same does not hold for monetary choices. Overall, in addition to fleshing out key differences in the processing of monetary and nonmonetary risky options, our work demonstrates how affective processes can be formally integrated within classical theories of decision making under risk.

Fragile states struggle to develop basic and essential capacities like tracking and paying employees. Developing these capabilities presumes confidence that the state will continue to exist, which is not assured in fragile states. We present the results of a randomized evaluation in Afghanistan, where the government implemented a major reform to modernize its teacher payroll system. This required teachers to biometrically register and receive digital salaries. This reform made employees legible to the state, reduced payment delays, and increased financial inclusion for teachers. We also find suggestive evidence that the reform increased student learning. Notably, implementation was most successful in districts with higher public confidence in the government’s ability to prevail against the Taliban. This result points to a complementarity between confidence in future stability and the development of basic state capacity.

In the impunity game, responders, unlike the ultimatum game, cannot affect proposers' outcomes. Proposers in this game, like in the dictator game, have full control over their own outcome, as rejection from the responder has no effect on their payoff. Thus, the theoretical prediction of this game states that the responder should accept any offer. An experiment is designed aiming at analyzing both players’ behavior in the impunity game when subjects are aware of the gender of their partner. Additionally, we examine the effect of different stake sizes. In a first stage, an online experiment with eight different treatments is implemented, with a total number of 1,210 observations. From this data, the main findings are that proposers give to responders an important (around 35%) share on average, and that the stake, size as well as the gender identification, affect subjects' decisions. Moreover, responders’ rejection patterns follow the game theoretical prediction, although the hypothesis that knowing your counterpart sex/gender affects responders’ behavior cannot be rejected. Several sociodemographic characteristics are found to affect subjects’ behavior in this game, like personality and psychopathy traits, as well as their emotional intelligence level. New data for a second stage of this experiment where doodly vignettes tacitly transmit gender features and overcome other limitations of the previous stage, reinforce some of the results mentioned above.

We present experimental evidence that enabling access to universal early child care for families with lower socioeconomic status (SES) increases maternal labor supply. Our intervention provides families with customized help for child care applications, resulting in a large increase in enrollment among lower-SES families. The treatment increases lower-SES mothers’ full-time employment rates by 9 percentage points (+160%), household income by 10%, and mothers’ earnings by 22%. The effect on full-time employment is largely driven by increased care hours provided by child care centers and fathers. Overall, the treatment substantially improves intra-household gender equality in terms of child care duties and earnings.

We examine the impact of ideological alignment between institutions disseminating research findings and policymakers on the adoption of evidence-based policies. We conduct a country-wide RCT in which two ideologically opposite prominent think tanks, two major newspapers, and a research institution with nonsalient ideology communicate identical information about a low-cost, non-ideological, and effective policy based on published research findings to a large sample of Spanish local policymakers. We measure the direct impact of information on policy adoption. We find that the mere provision of information is insufficient for policy adoption. However, when the informing institution aligns ideologically with policymakers, the communication of research results leads to a more than 65% increase in policy adoption. In contrast, the effect is halved when the institution’s ideology is nonsalient and null when it is the opposite. Both communication instruments - policy briefs and newspaper articles- are equally effective in increasing policy adoption. We propose a three-stage conceptual framework of policy adoption processes - selective exposure to information, belief updating, and policy implementation. We show that ideological alignment does not influence selective exposure to information. However, evidence from a post-intervention online experiment suggests that ideological alignment affects belief updating regarding the recommended policy’s effectiveness and subsequent implementation. Finally, we discuss the trade-offs between effectiveness and outreach when using ideologically aligned and nonsalient institutions to disseminate research evidence and comment on the economic impact of ideological alignment for policy implementation.

The promise of generative AI to increase human productivity relies on developing skills to become proficient at it. There is reason to suspect that women and men use AI tools differently, which could result in productivity and payoff gaps in a labor market increasingly demanding knowledge in AI. Thus, it is important to understand if there are gender differences in AI-usage among current students. We conduct a survey at the Norwegian School of Economics collecting use and attitudes towards ChatGPT, a measure of AI proficiency, and responses to policies allowing or forbidding ChatGPT use. Three key findings emerge: first, female students report a significantly lower use of ChatGPT compared to their male counterparts. Second, male students are more skilled at writing successful prompts, even after accounting for higher ChatGPT usage. Third, imposing university bans on ChatGPT use widens the gender gap in intended use substantially. We provide insights into potential factors influencing the AI adoption gender gap and highlight the role of appropriate encouragement and policies in allowing female students to benefit from AI usage, thereby mitigating potential impacts on later labor market outcomes.

We empirically examine the value of modern digital communication tools for inducing prosocial behavior. In our online experiment (N = 594), charity members transmit a standardized message to potential donors through alternative digital communication channels varying the amount of nonverbal content (written message in the baseline TEXT condition vs. voice recording in AUDIO vs. video-recorded discourse in VIDEO). We find partial support for the initial conjecture that individuals are more cooperative towards strangers the less anonymous the latter become to the former. Compared to the baseline TEXT condition, our AUDIO treatment induces a nearly 40% increase in the average donation. However, the transmission of nonverbal cues may backfire: the effect observed in the richest VIDEO condition has only half the magnitude of the one in AUDIO. We attribute this phenomenon to the “avoiding the ask” behavior previously documented in the charity giving literature. We also rule out the possibility that these treatment effects stem from perceptual mechanisms by which these changes in prosociality are driven by the differences in the perception of charity members in the stimuli, suggesting that the treatment effects capture the intrinsic value of reducing anonymity for promoting prosociality in the digital world.

This paper studies whether malfunctioning institutions erode good behavior. We use a large-scale online experiment, in which participants play a repeated observed cheating game. When we ask participants to report honestly and promise no control, we find low cheating rates. When control of truthful reporting is introduced, low cheating rates remain. In our main treatment with a malfunctioning institution, participants do not know whether they are in the treatment with or without control. In this treatment, participants who do not face control for some rounds start cheating significantly more often, reaching highest cheating rates. That is, a malfunctioning institution leads to more cheating than no institution at all, which indicates that the development of cheating behavior is endogenous to the institutions. Our findings suggest a novel negative effect of unenforced laws.

Unconditional cash transfers are a popular social protection policy, but their impact is often short-lived. One plausible explanation is that recipients struggle to budget and plan, leading to insufficient savings for future productive investments. Through an RCT involving 861 refugee households in Uganda, we test whether a mental accounting intervention—where households divide their monthly cash transfer across four labeled envelopes (Education, Health, Investments, Others), instead of the status quo of one unlabeled envelope—can help recipients overcome commitment challenges. Immediately after the cash transfer program ends, Treatment households exhibit greater savings and have taken out larger loans. One year later, these households have made more productive investments, particularly in lumpy investments, resulting in higher monthly income and savings, while also paying off their debt. These outcomes are driven by households that continue using the labeled envelopes one year after the cash transfer program ends, who are also more likely to report that the envelopes help with budgeting, planning, and disciplining spending.

Do land market frictions cause misallocation in agriculture? In a field experiment in Western Kenya, we randomly subsidize owners to rent out land. Induced rentals mostly persist after the subsidy ends and increase output and value added, consistent with misallocation. Gains from trade arise from renters choosing higher-value crops, having higher productivity, and adopting more non-labor inputs, while, perhaps surprisingly, renters use similar quantities of labor as owners. Induced rentals are not those with the largest predicted gains, underlining the importance of the joint distribution of gains and frictions, with frictions arising from search, risk, and learning.