|
on Experimental Economics |
| By: | Max Cytrynbaum; Fredrik S\"avje |
| Abstract: | We describe a new family of coupling designs, extending the basic principle of stratified randomization to experiments with continuous, constrained multivariate, text/image and other irregular treatment spaces. Our approach is to first match units into homogeneous groups, then use Monte Carlo coupling techniques to assign within-group treatments that are highly dispersed over the treatment space. We show that ensuring similar experimental units receive highly dissimilar treatments generically improves estimation efficiency. In particular, the efficiency gains from a coupling design are proportional to the product of dispersion and match quality, where dispersion measures how spread out the treatment assignments are under a given coupling relative to independent randomization. We develop a new spectral analysis, revealing how efficiency depends on a match between the smoothness and shape of the estimator's influence function and the principal directions of a given coupling. We illustrate how coupling designs work in practice using a cash transfer experiment in development economics and a discrete-choice experiment in two-sided marketplaces. |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2604.09858 |
| By: | Alex Farach; Alexia Cambon; Lev Tankelevitch; Connie Hsueh; Rebecca Janssen |
| Abstract: | Organizations have widely deployed generative AI tools, yet productivity gains remain uneven, suggesting that how people use AI matters as much as whether they have access. We conducted a field experiment with 388 employees at a Fortune 500 retailer to test two scaffolding interventions for human-AI collaboration. All participants had access to the same AI tool; we varied only the structure surrounding its use. A behavioral scaffolding intervention (a structured protocol requiring joint AI use within pairs) was associated with lower document quality relative to unstructured use and substantially lower document production. A cognitive scaffolding intervention (partnership training that reframed AI as a thought partner) was associated with higher individual document quality at the top of the distribution. Treatment participants also showed greater positive belief change across the session, though sensitivity analyses suggest this likely reflects recovery from carry-over effects rather than genuine training-induced shifts. Both findings are subject to design limitations including an AM/PM session confound, differential attrition, and LLM grading sensitivity to document length. |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2604.08678 |
| By: | Minseo Choi (KAIST College of Business); SeEun Jung (Inha University); Duk Gyoo Kim (Yonsei University) |
| Abstract: | We investigate whether behavioral norms formed in same-gender environments persist when individuals later interact in mixed-gender groups, focusing on willingness to volunteer for low-promotability tasks (LPTs). Using a two-stage laboratory experiment that varies group gender composition over time, we find that initial exposure to same-gender groups generally reduces subsequent volunteering in mixed-gender settings. However, women who transition from same-gender to mixed-gender groups volunteer more than men, a pattern traditionally attributed to gendered social expectations. While prior literature attributes such gaps to gendered social expectations, our data challenge the universality of this mechanism. In our context, participants overwhelmingly assign the LPT to a male, rather than a female, peer in hypothetical supervisor scenarios, suggesting that expectation-based mechanisms do not drive the observed gender gap. We propose that women’s higher volunteering instead reflects greater aversion to strategic uncertainty, which becomes more salient in mixed-gender environments. Consistent with this interpretation, women with single-sex schooling backgrounds, accustomed to more predictable peer environments, exhibit especially high volunteering rates in mixed-gender groups. These results indicate that same-gender experiences shape later LPT behavior and that women may volunteer for LPTs not only to comply with social norms but also to mitigate strategic uncertainty. |
| Keywords: | Low-promotability tasks, Same-gender environments, Gender differences, Strategic uncertainty |
| JEL: | C92 J16 I21 |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:yon:wpaper:2026rwp-284 |
| By: | Amil Camilo Moore; Rosemarie Nagel; Fabrizio Germano |
| Abstract: | We study similarity in the complete set of one-shot 2×2 games with payoffs from {1, 2, 3, 4} without replacement. Similarity is defined geometrically via a neighborhood structure on games and continuity of behavior, and is applied to both theoretical rules (e.g., Nash equilibrium, level-k reasoning) and experimental data. This produces a partition of the games into (theoretical or empirical) similarity classes. We run a large-scale experiment in which each subject plays all 78 games within our class without feedback. We find that empirically inferred similarity classes diverge sharply from those predicted by Nash equilibrium and dominance reasoning. Instead, the empirical similarity classes align closely with the theoretical classes of a level-k variant, with deviations reflecting fairness and efficiency concerns. At the individual level, subjects' play can be classified according to primary and secondary rules, conforming with either level-k variant (0 ≤ k ≤ 5) or a fairness and efficiency-based heuristic. The main insights extend to strategic settings beyond our 2 × 2 games. |
| Keywords: | equity and efficiency, experiments, level-k reasoning, similarity of games, topology of games |
| JEL: | C52 C70 C72 C81 C90 C93 D91 |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:bge:wpaper:1571 |
| By: | Shuhei Kitamura; Ryo Takahashi; Katsunori Yamada |
| Abstract: | Elections can deter corruption only if voters punish tainted incumbents. We study whether punishment depends on second-order beliefs—beliefs about how other voters will react. Before Japan’s October 2024 general election amid a funding scandal, we ran a pre-registered online survey experiment. To study this channel, we provided no new factual information about the scandal itself and instead reported a baseline statistic about perceived public intolerance of the underlying corruption: treated respondents learned that, in our baseline survey, the average respondent estimated that 67% of other respondents viewed the conduct as unacceptable. The message increased turnout by 6 percentage points and support for opposition challengers by 7 percentage points. Effects were sharply heterogeneous. Swing voters, especially those who initially overestimated how widely others would punish, became more likely to vote and back challengers. By contrast, ruling-party supporters, especially those who initially underestimated how widely others would punish, shifted toward the incumbent when they learned that intolerance of the corruption was higher than expected. More broadly, anti-corruption messages may affect voting not only by changing beliefs about wrongdoing, but also by changing beliefs about others’ reactions, helping explain why such campaigns often have mixed effects. |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:dpr:wpaper:1289rr |
| By: | Antoinette Baujard (GATE Lyon Saint-Étienne - Groupe d'Analyse et de Théorie Economique Lyon - Saint-Etienne - UL2 - Université Lumière - Lyon 2 - UJM - Université Jean Monnet - Saint-Étienne - EM - EMLyon Business School - CNRS - Centre National de la Recherche Scientifique); Roberto Brunetti (GATE Lyon Saint-Étienne - Groupe d'Analyse et de Théorie Economique Lyon - Saint-Etienne - UL2 - Université Lumière - Lyon 2 - UJM - Université Jean Monnet - Saint-Étienne - EM - EMLyon Business School - CNRS - Centre National de la Recherche Scientifique, CREM - Centre de recherche en économie et management - UNICAEN - Université de Caen Normandie - NU - Normandie Université - UR - Université de Rennes - CNRS - Centre National de la Recherche Scientifique); Isabelle Lebon (CREM - Centre de recherche en économie et management - UNICAEN - Université de Caen Normandie - NU - Normandie Université - UR - Université de Rennes - CNRS - Centre National de la Recherche Scientifique); Simone Marsilio (UniSR - Universita Vita Salute San Raffaele = Vita-Salute San Raffaele University [Milan, Italie]) |
| Abstract: | If individuals are to be empowered in their selection or use of a voting rule, it is necessary that they understand it. This paper analyzes people's understanding of two voting rules: evaluative voting and majority judgment. We first distinguish three components of understanding in this context: how to fill in the ballot; how votes are aggregated; and how to vote strategically. To measure each component, we draw on results from a lab experiment on incentivized voting where participants are exogenously assigned single-peaked preferences and answer comprehension questions on the rules employed. We find that most participants understand how to fill in the ballot with both voting rules. However, participants' understanding of vote aggregation under majority judgment is lower and, crucially, more heterogeneous. While some participants correctly understand its aggregation property, a sizable group fails to grasp it. We also observe no difference in voting behavior between evaluative voting and majority judgment: the data confirm the theoretical prediction that under evaluative voting there will be a high incidence of strategic voting through the use of extreme grades, but contradict the prediction that under majority judgment voters will vote less strategically. Finally, we find that with majority judgment, the better voters understand how votes are aggregated, the more they vote strategically, hence resulting in inequality in voter agency. |
| Keywords: | laboratory experiment, majority judgment, evaluative voting, understanding, voting rules |
| Date: | 2024–07–11 |
| URL: | https://d.repec.org/n?u=RePEc:hal:journl:hal-04653702 |
| By: | Brett R. Gordon; Robert Moakler; Florian Zettelmeyer |
| Abstract: | Randomized controlled trials (RCTs) provide the most credible estimates of advertising incrementality but are difficult to scale. We propose Predicted Incrementality by Experimentation (PIE), which reframes ad measurement as a campaign-level prediction problem. PIE uses a sample of RCTs to learn a mapping from campaign features to causal effects, then applies it to campaigns not run as RCTs. Because the RCTs identify the causal effects, PIE can incorporate post-determined features—campaign-level aggregates such as test-group outcomes, exposure rates, and last-click conversions, computed after campaign completion. These metrics reflect the consumer behaviors that generate treatment effects, so they carry predictive information about incrementality even though they would be invalid controls in a causal model. Using 2, 226 Meta ad experiments, PIE achieves an out-of-sample R2 = 0.88 for incremental conversions per dollar, compared to R2 = 0.19 for industry-standard 7-day last-click attribution. In a decision-making framework, PIE disagrees with RCT-based decisions in only 8–12% of campaigns, compared to 12–20% for last-click attribution. We conclude that PIE can help scale causal measurement from a limited number of RCTs to a large set of non-experimental campaigns. |
| JEL: | L81 L82 L86 M3 M30 M31 M37 |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:35044 |
| By: | Michaelides, Marios (Actus Policy Research); Mueser, Peter; Poe-Yamagata, Eileen; Nearchou, Paris; Ciobanu, Iuliana |
| Abstract: | The Reemployment Services and Eligibility Assessment (RESEA) program is a job-search assistance intervention targeting Unemployment Insurance (UI) claimants in the United States. The program requires new UI claimants to attend a counseling session at the start of their UI claims to: 1) undergo an eligibility review to confirm their compliance with UI work search requirements, and 2) receive customized reemployment services. This study reports the results of a large-scale randomized controlled trial (RCT) of the Iowa RESEA/RCM program conducted in 2022-2023, a period of strong labor market conditions. The program required participants to attend regular case management and job counseling meetings for the duration of their UI claims. Results show that the program increased take-up of job counseling services and significantly reduced UI duration and benefit amounts collected, generating substantial savings for the UI system. |
| Date: | 2026–04–07 |
| URL: | https://d.repec.org/n?u=RePEc:osf:socarx:z796a_v1 |
| By: | Gonzalo Ballestero; Hadi Hosseini; Samarth Khanna; Ran I. Shorrer |
| Abstract: | AI agents increasingly operate in multi-agent environments where outcomes depend on coordination. We distinguish primary algorithmic monoculture -- baseline action similarity -- from strategic algorithmic monoculture, whereby agents adjust similarity in response to incentives. We implement a simple experimental design that cleanly separates these forces, and deploy it on human and large language model (LLM) subjects. LLMs exhibit high levels of baseline similarity (primary monoculture) and, like humans, they regulate it in response to coordination incentives (strategic monoculture). While LLMs coordinate extremely well on similar actions, they lag behind humans in sustaining heterogeneity when divergence is rewarded. |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2604.09502 |
| By: | Asako Chiba (The University of Tokyo); Kazuya Haganuma (The University of Tokyo); Taisuke Nakata (The University of Tokyo); Thuy Linh Nguyen (The University of Tokyo); Reo Takaku (Hitotsubashi University) |
| Abstract: | We conducted an information provision experiment in April 2023 in Japan to investigate how different types of information affect people’s subjective assessment of COVID-19 related risks. The majority of respondents overestimate infection and fatality risks. Recent infection-related statistics lower risk perceptions if presented in percentage, but do not lower them if presented in levels. Providing pessimistic outlooks raises risk perceptions. We also find substantial heterogeneity in the response to information provision across various individual characteristics, such as age, gender, education, marital status, health status, COVID-19-related experiences, and vaccination status. |
| Date: | 2026–01 |
| URL: | https://d.repec.org/n?u=RePEc:cfi:fseres:cf616 |
| By: | Markus Lill (LMU Munich); Nastasia Gallitz (LMU Munich); Lucas Stich (University of Wuerzburg); Martin Spann (LMU Munich) |
| Abstract: | Platform endorsement badges (e.g., Amazon's Choice) are widely believed to shape consumer decisions, yet their effectiveness in complex retail environments---where endorsements compete with rankings, ratings, and other signals---remains not well understood. This article examines Amazon's product-level endorsements using a multi-method approach combining (1) a 50-day large-scale audit of more than 200, 000 search results spanning over 90, 000 products and (2) a lab-in-the-field experiment that manipulates badge visibility and placement in consumers' natural shopping context. The audit reveals that endorsements are rare (~1.3% of products) and disproportionately assigned to products with lower prices, higher ratings, and those sold or fulfilled by Amazon; receiving a badge is associated with greater search visibility and improved sales performance. The experiment shows that displaying the badge tends to increase click-through and add-to-cart likelihoods, whereas reassigning or masking it tends to reduce these behaviors; however, these effects---while economically meaningful---are estimated with uncertainty, consistent with a multi-cue environment in which endorsement competes with other signals such as search rank and Prime eligibility. Together, the findings indicate that platform endorsement badges shape consumer search and choice behavior even in information-rich retail settings. Implications are discussed for platform design, seller strategy, and regulatory oversight. |
| Keywords: | platform endorsements; consumer decision-making; digtial platforms; e-commerce experimentation; |
| JEL: | D12 D83 L86 M31 |
| Date: | 2026–04–01 |
| URL: | https://d.repec.org/n?u=RePEc:rco:dpaper:569 |
| By: | Louis-Pierre Lepage; Heather Sarsons; Michael Thaler |
| Abstract: | There is widespread opposition to affirmative action policies. We study whether personal disappointments shape preferences for such policies. Specifically, we test whether individuals' college admissions outcomes, relative to their expectations, influence their attitudes toward affirmative action policies. Using a retrospective survey among recent White and Asian college applicants, we find that disappointed individuals - those who were admitted to fewer schools than anticipated - action policies, and are more willing to donate to an anti-affirmative action organization. They also hold more negative views about the academic qualifications of under-represented minorities. To isolate the causal effect of "bad news" from selection, we conduct a complementary survey experiment with parents of future college applicants. We randomize whether parents receive information about their child's admissions prospects. Providing bad news to overconfident parents causes them to increase opposition to affirmative action and donate to an anti-affirmative action organization. Results suggest that some individuals attribute bad news to external factors, specifically policies that benefit out-groups. |
| Keywords: | affirmative action, policy views, disappointment |
| JEL: | I23 I28 D91 D83 J15 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:ces:ceswps:_12593 |
| By: | Johannes Wachs; Leonore R\"oseler; Tobias Gesche; Elliott Ash; Anik\'o Hann\'ak |
| Abstract: | Online platforms where volunteers answer each other's questions are important sources of knowledge, yet participation is declining. We ran a pre-registered experiment on Stack Overflow, one of the largest Q&A communities for software development (N = 22, 856), randomly assigning newly posted questions to receive an anonymous upvote. Within four weeks, treated users were 6.3% more likely to ask another question and 12.9% more likely to answer someone else's question. A second upvote produced no additional effect. The effect on answering was larger, more persistent, and still significant at twelve weeks. Next, we examine how much of these effects are due to algorithmic amplification, since upvotes also raise a question's rank and visibility. Algorithmic amplification is not important for the effect on asking additional questions, but it matters a lot for the effect on answering other questions. The increase in visibility increases the probability that another user provides an answer, and that experience appears to shift the poster toward broader community participation. |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2604.10360 |
| By: | Louis-Pierre Lepage; Heather Sarsons; Michael Thaler |
| Abstract: | There is widespread opposition to affirmative action policies. We study whether personal disappointments shape preferences for such policies. Specifically, we test whether individuals' college admissions outcomes, relative to their expectations, influence their attitudes toward affirmative action policies. Using a retrospective survey among recent White and Asian college applicants, we find that disappointed individuals—those who were admitted to fewer schools than anticipated—are relatively more likely to believe that affirmative action played an important role in their admissions outcomes, have the lowest support for affirmative action policies, and are more willing to donate to an anti-affirmative action organization. They also hold more negative views about the academic qualifications of under-represented minorities. To isolate the causal effect of "bad news" from selection, we conduct a complementary survey experiment with parents of future college applicants. We randomize whether parents receive information about their child's admissions prospects. Providing bad news to overconfident parents causes them to increase opposition to affirmative action and donate to an anti-affirmative action organization. Results suggest that some individuals attribute bad news to external factors, specifically policies that benefit out-groups. |
| JEL: | D83 J78 |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:35045 |
| By: | Terence Highsmith |
| Abstract: | Many matching markets feature unknown, dynamic arrivals of agents that must match immediately. A caseworker must match an abused child to a foster home, a hospital must assign a patient in critical condition to a room, or a city must place a homeless individual into a shelter. We design an online matching algorithm -- the Sequential Equilibrium Mechanism (SEM) -- that approximates large market equilibria to match arriving agents to objects. SEM is asymptotically efficient, fair, and strategy-proof with probability one. Our application plans to deploy a lab-in-the-field experiment where real caseworkers match vulnerable children to host homes, and we provide simulation evidence that SEM can substantially improve welfare. |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2604.12181 |
| By: | Jiawei Fu; Donald P. Green |
| Abstract: | How should researchers conduct causal inference when the outcome of interest is latent and measured imperfectly by multiple indicators? We develop a general nonparametric framework for identifying and estimating average treatment effects on latent outcomes in randomized experiments. We show that latent-outcome estimation faces two distinct noncomparability challenges. First, across studies, different measurement systems may cause estimators to target different empirical quantities even when the underlying latent treatment effect is the same. Second, within a study, different indicators may have different and possibly nonlinear relationships with the same latent outcome, making them not directly comparable. To address these challenges, we propose a design-based approach built around nonparametric bridge functions. We show that these bridge functions can be characterized and identified. Estimation relies on a debiasing procedure that permits valid inference even when the bridge functions are weakly identified. Simulations demonstrate that standard methods, such as principal components analysis and inverse covariance weighting, can generate spurious cross-study differences, whereas our approach recovers comparable latent treatment effects. Overall, the framework provides both a general strategy for causal inference with latent outcomes and practical guidance for designing measurements that support identification, comparability, and efficient estimation. |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2604.08681 |
| By: | Wilfried Youmbi Fotso; Xun Chen |
| Abstract: | We study Bayesian persuasion when information design is delegated to an intermediary who privately chooses the experiment subject to convex costs and would be incentivized by the principal via outcome-dependent transfers. We provide a sharp characterization of first-best implementability: implementing the first-best requires local affine alignment between the principal's and intermediary's reduced-form payoff indices on the posteriors induced by the target experiment, while a stronger global alignment condition guarantees implementability. Outside the global alignment condition, moral hazard typically prevents first-best implementation. We then characterize the second best: the principal's problem admits a virtual Bayesian persuasion representation in which the objective is distorted by a shadow cost proportional to the intermediary's valuation of posteriors. Under entropy costs, moral hazard compresses posterior dispersion relative to the first-best benchmark. In two-state environments with a binary-action receiver, the optimal second-best experiment has a tractable two-posterior form with explicit formulas for posterior endpoints and mixing weights, and the optimal transfer schedule is characterized in closed form as a triangular system in the shadow price, transfer gap, and participation constraint. A numerical example quantifies the compression: moral hazard reduces posterior spread by approximately 28 percent relative to first best under the baseline parameterization. |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2604.10006 |
| By: | Peter Andre (SAFE & Goethe University Frankfurt); Felix Chopra (Frankfurt School of Finance & Management); Luca Michels (University of Bonn); Johannes Wohlfart (University of Cologne & Max Planck Institute for Behavioral Economics Bonn) |
| Abstract: | Expectations are central to models of economic and financial decision-making. Yet in practice, individuals are often inattentive and, when asked, report fragile, context-dependent expectations that are only weakly linked to decisions. This raises the question to what extent they hold such expectations in the first place. Against this backdrop, we ask two questions: When people think about an economic issue, can they build on expectations they formed before? And does it matter if they cannot? We develop and validate a survey measure that distinguishes between individuals who can recall expectations formed in the past and those who must form expectations from scratch. We show that while many households have expectations about key economic variables, a large share of households do not — even among those close to decisions for which the expectation should be relevant. This matters: individuals without a previously-formed expectation (i) express expectations that are more context-dependent, (ii) update expectations more strongly but less persistently in response to new information, (iii) report expectations that are less relevant to decisions, and (iv) rely more on heuristics that do not require expectations when making economic decisions. |
| Keywords: | Expectations, Belief Formation, Previously-Formed, Context-Dependence, Learning, Decision Relevance, Heuristics |
| JEL: | C83 C91 D83 D84 D91 E71 G41 G53 |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:ajk:ajkdps:402 |