|
on Discrete Choice Models |
| By: | Contreras, Ivette; Dinarte, Lelys; Palacios-Lopez, Amparo; Costa, Valentina; Romero, Steffanny |
| Abstract: | This paper studies job preferences among women in rural and peri-urban areas in El Salvador using a discrete choice experiment. Drawing on focus group insights, the analysis varies wages and five non‑wage job attributes—contract status, experience requirements, commute safety, residential address disclosure, and childcare availability—and estimates preferences using a mixed logit model. Women are willing to forgo substantial earnings for jobs that offer a safe commute, accessible childcare, and lower barriers to entry. Formal contracts play a limited role in job choice in this high informality context. Preferences are heterogeneous: risk averse and rural women place a particularly high premium on safety and childcare, while younger and less risk averse women are more sensitive to entry barriers and address related stigma. The results highlight the importance of labor market frictions that prevent wages from compensating for job disamenities and suggest that policies targeting safety, childcare, and access may be more effective than contract formalization in expanding women’s employment opportunities. |
| Date: | 2026–03–16 |
| URL: | https://d.repec.org/n?u=RePEc:wbk:wbrwps:11333 |
| By: | Duncan Mortimer (Centre for Health Economics, Monash Business School, Monash University, Australia); Rohan Sweeney (Centre for Health Economics, Monash Business School, Monash University, Australia); Amelia Turagabeci (College of Medicine, Nursing and Health Sciences, Fiji National University); Sepesa Rasili (Consultant & Fiji Council of Social Services member) |
| Abstract: | Climate change is forcing difficult choices between in-place adaptation and relocation for Pacific Island communities, yet policy responses often rely on participatory planning frameworks that privilege louder voices or implicitly assume a consensus of preferences. We surveyed 476 adults across 25 at- risk Fijian villages using a discrete choice experiment to understand how individuals evaluate trade- offs between alternative future living arrangements, including location, services, housing, income opportunities, climate risk, and cultural connection. Our analysis identifies three distinct preference types—movers, stayers, and adapters—with sometimes conflicting priorities. While movers and adapters are generally willing to relocate to climate-resilient locations, stayers prefer to remain in their existing villages even in the absence of significant adaptation investment. These divergent preferences reveal relocation and in-place adaptation as spatially constrained and contested choices. Uncoordinated household-level decisions by movers and adapters risk redistributing rural populations across to urban centres and fragmenting communities. Preservation of connection to community and place may therefore require deliberate coordination and compromise at the community level, including the design of new climate-resilient settlements that accommodate the preferences of stayers. Recognising heterogeneous preferences and the limits of consensus-based participation is essential for designing community adaptation pathways that are socially, culturally, and spatially just and acceptable. |
| Keywords: | climate change adaptation, climate-induced relocation, Pacific Island communities, community preferences, discrete choice experiment |
| JEL: | Q54 R23 Q51 O15 D71 |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:mhe:chemon:2026-04 |
| By: | Gokan, Toshitaka; Thisse, Jacques-François (Université catholique de Louvain, LIDAM/CORE, Belgium); Zhu, Xiwei |
| Abstract: | We develop a model in which offline and online transportation costs, online shopping disutility cost and consumer taste heterogeneity, and online and offline retailers' pricing policies interact to determine the equilibrium retail format that emerges from firms' and consumers' choices. This is done by combining spatial pricing and discrete choice theory within a unified game-theoretic framework. We study the industry equilibrium, as well as the corresponding consumer surplus and total welfare. Our results show that firms' choices of a retail format and consumers' decision to buy from an offline or online firm often depend on consumers' locations relative to firms'. Comparing aggregate consumer surpluses shows that consumers prefer online to alternative channels when they are sufficiently heterogeneous, but this need not be so when heterogeneity is weak. When consumers' tastes are heterogenous enough, the retail format maximizing total welfare depends on the value of the distaste costs of online purchase. Thus, the nature of products supplied by retailers is likely to affect the socially desirable retailing system through the degree of product differentiation. |
| Keywords: | Offline retailing ; online retailing ; spatial price policies |
| JEL: | L13 L81 R10 |
| Date: | 2025–03–13 |
| URL: | https://d.repec.org/n?u=RePEc:cor:louvco:2025008 |
| By: | Gschwendt, Christian (University of Bern); Viarengo, Martina (Graduate Institute of International and Development Studies, Geneva); Zollner, Thea S. (University of Bern) |
| Abstract: | The economic impact of technological change will critically depend on how future workers invest in their human capital. Yet, little is known about how future workers themselves evaluate and choose their educational and occupational paths in light of emerging technologies. This paper examines how adolescents currently at the school-to-work transition stage value working with generative artificial intelligence (GenAI) in their future occupations, and how automation risk and opportunities for continuing education shape these preferences. We field a discrete-choice experiment among a nationally representative sample of over 7, 000 Swiss adolescents aged around 15. We find that adolescents generally exhibit an aversion to collaborating with GenAI at work, with females consistently more averse than males. However, preferences are nuanced: adolescents welcome greater GenAI collaboration, provided that GenAI usage levels remain moderate and that it is not accompanied by increases in job automation risk. Finally, our findings suggest that AI-related educational opportunities in occupations improve attitudes towards working with GenAI across genders. |
| Keywords: | occupational choice, gender gaps, GenAI, choice experiment, continuing education, automation risk |
| JEL: | I24 J24 O33 |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:iza:izadps:dp18456 |
| By: | Fuchs, Anna; Haensch, Anna-Carolina; Weber, Wiebke |
| Abstract: | Designing survey questions is easy; however designing good survey questions is a complex task. Large language models (LLMs) have the potential to support this task by automating parts of the item-generation process, but their suitability for survey research has not yet been systematically evaluated. Published research in this area remains sparse, and little is known about the quality and characteristics of survey items generated by LLMs or the factors influencing their performance. This work provides the first in-depth analysis of LLM-based survey item generation and systematically evaluates how different design choices affect item quality. Five LLMs, namely GPT-4o, GPT-4o-mini, GPT-oss-20B, LLaMA 3.1 8B, and LLaMA 3.1 70B, were used to generate survey items on four substantive domains: work, living conditions, national politics, and recent politics. We additionally evaluate three prompting strategies: zero-shot, role, and chain-of-thought prompting. To assess the quality of the generated survey items, we use the Survey Quality Predictor (SQP), a tool for estimating the quality of attitudinal survey items based on codings of their formal and linguistic characteristics. To code these characteristics, we used an LLM-assisted procedure. The findings show striking differences in survey item characteristics across the different models and prompting techniques. Both the choice of model and the prompting technique employed influence the quality of LLM-generated survey items. Closed-source GPT models generally produce more consistent items than open-source LLaMA models. Overall, chain-of-thought prompting achieved the best results. GPT-4o, GPT-4o-mini, and LLaMA 3.1 70B achieved similar item quality, while the LLaMA model showed greater variability. |
| Date: | 2026–03–12 |
| URL: | https://d.repec.org/n?u=RePEc:osf:socarx:fzn7t_v1 |
| By: | De Munck, Thomas (Université catholique de Louvain, LIDAM/CORE, Belgium); Tancrez, Jean-Sébastien (Université catholique de Louvain, LIDAM/CORE, Belgium); Chevalier, Philippe (Université catholique de Louvain, LIDAM/CORE, Belgium) |
| Abstract: | We consider the problem of a ride-hailing platform (e.g., Uber, Lyft) that connects supply with demand over a network of locations. To this aim, the platform makes pricing, driver repositioning, and customer admission decisions. Customers are impatient and have distinct willingness to pay. Drivers can be repositioned by the platform, or can choose to relocate to other locations by themselves. We formulate this problem as a discrete-time Markov decision process and propose a transfer learning approach to find an efficient policy. Our approach first derives a rolling-horizon strategy by repeatedly solving a deterministic optimization problem. Then, two neural networks are pretrained to replicate the strategy and learn the associated value function. Finally, the policy is further improved through deep reinforcement learning (DRL). Using data from New York City, we apply our approach to networks of up to 20 locations. The results show that our approach outperforms alternative DRL algorithms and rolling-horizon strategies while reducing computation time and stabilizing learning. We also explore the interplay between pricing, driver repositioning, and customer admission, providing insights into their respective roles. |
| Keywords: | Transportation ; Ride-hailing platforms ; Pricing and repositioning decisions ; Transfer learning ; Deep reinforcement learning |
| Date: | 2025–02–01 |
| URL: | https://d.repec.org/n?u=RePEc:cor:louvco:2025004 |
| By: | Mauleon, Ana (Université catholique de Louvain, LIDAM/CORE, Belgium); Vannetelbosch, Vincent (Université catholique de Louvain, LIDAM/CORE, Belgium) |
| Abstract: | We consider priority-based school choice problems where students can form two types of matches: public matches observed by everyone and private matches generally not observed by others. We introduce the notion of rationalizable conjectural stable (RCS) matching, which generalizes Gale and Shapley (1962)’s stability notion to accommodate private matches. We partially characterize RCS matchings and we show that the Efficiency-Adjusted Deferred Acceptance (EADA) matching is RCS when private and public matches are perfect substitutes. Finally, we extend the definition of RCS matching to school choice problems when private and public matches are imperfect substitutes. |
| Keywords: | School choice ; Private information ; Private matchings ; Stability ; Rationalizability |
| JEL: | C70 C78 D82 |
| Date: | 2025–03–20 |
| URL: | https://d.repec.org/n?u=RePEc:cor:louvco:2025007 |
| By: | Marcella Alsan; Joshua Schwartzstein; Stefanie Stantcheva |
| Abstract: | Lethal firearm ownership is deeply polarizing in the United States. We show that beneath this polarization, owners and non-owners share a common objective — safety — but disagree sharply about whether lethal firearms achieve it. Using an original survey of more than 5, 400 respondents combined with randomized experiments, we document that owners feel safe and confident with firearms, while non-owners on balance feel less safe around them and perceive large private costs and social harms. Demand for lethal firearms is nonetheless potentially large and growing: one-third of non-owners express interest in acquiring one — these individuals report the lowest day-to-day safety — while very few owners would consider reducing their holdings. Persuading owners to relinquish firearms without any replacement appears unrealistic; the more tractable margins may be safe storage and non-lethal substitution for additional purchases. We organize these patterns through a framework centered on a perceived safety possibilities frontier (SPF) — the safety outcomes a household believes achievable with different combinations of lethal and non-lethal tools. Households may differ in firearm demand because they face different risk environments, weigh protective benefits against harms differently, or hold different beliefs about the frontier. Our descriptive evidence points to heterogeneous beliefs as important drivers, suggesting that levers such as information could shift the perceived frontier. These patterns motivate three experimental treatments: one on the private legal/medical costs of lethal firearm ownership, and two on a non-lethal firearm (NLFA), with and without a conservative pundit’s endorsement. The private-cost treatment increases concern about harms among all respondents and support for safe storage policies, and modestly raises stated willingness to keep lethal firearms locked. NLFA treatments raise willingness to pay for an NLFA, to keep lethal firearms locked, and support for incapacitating over lethal firearms and for policies encouraging NLFAs. These effects are largely persistent. Importantly, NLFA information does not increase willingness to reduce lethal firearm ownership but does increase willingness to store lethal firearms safely. Our results suggest that many owners perceive the SPF differently from nonowners, neglecting harms or less-lethal alternatives, yet remain open to such tools. Overall, individuals share a common goal — safety — yet disagree about the means. Although these disagreements appear entrenched, people remain receptive to alternatives that might command broader agreement. |
| JEL: | H0 H80 I1 K0 |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:34962 |
| By: | Zhaoqi Zang; David Z. W. Wang; Xiangdong Xu; Shaojun Liu |
| Abstract: | Time variability is a pervasive feature of mobility services and a major source of welfare loss. Although literature has quantified the cost of time variability (COTV), it remains theoretically unclear how bad time variability can be in the worst case. Without such a benchmark, quantified variability costs lack a principled reference for assessing whether they are economically meaningful. Meanwhile, this benchmark is critical for strategic prioritization in transport appraisal, service design, and pricing -- particularly in early-stage decision making where detailed valuation is often infeasible. To fill this gap, this paper develops an expected utility (EU) framework to quantify the cost of time (COT) and COTV, establishing theoretical upper bounds on the ratio $COTV/COT$. For users with quadratic utility, we show $COTV/COT \le 1/2 CV^2$, where $CV$ is the coefficient of variation of service time. For Poisson processes, a common assumption, this bound simplifies to $COTV/COT \le 1/2$, implying the total cost of a stochastic service is at most 1.5 times that of an otherwise identical deterministic service. In more general settings, the ratio depends on three interpretable factors: $CV$ and users' second- and third-order risk preferences, captured by relative risk aversion (RRA) and relative prudence (RP). We identify benchmark values of RRA and RP that characterize preferences over mean-, variance-, and skewness-related reductions. Our analysis extends to non-EU frameworks, including dual theory and rank dependent utility, showing that key structural insights remain robust. By quantifying the cost induced by time variability and the $COTV/COT$ ratio, this study provides a data-light benchmark for early-stage decision making and a principled upper bound on users' willingness to pay for reliability improvements, informing the pricing and design of reliability-oriented services. |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2603.09142 |
| By: | Conti, Alice; Di Matteo, Fabio Lokwani; Candeloro, Giulia; Lancisi, Lorenzo; Sacco, Pier Luigi |
| Abstract: | Most economic models of decision-making assume that individuals maintain comprehensive mental representations of possible world states and compute expected utilities across these representations. The cognitive demands of such exhaustive state-space evaluation appear difficult to reconcile with known constraints on working memory, attention, and neural computation, motivating alternative accounts of how humans navigate complex choice environments. This review examines the hypothesis that neural systems supporting narrative processing contribute substantially to real-world decision-making. We synthesize evidence from three converging lines of research: hierarchical temporal processing in the cortex, the integrative functions of the default mode network, and inter-brain synchronization during naturalistic communication. We acknowledge the ongoing debates about the interpretation of these findings, but we argue that the neural architecture supporting narrative comprehension and generation offers cognitive efficiency advantages relevant to decision-making, including dimensionality reduction through causal structuring, integration of emotional and contextual information, and facilitation of social coordination through shared mental models. Rather than claiming that narrative processing constitutes the exclusive mechanism for decision-making, we propose that it represents one important component of a broader cognitive toolkit that also includes heuristic strategies, model-based planning, and habitual responses. We examine how this perspective relates to phenomena in cognitive economics, including context-dependent preferences, framing effects, and the propagation of economic beliefs through populations. By integrating findings from cognitive neuroscience with decision science, we aim to contribute toward a more biologically informed understanding of human choice behavior and identify key questions for future empirical investigation. |
| Date: | 2026–02–28 |
| URL: | https://d.repec.org/n?u=RePEc:osf:socarx:zxcvg_v1 |
| By: | Jason Delaney (Georgia Gwinnett College); Sarah Jacobson (Williams College); Thorsten Moenig (Temple University); |
| Abstract: | "Is the assumption that people automatically know their own preferences innocuous? We present a theory that explores the implications of having to discover one's preferences. We show that if tastes must be learned through experience, preferences for some goods will be learned over time, but preferences for other goods will never be learned. This is because sampling a new item has an opportunity cost. Learning is less likely for people who are impatient, risk averse, low income, or short-lived, and for consumption items that are rare, are expensive, must be bought in large quantities, or are initially judged negatively relative to other items. Choices will eventually stabilize, but they need not stabilize at true preferences. A pessimistic bias about untried goods should increase with time. Agents will make choice reversals during the learning process. Welfare loss from suboptimal choices will decline over time but need not approach zero. Overall, our results imply that undiscovered preferences could confound interpretation of choice data of all kinds and could have significant welfare and policy implications." |
| Keywords: | discovered preferences, preference stability, learning |
| JEL: | D83 |
| Date: | 2025–01–01 |
| URL: | https://d.repec.org/n?u=RePEc:wil:wileco:2025_113 |
| By: | Millimet, Daniel (Southern Methodist University); Paloyo, Alfredo (University of Wollongong) |
| Abstract: | Empirical researchers often replace latent constructs with composite indices, treating this as a neutral data-processing step. We prove this is a consequential identification choice. Our (near) impossibility theorem demonstrates that no linear index can guarantee consistent estimates of all parameters in multiple regressions. We show that proxy indices induce residual confounding, while popular weighting schemes introduce nonclassical measurement error with method-dependent biases. Using simulations and 2024 U.S. election data, we reveal that substantive conclusions are often artifacts of the chosen index. We argue that measurement models require the same formal scrutiny as other identification strategies. |
| Keywords: | latent variables, composite indices, measurement models, errors-in-variables, proxy variables, political economy, community health |
| JEL: | C13 C18 C43 D72 |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:iza:izadps:dp18454 |
| By: | Michael Finus (University of Graz, Austria; University of Bath, United Kingdom); Paolo Zeppini (Université Côte d'Azur, CNRS, GREDEG, France) |
| Abstract: | We introduce the concept of green lifestyles in an economic discrete choice model of consumption behaviour. Agents behave in either a ‘selfish’ or ‘pro-social’ way by choosing different degrees of internalisation of environmental damage from the consumption of an environmentally harmful good. Pro-social behaviour means lower consumption, and is rewarded with warm-glow. Moreover, the agents’ decision is influenced by social norms, which endogenously depend on aggregate choices. The model is developed in a dynamic framework, allowing agents to switch behaviour. Our results show that conventional measures limiting consumption at an individual level may increase consumption at the aggregate level. We characterise social tipping points for sustainability transitions in terms of equilibria bifurcations and hysteresis of population dynamics. The model is extended in different directions, with different types of social influence and with a state dependent warm-glow. This more complicated decision environment gives alternative regimes with either dampening or self-reinforcing feedback in decisions. Three scenarios are identified: for strong social norms positive feedback leads to multiple equilibria. For moderate social norms there is a unique equilibrium. For weak social norms, we obtain periodic dynamics of behaviours. In particular, more informed choices and lower variability across agents are ‘destabilising’, leading to periodic dynamics or multiple equilibria. |
| Keywords: | discrete choice; social interactions; sustainable consumption; transitions; warm-glow |
| JEL: | C62 D62 Q56 |
| Date: | 2026–03 |
| URL: | https://d.repec.org/n?u=RePEc:gre:wpaper:2026-08 |