|
on Discrete Choice Models |
| By: | Datta, Nikhil |
| Abstract: | This paper investigates workers’ valuations of the job attributes that distinguish atypical work arrangements from traditional employee relationships. Using a vignette-based discrete-choice module fielded to samples of UK and US workers, I estimate willingness-to-pay distributions for contract security, work-related benefits, flexibility, autonomy, and tax treatment. Across both countries, workers place substantially higher value on job security and employment protections than on flexibility-related attributes. Interestingly, this pattern holds even among the self-employed: almost two-thirds of UK self-employed workers and just under half of US self-employed workers are estimated to prefer the traditional-employment bundle. Despite institutional differences between the UK and US, average valuations are closely aligned. The heterogeneity analysis shows that most observable characteristics account for only limited variation in preferences, but risk aversion and parenthood are notable exceptions, with both substantially increasing the value workers attach to contract security. |
| Keywords: | atypical work;discrete choice experiments;flexibility;job security;labour supply;self-employment;willingness to pay |
| JEL: | R14 J01 J1 |
| Date: | 2026–09–18 |
| URL: | https://d.repec.org/n?u=RePEc:ehl:lserod:141190 |
| By: | Kowalik, Zuzanna (Institute for Structural Research (IBS), University of Warsaw); Lewandowski, Piotr (Institute for Structural Research (IBS)); Pompei, Fabrizio (Università degli Studi di Perugia); Stichnoth, Holger (Zentrum für Europäische Wirtschaftsforschung (ZEW)); Szymczak, Wojciech (Uppsala University); Thil, Laurène (KU Leuven) |
| Abstract: | We study platform workers' preferences for social protection — in the form of sick pay and paid annual leave — distinguishing between workers who entered platform work out of economic necessity (the "pushed") and those attracted by its flexibility and advantages (the "pulled"). Using a discrete-choice experiment administered to nearly 800 taxi, delivery, and cleaning platform workers in Germany, Italy, and Poland, we find an average willingness to pay at 3.7% of earnings for sick pay coverage and 4.6% for paid annual leave. Preferences, however, differ sharply by motivation: pushed workers are willing to forgo on average 5.3% of earnings for sick pay, and 5.6% for paid annual leave, while pulled workers' valuations are substantially lower (3.1% and 4.2%, respectively). This difference is most pronounced among workers without access to these benefits, suggesting that the absence of protections is particularly consequential for those who entered platform work under greater economic constraint. The findings indicate that the welfare impact of extending social protections to platform workers — as proposed by the EU Platform Work Directive — might depend on the share of necessity-driven workers in the platform workforce. |
| Keywords: | platform work, willingness to pay, discrete choice experiment, social protection, job quality |
| JEL: | J32 J28 J81 |
| Date: | 2026–10 |
| URL: | https://d.repec.org/n?u=RePEc:iza:izadps:dp18967 |
| By: | Irene Botosaru; Isaac Loh; Chris Muris |
| Abstract: | We develop a framework for identification, computation, and inference in econometric models with a linear-in-measures representation. These models express maintained restrictions as moment conditions linear in the joint probability measure of observed and latent inputs, and map that measure linearly to the distribution of outputs, even with nonlinear outcome equations. We construct an adversarial discrepancy function whose zeros characterize the identified set for structural and counterfactual parameters. With finite output support, finite linear programs compute the discrepancy function or provide certified bounds even when latent inputs have infinite support, and a penalized bootstrap yields confidence sets with uniform per-point coverage. We apply the framework to two open cases in binary choice panels with fixed effects and discrete covariates: sequential exogeneity with unspecified conditional marginal error distributions, and known conditional marginal error distributions with unrestricted serial dependence. In an entry game with multiple equilibria, the framework recovers the known sharp identification region. |
| Date: | 2026–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2609.32076 |
| By: | Anna B. Schmidt |
| Abstract: | This work studies how Bayesian machine learning methods can be used for large-scale demand estimation with many product categories. I compare two model classes, a latent factorization model and a mixed logit model and two Bayesian estimation approaches, Markov Chain Monte Carlo (MCMC) and Variational Inference (VI). The analysis combines a simulation study with an application to supermarket scanner data. The results show that the latent factorization model benefits from information across categories and improves its predictive performance as the dimensionality of the choice environment increases, whereas the mixed logit model does not exhibit the same pattern. MCMC delivers the highest predictive accuracy but is computationally intensive. VI achieves slightly lower predictive performance while substantially reducing runtime. In the empirical application, VI also outperforms the mixed logit benchmark. These findings highlight a trade-off between accuracy and computational feasibility in multi-category demand estimation. |
| Date: | 2026–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2610.08409 |
| By: | Julien Monardo |
| Abstract: | The answers to many economic questions depend on the slope and curvature of demand. Estimating demand for differentiated products entails a trade-off between flexibility and scalability. Existing nonparametric approaches face a curse of dimensionality in the number of products. I develop a scalable nonparametric approach using market-level data that overcomes this curse through economically motivated restrictions embedded in many standard models. Simulations and an application to the U.S. beer market demonstrate flexibility in demand slope and curvature, with consequential differences in counterfactual price responses. Estimation in the application takes seconds, versus several hours for a random coefficient nested logit model. |
| Date: | 2026–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2610.08467 |
| By: | Carmelo Rodríguez à lvarez (ICAE Instituto Complutense de Análisis Económico, Depto. Análisis Económico y Economía Cuantitativa, Universidad Complutense de Madrid) |
| Abstract: | We study social choice functions that select sets of alternatives, taking voters' preferences over sets as input, when voters evaluate sets according to conditional expected utility obtained from Bayesian updating of a prior. Barberà et al. (2001) shows that under a uniform common prior both dictatorial and bidictatorial rules satisfy strategy-proofness and unanimity. We show that this multiplicity is a knife-edge case. If at most one voter holds a uniform initial prior, dictatorship is the unique rule satisfying these properties. The argument links set-valued rules to random dictatorship and shows that the probability ratios imposed by Bayesian updating from non-uniform priors are incompatible with combining the top choices of multiple voters. |
| Keywords: | Strategy-proofness; Social choice functions over sets; Random dictatorship; Probability assessments. |
| JEL: | C71 D82 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:ucm:doicae:2602 |
| By: | Herbert Dawid; Ali Kashefi; Gerd Muehlheusser |
| Abstract: | Should platforms be held legally responsible for harm caused by products or services offered by third-party sellers? We address this question in a novel setting in which product safety (the likelihood of accidents) is determined by third-party sellers, while the consumer price is set by a monopolistic platform that pays sellers a commission. Relevant examples include platforms providing ride-hailing services (e.g. Uber and Lyft) or health care services (e.g. Teladoc). Sellers decide whether or not to become active on the platform, and consumers' willingness to pay for the platform good increases with the number of active sellers (platform externality). When an accident occurs, the harm is divided between the platform and the respective seller according to a liability rule set ex ante by a welfare-maximizing policymaker. Sellers' incentive to provide high product safety is driven by market demand, which in turn is determined by the platform's pricing scheme, balancing the trade-off between market revenue and expected liability costs. The platform may optimally set a price below the monopoly level to boost demand, thereby triggering active sellers to provide high product safety (market channel). The socially optimal liability rule is dichotomous: it either assigns no liability to the platform, or the maximum liability share for which the platform exploits the market channel to induce high product safety from the sellers. Holding the platform liable is socially desirable only when it induces lower consumer prices. |
| Keywords: | platform liability, consumer pricing, product safety, product liability |
| JEL: | K13 L13 L15 L86 D42 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:ces:ceswps:_13033 |
| By: | Joachim Freyberger; Julius Kappenberg |
| Abstract: | Economic theory frequently implies linear inequality restrictions on parameters or functions of interest. A common way to impose such restrictions is to project an unrestricted estimator onto the feasible set. Projection estimators arise naturally from constrained least squares, instrumental variables, generalized method of moments, maximum likelihood, and related extremum procedures. When the sampling covariance, loss function, and projection criterion induce different geometries, the choice of projection geometry can substantially affect risk. We study this choice in a fixed-dimensional local Gaussian experiment under quadratic loss. At exact-boundary configurations where only one maintained inequality binds, inverse-covariance projection is pointwise optimal. When at most two inequalities are locally relevant, it weakly improves on the unrestricted estimator throughout the corresponding local experiment and is minimax over exact-boundary configurations. For an arbitrary number of inequalities, we provide a sufficient condition for boundary minimaxity, but show by counterexample that inverse-covariance projection need not be boundary minimax once three inequalities can bind. Motivated by these results, we propose selecting the projection geometry to minimize worst-case boundary risk subject to a local no-harm condition relative to unrestricted estimation. We develop a feasible implementation and study its finite-sample performance in simulations and an application to gasoline demand. |
| Date: | 2026–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2609.32725 |
| By: | Bertrand Achou; Mehdi Ammi; Minchung Hsu; Minjoon Lee; Tomoe Naito; Keisuke Otsu |
| Abstract: | Population aging and shortages of long-term care workers have increased interest in caregiving robots and information and communication technology (ICT). This paper provides novel large-scale evidence on older individuals’ perceptions of such technologies, drawing on a custom-designed internet survey of 4, 314 Japanese individuals aged 55 to 75. Respondents choose between two otherwise identical nursing homes: one in which all care is provided by human caregivers, and one in which robots and ICT are used to provide a substantial share of care in one randomly assigned domain (communication, monitoring, or mobility assistance), while other care remains human-provided. A majority of respondents (65 percent) prefer nursing homes where robots and ICT are used. Acceptance differs substantially across care domains: It is highest for mobility assistance, intermediate for monitoring, and lowest for communication. The average willingness to pay (WTP) for a nursing home where robots and ICT provide a substantial share of care is sizable, at 8 percent above the typical nursing home fee. Prior awareness of caregiving robots is positively associated with both acceptance and WTP, pointing to the potential role of information in shaping older individuals’ preferences. Acceptance and WTP are also strongly associated with sentiments toward robots and ICT expressed in attitude questions, providing support for the internal consistency of survey responses. A back-of-the-envelope calculation offers suggestive evidence that the benefits of introducing robots and ICT can be much larger than the costs, in particular for mobility-aid robots. |
| Keywords: | Long-term care, caregiving robots and ICT, nursing homes, survey. |
| JEL: | I10 J14 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:rsi:irersi:27 |
| By: | Federico Boffa (Faculty of Economics and Management, Free University of Bozen-Bolzano, Bolzano, Italy); Chiara Farronato (Harvard Business School, Boston, MA, USA); Alessandro Fedele (Faculty of Economics and Management, Free University of Bozen-Bolzano, Bolzano, Italy); Eugenio Levi (Link Campus University, Rome, Italy); Andrea Mantovani (TBS Business School, Toulouse, France) |
| Abstract: | Large language models (LLMs) are increasingly used by consumers to search for information, compare alternatives, and obtain purchasing advice. As conversational AI systems progressively complement traditional search engines and recommender systems, they may influence consumer choice by determining which products receive visibility and attention. Despite their growing importance, little is known about how LLMs construct product recommendations, whether they concentrate attention on a narrow set of products, or whether vertically integrated providers favor affiliated products and services. This paper conducts an audit study of leading LLMs, including ChatGPT, Gemini, Claude, DeepSeek, Grok, and Mistral. Using standardized prompts and repeated queries, we collect recommendations across a broad range of products and service categories that differ in market characteristics and the presence of vertically integrated products. We characterize recommendations along several dimensions, including breadth, first-position concentration, prominence, best-overall selections, and personalization. We find substantial heterogeneity across models in the breadth and structure of their recommendations. At the same time, first-position recommendations are highly concentrated in a substantial subset of markets. In 20 of the 100 product and service categories, a single product is listed first in more than three quarters of responses. We also find that some LLMs are likely to recommend products affiliated with their providers. |
| Keywords: | large language models; empirical audit; product recommendations; AI; algorithmic bias |
| JEL: | L15 L41 L86 D83 |
| Date: | 2026–08 |
| URL: | https://d.repec.org/n?u=RePEc:net:wpaper:2609 |
| By: | Luigi Montrucchio; Lorenzo Stanca |
| Abstract: | We derive sufficient conditions for the existence, non-existence, uniqueness and global attractivity of stochastic recursive utilities. Our approach uses abstract order-theoretic fixed-point arguments, allowing us to accommodate a large class of Markov consumption-growth models with unbounded per-period consumption and a broad range of recursive preferences, including non-expected utility specifications. We illustrate the usefulness of the framework through macro-finance applications to widely used consumption-growth models under several preference specifications, including Epstein--Zin preferences, multiple priors, smooth ambiguity, and Hansen--Sargent robust-control preferences. In each application, our general conditions reduce to a simple, economically interpretable parameter test that shows how fundamental parameters influence existence and uniqueness. While uniqueness of recursive utility can be difficult to guarantee in unbounded environments, in a Markov setting with non-unit elasticity we establish uniqueness within the class of recursive utilities having a so-called separability property. |
| Date: | 2026–10 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2610.03157 |
| By: | Kexin Chen; Tao Lin; Jianwei Huang |
| Abstract: | AI advice increasingly enters human decision workflows before costly information acquisition and final action. This paper studies how such advice affects decision quality, measured by the gross utility of the final action. Advice shifts beliefs before the decision maker chooses an acquisition experiment with uniformly posterior-separable (UPS) costs. The analysis characterizes gross effects through the continuation gross value generated by net-optimal acquisition. Convexity of this value function is necessary and sufficient for every AI signal to weakly improve gross utility at every prior, while nonconvexity allows a binary signal to cause gross harm. The criterion gives tight state-space results. Every AI signal is weakly gross-improving in binary-state problems under arbitrary actions, payoffs, and UPS costs. A minimal three-state problem can generate gross harm through concentrated shutdown, where valuable acquisition stops. For classification problems with Shannon acquisition costs, the criterion becomes an active-set slope test. It guarantees weak gross improvement for homogeneous classification and all three-state weighted classification problems, while a minimal four-state failure arises from attention dilution, where acquisition shifts away from high-stake distinctions. Along a Blackwell-increasing mixture path, gross utility can first decrease and then increase. With endogenous interpretation, cheaper interpretation can reduce gross utility. |
| Date: | 2026–09 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2609.35137 |