|
on Discrete Choice Models |
| By: | Fitzsimmons, Jill |
| Abstract: | The expansion of utility-scale solar energy is expected to require millions of acres of land in the United States, increasing pressure on agricultural landscapes and raising concerns regarding farmland preservation, rural identity, and food production. Agrivoltaics, which combines agricultural production and solar energy generation on the same parcel of land, has been proposed as a strategy to reduce land-use conflicts between renewable energy development and agriculture. While previous studies find that the public generally prefers agrivoltaics to conventional utility-scale solar, little evidence exists regarding how consumers value agrivoltaics relative to continued agricultural production or other competing land uses. This study evaluates public acceptance of agrivoltaics using a discrete choice experiment administered to 6, 332 adult consumers across thirty-seven U.S. states with commercial apple production. Respondents repeatedly chose whether to purchase apples from farm stands associated with traditional apple production, agrivoltaics, utility-scale solar, or residential development. Choice attributes also varied the share of land-use revenues accruing to farmers, subsidy status, travel time, and apple price. The survey incorporated randomized information treatments and elicited measures of risk preferences and social preferences. Results indicate substantial heterogeneity in willingness to pay for agrivoltaics. On average, respondents prefer maintaining agricultural production relative to converting farmland to utility scale solar or residential development. Willingness to pay for agrivoltaics is not statistically different from continued agricultural production in the full sample. However, urban respondents exhibit a positive willingness to pay for agrivoltaics, whereas rural respondents exhibit a negative willingness to pay. Higher farmer revenue shares significantly increase support for land-use change, while the absence of subsidies reduces acceptance of solar-related development. Additional analyses indicate that support for agrivoltaics varies systematically with political ideology and risk preferences. The findings suggest that public acceptance of renewable-energy development on agricultural land depends not only on energy production outcomes but also on perceptions of agricultural land as a provider of public goods and on the distribution of economic benefits associated with land-use change. Policies that preserve agricultural functions, support increased and transparent revenues for farmers, and account for place-based differences in public preferences may improve acceptance of agrivoltaic development during the renewable-energy transition. |
| Keywords: | Environmental Economics and Policy |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:ags:aaea26:404505 |
| By: | Liu Liu; Dan Zhang |
| Abstract: | Tabular foundation models (TFMs) generate predictions on structured data via in-context learning, without task-specific estimation. We ask whether TFMs can be effectively applied to discrete choice, a central demand estimation framework in marketing and operations, and find that directly applying TFMs yields limited performance. The gap is structural: TFMs assume row-independent observations, whereas discrete choice is inherently set-valued and subject to persistent consumer preference heterogeneity. We propose a reformulation that encodes both choice-set dependence and individual heterogeneity within a row-based learning framework. Evaluated on a yogurt scanner panel, individual-level heterogeneity encoding is the dominant driver of predictive accuracy. The best reformulation outperforms hierarchical Bayesian estimation by 8\% in holdout log-likelihood and 3.6\% in hit rate, running 16 times faster, a practical advantage for large-scale demand estimation. The advantage is largest in the medium-data regime (10--40 purchase occasions per consumer), where parametric Bayesian shrinkage most distorts estimates for atypical consumers. Fine-tuning on population choice data provides additional gains for consumers with shallow purchase histories, where in-context learning has limited individual-specific signal to condition on. These results establish a principled approach for applying foundation models to consumer choice problems more broadly. |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2607.13314 |
| By: | Shrestha, Aavash; Zhang, Qi; Etienne, Xiaoli; Tejeda, Hernan; Trujillo-Barrera, Andres |
| Abstract: | The food sector faces growing pressure to reduce greenhouse gas emissions, and food companies increasingly communicate climate actions through product labels and sustainability claims. Yet low-carbon claims can be achieved through substantively different strategies. Some firms directly reduce emissions in production or along the supply chain, while others rely on carbon offsets that compensate for emissions through external projects. This study examines whether U.S. consumers distinguish among direct emissions reduction, carbon offsetting, and a combined reduction-and-offset approach in the context of dairy products. We conduct an online discrete choice experiment in which consumers choose between two 8-ounce blocks of sharp cheddar cheese and a neither option. Product attributes include brand type, price, climate mitigation claim, and certification source. Using a conditional logit model with respondent-clustered standard errors, we estimate willingness to pay for alternative climate claims and certification sources. Results show that consumers value all three climate mitigation claims relative to no claim, but they do not treat them as equivalent. Under both certification sources, the combined reduction-and-offset claim receives the highest willingness to pay, followed by direct emissions reduction and carbon offsetting alone. Similarly, across all the climate mitigation claims, government certification generates an additional premium relative to independent third-party certification. These findings suggest that the market value of climate-related food labels depends not only on the presence of a low-carbon claim, but also on the substantive mitigation strategy and the credibility of verification. |
| Keywords: | Institutional and Behavioral Economics |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:ags:aaea26:404432 |
| By: | Youngjin Hong; In Kyung Kim; Frank Verboven |
| Abstract: | We develop a framework to compare the relative effectiveness of subsidizing alternative emission-reducing technologies. We show that an intermediate technology may reduce emissions more effectively than the cleanest technology if it induces sufficiently greater substitution away from the prevailing high-emission technology. We apply the framework to the South Korean passenger vehicle market using a demand model that incorporates mileage heterogeneity, an important determinant of fuel-type choice. First, reallocating existing subsidies from battery electric vehicles (BEVs), the cleanest technology, to hybrid electric vehicles (HEVs), an intermediate technology, would reduce total greenhouse gas emissions by an additional 47%. Second, for a BEV-focused subsidy policy to outperform an HEV-focused policy, the carbon intensity of electricity generation would need to fall by approximately 45%. Our findings suggest that HEV subsidies remain more effective than BEV subsidies until consumers become sufficiently willing to switch to BEVs or electricity generation becomes sufficiently decarbonized. |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2607.14446 |
| By: | Jennifer Zou |
| Abstract: | We study how a representative sample of United States adult AI-assistant users (n=1, 999; June 2026) choose among platforms, allocate tasks across them, evaluate provider trustworthiness, and value data-handling features. Estimates are weighted to the AI-user population using external adoption benchmarks. Four patterns emerge. The market is concentrated but internally differentiated: ChatGPT is the primary assistant for 58% of users and Gemini for 25%, yet smaller platforms hold defensible task niches--Claude captures a third of coding tasks despite a 7% overall share. Task allocation is thus organized by platform far more than by user, and technical use falls steeply with age. Trust is earned through use rather than reputation: Claude is ranked most trustworthy in every head-to-head among users of both platforms, and shows by far the largest gap between how its users and non-users rate it. Finally, privacy concern is near-universal but action is gated by knowledge, not concern; in a choice experiment users pay most to keep humans--not models--out of their conversations ($11.20/month), with valuations rising in task sensitivity. |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2607.15134 |
| By: | Mwebaze, Paul; Khanna, Madhu; Majeed, Fahd; Mishra, Bijesh; Miao, Ruiqing |
| Abstract: | Utility-scale solar expansion across U.S. croplands is intensifying food–energy land-use conflicts, particularly in the Midwest. Agrivoltaics (AV), the co-location of photovoltaic systems with agricultural production, offers a potential land-use solution, but its scalability depends on farmer adoption. This study examines how financial incentives, operational requirements, and behavioral factors shape AV adoption decisions using a discrete choice experiment with 177 farmers across 12 Midwestern states (708 choice observations). The experiment evaluates attributes that directly affect farm profitability, risk exposure, and management complexity, including lease payments, income variability, land share under panels, and requirements for new equipment or crop switching. We model adoption as a joint discrete–continuous decision, capturing both the likelihood of adoption and the extent of land allocation to AV. Results show that higher lease payments significantly increase both the probability of adoption and land allocation, highlighting the importance of stable and attractive compensation. In contrast, operational complexity, particularly requirements for new equipment, reduces adoption and generates substantial implicit costs in willingness-to-accept (WTA) estimates. Behavioral characteristics also matter: farmers with higher discount rates are less likely to adopt, whereas risk-tolerant farmers are more willing to participate. Latent class analysis further reveals substantial heterogeneity, distinguishing a majority of cautious “moderate adopters” from a smaller group of “high-intensity adopters” that is more responsive to stable income streams and higher lease compensation. Overall, the findings suggest that financially attractive and operationally simple AV contracts will be critical for scaling adoption while preserving agricultural production. |
| Keywords: | Environmental Economics and Policy |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:ags:aaea26:404508 |
| By: | Andreas Makoto Fukuda Andersen; Gwen-Jiro Clochard; Guillaume Hollard; Andreas Kotsadam |
| Abstract: | People pay to discriminate. We replicate the price-of-prejudice paradigm of Hedegaard and Tyran (2018) in Japan and test whether brief cooperative contact with an outgroup member reduces the willingness to pay (WTP) for same-ethnicity coworkers. In a pre-registered laboratory experiment at the University of Osaka (N = 164 Japanese subjects), participants first performed a real-effort task, were then randomly paired with either a Japanese or an international student for a joint production round, and finally made incentivized binary choices between future partners whose name and productivity were revealed. Discrimination is clear in the data: one half of subjects choose the Japanese partner strictly more than 50% of the time in productivity-symmetric choices, and the average subject forgoes a productivity advantage of 2.4 envelopes, about 6% of pair earnings, to work with a Japanese rather than a foreign partner. Discrimination is also strongly price-sensitive, as in Hedegaard and Tyran (2018). The contact treatment reduces estimated WTP from 2.39 to 1.82, a 24% reduction, but the difference is statistically insignificant: the experiment validates the measurement paradigm while remaining underpowered for plausible effects of a single, brief contact intervention. In exploratory analyses we find that both men and women discriminate against women, at a price exceeding that of anti-foreign discrimination. |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:dpr:wpaper:1317 |
| By: | Wei Ma |
| Abstract: | We present a revealed preference characterization of the discounted expected utility model with a concave utility function. The characterization offers a nonparametric test of the model. We apply the test to an experimental data set in the literature and find that the model is almost always rejected even when all payments involved are subject to risk. |
| Date: | 2026–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2606.29779 |
| By: | Adabrah-Danquah, Vera; Hobbs, Lonnie; Britton, Logan |
| Abstract: | Foot-and-mouth disease (FMD) outbreaks can generate severe economic losses through supply chain disruptions, trade restrictions, carcass disposal costs, and consumer demand shocks, even though FMD poses no direct risk to human health. One proposed strategy for reducing these losses is to redirect recovered livestock protein into alternative downstream outlets, including pet food. This study evaluates U.S. consumer acceptance of pet food products containing livestock protein associated with a hypothetical FMD outbreak and compares pet food with other potential value-recovery outlets. Using an online stated-preference survey, we estimate willingness to pay for pet food attributes related to protein formulation, infection-zone status, country origin, vaccination disclosure, and price. We also compare these findings with consumer valuation of beef products and a two-stage ranking exercise measuring preferences for alternative downstream uses of meat from animals recovered from FMD. Results indicate that acceptance of recovered livestock protein is highly conditional. Pet owners required a substantial discount for pet food products sourced from a U.S. infected zone, while products sourced from U.S. noninfected zones received the highest valuation. Vaccination disclosure generated a positive willingness-to-pay premium, suggesting that consumers value information signaling disease management and oversight. Pet owners also preferred single-protein beef formulations over blended alternatives. The ranking exercise showed that pet food was viewed as a moderately acceptable downstream use, though indirect uses such as industrial applications and feed for nonfood animals received stronger support. These findings suggest that pet food may serve as a supplementary value-recovery outlet during FMD events, but its effectiveness would depend on transparent labeling, credible safety assurances, price discounts, consumer trust, and the ability of supply chains to manage segregation, verification, and coordination costs. |
| Keywords: | Food Consumption/Nutrition/Food Safety |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:ags:aaea26:404582 |
| By: | \"Ozg\"ur Akg\"un; Georgios Gerasimou |
| Abstract: | We provide an analytical framework that allows for distilling the full explanatory and welfare-relevant content of influential yet computationally hard models of bounded-rational general choice. We do so by introducing constraint programming methods and tools from the optimization literature. We focus on the prominent "shortlisting" and "limited-attention" models. Applying our framework on imperfectly rational human choice data, we find that these models jointly account for nearly all behaviors, with limited-attention ones explaining better while being more permissive. Selection criteria that we introduce narrow down the models' welfare-relevant predictions, considerably alleviating their indeterminacy and contributing toward their practical applicability. |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2607.03962 |
| By: | Avishek Bhandari |
| Abstract: | Standard economics assumes the consumer as a flawless calculator who always buys the best basket it can afford. This paper models the shopper instead as a limited information channel: it compresses its world to the detail its attention affords, so its choice is a probability distribution, not a single basket. The textbook consumer returns exactly as the unlimited-attention limit, while at the zero-attention end the shopper falls back on pure habit. The central result is about how this shopper's demand responds to price changes. That pattern of responses is just a rescaling of how the shopper's own choices vary and move together, so it comes out symmetric. And provided the budget really binds, because the shopper wants more than it can afford, raising a good's own price lowers demand for it once buying power is held fixed. So the downward pull comes from the budget and from compression, not from rationality. The framework also covers an artificial agent running a limited-capacity policy. A worked two-good quadratic consumer carries every quantity in closed form. |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2607.12412 |
| By: | Bervoets, Sebastian; Faure, Mathieu; Renou, Ludovic |
| Abstract: | Deviations from Bayesian updating are traditionally categorized as biases, errors, or fallacies, thus implying their inherent ``sub-optimality.'' We offer a more nuanced view. We demonstrate that, in learning problems with misspecified models, non-Bayesian updating can outperform Bayesian updating. |
| Keywords: | Learning |
| JEL: | C72 D83 |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:cpr:ceprdp:20114 |