Conference Agenda

This paper utilises a Latent Class model estimated using the Expectation-Maximisation (EM) algorithm to investigate the impacts of financial literacy and energy literacy on households’ investment decisions for a low-carbon heating system across income classes. This is done utilising a Discrete Choice Experiment (DCE) for low-carbon heating alternatives; geothermal district heating, hydrogen boilers, solar electric boilers and air source heat pumps. Implicit Discount Rates (IDRs) were extracted from the Latent Class model parameter estimates to understand the investment analysis conducted by the respondents. The low-income class in the Latent Class model displayed consistently higher implicit discount rates in comparison to the middle to high-income class, illustrating that future costs are more heavily discounted within the implicit net present value determination for an investment in a heating system. Heterogeneity in price sensitivity and implicit discount rates across income groups highlights the need for tailored policy interventions in order to effectively incentivise household investment in low-carbon heating systems, and to minimise the potential for creating regressivity in policies aimed at households.

Carbon labels are increasingly used to promote sustainable food consumption, yet evidence on which label formats best encourage low-carbon choices remains limited. This study applies an online survey and a laboratory experiment across three studies to examine how label formats and visual cues shape consumer preferences for low-carbon milk. Study 1 employs a between-subjects choice experiment to compare the effects of carbon footprint, carbon cost, and carbon reduction labels. Study 2 uses a within-subjects experiment to assess whether adding traffic light color coding enhances label impact. Study 3 integrates eye-tracking technology to explore the relationship between consumers’ visual attention and preferences from a physiological perspective. Results demonstrate that all three formats of carbon labels significantly increase consumers’ willingness to pay for low-carbon milk, with carbon footprint labels exhibiting the strongest effect. Traffic light color coding further improves guidance towards low-carbon options. Eye-tracking reveals a positive association between consumers’ visual attention to label attributes and product preference; carbon footprint labels attract the most gaze, and traffic light coding shifts attention toward low carbon information and increases preference. These findings suggest combining carbon footprint labels with traffic light cues to enhance visual engagement, improve label comprehension, and promote sustainable choices.

Time-of-use (ToU) tariffs are an economically appealing mechanism for managing peak electricity demand, but low consumer adoption and suboptimal usage tend to limit their real-world effectiveness. This paper examines four potential economic and behavioural barriers that may underpin the issue: (1) Potential savings from ToU tariffs may be insufficient given the electricity-use changes required to achieve them. (2) Consumers may have imperfect knowledge of how time-of-use tariffs work. (3) Consumers may lack understanding about which electricity-use changes would be financially beneficial under a ToU plan. (4) Consumers may hold incorrect perceptions about how ToU pricing impacts their total electricity costs. We develop an online survey experiment to assess consumer comprehension of ToU tariffs, preferences for adopting the plans, and the role relevant information disclosures can play in encouraging their uptake. Using a multiple price list, we derive novel estimates of consumers’ willingness to accept (WTA) ToU tariffs specifically designed to provide bill discounts conditional on consumers making changes to the timing of their electricity usage. Randomised information treatments are used to experimentally evaluate the impacts of tariff knowledge and price salience on WTA. We find that the compensation consumers require to change behaviour under ToU tariffs can be substantial but varies significantly with the information presented. Our results suggest that the salience of the marginal per-unit electricity prices that ToU tariffs entail are a particular deterrent to adoption. In contrast, our knowledge interventions do not significantly affect willingness-to-accept. Furthermore, we find that baseline tariff understanding is high in our sample and is not improved by our interventions. The findings highlight the importance of understanding the nuances of the trade-offs consumers face when selecting an electricity tariff.

Market digitalization has changed the role of consumers who can now be both sellers and buyers on platform-based secondary markets. This evolution reinforces the connections between primary and secondary markets as well as extends the choice context of consumers. The online platforms have expanded consumer options by increasing the supply of secondhand goods through their geographical reach, facilitating peer-to-peer exchange and the wide variety of goods on offer. Thus, the rise of secondhand markets highlights the importance of acknowledging the heterogeneity of the secondhand offers and the associated heterogeneity in consumer preferences. This paper provides empirical evidence of the three aspects of heterogeneity in the market: seller-related, buyer-related and product-related. We use a discrete choice experiment to study the choice between new and secondhand goods from the incumbent or an online independent platform in Spain. Expected quality is an important factor in consumer choices in this context. We observe a brand effect where consumers prefer to buy new and secondhand goods from the incumbent, as this may seem to be a more reliable choice. In addition, we find that consumers value uncertainty-reducing features such as an extended warranty and previous knowledge of the seller’s reputation.

Despite widespread policy support for climate-smart agriculture, adoption of conservation practices remains constrained by an intention-action gap. We test whether spatially targeted, AI-generated advisory can close this gap by reinforcing farmers’ existing intentions during critical decision windows. We implement a framed field experiment with 1,488 corn, soybean, and wheat farmers in the U.S. Midwest, randomly assigning farmers to receive either their usual information sources or four AI-generated advisory emails delivered during the cover crop decision window. The AI advice was tailored using county-level climate, soil, and agronomic data and focused on practical management guidance. Using a Cragg-type double-hurdle model, we estimate intent-to-treat effects on both adoption and acreage intensity. Assignment to the AI-information treatment increased the probability of cover crop adoption by 4-5 percentage points, with no statistically significant effect on the share of acreage planted among adopters. These results indicate that AI-generated advice primarily reinforces existing intentions by reducing implementation frictions rather than persuading new adopters or expanding adoption intensity. The findings suggest that timely, targeted AI advisory can serve as a scalable, low-cost complement to extension services in improving behavioural follow-through in conservation programs.

Men and women often value social recognition over money when donating to prosocial causes, sometimes sacrificing material gains to boost their reputation. This study uses a lab-in-the-field experiment in rural India to examine gender differences in valuing public recognition for generosity and productivity. Participants earned money in a task, donated privately to an environmental NGO, and then stated how much they would pay for public recognition. Results show: women donate more, men pay more for public recognition of generosity and productivity, and women pay more for anonymity in both domains. Further, as the donation increases, women pay more for anonymity and men pay more for public recognition of generosity. These findings highlight gendered patterns in how social image shapes prosocial behavior and underscore the importance of considering such differences in norm-based policy design.

I study the retailer-optimal electricity pricing problem. I formulate the problem as a screening problem, in which households are price elastic when exposed to wholesale prices and have private information regarding their heterogeneous risk aversion. The model shows how risk aversion, in particular first order risk aversion, is necessary to rationalize the large share of fixed price contracts observed in reality. This follows from the observation that price elasticity acts as partial self-insurance for the households, making them risk-loving without further assumptions on risk aversion. To showcase the applicability of first order risk averse preferences, I solve the retailer-optimal price schedule with Yaari dual utility preferences (Yaari 1987) for households and binary wholesale prices, and show that the most risk averse agents are pooled together at the fixed price contract.