Conference Agenda

Changing diets toward plant-based foods (PBF) is widely viewed as an important complement to production-side climate mitigation policies. Yet despite growing awareness of the environmental and health benefits of plant-based foods, consumer adoption remains limited. This paper examines whether scalable consumer-facing interventions can bridge this gap. We conduct a pre-registered randomized natural field experiment using a smartphone household-accounting application in Japan. The study combines survey responses from 4,863 consumers with linked receipt-level purchase data from 3,777 individuals, generating approximately 150,000 individual-week observations and more than one million purchase records.　Participants were exposed to informational interventions emphasizing environmental impacts, health benefits, or personalized purchase-history feedback, with or without cash-back incentives for plant-based food purchases. Survey results indicate that several interventions substantially reduced uncertainty about plant-based foods and increased consumers’ stated willingness to substitute away from conventional meat products. However, these attitudinal changes translated only weakly into actual purchasing behavior. Difference-in-differences and event-study analyses show little evidence of sustained increases in either the probability of purchasing plant-based foods or expenditure on such products. Positive effects, when observed, were modest and concentrated among a small subset of treatments, particularly those combining financial incentives with personalized feedback. Most effects attenuated rapidly and disappeared after the interventions ended. These findings suggest that changing consumer beliefs may be easier than changing consumer purchases. Although several interventions altered perceptions of plant-based foods and increased stated substitution intentions, their effects on realized purchases were small and short-lived. The results therefore provide field-experimental evidence of a substantial intention–behavior gap in sustainable food consumption and raise questions about the effectiveness of low-cost consumer-facing policies as a tool for large-scale dietary change.

In this paper, we estimate the causal effect of improving consumers’ climate impact knowledge on food purchases and associated emissions. In collaboration with Norway’s largest grocery chain, we implement a randomized information intervention in a survey of almost 7,000 individuals, in which we update the treatment group’s knowledge about the climate impacts of beef and other substitutes (including plant-based alternatives).

We then link respondents to loyalty-card scanner data from the same retailer, covering nearly two years around the intervention. This design allows us to measure (i) baseline climate-impact knowledge, (ii) the causal effect of improved knowledge on stated purchase intentions in the survey, and (iii) the causal effect of improved knowledge on actual grocery purchases over several months after the intervention.

We document widespread underestimation of beef’s emissions: fewer than 1% of respondents correctly assess its climate impact relative to substitutes such as pork, poultry, or plant-based alternatives. The information treatment improves treated respondents’ knowledge and increases stated intentions to shift toward lower-emission options, particularly plant-based alternatives. In the transaction data, it leads to a statistically and economically significant reduction in beef consumption over the subsequent five months. However, contrary to stated intentions, we do not find an increase in plant-based purchases.

We evaluate the impact of curbside recycling collection service on the recycling and sorting of plastics, with a focus on plastic bottles. We conducted a non-randomized field experiment with a three-stage intervention in two middle-income neighborhoods in Osorno, a city in southern Chile. First, we introduced open-access street recycling bins in both neighborhoods to establish a baseline scenario. Next, we implemented a curbside recycling collection system in the treatment sector, which involves door-to-door collection. Finally, we distributed normative appeal brochures in the treatment area. Special attention is given to the effects of the interactions between both types of recycling services. Furthermore, we investigate whether appealing to personal norms reinforces the effect of the curbside recycling collection system. Our findings show that a curbside recycling collection service increases plastic recycling, and that both types of recycling options are complementary. Our results suggest the absence of reinforcement effect when normative appeals are provided, and that the curbside recycling collection service increases the amount of non-recyclable waste in recycling bins. This may most likely be due to the higher chance that people recycling through the curbside recycling collection service can be exposed as recycling improperly.

For at least the last decade, food waste has a been a growing concern for policymakers in OECD countrie. One policy emphasized for addressing food waste challenges and their associated environmental impacts is the expansion of municipal composting services. Important economic feedbacks may confound the predicted impacts of policy-induced composting service expansion on food waste generation. One key feedback is behaviorally analogous to the ‘rebound effect’ studied in the energy efficiency literature: To the extent that households face ethical or moral costs from throwing away food with the rest of their trash, composting offers a means to feel less bad about throwing away food. In this paper, we analyze data from a recent largescale household survey conducted by the OECD across nine countries (N = 8,569), to study the effects of composting services on food waste generation behavior. The survey focused on eliciting a variety of household behaviors and attitudes across several environmental domains, including waste generation, separation of recyclables, and composting. Using clustered matching and instrumental variable metthods, we estimate the effect of curbside composting on food waste behavior and then translate them into volumetric terms using an auxiliary regression of non-composting households’ mixed waste volume on their food waste behaviors. We find that curbside composting significantly increases the propensity with which households throw away food. From the auxiliary volumetric regression, we reckon this effect translates into an average treatment effect on treated households of a +1.3% increase in food waste volume in response to curbside composting service. Our results indicate that there may be a modest increase in food waste when it is easier for households to compost. Policymakers should be aware of this rebound effect on overall waste volumes.