Measuring what works: quantifying greenhouse gas emission reductions of behavioural interventions to reduce driving, meat consumption, and household energy use

Abstract

Interventions that promote pro-environmental behaviours are increasingly necessary in reducing use of high-emissions goods and services to meet international climate change targets. Here we assess the greenhouse gas emissions reductions associated with behavioural interventions in three high-emitting domains (personal vehicle use, meat consumption, and household energy use) based on an analysis of the peer-reviewed literature. We examined 40 studies from 1991-2018 involving 886 576 subjects. We found that some of the most robust interventions shown to reduce emissions were financial incentives for personal vehicles, defaults for reduced meat consumption, and feedback for home energy use. We estimate mean annual reductions of 571 kgCO 2 e per vehicle driver for reduced vehicle use, 51 kgCO 2 e per individual for reduced meat consumption, and 149 kgCO 2 e per household for reduced energy use. Despite substantial attention to behavioural interventions in the literature, we find that few studies are suitable for quantifying emissions reductions (N=6 for diet, N=5 for personal vehicles) and few (N=3) are conducted outside OECD countries. Due to this imbalance in the literature, we focus our findings on western economies. We find substantial variation in the emission reductions achieved with different interventions within each domain; interventions in diet ranged from reductions of 231 kgCO 2 e to increases of 116 kgCO 2 e per person per year (both statistically significant). Further, emissions reductions are sensitive to external factors, such as the emissions intensity of the electrical grid, which may change over time. Key gaps in the literature include a lack of studies conducted using randomized controlled trials or follow-ups, and in high-impact areas for emissions reductions including air travel. We highlight promising areas of intervention, such as habit changes to promote mode shifts in personal transportation, which would benefit from an analysis of greenhouse gas emissions reductions in future research.

How psychology can help limit climate change.

“…When psychologists interested in pro-environmental behaviors neglect looking in detail at impacts, they risk reinforcing existing popular misperceptions of TP. For example, a recent review of behavioral interventions that promote pro-environmental behavior was unable to identify any interventions aiming to reduce air travel despite it being one of the highest GHG-emitting individual behaviors and one that is rapidly increasing (Wynes et al, 2018).…”

Developing a Scalable Dynamic Norm Menu-Based Intervention to Reduce Meat Consumption

“…How would the cost and benefits of such an intervention compare to similar interventions targeting meat consumption in broader dietary contexts? Recent research has begun to compare different experimental interventions aimed at encouraging individuals to reduce their meat consumption [20]. One study, implemented in a student catering facility, found that providing information about the carbon footprint of various food choices resulted in a 4.8 percentage point reduction in meat dishes labeled with high emissions-but an 11.5 percentage point increase in meat dishes labeled as having low emissions-leading to an estimated 3.6 percentage point reduction in emissions [21].…”

Becoming FEW Conscious: A Conceptual Typology of Household Behavior Change Interventions Targeting the Food-Energy-Water (FEW) Nexus

“…Without systematic logging and tracking of study design and results, however, academics and practitioners will continue to face the same challenges of synthesizing the vast amount of existing information. Wynes et al (2018) provided an example of a better way to understand impacts across studies by quantifying their greenhouse gas emissions reductions [97]. Providing a weight in kilograms, of reduced carbon dioxide, goes a step further than effect sizes to measure intervention impacts.…”

“…Logging the results of behavioral intervention studies on a publicly available database would advance accumulation of knowledge in the field. Studies like Wynes et al (2018) would have a much larger pool of data to draw from when conducting quantitative analyses. Additionally, tracing unpublished interventions in the database may work to combat publication bias that favors positive results, much like the way clinical trials must be registered before they are completed [98].…”