Defaults, prices, and peer comparisons in environmental behavior.
Energy and environment has produced some of the cleanest experiments on default effects and social norms at scale. The Opower home energy report — a simple letter comparing your energy use to your neighbors' — produced durable energy reductions in 8 million households. Carbon pricing experiments in British Columbia and Stockholm provide the strongest causal evidence on price signals and emissions.
6
experiments
6
positive results
0
null or negative
4
replicated
Key Findings
01
Social comparison reports — showing households their energy use relative to neighbors — produce persistent 2% energy reductions at very low cost.
Opower's randomized experiments across 85+ utilities found that home energy reports showing households their consumption relative to efficient neighbors produced average reductions of 2.0–2.5% in energy use. Effects persisted for years after initial contact and survived changes in season, home ownership, and utility pricing. A 2014 analysis of 8 million households found the intervention was more cost-effective than most utility efficiency programs. The mechanism is normative comparison — people adjust consumption toward what they perceive as the norm.
02
Switching electricity defaults to green energy dramatically increases adoption without reducing customer satisfaction.
Germany's green electricity default experiment found that switching the default from conventional to renewable electricity for new residential customers increased green energy adoption from approximately 7% (opt-in) to 69% (opt-out). Customer satisfaction and retention were unaffected. The default effect — the tendency to stay with whatever option is presented as the status quo — operates powerfully even for decisions with real cost implications.
03
Carbon taxes produce measurable emissions reductions without the economic harm opponents predict — but political durability requires revenue recycling.
British Columbia's carbon tax — introduced in 2008 at C$10/tonne and rising to C$50/tonne — produced a 15% reduction in fuel consumption relative to the rest of Canada, with no measurable GDP effect. The tax was revenue-neutral, with proceeds returned as tax cuts and rebates. Seattle's minimum wage study and London's congestion charge provide parallel evidence that price signals produce behavioral change without the employment and congestion effects critics predicted.
Important Null & Negative Results
Programs that failed to produce expected outcomes under rigorous evaluation.
Solar Panel Adoption Peer Effects
Connecticut, United States · 2014
Peer effects in solar adoption — the phenomenon where one household adopting solar increases the probability that nearby households adopt — are real but modest. Simply showing people that their neighbors have solar does not reliably increase adoption; the barrier for most households is financial, not informational.
All Experiments in the Registry
Opower Home Energy Reports
positiveUnited States (multi-site) · 2012
Green Electricity Opt-Out Default
positiveGermany · 2016
Social Comparison Home Water Reports
positiveUnited States (multi-site) · 2014
Green Investment Default in Retirement Plans
positiveUnited Kingdom · 2020
British Columbia Carbon Tax
positiveBritish Columbia, Canada · 2008
Solar Panel Adoption Peer Effects
positiveConnecticut, United States · 2014
What the Evidence Cannot Yet Tell Us
Do energy conservation effects from social comparison reports persist indefinitely, or do they decay as the novelty of comparison wears off?
What carbon price level is required to shift major transportation and industrial behavior, not just household energy use?
How do green defaults interact with energy prices — are effects larger when energy is expensive (financial stakes) or when it is cheap (lower barrier to adoption)?
Can the Opower model be extended to water, meat consumption, or other high-impact household behaviors, or is energy consumption uniquely salient?