How Effective are Energy Labels? Evidence from the 2021 EU Reform

Energy labels are often used to promote the adoption of more energy-efficient domestic appliances. This paper studies the 2021 EU energy labeling reform using product-level sales data on refrigerators in four European countries (Belgium, France, Germany, and Poland). Descriptive evidence suggests that the average energy consumption of refrigerators sold, adjusted for product characteristics, fell by between 3.2% in France and 3.9% in Belgium between March 2021 and December 2022. We then estimate a product-level demand model that incorporates energy labels in utility either directly or through their associated 15-year electricity costs. Consumers undervalue future electricity costs under both labeling regimes. Estimated undervaluation declines under the new labels in Belgium, France, and Germany, but increases slightly in Poland. Counterfactual simulations suggest that the labeling reform generated only modest reductions in energy consumption. Larger reductions could be achieved through alternative policies: stricter minimum standards (banning the least efficient appliances) or fiscal incentives (a feebate scheme subsidizing efficient appliances while taxing inefficient ones). The minimum standard generates larger total welfare gains than the feebate schemes. These gains arise primarily from correcting internalities associated with consumers’ undervaluation of electricity costs rather than externalities associated with emissions.