A Multi-Level Framework for Persistent Savings in The Built Environment

Abstract

People often say that energy efficiency is the cheapest way to cut down on carbon emissions, but the gap between the savings that were expected and the savings that were made (the efficiency performance gap) is still very big and not well understood. This article synthesises insights from engineering, behavioural economics, and science and technology studies (STS) to propose a holistic framework for comprehending and managing sustained energy savings. Based on a systematic review of 210 peer-reviewed studies (2015–2025) and four longitudinal case studies (two commercial office buildings, one industrial facility, one multi-family housing cooperative), we find three overlapping mechanisms that undermine efficiency persistence: (1) technical drift (equipment degradation, improper commissioning, control logic override), (2) behavioural rebound (direct, indirect, and economy-wide rebound effects that offset 10–40% of predicted savings in buildings). When we compare them to actual metered data instead of engineering models, we find that even the best efficiency projects lose an average of 34% of the savings they expected to see within five years of being put into place. We get our information from energy metering (15-minute interval data over 2–8 years), semi-structured interviews with 62 stakeholders (facility managers, energy auditors, occupants, maintenance staff), and document analysis of commissioning reports and maintenance logs. We recommend a dynamic efficiency contract that links payment to savings over time, needs to be recommissioned on a regular basis, and has feedback loops for behaviour. Policy changes: efficiency resource standards need to move from modelled savings to measured savings, and green leases should make it mandatory for tenants and owners to share data.