Space heating and cooling account for 41% of total commercial energy use nationwide, and in California, most of it still runs through gas-fired rooftop units. New research presented at the 2026 ACEEE Summer Study on Energy Efficiency in Buildings, based on two CalNEXT pilot incentive programs, finds that higher efficiency heat pump rooftop units, known as HP RTUs, are increasingly available on the market. The bigger obstacle to adoption is that current incentive levels don’t cover enough of the added cost.
The CalNEXT pilots ran from October 2024 through October 2025 and tested two different incentive structures for HP RTUs. The first focused on simplifying program participation for distributors. The second tested a three-tier measure design that pays higher incentives for units with variable speed compressors, energy recovery ventilators, or advanced rooftop controls. Researchers paired those pilot results with market interviews, product catalog data, and national equipment listings to see where the technology stands and where program design is falling short.
One clear finding is that variable speed compressors remain rare despite their efficiency benefits, largely because they add an estimated $500 to $750 per ton to equipment cost. Two stage and multistage compressors, which cost less, have become far more common in the market instead. Supply fans tell a similar story. They account for roughly 30% of RTU’s total energy use, and variable speed fans can cut that significantly, but adoption still lags where compressors are single speed.
Electric resistance auxiliary heat, the backup coil that kicks in when a heat pump can’t keep up, came up as a bigger cost driver than expected. These coils have a maximum coefficient of performance of 1, compared with over 3 for the heat pump itself, and they can trigger expensive electrical panel upgrades during retrofits. An NREL analysis cited in the paper found California relies on that backup heat for only 6% of heating hours, less than any other state, suggesting many buildings could skip it entirely and avoid the added installation cost.
The pilots also exposed a participation problem with California’s existing fuel substitution measure, which pays for replacing gas RTUs with heat pumps. That measure made up just 7% of the state program’s total volume in 2025, its best year yet, while the standard heat pump replacement measure made up the rest. Distributors told researchers the fuel substitution measure requires them to confirm what type of equipment a building currently has, information they often can’t get with confidence, and that current incentive levels don’t cover enough of the incremental cost to make the sale easy.
By comparison, the CalNEXT pilots moved far more volume per year than the long-running statewide program, which researchers attribute to a simpler midstream design and incentives that better matched actual equipment costs.
The paper recommends that California expand its measure package with a higher efficiency tier and a dedicated variable speed measure, size incentives to cover at least 70% of incremental cost, and reduce the data burden on distributors, potentially through a baseline study that estimates existing equipment types by building type and climate zone rather than requiring each participant to report it. It also points to mild coastal climate zones and building types like manufacturing and higher education, which already show strong interest, as good near-term targets for expanding adoption ahead of an anticipated jump in higher tier equipment around 2027.