Is the Heat Pump Market Shift Structural or Just a Subsidy Story?

The federal heat pump tax credit expired at the end of 2025. Heat pump shipments then outpaced natural gas furnaces by 32% in Q1 2026. That sequencing matters for every investor who assumed the subsidy was the whole story.

Across virtually every month of reported 2026 data from the Air-Conditioning, Heating, and Refrigeration Institute (AHRI), U.S. air-source heat pump shipments have run ahead of gas warm-air furnaces. Year-to-date through June 2026, heat pump volumes climbed 3.8% year-over-year while gas furnace volumes fell 6.2%. The heat pump market shift underway is telling a structurally different story than it was five years ago, and the implications reach well beyond equipment manufacturers into venture capital, utility strategy, and building-sector equities.

The AHRI monthly shipment data underlying this analysis is drawn from the institute’s continuously updated statistics portal, which publishes unit-level figures for heat pumps, gas furnaces, and other HVAC categories across all U.S. regions, making it the primary verification source for any investor tracking this transition.

This analysis gives you a framework for separating durable structural momentum from the residual policy tailwind that is now fading. The question that matters for capital allocation is simple: is the shipment lead an entry signal, or a false summit before the subsidy hangover arrives?

What the shipment data actually shows, and what it does not

Start with the one month that cuts against the thesis. In November 2025, air-source heat pumps shipped 198,247 units, down a steep 37.2% year-over-year, while gas furnaces shipped 238,810 units, down a milder 8.4%. That is the only reported month in the current window where gas furnaces led. Any honest reading of the data has to hold it in view.

Then the pattern begins to turn. In January 2026, heat pumps shipped 229,957 units (down 16.5% YoY) against gas furnaces at 217,762 units (down 19.8% YoY). Heat pumps edged ahead, but only because both categories were falling and gas fell faster. This is not yet a growth story; it is a relative one.

By spring the picture sharpens. April 2026 heat pump shipments reached 380,888 units, up 1.8% YoY, while gas furnaces slid to 251,401 units, down 4.9%. May 2026 extended the lead: heat pumps at 396,553 units (up 2.3%), gas furnaces at 265,734 units (down 6.3%). Heat pumps were the only cooling category in year-to-date growth that month.

June 2026 delivered the strongest absolute figures: 418,538 heat pump units (up 15.1% YoY) against 320,281 gas furnace units (up 15.2%). Both rose sharply, a reminder that seasonal demand lifts the entire market at once.

The monthly figures swing enough that no single one settles the argument. The year-to-date aggregate does.

The signal that individual months cannot explain away: Through June 2026, heat pumps shipped 2,156,549 units (up 3.8% YoY) while gas furnaces shipped 1,552,934 units (down 6.2% YoY). The two categories are moving in opposite directions.

That divergence is the number to price positioning against. It tells you the Q1 32% lead was not a subsidy-expiry artefact or a one-quarter fluke; it was the culmination of a reallocation of volume that the full first half confirms. The direction is set. What remains contested is how durable it is, which is where the next section goes.

Why the market kept moving after the subsidies stopped

If demand held up after the credit lapsed, the operating economics have to be doing the heavy lifting. They are.

Heat pumps run at roughly 3-4 times the efficiency of resistance heating because they move existing heat rather than generate it from scratch. The gap is even starker in water heating. According to Asymmetric Capital Partners and ValueAddVC, heat-pump water heaters operate at approximately 400% efficiency against 92-95% for conventional gas and electric resistance units. In regions with moderate electricity prices, that bill-savings maths pencils out without any tax credit attached.

Lucas Davis, an energy economist at UC Berkeley, frames the longer arc directly. Research cited via Grist shows U.S. heat pump sales roughly doubled over the 15 years through 2026, and Davis reads that trajectory as evidence the market reached maturity before the Inflation Reduction Act. On his view, the credit accelerated an existing trend rather than manufacturing one.

Davis’s research on heat pump adoption determinants finds that geography, climate, and electricity prices are stronger predictors of adoption rates than tax credit availability, lending academic weight to the article’s argument that the subsidy accelerated an existing trend rather than creating one.

Beneath the economics sit three reinforcing mechanisms that make adoption path-dependent:

• Operating efficiency economics that deliver bill savings independent of federal incentives in moderate-price markets.
• Grid decarbonisation and utility strategy, as utilities increasingly value controllable electric heating loads for managing peak demand and integrating renewables.
• Building codes and local gas bans that phase out new gas hookups, mathematically shifting the default to electric systems in new construction.
• Installer network effects, where trained contractors default to recommending heat pumps once they are comfortable with the technology.

The code mechanism is the structural floor. State and municipal restrictions on new gas connections, combined with the fact that Northeastern programmes in Maine and Massachusetts paired rebates with contractor training and public outreach, create reinforcing cycles. Once developers standardise on electric systems, that behaviour persists regardless of what Congress does with incentives next.

The code mechanism is the structural floor, but its reach has limits: heat decarbonisation alternatives including hydrogen blending, district heating networks, and high-efficiency gas systems remain competitive in cold-climate and dense-urban contexts where all-electric retrofits face the steepest installation economics.

For you as an investor, the read is specific. The secular demand curve is real, but it is real in segments where the electricity price environment and building stock make the numbers work unaided. That is a narrower claim than a blanket sector bull case, and the distinction is where capital gets allocated well or poorly.

The installer network effect as a durable moat

Once HVAC contractors are trained and confident with heat-pump installations, they tend to propose them by default, particularly for customers replacing an entire system rather than repairing one. That converts a technology preference into a channel habit, and channel habits are not tied to any single rebate programme.

The countervailing risk is equally documented. Training gaps and labour shortages remain a genuine drag, especially in smaller markets where the pool of qualified technicians is thin. The moat is real where the installer base has been built; where it has not, adoption still stalls.

Reservoir and the venture thesis taking shape around grid-integrated heat pumps

If you want to see the investment logic made physical, look at what Reservoir builds. The Boston-based startup closed an $8 million seed round on 12 August 2026, and its product, a smart, grid-connected heat-pump water heater, embodies the thesis that heat pumps are becoming distributed grid assets rather than passive appliances.

The round was led by Asymmetric Capital Partners, with Founder Collective and the climate-focused MCJ Collective participating. That syndicate is betting on three overlapping value pools: household bill savings from the unit’s roughly 400% efficiency (versus 92-95% for conventional heaters), utility grid-services revenue from controllable load, and climate-aligned demand response.

The product features are what make the grid framing credible rather than aspirational:

• Learning and optimisation, using the first month of household usage data to tune operating schedules for cost and efficiency.
• Ultrasonic leak detection built into the flow sensor.
• A recirculation valve for near-instant hot water and a mixing valve to guard against freeze damage.
• Full mobile control of all device functions.
• A 10-year warranty supporting long-term deployment.

The investor framing: Asymmetric Capital Partners positions residential water heaters as AI-orchestrated, grid-responsive thermal batteries, distributed storage that can be dispatched at scale as installations grow.

That framing is what gives the business model scalability beyond a local install count. The economics of a single water heater are modest; the economics of thousands of coordinated thermal batteries selling flexibility to a grid operator are a different proposition entirely.

Reservoir’s grid-services framing connects to a broader investment thesis around grid infrastructure transformation, where AI-orchestrated demand response and distributed controllable loads are becoming a structural input to utility planning rather than a pilot-programme novelty.

Be honest about scale. Reservoir has roughly 100 units installed around Boston and targets approximately 1,000 units by the end of 2027. That gap is where the execution risk concentrates: hardware scaling, installer alignment, and utility partnership timelines. The product concept and market thesis are not the question marks; the manufacturing, field operations, and contract cadence are.

For you, Reservoir functions as a template for how climate-tech venture now structures building-electrification bets: hardware plus controls software plus grid-services revenue. That stack demands different due diligence than a pure-software investment, and a longer, more capital-intensive risk-return horizon.

Where the real risks sit for investors betting on this transition

Momentum in the shipment data does not mean smooth compounding, and three risk clusters could stall it.

1. Supply-chain and shipment volatility. The 37.2% year-over-year heat pump decline in November 2025 is a live data point, sitting in the same dataset that supports the bull case. It shows a market still adjusting through inventory and production cycles rather than compounding cleanly.
2. Policy and incentive uncertainty. The prior federal credit of up to $2,000 per qualifying system expired at the end of 2025. The current legislative status of IRA Section 25C could not be verified from available sources, and local gas-ban reversals add further political exposure.
3. Segment-level dependency. Cold-climate retrofits, low-income households, and markets lacking strong utility programmes remain materially policy-sensitive, a different risk category from the structurally robust segments.

Supply-chain exposure extends beyond installer labour into the coil and refrigerant material stack; HVAC manufacturing economics are being reshaped in 2026 by the aluminium-copper substitution trend, a cost dynamic that affects heat pump bill-of-materials and margin profiles across the equipment OEM layer.

The analyst community reflects this split rather than resolving it. BloombergNEF and Rocky Mountain Institute point to falling costs, improving performance, and tightening building targets as reasons adoption can self-sustain in key segments. MIT Technology Review and Grist counter that adoption spikes have historically tracked strong incentives and utility programmes, implying continued policy dependence in harder-to-reach markets. Labour and installer shortages, cited across industry surveys, act as a drag on both camps’ timelines.

The 37.2% single-month drop is your calibration check. Structural momentum and volatile compounding can coexist, and position sizing should reflect the sector’s demonstrated swings rather than the tidy year-to-date trend line.

Which segments are structurally positioned, and which are not

The durable exposures cluster in new construction, higher-income households, and moderate-electricity-price regions, where the operating economics and code mechanisms carry adoption without a credit. These are the segments to treat as genuine structural growth.

The policy-sensitive exposures sit in cold-climate retrofits, low-income markets, and jurisdictions without deep utility programmes. Here, a rebate cut or interconnection hurdle can stall demand, and the exposure should be sized accordingly.

The bridge between the two is emerging in the form of utility grid-services revenue, precisely the model Reservoir is building toward. Where a controllable load can earn demand-response payments, the economics improve enough to pull a policy-dependent segment toward structural viability. That mechanism is the one to watch for signs the harder segments are crossing over.

Positioning for a transition that is real but uneven

The market-level question is settled. Heat pumps outshipped gas furnaces by 2,156,549 units to 1,552,934 units through June 2026, and the 15-year doubling in sales that Lucas Davis of UC Berkeley points to gives that lead a structural baseline. The transition away from gas furnaces is happening.

The heat pump transition is one current within a broader reorientation of energy security capital allocation, where record diversification spend in 2026 reflects institutional investors moving away from single-fuel-source exposure across residential, commercial, and grid-scale energy infrastructure simultaneously.

What is not settled is where the commercial opportunity concentrates. It is not distributed evenly across the sector; it sits in identifiable segments, and the companies best positioned are those, like Reservoir, that stack household economics, utility grid services, and software controls into a single platform designed to outlast any one incentive cycle.

Three variables will confirm or complicate the thesis over the next six to twelve months:

• Whether the AHRI year-to-date gap between heat pumps and gas furnaces widens or narrows through the second half of 2026.
• The pace of utility grid-services contract announcements from building-electrification startups.
• The legislative status of IRA Section 25C, which readers should independently confirm before making any tax-sensitive decisions.

The question is no longer whether the heat pump transition is real. It is which entry points, companies, and segments represent durable exposure versus policy-dependent bets that reprice whenever Washington changes direction.

This article is for informational purposes only and should not be considered financial advice. Investors should conduct their own research and consult with financial professionals before making investment decisions. Past performance does not guarantee future results, and forward-looking statements are speculative and subject to market and policy developments.