Research
Do cold-climate heat pumps actually work?
The certification data says yes on heat output and no on running cost. That is why both sides of this argument have sounded right.
Written by HYRE HVAC Research Desk Primary-source research, data analysis and fact checking
The finding
Yes on heat, with a catch on cost. In ENERGY STAR test data, cold-climate models keep a median 83.3% of their 47°F heating capacity at 5°F, against 55.8% for standard models.
But their median efficiency (COP) at 5°F is 1.97. Only 0.8% reach 2.5 and none reach 3.0. In the 30 states where gas already costs less to run, the heat pump loses by more on the coldest days.
Can this question be settled with data?
Yes. "Heat pumps do not work in the cold" is the most durable objection in home heating, and it is usually met with a testimonial, not a number.
Every ENERGY STAR certified heat pump must report its measured heating capacity at 47°F, 17°F and 5°F, and its coefficient of performance (COP) at 5°F, all to the same test. The whole certified market is public record.
We downloaded it: 264,171 certified ducted combinations and 17,598 mini-split combinations. A row in that file is not a product: one outdoor unit sold with forty air handlers appears forty times. So everything below is computed over 12,562 distinct outdoor units, of which 9,271 (73.8%) carry the cold-climate designation.
Two measures matter, and people mix them up. Capacity is whether the machine makes enough heat to keep the house warm. Coefficient of performance is how much electricity it uses to do it.
A heat pump can be excellent at the first and poor at the second at the same temperature. That is exactly what the data shows.
Do cold-climate heat pumps keep their heat output at 5°F?
Mostly, yes. Cold-climate certified units keep a median 83.3% of their 47°F heating capacity when it falls to 5°F outside. Even the weakest tenth of the cold-climate market (the 10th percentile) keeps 70.1%.
Standard certified models are a different machine: a median of 55.8% retention, with the weakest tenth at 45.0%. That gap is the whole content of the cold-climate designation, and it is large.
At 17°F, closer to the design temperature much of the country actually sizes for, cold-climate units retain a median 80.4%.
HyreHVAC analysis: the old objection dates from equipment two product generations back. Against current cold-climate models it is wrong. Some units exceed 100% at 5°F, and that is not a data error. Variable-capacity compressors can ramp above their rated output as it gets colder, which is what they are built to do.
How efficient are cold-climate heat pumps at 5°F?
Much less efficient than at mild temperatures. This is the half of the subject that marketing does not lead with.
Across cold-climate certified units the median COP at 5°F is 1.97. The 10th to 90th percentile range is 1.80 to 2.20, a very tight band. Only 0.8% of cold-climate models achieve a COP of 2.5 or better at 5°F, and effectively none reach 3.0.
Set that against the breakeven thresholds we computed from EIA fuel prices. Michigan needs a seasonal COP of 5.18 for a heat pump to undercut a gas furnace.
A cold-climate heat pump in Michigan at 5°F is running at roughly 1.97. It is not close, and the shortfall is worst on exactly the days when the heating load is largest.
HyreHVAC analysis: both camps in the public argument have been half right. Advocates are correct that modern cold-climate heat pumps heat a house well in severe cold; the capacity is there.
Skeptics are correct that in a cheap-gas state they cost more to run when it is truly cold. These claims do not conflict. They answer two different questions asked in the same sentence.
How do cold-climate and standard models compare?
| Measure | Cold-climate certified | Standard certified | All units |
|---|---|---|---|
| Distinct outdoor units | 9,271 | 3,291 | 12,562 |
| Capacity retained at 17°F — median | 80.4% | 65.1% | 76.6% |
| Capacity retained at 5°F — median | 83.3% | 55.8% | 75.9% |
| Capacity retained at 5°F — 10th percentile | 70.1% | 45.0% | 51.7% |
| COP at 5°F — median | 1.97 | 2.00 | 2.00 |
| COP at 5°F — 10th to 90th percentile | 1.80 – 2.20 | 1.77 – 2.26 | 1.80 – 2.23 |
| HSPF2 — median | 9.80 | 8.20 | 9.50 |
| SEER2 — median | 20.00 | 17.10 | 19.50 |
ENERGY STAR certified heat pump performance, September 4, 2026. Computed over distinct outdoor units rather than certified combinations.
Capacity retention above 100% is physically real for variable-capacity equipment, which can ramp its compressor above nominal output as the outdoor temperature falls.
What should you check before you buy one?
It is not marketing. The median cold-climate unit holds 83.3% of capacity at 5°F against 55.8% for a standard unit. In any climate that sees single digits, this is the difference between a heat pump that heats the house and one that hands the job to its backup strips.
Equipment is normally quoted at its 47°F rating. That figure tells you almost nothing about the coldest fortnight of the year. The certified 17°F and 5°F capacities are public for every model, and an installer should size against them. Our sizing estimator shows the load side of that math.
When a heat pump cannot meet the load it falls back on electric resistance heat, which runs at a COP of 1. Every hour on backup is heat at the full electricity price. In a cheap-gas state that is the single most expensive way to heat a house.
The same equipment that saves money in one state raises the bill in another, purely on fuel prices. Our Heat Pump Readiness Index and price study compute the threshold for all 51 jurisdictions.
Most of a heating season is not spent at 5°F. Seasonal performance is materially better than the 5°F figure in almost every climate, which is why HSPF2 exists. Comparing a 5°F COP against an annual breakeven overstates the problem, just as comparing a 47°F COP against it understates the problem.
How did we measure this, and what are the limits?
ENERGY STAR certified product data for ducted and mini-split heat pumps, downloaded from EPA’s Socrata endpoint on September 4, 2026. 264,171 ducted and 17,598 mini-split certified combinations.
The files list one row per outdoor/indoor combination, so an outdoor unit sold against many air handlers appears many times. Counting rows would weight multi-combination products absurdly. Every figure here is computed over distinct outdoor units, keyed on brand plus outdoor model number: 12,562 of them.
This is the certified market, not the installed market. It is a census of what can be bought, weighted equally across models regardless of how many units each sells. A sales-weighted picture would look different and no public dataset supports one.
Uncertified equipment exists and is generally less efficient. This analysis therefore describes the better end of what is available, and the real installed fleet performs worse than these figures.
These are test-procedure measurements at steady state. A real installation loses performance to defrost cycles, duct losses, oversizing, undersized returns and low refrigerant charge. Field performance is below rated performance, sometimes substantially.
About 3% of units report capacity retention above 110% at 5°F. Most of that is legitimate variable-capacity behavior, but a small number are clearly data errors (one record reports 383%).
Excluding everything above 110% moves the cold-climate median from 83.3% to 82.2%, which is to say it changes nothing. Both figures are stated here rather than picking the flattering one.
Questions
Do heat pumps work in cold weather?
How efficient is a heat pump at 5°F?
What does the cold-climate designation actually mean?
Will a heat pump save me money in a cold state?
What happens when it gets colder than 5°F?
Is a mini-split better in the cold than a ducted system?
Written and audited by
HYRE HVAC Research Desk
Primary-source research, data analysis and fact checking
We are a research desk, not a sales floor. We read the federal microdata file, the statute or the manufacturer data sheet ourselves, and we publish the figure with the document it came from and the date we retrieved it.
Where a number cannot be traced to a primary source, we publish the shorter page and say what we could not verify.
The counts below are generated from the published pages themselves, last counted September 28, 2026, and they are what we have actually published rather than what we intend to.
- 13
- studies published
- 12
- federal sources read and cited
- 8
- studies published with their full dataset as CSV
- 51
- jurisdictions reproduced against EIA’s own tables
How this desk works
- Primary sources only. Federal data comes from the agency that collects it, in the file that agency publishes. We do not cite an article that cites a source; we download the source and compute the figure ourselves.
- We validate against the agency before we publish. First, we use each federal microdata file to reproduce the agency’s own published tables. Our cooling research reproduces EIA’s state estimates and standard errors for all 51 jurisdictions. That check caught a variance formula that was off by a factor of four.
- Every estimate carries its uncertainty. These are survey figures, not counts. Standard errors are computed from the replicate weights the federal file supplies and printed beside the estimate. An estimate too imprecise to publish is reported as such rather than printed.
- Nothing is typed by hand. Prose, tables and charts all read from one dataset built by script, so a number in a sentence and the same number in the table below it cannot disagree.
- We publish the data, not just the conclusion. 8 of our 13 studies offer the full computed table as a CSV download on the page, so you can check the analysis or disagree with it. A study without a row-level dataset gets no download link and claims none in its structured data.
- We correct in public. Where we have published a figure wrongly we fix the figure, rewrite any analysis that rested on it rather than patching the number underneath it, and leave a dated correction note on the page.
- We do not install or sell HVAC equipment, and we take no payment for placement, ranking or a favorable mention. Nobody buys a position on this site.
Data as of ENERGY STAR certified product data, September 4, 2026. Authorship on this site is organizational: the analysis belongs to the desk rather than to a named individual, and we do not publish credentials we do not hold. Our editorial policy sets out how we source, date and correct what we publish.
The data behind this page
Every figure on this page is computed from one file, and that file is published here so the analysis can be checked, disagreed with, or reused.
Sources & retrieval dates
Check the numbers for your own house
The certified data describes a market. Your design temperature, your rates and your ductwork decide the answer for one building.