Heat pump running cost calculator
Your state's real heating season prices, every assumption shown.
Your current system costs less to run
The heat pump would cost about 94% more to deliver the same heat.
| Cost to deliver one kWh of heat | $ |
|---|---|
| Natural gas furnace at 95% | 0.0560 |
| Heat pump, effective COP 2.64 | 0.1084 |
Effective COP includes 10% of heat coming from electric resistance backup at COP 1.0. Running cost only. Equipment, installation, rebates, and summer cooling are excluded.
What that costs per month
Cost per unit of heat is the honest way to compare two machines. It is not how anyone reads their own bill. Add your home size and this turns it into dollars.
How to get a real answer out of this
The defaults above are a state average applied to a typical machine. They will tell you roughly which side of the line you are on. Four changes turn that into an answer about your house, in the order they matter.
1. Change the backup heat figure first
It is the input almost nobody thinks about and it moves the answer more than anything else on the page. Electric resistance backup runs at a coefficient of performance of exactly 1.0, so a system taking a quarter of its heat from the strips gives up roughly a quarter of its efficiency advantage before you have compared a single price.
Ten percent is a reasonable guess for a well commissioned cold climate system. If you do not know yours, put in 0% and then 25% and look at the gap between the two answers. That gap is what commissioning is worth. What federal field measurements found real systems actually use is the honest starting point, and it is higher than most people expect.
2. Use your own two prices, not the state average
Both price fields are editable and both are worth editing. State averages hide large differences between utilities: one household in the dataset pays $0.193 per kWh while another ten miles away across a state line pays $0.097, and the difference is whether the utility is municipal or investor owned.
Take the electricity figure from your own bill by dividing the total by the kilowatt hours, not by reading the supply line. On a real Massachusetts bill the supply line was $354 and the bill was $760, and delivery is roughly half of what you pay across New England.
Several utilities also run discounted winter rates for heat pump customers that appear in no state average, including the one on this page. That is worth a phone call before you trust any of this.
3. Check the HSPF2 against the actual model
8.0 is close to the regulatory floor, 11.0 is a good cold climate unit. The figure is on the AHRI certificate for the specific indoor and outdoor pairing rather than on the brochure, and a rating for a different indoor unit does not describe what you are buying. What HSPF2 means, and where it stops describing your house covers why a seasonal average measured in one climate region is not your climate.
4. Read the output as cost per unit of heat
The comparison is dollars per kilowatt hour of heat delivered into the house, which is the only way to line up fuels sold in therms, gallons and kilowatt hours. Add your floor area in the panel below and it becomes a monthly figure, though that step brings in a rule of thumb about how fast your house loses heat, which can be wrong by half in either direction. A better method, if you have a year of bills, needs no assumption about the building at all.
What this is built on
Whether a heat pump costs less to run than what you have now depends on two things: the price of electricity where you live relative to your current fuel, and the efficiency of both machines. Everything on this page comes from those two, computed from published prices rather than from a national average or a manufacturer's claim.
The one equation behind this
A heat pump costs less to run than a combustion furnace when:
The right side is set entirely by your equipment. This is why the common rule of thumb, that heat pumps win below a price ratio of 3.5 to 1, is a description of one specific pairing rather than a general law. A ratio of 3.5 is what the equation returns for an HSPF2 11.0 unit against a 95% AFUE furnace with no backup heat. Change either machine and the threshold moves.
| Pairing | Breakeven ratio |
|---|---|
| HSPF2 11.0 against 95% AFUE | 3.39 |
| HSPF2 8.0 against 95% AFUE | 2.47 |
| HSPF2 11.0 against 80% AFUE | 4.03 |
| HSPF2 11.0 against 95% AFUE, 15% backup heat | 2.54 |
One realistic backup heat assumption moves the threshold by 25%. Most calculators leave it out entirely.
Four assumptions, all of them stated
- Natural gas carries 1,037 BTU per cubic foot. This is the EIA convention. Delivered gas ranges roughly 1,020 to 1,050 depending on utility and season.
- HSPF2 converts to seasonal COP by dividing by 3.41214. HSPF2 is measured in one AHRI climate region, so your real seasonal COP will differ with climate.
- Electric resistance backup runs at COP 1.0, weighted by share of heat delivered rather than by runtime.
- No winner is declared inside a 10% margin. Fuel prices move more than that between seasons.
Assumption three is the shakiest and it matters most. Backup heat depends on the balance point of your specific installation, which a state level model cannot know. It is an input here rather than a guess. Full detail is in the methodology.
What this page does not answer
Running cost is one part of the decision. This page leaves out purchase price, installation, rebates, and the fact that a heat pump also does the job of an air conditioner in summer. A system that costs more to run can still be the better purchase once those are counted. What you get here is the number most calculators get wrong, computed from your state's real heating season prices.
Where these numbers come from
I want to be specific about this, because most sites that publish heating cost figures do not say where they got them.
I pull residential electricity and natural gas prices directly from the U.S. Energy Information Administration through their public API, using a script that lives in this project and that I wrote. It requests the same two series for every state: retail electricity price for the residential sector, and residential natural gas price in dollars per thousand cubic feet. Nothing is typed in by hand, and nothing is adjusted after it arrives.
I average those over the most recent complete December to February window, currently 2025-12, 2026-01, 2026-02, using the identical three months for every state, and this site covers 51 of them including the District of Columbia. Where EIA has not published a complete gas season for a state, the gas figure is left empty rather than estimated and every other fuel for that state is still priced. The state index names the six that affects.
I made a real mistake here on the first build. I took the most recent month EIA had published, which for gas resolved to May or June. EIA's residential price is total revenue divided by total volume, and in summer the fixed monthly customer charge gets divided by almost no gas, so the apparent price roughly doubles. I was answering a winter question with a summer number and it biased every verdict on the site toward the heat pump. Georgia's gas price was overstated by 114%. That is written up in full in the corrections log, and it is the reason I now average a fixed heating season window instead.
Everything downstream, the calculator, every state page, and every table in the comparison guides, is generated from that one data file. There is no second copy of the numbers and no hand editing, so the tool and the writing cannot disagree with each other.
Everything on this site, and which part answers what
This site covers one subject, what a heat pump costs to run, and it tries to cover it completely rather than broadly. Here is all of it, grouped by the question you probably arrived with.
What will it cost me?
The calculator above gives a figure. The running cost guide explains what that figure means, where it comes from, and gives a better method if you have a year of bills to work from, which needs no assumption about your house at all. Running cost by state applies the same model to every state with complete EIA data, and each state page carries its own prices, its own verdict and the households reported from that state. The methodology lists every constant and every condition under which this model would give the wrong answer.
Is it cheaper than what I have now?
That depends entirely on what you are replacing, and the answer is genuinely different for each one. Against a gas furnace the heat pump loses on running cost in most of the country, and the reason is the price ratio rather than the climate. Against propane and heating oil it usually wins. Against electric resistance it wins everywhere, by roughly the coefficient of performance, and it is the one case where nobody needs a calculator. A dual fuel system beats both machines in most states, and the page works out where the crossover temperature actually belongs from your own two prices rather than from a default. The comparison hub puts all five on one scale.
My bill has already arrived and it is too high
Start with the four things that make a winter bill run high and how to tell which one is yours, including a line by line breakdown of a real bill showing why the rate you think you pay is usually half the rate you actually pay. The single most common cause is auxiliary heat running when it does not need to, and the lockout test will tell you whether that is happening using two numbers off your own bill. The questions people actually ask when the bill arrives covers the rest, including whether switching the system off while you are out saves money, which has an answer and it is not the one everybody gives.
To find out whether your usage is normal at all, the household dataset holds real winter consumption from people who published their own numbers, normalised so houses of different sizes in different climates can be compared. Every household in it above the normal band turned out to have a specific and mostly fixable cause. How much heat resistance strips actually deliver and what the measured studies show sit underneath that.
Why should I believe any of this?
Reasonable question, and the honest answer is that you should check. Three contractor interviews and a commissioned load calculation are the only first hand material here, and that page tests this site's shakiest assumption against a real document. The corrections log records what this site has got wrong, with the figures before and after, including a gas price error that flipped the verdict in fifteen states. The open items list says what is still unresolved, how this site is made covers where the data comes from and how it is rebuilt, and the editorial standards set out what happens when an error is found. The note on the 3.5 to 1 rule explains why the number everyone quotes is a special case rather than a law, and the glossary defines HSPF2, COP, AFUE and balance point if any of that was unfamiliar.
Common questions
Does a heat pump always cost less to run than a gas furnace?
No. It depends on the price of electricity where you live relative to the price of gas, and on the efficiency of both machines. Across the 45 states with a complete EIA gas season, a cold climate heat pump costs less to run than a 95% AFUE gas furnace in 4 states and costs more in 33.
Then why do so many calculators say heat pumps always save money?
Two reasons. Most use a single national average price instead of your state. Many also pull the most recent monthly gas price from EIA, which in summer runs roughly double the winter figure, because that price is revenue divided by volume and summer volume is near zero while the fixed customer charge stays.
What about propane, heating oil, and electric baseboard?
A heat pump wins comfortably against all three in almost every state. Propane and heating oil deliver heat at roughly two and a half times the cost of natural gas, and electric resistance heat is the most expensive of all. If you heat with any of those, the running cost case for a heat pump is strong.
Does this include the cost of buying and installing the equipment?
No. This calculator answers the running cost question only. Purchase price, installation, rebates, and the fact that a heat pump also replaces your air conditioner are all real considerations and all excluded here.
What is HSPF2?
Heating Seasonal Performance Factor 2, the number of BTU of heat a unit delivers per watt hour of electricity across a heating season under AHRI 210/240-2023. Divide it by 3.41214 to get a seasonal coefficient of performance. An HSPF2 of 11.0 is a COP of about 3.2.