Heat pump running cost in New York
In New York, a heat pump costs about 94% more to run than a 95% AFUE gas furnace. Against propane, heating oil and electric baseboard the answer is different, and all four are below.
New York energy prices, 2025-12, 2026-01, 2026-02
The December 2025 to February 2026 EIA heating season mean, pulled . Every figure on this page is in the CSV and JSON download, published under CC BY 4.0.
| Figure | Value | Source |
|---|---|---|
| Residential electricity | $0.286/kWh | EIA, by state |
| Residential natural gas | $1.56/therm | EIA, by state |
| Propane | $3.36/gal | EIA, PADD 1B, Central Atlantic |
| Heating oil | $3.92/gal | EIA, PADD 1B, Central Atlantic |
| Electricity to gas price ratio | 5.37 to 1 | derived |
| Breakeven ratio for this equipment | 2.78 to 1 | derived |
| Annual heating degree days | 5,399 | EIA, MAC division |
New York ranks 43 of 45 states by electricity to gas price ratio, where 1 is most favourable to a heat pump.
Cost to deliver one kWh of heat in New York
| Current system | Fuel price | $ per kWh heat | Heat pump verdict | Difference |
|---|---|---|---|---|
| Natural gas furnace | $1.56/therm | 0.0560 | More expensive to run | 94% |
| Propane furnace | $3.36/gallon | 0.1320 | Cheaper to run | 18% |
| Heating oil furnace or boiler | $3.92/gallon | 0.1163 | Too close to call | tie |
| Electric resistance (baseboard or furnace) | $0.286/kWh | 0.2858 | Cheaper to run | 62% |
| Heat pump, COP 2.64 | $0.286/kWh | 0.1084 |
Assumes a cold climate heat pump at HSPF2 11.0 with 10% of heat from electric resistance backup. Propane and heating oil are published by EIA regionally rather than by state, so those are the nearest regional figures. Change any of it on the calculator.
Rough monthly cost in New York
| System | Winter month | Per year |
|---|---|---|
| Gas furnace, 95% AFUE | $204 | $1,276 |
| Heat pump, effective COP 2.64 | $395 | $2,469 |
An 1,800 sq ft home of typical construction, at 5,399 annual heating degree days for the MAC census division. The square foot coefficient behind this is a rule of thumb rather than a measurement and can be wrong by half in either direction. How this is worked out, and a better method if you have a year of bills.
What this page does not tell you
Running cost only, at state average prices. It excludes equipment and installation cost, rebates, and the fact that a heat pump also replaces an air conditioner. State averages hide large differences between utilities inside New York, and this page cannot know your balance point or how much backup heat your system would actually use, which is the input that moves the answer most.
Has the answer in New York been stable?
Everything above is one heating season. Whether that is worth acting on depends on how much New York moves, so here is the same comparison run against every season since 2010-11.
| Since 2010-11 | Value |
|---|---|
| Heating seasons on record | 16 |
| Times the verdict changed | 0 |
| Typical move in the price ratio, year to year | 7.0% |
| Margin today | 94% |
Stable
The answer in New York has not changed once across 16 heating seasons, and today's margin is more than three times what this state's prices typically move in a year. It would take a move well outside anything in the record to flip it. How this was computed, and what the other states look like.
What New York households have actually reported
Everything above is a state average price applied to a model house. Below are real New York households who published their own winter figures, normalised so they can be compared with each other and with your own bill. There are 4 so far.
| Household | Floor area | kWh | Wh per sq ft per HDD | Worked through in |
|---|---|---|---|---|
| Upstate New York Ducted Mitsubishi, all electric |
3,800 | 2,300 January |
0.53 | the household dataset |
| Upstate New York A single 18,000 BTU/h mini split |
1,350 | 1,674 January |
0.84 | the household dataset |
| Queens, New York City Three GREE ductless mini-splits, one per room, installed by the landlord |
1,150 | 2,923 January |
2.24 | the section for renters |
| New York, 90 minutes north of NYC 12 year old Carrier heat pump with electric coil backup |
1,733 | not given | cannot compute | the household dataset |
Watt hours per square foot per heating degree day, which is how this site makes houses of different sizes in different climates comparable. Households with no identified fault sit between 0.53 and 0.92, and what that spread means in a northern climate is worked through separately. The full dataset, the method behind that figure, and what was wrong in every household above the band.
If your own figure lands well above these, that is worth chasing rather than accepting, because every household in this dataset above the normal band turned out to have a specific and mostly fixable cause.
The model against real New York houses
Every monthly figure on this page comes from one assumption: that a house loses 8 BTU per square foot per heating degree day. That is a rule of thumb, and this is the only page on the site where it can be checked against somebody's actual meter in the same state.
| Household | Model predicts | Actually used | Ratio | Verdict |
|---|---|---|---|---|
| Upstate New York Ducted Mitsubishi, all electric |
3,831 | 2,300 | 0.60x | Used less than predicted |
| Upstate New York A single 18,000 BTU/h mini split |
1,768 | 1,674 | 0.95x | Within 15% of predicted |
| Queens, New York City Three GREE ductless mini-splits, one per room, installed by the landlord |
1,159 | 2,923 | 2.52x | Used more than predicted Cause inferred by this site |
Prediction is this site's default model applied to that household's own floor area and heating degree days: 8 BTU per sq ft per HDD delivered, at an effective COP of 2.64. Their figure is what they reported using.
If your own house lands far above the prediction, the model is not usually the thing that is wrong. Auxiliary heat running when it does not need to and a compressor lockout left at a factory default account for most of the households above the band in this dataset, and both are settings rather than equipment.
Where to go from here
Because gas wins on running cost in New York, the question worth asking is not whether to switch but whether a dual fuel system beats both, which it usually does, and where its crossover temperature belongs.
If you already have a heat pump and the bill is higher than you expected, start with the four things that make a winter bill run high and how to tell which one is yours. If somebody has told you your usage is insane, the questions people actually ask when the bill arrives covers what that comparison is usually missing.
If you are still choosing, the comparison depends on what you are replacing: a gas furnace, propane, heating oil, or electric baseboards, where the heat pump wins by a wide margin in every state including this one.
And if you want to check the arithmetic rather than take it, every constant and limitation is listed, the data is rebuilt from EIA on a documented schedule, and the things this site has got wrong are published rather than quietly edited.
The reasoning behind these numbers
The verdict at the top of this page turns on one threshold, and that threshold is set by your equipment rather than by your climate, which is the single most misunderstood thing about heat pump running cost. The price it is compared against comes from EIA, averaged across December to February, and using any other window inflates the gas price enough to flip the verdict in fifteen states. This site published that mistake before catching it.