I’m kinda out of the loop on this stuff, but it seems like suddenly out of nowhere everyone around me is converting natural gas heat into electric heat pumps—specifically a thing called a “mini-split” which can be both a heat pump in the winter and an air conditioner in the summer. It consists of one or more indoor units connected to one or more outdoor units by a small-ish (well, small compared to an air duct) tube of flowing liquid refrigerant (which transfers heat between the indoor and outdoor unit).

An electric heater converts 1W of electricity into 1W of heat. A heat pump converts 1W of electricity into >1W of heat, because it pulls heat from the outside air—it’s basically an air conditioner pointed backwards. The ratio of heat to electricity is called Coefficient of Performance (COP).

The COP for heat pumps gets worse (albeit still >1) as the outdoor temperature goes down, basically because the outdoor air is colder so it’s harder to extract heat from it.

I live in a pretty cold climate; I figure the weighted-average outdoor temperature (weighted by how much heating I'm using) is 30°F (~0°C). A quick search suggests maybe I can expect COP of 3-3.5. The highest figure I found anywhere for that outdoor temperature was 5. I could be wrong here. (Update: in this comment I used a different method based on the "HSPF" spec and concluded that I should definitely expect weighted-average-COP below 3.5 where I live, unless I messed up the calculation, which is entirely possible.) 

Meanwhile I also need to account for my gas boiler not successfully transferring all the heat from "burning gas" to "warm interior" (as opposed to losing heat out the chimney or whatever). I think 80% is a reasonable guess for our not-particularly-well-tuned boiler. It should go up a bit, maybe towards 90%, when I get it tuned up or replaced at some point in the future, which I need to do anyway, it's super old and the plumber says it could catastrophically break any day.

I looked up my marginal electricity price and marginal gas price from my most recent utility bills, and did the unit conversions to a common unit of gigajoules (GJ). (I'm using marginal price because it's not feasible for me to rid of gas altogether, for other reasons.) It turns out that 1 GJ of electricity costs me almost 10× as much as 1 GJ of natural gas! (I fully expected the electricity to be more expensive but didn't know what ratio to expect.)

The COP>1 and boiler inefficiency help close that 10× gap, but they don't close it all the way, not even close. The existing natural gas boiler is still looking way cheaper to operate than an electric heat pump for me:

The top two rows are probably my best guess, and suggest that I'd be paying >2× more to operate a heat pump than a gas boiler. The bottom row is the most flattering-to-heat-pumps figure I can find, and gas still wins by a factor of 1.6.

It's looking like a mini-split would amount to many hundreds of dollars per year of higher operating costs compared to my status-quo natural gas boiler.

Some additional considerations:

  • Supposedly during summer the mini-split is a more efficient air conditioner than a normal (window) air conditioner. But I spend way more on heating than air conditioning (as assessed by the fact that my electricity bills are barely higher in the summer than the winter, whereas my gas bills have a blindingly obvious seasonality). So it doesn’t make a noticeable difference in the overall calculus.
  • I live in Massachusetts, USA. The warmer the environment you live in, the more attractive a mini-split would be compared to a natural gas boiler. That's because of both higher COP at higher outdoor temperature, and greater weight on the previous bullet point (air conditioning efficiency).
  • I’m not sure whether my most recent electric & gas bills reflect a typical price ratio. Maybe this is an unusual period (or I’m in an unusual area), with anomalously cheap gas and/or expensive electricity? (Update: A commenter points out that I live in an area with more expensive electricity than most of the USA—link.)
  • There are some nice things about mini-splits, like they’re quiet, they allow air-conditioning without blocking windows, they don’t involve pipes that can freeze and burst, etc. However, none of these things is worth "many hundreds of dollars a year" to me.
  • This table could even be overly flattering to mini-splits, because in cold climates like mine, there’s extra energy spent running the mini-split system not accounted for in COP, particularly there's an electric heater that keeps the outdoor unit from freezing over. Not sure how important that is.
  • I guess mini-splits could be more "green", depending on how "green" my electricity provider is (I forget). But I think that I wanted to spend many hundreds of dollars a year being more "green", I doubt that this would be a cost-effective way to do so.
  • Installation and maintenance costs are, I think, not radically different for mini-splits vs status quo in my particular situation, at least not compared to operating costs. My electricity provider has a rebate for mini-split installation, but I don't think that's enough to change the overall picture.
  • Update: A commenter adds that there's a cost savings from being able to set different temperatures in different rooms, especially if you have, say, a rarely-used guest room that you can allow to get very cold. Mini-splits let you do that, although there are other ways too, like baseboard radiators with adjustable air flow vents. Anyway this isn't a consideration for me personally; I'm already set up with as fine-grained temperature zones as I could possibly want to use.

So, switching to mini-splits seems like a bad idea in my particular situation. I don't know why everyone else is doing it. Maybe I'm missing something. I guess to be thorough I ought to call a local mini-split installer, explain my thinking, and let them try to talk me into it. And ask my friends and neighbors... ¯\_(ツ)_/¯

New Comment
13 comments, sorted by Click to highlight new comments since: Today at 10:38 PM

Few random thoughts from an energy economist; might be of some interest to you Steven and the odd reader:

  1. Your COP assumptions seem very reasonable indeed I think.
  2. I'm an energy engineer & economist studying practical energy and power systems. I see bias towards heat pumps all over the world, even in places where marginal implied GHG emissions with heat pumps seem higher than when directly heating with gas.
  3. Do many heat-pump people you talk about have PV panels on the roof? Given tariff asymmetries between feeding-in and drawing power locally to/from the grid, there tends to exist a strong self-consumption incentive; it can be highly lucrative to consume your PV power rather than to sell it to the grid.
  4. The largest share of your home electricity tariff  is often not at all the "energy" component but based on transmission costs and other fees/taxes (so what @Gerald Monroe writes is not necessarily a reason not to expect high electricity price differences)

Given tariff asymmetries between feeding-in and drawing power locally to/from the grid, there tends to exist a strong self-consumption incentive

MA has net metering, so unless you are producing more electricity than your house consumes averaged over the year, this isn't a consideration in favor of adding additional self-consumption.

In MA, it’s a bad idea (from a selfish monetary perspective) to produce more solar electricity than the amount of electricity you consume. So if you already have solar panels, then it’s plausible that you originally sized the panels to match your then-current electricity consumption. In that case, if you install mini-splits later on, then you’re consuming more electricity than you’re producing, and buying grid electricity.

However, another possibility is that you installed solar panels after already having mini-splits, or that you installed extra solar panels in the expectation that you were imminently going to also install mini-splits. In that case, I think FlorianH’s point is correct: we should be comparing to rooftop solar electricity costs not grid electricity price. (And this is a point in favor of mini-splits.)

I was getting bids from solar installers recently, and they absolutely have been saying “hey, are you planning to install mini-splits? or buy an electric car? if so, we should definitely quote you for more panels than one would think from your recent electricity bills.”

If the outside air temperature gets very low (or high?) I've heard that ground-source heat pumps might be the way to go: https://en.wikipedia.org/wiki/Geothermal_heat_pump

I watched this video a couple months ago. I came away thinking that the title of the video is roughly correct, but maybe not in all circumstances? However, I barely remember any of the details.  (One of the reasons I don't like watching videos for non-entertainment reasons!) You may want to see if it gives you any more information than what you have.

I will note that I think many people would pay hundreds of dollars per year for the advantages of mini splits that you do not weigh that highly and I think this could easily account for the reason you see many people switching to them. Personally, if the only advantage of mini-splits was getting rid of window A/C units I'd be very tempted to pay that hundreds/year.


For some people, another consideration is that you could use both. Thermostats frequently have a way to connect multiple heat sources, and there's presumably one fancy enough that you could decide between them based on the outdoor air temperature. I plan to do this whenever the one-way heat pump ("air conditioner") that came with my house breaks.

For that to be worth, you would need several things to be true though:

  • You want a heat pump for other reasons (air conditioning)
  • There are some days when the heat pump is better than the boiler despite electricity costs
  • The upfront cost of the slightly more expensive heat pump is made up for by the lower energy cost on the days you use it

You made a significant flaw in your calculations. https://www.eia.gov/electricity/monthly/epm_table_grapher.php?t=epmt_5_6_a  [average price of electricity].

The eia's data says 10.9 cents per kWh is the national average price per electricity.  Essentially for whatever reason (regulatory capture, a mishap with a nuclear plant, onerous local regulations) you are paying 2.4 times what you 'should' be paying, given the power company locally should be able to buy natural gas generators and fuel for around the same price as a power company anywhere else.  

Second, mini splits are significantly more often on the higher end than your numbers reflect.  https://www.energystar.gov/products/most_efficient/central_air_conditioners_and_air_source_heat_pumps

5.29 (COP is EER/3.4) is what the fujitus RLS3 gets, which is the bare minimum your neighbors would be installing.  There are more efficient models not listed on this chart, such as this 40 SEER model.

So the 'high end' estimate is actually the average and not high enough.

Anyways if you pay 2.4 times less for electricity, then a heat pump would be 65 percent as expensive as your boiler.  Combine that with a solar array, and remember the other advantages of mini splits: redundancy, zoning, and air conditioning as well.  Redundancy because a typical house will have 2-5 mini splits, so a failure of one is not a failure of climate control.  Zoning, aka turning on just the units in the occupied rooms, can add another factor of 2 energy savings on top of the above.  And you get air conditioning on the days you will need it.

the power company locally should be able to buy natural gas generators and fuel for around the same price as a power company anywhere else

I don't think this is true. Getting natural gas into this part of the country is very expensive. We don't have enough pipeline capacity, and voters are strongly against building more, so the marginal therm arrives on LNG tankers.

Thanks for your comment!

That's an interesting electricity price chart. It seems like I'm paying typical rates for my state, and I don't know why it's high compared to other parts of my country. I wouldn't say that's "a flaw in my calculations", since I'm calculating it for myself and I'm not planning to move, but it definitely sheds light on why mini-splits are more attractive for people in other places.

The "COP" in my chart is specifically "COP for heating the interior of a building when it's 30°F (~0°C) outside". I don't think it's true that COP under those conditions is equal to EER/3.4, because EER is not measured under those conditions. EER seems to be measured assuming a much smaller temperature difference between outside and inside. Heat pump COPs get worse and worse as the outdoor-indoor temperature difference gets larger.

There are other metrics such as HPSF meant to factor in aggregate performance. Since by choosing a fixed temperature you neglect all the days where the mini split has a huge efficiency advantage over combustion. Also you overlook the zoning. Larger houses that have extra rooms that are not always in use benefit from not heating those areas. And the solar. At your high local electric rates solar has a rapid payoff.

Hmm, well I was trying to ballpark "weighted average outdoor temperature", specifically weighted by how much heat I'm using. Like, if outdoor temperature is only slightly cooler than what I want inside, I need relatively little heat regardless, so the efficiency of that heat isn't all that important. My reference temperature of 30°F (~0°C) is very far from the lowest temperature we experience, it's close to a 24-hour-average temperature during the coldest three months.

I didn't know about HSPF, thanks for the tip! It seems to assume "climate region IV" (based on here for example), which I guess corresponds to this map which suggests that where I live (Massachusetts) is somewhat colder than climate region IV. Wiki says an electric heater is 3.41 and this says that 11.8 is about the highest HSPF out there (?), so if I divide them I get a weighted-average COP of 3.5, i.e. my initial ballpark guess was right on. But since I'm colder than "climate region IV" it would be even lower than 3.5. (To be clear, there are a bunch of things in this paragraph that I'm not sure about.)

Thanks for bringing up zoning, but we already have individually-settable radiators and do in fact keep unused rooms cold, so I don't think that's relevant to me personally. I'll add a bullet point for the benefit of other readers.

I just haven't gotten around to thinking about solar. One of these days...

Sure. For a new build in your climate zone, probably the most efficient setup is a tanked condensing natural gas water heater, ideally sorta centrally located. Then a hydronics air handler and vents that just cover the immediate area around the installation. This gives you the cost advantage of natural gas for most of the heating but you avoid the equipment cost of a second furnace. Tankless condensing is an option but in your biome there probably isn't a sufficient advantage.

Then mini splits around the periphery for heating/cooling during most days.

Wanted to say thank you for pulling this info together and sharing it - super helpful and appreciated!