Upgrading an old oil furnace and AC unit to a new high efficiency heat pump has been one of the two keys to making my family's home in PA produce more energy than it consumes (the other of course was adding a big fat solar array). Now that we've lived in the house for over a year I thought it would be a good time to look back at the data and see what sort of return on investment (ROI) this heat pump has given us.

UPDATE: After living in my house for a couple more years I can attest that the heat pump has handled each winter flawlessly. In 2019 it used a total of 4,375 kWh (around $300 worth of solar energy) to heat and cool my home and in 2020 it used 5,106 (around $350). Check out this thread to see why my use increased. Coincidentally 5,106 is the exact amount of energy my resistance hot water heater is estimated to use in a year. Maybe I should upgrade it.

After one year we have ended up only spending $477 total to power our heat pump for both heating and cooling. This low $477 annual HVAC spend assumes we pay 10.5 cents per kwh like our neighbors do. In reality though our solar panel system gives us electricity for less than this. I like to calculate that price difference as part of the solar ROI and leave heat pump ROI based on the 10.5 cents all our neighbors pay for power. The estimate for how many kWh's our heat pump uses comes right from our Bryant thermostat display. I've compared the thermostat's numbers to our total monthly electricity bills to double check it and it seems about right. It also nicely breaks down the estimate based on how it is using the energy, and if you write down the totals every month you can make a nice graph of them all.

Looking at this graph you'll notice a few things immediately. First off, we didn't turn our heat pump on a single time during the month of September. This is because the weather here in central PA was very nice that month. Opening our windows at the right times and wearing the right clothing was all it took for us to be comfortable all month long. I was hoping that May and June would be the same, but as you can see we did use a tiny bit of electricity on heating and cooling during those months. Maybe we're just getting weak in our old age?

Next off, you'll see that we use far more electricity for heating than for cooling and it has a big spike in January. This spike contains a significant amount of "resistive heat" which comes from old style heating coils like the kind that glow in a toaster. When the outside air gets cold enough that the heat pump alone can't maintain the thermostat's set temperature these turn on. Getting heat out of these coils can take 4 times as much electricity than it would take to produce the same heat from a heat pump. As the temperature drops a heat pumps efficiency drops as well, so when its really cold out resistive heat coils may only use twice as much electricity to produce the same amount of heat, but you still want to avoid running them as much as you can.

The above graph shows that while the resistive heat is a significant portion of our total energy use in January it wasn't all that bad, only about 1/4 of the total electricity we used to heat during that month. This means resistive heat produced somewhere between 1/8th and 1/16th of the total BTUs of heat we used that month. I'd say a "bad" heat pump would be one that puts more energy into resistive heat coils than the heat pump during a cold month. What is "bad" of course depends on how cold a "cold month" in your area really is and it's still important to compare this to how much the unit costs to install. What may be a "bad" heat pump in Pennsylvania might be a great heat pump in Florida.

The Key Metric To Look For In A Heat Pump

When looking at heat pumps there are are few key things to look at. First, and most obviously is the price of the unit. After this you can look at something called the HSPF, which tells you roughly how efficient the heat pump will be over a season. The problem with HSPF is that the weather you have is likely different than the weather HSPF was calculated for. If you live in a cold place like central PA I recommend looking at how many BTUs the unit puts out at 47F and at 17F. Any reputable heat pump should have ah AHRI certification with this information.

Our Bryant unit is rated at 33.4k BTUs at 47F and 31.6K BTUs at 17F which is a nice small drop. I've seen a similarly priced Lennox model that produces 32.2k BTUs at 47F but drops way down to just 21k BTUs at 17F. It will use resistive heat to pump out the 11.2k of BTUs the heat pump can no longer generate when the outside air is 17F. As I explained above resistive heat is so much less efficient than a heat pump use 2-4 times as much electricity to produce the same amount of heat. This means that the resistive heat elements in this Lennox unit can use up to twice the electricity as its heat pump when temps drop very low. Imagine if the orange top to my red bar in January was twice as thick as the red bar below it and you start to understand why I call heat pumps like this "bad" for climates like Pennsylvania.

What About A Ground Source Heat Pump?

One way to avoid the low 17F performance issue entirely is to buy a ground source heat pump instead of an air source one. These take heat from a fluid that runs through a pipe buried underground rather than the outside air. Since the ground has a pretty stable temperature once you get down far enough these units perform about the same no matter what the outside air temperature is. The other good thing about ground source heat pumps is that since they are located inside your home they never need to run a deicing cycle like an air source unit does. These cycles use energy and can sound like someone running a small chainsaw outside your house (though they only run on the coldest days when you're unlikely to be outside).

The problem with ground source is that burying a long pipe deep in the ground is pretty dang expensive. It can easily add $10k to the installation cost of your installation. As I mentioned above I spent less than $500 running my air source heat pump last year. A ground source unit will probably only use 10-20% less energy than my unit so I could save at most $100/year.

If you invested $10k in a CD with a 1.5% interest rate it would return $150/year so a ground source heat pump would essentially never beat even a very low returning investment like a CD. If pipe installation costs are much lower where you are, or you live in a much colder area (like say, Canada) then it might make sense. But, for most of America I think that air source units have gotten so good that ground source heat pumps should be ignored.

How Much Did The Heat Pump Save In A Year?

I started out noting that we spent only $477 to power our heat pump last year, but how much did this save us? Since we had the heat pump installed before we moved in we can't answer this for sure. A home energy audit we had done before we moved in estimated that it would cost $1,756/year to heat and cool our 2,000 square foot house with it's old oil furnace and AC. By that estimate we're saving $1279/year. But, it was using the average American's thermostat settings, which are far higher than our own. We set our thermostat lower in winter (65) and higher in summer (79) and do some other tricks like opening our windows at the right moments during the day to save more energy. Taking this into account and looking at our heat pumps data I think we saved more like $1,000 this last year, or around $83 per month.

What's the total ROI?

To calculate the total ROI you need to compare the up front install cost to the monthly savings. Our high efficiency heat pump had a total install cost of around $15k after rebates. This seems like a lot, but based on the age of the furnace and AC we were going to have to replace them anyways. The median install cost for a new furnace in PA is $4500 while an AC is $4700. That's over $9k right there. Of course, given how little we use the AC we could probably have gotten away with just using a few cheaper window units instead of a whole house setup. Still it seems unlikely that we could have replaced both for under $5k total. That means that our heat pump cost at most $10k extra. If I invested this $10k in some stock/bond/CD that earned 6% each year then after 15 years I'd have $23,965. If instead I invest the $83.33 I save each month in the same stock/bond/CD that earns 6% each year then after 15 years I'd have $23,275. These are pretty close, and if you're paying taxes on your capital gains then the $83.33/month heat pump actually wins because with it you are only paying taxes on $8,275 worth of gains, instead of $13,965 (at our current 15% long term capital gains rates this amounts to $853 more in taxes).

The take away from all the numbers above is that if my investments return less than 6% per year then I am better off having paid extra for my high efficiency heat pump and if my investments return more than 6% then I'm losing. If you look at the return of the S&P 500 you'll see many recent years where it returned over 10% and maybe you'll think that a heat pump is a bad investment. If you go back further you'll also see years where it dropped, 10, 20, even 30%. This volatility is why people diversify their portfolios with safer, lower returning assets like bonds. If you look at the returns of a bond index you'll see them only returning over 6% a few times over the years. The return from this high efficiency heat pump is both more stable than the return from bonds and higher, what's not to love?

In the end the exact math on this will depend on the weather where you live, the sort of house you live in, the price you pay for power and your personal thermostat habits. Run the numbers yourself (or leave them in a comment here and I'll try to help) and see what sort of return you'll get on your own heat pump. If you're the sort of person that wants safe investments in your portfolio then you should definitely consider shifting money from those investments into a high efficiency heat pump when your current HVAC nears the end of its life. If you're a risk taker who owns all stocks then you should compare current mortgage interest rates (around 4% as I write this) to the return you'd get from upgrading to a heat pump. You'll probably find that expanding your mortgage to include the price of a new heat pump will actually lower your monthly payments. Whatever you do, I hope the real world numbers and analysis I've provided in this article can help you to profit greenly yourself.

An Aside on Gas Furnaces

In this post I compare the price of a heat pump to an oil furnace. Per unit of energy natural gas currently costs about half of what oil does. A gas furnace can also be installed quite cheaply if you're willing to DIY. One thing you will face with a gas furnace is a gas connection charge of around $15/month. You'll have to pay this all year round, even in the summer months when you are using next to no gas.

I ran the numbers for gas service I had at an old house in upstate NY and the monthly connection fee increased the cost per ccf of gas by nearly 35%! This was for an annual gas use of nearly 550 ccf, which I consider fairly high. If you are in a warmer state, or have a better insulated/air sealed home and use less than this then the % price increase will be even higher. For example, if I'd used only 200 ccf per year then the monthly usage fee would have nearly doubled my cost per ccf. In that case shutting off gas to my house and going all electric could allow a heat pump to provide a higher ROI than gas.

Further Reading: Now that you've learned a bit about heat pumps, check out my post on Solar Panels then put it all together with my post on Going Net Zero