Skip to content

Retrofit This

Should we be aiming the money hose at retrofits?

Let me just come out and say it. Phew! Sanity in environmental policy is on its way to being restored. The Biden administration has not only managed to vaccinate 67 million Americans in under 100 days (I will be back at those Costco sampling stations in no time!), but across the board there are efforts underway to fix the consequences of broken markets from sea to shining sea. Our own master blogger and all star economist Catherine Wolfram is hard at work as Deputy Assistant Secretary for Climate and Energy Economics at the Treasury. Basically our entire graduating class of Ph.D. students has accepted jobs in DC. The efforts underway range from significant headway in global climate policy, to renewable energy policy, regulation of public lands, protecting our waters and coastlines to name but a few. What is great about the current push is that the administration is using good old Keynesian stimulus thinking during a time when interest rates are essentially zero, leaving little room for monetary policy. They are pointing that hose of government money straight at the energy sector – in the hopes that it will both make the economy greener and cleaner and generate lots of jobs – as we need folks to get back to work after this horror show of a year. 

While reading through the policy proposals – and again, I liked most of them – I nearly choked on my popcorn when I saw that we are going to try to retrofit 4 million buildings. I understand that fixing up homes of course requires lots of labor from contractors, and quickly injects capital into many local economies. The rationale is that it will also save a lot of energy. But….we tried to do this last time around. We have solid evidence from a gold standard randomized controlled trial (which is the way we learned that the COVID-19 vaccines work), showing that most folks don’t necessarily jump on the opportunity to get their homes retrofitted – even when it’s free to them. Further, the costly retrofits don’t save nearly as much energy in practice as engineering calculations suggested and are a very expensive way to reduce the emissions of greenhouse gases. 

Further, I just don’t understand why this should be a priority currently. If you take the long view, we are going to take a swing at making renewables + storage so cheap that fossils will only come online when it’s really hot or really really really cold. This switch to a close to carbon neutral grid is going to be plenty expensive and will generate a significant number of jobs in the short and medium run. Building retrofits are a bet on a complementary long view – namely driving down the energy demand from those buildings over a similar horizon. Whether one believes that most electricity will come to us at or close to zero marginal cost generated by gadgets that have no carbon footprint and no emissions of local air pollutants or not – these retrofits need to “beat” the very low long run cost of renewables. So if electricity will be cheap and clean, why spend so much money on something that we know does not achieve its stated goal and in the long run may not really be needed? The answer I got from my even further to the left set of friends and energy nerds is: comfort. A properly sealed house is so much more comfortable. So I wanted to try this out. 

My house was built in 1946. It’s a typical California Rancher. 1500 square feet of wooden sticks, crappy dry wall, all lacking insulation. As we are not spending money on vacations, I called my contractor and he said my best bet would be to put a foot of cellulose (blown in) insulation into the attic and insulate my floors from underneath with the pink fluffy stuff that makes you itchy, plus put a vapor barrier under the house (which really is just a big sheet of plastic to prevent humidity from leaking into the house). So we did this. The cool thing about my house is that I have installed three outdoor high frequency weather stations, and an indoor high frequency temperature and relative humidity sensor (which I had used for a very cool paper). I also like numbers.

So I recorded my natural gas consumption (I have a natural gas furnace) for the two weeks before the installation and the two weeks after the installation along with temperature indoors and outdoors to see what happened. I then fired up my old trusty computer and chased this through my statistical software. What I found surprised even me! My natural gas consumption went down by 0.58 therms per day. That is a 25% decrease! What was even more surprising is that my electricity consumption went down by 4.55 kWh per day, which is an 18% decrease! So that is pretty good. And I felt amazing about myself. If I kept up these savings for the heating season, I would save 58 cents per day on gas. The electricity consumption savings are tougher to calculate. The savings here come from the natural gas heat not turning on so frequently requiring a fan to blow the hot air around the house. The real savings probably do come from the fact that I stopped running the electric space heater in my office. But let’s call it a dollar a day, bringing me to somewhere around 400-500 dollars a year in savings. That is a payback period of 12 years. Not great, yet not terrible. But if you understand science, you should ignore my calculation and read about what Catherine and Meredith did with their weatherization RCT, which gives you credible causally valid numbers. 

But am I more comfortable? I am not so sure. When I look at the temperature measurements indoors and measures of relative humidity, things are exactly the same as before. Statistically. With data. No matter how I slice and dice the data. I also surveyed the members of my household to see whether they detected any differences in comfort. The answer was a solid “maybe – but I’m not sure”. But it definitely was not a resounding yes. And in order for this increase in comfort to be welfare improving, one should feel it. 

So what’s the takeaway? Max is going to make his money back in a decade or so and the climate is better off as Max is burning less natural gas and using less electricity. At a high cost. You’re welcome. But Max is well paid and lives in a very nice neighborhood. What about folks that are not as well off and live in much lower quality housing stock? I question whether energy efficient building retrofits are the best use of scarce public dollars. I would argue that meaningful rate reform would probably be a much more effective way of addressing inequalities while affecting a much larger population than the proposed retrofits. But what about all of those construction jobs, Max? Why don’t we point the Keynesian hose at building new, energy efficient, possibly low income housing near public transportation infrastructure instead? Seems like a much smarter bet than trying to push retrofits on folks, who are not very excited about them.

Keep up with Energy Institute blogs, research, and events on Twitter @energyathaas.

Suggested citation: Auffhammer, Maximilian“Retrofit This” Energy Institute Blog, UC Berkeley, April 19, 2021,




Maximilian Auffhammer View All

Maximilian Auffhammer is the George Pardee Professor of International Sustainable Development at the University of California Berkeley. His fields of expertise are environmental and energy economics, with a specific focus on the impacts and regulation of climate change and air pollution.

19 thoughts on “Retrofit This Leave a comment

  1. Max, a 12-year payback on your home investment is similar to the annual long-term rate of return on stock market investments – about 8%. Plus this perhaps improved your comfort a little. If we are going to fully decarbonize our economy, at some point we’re going to have to make investments with lower returns, and perhaps even some investments with negative returns; 8% is not bad.

    For low-income households, the benefits in terms of comfort and health will on average be greater than yours. For example, a survey by the research firm APPRISE of participants in the low-income weatherization program found improvements of 12-23 percentage points in various measures of comfort following weatherization ( But to achieve these benefits is not cheap, as many of these homes need basic home repairs before things like insulation can be installed. To pay for these improvements, both climate and housing funds will be needed. You cited the 2015 Haas study on the low-income weatherization program in Michigan, but as you no doubt know, this study has been extensively critiqued by others (a sample of feedback can be found here: ). In particular, I would note that the Michigan program did not follow good practice and calibrate their energy audits to actual home energy use (

    In your blog you suggest that a key metric is the relative cost of energy efficiency and renewable energy. While I agree at a very high level, details are essential. First, the comparison needs to factor in the cost of meeting peak demand when the sun isn’t shining and the wind isn’t blowing. In the future, for much of the country, peaks are likely to be driven by winter polar vortex events when temperatures are so low that even cold climate heat pumps may not work well. The cost of heat during these events need to be factored in, whether it’s maintaining a gas network despite plummeting demand driven by electrification or an electric network that has to be expanded to serve electric resistance heat in operation for just hours a year. Often in these situations, energy efficiency to reduce peak heating needs will be part of the lowest-cost solution (see here for example: ). Second, all of the benefits of both efficiency and renewables need to be included, including comfort and health. In the recent Texas cold-wave for example, inefficiency was a substantial contributor to both discomfort and frozen pipes.

  2. I’m with you on the expense-versus-return of retrofitting so many homes. I had my home in Pleasant Hill, CA airsealed and insulated several years ago and couldn’t believe the difference in comfort but I didn’t see a huge difference in my utility usage. The Building Performance Institute has made such retrofits a fairly simple science that can indeed be measured — but even they’ll caution that utility consumption and energy savings are not necessarily the same side of the coin. It seems to me that the Biden Administration’s best bang for the buck would be an all-out effort to educate the American public and change the behavior side of our energy efficiency discussion — think the “Only YOU Can Prevent Forest Fires” or “Don’t Be a Litterbug” campaigns of bygone days. Energy Upgrade California is trying but the messaging is just not really clear.

  3. I’m not sure what’s your (or the federal government’s) definition of “retrofit.” In the context of climate change policy it covers much more such as the changeout of the HVAC, water heater and stove, and even adding solar + storage to relieve the grid and avoid environmental damages in the desert (having worked on the DRECP). The Acting now to truly retrofit in this way can avoid emissions on the magnitude of decades.

    To gain the benefits of renewable electricity requires that the end uses be converted to electricity. Given how slowly we are building new houses (which are being pushed to be all-electric across the state), retrofits a necessary means of gaining those benefits. How we fund those retrofits is an important question.

    In fact, I don’t see adding insulation (which needs to be done as periodic maintenance in any case) as “retrofit”–it’s just the normal course of action. This site says it should be done about every 15 years. ( The actual cost calculation should account for the depreciation of the existing insulation and the need for future replacement.

    As for comfort, having lived in Village Homes in Davis where we didn’t have to turn on the AC until it was a 100 degrees, and our furnace ran only in the morning during the winter, I can attest to the comfort gain. We currently live in a 1956 house with a crawlspace that we added a number of retrofits to 4 years ago. We find our baffled blinds make a tremendous difference, but we still have an uninsulated wall (built with no dead space) and floor and I can tell the difference during the winter. And we need to turn on the AC in the mid 90s instead.

    As to lowering electricity bills, we don’t need federal money–we need the State Legislature and the California Public Utilities Commission to hold PG&E and the other utilities accountable for the poor management of their generation portfolios. PG&E’s cost of generation could be at least 20% or more lower if it had made prudent decisions over the last decade.

    (BTW, how did you account for outside conditions when you measured your energy changes?)

  4. Thanks for tackling this with data albeit on one house! The benefit missing in your analysis is that substantial conservation/energy efficiency investments also contribute to our clean energy future by minimizing the amount of new electricity generation needed. When one looks at generation increases needed to support building and transportation electrification over time, conservation and EE are critical tools for optimizing feasibility, land use, transmission increases etc. BTW, my personal anecdote is that the major retrofit contributions to my mid-century modern home’s comfort were replacing single pane windows and moving to electric heat pump space heating from gas.

  5. Try adjusting your control strategy to see if acceptable comfort can be maintained with less equipment operating hours. I’ve done a lot of shell improvements to my house too, i.e. I’m an energy nerd too. One of the biggest benefits I notice is that turning everything off at night still leaves the house acceptably warm the following morning. A big savings. Like you I’m sure, I have no mechanical cooling, just awnings, shades, and occasionally, fans. Nonetheless, our house is noticeably more comfortable on our rare hot days. One benefit, we’ve been able to sacrifice some shade trees for better solar insolation without any discomfort. We also enjoy better solar gain in shoulder and winter months. Of course you’re strictly right, the correct technical comparison is at exactly the same amenity level, but slackening that requirement might open up some big gains.

  6. Dear Max,
    I am not sure I agree with your points, on two counts. First, storage of electricity remains and will likely remain an expensive way to adapt to renewable fluctuations. Second, another way to adapt to renewable fluctuations is to use building thermal storage. For example, pre-cool a house in a hot climate in the afternoon to reduce the effect of the duck curve. So, in addition to the energy savings benefits of increased insulation, one should also think of the ability to better insulate internal thermal mass from the outside and therefore be able to more opportunistically make use of renewable energy.
    Ross Baldick

  7. Nice post Max!

    One quibble about this claim:

    ” If you take the long view, we are going to take a swing at making renewables + storage so cheap that fossils will only come online when it’s really hot or really really really cold.”

    The toughest demand to serve in the future may not be the peak loads we struggle with today. Tough days will be when demand is high *relative* to renewable supply, and based on analysis we’ve done so far, these can be very different days. It’s also the length of the spell that matters and the amount of stored energy — inventories will rule electricity prices much like they do for other stored commodities today.

    Somewhat relatedly, and more on point: a well-insulated house can do much better doing very lean-energy times. Not all kWh have the same opportunity cost, and this will be especially true in the future. Consider the recent episode in Texas: if folks had well-insulated houses, turning down the heat a few degrees during the tough spell could have saved a bundle.

    I see and appreciate your point that retrofits may not be the highest priority right now, but the calculations are a lot more complex in a near 100% renewable system.

  8. Here’s the deal, Max: 70% of the buildings that will be here in 2050 are here today, and if we don’t improve their efficiency, we can’t be within 2 degrees as a planet. Remember that the Waxman-Markey bill analysis by EPA predicted that cap-and-trade prices would get to $150/ton by 2050. Several other analyses said the same. If you drop $150/ton into your non-felicific calculus, you may get a better return.

    Also, remember that utility integrated resource (least cost) planning is all based on getting a fair return over the life-cycle of assets acquired. Your insulation will last longer than 12 years.

    You need to look at the full layer cake of savings, not just the simple savings on the heating bill. You will also save on your cooling bill. Even on your water bill (if you use evap cooling).

    Also, if you looked at the cost of your insulation against the cost of a heat pump retrofit, you might well find that the insulation you added reduced the size of heat pump required, and would save you $3,000 or so up front on the capital costs of retrofit by reducing the design day heat loss of your house. Add that to the savings over time, and you have a decent return.

    On Sunday, when you put money on the collection plate, do you ask what the payback period is? If you do, you are given a solemn answer: all eternity. That’s the right time scale when taking actions that may preserve the habitability of the planet.

  9. Hi Max,
    Thank you for the interesting read. I was curious if you had considered the difference in climates around the United States. The climate here in the Bay Area is relatively mild while other areas such as in New England, Southeast, and Southwest U.S. experience significantly low and/or high temperatures throughout the year. I expect their energy and cost savings to be greater (especially since New England states use a lot of fuel oil and propane) than those you experienced after your home retrofit (I assume you live here in the Bay). Additionally, the difference in indoor versus outdoor temperatures would be greater if they are experiencing typical summer and winter weather. Do you think the emphasis on building retrofits should be geographically focused, such as in New England where a bulk of the building stock are 100+ years old and have high electricity costs? Building retrofits might not make so much sense here in the Bay Area, but it might be a completely different story in colder and hotter climates.

  10. Subsidizing retrofits at a “Max” socio-economic level saves energy, but does not particularly do so at a sub-Max level.
    Those of us at the Max level, will use energy to make ourselves comfortable because we can afford it. Investement into retrofits becomes an economic investement and one can argue if is of public benefit to reduce energy consumption by subsidizing that investement.
    The “make renewables too cheap to meter” crowd would not think so.

    At lower socioeconomic levels, the situation is quite different. Often occupants can’t afford the energy to keep themselves comfortable. We already subsidize their use through rate structures and similar LIHEAP-like programs. That Energy Justice is paying to fill a leaky bucket, that is making an energy mess for everyone. Fixing sub-Max houses will not save as much energy as fixing Max-level homes, but it will allow more people to be able to heat and cool thier homes and potentially save money in subsidy programs.

    • I like MAX!!!’s sentiment (to continue to play on his words).

      His observation highlights an issue that I find more salient–the apparent overemphasis by economists on “efficiency” as measured through an overall monetary metric as the singular benchmark to be applied to determining policy. As an economist myself, my view is changing, particularly when assessing grand policy choices that involve billions of dollars across millions of people. We seem to forget that the elusive “util” isn’t directly equivalent to dollars, and that utils most likely have a non linear relationship to cumulative well being (along with the fact that not all aspects of well being have a monetary measure.) We need to be much more aware of the distribution of those costs and benefits, both across current well being and resources, and over time. Understanding how technologies that affect our well being evolve over time also needs to enter that equation.

%d bloggers like this: