What Can Building Retrofits Achieve?

April 10, 2009 | Michael Shellenberger,

Over the last few years we've heard a lot about how if just invested in efficiency, like retrofitting old homes, we would save huge amounts of energy and massively reduce our emissions (not to mention create millions of jobs, reduce inner-city poverty, stimulate the economy, etc.). This was the subject of an op-ed in the New York Times on Monday.

But today, the Times ran an interesting correction on the op-ed page:

An Op-Ed article on Monday, about renovating older houses to save energy, included an incorrect figure for the number of homes that would need to be retrofitted every year to save 200 million barrels of oil over a decade. The figure is 300,000, not 3,000.

I went and looked up some of these numbers. According to the U.S. Energy Information Administration, Americans consumed over 20 million barrels of petroleum per day in 2007. About half of that petroleum is in the form of gasoline and the rest is heating oil, industrial distillates, kerosene, jet fuel, etc.

Few houses, of course, are heated by petroleum, so the comparison is inexact. Moreover, I doubt the assumptions embedded recognize that people tend to use more electricity when efficiency lowers the total monthly bill.

The numbers provided by the Times offer a sobering view of how little retrofits will actually do. Here's what we find: If we were to retrofit 3 million homes over the next decade, we would only consume 10 days less petroleum (.27% less). If we were to retrofit 10 times as many homes, 30 million, (which is almost certainly not possible to do in 10 years), we would save just 100 days of oil over a 10 year period (2.7% less).

If these numbers are basically right, then it's safe to conclude that building retrofits will not contribute significantly to reducing overall global emissions. They might be good for other reasons, but they are hardly the basis for a strategy to mitigate warming.


Christina, thanks for the comment. As far as CO2 goes, we can convert relatively easily there. It appears that the CO2 intensity of a Quad of crude oil (about 75 million metric tons/Quadrillion BTUs) is just about a third of the carbon intensity of the average kilowatt-hour from the U.S. grid (about 1470 lbs/KWh or 215 million metric tons/Quadrillion BTU). So, if all of that at-home energy consumption saved was electricity, it would have a higher impact on CO2 emissions than the comparison to barrels of oil energy equivalent would imply. However, even in this scenario, if you retrofit one third of all U.S. homes to save one third of their energy on average, you'd save about 8 percent of total U.S. CO2 emissions from energy (currently around 6 billion metric tons/year, that's for CO2 only, not including other GHGs or emissions from non-energy related activities). In reality, it'd likely be much less than that, since homes also consume natural gas and fuel oil, both of which are much less carbon intensive than electricity in the U.S. average mix (maybe the savings would be around 6%). Again, not insignificant savings, but that's from a massive effort to retrofit over 30 million homes over a decade. That's also a decade in which energy demand is expected to rise under business as usual scenarios, so you are fighting to overcome that as well. There's no reason to not get started on that effort, but we shouldn't consider it reason to delay action to rapidly develop and deploy a whole portfolio of affordable and abundant clean energy sources. There's no "time bought" by this kind of effort to retrofit millions of homes.

By Jesse Jenkins on 2009 04 15

Seeing these numbers in terms of implications for global warming requires considering the decrease in carbon emissions associated with retrofitting home. About half of electricity consumption is met through burning coal, a much dirtier fuel than oil. Thus, carbon emissions are likely being decreased by a larger percentage than energy consumption.

Among the range of possibilities for decreasing carbon emissions and energy consumption, efficiency programs stand out as the most economically viable option. In most cases, efficiency improvements produce a net savings both financially and in terms of consumption. Because the technology is available and affordable, these programs can get off the ground immediately while the details of more complex programs such as cap and trade are being debated.

Retrofits can be costly and cumbersome. However, there is great potential for encouraging the use of best available technologies through regulation. The EPA already has mandated the use best available technologies in relation to other harmful gases through its administering of the Clean Air Act. Retrofitting 3 millions homes in a decade is small compared to the roughly 2 million new homes per year predicted to be built per year by a report of the Homeownship Alliance. This is without including figures on retrofitting existing commercial structures, which account for a large proportion of our energy use. Best technologies could also be mandated for future commericial construction to increase this savings figure.

By Christina H on 2009 04 15

After double and triple checking these numbers, it really does seem clear that while retrofitting homes to save energy is a no-brainer and clearly has some impact, it simply isn't (and cannot be) the cornerstone of our national energy strategy. Efficiency retrofits should be pursued as an adjunct to (and complementary to) a concerted energy innovation and deployment strategy which can make clean energy cheap and bring a whole portfolio of clean energy technologies to scale as quickly as possible.

This post clearly points out the scale issues involved with efficiency retrofits and makes it clear that those who propose efficiency as a way to "buy us some time" are really misleading us. If we launched an all-out effort to retrofit 3 million homes annually, we'd save just a 18 days worth of oil consumption equivalent each year. Not a lot of time bought there!

We should pursue efficiency, but it CANNOT and MUST NOT be approached as an alternative to a cohesive, strategic and well-funded effort to develop clean, cheap and massively scalable energy technologies to sustainably power the 21st century global economy.

By Jesse Jenkins on 2009 04 14

Those numbers look pretty much dead-on. The U.S. Residential sector consumed 20.77 Quadrillion BTUs (Quads) of primary energy in 2006 (from EIA's Annual Energy Outlook 2009). That's about 1/5th of total US energy consumption (which is at right about 100 Quads in 2006).

If we cumulatively retrofitted 1/3rd of all US homes to save 1/3rd of their energy, we'd save 20.77*.33*.33 = 2.26 Quads of energy annually. That's the energy equivalent of 373 million barrels of oil each year, or about 18 and a half days of oil consumption each year, 186 days over 10 years. That's also just 5% of annual oil consumption and about 2.25% of total US primary energy consumption.

So that puts the numbers Michael estimates above in the right ballpark. If 30 million homes is about 1/3rd of all US homes and they get about a 15-30% energy savings when retrofit (who knows how much the Times assumes you save when a home is retrofitted), you'd save on the order of 90-180 days worth of oil consumption equivalent, or just a couple percent of total U.S. energy demand for all of that effort. Helping save customers money on their energy bills is great, and there are plenty of reasons to want to do that, but we're not going to get to a sustainable and prosperous 21st century energy economy by retrofitting homes. It's just not the cornerstone of a smart energy policy.

By Jesse Jenkins on 2009 04 11