Feature Article

The Challenge of Existing Homes: Retrofitting for Dramatic Energy Savings

July 10, 2007

One can hardly pick up a magazine or turn on the television today without hearing something about climate change. The issue finally appears to be gaining traction in our nation’s collective consciousness. Much of the focus of reducing greenhouse gas emissions rightly centers on how we design and construct buildings. Indeed, the 2030 Challenge, adopted by The American Institute of Architects, the U.S. Green Building Council, and others, calls for commercial and residential buildings being built today to use half the fossil fuel of average existing buildings and a gradual increase in performance so that new buildings are carbon-neutral by 2030 (see EBN Vol. 16, No. 6).

This goal is critically important. But it isn’t enough. If we as a society are to achieve the sorts of reductions in greenhouse gas emissions that climate scientists tell us are needed to prevent catastrophic climate change, we also need to tackle our existing building stock, including houses. The 2030 Challenge calls for the renovation of existing building stock equal in square footage to that of newly constructed buildings and achieving the same 50% reduction in fossil fuel use. Achieving this goal with residential buildings would require leading-edge energy retrofits on about 1.5 million existing homes per year—while today there are at most a few thousand energy retrofits per year that achieve such a target, and probably far fewer. This article examines the challenge of existing houses and the potential for dramatically reducing the energy consumption (and greenhouse gas emissions) of these buildings through major renovations.

The Scale of the Challenge

There are 124 million housing units in the U.S., according to the U.S. Census Bureau. Of these, about 85 million are site-built single-family homes, 8.6 million are manufactured or mobile homes, and the rest are units in buildings that house multiple families. If we include single-family, multifamily up to four units (which are similar to single-family in design and performance), and manufactured and mobile homes, there are about 103 million existing houses in the country, and it’s these that we focus on in this article.

The median age of all housing units in the U.S. (including multifamily) is 34 years (built in 1973), meaning that 50% are older and 50% are newer; 77% were built in 1950 or later.

In 2004, residential buildings in the U.S. used 21.1 quads (22 million terajoules) of energy, or 21% of the nation’s total primary energy consumption and 36% of total electricity consumption. (Primary energy is that energy contained in, or used to produce, the fuels and electricity we consume; it includes the fuel that utility companies use to generate electricity rather than the electricity itself.) On a per-household basis in the U.S., total primary energy consumption in 2004 was 185 million Btus (195 gigajoules), and total delivered energy consumption was 101 million Btus (107 gigajoules). Within the residential sector, the largest users of primary energy are shown in the table below. In different regions of the country, this split among energy end-uses is quite different; the share for space cooling, for example, is significantly higher in the South than it is in the rest of the country.

As for impact on global warming, the greenhouse gas emissions associated with residential energy use in the U.S. in 2005 totaled 330 million metric tons carbon equivalent. This represents 20.5% of total carbon emissions for the U.S. and nearly equals the total carbon emissions of Japan, whose economy ranks second to the U.S. but whose carbon emissions rank fourth—well behind China (now number one), the U.S., and Russia. Interestingly, of the carbon dioxide production attributed to our residential buildings, 70% is from electricity use and only 30% is from direct combustion of fossil fuels.

What is Achievable?

Major Residential Energy End-Uses in the U.S. (2004)

Totals may not add up to 100% due to rounding and data-collection complexities.

Source: 2006 Buildings Energy Data Book, U.S. Department of Energy, September 2006

Climate researcher James Hansen, Ph.D., of NASA argued in testimony before the U.S. House of Representatives in April 2007 that the level of carbon dioxide in the atmosphere at which catastrophic climate change would occur is likely about 450 parts per million (ppm), and could be lower. (The current CO2 level is about 385 ppm, up from about 315 ppm in 1958.) Although Hansen did not present any specific reduction in carbon emissions needed to stabilize carbon dioxide levels during that testimony, he has elsewhere suggested that a 60% to 80% reduction would be required. A number of prominent scientific and environmental organizations, including the Union of Concerned Scientists, have called for an 80% reduction in U.S. greenhouse gas emissions by 2050. Both California and New Jersey have adopted that target as a state goal, and legislation recently entered in the U.S. Senate and House of Representatives sets such a target.

The United Nations’ Intergovernmental Panel on Climate Change (IPCC), in a draft of a report to be released in October 2007, finds significant opportunities for carbon emissions mitigation in both new and existing buildings. According to IPCC, 30% of the expected global growth in emissions related to buildings before 2030 could be avoided with economic benefit. The report also finds that, although new buildings present opportunities for the most energy savings per building, existing buildings represent a greater opportunity for energy savings overall.

According to a 1998 study prepared for the U.S. Environmental Protection Agency, about 290,000 buildings are demolished every year, 245,000 of which are residential (about 0.2% of all residential buildings). If all of these demolished houses were replaced with new, carbon-neutral houses and if the energy performance of existing houses was not improved, this turnover rate of housing (0.2% per year) would achieve only about a 6% reduction in total residential carbon emissions by 2030 (10% by 2050)—far less than that called for by climate scientists.

To achieve the carbon emissions reductions called for by Hansen and others, we need to aggressively address existing buildings, including homes. In the 2005 report of the Pew Center on Global Climate Change, Towards a Climate-Friendly Built Environment, Marilyn Brown, Frank Southworth, and Therese Stovall, of Oak Ridge National Laboratory, suggest that with new residential buildings, energy consumption can be reduced 60% to 70% by 2020 compared with conventional practice. With existing buildings, achieving similar energy reduction targets will rarely be possible, but the starting point is often worse, so a reasonable target is to reduce energy use by one-half to two-thirds (the greater savings achievable only with homes that start out with very poor performance). Achieving such a reduction may be both challenging and expensive, but it is possible—and it would reduce loads enough that, with good solar exposure, rooftop solar panels (photovoltaic and solar-thermal) could enable such homes to achieve zero-net-energy performance.

Achieving a Three-Fold Reduction in Energy Use

Taking on such a robust challenge to reduce energy use in an existing home requires significant efforts on multiple levels. Cutting energy consumption by two-thirds or more necessitates addressing equipment for heating, cooling, and water heating as well as lighting, appliances, and electronics (including entertainment and computer equipment). However, one can’t get anywhere close to such a goal without tackling the building envelope: foundation, walls, windows, and roof (or attic floor). We will start there.

How far to go in upgrading the house envelope depends, to a significant degree, on the climate. In cold climates, where heating dominates energy use, it makes sense to start with measures that reduce heating loads: air sealing, adding insulation, and improving the energy performance of windows. According to building energy expert Ambrose Spencer, “a superinsulation retrofit in the Northeast can cut the heating load in half.” In warmer climates, where air conditioning accounts for a significant fraction of total energy use, reducing cooling loads is proportionally more important, with solar control measures and air sealing as key strategies. In any climate, however, tightening a house with insulation and air sealing can cause moisture levels inside the house to rise, so paying attention to the vapor profile of wall and roof assemblies is critically important.

The cost of such retrofits is an impediment—running tens of thousands of dollars per house. As Bruce Harley, technical director for Conservation Services Group, told EBN, “I’d be hard-pressed to find a house where you could do [a major energy retrofit] for less than $50,000.” Other experts EBN spoke with, including builder John Abrams, agree. Using the estimate of $50,000 per retrofit, achieving the near-term 2030 Challenge of retrofitting 1.5 million homes per year in the U.S. would cost $75 billion per year.

In any climate, major improvements to the envelope are easiest—and cheapest—when other major renovations are happening at the same time. “If someone’s planning on replacing their siding anyway,” said Harley, “that’s the time to add one, two, or three inches of foam insulation.” Adding insulation to the exterior of a house also allows occupants to live in the house while the work is being done, which can reduce costs. Air sealing, which also improves performance, can occur at the same time. Spencer argues that this is one of the easier ways to improve existing houses. “Every time a siding replacement is done, I’d like to see extra exterior insulation added,” he said.

Meeting Energy Loads with Renewables

If energy loads have been reduced enough, or if the house is located in an extremely sunny climate, it may be possible to meet those loads with renewable energy systems—achieving zero-net-energy performance. Solar-thermal heating systems, solar water-heating systems, and photovoltaic (PV) power systems are typically the most viable options for homes, and true net-zero-energy homes generally require a combination of these systems.

Spencer and others argue that renewable energy systems should be considered only after the building envelope has been addressed. “If you have the opportunity to deal with the envelope and you spend your money on photovoltaics instead,” he told EBN, “you’ve made it nearly impossible to improve the building later.” If the energy loads are brought low enough, however, a renewable energy system may make sense financially.

Occasionally, the economics point towards putting money into renewable energy rather than lowering demand. Eric Doub, a builder of low-energy homes in Boulder, Colorado, is working on a zero-net-energy retrofit that is nearing completion. According to Doub, “we didn’t have far to go to get this home to be superinsulated,” which meant that more money could be put into renewable energy sources. Doub added a $30,000 evacuated-tube solar thermal system and a 6.5-kW PV system that cost $25,000 after rebates. By comparison, the envelope improvements to the house cost $15,000. Although Doub admitted that “the PV system will be nowhere near cost-neutral at current electricity costs,” his clients’ desire to be able to provide all of their energy needs, including those for a plug-in electric car when one becomes available, with renewable energy made the PV system worth the money.

Case Study: Pettit Four-Square House, Massachusetts

When Betsy Pettit, AIA, of Building Science Consulting, bought a house for her daughter, she wanted to make it easy to live in. “I wanted to make a house where there would be no energy bills to worry about, and little maintenance,” she said. Pettit also saw an opportunity to use the retrofit project as a pattern for others: the 2,000 ft2 (190 m2) house, built around 1916 and located near Boston, follows the common “four-square” pattern made popular by Sears Roebuck kit homes, so many of the techniques used by Pettit could be replicated in similar houses throughout the U.S.

“Most houses that are 100 years old need all of their systems replaced,” said Pettit, “and this house hadn’t been touched since 1916.” The $300,000 renovation included added living space (finished basement and attic) and upgrades to existing spaces as well as added insulation throughout the house, new windows, a new roof, a new boiler and hot water heater, and a new ventilation system. The renovation increased the living space by 80%—to 3,600 ft2 (330 m2).

The work started in the basement, where Pettit installed a drainage mat and a layer of extruded polystyrene (XPS) rigid foam insulation on the existing slab, followed by a new reinforced concrete slab. Four inches (10 cm) of high-density polyurethane foam were sprayed onto the basement walls, then covered with non-paper-faced gypsum board to create the finished basement.

On the upper floors, the aluminum siding was stripped down to the existing board sheathing, and cellulose insulation was blown into the uninsulated wall cavities. Pettit chose to have the insulation blown in from the exterior, both to preserve the historic trim and plaster on the interior and to allow her daughter to live in the house while the work was being done. Because she was working with a trusted contractor, she told EBN, she did not feel the need to perform a thermographic analysis to ensure that the walls were properly insulated—but that can be a useful step with blown-in insulation and many insulation contractors offer the service. In the attic, 4” (10 cm) of high-density polyurethane was sprayed into the exposed rafters and covered with gypsum board. On the exterior walls, Pettit added a draining housewrap and two staggered layers of 2” (5 cm) foil-faced polyisocyanurate rigid insulation, covered by cedar siding. On the roof, the old shingles were removed and the same thickness of polyiso insulation was installed as on the walls, followed by plywood sheathing and roofing (without an air space). Pettit prevented moisture problems by installing a continuous air barrier to limit air leakage and a continuous drainage plane on the exterior wall.

For windows, Pettit chose double-glazed, argon-filled, low-emissivity (low-e) windows from Andersen. Pettit kept the existing trim on the interior, extending the exterior windowsills around the added insulation and protecting the assembly from moisture with metal flashing and housewrap. Expanding foam sealant prevents air leakage around the perimeter of the window. “We didn’t go to the very best window,” Pettit said, because the glazing accounted for only about 12% of the surface area of the house and she felt that its contribution to heat loss did not warrant the expense of a further performance upgrade. Although Pettit chose the Andersen windows because she could order them in custom sizes, she told EBN that she “wasn’t as thrilled with them as I should be for the price.” If approaching the same decision again, she said she would probably put the cost premium toward windows with better energy performance.

Overall, Pettit is happy with the results: the house performs well (the energy loads dropped by two-thirds: from 360 million Btu to 120 million Btu per year) and is comfortable. While Pettit originally planned to add solar systems to achieve net-zero-energy, she has not yet done so. Parsing out the costs of the project, Pettit estimates that the extra insulation cost about $10/ft2 ($100/m2) in materials and $10/ft2 ($100/m2) in labor, or 10%–15% of the total project cost. Because the house needed major renovation anyway and the energy improvements could be done as part of that work, Pettit felt that the cost of the added insulation was reasonable. She cautioned, however, that “for someone who doesn’t have to do any other work, it would be a different story.”

Case Study: The Raritan Inn, New Jersey

Bill Asdal of Asdal Builders in Chester, New Jersey, bought a 24-acre (10 ha) property in nearby Califon, knowing the two 100-year-old buildings (the inn and the cottage) on the property would need extensive renovations. Neither house had been occupied for at least ten years, and both had some structural damage. Since the renovations would be extensive anyway, moving to a net-zero-energy retrofit seemed feasible. The National Association of Home Builders Research Center expressed interest in one of the buildings, the cottage, for a case study for its Strategies for Energy Efficiency in Remodeling program. (Several manufacturers donated materials and expertise, but, according to Asdal, this did not affect the decisions he made about materials or technologies.)

After repairing some structural damage to the buildings’ foundations, Asdal began the work of insulating them. He tore down the interior wallboard and added blown cellulose to the walls. He stripped the existing siding before adding sheathing and housewrap and covering it with R-4 foam-backed insulated vinyl siding. This work brought the total insulation value of the wall assembly to R-22. Simply adding insulation does not go far enough, however: as Asdal points out, “the magic is not always in the R-value, but in the tightness of the house.” Therefore, Asdal’s crew used air-sealing techniques throughout the house, caulking the window openings, foaming around floor and wall penetrations, and gluing both wallboard and exterior sheathing to the framing.

Asdal added a room to the cottage, and went with structural insulated panel (SIP) construction. He worked with Techbuilt of Cleveland, Ohio, to design the addition using expanded polystyrene (EPS) foam panels with steel supports. Techbuilt created the panels and worked with Asdal and his crew to assemble the addition in under six hours. The 7” (20 cm) wall panels achieve an R-38 insulation value, and the 12” (30 cm) roof panels achieve R-50. Ground-source heat pumps were added to both houses, feeding off of a single horizontal piping loop installed on the property.

After thoroughly insulating the houses, the next step for Asdal was to introduce several renewable energy systems to achieve net-zero site-energy performance (see EBN Vol. 14, No. 10 for more on defining net-zero energy). In the cottage, a solar-thermal collector preheats domestic hot water that is then fed through a tankless water heater and delivered to the fixtures. In the inn, a desuperheater (an optional energy-saving feature on many heat pumps) extracts waste heat from the heat-pump cooling cycle to preheat the water. Asdal also installed a 14.2 kW PV system; the panels are divided among the cottage roof, a garage roof, and a barn roof. Because the onsite PV system generates more power than the houses need, Asdal received $232 back from the utility in 2006, in addition to funds from the utility that were provided to replace government rebates on the PV system (which New Jersey no longer offers).

According to Asdal, both buildings are “overperforming, in a number of ways.” The energy savings are greater than expected, he said, and public interest in the project continues to grow. He cautions, however, that such an extensive retrofit might not work for every project: “consumers and contractors should not touch renewables” until they’ve changed occupant behavior and performed an extensive weatherization job, he said. Despite the challenges of such extensive renovations, Asdal is hopeful about the potential for net-zero retrofits. “If we can go out and do a full gut rehab on a lot of these houses, we can avoid some of the barriers of new building,” he said. “I see this as the ultimate form of recycling.”

Case Study: 1926 Austin Bungalow

In February 2007, a crew from the television show This Old House began work on a 1926 bungalow in Austin, Texas. The homeowners wanted to expand the 1,500 ft2 (140 m2) house to make space for their growing family, and they wanted a green renovation. Architect David Webber, AIA, of Webber + Studio in Austin, Texas, designed a second-floor addition that added two bedrooms and only 6’ (2 m) to the roofline. With such major renovations happening, making improvements to the energy performance of the house was fairly easy.

Adding a second floor gave builder Bill Moore a chance to add 7.5” (20 cm) of closed-cell sprayed polyurethane insulation between the attic rafters. Moore decided to insulate the entire attic, including the unfinished portion that would house the air-conditioning equipment. He also added TechShield, a radiant barrier roof decking product from LP Building Products, to keep the attic cool. Although the addition offered the greatest opportunity for envelope improvements, Moore also added dry-pack cellulose insulation to the existing wall cavities on the first floor.

Adding large amounts of extra insulation to the house did not make sense for the climate, according to Moore. In the hot, humid climate of Austin, he said, “the biggest bang for the buck is to seal the air leaks, because that’s effective for heating and cooling.” Similarly, replacing all of the original single-pane windows with high-performance windows didn’t make sense financially. “There’s not a payback down here for high-performance windows” in renovation situations, Moore said, because the energy savings are not great enough to justify the cost of replacement. He admitted that “comfort is a different issue,” saying that insulating windows make a house more comfortable in the winter. He used double-glazed, argon-filled, low-e windows from Andersen in the second-floor addition.

Moore focused on other ways to lower the energy demand of the house, replacing the old furnace with a high-efficiency model from Trane and adding a gas-fired tankless water heater with an on-demand recirculator. The air-conditioning system was upgraded from a 10 SEER unit to a 16 SEER unit, and all of the ductwork was placed in conditioned space. The largest energy savings for cooling, however, came from occupant behavior: the homeowners wait as long as possible to turn on the air-conditioning, according to Moore, and rely on natural ventilation and shading to cool the house, making even their pre-renovation energy use low compared to other area homes.

According to Moore, a computer simulation showed that, despite increasing the square footage of the house 50%, the energy loads were cut by 10%. Rebates for a 2.45-kW PV system, which is expected to provide 40% of the home’s energy needs, made it financially feasible to offset the small gains in efficiency with onsite renewable generation.

Final Thoughts

Significantly reducing the energy consumption of existing homes is a critical priority if we are to stem greenhouse gas emissions and global warming. It is also very challenging. Achieving 50% energy savings in an existing home is not as easy as it might seem—but it can be done.

Checklist:Low-Energy Retrofits–Priority Checklist

The biggest challenges are cost and skilled labor. To address these, EBN’s editorial this month (see An Environmental Service Corps for America) presents the concept of an Environmental Service Corps to focus on home energy retrofits along with such other activities as ecological restoration and invasive species control. A program like this—as bold and unlikely as it seems—would provide only part of the answer to bring about widespread energy retrofitting. We will also need to encourage the use of energy-efficient mortgages and home loans, subsidized low-interest loans, outright grants, and performance-based tax incentives.

The costs of committing our nation to a cutting-edge program to retrofit a significant percent of our nation’s housing stock would be huge—hundreds of billions of dollars, if not trillions of dollars, over the coming decades. The cost would be on the scale of what we are spending on the Iraq War but arguably with more return on investment for the people footing the bill. The hope is that our political leaders would see the benefits that this investment would provide in the way of energy security, protection against rising energy costs, and environmental stewardship.

For more information:

Architecture 2030

Santa Fe, New Mexico


Asdal Builders

Chester, New Jersey


Building Science Consulting

Boston, Massachusetts


Wm. T. Moore Construction, Inc.

Austin, Texas






July 9, 2012 - 11:22 am

Alex and Allyson make some good points but over look the biggest opportunity to perform single-family deep energy retrofits (DERs) : re-sale. Quite simply, more than 4 million homes are sold each year (even during our current real estate depression) and each sale represents an almost once in a lifetime opportunity to upgrade the energy performance of the home. A&A are correct in stating the DERs are expensive and that affordable financing is key for a successful DER (or really deep pockets). But the best DER financing tool - the Energy Efficient Mortgage (EEM)- is a perfect match for the re-sale market. Granted there are problems w/ the EEM none more important than getting realtors on board. However, the efficiency community, especially utilities with active efficiency and/or home performance programs, have largely ignored the re-sale market and the EEM. In fact, I have only been able to find one example of a utility sponsed EEM program, a program sponsed by PG&E in 1998. These points were proven through DER R&D program I managed for SMUD. Six foreclosed, abandoned homes underwent DERs and costs were tracked. More detailed info can be found at the SMUD.org website under the residential/success stories tabs. In brief, costs were high but the homes required major rehabilitation work to make them market ready. AS a result, the incremental costs of making envelope improvemenets and adding high efficiency equipment were a fraction of the toal costs of the project. Another example of how a re-sale program would work can be found in the work of David Robinson flipping high performance homes in Fresno. The fact of the matter is that there is no better time to do the upgrades and finance them at re-sale. What's required to get the EEM off the ground is launch a multi-facted, coordinated utility EEM program that informs homebuyers and realtors of the benefits of the EEM while aligning HERS and home performance programs to support the EEM. I call this the "new" new construction program. I'm interested in knowing what others think and are doing.

September 12, 2007 - 9:52 am

You're right--the replacement rate is pretty tiny. We derived that number from the number of tear-downs in the country each year, and I was pretty surprised that it was so small. To answer your question, a house life of 250 years isn't entirely unheard of, at least in the Northeast, where there are some houses still standing from the 1700s. But it's not typical. My guess is that the amount of new construction and major renovations far outweighs the actual tear-down-and-replace rate right now. That might change, if demand for housing increases in places where there is no available land. Then we might see the tear-down rate increase.

August 31, 2007 - 10:00 am

The article quotes 0.2% as the turnover rate for existing residential buildings. This implies a 500 year time span for replacing them all, or say 250 average lifetime. Are you (and the EPA) really saying residences will last that long?

July 12, 2007 - 9:43 am

This is a good article on a difficult subject but I can't read the checklist!