The Committee on Energy-Efficiency and Sustainability Standards Used by the Department of Defense for Military Construction and Repair was tasked to conduct a literature review that synthesizes the state-of-the-knowledge about the costs and benefits, return on investment, and long-term payback of specified American Society of Heating, Refrigerating and Air-Conditioning Engineers (ASHRAE) design standards and green building certification systems. The committee identified numerous publications, ranging from studies with clearly outlined design objectives and methodologies and empirical information, to individual case studies and opinion editorials. In Chapters 1 and 3, the committee identified factors that made its task more complex. Additional factors became apparent as the committee reviewed the literature, as outlined below.
• Baselines and definitions. As noted in Chapter 3, baselines for measuring the energy and water use and operations and maintenance costs for buildings are limited, and it is difficult to quantify the benefits and costs of those factors. The equally important but more difficult to quantify effects, such as worker health, productivity, and well being, are typically treated qualitatively, although quantitative measures are sometimes developed for these factors. Typically information about indoor environmental quality (IEQ),1 which relates to health, well being, and productivity, is gathered through surveys of building occupants, introducing a level of subjectivity that is not present when resources are monitored through engineered systems.
There are no national baselines for measuring occupant satisfaction with indoor environmental quality or for measuring worker productivity related to building design. Standard survey forms to collect data from building users have been developed by the Center for the Built Environment (CBE) at the University of California, Berkeley, and the Usable Buildings Trust in the United Kingdom.2 Data gathered from the CBE surveys have been collected in a single database from which baselines can be developed for comparative studies. As of October 2009, the CBE database included 51,000 individual
1 Indoor environmental quality typically refers to factors such as temperature, humidity, ventilation, lighting, and noise.
responses from occupants of 475 buildings (CBE, 2012). As of 2011, the Usable Buildings database contained surveys of occupants of 500 buildings in 17 countries (Baird et al., 2012).
Studies on high-performance or green buildings use a wide range of definitions to describe the criteria/attributes of the buildings being evaluated. In some studies, green buildings are defined as Leadership in Energy and Environmental Design (LEED)-certified. In others, the green building sample may include a mix of LEED-certified buildings, LEED-registered buildings, buildings receiving industry awards, and buildings designed with energy efficiency as an objective. This variance in definitions, like the variance in baselines, makes it difficult to objectively compare the results of one study to another.
• Types of buildings and sizes. Each of the studies reviewed included a variety of building types in the sample sets for green buildings, ranging from office buildings to schools, hospitals, and laboratories to courthouses. Different building types and different building sizes incorporate different types of mechanical and other systems to meet differing needs in terms of hours of operation (24/7 or weekdays only), use, intensity of use, number of floors, and other factors. Generalizing findings across a mix of building types and sizes introduces another set of confounding factors that prevent an apples-to-apples comparison across studies.
Given those factors, the factors identified in Chapters 1 and 3, and a 6-month time frame to complete its work, the committee determined it would need to focus solely on the main purposes of the statement of task. For its evaluation of the research literature, the committee determined that it would rely on studies that met the following criteria:
• Time frame. The committee relied on studies published in 2004 or later because the first studies evaluating the incremental costs of LEED-certified buildings were published in 2004. The first evaluations of a sample of at least six high-performance or green buildings were published in 2006.
• Robustness. The committee focused on studies with clearly stated objectives, a clearly defined methodology, findings based on empirical data, and a sample size of at least six buildings. The committee relied more heavily on those studies that reported measured results for energy (utility bills) than on modeled or predicted results, because the committee believes that data from actual buildings will be more reflective of the type of results that DOD can expect from its high-performance buildings.
Because the number of green buildings is increasing each year, more recent studies can incorporate larger sample sizes from which to make comparisons. Larger sample sizes can help to eliminate some factors of bias, error, and chance that are prevalent in individual case studies, although such factors may still be present.
• Relevance to the DOD operating environment. The research literature on high-performance and green buildings includes a number of reports that analyze the market and price effects of LEED or ENERGY STAR®3-certified buildings (primarily office buildings) compared to conventional buildings in terms of rental rates, vacancy rates, turnover ratios, appraised value, and other factors (Miller et al., 2008; Chappell and Corps, 2009; Dermisi, 2009; Fuerst, 2009; Fuerst and McAllister, 2008; Fuerst et al., 2010; Conlan and Glavis, 2012; Eicholz et al., 2009, 2011). These studies are of value, particularly to the private sector and to federal agencies such as the General Services Administration (GSA), which secures commercial space for other agencies. However, because DOD primarily owns and operates its facilities for 30 years or longer, the committee did not analyze these studies in detail, because market-related factors such as rental premiums and appraised value are not directly relevant to the DOD operating
3 ENERGY STAR® is a labeling program for energy-efficient building-related products and equipment. It is not a green building certification system.
environment. The committee instead relied on studies that focused on energy and water use, indoor environmental quality, and other factors that DOD is required to address through the Energy Independence and Security Act of 2007 and other mandates.
The committee did not identify any studies that conducted a traditional benefit-cost analysis to determine the long-term net present value savings, return on investment, or long-term payback related to the use of ASHRAE standards 90.1-2010 and 189.1-2011 and the LEED or Green Globes green building certification systems. Only two studies (Turner, 2006; Kats, 2010) compared the performance of green buildings (defined differently) to conventional buildings (different baselines) and assigned some measure of net present value (NPV) to different categories of costs and benefits.
The committee also did not identify any studies that analyzed the performance of samples of six or more Green Globes-certified buildings; the only evaluations of the performance of Green Globes-certified buildings were individual case studies.
The data cited for the 25 studies that met the committee’s criteria are not universally unique; that is, some studies evaluated all or portions of the same data sets. For example, one of the most robust studies of green buildings conducted to date is Energy Performance of LEED for New Construction Buildings (Turner and Frankel, 2008). Turner and Frankel gave other researchers access to their data set for green buildings. Thus, studies published by Newsham et al. (2009) and by Scofield (2009a, 2009b) used the same data set but applied different analytical tests and arrived at different conclusions. In a different instance, Fowler and Rauch (2008) analyzed 12 green buildings owned and operated by the GSA. Fowler et al. (2010) reanalyzed the original 12 buildings, updated the available data, and also included 10 additional GSA green buildings in their analysis.
For the ease of the reader, the findings from the 25 studies are organized by specific topic area— energy use, water use, operations and maintenance costs, indoor environmental quality and productivity, and incremental costs to design and construct high-performance buildings. Where studies are cited more than once, the first reference includes some basic information about the sample size, definitions, methodology, and other factors. This information is not repeated if the study is cited multiple times. Table 4.1 contains summary information about the studies cited. They are arranged in the order that they first appear in Chapter 4. More detailed information about each of the studies is contained in Appendix D.
Sixteen studies focused solely or in part on the site energy use in high-performance or green buildings. They are organized below into three categories: studies of energy use in commercial buildings; studies of energy use in federal buildings; and regional studies of energy use.
The majority of the studies measured energy use intensity (EUI), typically calculated by taking the total energy consumed in 1 year (measured in kBtu) and divided by total building floor area to compare the performance of green to conventional buildings. Most studies measured site energy, although a few measured source energy. Measurement of source energy brings into play issues and policies related to the reduction of greenhouse gas emissions, which is beyond the scope of the committee’s statement of task. For that reason, the committee reports study results in terms of site energy.
|Authors and Study Title||Characteristics of High-Performance or Green Building Sample||Variables Measured||Methodology|
|Torcellini et al. (2006) Lessons Learned from Case Studies of Six High-Performance Buildings||Six high-performance buildings defined as designed to achieve aggressive energy goals; buildings constructed between 1996 and 2005; six building types; a range of locations||Net source energy used; net site energy used (both measured as energy use intensity (EUI); energy costs||Monitored six buildings intensively over a 4-year period; gathered at least 1 year of energy use and costs for each building; compared actual costs to baseline energy models for the buildings and to energy-code compliant baseline buildings|
|Diamond et al. (2006) Evaluating the Energy Performance of the First Generation of LEED-Certified Commercial Buildings||21 LEED-NC buildings certified between 2001 and 2005; 14 federal and 7 non-federal; 8 office, 4 laboratories, 1 library, 3 multifamily, 4 mixed use, 1 education||Baseline energy modeled; design energy modeled; actual energy use, all expressed as EUI; ENERGY STAR® scores (illustrative); LEED energy-efficiency-related points||Compared the actual site energy use of 18 of the 21 buildings, based on utility bills, to the baseline energy and design energy models submitted for the LEED certification process; also compared simulated whole building energy to actual billed energy|
|Turner and Frankel (2008) Energy Performance of LEED®for New Construction Buildings||121 LEED-NC-certified buildings; 100 buildings classified as “medium energy use activities” (office and similar); 21 buildings as “high energy use activities” (data centers, laboratories, and similar)||Site energy (EUI) actual, modeled for total sample and subsets of sample, including office (35 buildings) and LEED certification levels (Certified, Silver, Gold/Platinum; also collected data on occupant satisfaction||Compared the actual site energy use of 121 LEED-certified buildings to CBECS national averages, ENERGY STAR® ratings, and LEED baseline energy models; evaluated energy use of medium-energy-use buildings, high-energy-use buildings, 35 office buildings, and for buildings at different LEED certification levels|
|Newsham et al. (2009) Do LEED-certified buildings save energy? Yes, but . . .||100 LEED-certified buildings categorized as “medium energy use activities”; same data subset as Turner and Frankel (2008)||Site energy (EUI) actual, modeled; site energy use for buildings certified as LEED-certified, LEED-Silver, and LEED-Gold/ Platinum||Reanalyzed a data set from Turner and Frankel (2008), applying t-tests and other statistical measures to provide more rigor; individually matched LEED-certified buildings in data set to similar non-LEED-certified buildings in the CBECS database|
|Scofield (2009a) A Re-Examination of the NBI LEED Building Energy Consumption Study||100 LEED-certified buildings categorized as “medium energy use activities”; same data subset as Turner and Frankel (2008) and Newsham et al. (2009)||Site energy use; source energy use; energy use by LEED certification level||Defined mean EUI differently than two other studies; measured source energy as well as site energy and conducted statistical tests|
|Authors and Study Title||Characteristics of High-Performance or Green Building Sample||Variables Measured||Methodology|
|Scofield (2009b) Do LEED-Certified Buildings Save Energy? Not Really||35 LEED-certified office buildings; same subset of data used by Turner and Frankel (2008)||Site energy; source energy||Weighted EUI of each building by its gross square feet; used different averaging methods than other studies|
|Kats (2010) Greening Our Built World: Costs, Benefits, and Strategies||170 green buildings of a wide range of types, located in 33 states and 8 countries; green buildings defined as LEED-certified, anticipating LEED certification or certified under another similar system (none certified under Green Globes)||Incremental costs of green design and construction; energy use and costs; water use and costs; data reported for all buildings in sample and by LEED-certification level||Conducted benefit-cost analysis and payback analyses for energy use and water use of green versus conventional buildings; data for green buildings primarily based on models, not actual measured data|
|Fowler and Rauch (2008) Assessing Green Building Performance: A Post Occupancy Evaluation of 12 GSA Buildings||12 General Services Administration (GSA) buildings designed to be LEED-certified or otherwise designated green; 6 office, 4 courthouses, 2 combination office/courthouse||Site energy use, water use, operating costs, occupant satisfaction||Measured energy use based on utility bills and compared to CBECS national and regional averages and GSA baselines; measured water use based on utility bills and compared to a derived baseline for domestic water use; compared operating costs to industry sources; distributed CBE survey to measure occupant satisfaction|
|Fowler et al. (2010) Re-Assessing Green Building Performance: A Post Occupancy Evaluation of 22 GSA Buildings||Updated data for 12 GSA green buildings studies by Fowler and Rauch (2008); expanded data set to include 10 additional GSA green buildings; total sample included 8 courthouses, 12 office buildings, and 2 mixed office/courthouse||Same measures as Fowler and Rauch (2008)||Same methodology as Fowler and Rauch (2008); also provided analyses of a subset of 15 LEED-certified buildings by certification level (Certified, Silver, Gold/Platinum)|
|Menassa et al. (2012) Energy Consumption Evaluation of U.S. Navy LEED-Certified Buildings||11 Naval Facilities Engineering Command (NAVFAC) buildings LEED-certified by 2008, included 3 LEED-certified, 5 LEED-Silver, 3 LEED-Gold buildings; 1 drill hall, 3 maintenance facilities, 1 laboratory, 1 child care center, 2 barracks, 1 golf course clubhouse, 2 administration buildings||Site energy use for 11 buildings; water use for 9 buildings (2 LEED-certified, 4 LEED-Silver, 3 LEED-Gold)||Compared measured site and water use for the LEED-certified buildings to measured energy and water use for 11 similar NAVFAC buildings that were not LEED-certified|
|Authors and Study Title||Characteristics of High-Performance or Green Building Sample||Variables Measured||Methodology|
|Turner (2006) LEED Building Performance in the Cascadia Region: A Post Occupancy Evaluation Report||11 LEED-certified buildings in the Pacific Northwest; sample included 7 offices or libraries and 4 multi-family buildings; 3 LEED-NC-certified, 4 LEED-NC-Silver, 3 LEED-NC-Gold, 1 LEED-EB-Gold||Site energy use; indoor water use; NPV benefits for energy and water; occupant satisfaction||Compared actual energy use (utility bills) and water use to three baselines: initial model projections, baseline approximate to code, and ENERGY STAR® median; NPV calculations assumed a 25-year time period, discount rate of 3 percent, and utility rate increases equal to rate of inflation|
|Baylon and Storm (2008) Comparison of Commercial LEED Buildings and Non-LEED Buildings within the 2002-2004 Pacific Northwest Commercial Building Stock||24 LEED-certified buildings constructed between 2002 and 2005 in the Pacific Northwest. 8 different building types; most buildings had been occupied at least 2 years||Site energy use (EUI)||Compared the characteristics of the LEED-certified buildings to a larger sample of contemporary buildings built to local standard codes; characteristics studied included lighting, HVAC systems, building envelope, glazing, and control systems|
|Sacari et al. (2007) Green Buildings in Massachusetts: Comparison Between Actual and Predicted Energy Performance||19 new or renovated green buildings in Massachusetts, including 12 green schools and 6 other buildings that were LEED-certified||Site energy use||Compared actual site energy use in the green buildings to the energy use predicted by design models and to energy use in buildings constructed to Massachusetts code|
|Widener (2009) Regional Green Building Case Study Project: A Post-Occupancy Study of LEED Projects in Illinois||25 LEED-certified projects in Illinois including projects certified under a variety of LEED programs; at least 6 different building types; most certified under LEED versions 2.0 or 2.1||Site energy use (EUI); greenhouse gas emissions; water use (indoor and outdoor); commute transportation; construction and operating costs; green premium, health and other benefits; occupant comfort||Compared data for the LEED-certified projects to three other data sets/baselines: Turner and Frankel (2008), CBECs national averages, and ENERGY STAR®; single data element that was mandatory for inclusion in the sample was post-occupancy measured energy use|
|Authors and Study Title||Characteristics of High-Performance or Green Building Sample||Variables Measured||Methodology|
|Oates and Sullivan (2012) Postoccupancy Energy Consumption Survey of Arizona’s LEED New Construction Population||25 LEED-NC-certified buildings in Arizona; 7 building types certified under LEED versions 2.0, 2.1, and 2.2; all had been in operation at least 1 year as of October 2009; sample broken into 19 buildings with medium energy intensity (offices and similar) and 6 buildings of high energy intensity (laboratories)||Site energy (EUI); source energy (EUI)||Actual energy performance of the LEED-certified buildings was compared to national averages from the CBECs database; CBECS data normalized to match the gross square feet weights for each building type in the LEED sample|
|Leonardo Academy (2008) The Economics of LEED for Existing Buildings for Individual Buildings||11 to 13 buildings certified under LEED-EB program as of 2007||LEED-EB certification costs; operating costs||Gathered data from building owners on costs to certify buildings under the LEED-EB program (13 buildings); collected data on operating costs for 11 buildings and compared them to industry sources|
|Abbaszadeh et al. (2006) Occupant Satisfaction with Indoor Environmental Quality in Green Buildings||21 green office buildings of which 15 were LEED-certified and 6 had received green or energy efficiency awards||Overall occupant satisfaction; thermal comfort, air quality, lighting, and acoustics/ noise||Surveyed occupants of green buildings directly using questionnaire developed by CBE; compared results to remaining buildings in CBE database (conventional)|
|Miller et al. (2009) Green Buildings and Productivity||154 buildings that were LEED-certified or had an ENERGY STAR® label; located across the country||Productivity measured as sick days and self-reported productivity percentage after moving into a green building||Conducted a survey of more than 2,000 tenants in 154 buildings; also calculated the economic impacts of those tenants who claimed an increase in productivity (report summarized a literature review as well)|
|Baird et al. (2012) A Comparison of the Performance of Sustainable Buildings with Conventional Buildings from the Point of View of the Users||31 sustainably designed commercial or institutional buildings located in 11 countries; occupied by 15 to 350 staff; 15 office, 10 education, 4 laboratories, 2 mixed use||Occupant satisfaction overall; occupant satisfaction with temperature, lighting, and acoustics/noise||Distributed a questionnaire developed for the Buildings In Use (BIU) studies to 2,035 tenants in 31 green buildings; compared results to data for occupants in 109 conventionally designed buildings from the BIU database that had been surveyed during a similar time period|
|Authors and Study Title||Characteristics of High-Performance or Green Building Sample||Variables Measured||Methodology|
|Matthiessen and Morris (2004) Costing Green: A Comprehensive Cost Database and Budgeting Methodology||45 LEED-seeking buildings from the database of the Davis Langdon Company||Incremental construction costs of green buildings||Compared the construction costs of 45 LEED-seeking buildings to the construction costs of 93 non-LEED-seeking buildings; all costs were normalized for time and location to ensure consistency for the comparisons|
|Matthiessen and Morris (2007) Cost of Green Revisited: Reexamining the Feasibility and Cost Impact of Sustainable Design in Light of Increased Market Adoption||83 buildings seeking LEED certification under versions 2.1 and 2.2; building types included academic classrooms, laboratories, libraries, community centers, and ambulatory care facilities||Incremental construction costs of green buildings||Compared the construction costs of 83 LEED-seeking buildings to the construction costs of 138 non-LEED-seeking buildings; all costs were normalized for time and location to ensure consistency for the comparisons|
|Steven Winter Associates (2004) GSA LEED Cost Study||Study undertaken to estimate the costs to develop green federal buildings using LEED 2.1; examined a 5-story courthouse and a mid-rise federal office building||Incremental construction costs for federal courthouses and office buildings||Individual LEED credit assessments and cost estimates were completed for six different scenarios to create a cost range for LEED-certified, LEED-Silver, and LEED-Gold levels|
|Indian Health Service (IHS) (2006) LEED Cost Evaluation Study||Study undertaken to evaluate potential cost impacts of achieving LEED-NC and LEED-NC-Silver certification on IHS facilities||Incremental construction costs for hospitals and other healthcare-related buildings||Evaluated initial capital cost investments and life-cycle costs (20-year period); LEED credits were evaluated against standard practices of the IHS as outlined in the IHS design guide|
|Caprio and Soulek (2011) MILCON Energy Efficiency and Sustainability Study of Five Types of Army Building||Five standard building types most commonly constructed by the U.S. Army: barracks, tactical equipment and maintenance facility, government office and other public assembly, brigade headquarters, and dining facility||Incremental construction costs; total energy use (modeled)||Study undertaken to identify incremental construction costs for building energy efficiency enhancements intended to meet federal mandates|
NOTE: HVAC = heating, ventilation, and air-conditioning.
Studies of Energy Use in Commercial Buildings
Torcellini et al. (2006) conducted field evaluations over a 4-year period for six high-performance buildings of different types and in different geographic areas. High-performance buildings were defined as those that were designed to meet energy-savings goals ranging from 40 percent better than energy-code-compliant buildings to net-zero-energy buildings. All used innovative technologies and a whole building design process to look at the interrelationships of each building’s technologies, materials, and design. The researchers compared source and site energy performance (at least 1 year of measured performance) and the energy costs of each of the buildings to energy-code-compliant base-case buildings. They found that the six high-performance buildings used between 25 percent and 79 percent less site energy than the baseline buildings. Site energy costs were 12 to 67 percent lower than the energy costs for the baseline buildings. The variability in energy cost savings was attributed to differences in utility rate structures, fuel types, and peak demand profiles, among other factors.
Diamond et al. (2006) measured the actual energy use of 21 LEED-certified buildings (utility bills for the first year of operation) against the energy use predicted by the energy-use baseline and design models submitted for the same buildings for LEED certification. For the 18 buildings in the sample for which the researchers had both simulated whole building energy use and actual purchased energy data, the actual energy use was 28 percent lower than for the baseline model. However, there was significant variation among individual buildings, with some being more energy efficient than predicted and some being less efficient. For a subset of nine federal buildings, the actual energy use was lower than the modeled use.
Turner and Frankel (2008) reviewed the post-occupancy energy performance of 121 LEED-New Construction (LEED-NC)-certified buildings, of which 100 were classified as “medium energy use activities” and defined as buildings that had EUIs in a range similar to office buildings. (Total EUI was derived by summing the purchased energy for all fuel types.) Twenty-one buildings were classified as “high-energy use activities,” which included buildings with very high process loads, such as laboratories, data centers, and recreation facilities. Most of the analyses in the report focused on the 100 medium-energy-use buildings. Within the 100-building sample, at least eight building types were included, and office was the predominant use. Thirty-eight of those buildings were certified as LEED-NC-certified; 35 as LEED-NC-Silver; and 27 as LEED-NC-Gold or -Platinum.
The report compared measured energy use (1 full year of post-occupancy energy use) to several different benchmarks, including CBECS national averages, ENERGY STAR® ratings, and modeled energy performance predictions provided as part of the submittals for LEED certification. They found that for all 121 LEED-certified buildings, the median measured site EUI was 24 percent lower than the CBECS national average (as of 2003) for all commercial building stock. For 35 office buildings in the LEED-certified sample, the average energy use was 33 percent lower than the CBECS national average for office buildings. The authors found that project types classified as high-energy-use activities with high process loads, such as laboratories, were problematic, because the energy use of high-energy-use building types is not well understood by designers.
Within the sample of 100 medium-energy-use activities, Turner and Frankel found that LEED-NC-certified buildings used 26 percent less site energy than the CBECS national average, LEED-NC-Silver buildings used 32 percent less energy, and LEED-NC-Gold/Platinum-certified buildings used 44 percent less energy on average than the CBECS national average. The authors also compared the actual energy use in the LEED-certified buildings to the energy use predicted by baseline and design models submitted for the buildings as part of the LEED certification process. In this instance, measured energy use for the buildings was 28 percent less on average than the energy baseline models (most used ASHRAE Standard 90.1-1999) and 25 percent less on average than the levels predicted by the design models. However, the
energy use for more than half of the projects deviated by more than 25 percent from design projections, with 30 percent significantly better and 25 percent significantly worse.
For all but the warm-to-hot zones, LEED-NC buildings used significantly less energy than the CBECS national average, with median LEED EUIs 36 to 49 percent lower than the CBECS average for those zones. For the warm-to-hot zones, the median LEED EUI was virtually the same as CBECS. The authors stated that “the current variability between predicted and measured performance has significant implications for the accuracy of prospective life-cycle cost valuations for any given building” (Turner and Frankel, 2008, p. 5).
Newsham et al. (2009) re-analyzed the data used in Turner and Frankel (2008) for the 100 LEED-NC buildings categorized as “medium energy use activities.” They employed a range of statistical tests to improve the rigor of the analysis. In the tests, Newsham et al. sought to pair each LEED building with a single matched building from the CBECS database. Newsham et al. noted that a limitation of the Turner and Frankel study was that the comparisons to the CBECS data were somewhat crude:
The median EUI of all LEED buildings was compared to the mean EUI of all CBECS buildings, by activity type, thus confounding two different metrics of central tendency. Little specific account was made of differences in the two datasets related to climate zone, building size, or building age (Newsham et al., 2009, p. 5).
Nonetheless, Newsham et al. found that:
- No matter the basis of comparison, the LEED-certified buildings used statistically significant less energy per floor area than the CBECS averages. On average, the LEED-certified buildings used 18 to 39 percent less energy per floor area.
- Twenty-eight to 35 percent of LEED-certified buildings used more energy per floor area than their individually matched buildings from the CBECS database.
- There was no statistically significant relationship between LEED-NC certification level and energy use intensity or percent energy saved versus the baseline. LEED-NC-Silver buildings did not exhibit better energy performance than LEED-NC-certified buildings and LEED-NC-Gold/Platinum buildings did not exhibit better energy performance than LEED-NC-Silver buildings. This finding was the opposite of the finding from the Turner and Frankel (2008) study.
Scofield published two separate papers that reanalyzed subsets of the Turner and Frankel data (Scofield, 2009a, b). In both cases, Scofield’s major focus was source energy, although he also analyzed site energy. Turner and Frankel (2008) and Newsham et al. (2009) used site energy only.
In the report A Re-examination of the NBI LEED Building Energy Consumption Study (Scofield, 2009a), Scofield pointed out Turner and Frankel’s comparison of the mean of one distribution to the median of another and stated that “to compare the mean of one with the median of the other introduces bias by compensating for skew in only one distribution” (Scofield, 2009a, p. 765). Scofield also defined mean energy intensity differently, using a gross square foot averaging method, and conducted statistical tests of the data for several subsets of the Turner and Frankel database. Scofield compared data from some of the LEED-certified buildings to the CBECS database and also to a subset of buildings from CBECS constructed between 2000 and 2003. His conclusions included the following:
- LEED-certified medium-energy-use buildings, on average, used 10 percent less site energy but no less source (or primary) energy than did comparable conventional buildings, whether restricted to new vintage (constructed between 2000 and 2003) or not.
- LEED-NC-certified buildings used slightly more site energy than the CBECS comparison group, while LEED-Silver and LEED-Gold or -Platinum buildings used 23 percent and 31 percent less site energy, respectively, than the CBECS comparison group.
- LEED office buildings used 17 percent less site energy than that of the CBECS comparison group of all vintages; there was no significant reduction in primary (source) energy use relative to non-LEED office buildings.
The paper “Do LEED-certified buildings save energy? Not really…” (Scofield, 2009b) was written as a direct rebuttal to Newsham et al. Scofield reanalyzed data from Turner and Frankel (2008) for a subset of 35 LEED-certified office buildings. Scofield weighted the energy intensity of each building in the LEED sample by its gross square footage, which he stated was exactly equal to the total energy use by all buildings divided by their total gross square feet. In doing so, Scofield pointed out that different averaging methods would yield different means and different conclusions. Nonetheless, Scofield found that:
- LEED-NC-certified office buildings used, on average, 10 to 17 percent less site energy than comparable non-LEED buildings.
- LEED-certified commercial buildings, on average, show no significant primary energy savings over comparable non-LEED buildings.
- Smaller LEED office buildings had relatively lower purchased EUI (relative to non-LEED), while larger LEED office buildings showed less savings in comparison to non-LEED buildings.
Kats (2010) analyzed data for 170 green buildings representing of a wide range of building types located in 33 states and 8 countries. The primary emphasis of this study was on the financial benefits and costs of green buildings in comparison to conventional buildings. Data related to the incremental costs of green construction, energy use and water use, and other measures were gathered directly from building owners, architects, and developers. The results of the survey were synthesized with the findings from other studies to develop estimates of the NPV of benefits and costs. Other studies used in the synthesis included surveys, case studies, and market research.
The buildings in the sample were completed between 1998 and 2009. Green buildings were defined as those that were LEED-certified or anticipating LEED certification or certification under another similar rating system. Approximately 15 percent of the 170 buildings were certified under systems such as the Massachusetts green schools guidelines, Enterprise Green Communities, or the Green Guide for Healthcare Facilities.
Reported reductions in energy use for the green buildings were measured as EUI and largely based on computer design and baseline models submitted as part of the LEED certification process, not on actual measured energy use (utility bills) for the buildings. Kats (2010) reported that the buildings in the data set had projected reductions in energy use, from less than 10 percent to more than 100 percent (meaning that the building generated more power than it used), with a median reduction of 34 percent. However, Kats also noted that even within a single building type and region, green and conventional buildings showed a wide range of energy intensities depending on factors such as building design, mechanical systems and appliances, operations and maintenance practices, and occupancy.
For the benefit-cost analyses to calculate NPV benefits, Kats used a time period of 20 years, a discount rate of 7 percent, and assumed annual inflation rates of 2 percent, and used the median savings of 34 percent for the green buildings comparison. Kats calculated that the NPV of 20 years of energy savings in a typical green building ranged from $4 per square foot to $16 per square foot, depending on
building type and LEED level of certification. Kats found that “when compared with an ASHRAE 90.1 baseline building, LEED-certified buildings in the data set reported median savings of 23 percent; for Silver, the figure was 31 percent; for Gold, 40 percent; and for Platinum, 50 percent” (Kats, 2010, p. 16).
Studies of Energy Use in Federal Buildings
Fowler and Rauch (2008) looked at 12 green buildings owned by GSA located in half of its national regions. The sample included seven LEED-certified buildings, one LEED-registered building, one building constructed to meet the Living Building Challenge, and three buildings designed to achieve energy efficiency. The building sample included six office buildings, four courthouses, and two combination office/courthouse buildings. Fowler and Rauch measured the actual energy use of these buildings based on utility bills. They found that on average the 12 GSA green buildings used 29 percent less energy than the CBECS national average, 29 percent less energy than the CBECS regional average, and 14 percent less energy than the GSA energy goal for its portfolio of facilities.
In 2010, Fowler et al. studied 22 green buildings in the GSA’s portfolio. The sample included updated data from the 12 buildings included in the 2008 study and 10 additional GSA LEED-certified buildings. In all, the study included 8 courthouses, 12 federal buildings (office space), and 2 courthouse/federal buildings. Thirteen of the buildings were LEED-certified, three were LEED-registered (one of these buildings did not specify the proposed level of certification), while the others emphasized energy efficiency during the design phase. The methodology used was generally the same. Fowler et al. found that energy use in the 22 GSA green buildings, on average, was 25 percent lower than the CBECS national average, 18 percent lower than CBECS regional averages, and 10 percent lower than GSA regional averages for fiscal year (FY) 2009.
Data were available for 15 LEED-certified buildings. For five of the seven LEED-Silver buildings, energy use was lower for all three baselines (CBECS regional, GSA target, GSA regional). The energy use in two LEED-Silver buildings was higher than the CBECS regional average. The LEED-Gold buildings used consistently less energy than the baseline for all buildings.
Menassa et al. (2012) looked at the energy use of 11 buildings operated by the Naval Facilities Engineering Command (NAVFAC) that had achieved various levels of LEED certification (three Certified, five Silver, three Gold) by 2008. The study compared the site energy of the LEED-certified buildings to 11 NAVFAC buildings of similar size, function, and location that were not LEED-certified. Menassa et al. found that 7 of 11 LEED-certified buildings reduced their electricity use when compared to their non-LEED-certified counterparts, with reductions ranging from 3 to 60 percent less electricity. However, 4 of the 11 NAVFAC LEED-certified buildings used more energy than their non-LEED counterparts, ranging from 11 to 200 percent more energy. Four of five LEED-Silver buildings used 3 to 49 percent less energy than their non-LEED counterparts, while one LEED-Silver building used 128 percent more energy than its non-LEED counterpart. Two of the three LEED-Gold-certified buildings used 6 percent and 15 percent less energy than their non-LEED counterparts, while the third used twice as much energy as its non-LEED counterpart. Only 3 of the 11 NAVFAC LEED-certified buildings used less energy than the CBECS national average.
Regional Studies of Energy Use
Turner (2006) looked at measured energy usage (at least 1 year of utility bills) of 11 LEED-certified buildings (three building types) in relation to initial modeling predictions and to a baseline approximate to code in the Pacific Northwest. Energy was measured as per conditioned square feet, and savings esti-
mates were made by comparing actual energy to the energy use predicted by models. Turner found that all of the buildings used less energy than the baseline approximate to code, averaging nearly 40 percent below that baseline. Nine of the eleven buildings achieved energy savings when compared to a baseline similar building in the region. The author calculated NPV benefits for energy assuming a 25-year time period, a discount rate of 3 percent, constant use of energy, and energy price increases at the rate of inflation. Based on those parameters, Turner estimated that the cost savings per year for energy for the LEED-certified buildings would range from $0 to $26 per square foot, with an average savings of $2 per square foot when compared to the regional median.
In the Turner study, four LEED-NC-Silver buildings used 39 to 57 percent less energy than their approximate to code baseline model. The two LEED-NC-Gold buildings for which data were available used 43 to 86 percent less energy than the baseline approximate to code. For the four LEED-NC-Silver buildings, Turner estimated that the long-term cost savings would be $7 to $26 per square foot; for the three LEED-Gold buildings the savings would range from $0 to $8 per square foot.
Baylon and Storm (2008) compared the actual site energy performance of 24 LEED-certified buildings in the Pacific Northwest to the actual site energy performance of a larger sample of contemporary buildings constructed to local codes. Most of the buildings in the study had been occupied for at least 2 years. The LEED buildings in the sample saved 12 percent more energy than the comparison group. The authors noted that energy codes in Washington and Oregon were more stringent than ASHRAE 90.1-1999, which was the basis for LEED at that time.
Sacari et al. (2007) compared the predicted energy use (estimated during the preconstruction, design phase) to the actual energy use (utility bills for electricity and natural gas) in 19 new or renovated green buildings in Massachusetts compared to buildings designed to the Massachusetts baseline building code. The sample included 12 schools and 7 other buildings. Sacari et al. found that most of the green buildings were consuming less energy than a building designed to Massachusetts baseline code, although they were also consuming 40 percent more energy on average than predicted by design models.
Widener (2009) analyzed the post-occupancy performance and costs and benefits of 25 LEED-certified projects in Illinois. Most projects were certified under LEED versions 2.0 and 2.1. The sample included more than six building types certified under different LEED programs (e.g., LEED-NC, LEED-CI) and at all LEED certification levels. All projects provided at least 1 year of post-occupancy energy use; 17 of the 25 projects provided “whole project energy use data,” where complete energy data were provided. The performance of all the LEED-certified buildings was compared to three other data sets: the Turner and Frankel study published in 2008; the 2003 CBECS; and ENERGY STAR®. Widener found that the 17 LEED-certified projects for which complete energy data were available used 5 percent less energy than the CBECS comparison group. Widener also noted that there was a large variation in the energy performance among projects.
Oates and Sullivan (2012) conducted post-occupancy energy consumption surveys for 25 LEED-NC buildings in Arizona. The sample included various types of buildings that had been certified under LEED versions 2.0, 2.1, and 2.2 and that had been in operation for at least 1 year as of October 2009. Actual energy performance of those buildings as measured by EUI for source and site energy was compared to CBECS data. The CBECS data were normalized to match the gross square feet weights for each building type in the LEED sample. The LEED building sample was also characterized by medium energy intensity (19) and high energy intensity (6) structures. The authors noted that two buildings accounted for 40 percent of the total data set’s gross square footage and 51 percent of the gross square footage in the medium energy intensity subset.
The authors found that the 19 medium-energy-intensity LEED-certified buildings used 13 percent less energy than the CBECS comparison group. (The high-energy-intensity subset was not analyzed,
because the sample size was too small.) Of the 19 buildings (both medium- and high-energy-intensity use) with design and baseline model simulations, only one used less energy than had been predicted in the design case, and only four used less energy than the baseline simulation.
Six of the studies cited under energy use also studied water use. No studies were identified that focused only on water use in high-performance or green buildings.
Kats (2010) looked at 170 green buildings across the country. Of these, 119 reported projected reductions (from models) in indoor potable water use when compared to conventional buildings. The reductions ranged from 0 percent to more than 80 percent, with a median of 39 percent. Kats also found that water savings generally increased with LEED level of certification. Kats estimated the NPV benefits of water savings in typical green buildings ranged from $.50 per square foot to $2 per square foot, depending on building type and LEED level of certification.
Fowler and Rauch (2008) measured water use for 12 GSA green buildings. They established a baseline for domestic water use as the base load revealed from monthly water use data. Given these estimates, the average water use for the GSA green buildings was 3 percent less than the baseline.
Fowler et al. (2010) measured water use for 22 GSA green buildings and found that two-thirds of the buildings used less water than the GSA baseline, with the average being 11 percent lower. Of the 6 buildings with higher water use than the baseline, 5 had cooling towers or evaporative cooling, 2 had exterior fountains in a hot, dry climate, and 3 had non-typical operating schedules. For 5 of the 7 LEED-Silver buildings, water use was below the national and regional averages and the GSA baseline. Two LEED-Silver buildings (one with a cooling tower and one with evaporative cooling) had significantly higher water use than the average. Two of the 3 LEED-Gold buildings performed better than the baselines, but one used significantly more water than the baselines in both the 2008 and 2010 studies.
Menassa et al. (2012) found that 7 of 9 LEED-certified buildings used by NAVFAC reduced their water consumption by more than 15 percent when compared to NAVFAC non-LEED-certified similar buildings. Four of the LEED-certified buildings reduced their water use by 50 to 75 percent. Seven of 9 LEED-certified buildings reduced their water consumption between 18 and 72 percent. For the 4 LEED-Silver buildings for which water data were available, water use was 18 to 61 percent lower than their non-LEED counterparts. Two of the 3 LEED-Gold-certified buildings showed water savings of 56 and 60 percent, while the third used 90 percent more than its non-LEED counterpart.
Turner (2006) compared actual water use to modeled water use and to baseline code buildings in the Pacific Northwest. When compared to the baseline code buildings, 4 of the 7 buildings were using 8 percent less water. For the 7 buildings for which water use projections (models) were available, 6 buildings used at least slightly more water than projected.
Widener (2009) collected data on water use for 12 LEED-certified projects in Illinois. Widener found a wide range in annual water use and attributed it to individual project size, principal activity, and occupancy.
The committee identified three studies that attempted to compare operations and maintenance costs for high-performance or green buildings to other baselines.4
4 A study by Miller et al. (2010) looked at operations and maintenance costs for ENERGY STAR® buildings and was not reviewed by the committee because the Energy Star labeling program was not included as part of the statement of task.
A 2008 study by the Leonardo Academy measured operating costs for 11 buildings certified under the LEED-Existing Buildings (EB) program, each of which had a significant component of office space (Leonardo Academy, 2008). Operating costs included cleaning expenses, repair and maintenance expenses, roads/grounds expenses, security expenses, and administrative and utility expenses. Data for the LEED-EB-certified buildings were collected and compared to the operating costs in BOMA’s (Building Owners and Managers Association) Experience Exchange Report, an industry standard. The authors found that “in all categories of operating costs, more than 50% of the LEED-EB buildings have expenses less than the BOMA average for the region. Total expenses per square foot of the LEED-EB buildings are less than the BOMA average for 7 of the 11 buildings” (p. 21).
Fowler and Rauch (2008) calculated aggregate operating costs for 12 GSA green buildings and compared those costs to industry baselines. The baselines were developed from a number of sources, including data from BOMA and the International Facility Management Association (IFMA). Aggregate operating costs included water and energy utilities, general maintenance, grounds maintenance, waste and recycling, and janitorial costs. They found that, on average, aggregate operating costs were 13 percent lower than average costs than the industry baselines. However, several of the buildings had consistently higher operating costs in each category.
Fowler et al. (2010) analyzed operating costs for 22 GSA green buildings using the same definition of operating cost as Fowler and Rauch (2008). Fowler et al. found that, on average, aggregate operating costs were 19 percent lower for the green buildings than the baseline. Aggregate operating costs for 17 of the buildings were 2 to 53 percent lower than the industry baselines. Five of the 22 buildings had higher aggregate operating costs than the baselines, ranging from 1 to 27 percent higher.
The committee identified five studies that met its criteria for time frame, robustness, and relevancy, and that compared IEQ and the health and productivity of workers in high-performance or green buildings to that of workers in conventional buildings.5 It should be noted that a body of well-designed, empirical studies evaluating various factors related to IEQ in all buildings is available. However, in keeping with its narrow focus on the statement of task, the committee evaluated only studies specifically related to IEQ and high-performance or green buildings.
Abbaszadeh et al. (2006) looked at the satisfaction of occupants in green buildings compared to the satisfaction of occupants in conventional buildings, using information from the CBE database. They compared surveys from occupants in 21 green buildings (15 were LEED-certified and 6 additional buildings were reported as green, based on the receipt of national or local green building or energy efficiency awards) to CBE surveys from occupants in conventional buildings.
The study focused on occupant satisfaction with thermal comfort, air quality, lighting, and acoustics. The authors noted that “self-reported productivity scores follow the same pattern as those of satisfaction—productivity scores are high where satisfaction is high and low where satisfaction scores are low” (Abbaszadeh et al., 2006, p. 366). Other findings included the following:
- On average, occupants in LEED-certified green buildings were more satisfied than occupants of conventional buildings when it came to thermal comfort, air quality, and overall satisfaction with workspace and building.
- The mean satisfaction score in LEED-rated/green buildings was significantly higher than that for conventional buildings (1.47 versus 0.93).
5 Three additional studies, Birkenfeld et al. (2011), Singh et al. (2009), and Cook (2005), analyzed only one or two buildings each, and for that reason were not included in the review of studies by the committee.
- Occupants in LEED-rated/green buildings were more satisfied with thermal comfort compared to occupants in conventional buildings (0.36 versus −0.16) and more satisfied with air quality in their workspace (1.14 versus 0.21).
- Even when considering only conventional buildings that were less than 15 years old, the mean satisfaction score with air quality was significantly higher for LEED-rated/green buildings (1.14 versus 0.52).
- When including only buildings 15 years old or newer in the conventional category, no statistically significant relationship was found for the IEQ categories of lighting and acoustics.
Fowler and Rauch (2008) used the CBE questionnaire to survey the occupants of 12 GSA green buildings. All of the green buildings scored above the CBE median for general occupant satisfaction, with the average being 22 percent higher than the CBE median.
Fowler et al. (2010) assessed 22 GSA green buildings and also used the CBE questionnaire. They found that, on average, occupant satisfaction with the green buildings in general was 27 percent higher than the CBE baseline, except for lighting, where it was the same as the baseline.
Miller et al. (2009) conducted a survey of 154 buildings that were deemed green by virtue of either an ENERGY STAR® label or LEED certification (any level) to determine if green buildings provided more productive environments. They gathered data for sick days and self-reported productivity percentages from building occupants who had moved to a new green building. Some 534 tenant responses were collected from buildings located across the United States. They found that 55 percent of the respondents agreed or strongly agreed that employees in green buildings were more productive, while 45 percent suggested no change. They also found that 45 percent of the respondents agreed that workers were taking fewer sick days than before moving to a green building, while 45 percent found it was the same as before, and 10 percent reported more sick days (the 10 percent were all in ENERGY STAR®-labeled buildings).
Baird et al. (2012) sought to determine whether there were any significant differences in the users’ perceptions of a range of factors concerned with the operation, environmental conditions, control, and degree of satisfaction between sustainable and conventionally designed buildings.
The set of sustainably designed buildings (defined as either recipients of national awards for sustainable design or highly rated in terms of their country’s buildings sustainability rating tool(s) or had pioneered some aspect of green architecture) included 31 commercial and institutional buildings (at least six different building types) located in 11 different countries. Surveys were gathered from 2,035 occupants. The survey questionnaire and baselines for comparison were from the Buildings in Use (BIU) database. The comparison sample of 109 conventionally designed buildings was compiled from the BIU database and included buildings that had been surveyed during a similar time period as the sustainable buildings were surveyed. Baird et al. (2012) found the following:
- An overall improvement in temperature and air quality in sustainably designed buildings was statistically significant. The sustainable buildings were perceived to be colder on average in winter but much the same (still on the hot side) in summer, whereas their air was perceived to be both fresher and less smelly year round.
- Lighting also showed a considerable and statistically significant improvement in the sustainably designed buildings when compared to the conventional buildings.
- No significant difference for noise was found in the sustainable buildings compared to the conventional buildings. There was a perception of slightly too much noise from various internal sources (e.g., conversations, telephones) in both samples.
- For the sustainable buildings, all of the factors in the satisfaction category showed a significant improvement over the conventional buildings. Occupants of sustainable buildings perceived
that they were 4 percent more productive than did occupants of conventional buildings. The improvement in perceived health among occupants in sustainable buildings (4.25) in comparison to occupants in conventionally designed buildings (3.29) was also statistically significant.
Widener (2009) found that most of the 21 LEED-certified projects in Illinois were not tracking health-related benefits. Survey results related to occupant overall satisfaction with building comfort (light level, noise, temperature, air quality/ventilation) were available for 11 LEED-certified projects. Widener found that, overall, occupant satisfaction was high, with the highest-rated categories being lighting and air quality/ventilation. The lowest-rated category was temperature.
Studies that seek to compare the difference in design and construction costs, the so-called first costs, or the “green premium,” between high-performance or green and conventional buildings typically discuss four different types of costs: (1) the baseline costs of the project itself; (2) the marginal capital costs of some (but not all) green improvements to the project itself, such as more expensive technologies or materials, which may be offset by savings in other systems; (3) the soft costs associated with additional documentation, analysis, and evaluation, such as energy modeling; and (4) the direct costs associated with third-party certification. Those studies, however, use different methods to define the comparison group. The different methods result in different types of findings. Some studies are specific in evaluating the cost of individual green strategies on a given building, in effect using a hypothetical baseline model for the self-same building, much as energy models do. Studies conducted for the GSA and the Indian Health Service (IHS) to look at the cost differential between LEED-certified and non-LEED-certified buildings used this approach (SWA, 2004; IHS, 2006). Caprio and Soulek (2011) looked at the cost-effectiveness of various energy efficiency improvements in Army standard designs. Others reference building budgets, asking whether the green project cost more than budgeted or anticipated for the conentional equivalent; Kats (2010) used this approach. Two studies by Mattheissen and Morris (2004, 2007) used the population approach, aiming to identify whether the population of green buildings was distinguished by cost when compared to the building stock in general. The latter approach is typically used in valuation studies that identify whether green buildings sell or lease for more than the building stock in general. The different methods for calculating incremental construction costs are valid, but should not be combined.
Matthiessen and Morris (2004) undertook a study with the goal of comparing construction costs of buildings where LEED certification was a primary goal to the costs of similar buildings where LEED was not considered during design. The authors studied 93 non-LEED-seeking and 45 LEED-seeking buildings for which data were gathered from the database of the Davis Langdon Company. All costs were normalized for time and location to ensure consistency for the comparisons. Among their conclusions were the following:
- Many projects achieve sustainable design within their initial budget or with very small supplemental funding, suggesting that owners are finding ways to incorporate project goals and values, regardless of budget, by making choices.
- There was no statistically significant difference [in cost per square foot] between the LEED-seeking and the non-LEED seeking buildings. The cost per square foot for the LEED-seeking buildings was scattered throughout the range of costs for all buildings studied, with no apparent pattern to the distribution.
A second report using the Davis Langdon database (Matthiessen and Morris, 2007) compared the construction costs of 83 buildings seeking LEED 2.1 and 2.2 New Construction certification to 138 non-LEED-seeking buildings (the samples included five different building types). Findings from the study were the following:
- Many projects were achieving LEED certification within their budgets and in the same cost range as non-LEED-seeking projects.
- While there appeared to be a general perception that sustainable design features added to the overall cost of the building, the data did not show a significant difference in the average costs of LEED-seeking and non-LEED-seeking buildings.
Kats (2010) found that the owners or owner’s representatives of 170 green buildings reported the median additional cost was 1.5 percent more to build a green building compared to a conventional building. The large majority of green building owners reported additional incremental costs between 0 and 4 percent, although the total range was 0 to 18 percent. The author concluded that most green buildings cost slightly more than similar conventional buildings to construct. Generally, the higher the certification level, the greater the cost premium, but all LEED levels could be achieved for minimal additional cost.
Three studies looked at the incremental costs associated with energy efficiency or LEED certification of federal buildings. Stephen Winter Associates (SWA, 2004) provided a detailed and structured review of both the capital and soft cost implications of achieving Certified, Silver, or Gold LEED ratings for the two building types most commonly constructed by the GSA: a five-story courthouse and a mid-rise federal office building. The study indicated that there was an inherent degree of variability to LEED construction cost impacts. However, the authors concluded that many Silver-certified projects could be built at a cost that was within 4 percent of the cost for a similar non-LEED-certified courthouse or office building, as well as occasional LEED-Gold-certified projects.
The IHS conducted a study (IHS, 2006) to evaluate the potential cost impacts of achieving a LEED-certified or a LEED-Silver certification on its facilities, which are primarily hospitals and other healthcare-related buildings. Among the study findings were the following:
- Initial capital construction costs (design and construction) would require a 1 to 3 percent increase in the budget to meet the Certification level and a 3.5 to 7.6 percent increase in the budget to meet LEED-Silver certification.
- Energy savings over 20 years of operation have the potential to significantly mitigate the initial capital cost impacts. Given the potential margin of error inherent in these types of calculations and the uncertainty of future energy prices, life-cycle cost savings may completely offset or even exceed initial capital costs.
Caprio and Soulek (2011) sought to determine the difference in initial investment (incremental construction costs) for building energy-efficiency enhancements intended to meet federal mandates. Benefit-cost analyses were conducted for the U.S. Army’s new construction standard designs for FY 2013 for the five most commonly constructed Army building types. The results were based on total energy use and were modeled, not measured. The authors noted that the study was able to show the energy effectiveness of a range of efficiency measures, but it was not able to show the cost-effectiveness of individual measures, nor was it able to optimize the designs for the highest energy performance at the lowest costs. They concluded, however, that (1) significant energy savings were possible for all climates, and (2) buildings achieving 25 to 35 percent energy savings would yield the maximum energy savings for
the lowest cost. For buildings achieving 35 to 60 percent energy savings, each increment of energy saved came at an increasingly higher cost (plug load reduction, small-scale renewable energy, building orientation, site-specific design).
Widener (2009) found wide variation among the 15 Illinois LEED-certified projects that submitted information on construction costs. Widener concluded that similar to conventional buildings, the variation in construction costs for the LEED-certified buildings may be attributed to principal building activity and the individual project’s goals and specifications.
The Kats (2010) finding that the median premium is 1.5 percent, as compared to a notional budget, is not incompatible with the IHS finding that adding green features to a reference conventional building results in a premium of 1 to 8 percent, nor is it incompatible with the Matthiessen and Morris (2004) finding that there was no statistically significant difference between the LEED-seeking and non-LEED-seeking buildings.
The committee did not identify any research studies that met its criteria and that conducted a traditional benefit-cost analysis to determine the long-term net present value savings, return on investment, or long-term payback related to the use of ASHRAE standards 90.1-2010 or 189.1-2011, the LEED or Green Globes green building certification systems, or the LEED Volume certification program.
The committee did identify 15 studies that compared the energy use of high-performance or green buildings to conventional buildings. Those studies incorporated different methods, baselines, types of buildings, and sample sizes; some applied to large areas of the country, and some were specific to regions or states. Despite these variations, the 13 studies that measured actual energy used (not modeled energy) found that high-performance or green buildings, on average, used 5 to 30 percent less site energy than conventional buildings.
There was also some evidence that high-performance or green buildings used less water than conventional buildings, with average water-use reductions in the range of 8 to 11 percent.
On a building-by-building basis, however, not all green buildings achieved energy or water savings in comparison to conventional buildings. Because there was significant variability within sample sets in terms of the types, numbers, and locations of buildings, the committee could not determine with certainty why individual buildings succeeded or failed to meet the average. For those studies that looked at buildings certified at different levels of LEED, the evidence that is available is inconclusive regarding whether LEED-Silver-certified buildings outperformed LEED-certified buildings, or whether LEED-Gold buildings outperformed LEED-Silver buildings.
There was also suggestive evidence that operations and maintenance costs may be lower for green buildings, but the very limited sample size leaves the analysis results outside the range of certainty. The three studies evaluated all included utility costs (energy and water) in operating costs, so it is not possible to determine how significant the other factors were in total operating costs.
Additionally, there was suggestive evidence that high-performance buildings result in improvements in some aspects of indoor environmental quality (air quality, thermal comfort, and overall satisfaction with workspace).
Regarding the differences in costs to design and construct green buildings in comparison to conventional buildings, the studies reviewed used different methods to identify those costs. The results from the studies indicated that design and construction cost (variously defined) would range from 0 to 8 percent higher for green versus conventional buildings, depending on the method used to calculate the costs and the type of building.