Economic and environmental impacts of community-based residential building energy efficiency investment

Energy xxx (2014) 1e10

Contents lists available at ScienceDirect

Energy journal homepage: www.elsevier.com/locate/energy

Economic and environmental impacts of community-based residential building energy efficiency investment Jun-Ki Choi a, b, *, Drew Morrison a, Kevin P. Hallinan a, b, Robert J. Brecha a, c a

Renewable and Clean Energy, University of Dayton, Dayton, OH 45469-0238, USA Mechanical and Aerospace Engineering, University of Dayton, Dayton, OH 45469, USA c Physics, University of Dayton, Dayton, OH 45469, USA b

a r t i c l e i n f o

a b s t r a c t

Article history: Received 24 July 2014 Received in revised form 28 September 2014 Accepted 26 October 2014 Available online xxx

A systematic framework for evaluating the local economic and environmental impacts of investment in building energy efficiency is developed. Historical residential building energy data, community-wide economic inputeoutput data, and emission intensity data are utilized. The aim of this study is to show the comprehensive insights and connection among achieving variable target reductions for a residential building energy use, economic and environmental impacts. Central to this approach for the building energy reduction goal is the creation of individual energy models for each building based upon historical energy data and available building data. From these models, savings estimates and cost implications can be estimated for various conservation measures. A ‘worst to first’ (WF) energy efficient investment strategy is adopted to optimize the level of various direct, indirect, and induced economic impacts on the local community. This evaluation helps to illumine opportunities to establish specific energy reduction targets having greatest economic impact in the community. From an environmental perspective, short term economy-wide CO2 emissions increase because of the increased communitywide economic activities spurred by the production and installation of energy efficiency measures, however the resulting energy savings provide continuous CO2 reduction for various target savings. © 2014 Elsevier Ltd. All rights reserved.

Keywords: Community based energy efficiency Inputeoutput analysis Local economic impacts Worst-first energy efficient measure Life cycle assessment

1. Introduction Cities are responsible for anywhere from 50 to 75% of climate change, where the lower bound assigns climate impact to the producers and the upper bound is assigned to consumer of climate impact contributions [1,2]. It is clear that research which has the potential to diminish this impact is vital. What is also clear, especially in the U.S., is that federal-level policy has not emerged to reduce this impact [3]. Yet despite the dearth of large-scale policies, numerous cities throughout the U.S. and internationally have committed to sustainability. In 2005 more than 1000 U.S. mayors signed a pledge to meet the goals stated in the Kyoto protocol and to lobby their states and the federal government to act on climate protection. Studies [4,5] found that cities that are more fiscally strapped are more likely to commit to sustainability, especially if they are mayorled cities. The research presented in this paper is established in this

* Corresponding author. Renewable and Clean Energy, University of Dayton, 300 College Park Ave., Dayton, OH 45469-0238, USA. Tel.: þ1 937 229 5344; fax: þ1 937 229 4766. E-mail address: [email protected] (J.-K. Choi).

context. For those cities and towns showing commitment to sustainability, many have adopted goals relative to a variety of sustainability indicators. For example, the City of Portland's Climate Action Plan has put Portland on a path to achieve a 40 percent reduction in carbon emissions by 2030, and an 80 percent reduction by 2050 [6,7]. In August 2013, the city of Cincinnati adopted a green energy opt out for all residents [8]. This initiative has provided a jump start toward achieving greenhouse gas emission reduction goals of 8% within four years, 40% within 20 years, and 84% by 2050. Boulder Colorado's 2012 plan calls for carbon neutrality by 2020 [9]. With its PlaNYC, New York City set an ambitious goal to reduce citywide greenhouse gas (GHG) emissions 30 percent by the year 2030 [10]. We could go on and on. But, the exciting thing is that this is not just the domain of large cities. In Ohio alone, three smaller communities have committed to 100% renewable energy and are close to achieving their goal (Oberlin, Hamilton, and Yellow Springs). While the fervor for goal setting is strong, the reality is that many of those cities and towns committing to goals do not have an etched path to get there and don't understand the economic implications of their goals. Moreover, there remain many more cities, towns, and counties which aren't acting. There is a strong need to motivate new

http://dx.doi.org/10.1016/j.energy.2014.10.082 0360-5442/© 2014 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Choi J-K, et al., Economic and environmental impacts of community-based residential building energy efficiency investment, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.10.082

2

J.-K. Choi et al. / Energy xxx (2014) 1e10

communities to begin a path toward sustainability. It is imperative that communities be provided resources to help sell sustainability to their local constituents. Linking sustainability to community-wide economic impact is essential, because even sustainability skeptics may be persuaded to see value from it if there is positive economic impact in their community. Further it is essential that communities understand the value of achievement relative to the various sustainability goals so that priorities can be established. Communities should first investigate how sustainability might be conceptualized for their community and how this understanding could lead to distinct programmatic priorities [11e13]. Particularly important for communities which are just now considering sustainability is an understanding of the economic impact locally [14e16]. Finally, research strongly suggests that to be effective, measurement of sustainability impact must be easy [17e19]. Community Sustainability Assessment Tools (LEED (Leadership in Energy and Environmental Design), HQE (High Quality Environmental standard), etc.) are highly complicated to use and manage and they are not working [20]. A second problem is that communities are unlikely to invest in collecting data on sustainability indicators unless monitoring is linked to action that provides immediate and clear local benefits [21,22]. In this context, this paper, using one sustainability metric and associated target improvement levels (energy reduction), seeks to not only demonstrate the significant economic and environmental value of sustainability to the local community, but more importantly to show that specific sustainability goals can be tailored to a local community for greatest impact. This optimization of local sustainability goals/targets is shown to be enabled from greater data granularity; in this case, from use of building specific historical energy and building data from throughout the community. Additionally, this paper seeks to show that greater data granularity illuminates nonlinear community-wide economic and environmental impact with target improvements, and by doing so, shows the value in strategic worst-to-first (WF) investment strategies (see Section 2.2 for more detailed explanation about WF) to realize greatest impact. 2. Methodology Two main research questions are posed for this study; 1) If we know the spectrum of building energy effectiveness and cost effectiveness of energy reduction as well as the manufacturing base within a community, can a more accurate assessment of economic impact of energy reduction in a community be developed? 2) What is the community-wide economic and environmental impact of a targeted energy reduction based upon a ‘worst to first’ (WF) strategy, and will the local manufacturing benefit from the

strategy? Fig. 1 shows the general methodology employed, combining historical community-wide building energy and building data with local economic inputeoutput data, thus enabling estimate of economic impact due to energy reduction for targeted measures having the greatest local economic impact. The following sub-sections detail each block described in Fig. 1. 2.1. Historical community-wide energy data and an energy model Historical energy data can be merged with county maintained building databases for all residential, commercial, governmental, and industrial buildings. This data includes use type (residential, office, etc.), square footage, and number of floors. Weather-normalization regression approaches based upon the PRISM (PRInceton Scorekeeping Method) approach [23] can be used to disaggregate energy use into annual heating, cooling, lighting/appliances, and water heating energy use. With known square footage, the energy intensity in each category can be determined. Local benchmarks can be established in each energy category for each building type. Each building can be compared to the appropriate benchmark. Simple energy models can then be established for each building and residence [24e26]. These models are based upon the following framework. With heating/cooling energy intensity (MJ/m2) estimated, the residences can be compared against benchmark heating/cooling energy intensity for comparable residences. High heating/cooling energy intensity residences in comparison to the benchmark are assumed to have both poor heating/cooling system efficiencies and envelope thermal characteristics. Whereas, low heating/cooling energy intensity residences are assumed to have high efficiency heating/cooling systems and low energy envelope characteristics. The residence models developed assume a functional relationship between the residence's energy characteristics and the difference between the heating/cooling intensity to the benchmark heating/ cooling intensity. These characteristics are bound minimally and maximally to the worst and best characteristics in practice. Ultimately, the models developed for each residence yields a predicted typical weather year energy intensity equal to that measured. The savings potential in all energy categories for each residence, along with the simple payback, can then be estimated. 2.2. A worst-to-first (WF) investment strategy Generally, analysis of the savings benefits from investment is restricted to the energy costs savings each individual residence yields from investment. Utility rebate programs for example do not differentiate between customers. Rebates are offered equally to all

Fig. 1. Flow diagram of proposed research methodology (each block explained in the following subsections).

Please cite this article in press as: Choi J-K, et al., Economic and environmental impacts of community-based residential building energy efficiency investment, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.10.082

J.-K. Choi et al. / Energy xxx (2014) 1e10

irrespective of the need for investment of individual facilities. Utilities thus employ statistical average savings from actual investments to estimate individual savings. Thus, if a community wanted to get greater savings penetration, they would need only to scale their investment. The reality is that deeper penetration will inevitably require a greater than linear investment. The reason is that increased investment generally will capture more efficient residences where efficiency upgrades will not realize as much savings. A more cost effective strategy for investment in energy reduction in a community could be realized if investments were allocated to support actions in the worst residences first. In other words, the most cost effective community energy savings would be derived if the measures having the lowest levelized cost from among all possible measures within the whole community were implemented first; the next best second; and so on. For example, in the case study we presented in the Section 3, energy models were developed for all residences based upon historical energy use and available residential building information. From these energy models, savings from and investment costs required to implement various heating and cooling energy reduction measures are estimable for each residence. The corresponding community-wide investment and savings could be determined based upon a ‘worst to first (WF)’ implementation strategy. The next section describes the framework for assessing the local economic impact of energy saving actions.

2.3. Community economic inputeoutput model Different macro-economic models have different strengths and limitations in terms of economic rationale, level of disaggregation in decision variables, and the spatial, temporal, and geographical scopes of application. One of the most significant differentiators is the degree of detail with which commodities and technologies are represented, ranging from top-down models to bottom-up models. Top-down models evaluate an entire economic system via a relatively small number of aggregated economic variables. These models generally focus on the economic description of interactions and relations between aggregate economic sectors. A representative top-down approach employing economic inputeoutput models which evaluate the environmental and economic impacts in an aggregated macroeconomic level [27e31]. Bottom-up models utilize detailed information about different technologies, and relate consumption or supply to technical performance. In bottom-up models, important technologies are identified by a detailed descriptions of its inputs, outputs, unit costs, and several other technical and economic characteristics [32e34]. This research integrates a top-down economic framework (i.e. InputeOutput Analysis) with detailed local data using IMPLAN (Impact Analysis for PLANning) data. There are two types of inputeoutput study. A primary inputeoutput study is based on data collected directly from industries. An example is the United State's Benchmark Study on InputeOutput Account [35]. Other countries have done primary national level inputeoutput studies as well. However, primary state or local level inputeoutput studies are not common due to the high cost of data collection. IMPLAN is an example of a secondary inputeoutput modeling system which allows the inputeoutput analysis in state and local levels as well. IMPLAN was originally developed by the USDA Forest Service [36]. The Leontief InputeOutput model [37e39] is the backbone of this model. It takes user specific inputs and generates economic impact output through matrices based on actual historical economic data. It is usually utilized to analyze the change of the total production vector (x) by a change of the final demand vector (y) by equation (1):

Dx ¼ ðI  AÞ1 Dy;

3

(1)

where A is the n  n direct requirement matrix, y is the monetary amount of the final demand column vector, and x is the total monetary industry output column vector. The magnitude of the direct, indirect, and total effects is completely dependent on the values of the A matrix. The uniqueness of IMPLAN modeling is the breath and granularity of the database. IMPLAN output includes direct, indirect, and induced effects from production changes brought by increases in final demand. For example, the addition of attic insulation to reduce heating and cooling energy in a community increases final demand for insulation materials, impacting directly local production (if it exists) and local contractual services for installation. Indirect effects result from changes in the demand for the main goods and services necessary for production of a sector's output. A special feature of the IMPLAN model is the ability to capture induced effects through the Social Accounting Matrix (SAM) [40]. Unlike traditional inputeoutput analysis, SAM captures inter-institutional transactions, which includes cash flows from business to households, people to government, and government to people. Induced effects result from expenditures made by employees of the directly and indirectly effected industries on general consumer goods and services in a geographical area. For example, households may spend their energy bill savings on entertainment industries or food service industries located in a geographical reason of the consideration. IMPLAN generates key indicators including job creation, total production output, labor income, and total value added. The IMPLAN approach has been applied to estimate economic impact of sustainability goals relative to energy efficiency at state levels [41,42], green energy at a national level [43], recycling at national and state levels [44e47], green enterprise creation [48,49], green buildings at a national level [50], local food at state levels [51,52], use of mass transportation at a national level [47,53], and water conservation [54,55].

2.4. Environmental impact model Life Cycle Assessment (LCA) aims to determine the inputs, outputs and their impacts from the complex network of economic and industrial activities that constitute the life cycle of a selected product or process. LCA methodology is differentiated by details in the Life Cycle Inventory (LCI) data and the level of the system boundary settings. Bottom-up models are process-specific (i.e. process-LCA) [56], relying on a detailed inventory of the inputs and emissions of the selected processes. The main purpose of this type of LCA is to quantify the amount of resource consumption and emissions from industrial products and processes [57e59]. However, it often uses arbitrary system boundaries which can generate significant errors [60]. In order to address this system boundaryrelated issue, some study have utilized a meta-data analysis approach to investigate these errors [61,62]. These studies compared a wide range of data from previous LCA studies and quantified the variation in the reported results in order to harmonize the system boundary variations. Top-down models consider the entire economy and, unlike the process-LCA, avoid defining an arbitrary boundary around selected processes. Previous studies have used an integrated top-down LCA framework for assessing the economic and environmental implication of policy by integrating multiple modeling schemes [63e65]. Various types of resource consumption and emissions change can be estimated with the inputeoutput based life cycle assessment (LCA) and the economic output data generated using the methodology described in the previous section. Equation (2) is used

Please cite this article in press as: Choi J-K, et al., Economic and environmental impacts of community-based residential building energy efficiency investment, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.10.082

4

J.-K. Choi et al. / Energy xxx (2014) 1e10

to capture the economy-wide CO2 change generated from considered actions.

^
DX DCO2 ¼ RCO i 2 i

Table 1 Top industry sectors of Hamilton County, OH (sorted by the total production). Sector

(2) 381

where, R is the CO2 intensity matrix and sector specific values (i.e. average CO2 generation per $1 production from each industry) are listed in the diagonal of the matrix (see Electronic supplementary information), DX is a column vector with values representing the economic output changes in each sector after the investment on the furnace upgrades. The R values are collected from the previous research [27,66]. As described in the previous section, DX is composed of the direct, indirect, and induced portions of the economic activities, therefore, the change of emission vector (i.e. DCO2) also can be represented in the same manner to identify how much emission changed directly, indirectly, or from induced impacts to the base year. Although only CO2 emission is presented in this paper, change of other emissions and resource consumption can be readily analyzed with the same methodology as well. 3. Case studies In this section, the economic impact of Hamilton County, Ohio, United States resulting from specific energy efficient investments is established as a function of energy reduction targets. Two specific energy reduction measures are considered; attic insulation addition and furnace upgrade. Hamilton County, Ohio was chosen, because the City of Hamilton Municipal Utility has agreed to provide historical energy data (natural gas and electric) for every residential, commercial, and industrial building in the city. Hamilton County is located in the southwest corner of the state of Ohio. As of year 2011, the estimated total population is 800,362, the total number of households is 341,483 and there are 324 industries in this location [67]. Estimated gross regional product is $53.7 Billion, total personal income is $38.9 Billion, and total employment is around 594,000 [68]. 3.1. Current economic status of Hamilton County, Ohio In order to evaluate economic impact in a community, it is important to understand its economic profile. Table 1 shows the top ten economic sectors of Hamilton County in descending order of total production output (i.e. the dollar amount produced from each economic sector). It provides some sense of the magnitude of the Hamilton County's baseline economy and shows the type of industries that have the highest local economic impacts in terms of total production output. 3.2. Level of energy efficiency investment In order to estimate economy-wide impacts for Hamilton County resulting from a specified investment amount in various residential heating energy conservation measures, the first step is to allocate the specified investment dollars to the appropriate economic sectors. The Hamilton County census information indicates that there are 234,375 residences (i.e. multi-unit housing 37% is not included) [67]. Given that size of the average residence is approximately 186 sm and the average energy use is 365 MJ/sm/ annum [69], the average annual heating energy per house is estimated to be a 67 GJ/year. Therefore, total residential heating energy for the county is estimated about 16 PJ. Based upon this estimate and using average investment costs for just a couple of measures [70], attic insulation addition ($5.7/GJ, annual savings over the investment lifetime) and furnace/boiler upgrades ($7.5/GJ, annual savings), and given a 15 year life cycle cost, a 10% community-wide

319 397 285 360 354

126 394

357 413 351 356 139 133

Management of companies and enterprises Wholesale trade businesses Private hospitals Aircraft engine and engine parts manufacturing Real estate establishments Monetary authorities and depository credit intermediation activities Other basic organic chemical manufacturing Offices of physicians, dentists, and other health practitioners Insurance carriers Food services and drinking places Telecommunications Securities, commodity contracts, investments, and related activities Toilet preparation manufacturing Pharmaceutical preparation manufacturing 314 þ Sectors …… Total

Output ($million)

Employment (person)

$6845

30,694

$4975 $4201 $3625

26,058 32,511 6516

$3546 $2980

20,440 8746

$2429

1396

$2425

17,577

$2248 $2142 $1,996 $1754

8088 38,749 3814 11,536

$1730 $1612

2340 1549

… $95,014

… 594,010

heating energy reduction in these categories would realize respectively investments of $9M and $12M for attic insulation addition and furnace upgrades. 3.3. Economic impact of community-wide attic insulation addition and furnace upgrades The first scenario focuses on the local economic impacts of a $9M energy efficient investment to the attic insulation material manufacturing and the installation services to achieve 10% community-wide heating energy reduction. Out of the total $9M investment, 25 percent (i.e. $2.25M) and 75 percent (i.e. $6.75M) are allocated to the “Insulation material and foam product manufacturing” sector and to the “maintenance and repair construction of residential structures” sector respectively. The latter is associated with installation services. A $9M investment in increased insulation to reduce heating energy and cost in the economy of the Hamilton County can result in a total local economic impact of $12.3M, stemming from the $9M coming from direct impact, $1.6M coming from indirect impact, and $1.7M coming from induced impacts. Job creation over the investment period yields a total of 66 jobs, with 43, 10, and 13 coming respectively from direct, indirect, and induced impacts. Inside of IMPLAN software, information about new household income in each economic sector is used to estimate the job creation in the region. That is, increase in household income data are used to calculate the direct, indirect, and induced portion of the job creation. Direct effects are those jobs created by investing to the specific economic sectors (i.e., expected job created by furnace industry or attic manufacturer in the region). The indirect effect represents jobs created by businesses associated with those manufacturers (i.e. local supplier) when their sales rise indirectly. For example, a furnace/attic manufacturer might order more parts to suppliers in the region. The induced effect represents the additional household spending because income has risen throughout the local economy. Jobs in the community businesses not directly/indirectly related to the furnace/attic insulation manufacturing also increase opportunities since the workers of manufacturers/suppliers would spend

Please cite this article in press as: Choi J-K, et al., Economic and environmental impacts of community-based residential building energy efficiency investment, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.10.082

J.-K. Choi et al. / Energy xxx (2014) 1e10

the income they earn from employment on various goods and services in community. A more detailed level of the breakdown of economic effects by sectors is available in the Electronic supplementary information. Fig. 2 summarizes the combined economic effects in descending order of total economic impact. It maps the percentile contribution of direct, indirect, and induced effects of some sectors caused by the $9M community-wide investment in attic insulation addition. It provides insight into the most important types of economic effects and the percentile contribution of those to the local industries. For example, economic sectors such as “maintenance and repair construction of residential” and “Urethane and other foam product manufacturing” have almost 99% of the direct economic effect. The following sectors are most strongly effected indirectly; basic organic chemical manufacturing, architectural, engineering, and related services, adhesive manufacturing, retail stores, transportation, plastic packaging materials and unlaminated film and sheet manufacturing, and management of companies and enterprises. These sectors most likely provide goods and services to the insulation manufacturing and installation related services directly and indirectly (i.e. inter-industry transactions). Finally, the most important induced economic impacts are associated with imputed rental activity for owner-occupied dwellings, private hospitals, health services, food services/drinking places, and insurance carriers. These sectors have no inter-industry transactions with the insulation related activities directly, but households may spend the increased income salaries for activities in these non-production related industries, thereby boosting the local economy. In a similar manner, the second scenario assesses the local economic impact of a $12M investment for furnace upgrades to achieve 10% community-wide heating energy reduction. Out of a total $12M investment, 50 percent (i.e. $6M) is allocated to the “heating equipment manufacturing” sector and the other 50 percent is allocated to the “engineering service” sector for upgrades respectively. This investment results in a total estimated local economic impact of $17.9M, stemming from the $12M coming from direct impact, $2.4M coming from indirect impact, and $3.5M coming from induced impacts. Job creation over the investment

5

period yields a total of 106 jobs, with 63, 17, and 26 coming respectively from direct, indirect, and induced impacts. Tables and a figure listing more detailed breakdown of economic effects by sectors can be found in the Electronic supplementary information. While both measures yield the same community-wide energy reduction, the short term economic impact from furnace upgrades is larger since: 1) it requires greater investment than for attic insulation addition; and 2) it leverages greater amplification (the ratio of total economic impact divided by investment is 1.5 as compared to 1.36 for the attic insulation investment). 3.4. Adopting a worst-to-first (WF) investment strategy The previous analysis is based upon an assumed constant cost of investment relative to savings, e.g., the cost and economic impact are linear with savings. Thus, if a community wanted to get greater savings penetration, they would need only to scale their investment. We begin by considering natural gas heating energy use in a nearby town to Hamilton County, with just over 2000 residences. Fig. 3 shows the results from the study [24]. Shown in a) is a plot of investment cost per GJ saved annually for the hierarchy of residences (from worst to best) in this town from attic insulation addition. Shown in b) is a plot of investment cost per GJ saved annually for furnace upgrades versus collectively community-wide HVAC (Heating, Ventilating, and Air Conditioning) savings. Clear from these figures is that the investment cost to achieve unitary savings ($/GJ) increases through the hierarchy of energy effectiveness in residences, moving from the worst to first houses although the investment per residence remains roughly constant. Additionally, it is clear that an increasing cost per savings is required to gain greater savings penetration in the community. These results help to improve the estimate of a local economic impact of an energy reduction projects from specific measures, using the WF strategy for Hamilton, Ohio. Assumed is that the housing stock considered in the nearby town mirrors that of Hamilton County, so that the investment costs per savings estimated in the nearby community reflect that of Hamilton County. This assumption is reasonable as the age of the housing stock and affluence in the communities are nearly the same.

Fig. 2. Major economic sectors affected by the investment on attic insulation in the Hamilton County, OH.

Please cite this article in press as: Choi J-K, et al., Economic and environmental impacts of community-based residential building energy efficiency investment, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.10.082

6

J.-K. Choi et al. / Energy xxx (2014) 1e10

Fig. 3. Levelized cost for attic insulation addition for: (a) individual houses presented in ascending order of cost effectiveness and (b) for the collective grouping of houses presuming that houses are improved in order of levelized cost as a function of cumulative HVAC energy savings [24].

Table 2 Granular energy saving for attic insulation and furnace upgrades. Savings (%)

1 2 3 4 5 6 7 8 9 10 11

Attic insulation

Furnace upgrade

Ann. savings ($)

Investment ($)

Simple payback (years)

Ann. savings ($)

Investment ($)

Simple payback (years)

91,879 183,757 275,636 367,515 459,393 551,272

306,262 918,786 1,837,573 3,062,621 4,593,931 6,431,504

3.3 5.0 6.7 8.3 10.0 11.7

91,879 183,757 275,636 367,515 459,393 551,272 643,150 735,029 826,908 918,786 1,010,665

612,524 1,408,806 2,388,844 3,552,640 4,900,193 6,431,504 8,146,572 10,045,397 12,127,979 14,394,318 16,844,415

6.7 7.7 8.7 9.7 10.7 11.7 12.7 13.7 14.7 15.7 16.7

Table 2 presents the percentage community-wide heating energy reduction versus annual energy cost savings, investment, and simple-payback for both improvement measures considered. This table shows that lower savings for the community can be realized with attractive simple paybacks from investment. Deeper savings are possible, though, but with less attractive economic benefit to the investors. However, the economic benefit derived from savings alone does not begin to measure the complete economic impact to the community. 3.5. Short terms community-wide economic impacts of the investments considering WF strategies Table 3 illustrates the breakdown of the economic impacts from investment from adoption of these measures as direct, indirect and induced effects for a $6.43M investment in attic insulation addition and furnace upgrades in order to achieve a 6% of heating energy reduction (i.e. Table 2). Interestingly, the simple payback period for these two measures is the same when using a WF implementation methodology at this savings level. For attic insulation, 25 percent of investment (i.e. $1.6M) is allocated to insulation manufacturing and 75 percent (i.e. $4.83M) is allocated to the possible sector of installation of insulation materials (i.e. 369 e Architectural, engineering, and related service). For furnace upgrades, 50 percent (i.e. $3.2) of investment is allocated to each of the furnace manufacturing (216 e warm air equipment manufacturing) and the upgrade services (369 e Heating engineering consulting services) respectively. Table 3 shows that a $6.43M investment in attic insulation can result in a total local economic impact around $8.8M, with 73, 13, and 14 percent of this impact coming from direct, indirect, and induced effects respectively. It also created 47 jobs in the

year of investment from the increased economic activities; 31 directly, 7 indirectly, and 9 from induced impacts. Note that if Hamilton County did not have manufacturing related to these activities, then it might be possible to add such a manufacturing capability were there to be a certain local market given a commitment to investment significantly in this energy conservation measure for the community. Most notable is that for achieving same target saving with the equivalent investment for furnace upgrades results somewhat greater local economic impact than attic insulation addition: $9.6M total economic effect, with 67, 13, and 20 percent from direct, indirect, and induced effects respectively. Number of job created in the region is 57. Thus, a general conclusion can be drawn. For each community, priority actions toward energy reduction can be established to maximize economic impact. Fig. 4 shows the amount of the economy-wide total production change resulting from investments in attic insulation (Fig. 4(a)) and furnace upgrades (Fig. 4(b)) using the WF strategy for achieving energy reduction target of 1e6 percent. The higher indirect and

Table 3 Breakdown of total economic impact resulting from investment on attic insulation and furnace upgrade. Economic impact type

Attic insulation

Furnace upgrade

Employment (job)

Total economic output

Employment (job)

Total economic output

Direct effect Indirect effect Induced effect Total effect

31 7 9 47

$6,431,503 $1,161,511 $1,240,482 $8,833,497

34 9 14 57

$6,431,502 $1,264,545 $1,869,691 $9,565,739

Please cite this article in press as: Choi J-K, et al., Economic and environmental impacts of community-based residential building energy efficiency investment, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.10.082

J.-K. Choi et al. / Energy xxx (2014) 1e10

7

Fig. 4. Total economic impact of Hamilton County for various level of energy reduction targets for: (a) attic insulation and (b) furnace upgrades.

induced effects for furnace upgrades reflects the fact that local business rely on other local businesses more strongly and, combined, their wages stay local. Overall, a deeper saving target yields an increasingly greater short term economic impact in the community, as the investment required to achieve additional savings increases non-linearly. Fig. 5 illustrates the expected incremental economic growth in Hamilton County for different heating savings goals for only the furnace upgrade measure based upon the WF strategy (note that this is not showing the total economic effect but incremental changes). A similar figure could be presented for attic insulation addition. There are some very interesting points to be made about this plot. First, for low target savings, the community wide economic impact far surpasses the investment. Deeper energy savings yield increasingly higher community-wide economic impacts, however, the amplification effect from the investment diminishes. Finally, it is clear that as the savings target surpasses 10%, the economy-wide “incremental” growth actually decreases. This happens because the worst household's furnaces are upgraded first and the incremental investment decreases to achieve deeper saving goal. Because lesser incremental investments are necessary to upgrade furnaces in more efficient residential buildings, the community wide economic impact decreases (e.g., smaller furnaces are needed). In summary, these results suggest that the maximum short-term local economy-wide growth can be achieved from

Fig. 5. Incremental economic growth vs. heating energy saving target.

furnace upgrade goals for the community at a 10 percent heat energy target savings goal. 3.6. Long-term community economic impact from energy savings Fig. 6 illustrates the cumulative long-term total economic impact in Hamilton County considering a 15 year project life for the furnace upgrade case and a 6% community-wide energy reduction using a WF implementation methodology. The short term economic impacts (direct, indirect, and induced) are shown in the first year. In each year of the project life, including the first year, the annual energy cost savings derived from the upgrade and the associated induced impact of these savings are also added. IMPLAN is used to estimate the latter with a 3.3 percent energy escalation rate. Clear from this figure is that investment costs are rapidly recovered when the system boundary for economic impact expands beyond the residence and includes the community as a whole. In this case the investment of $6.4M yields an immediate local impact of $9.6M when considering the local direct, indirect, and induced impacts of the investment. The energy savings then yield induced impacts for each year of the investment life cycle. Fig. 7 presents similar results for various savings targets over the life of the upgrade. Not surprisingly, deeper savings targets are

Fig. 6. Community-wide economic impact for 6% energy saving target from the furnace upgrade in Hamilton County, OH.

Please cite this article in press as: Choi J-K, et al., Economic and environmental impacts of community-based residential building energy efficiency investment, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.10.082

8

J.-K. Choi et al. / Energy xxx (2014) 1e10

Fig. 7. Long term economic impact for various level of energy saving target from the furnace upgrade in Hamilton County, OH.

associated with significant community-wide economic stimulus and greater annual stimulus in each year thereafter. 3.7. Environmental impact of energy efficient investment Using the methodology described in Section 2.3 and the results from Sections 3.5 and 3.6, the short-term and long-term environmental impact associated with furnace upgrade (and any other measure) for Hamilton County, OH can be estimated. Fig. 8 illustrates a simplified view of the increased amount of the CO2 emission throughout the 434 economic sectors for a $12M investment on the furnace upgrade is made. Details of emissions results for each sector can be found in the Electronic supplementary information. For furnace upgrades, the most important direct impact sectors are Sectors 216 (i.e., air conditioning, refrigeration, and warm air heating equipment manufacturing) and 229 (i.e.

Fig. 9. Economy-wide total CO2 emission in Hamilton County, OH for projected 15 years with the furnace upgrades for total 6% residential energy savings.

engineering services) which respectively realize 1.17  104 t-CO2 and 1.96  103 t-CO2. All other sectors generate additional amount of CO2 emission indirectly. The total short-term economy-wide CO2 emission increase is 2.23  104 t-CO2, with 1.5  104 t-CO2., 2.95  103 t-CO2 and 4.36  103 t-CO2. coming from direct, indirect, and induced impacts respectively. The power generation sector's CO2 emissions increased quite significantly (2.74  103 t-CO2) by indirect and induced impacts. There are two main reasons: i) many of the effected economic sectors utilize electricity indirectly; and ii) the carbon intensity of the power generation sector is relatively higher than most of other economic sectors. Although, CO2 emissions are increased from the investment, energy savings from the furnace upgrade will provide continuous reduction of the CO2 emission to the community over the life of the investment. In the base year (2013), the total community residential energy consumption in the area is estimated to 16 PJ as shown in the previous section. Therefore, the estimated CO2 emission generated

Fig. 8. Economic sector's CO2 emission generated from the $12M investment on the furnace upgrades in the Hamilton County, OH at the first year of project (full name of the economic sector and values can be found in the Electronic supplementary information).

Please cite this article in press as: Choi J-K, et al., Economic and environmental impacts of community-based residential building energy efficiency investment, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.10.082

J.-K. Choi et al. / Energy xxx (2014) 1e10

9

from the residential housings in the community is 7.96  105 tons in the base year as shown in Fig. 9, assuming that a natural gas carbon intensity of 50 g CO2/MJ. In the year of investment (2014), the total economy-wide short-term CO2 increased about 2.23  104 t-CO2 because of the $12M investment on the furnace upgrades. For the following years, considering the 6% target residential energy savings and no efficiency degradation from the furnace upgrades, community heating emissions will be reduced by 4.78  104 t-CO2 every year as shown in Fig. 9. Therefore, the CO2 emission payback time is less than a year (i.e. amount of the economy wide CO2 gain e 2.23  104 t-CO2 vs. CO2 reduction by 6% residential energy savings). Similar results are obtained for attic insulation addition.

country which enables local optimization of sustainability goals and improved opportunities for cities/towns/communities to communicate the value of sustainability to their communities.

4. Conclusion

Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.energy.2014.10.082.

In this paper, we integrate historical community residential energy consumption data, economic inputeoutput analysis, and emission intensity of economic sectors to estimate the economic and environmental impacts of the potential investment on energy efficiency measures such as attic insulations and furnace upgrades. Most interesting finding is that the total economic impact for the community is much better if all economic impacts (i.e. direct, indirect, and induced) are considered; not just the economic benefit associated with savings e which is the criterion used to evaluate the potential for investment in energy efficiency. We demonstrate the importance of greater data granularity on estimating community wide economic impact from strategic energy reduction utilizing a worst-first incentivation approach (i.e. targeting the least efficient buildings first). Data granularity has two very important implications. First, increased granularity can enable estimation of investment cost as a function of communitywide penetration of energy efficiency. Second, it permits estimation of community wide savings and economic impacts as a function of investment for different energy conservation measures. This ability is an enabler for determining which measures add greatest economic value to a community. It also offers to evaluate the potential economic impact achievable by attracting a manufacturer connected to specific energy efficiency measures and can guarantee local purchase in order to support community wide energy reduction goals. The payback to a community is much greater if they have manufacturing integral to the energy conservation measure. The direct economic impacts are essential. If the manufacturing doesn't exist, the local economic commitment can be attractive to potential local manufacturers. The worst-to-first (WF) strategy may not be practical yet; however with the Property Assessed Clean Energy (PACE) financing [71] beginning to take off nationally in the U.S., this strategy has much promise for communities interested in deep savings with significant economic impact to a community. Additionally, increasing amounts of readily accessible data are becoming available for buildings. Increased building data combined with residential energy data can enable greater strategic implementation of energy reduction. Each economic sector's emission intensity values (e.g., CO2 emission per $1 production of sector's product/service) adopted for this study is based on the national average since the local level environmental emission data is not currently available. There is a strong need to integrate the community specific life cycle inventory (LCI) data (i.e. using a local electricity generation mix instead of using the national level electricity generation mix) to capture accurately the environmental impacts generated from various local scenarios (i.e. certain community consumes much greener source of electricity than other community). Our ultimate goal is to develop an economic and environmental impacts resource to cities/towns/communities throughout the

Acknowledgments Authors appreciate the University of Dayton for the research council seed grant (Grant No. IGRQ14). Part of this research was supported by the United States National Science Foundation (NSF0850050). Appendix A. Supplementary data

References [1] Satterthwaite D. Cities' contribution to global warming: notes on the allocation of greenhouse gas emissions. Environ Urban 2008;20(2):539e49. [2] UNEP. In: Lemmet S, editor. Buildings and climate change: summary of decision makers. United Nations Sustainable Buildings and Climate Initiative; 2009. [3] Rabe BG. Beyond Kyoto: climate change policy in multilevel governance systems. Gov Int J Police Adm Inst 2007;20(3):423e44. [4] Sharp EB, Daley Dorothy M, Lynch Michael S. Understanding local adoption and implementation of climate change mitigation policy. Urban Aff Rev 2011;47(3):433e57. [5] Goodyear S. Why the federal government should give more power to mayors. The Atlantic Media 2012. Available from: http://www.citylab.com/work/2012/ 04/why-federal-government-should-give-more-power-mayors/1793/. [6] Portney KE. Taking sustainable cities seriously: economic development, the environment, and quality of life in American cities. Cambridge, MA: MIT Press; 2003. [7] Portney KE. Civic engagement and sustainable cities in the United States. Public Adm Rev 2005;65(5):579e91. [8] Heeter Jenny, McLaren Joyce. Innovations in voluntary renewable energy procurement. National Renewable Energy Laboratory; 2012. [9] Cindy Domenico WT, Gardner Deb. Environmental sustainability plan. Boulder County Commissioners; 2012. [10] New York City Mayor's Office. Overview of the greener, greater buildings plan. New York City Mayor's Office of Long-Term Planning and Sustainability; 2012. [11] Zeemering ES. What does sustainability mean to city officials? Urban Aff Rev 2009;45(2):247e73. [12] Valentin A, Spangenberg JH. A guide to community sustainability indicators. Environ Impact Assess Rev 2000;20(3):381e92. [13] Fraser E, Dougill Andrew, Mabee Warren, Reed Mark, McAlpine Patrick. Bottom up and top down: analysis of participatory processes for sustainability indicator identification as a pathway to community empowerment and sustainable environmental management. J Environ Manag 2006;78(2):114e27. [14] Sharifi A, Murayama Akito. A critical review of seven selected neighborhood sustainability assessment tools. Environ Impact Assess Rev 2013;38:73e87. [15] Tanguay GA, Rajaonson J, Lefebvre J-F, Lanoie P. Measuring the sustainability of cities: an analysis of the use of local indicators. Ecol Indic 2010;10(2): 407e18. [16] Turcu C. Re-thinking sustainability indicators: local perspectives of urban sustainability. J Environ Plan Manag 2012;56(5):695e719. [17] Mitchell G, May A, McDonald A. PICABUE: a methodological framework for the development of indicators of sustainable development. Int J Sustain Dev World Ecol 1995;2(2):104e23. [18] Epstein M. Making sustainability work: best practices in managing and measuring corporate social, environmental, and economic impacts. United Kingdom: Greenleaf Publishing Ltd.; 2008. [19] Reed MS, Fraser EDG, Dougill AJ. An adaptive learning process for developing and applying sustainability indicators with local communities. Ecol Econ 2006;59(4):406e18. [20] Scofield JH. Do LEED-certified buildings save energy? Not really…. Energy Build 2009;41(12):1386e90. [21] Freebairn D, King C. Reflections on collectively working toward sustainability: indicators for indicators! Anim Prod Sci 2003;43(3):223e38. [22] Rydin Y, Holman N, Wolff E. Local sustainability indicators. Local Environ 2003;8(6):581e9. [23] Fels M. PRISM: an introduction. Energy Build 1986;9:5e18. [24] Villoria-Siegert R, Brodrick P, Hallinan K, Brecha RJ. Cost-availability curves for hierarchical implementation of residential energy-efficiency measures. Energy Efficiency 2014:1e13. [25] Hallinan PK, Northridge Jessica, Kissock J Kelly, Brecha Robert. Establishing building recommissioning priorities and potential energy savings from utility energy data. ASHRAE Trans 2011;117(2):495e505.

Please cite this article in press as: Choi J-K, et al., Economic and environmental impacts of community-based residential building energy efficiency investment, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.10.082

10

J.-K. Choi et al. / Energy xxx (2014) 1e10

[26] Hallinan PK, Brecha R, Mitchell A, Kissock K. Targeting residential energy reduction for city utilities using historical electrical utility data and readily available building data. ASHRAE Trans 2011;117(2):577e84. [27] Choi J-K, Bakshi BR, Haab T. Effects of a carbon price in the U.S. on economic sectors, resource use, and emissions: an inputeoutput approach. Energy Policy 2010;38(7):3527e36. [28] Meng FY, Zhou DQ, Zhou P, Bai Y. Sectoral comparison of electricity-saving potentials in China: an analysis based on provincial inputeoutput tables. Energy 2014;72:772e82. [29] Kuswardhani N, Soni P, Shivakoti GP. Comparative energy inputeoutput and financial analyses of greenhouse and open field vegetables production in West Java, Indonesia. Energy 2013;53:83e92. [30] Qi T, Zhou L, Zhang X, Ren X. Regional economic output and employment impact of coal-to-liquids (CTL) industry in China: an inputeoutput analysis. Energy 2012;46(1):259e63. €sse n J, Holmberg J, Wadeskog A, Nyman M. Direct and indirect energy use [31] Na and carbon emissions in the production phase of buildings: an inputeoutput analysis. Energy 2007;32(9):1593e602. [32] Van der Voort E, Thonet D, Dechamps D, Guilmot J. Energy supply modeling package EFOM-12C mark 1 mathematical description. 1984. Louvain-laNeuve, France. [33] Lydick JK, Morris SC, Lee J, Goldstein G. Demo MARKAL, abbreviated version of the US MARKAL energy systems model, created for demonstration: abilities, limitations, and demonstration. Brookhaven, NY: Brookhaven National Laboratory; 1990. [34] Messner S, Strubegger M. User's guide for MESSAGE. Laxenburg: International Institute for Applied Systems Analysis; 1995. [35] U.S. BEA. National economic accounts in the United States. Bureau of Economic Analysis; 2014. Available from: http://www.bea.gov/. [36] Alward GS, Palmer CJ. IMPLAN: an inputeoutput analysis system for forest service planning. Food and Agriculture Organization of the United Nations; 1983. [37] Leontief W. Quantitative inputeoutput relations in the economic system of the United States. Rev Econ Stat 1986;18(3):105e25. [38] Leontief W. Inputeoutput economics. 2nd ed. Oxford: Oxford University Press; 1986. [39] Leontief W, Duchin F, Szyld D. New approaches in economic analysis. Science 1985;228:419e22. [40] Roland-Holst DW, Sancho Ferran. Relative income determination in the United States: a social accounting perspective. Rev Income Wealth 1992;38(3):311e27. [41] ACEEE. How does energy efficiency create jobs?. 2012. Available from: http:// aceee.org/files/pdf/fact-sheet/ee-job-creation.pdf. [42] Angelou Economics. The Texas economic impact of energy efficiency programs administered by investor-owned utilities. 2011. [43] Grover S. Energy, economic, and environmental benefits of the Solar America Initiative. 2007. Available from: http://www.nrel.gov/docs/fy07osti/41998.pdf. [44] Choi J-K, Fthenakis V. Crystalline silicon photovoltaic recycling planning: macro and micro perspectives. J Clean Prod 2014;66:443e9. [45] Choi J-K, Fthenakis V. Economic feasibility of recycling photovoltaic modules. J Ind Ecol 2010;14(6):947e64. [46] Nakamura S, Kondo Yasushi. Waste inputeoutput analysis: concepts and application to industrial ecology. Eco-Effic Ind Sci 2009;26. [47] Choi J-K, Fthenakis V. Design and optimization of photovoltaics recycling infrastructure. Environ Sci Technol 2010;44(22):8678e83. [48] Edgeman R, Eskildsen J. Modeling and assessing sustainable enterprise excellence. Bus Strategy Environ 2014;23(3):173e87.

[49] vom Brocke Jan, Seidel S, Recker Jan. Green business process management: towards the sustainable enterprise. New York: Springer; 2012. [50] Booz Allen Hamilton. U.S. Green Building Council green jobs study. Available from: http://www.usgbc.org/Docs/Archive/General/Docs6435.pdf. [51] Hodges AW, Stevens Thomas J. Local food systems in Florida: consumer characteristics and economic impacts. 2013. [52] Seung C, Waters E. The role of the Alaska seafood industry: a social accounting matrix (SAM) model approach to economic base analysis. Ann Reg Sci 2006;40(2):335e50. [53] Weisbrod G, Reno Arlee. Economic impact of public transportation investment. 2009. [54] Velazquez E. An inputeoutput model of water consumption: analysing intersectoral water relationships in Andalusia. Ecol Econ 2006;56:226e40. [55] Hristoc J, Martinovska-Stojceska Alexsandra, Surry Yves. Inputeoutput analysis for water consumption in Macedonia. In: European summer school in resource and environmental economics: management of international water; 2012. Venice, Italy. [56] Udo de Haes H, Reinout H, Huppes Gjalt, van der Voet Ester, Hettelingh JeanPaul. Full mode and attribution mode in environmental analysis. J Ind Ecol 2000;4(1):45e56. [57] Baumann H, Tillman AM. The Hitch Hiker's guide to LCA e an orientation in life cycle assessment methodology and application. Lund, Sweden: Student Literature; 2004. [58] Kittner N, Gheewala SH, Kamens RM. An environmental life cycle comparison of single-crystalline and amorphous-silicon thin-film photovoltaic systems in Thailand. Energy Sustain Dev 2013;17(6):605e14. [59] SETAC. Life cycle assessment. 2008 [cited], http://www.setac.org/node/32#. [60] Joshi S. Product environmental life-cycle assessment using inputeoutput techniques. J Ind Ecol 2000;3(2):95e120. [61] Kim HC, Fthenakis Vasilis, Choi Jun-Ki, Turney Damon E. Life cycle greenhouse gas emissions of thin-film photovoltaic electricity generation. J Ind Ecol 2012;16:S110e21. [62] Hsu DD, O'Donoughue P, Fthenakis V, Heath GA, Kim HC, Sawyer P, et al. Life cycle greenhouse gas emissions of crystalline silicon photovoltaic electricity generation. J Ind Ecol 2012;16:S122e35. [63] Arpad Horvath CTH, Lave Lester B, McMichael Francis C, Wu Tse-Sung. Toxic emissions indices for green design and inventory. Environ Sci Technol 1995;29(2):86Ae90A. [64] Choi J-K, Friley Paul, Alfstad Thomas. Implications of energy policy on a product system's dynamic life-cycle environmental impact: survey and model. Renew Sustain Energy Rev 2012;16(7):4744e52. [65] Chris Hendrickson AH, Joshi Satish, Lave Lester. Green design: economic inputeoutput models for environmental life-cycle assessment. Am Chem Soc 1998;32(7):184Ae91A. [66] Zhang Y, Baral A, Bakshi BR. Accounting for ecosystem services in life cycle assessment, part II: toward an ecologically based LCA. Environ Sci Technol 2010;44(7):2624e31. [67] United States Census Bureau. State & county quick facts. 2011. [68] IMPLAN Group. In: Choi Jun-Ki, editor. IMPLAN Hamilton County, OH economic dataset; 2013. [69] EIA. Residential energy consumption survey (RECS). 2013. Available from: http://www.eia.gov/consumption/residential/ [cited 05.06.14]. [70] Granade HC, Creyts J, Derkach A. Unlocking energy efficiency in the U.S. economy. McKinsey & Company; 2009. [71] DSIRE. PACE financing. 2008. Available from: http://www.dsireusa.org/solar/ solarpolicyguide/?id¼26.

Please cite this article in press as: Choi J-K, et al., Economic and environmental impacts of community-based residential building energy efficiency investment, Energy (2014), http://dx.doi.org/10.1016/j.energy.2014.10.082