Journal of Cleaner Production 81 (2014) 234e243

Contents lists available at ScienceDirect

Journal of Cleaner Production

journal homepage: www.elsevier.com/locate/jclepro

Sustainability assessment of U.S. final consumption and investments:

triple-bottom-line inputeoutput analysis
Murat Kucukvar a, Gokhan Egilmez b, *, Omer Tatari a
a
Department of Civil, Environmental, and Construction Engineering, University of Central Florida, Orlando, FL 32816, USA
b
Department of Industrial and Manufacturing Engineering, North Dakota State University, Fargo, ND 58102, USA

a r t i c l e i n f o a b s t r a c t

Article history: The U.S. final demand categories such as household consumption, private fixed investments, government
Received 17 May 2013 purchases and investments, and export of goods and services have a wide range of environmental,
Received in revised form economic, and social impacts. Analysis of these impacts, termed as the Triple Bottom Line (TBL), stim-
7 June 2014
ulated a tremendous interest by policy makers over the last decade. Therefore, current research aims to
Accepted 10 June 2014
Available online 27 June 2014
analyze the TBL of U.S. final demands from a systems perspective. To accomplish this goal, the supply and
use tables published by the U.S. Bureau of Economic Analysis are merged with a range of environmental,
economic, and social metrics. The results show that household consumption has the largest indirect TBL
Keywords:
Inputeoutput analysis
sustainability impacts compared to other final demand categories with shares that range between 43%
Triple bottom line and 88%. Industrial sectors including manufacturing, utilities, agriculture, construction, transportation,
Final consumption and mining are generally found to be responsible for the highest impacts for most of the environmental
Sustainability assessment impact categories. Service sectors generally have the highest impacts on the economic and social in-
Cradle to gate life cycle assessment dicators of sustainability. Analysis results also indicate that while meeting the household demand,
agriculture, utilities, and manufacturing sectors have relatively more environmental impacts than their
contributions to gross domestic product (GDP), whereas service sectors contribute to GDP with a higher
share than their environmental burdens. Furthermore, it is envisioned that significant reductions in
environmental footprints of U.S. households can be achieved if environmental policies that aim to reduce
the household consumption are also supported with sustainable growth through greener and resource
efficient economy.
Published by Elsevier Ltd.

1. Introduction problems. In the plan of implementation of this international

The World Summit for Sustainable Development (WSSD) in
Johannesburg identified the sustainable consumption of goods and
services, as well as the protection of social and economic devel-
opment as the primary objectives to promote global sustainability
(WSSD, 2002). They identified the unsustainable way of production
and consumption as the major cause of today's complex ecological
Abbreviations: EIO-LCA, Economic Input Output-based Life Cycle Assessment;
GDP, Gross Domestic Product; GFN, Global Footprint Network; GHG, Greenhouse
Gas; GOS, Gross Operating Surplus; GPI, Genuine Progress Indicator; GTAP, The
Global Trade Analysis Project; MRIO, Multi Region Input Output; P-LCA, Process-
based LCA; TBL, Triple Bottom Line; UNFAO, United Nation's Food and Agriculture
Organization; WSSD, The World Summit for Sustainable Development.
* Corresponding author. Industrial and Manufacturing Engineering, North Dakota
State University, Civil and Industrial Engineering, 202K, 1410 14th Avenue North,
Fargo, ND 58102, USA. Tel.: þ1 701 231 7286; fax: þ1 701 231 7195.
E-mail addresses: gokhan.egilmez@ndsu.edu, gokhanegilmez@gmail.com
(G. Egilmez).
summit, the importance of adoption of effective tools, policies, and
assessment models based on life cycle assessment (LCA) was
highlighted to promote sustainable patterns of production and
consumption, as well as increase the eco-efficiency of products and
services (Hertwich and Peters, 2009). In this regard, LCA gained a
tremendous interest worldwide to quantify the environmental
burdens of production and consumption in a systematic way. LCA
aims to quantify the environmental impacts of products or pro-
cesses from cradle to grave. It primarily consists of goal and scope
definition, life-cycle inventory analysis, life-cycle impact assess-
ment, and interpretation of results (Gradel and Allenby, 2003).
Process-based LCA (P-LCA), economic input output-based LCA (EIO-
LCA), and hybrid LCA are the most widely used LCA approaches in
the literature (Suh and Huppes, 2005).
EIO analysis is a widely used methodology, which was theorized
and developed by Wassily Leontief in the 1970s, based on his earlier
works in the late 1930s, for which he received the Nobel Prize
(Leontief, 1970). The EIO analysis consists primarily of financial

http://dx.doi.org/10.1016/j.jclepro.2014.06.033
0959-6526/Published by Elsevier Ltd.
 M. Kucukvar et al. / Journal of Cleaner Production 81 (2014) 234e243
flows and interdependencies between different sectors of the na-
tions' economy (Suh et al., 2004). In the literature, single and multi-
region EIO models are widely used in order to analyze the envi-
ronmental impacts of consumption activities (Huppes et al., 2008;
Hubacek et al., 2009; Wiedmann, 2009; Steen-Olsen et al., 2012). In
this research, the system boundaries cover all cradle-to-gate envi-
ronmental, economic, and social interventions for U.S. final demand
categories such as household consumption, private fixed in-
vestments, government purchases and investments, and export of
goods and services. Since the primary goal of this research is an
analysis of the total sustainability impacts of U.S. final consumption
and investments at the most detailed level practically achievable
(discerning 426 sectors of U.S economy), we used a single-region
inputeoutput framework which provides the most detailed anal-
ysis for U.S final demands. In this model, products imported into U.S
are assumed to be produced with the same technology as in U.S.
The importance of using highly disaggregated inputeoutput tables
for environmental analysis of production and consumption has
already been discussed in previous studies (Huppes et al., 2008;
Lenzen, 2011).
Environmentally extended inputeoutput analysis has been used
for analyzing the environmental impacts of household demands
(Kok et al., 2006). For example, Cohen et al. (2005) analyzed the
energy requirements of twelve household consumption categories
with different income levels in Brazil. Huppes et al. (2008) inves-
tigated the environmental impacts of consumption in European
Union using an environmentally extended inputeoutput model.
Munksgaard et al. (2008) studied CO2 requirement of eight
household consumption categories in Denmark from 1966 to 1992.
Masanet et al. (2009) developed a carbon footprint estimation
model to analyze the direct and indirect carbon emission associated
with energy, goods, and services consumption of households in
California and the United States. In another research, Hubacek et al.
(2009) studied environmental impacts of urbanization and life style
change in China using inputeoutput analysis in combination with
the ecological and water footprint analysis. Cellura et al. (2011)
developed an environmentally extended inputeoutput model in
order to quantify the energy and environmental impacts related to
the consumption of the Italian households in the period
1999e2006 and to identify the economic sectors involving the
highest impacts.
Triple Bottom Line (TBL) concept focuses on the three main
pillars of sustainability such as environment, economy, and society
(Elkington, 1998). In addition to the environment, social and eco-
nomic dimensions of the sustainability have been integrated with
the EIO analysis in order to develop a comprehensive sustainability
accounting framework. Foran et al. (2005a) applied an approach for
EIO based TBL analysis for environmental, economic, and social
indicators for the industrial sectors of an entire Australian econ-
omy. This model has been named as Balancing Act that integrates
the supply and use tables with environmental, economic, and social
metrics for 135 sectors. Researchers from the University of Sydney
established the foundation of the EIO model for the Balancing Act
study and created a TBL software tool for the Australia, United
Kingdom, and Japan economies. Several studies were conducted
using the TBL version of the EIO by presenting first examples of TBL
accounting in the EIO literature (Foran et al., 2005b; Wiedmann and
Lenzen, 2009; Wiedmann et al., 2009). TBL model of the 426 U.S
economic sectors, named as TBL-LCA, was also created by Kucukvar
and Tatari (2013). This model was initially used to quantify the TBL
implications of the seven U.S. construction sectors. Later on, the
TBL-LCA model was jointly used with multi-criteria decision mak-
ing tools to analyze the sustainability of food manufacturing sectors
and pavement designs in U.S. (Tatari et al., 2012; Egilmez et al.,
2014; Kucukvar et al., 2014). In a recent study, Onat et al. (2014)
235
employed a hybrid approach by integrating TBL inputeoutput
analysis into life cycle sustainability assessment framework for the
U.S. residential and commercial buildings.
Among the TBL sustainability indicators, employment, income,
tax, and work-related injuries were considered as social indicators,
while gross domestic product, gross operating surplus, and imports
were categorized as key economic indicators (Hendrickson et al.,
2005; GTAP, 2008; Wiedmann et al., 2009; Wood and Garnett,
2010; WIOD, 2012). Tax, which is also referred to as government
revenue, is considered as a positive social indicator since collected
tax are used for supporting the national health and education
systems, public transportation, highways, and other civil in-
frastructures (Foran et al., 2005a, 2005b). In addition, the ecological
footprint is measured in terms of global hectares for several land
types such as fishery, grazing, forest land, cropland, built-up land,
and CO2 uptake land (Kitzes et al., 2009). In the literature, the
ecological, energy, water, and carbon footprint indicators are also
considered as a part of the environmental dimension of TBL, and
these indicators have already been used as a measure of environ-
mental sustainability in previous inputeoutput studies (Turner
et al., 2007; Huang et al., 2009; Blackhurst et al., 2010; Ewing
et al., 2012; Steen-Olsen et al., 2012; Egilmez et al., 2013; Galli
et al., 2013; Weinzettel et al., 2013).
The Global Trade Analysis Project (GTAP) database version 8.1
produces an extensive database of trade-linked inputeoutput ta-
bles for the world economy, which involves about 57 sectors and
134 regions in the world (Aguiar, 2013). GTAP is a widely used
database for the modeling of the role of international trade in goods
and services. In recent studies, GTAP database is combined with
environmentally extended inputeoutput analysis in order to
develop global MRIO models. For example, Weber and Matthews
(2008) analyzed American household carbon footprint developing
a MRIO model considering the impacts of international trade.
Ewing et al. (2012) introduced a new MRIO method for conserving
the high degree of product detail for calculating ecological and
water footprints. Steen-Olsen et al. (2012) used a MRIO model to
assess three environmental footprint categories (ecological, water,
and carbon) for the member states of the European Union. Feng
et al. (2011) used a global MRIO model to calculate water foot-
print for China. In other study, Weinzettel et al. (2013) developed a
global MRIO model in order to understand the land utilization by
humans for production and consumption.
As can be seen from environmentally extended inputeoutput
studies reviewed here, the land, carbon, energy, and water footprint
of final consumption have been analyzed from a holistic perspec-
tive. The same inputeoutput methodology can also be utilized to
quantify the TBL sustainability impacts of different final demand
categories by quantifying the economic and social impacts in
addition to environment; however there has been a limited
research considering all of the three dimensions of the sustain-
ability, simultaneously. In this regard, this research has several
novel elements such as:
(i) inclusion of several macro-level socio-economic indicators,
(ii) analysis of consumption and investments using high resolution
U.S. inputeoutput tables with detailed TBL extensions, (iii) inte-
gration of ecological footprint categories (defined by the Global
Footprint Network) with detailed U.S. inputeoutput tables, and (iv)
sustainability analysis of U.S. final demands, including not only
household consumption, but also private fixed investments, gov-
ernment purchases and investments, and export of goods and
services. Consequently, this paper aims to answer the following
research questions:
▪ What are the environmental burdens of final demands in terms
of energy, water, and carbon footprints?
236 M. Kucukvar et al. / Journal of Cleaner Production 81 (2014) 234e243
▪ What are the land use impacts of final demands in terms of
cropland, forestland, fishery land, and grazing land?
▪ What are the contributions of final demand categories to
economy in terms of gross domestic product, gross operating
surplus, and import?
▪ What are the contributions of final demand categories to society
in terms of income, government revenue, and work-related
injuries?
▪ What are the major sectors that contribute the most to the in-
direct TBL sustainability impacts of final demand categories?
2. Methodology
2.1. EIO Framework
In this study, the supply and use tables published by the U.S.
Bureau of Economic Analysis (2002), as part of the International
System of National Accounts, are merged with a range of TBL sus-
tainability metrics to develop a holistic EIO framework for the U.S.
final demand categories. We used a single region industry-by-
commodity model in which imported commodities are assumed to
be produced with domestic technology. Similar modeling approach
and domestic technology assumption were also employed for
analyzing the ecological footprints of U.K's final consumption
(Wiedmann et al., 2006). In this model, the inputoutput multipliers
represent the total TBL impacts based on per unit of final demand of
commodities produced by industrial sectors. This format is utilized
since the basic inputeoutput model presents the financial flows be-
tween industrial sectors without distinguishing between primary and
secondary products. By using this format, it is possible to accounts for
the fact that an industry can produce more than one commodity, such
as secondary products and by-products (Wachsmann et al., 2009).
The Eurostat manual provides a comprehensive and detailed discus-
sion on the use of this framework (Eurostat, 2008).
In this model, the Use matrix, which is usually represented by U,
gives information on the consumption of commodities by in-
dustries or by final demand categories, such as households, gov-
ernment, investment or export. As a component of U, uij denotes the
value of commodity purchase of commodity i by industry j and xj
represent the total output of industry j including imports. There-
fore, bij is the amount of commodity i required for producing one-
dollar output of industry j. By using the total industrial output of
industry j, the technical coefficient matrix B can be written as
(Miller and Blair, 2009):
" #

 uij
B ¼ bij ¼
xj
In addition, the Make matrix, which is denoted as V, provides
detailed information on production of commodities by industries. In
the make table, each row represents the production of commodity by
different industries. As an element of V, vji is the value of the output of
commodity i by industry j and qi accounts for the total (domestic plus
imported) output of commodity i. Therefore, dji represents the frac-
tion of total commodity i output which is produced by industries,
both as main product as well as by-product. Using the total output of
commodity i, the industry-based technology coefficient matrix D can
be written as (Miller and Blair, 2009):
 

 vji
D ¼ dji ¼
qi
After defining B and D, the industry-by-commodity
inputeoutput model can be formulated as follows (Miller and
Blair, 2009):
h i
k ¼ DðI  BDÞ1 f (3)
where k represents the total industry output vector, I is the identity
matrix, and f refers to the total final demand vector for commod-
ities. In addition, B is the input requirements for products per unit
of output of an industry matrix, and D is sometimes called as
market-share matrix. Also, the term [D(I-BD)¡1] represents the
total requirement matrix, which is also known as the Leontief in-
verse. Also, BD is the direct requirement matrix, which is repre-
sented by A matrix in Leontief's model (Leontief, 1970). After
industry-by-commodity inputeoutput framework has been estab-
lished, indirect sustainability impacts can be quantified by multi-
plying the final demand of a sector with the multiplier matrix.
Then, a vector of total sustainability impacts can be formulated as
follows:
h i
r ¼ Edir k ¼ Edir DðI  BDÞ1 f (4)
where r denotes the total impacts vector which represents overall
sustainability impacts per unit of final demand, and Edir matrix
consists of the direct environmental, economic, and social impact
values per dollar of output for each industrial sector. Each element
of this matrix is calculated by dividing the total direct sectoral
impact (e.g. carbon footprint, income, tax) with the total economic
output of that sector. In addition, the product of Edir and bracketed
term [D(I-BD)¡1] represents the multiplier matrix.
2.2. Sustainability indicators
This research utilizes the aforementioned inputeoutput model
to build a TBL sustainability accounting framework of the U.S.
economy using numerous environmental, economic, and social
indicators. These indicators are described in Table 1 with their
corresponding units and data sources.
3. Sustainability assessment of final demand categories
After determining the aforementioned sustainability assess-
ment metrics, we quantify the indirect sustainability impacts of the
U.S. final demand categories from environmental, economic, and
social perspectives. The government-related final demand cate-
gories, including government defense consumption and invest-
ment, government non-defense consumption and investments, and
state and local government education consumption and in-
vestments are analyzed under the government purchases and in-
(1)
vestments. The economic values of each final demand category are
obtained from the Use table of 2002 benchmark inputeoutput ac-
counts (BEA, 2002). First, indirect TBL sustainability impacts of
different final demand categories are calculated by using proposed
EIO model. These impacts are then grouped under 15 main U.S.
sectors based on their relative contributions. The following sub-
sections discuss the environmental, economic, and social impacts
with details.
3.1. Environmental impacts
This subsection presents the environmental impacts of each
final demand category in terms of energy consumption, GHG
(2) emissions, water withdrawal, and ecological footprint.
3.1.1. Energy consumption
The energy consumption of each sector is calculated by sum-
ming the energy content of different fossil fuels and electricity from
 M. Kucukvar et al. / Journal of Cleaner Production 81 (2014) 234e243 237

Table 1
Summary of the sustainability indicators.

TBL indicator Unit Description Data source

Environmental

GHG Emissions t CO2-eqv The total greenhouse gas (GHG) emissions of each sector in terms of metric tons (CMU, 2008)
of CO2 equivalent.
3
Water Withdrawal m The total amount of water withdrawals of each sector. (CMU, 2008)
Energy consumption TJ The total energy (fossil plus electricity) consumption of each sector. (CMU, 2008)
Fishery gha The estimated primary production required to support the fish caught. (GFN, 2010)
Grazing gha The amount of livestock feed available in a country with the amount of feed (GFN, 2010)
required for the livestock produced.
Forest Land gha The amount of lumber, pulp, timber products, and fuel wood consumed by a (GFN, 2010)
country on a yearly basis.
Cropland gha The most bio-productive of all the land use types and includes areas used to (GFN, 2010)
produce food and fiber for human consumption.
Economic
Gross Operating Surplus (GOS) $M The capital available to corporations, which allows them to repay their creditors, (BEA, 2002)
to pay taxes and to finance their investments.
Gross Domestic Product (GDP) $M The market value of goods and services produced within the country in a given (BEA, 2002)
period of time.
Import $M The value of goods and services purchased from foreign countries to produce (BEA, 2002)
domestic commodities by industries.
Social
Tax $M The government revenue, which includes the taxes on production and imports. (BEA, 2002)
Income $M The compensation of employees, including wages and salaries (BEA, 2002)
Injuries employee The total number of non-fatal injuries related to industrial places. (BLS, 2002)

non-fossil sources. The results show that household category is
responsible for the highest amounts of total indirect energy with
the value of 5.17  107 TJ. This final demand category is followed by
government purchases and investments and private fixed in-
vestments, respectively (see Table 2). Within the main industrial
sectors, utilities, manufacturing, transportation, and mining have
the highest contributions to the net energy consumption related to
household consumption. In addition, manufacturing and utilities
sectors are found to be responsible for the highest energy con-
sumption associated with private fixed investments and export of
goods and services. For government purchases and investments,
direct energy use from the government industries has the highest
values compared to other major industries (see Fig. 1).
compared to other final demand categories. This final demand
category is followed by export of goods and services with water
withdrawal values of 8.28  1010 m3 (see Table 2). For household
consumption and private fixed investments, agriculture and utili-
ties sectors are found to be responsible for over 90% of the total
water withdrawal. In addition, about 75% of total water consump-
tion can be attributed to agriculture and utilities sectors related to
private fixed investments. On the other hand, for government
purchases and investments, these sectors are responsible for
approximately 80% of total water withdrawals, while around 20% of
total water is directly consumed by the government industry (see
Fig. 1).

3.1.4. Cropland
3.1.2. GHG Emissions
The cropland represents the most bio-productive of all the land
The emissions of carbon dioxide (CO2), nitrous oxides, methane,
use types and includes areas used to produce food and fiber for
and fluorinated gases from fossil fuel combustion (coal, natural gas,
human consumption, feed for livestock, crops, and rubber. Based on
petroleum, etc.) are also quantified for each sector, and total GHG
the analysis results, household consumption utilizes the largest
emissions are expressed in terms of metric tons of CO2 equivalents
amount of cropland in comparison with other final demand cate-
(CMU, 2008). The results indicate that household consumption
gories. This final demand category is followed by export of good and
results in the highest indirect GHG emissions with the value of
services with cropland consumption value of 4.42  107 (see
4.09  109 t CO2-eqv. This final demand category is followed by
Table 2). For household consumption and private fixed in-
government purchases and investments and private fixed in-
vestments, agriculture and manufacturing sectors are found to be
vestments with GHG emission values of 9.07  108 and 8.16  108,
responsible for over 95% of the total cropland use. Furthermore, the
respectively (see Table 2). For GHG emissions related to household
consumption, industrial sectors, such as utilities, manufacturing,
agriculture, transportation and mining show the highest percent- Table 2
age contributions in comparison to other U.S. sectors, as shown in Environmental impacts of U.S. final demand categories.
Fig. 1. In addition, manufacturing, utilities, construction, trans- Environmental Units HC PFI EGS GPI
portation, and mining sectors are also found to be responsible for indicators
the highest GHG emissions related to private fixed investments. For
Energy TJ 5.17  107 1.19  107 9.00  106 1.28  107
government purchases and investments, direct emissions from Consumption
government industry plus indirect emissions from manufacturing GHG Emissions t CO2-eqv 4.09  109 8.16  108 6.62  108 9.07  108
and utilities make up to 80% of net carbon footprint of this final Water m3 4.04  1011 3.89  1010 8.28  1010 4.04  1010
Withdrawal
demand category.
Fishery gha 1.39  107 3.31  105 5.42  105 1.02  106
Grazing gha 3.89  107 5.76  105 4.90  106 2.37  106
3.1.3. Water withdrawal Forest Land gha 2.01  107 1.46  107 5.79  106 5.54  106
The water withdrawal associated with U.S. final demands are Cropland gha 2.90  108 1.84  107 4.42  107 1.93  107
also analyzed. The findings of our research show that household HC: Household Consumption; PFI: Private Fixed Investments; EGS: Export of Goods
consumption utilizes the highest amounts of indirect water and Services; GPI: Government Purchases and Investments.
238 M. Kucukvar et al. / Journal of Cleaner Production 81 (2014) 234e243

Fig. 1. Percentage contribution of U.S. industries to energy consumption, GHG emissions, and water withdrawal categories.

construction industry contributes to approximately 50% of total 3.1.5. Grazing land

cropland use for private fixed investments category. When we look The grazing land footprint is calculated by comparing the
at the government purchases and investments related cropland amount of livestock feed available in the country with the amount
use, agriculture, manufacturing, government, and construction in- of feed required for the livestock produced in that year, with the
dustries account for 90% of total cropland use (see Fig. 2). remainder of feed demand assumed to come from grazing land. The

Fig. 2. Percentage contribution of U.S. industries to ecological footprint categories.

 M. Kucukvar et al. / Journal of Cleaner Production 81 (2014) 234e243 239

Table 3 fish catch for direct human consumption and catch for fish meal
Economic impacts of U.S. final demand categories. (Kitzes et al., 2009). From Table 2, it is apparent that fishery land
Economic Units HC PFI EGS GPI footprint of household consumption is quite larger than the other
indicators final demand categories. After household demand, government
GOS $M 3.0  106 5.38  105 3.02  105 4.62  105 purchases and investments is the second largest final demand
GDP $M 7.4  106 1.67  106 7.99  105 1.95  106 category in terms of total fishery land use. As indicated in Fig. 2,
Imports $M 1.49  106 4.53  105 3.15  105 1.64  105 manufacturing and agriculture represent two main industrial sec-
tors with the highest contributions to overall fishery land footprint.
grazing land footprint results indicate that household demand In general, manufacturing and arts, entertainment, recreation, ac-
consumes the largest amount of grazing land compared to other commodation, and food services are two sectors with the highest
final demand types. This category is then followed by export of contributions to fishery land consumption. For example,
goods and services and government purchases and investments, manufacturing industry consumes nearly 80% of total fishery land
respectively (see Table 2). For all the final demand categories, related to export of goods and services, and the rest of fishery land
manufacturing and agriculture sectors dominate the total grazing is primarily utilized by arts, entertainment, recreation, accommo-
land consumption, and account approximately for 100% of overall dation, and food services sectors.
grazing land footprint, as indicated in Fig. 2.
3.2. Economic impacts
3.1.6. Forestland
The forestland footprint is calculated based on the amount of This subsection presents the economic impacts of each final
lumber, pulp, timber products, and fuel wood consumed by a demand category in terms of GOS, GDP, and imports.
country on a yearly basis. Household consumption is responsible
for the highest forestland use with value of 2.01  107 gha (see 3.2.1. GOS
Table 2). In addition, private fixed investments represent the final GOS, also known as profit, is a positive economic indicator. As
demand category, which has a second highest forestland con- shown in Table 3, household consumption has generated the
sumption value. When we analyzed the percentage contribution of highest amount of GOS compared to other final demands, and
each major industrial sector to overall forest land use, we see that finance, insurance, real estate, and rental accounts for 50% of total
manufacturing and agriculture sectors account approximately for GOS generated by this final demand category. In addition, private
100% of total forestland footprint for all final demand categories fixed investments have generated second largest GOS values
(see Fig. 2). among final demand types. As indicated in Fig. 3, manufacturing,
construction, and finance, insurance, real estate, and rental sectors
3.1.7. Fishery land have played a significant role in this result.
The fishery land footprint is calculated by using estimates of the
maximum sustainable catch for a variety of fish species. Global 3.2.2. GDP
Footprint Network (GFN) used the catch data from the United Na- Contribution of each sector to GDP is assumed to be as a positive
tion's Food and Agriculture Organization (FAO) which is used to economic indicator since it monitors the health of a nation's
estimate demand on fishing grounds. Current accounts track both economy and includes wages and salaries, indirect taxes less

Fig. 3. Percentage contribution of U.S. industries to economic impact categories.

240 M. Kucukvar et al. / Journal of Cleaner Production 81 (2014) 234e243

Table 4 can be seen from Fig. 3, the percentage contributions of other major
Social impacts of U.S. final demand categories. sectors to imports are found to be negligible in comparison with
Social Units HC PFI EGS GPI manufacturing and mining sectors.
indicators

Income $M 3.79  106 1.05  106 4.56  105 1.44  106 3.3. Social impacts
Tax $M 6.06  105 8.94  104 4.22  104 4.50  104
Injuries employee 4.51  106 1.33  106 5.56  105 6.80  105
This subsection presents the social impacts of each final demand
category in terms of income, tax, and work-related injuries.

subsidies, and gross operation surplus. In the production approach,

the total output of a particular industry minus its total intermediate
use equals to the contribution of that industry to GDP (Eurostat,
2008; Lenzen and Dey, 2002). According to analysis results,
household consumption has the highest contribution to GDP with
value of 7.40Eþ06 $M (see Table 3). Especially, wholesale trade
sector has contributed highly on the total GDP generated by
household demand. For all final demand categories, U.S. sectors,
which have the highest contributions to GOS, also have the largest
impacts on the GDP (see Fig. 3). Another important result is that
government industries are responsible for over 50% of total GDP
generated by the final demand category of government purchases
and investments.
3.2.3. Import
Import, which represents the value of goods and services pur-
chased from foreign countries to produce domestic commodities by
industries, is also selected as another indicator. Import can be
considered as a negative economic indicator since excess of imports
can increase the current deficit through the flow of money out of
the country. The final demand of U.S. households results in the
highest import values when compared to other categories (see
Table 3). Especially, manufacturing industry has dominantly
contributed to imports for all final demand categories. This industry
is then followed by mining and agriculture sectors, respectively. As
3.3.1. Income
Income is a positive social indicator and represents the general
compensation of employees including wages and salaries and
worker's compensation payments. Analysis results show that
household consumption generates the largest amount of income in
U.S. compared to other final demand categories (see Table 4).
Among the U.S. sectors, service sectors are found to be responsible
for over 80% of total income, which is generated by households. For
private fixed investments, manufacturing and construction sectors
have supported nearly 50% of overall income. Based on the gov-
ernment purchases and investments related income generation,
government industries are found to account for around 60% of total
income, and the rest is obtained from other indirect sectors (see
Fig. 4).
3.3.2. Tax
In addition to the income, tax which is referred to as govern-
ment revenue is chosen as another positive social indicator. First,
household consumption produces the highest amount of tax with
economic value of 6.06  105 $M. After household demand, private
fixed investments is found to be the second largest final demand
category in terms of contributions to tax (see Table 4). Based on
Fig. 4, it can be seen that the service sectors have the highest share
in tax for all final demand types. In addition, it should be noted that
government and agriculture industries have negative contribution

Fig. 4. Percentage contribution of U.S. industries to social impact categories.

 M. Kucukvar et al. / Journal of Cleaner Production 81 (2014) 234e243 241

Table 5
Normalized sustainability impacts of U.S. final demand categories.

TBL indicators Units HC PFI EGS GCI

Economic
GOS $M/$M 4.02  101 3.18  101 3.72  101 2.34  101
GDP $M/$M 9.91  101 9.87  101 9.84  101 9.82  101
Imports $M/$M 2.00  101 2.68  101 3.88  101 8.29  102
Social
Income $M/$M 5.08  101 6.21  101 5.61  101 7.30  101
Tax $M/$M 8.12  102 5.29  102 5.20  102 2.28  102
Injuries Employee/$M 6.04  101 7.86  101 6.85  101 3.44  101
Environmental
Energy consumption TJ/$M 6.92 7.04 11.1 6.47
GHG emissions t CO2-eqv/$M 5.36  102 4.82  102 8.15  102 4.60  102
Water Withdrawal m3/$M 1.43  104 6.09  103 2.70  104 5.41  103
Fishery gha/$M 1.86 1.96  101 6.67  101 5.17  101
Grazing gha/$M 5.21 3.41  101 6.03 1.20
Forest Land gha/$M 2.69 8.63 7.13 2.81
Cropland gha/$M 38.8 10.9 54.4 9.80

to the government revenue, since some of the government and
agriculture sectors show a high dependency on government sub-
sidies, which have resulted in negative tax values, as indicated in
Fig. 4.
3.3.3. Work-related injuries
Finally, as a negative social indicator, the number of total work-
related injuries is analyzed in order to investigate the contribution
of U.S. final demands to injuries. The analysis results show that
household consumption is responsible for the highest numbers of
work-related injuries with the value of 4.51  106. After households,
private fixed investments are found to be the second highest in
terms of injuries (see Table 4). For private fixed investments and
government purchases and investments, manufacturing and con-
structions sectors account approximately for over 50% of total work
place injuries. Furthermore, for export of goods and services,
almost half of the work-related injuries are caused by the
manufacturing industry (see Fig. 4).
4. Normalized TBL impacts of the U.S. final demands
In section 3, TBL impacts are presented. Based on analysis re-
sults, household consumption is found to have the largest impacts
for all impact categories. Table 5 shows the normalized TBL im-
pacts of each final demand category based on per $M economic
output. Overall, export of goods and services shows higher amount
of environmental impacts per $M output in comparison with other
categories. When we look at the social indicators, government
purchases and investments produce the largest amount of income,
while household consumption generates the highest amount of
government revenue per $M economic output. On the other hand,
private fixed investments are responsible for the highest number
of work-related injuries compared to others. For economic in-
dicators, household demand has the highest contribution to GDP
and GOS based on per $M economic activity. Export of goods and
services results in higher imports than any other final demand
category. Therefore, it is important to note that although house-
hold consumption is a dominant final demand category in terms of
overall TBL impacts, other final demand categories are also able to
show higher TBL impacts based on per $M economic activity (see
Table 5).
5. Conclusion and recommendations
This study proposes a comprehensive TBL sustainability
assessment framework which includes the social, economic, and
environmental impacts from a broader perspective using several
macro-level indicators. Due to the broader scope of analysis, EIO
analysis is utilized in order to provide a holistic framework to trace
the impacts across the supply chains related to U.S. final demands.
Furthermore, since recent trends also emphasize the inclusion of
three pillars of sustainability as economy, society and the envi-
ronment, the proposed sustainability scope fits to the needs of such
a comprehensive sustainability assessment understanding (Guinee
et al., 2010).
The findings of this study are critical in helping to propose
sound policies for sustainable consumption policy, and can be used
to give suggestions to decrease the amount of environmental im-
pacts. The results show that household consumption causes the
largest overall environmental impacts in the U.S. To give an
example, based on U.S. household consumption, utilities represent
the most significant industrial sector in terms of GHG emissions.
Especially, power generation and supply sector has the largest
share among the utilities sectors in respect to total GHG emissions.
Although there is an increasing trend in household energy de-
mands with increasing population, a higher utilization of renew-
able energy sources in electricity grid mix can be good for reducing
the net carbon footprint of household consumption. Moreover,
manufacturing industry is responsible of high energy utilization,
and therefore energy efficiency improvement in industrial pro-
cesses is necessary and several important energy efficient clean
production strategies are discussed in a research conducted by Dovì
et al. (2009). In addition, agriculture and utilities sectors are
responsible for high water footprints, and consume more than 90%
of overall water related to household demand. In particular, power
generation and supply sector primarily consumes water for cooling
purposes (Blackhurst et al., 2010). Increasing efficiency of cooling
technologies and water-conserving agriculture methods can
significantly minimize the water footprint of household demand for
future. Therefore, it is important to note that significant reductions
in environmental footprints of households can be achieved if
environmental polices aiming to reduce the household consump-
tion are also supported by technological improvements in indus-
trial facilities.
Analysis results also show that for household consumption, the
environmental impact share of manufacturing industries are
higher than their contribution to GDP, whereas most of the service
sectors have less environmental burdens than their contributions
to GDP. For example, percentage share of water consumption of
agriculture sectors are 37 times higher than their share in GDP. In
addition, utilities sector's water withdrawal share is found to be
21.5 times higher than its GDP contribution. For energy use,
242 M. Kucukvar et al. / Journal of Cleaner Production 81 (2014) 234e243
percentage share of manufacturing, utilities, and agriculture sec-
tors are found be two times higher than their share in GDP. In
terms of carbon emissions, utilities sector is a dominant sector, and
found to have 45 times higher contribution to GHG emissions than
GDP. On the other hand, service sectors, such as information,
finance, and education have less environmental impacts against
their contributions to GDP. Therefore, it was likely to conclude that
production industries, including agriculture, utilities, and
manufacturing have environmental impacts, which are not pro-
portional to their GDP share.
Although GDP has been widely used to monitor the economic
health of nations, it fails to reflect the cost of negative effects related
to economic growth such as cost of environmental pollution, the
depletion of nonrenewable resources, income inequality, etc. As an
alternative metric to measure the economic prosperity, Genuine
Progress Indicator (GPI) is able to evaluate the economic perfor-
mance and human well-being simultaneously and shows the de-
ficiencies of GDP in measuring the economic performance. For this
reason, GPI can be used to account for the social, economic, and
environmental sustainability while considering the positive and
negative consequences of the economic growth related to nations'
increasing consumption (Wen et al., 2007).
To have a better insight into the sustainability effects of final
demand categories, life-cycle investigations should go beyond
traditional P-LCA and EIO-LCA, which are generally used for
analyzing the environmental impacts, mainly energy, water, car-
bon, and land footprints. To accomplish this goal, we extended the
scope of the EIO methodology by adding several prominent sus-
tainability assessment metrics to answer the questions regarding
sustainable consumption and investments in U.S. For future studies,
an integrated approach of goal programming and EIO model can
provide alternative solutions for optimizing the TBL sustainability
implications of different final demand categories. Miller and Blair
(2009) developed a policy programming model in which a goal
programming based multi-criteria optimization model is combined
with the inputeoutput analysis to optimize the multiple environ-
mental, economic and social objectives, simultaneously. We believe
that using integrated approaches can provide sound policies
regarding the optimum output of different sectors based on final
consumption and investments. In terms of long term policy mak-
ing, system dynamics (SD) modeling is a crucial future research
direction, which can provide insights about mid and long term
policy making (Egilmez and Tatari, 2012). Therefore, integration of
EIO analysis and SD models is among the future research goals of
the authors.
Although this paper utilized detailed supply and use tables
which include 426 economic sectors, current results have signifi-
cant uncertainties related to using domestic technology assump-
tion for imported products. Therefore, environmental impact
results might be different due to the fact that imported products are
produced with foreign technology. In addition, production of these
materials may require less or more amounts of ecological resources.
For future studies, to have a trade-linked EIO model, a global MRIO
framework can be developed in order to account for the impacts of
international trade in a way that environmental, economic, and
social implications of the U.S. final demands will account for the
technological differences related to production of imported mate-
rials. The importance of applying MRIO models can be found in the
literature (Hertwich and Peters, 2009; Tukker and Dietzenbacher,
2013; Weinzettel et al., 2014).
References
Aguiar, A., 2013. Report on the Regional Input-Output Tables in the GTAP Data Base
Version 8.1.
Blackhurst, B.M., Hendrickson, C., Vidal, J.S.I., 2010. Direct and indirect water
withdrawals for U.S. industrial sectors. Environ. Sci. Technol. 44 (6), 2126e2130.
http://dx.doi.org/10.1021/es903147k.
Bureau of Economic Analysis, 2002. Benchmark Input-output data. Retrieved from.
http://www.bea.gov/industry/io_benchmark.html.
Carnegie Mellon University Green Design Institute, 2008. Economic Input-Output
Life Cycle Assessment (EIO-LCA). US 2002 Industry Benchmark model
[Internet], Available from: http://www.eiolca.net (accessed 5.02.13.).
Cellura, M., Longo, S., Mistretta, M., 2011. The energy and environmental impacts of
Italian households consumptions: an inputeoutput approach. Ren. Sust. Energ.
Rev. 15 (8), 3897e3908.
Cohen, C., Lenzen, M., Schaeffer, R., 2005. Energy requirements of households in
Brazil. Energ. Policy. 33, 555e562.
Dovì, V.G., Friedler, F., Huisingh, D., Klemes, J.J., 2009. Cleaner energy for sustainable
future. J. Clean. Prod. 17 (10), 889e895.
Egilmez, G., Tatari, O., 2012. A dynamic modeling approach to highway sustain-
ability: strategies to reduce overall impact. Transp. Res. Part A: Policy Pract. 46
(7), 1086e1096.
Egilmez, G., Kucukvar, M., Tatari, O., 2013. Sustainability assessment of U.S.
manufacturing sectors: an economic input output-based frontier approach.
J. Clean. Prod. 53, 91e102.
Egilmez, G., Kucukvar, M., Tatari, O., Bhutta, M.K.S., 2014. Supply chain sustainability
assessment of the US food manufacturing sectors: a life cycle-based frontier
approach. Res. Conser. Recyc. 82, 8e20.
Elkington, J., 1998. Partnerships from cannibals with forks: the triple bottom line of
21st-century business. J. Environ. Qual. Manage. 8 (1), 37e51.
Eurostat, 2008. Eurostat Manual of Supply, Use and Input-Output Tables. European
Communities, Luxembourg.
Ewing, B.R., Hawkins, T.R., Wiedmann, T.O., Galli, A., Ertug Ercin, A., Weinzettel, J.,
Steen-Olsen, K., 2012. Integrating ecological and water footprint accounting in a
multi-regional inputeoutput framework. Ecol. Ind. 23, 1e8.
Feng, K., Chapagain, A., Suh, S., Pfister, S., Hubacek, K., 2011. Comparison of bottom-
up and top-down approaches to calculating the water footprints of nations.
Econ. Syst. Res. 23 (4), 371e385.
Foran, B., Lenzen, M., Dey, C., 2005a. Balancing act a triple bottom line analysis of the
australian economy volume 1. In: Csiro (Ed.), Balancing Act, 358. CSIRO, p. 277.
Foran, B., Lenzen, M., Dey, C., Bilek, M., 2005b. Integrating sustainable chain man-
agement with triple bottom line accounting. Ecol. Econ. 52 (2), 143e157.
Galli, A., Weinzettel, J., Cranston, G., Ercin, E., 2013. A footprint family extended
MRIO model to support Europe's transition to a one planet economy. Sci. Tot.
Environ. 461, 813e818.
GFN, 2010. National Footprint Accounts: Ecological Footprint and Bio-capacity.
Retrieved February 15, 2013, from. http://www.footprintnetwork.org/en/index.
php/GFN/page/footprint_for_nations.
Gradel, T., Allenby, B., 2003. Industrial Ecology and Sustainable Engineering, third
ed. Prentice Hall, New Jersey.
GTAP, 2008. Global Trade Analysis Project. Version 7. Department of Agricultural
Economics, Purdue University. US. Retrieved from. https://www.gtap.agecon.
purdue.edu/databases/default.asp.
Guinee, J.B., Heijungs, R., Huppes, G., Zamagni, A., Masoni, P., Buonamici, R.,
Rydberg, T., 2010. Life cycle assessment: past, present, and future. Environ. Sci.
Technol. 45 (1), 90e96.
Hendrickson, C., Lave, L.B., Matthews, H., 2005. Environmental Life Cycle Assess-
ment of Goods and Services: an Input-output Approach, first ed. Washington
DC: Resources for the Future.
Hertwich, E.G., Peters, G.P., 2009. Carbon footprint of nations: a global, trade-linked
analysis. Environ. Sci. Technol. 43 (16), 6414e6420.
Huang, Y.A., Lenzen, M., Weber, C.L., Murray, J., Matthews, H.S., 2009. The role of
inputeoutput analysis for the screening of corporate carbon footprints. Econ.
Syst. Res. 21 (3), 217e242.
Hubacek, K., Guan, D., Barrett, J., Wiedmann, T., 2009. Environmental implications
of urbanization and lifestyle change in China: ecological and water footprints.
J. Clean. Prod. 17 (14), 1241e1248.
Huppes, G., Koning, A., Suh, S., Heijungs, R., Van Oers, L., Guinee, J.B., 2008. Envi-
ronmental impacts of consumption in the European Union: high-resolution
input-output tables with detailed environmental extensions. J. Ind. Ecol. 10,
129e146.
Kitzes, J., Galli, A., Bagliani, M., Barrett, J., Dige, G., Ede, S., Wiedmann, T., 2009.
A research agenda for improving national ecological footprint accounts. Ecol.
Econ. 68 (7), 1991e2007.
Kok, R., Benders, R., Moll, H.C., 2006. Measuring the environmental load of
household consumption using some methods based on inputeoutput energy
analysis: a comparison of methods and a discussion of results. Energ. Policy 34
(17), 2744e2761.
Kucukvar, M., Tatari, O., 2013. Towards a triple bottom-line sustainability assess-
ment of the U.S. construction industry. Int. J. Life. Cycl. Assess. 18 (5), 958e972.
Kucukvar, M., Noori, M., Egilmez, G., Tatari, O., 2014. Stochastic decision modeling
for sustainable pavement designs. Int. J. Life. Cycl. Assess., 1e15.
Lenzen, M., 2011. Aggregation versus disaggregation in inputeoutput analysis of the
environment. Econ. Syst. Res. 23 (1), 73e89.
Lenzen, M., Dey, C.J., 2002. Economic, energy and greenhouse emissions impacts of
some consumer choice, technology and government outlay options. Energy
Econ. 24 (4), 377e403.
Leontief, W., 1970. Environmental repercussions and the economic structure: an
input-output approach. Rev. Econ. Stat. 52 (3), 262e271.
 M. Kucukvar et al. / Journal of Cleaner Production 81 (2014) 234e243
Masanet, E., Kramer, K.J., Homan, G., Brown, R., Worrell, 2009. Assessment of
Household Carbon Footprint Reduction Potentials, p. 28.
Miller, R., Blair, P., 2009. Input-output Analysis: Foundations and Extensions.
Cambridge University Place.
Munksgaard, J., Wier, M., Lenzen, M., Dey, C., 2008. Using input-output analysis to
measure the environmental pressure of consumption at different spatial levels.
J. Ind. Ecol. 9 (1e2), 169e185.
Onat, C., Kucukvar, M., Tatari, O., May 2014. Integrating triple bottom line input-
output analysis into life-cycle sustainability assessment: the case for U.S
buildings. Int. J. Life. Cycl. Assess., 1e18.
Steen-Olsen, K., Weinzettel, J., Cranston, G., Ercin, A.E., Hertwich, E.G., 2012. Carbon,
land, and water footprint accounts for the european Union: consumption,
production, and displacements through international trade. Env. Sci. Tech. 46,
10883e10891.
Suh, S., Huppes, G., 2005. Methods for life cycle inventory of a product. J. Clean.
Prod. 13 (7), 687e697.
Suh, S., Lenzen, M., Treloar, G.J., Hondo, H., Horvath, A., Huppes, G., Norris, G., 2004.
System boundary selection in life-cycle inventories using hybrid approaches.
Environ. Sci. Technol. 34 (3), 657e664.
Tatari, O., Nazzal, M., Kucukvar, M., 2012. Comparative sustainability assessment of
warm-mix asphalts: a thermodynamic based hybrid life cycle analysis. Res.
Cons. Recycl 58, 18e24.
Tukker, A., Dietzenbacher, E., 2013. Global multiregional inputeoutput frameworks:
an introduction and outlook. Econ. Syst. Res. 25 (1), 1e19.
Turner, K., Lenzen, M., Wiedmann, T., Barrett, J., 2007. Examining the global envi-
ronmental impact of regional consumption activities e part 1: a technical note
on combining input-output and ecological footprint analysis. Ecol.Econ. 62,
37e44.
Wachsmann, U., Wood, R., Lenzen, M., Schaeffer, R., 2009. Structural decomposition
of energy use in Brazil from 1970 to 1996. App. Energ. 86, 578e587.
243
Weber, C.L., Mathews, H.S., 2008. Quantifying the global and distributional aspects
of American household carbon footprint. Ecol.Econ. 66, 379e391.
Weinzettel, J., Hertwich, E.G., Peters, G.P., Steen-Olsen, K., Galli, A., 2013. Affluence
drives the global displacement of land use. Glob. Env. Chang. 23 (2), 433e438.
Weinzettel, J., Steen-Olsen, K., Hertwich, E.G., Borucke, M., Galli, A., 2014. Ecological
footprint of nations: comparison of process analysis, and standard and hybrid
multiregional inputeoutput analysis. Ecol. Econ. 101, 115e126.
Wen, Z., Zhang, K., Du, B., Li, Y., Li, W., 2007. Case study on the use of genuine
progress indicator to measure urban economic welfare in China. Ecol. Econ. 63
(2), 463e475.
Wiedmann, T., Minx, J., Barrett, J., Wackernagel, M., 2006. Allocating ecological
footprints to final consumption categories with inputeoutput analysis. Ecol.
Econ. 56 (1), 28e48.
Wiedmann, T., 2009. A review of recent multi-region inputeoutput models used for
consumption-based emission and resource accounting. Ecol. Econ. 69 (2),
211e222.
Wiedmann, T., Lenzen, M., 2009. Unravelling the Impacts of Supply ChainsdA New
Triple-Bottom-Line Accounting Approach and Software Tool….Management
Accounting for Cleaner Production.
Wiedmann, T.O., Lenzen, M., Barrett, J.R., 2009. Companies on the Scale: comparing
and benchmarking the sustainability performance of businesses. J. Ind. Ecol. 13
(3), 361e383.
Wood, R., Garnett, S., 2010. Regional sustainability in Northern AustraliadA
quantitative assessment of social, economic and environmental impacts. Ecol.
Econ. 69 (9), 1877e1882.
WIOD, 2012. World Input-output Database: Construction and Applications. FP7
Research Project, 2009-2012. Retrieved from. www.wiod.org.
World Summit on Sustainable Development, 2002. Plan Of Implementation: the
action plan. Johannesburg, South Africa. Retrieved from. http://www.
worldsummit2002.org/index.htm. Johannesburg, South Africa.