Energy Policy 46 (2012) 460–472

Contents lists available at SciVerse ScienceDirect

Energy Policy

0301-42

http://d

n Corr

E-m

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

Overcoming energy efficiency barriers through systemsapproach—A conceptual framework

Kah-Hin Chai a,n, Catrina Yeo b

a Department of Industrial and Systems Engineering, National University of Singapore, Singaporeb Energy Studies Institute, National University of Singapore, Singapore

a r t i c l e i n f o

Article history:

Received 1 April 2011

Accepted 9 April 2012Available online 1 May 2012

Keywords:

Energy efficiency

Systems thinking

Barriers

15/$ - see front matter & 2012 Elsevier Ltd. A

x.doi.org/10.1016/j.enpol.2012.04.012

esponding author. Tel.: þ65 6516 2250; fax:

ail address: iseckh@nus.edu.sg (K.-H. Chai).

a b s t r a c t

In this paper we propose a framework which categorizes energy efficiency barriers based on the stage

at which the barriers exist. Barriers to energy efficiency have been widely studied but to our

knowledge, except for a few studies, we found inadequate consideration for barrier–barrier interactions

when proposing policy measures for improving energy efficiency. Leveraging systems thinking’s power

as a problem solver which identifies underlying structure that explains (similar) patterns of behavior in

a variety of different situations, we attempted to identify patterns of barriers to adoption of energy

efficiency measures in industrial companies. Inspired by systems thinking, the proposed framework has

four stages, namely, Motivation, Capability, Implementation and Results, as well as a feedback loop.

Using a case study, we show that following the four stages will lead to positive feedback for future

energy efficiency implementations. The framework highlights the interconnected nature of the barriers

and a need for policymakers to address these barriers in a holistic manner. We argue that the overall

effectiveness of energy efficiency policies is only as strong as the weakest link in the four-stage

framework. This differs from most prior research that addressed barriers in isolation, where a solution

is proposed for each of the barriers without considering the relationship between the barriers. Our

framework also offers a way to understand the roles and responsibilities of major stakeholders such as

governments and energy service companies (ESCOs) in driving energy efficiency. This allows the

assessment and identification of weak links in energy efficiency policies.

& 2012 Elsevier Ltd. All rights reserved.

1. Introduction

Industrial energy use accounts for approximately one-third ofthe world’s energy demand. In particular, the 1970s oil crises sawhow the efficient use of energy become a priority for policy-makers in many industrialized countries. Rising concerns aboutclimate change has heightened the importance of energy effi-ciency. Energy-related emission accounts for 9.9 Gtong of CO2 in2004, which is an increase of 65% from 1971 levels (Worrell et al.,2009). With the current best available technologies (BAT) andgiven the huge amount of energy wasted (Energetics Study, 2004,PNNL Study, 2004), energy efficiency is almost regarded as themost cost effective tool to battle carbon dioxide (CO2) emissionsand hence climate change. At the firm level, energy efficiency alsoreduces cost of production and increases firms’ competitiveness(Worrell et al., 2009). However, despite many years of trying,numerous researchers lament that the potential of energy effi-ciency remains untapped.

ll rights reserved.

þ65 6777 1434.

Much of the academic and policy research for industrial energypolicy had focused on improving energy efficiency by addressingthe infamous energy efficiency gap. This typically involves con-ducting studies to identify the barriers which inhibit the adoptionof cleaner equipments and manufacturing practices, as well aslearning from the experience of other countries such as Japan andthose in Europe (Hendel-Blackford et al., 2007).

Given the multi-disciplinary nature of energy efficiency, it isnot surprising that researchers with different backgrounds, ran-ging from ecology to economics, have engaged in this research.Due to this, advice on how to promote energy efficiency differsdepending on the adopted perspective. Mainstream economistshave argued that the main barriers to energy efficiency aremarket failures such as the principle-agent problem and imper-fect information. On the other hand, non-economic researchers,such as engineers and policymakers, have conducted surveys toidentify and rank the possible barriers. Based on the barriersidentified, solutions are proposed on how the barriers should beovercome.

Despite the myriad of studies, there remains no consensus onwhich barriers are the most important. The attempt to classifybarriers into different categories, while interesting, reveals

K.-H. Chai, C. Yeo / Energy Policy 46 (2012) 460–472 461

nothing substantially new on the nature of these barriers. Withfew exceptions, one commonality to these studies is that thebarriers (or groups of barriers) were usually treated in isolationwhere standalone solutions were recommended to tackle the(groups of) barriers without considering the possible relation-ships between the barriers. As will be explained later, accordingto the systems perspective, such a piecemeal approach neglectsthe interconnected nature of the barriers and is not likely to leadto a sustainable improvement in energy efficiency. Therefore, thepurpose of this paper is to address the energy efficiency problemfollowing the systems perspective which takes into account thepossible interactions between the various elements such asbarriers, stakeholders and policies.

The main objective of this paper is to introduce andillustrate—through a conceptual framework—the advantages ofaddressing energy efficiency adoption from a systems perspec-tive. The development of such a concept requires that we reviewand dwell into the existing literature in order to develop atheoretical framework that addresses some shortcomings of thetypical assumptions and conventional views on barriers to energyefficiency. One major commonality among the reviewed literatureis the lack of consideration for the relationships and interaction ofbarriers. Building on this and coupled with ‘‘hints’’ observed inthe case interviews, we are able to propose a novel perspective toenergy efficiency barriers.

The remainder of this paper is organized in the followingmanner. The next section presents an extensive literature reviewon energy efficiency. This is followed by a brief description ofsystems thinking and its merits. After that, we present theresearch approach adopted in this study. The section after thatis on data collection, analysis and results. The paper concludeswith the study’s contributions and implications for practice andresearch.

2. A literature review on energy efficiency barriers andpolicies

2.1. Barriers to energy efficiency

It is widely discussed and recognized that the presence ofcertain barriers is the reason for the ‘‘energy efficiency gap’’, aterm coined by Jaffe and Stavins to explain the ‘‘paradox ofgradual diffusion of apparently cost-effective energy efficienttechnologies’’ (Jaffe and Stavins, 1994). Weber (1997) proposeda methodological background to introduce the concept of barriermodels (for energy efficiency) in which three specific featureswere addressed. The three features were, namely, the objectiveobstacle, the subject hindered and the action hindered. Someyears later, Sorrell defined barriers to energy efficiency as ‘‘pos-tulated mechanisms that inhibit investment in technologies thatare both energy efficiency and economically efficient’’ (Sorrellet al., 2000, p. 27). In the following paragraphs of this section, wediscuss how barriers have been studied in the past and how theanalysis of barriers has been done from a more interdisciplinaryperspective in recent years.

A review of studies on barriers to energy efficiency shows thatcountry-specific (e.g. Nagesha and Balachandra, 2006; Rohdin andThollander, 2006; Thollander and Ottosson, 2008; Wang et al.,2008), region-specific (e.g. UNEP, 2006) and theoretical economicstudies (e.g. Howarth and Andersson, 1993; Brown, 2001) havebeen conducted. Country-specific studies are usually conductedfor major sub-sectors (e.g. Rohdin et al., 2007; Thollander andOttosson, 2008) or for industry clusters such as small industryclusters (e.g. Nagesha and Balachandra, 2006) and small-mediumenterprises (e.g. Onut and Soner, 2007; Thollander et al., 2007).

We can see different approaches to barrier analysis among theaforementioned studies. In the early years, barriers to energyefficiency were often explained using theories from the main-stream economics. The energy efficiency gap was largely attrib-uted to market failures, which occur due to flaws in the waymarkets operate. Mainstream economists argued that an imper-fect market is a major reason for slow adoption of energyefficiency technologies and suboptimal energy efficiency invest-ments. Commonly reported market failures include informationproblems, unpriced energy costs and the spillover nature ofresearch and development (R&D) (Brown, 2001; Gillinghamet al., 2009).

Information problems include a number of specific problemssuch as lack of information, asymmetric information and the well-documented principle-agent problem. Asymmetric informationproblems occur when one party involved in a transaction hasmore information than the other (Gillingham et al., 2009), whichmay lead to suboptimal energy efficiency decisions. The fact thatenergy efficiency cannot be observed (i.e. it is ‘‘invisible’’) furtherintensifies this asymmetric information barrier. Equipment sellerscan advocate the energy efficiency of a machine, but buyers oftendo not regard this as an important aspect. According to Andersonand Newell (2004), this is a prevalent problem in the industrialsector and they reported that managers are more concernedabout initial costs rather than annual savings when deciding toinvest in an energy efficiency program.

Economists also believe that a correctly priced energy costwould spur energy efficiency almost automatically. Mechanismsthat try to incorporate negative externalities into energy pricesinclude practices such as domestic carbon trading (in selected EUcountries) and the embodiment of some emission costs asrequired by the Clean Air Act enforced by the US EnvironmentProtection Agency. However, schemes such as domestic carbontrading are not problem free; they increase the business operatingcosts of the country concerned compared to the countries withoutsuch schemes. In addition, proper and efficient trading can onlytake place when the data involved is accurate and verifiable(Egenhofer, 2007).

Another frequently identified market failure is R&D spillover.R&D spillover occurs when organizations which develop or adoptenergy efficiency technologies absorb the market and technolo-gical risks associated with it, but the payback and benefits, whilebenefiting the organization involved, also flow to the public,competitors and other parts of the economy indirectly. Thisspillover makes energy efficiency R&D investments unattractive(Brown, 2001).

However, market failures can only account for part of theenergy efficiency gap. Increasingly, analysts as well as policy-makers are seeing industrial energy efficiency as a multi-facetedtopic entailing technical, economic and organizational challenges.In recent years, researchers have adopted a more inclusive andopen approach by conducting interviews and surveys (question-naires) and performing case studies to identify barriers present inthe industrial sector. Typically, barriers are identified, classifiedand discussed according to their nature (Rohdin and Thollander,2006). In addition to economic, organizational and behavioralclassifications (Sorrell et al., 2000), we saw UNEP (2006) classifyingbarriers into areas of management, information and knowledge,financing and government policy. Based on these classifications,suggestions may be offered on possible remedies to overcomethese barriers. Examples include energy labeling programs toovercome information problems and incentives or grants to alle-viate financial barriers. Some researchers have also attempted toidentify the most significant barrier in their respective areas ofstudy by ranking the barriers (Rohdin et al., 2007). The natureof such surveys is that the results are contingent, i.e. the degree of

K.-H. Chai, C. Yeo / Energy Policy 46 (2012) 460–472462

importance of the barriers is applicable only at the place and timeat which the survey was conducted, and therefore the findings maynot be applicable to other countries and/or industrial sectors.Nonetheless, the list of identified barriers remains fairly similardespite the different rankings and classifications by differentanalysts.

In recent years, researchers in other disciplines have also takenan interest in energy efficiency. In particular, we see an increasingnumber of other social science perspectives on barriers to indus-trial energy that largely discusses social barriers to technologyadoption and innovation diffusion. For instance, Owens and Driffill(2008) and Stephenson et al. (2010) argued that behavioral andattitude changes to energy consumption contribute to energyefficiency. Similar and newer perspectives on identifying andcreating socio-technical transition pathways to sustainable energysystems have also been introduced (Adamides and Mouzakitis,2009; Smith et al., 2010). These perspectives largely discuss socialbarriers to technology adoption and innovation diffusion. A recentstudy by Palm and Thollander (2010) highlighted the interdisci-plinary nature of energy efficiency and investigated the effects ofsocial networks and regimes on energy efficient technology diffu-sions. They emphasized the need for analysts to steer away fromtraditional approaches to barrier analysis.

Collectively, previous studies have identified a somewhatcomprehensive list of barriers to adoption of energy efficiencyin the industrial sector. However, there is no consensus on whichbarriers are the most important. While analysts such as Nageshaand Balachandra (2006) and Rohdin et al. (2007) concluded thatfinancial barriers are the most significant barriers, others haveidentified production risk and information barriers as the mostsignificant barriers (Kounetas et al., 2010 Rohdin and Thollander,2006). Perhaps more importantly, it is unclear whether over-coming the most significant barriers will automatically lead tobetter energy efficiency adoption, especially if the barriers areinterconnected.

It must be highlighted that, of the references cited in this study,only three studies implied that the barriers are interconnected.The first study, Wang et al. (2008), explored the interactions ofbarriers using Interpretive Structural Modeling (ISM) to map andrank the energy efficiency barriers in China. The second study,Nagesha and Balachandra (2006), employed Analytical HierarchyProcess (AHP) to identify the structure of energy efficiency barriersin several small sector industry (SSI) clusters in India. The resultssuggest that barriers may have a multi-structural level model or aform of hierarchy. The third study, Hasanbeigi et al. (2009),showed the connections between barriers in Thailand, upon whicha framework for the process of decision-making for investment inenergy efficiency was proposed. Together, these three studiesallude to the fact that there is some sort of connection betweenthe various barriers which needs to be recognized when over-coming energy efficiency barriers.

To aid subsequent data collection and discussion, weattempted to identify and sieve out key barriers from literature,since many barriers reported in different references are essen-tially similar but labeled differently by different authors (forexample, limited access to capital is similar to lack of fundingfrom management which was theoretically categorized under‘‘economic non-market failure’’). Table 1 shows how key barriersto energy efficiency were derived from the relevant literatures.

2.2. Policies for promoting industrial energy efficiency

Having reviewed the literature on barriers, we will nowexamine policies which aim to promote energy efficiency. Formany governments, energy efficiency is often a first measure inreducing energy intensity. Perhaps not surprisingly, the countries

which first drove energy efficiency like Japan and the UnitedKingdom (UK) have little or no indigenous energy resources andimport most of their energy (Hendel-Blackford et al., 2007). Theirvulnerability to energy supply and prices led to the need forhigher energy efficiency. In an era of climate change mitigationand adaptation, energy efficiency is further viewed as a practicalmeans to reduce CO2 emissions.

As prices of renewable energy are still uncompetitive, theindustrial sector is expected to rely heavily on conventional fuelsfor operations. Therefore, energy efficiency remains a highlyvalued government strategy for the industrial sector. Not onlydoes it drive down operating costs, energy efficiency also providesa practical means to meet CO2 emission reduction targets. Amyriad of government tools and policies to improve industrialenergy efficiency can be found in the literature. Broadly, energyefficiency policies and programs take the form of regulation andlegislation, economics and voluntary measures (UNEP, 2006).

Depending on the country’s culture and legal tradition, theextent of regulation and legislation measures varies. Of notablesuccess is Japan’s Energy Conservation Act of 1979 under whichan energy efficiency program took place (Hendel-Blackford et al.,2007). Japan has a history of strong legal tradition therefore whenthe Energy Conservation Act was passed, implementation wasrelatively smooth and results were effective: Japan saw a 37%reduction in its energy intensity during the 1979–2003 period(Hendel-Blackford et al., 2007). Some European Union (EU)countries like the United Kingdom (UK) also experienced somesuccess with industrial energy regulatory policies. Industrialregulations and legislation programs include minimum efficiencystandards for common equipment such as air compressors andcombined heat and power (CHP) plants, mandatory appointmentof an energy manager, mandatory energy audits and factoryenergy conservation plans (Geller et al., 2006).

On the other hand, more countries (including Netherlands andGermany) took to voluntary agreements (VA) and fiscal measures tostimulate industrial energy efficiency improvements. Voluntary mea-sures have been more popular with governments because, comparedwith regulations, voluntary measures have fewer negative impacts onindustrial competitiveness (Hendel-Blackford et al., 2007). The mainobjective of a voluntary agreement is to gather participation fromindustrial organizations to reduce energy consumption and CO2

emissions. The details and rigors of voluntary agreements differ indifferent countries but, generally, VA are complemented by fiscalmeasures such as tax incentives, subsidies or exemptions, andinvestment grants (Geller et al., 2006). The success of a VA usuallydepends on the incentives offered, the potential for energy efficiencyimprovement in the company as well as the social-culture nature ofthe sector. For example, it was found that VA was less effective withsmall enterprises compared to big enterprises (Rietbergen et al.,2002). Although fiscal measures alone already provide some formof motivation for organizations to adopt energy efficiency technology,voluntary agreements create an increased awareness about theavailable government financial incentives.

Alongside the abovementioned programs, educational andinformative programs are also commonly implemented. Onewell-known example is the energy labeling program which servesto inform consumers about the energy consumption of equip-ment. We are seeing an increase in the number of countries thatmake energy labels mandatory. US Energy Star labeling, forexample, has achieved great success over the past few decades.Energy audits, energy management systems and energy managertraining and certification are also awareness programs that areusually part of VA. Japan and Singapore are two examples ofcountries which employ energy auditing and energy managementsystems in the industrial sector to create awareness about energyconsumption.

Table 1Identifying key barriers to energy efficiency from reviewed literature.

Category Typical barriers References Key barriers identified

Economic non-market failure or

market barriers (Sorrell et al.,

2000; Brown, 2001)

Low priority of energy issues Brown (2001) � Fear of technical risk/

cost of production loss

� Perceived high cost of

energy investment

� Other capital

investments are more

important

� Uncertainty about future

energy price

� Lack of experience in

technology

� Lack of information in

energy efficiency and

savings technology

� Lack of trained

manpower/staff

� Lack of energy metering

� Lack of access to capital/

budget

� Lack of government

incentives

� Weak policies and

legislations

� Resistance to change

� Legacy system

Cost of production disruption Rohdin and Thollander (2006), Thollander and Ottosson

(2008), Thollander and Dotzauer (2010)

Other priorities for capital investments Rohdin and Thollander (2006), Thollander and Dotzauer

(2010), Sardinou (2008)

Lack of time/other priorities Rohdin and Thollander (2006), Nagesha and Balachandra

(2006), Thollander and Dotzauer (2010)

Reluctant to invest because of high risk Wang et al. (2008)

Technical risk such as risk of production

disruptions

Thollander and Ottosson (2008)

Competition from other projects Ren (2009)

Lack of management support UNEP (2006)

Limited access to capital Rohdin and Thollander (2006), UNEP (2006), Thollander

and Dotzauer (2010), Sardinou (2008)

Capital market barriers Brown (2001)

Lack of investment capability Nagesha and Balachandra (2006)

Lack of funding/financing capabilities Wang et al. (2008)

Uncertainty about future energy price Thollander and Dotzauer (2010), Sardinou (2008)

Increased perceived cost of energy

conservation measures

Sardinou (2008)

Cost of identifying opportunities,

analyzing cost effectiveness and

tendering

Thollander and Ottosson (2008), Thollander and

Dotzauer (2010), Rohdin and Thollander (2006)

Economic market failure (Sorrell

et al., 2000; Brown, 2001)

Split incentives Brown (2001)

Un-priced cost and benefits Brown (2001)

Insufficient and inaccurate information Brown (2001), Wang et al. (2008), Ren (2009), UNEP

(2006), Nagesha and Balachandra (2006), Thollander and

Ottosson (2008)

Lack of experience in technology and

management

Wang et al. (2008), Ren (2009)

Difficulties in obtaining information

about the energy consumption of

purchased equipment

Thollander and Dotzauer (2010)

Lack of technical skills Thollander and Dotzauer (2010), Sardinou (2008)

Lack of trained manpower Wang et al. (2008), Thollander and Dotzauer (2010),

Thollander and Ottosson (2008), Rohdin and Thollander

(2006), Sardinou (2008)

Lack of information on profitability of

energy saving measures

Sardinou (2008), Wang et al. (2008)

Lack of information with respect to

energy conservation opportunities

Sardinou (2008)

Behavioral (Sorrell et al., 2000) Resistance to change Nagasha and Balachandra (2006)

Institutional (Weber, 1997) Weak legislations and/or enforcement UNEP (2006), Nagesha and Balachandra (2006)

Lack of government incentives UNEP (2006)

Organizational (Sorrell et al., 2000;

Weber, 1997)

Lack of sense of corporate social

responsibility or environmental values

Rohdin and Thollander (2006)

Lack of environmental policies within

company

UNEP (2006)

Energy manager lacks influence Sardinou (2008)

Lack of sub-metering Thollander and Dotzauer (2010), Thollander and

Ottosson (2008)

Physical constraints Inappropriate technology at site Thollander and Dotzauer (2010), Wang et al. (2008)

Inappropriate industrial framework Wang et al. (2008)

K.-H. Chai, C. Yeo / Energy Policy 46 (2012) 460–472 463

A relatively new development in the arena of energy efficiencymeasures is energy efficiency financing where organizations(borrowers) can obtain financial support from energy servicecompanies (ESCOs) themselves or from a third party financersuch as large commercial banks and international financialinstitutions like the World Bank. The financial support occurs ina manner that allows the borrower to repay the lender from theenergy savings.

Clearly, various policies have been deployed in order topromote energy efficiency over the years in many countries.However, there is no established advice or theory on when andwhat policies should be applied. The disparity between promiseand actual progress suggests that there is an urgent need todevelop a framework which links these policies together. Thislack of an overarching framework may stem from the fact that

many energy efficiency studies, as discussed in the first part ofthis paper, neglect the possible relationships between the bar-riers. Developing a holistic framework which takes into accountthe relationships between the barriers is thus necessary in orderto achieve greater energy efficiency in industries.

3. The systems approach to problem solving

The systems approach or systems thinking is a perspectivewhich views an event or a system in a holistic manner by placingexplicit emphasis on the relationships and interactions betweenthe system’s elements and constituents (Senge, 1990). In the earlyyears, concepts and applications of systems thinking were recog-nized as general systems theory (Bertalanffy, 1950). The core

K.-H. Chai, C. Yeo / Energy Policy 46 (2012) 460–472464

concepts included parts/wholes/sub-systems, system/boundary/environment, structure/process, emergent properties, hierarchy ofsystems, feedback effects, information and control, open systemsand holism (Mingers and White, 2010). These fundamental con-cepts have not changed much throughout the years and theemphasis on relationships and interactions could not have beenmore valued. Much of systems thinking’s power lies in its abilityas a problem solver to identify underlying the system’s structurethat explains (similar) patterns of behavior in a variety ofdifferent situations. Systems thinking also requires that we shiftour mind from event orientation (linear causality) to focusing oninternal system structure (circular causality), as the underlyingsystem structure is often the root cause of the problems. Thisprobably explains why the systems approach is considered usefulfor dealing with complex, large scale and interdisciplinary pro-blems (Boulding, 1956).

Hawkesbury’s hierarchy (Bawden et al., 1985) presented typesof research approach to problems, from basic research to appliedresearch and to systems research. Basically there are two types ofsystems approach, the hard and soft systems approaches. Stephenand Hess (1999) illustrated the application of hard and soft systemsusing the concept of ‘‘level’’ and ‘‘output’’, where ‘‘level’’ can beloosely understood as the unit of analysis ranging from individualCEOs, companies or industrial subsectors. The level of system beingstudied has a direct implication on the choice of approach adoptedfor analysis. Naturally, the higher the level, the interplay of a largernumber of factors, the higher the degree of ‘‘subjectivity’’ and thelower the degree of ‘‘reductionism’’ (breaking it into components)(Bawden et al., 1985). To further illustrate, Checkland (1981) refersto a spectrum of systems approaches from those ‘relatively hardsystems characterized by easy-to-define objectives, clearly defineddecision taking procedures and quantitative measures of perfor-mance’ to soft systems in which objectives are hard to define,decision taking is uncertain, measures of performance are at bestqualitative and human behavior is irrational’.

Therefore, hard systems approaches are appropriate for lowerlevel (i.e. more well-defined system) of analysis which often leadsto quantitative modeling, where a simulation of the functioning ofthe system mathematically allows researchers to investigate theresponse of the system to alternative stimuli (Stephen and Hess,1999). Soft systems on the other hand are appropriate forproblems less clearly defined and takes into account the differentperspectives of all relevant valid stakeholders (Stephen and Hess,1999). In our case, a soft system approach would help to betterdefine the problem. Some examples of application of systemsapproaches to research are provided in Table 2. In addition,systems thinking is also applied quite extensively to policy and

Table 2Application of systems approach to problem analysis.

Application area Type of systems approach References

Water

management

Hard systems approach Stephens and Hess (1999),

Matthews et al. (1997),

Perry (1996)

Soft systems approach Uphoff (1996)

Energy

management

Soft systems approach Freeman and Tryfonas (2011),

Ngai et al. (2011)

Waste

management

Hard systems approach

(systems engineering)

Pires et al. (2011)

Shipbuilding

industry

Systems thinking Anh et al. (2009)

Product/project

management

Systems thinking Lin and Ng (2009)

Socio-technical

transitions

Systems thinking Bennett and Pearson (2009)

economic analysis due to its ability to model feedbacks (e.g. Chiet al., 2009; Qudrat-Ullah and Baek, 2010; Gielen et al., 2001).

Driscoll (2008) pointed out that we are unable to view systemlevel behaviors and interactions (or the system’s structure) whenwe decompose a system into its elements. Bearing that in mind, werecognized and considered a multifaceted energy efficiency adop-tion system in a company that takes into consideration the inter-play of barriers to energy efficiency internal and external to thecompany, as well as the influence of the actions of differentstakeholders in the process of energy efficiency adoption. Weargue that there is a lack of consideration for interactions amongbarriers, which is why barriers persisted despite the efforts oftrying to remove them. Fundamental to the holistic approach is theconcept of the ‘‘whole being greater than the sum of its parts’’ dueto interactions (Rountree, 1977). Barriers to energy efficiencycannot be properly studied by looking at them in isolation, whichis what we observed in many prior studies. Often, recommenda-tions were proposed for one barrier or a group of barriers withsimilar nature, disregarding the possible interactions betweenbarriers which may well render the recommendation ineffective.This we shall argue display a lack of systems thinking, which willenable us to identify possible relationships among the (groups of)barriers. Understanding the relationships is important in makingeffective recommendations.

In the context of this study on energy efficiency, our interest isthe removal or reduction of barriers to energy efficiency and werecognize and accept the validity and relevance of all actors orstakeholders (i.e. industrial organizations, manufacturers, govern-ment agencies, customers, and energy service companies), relatedpolices, energy efficient technologies and practices. As will beshown later, by adopting a systems thinking perspective, weavoid falling into the trap of assuming that barriers to energyefficiency are solely caused by external events such as marketfailures (a form of linear causality), and thinking that barriers areindependent of each other. We attempted to identify possibleinteractions, relationships, feedbacks and delays in the system todevelop a framework for improving industrial energy efficiency.

4. Research approach

Given the relative lack of established theories from a systemsperspective on barriers to energy efficiency, a theory-buildingapproach (Strauss and Corbin, 1998; Eisenhardt, 1989 wasadopted in this study). In contrast to the theory-testing approachwhich aims to test the hypotheses by using quantitative methods,theory-building or theory generation involves the formation ofabstract concepts and generation by observing and reflecting reallife experiences through an inductive and qualitative process. Thisapproach is commonly adopted when there is a lack of estab-lished theories in the area of research (Eisenhardt, 1989, Gill andJohnson, 1991). In such cases, framework and conceptual con-structs, rather than robust and rigorous models, are more usefulfor understanding the issue (Adler, 1989).

Fig. 1 shows the research design and implementation adopted inthis study. It begins with an extensive literature review of bothacademic and practitioners’ publications which was continuedthroughout the research. It is important to have an up-to-date

Fig. 1. Research approach.

K.-H. Chai, C. Yeo / Energy Policy 46 (2012) 460–472 465

understanding of approaches to barriers analysis and organizations’perceptions of energy efficiency. Preliminary findings were used toguide data collection and analysis. The second stage of this approachis data collection through semi-structured interviews with practi-tioners as well as by examining the relevant documents. Theinterview questions included what are the challenges or barriersfaced in implementing energy efficiency? How are they overcome?Are the current government measures adequate? Why? etc. A moredetailed case study was conducted with Glaxo Wellcome Manufac-turing Pte Ltd. Singapore, one of the companies interviewed, becauseof its success in energy efficiency efforts over a long period of time.We analyzed the multiple sources of data collected (i.e. interviewtranscripts and those listed in Table 3) and applied principles ofsystems approach in the third phase to develop a conceptualframework. It is worth mentioning that, while Fig. 1 depicts a linearresearch process, in reality the stages overlapped and were iterative;we refined the framework according to each new and relevantfinding during the process of our research.

A key strength of the qualitative method is the variety of dataand the ways it can be collected. This allows triangulation—theconfirmation of findings through the convergence of multipledata—to take place. There is more than one method of triangula-tion. Triangulation can happen by data source (persons, times,places, etc.), by method (observations, interviews, etc.), and bythe use of different researchers on the same subject, by theoryand by data type (texts, numbers, etc.) (Miles and Huberman,1994). In this research, triangulation by data source, method anddata type were adopted.

Triangulation by researcher and theory were not possible dueto time and resource constraints. In reality, as Miles andHuberman (1994) pointed out, triangulation is more a way ofresearch life than a tactic. When a researcher consciously double-checks findings by using multiple sources and modes of evidence,the verification processes is built into the data collection.Throughout this research, whenever possible, attempts weremade to obtain data from different sources (e.g. asking similarquestions to different managers in the same company), fromdifferent methods (e.g. formal interviews, observations, informalconversations, project reports) and from different data types (e.g.numbers, descriptions in interviews). In total, interviews wereconducted with eleven industrial organizations and five energyservice companies (ESCOs) which have extensive experience withenergy efficiency. Majority of the industrial organizations studiedare from the petrochemical industry, where energy cost is asubstantial component of their operating cost.

Table 3 lists the primary and secondary data collected andtriangulation used throughout the case studies. For confidentialitypurposes, the actual names of the organizations have beenreplaced by letters. In this study, based on the research objectives,the unit of analysis is industrial organizations that haveattempted energy efficiency improvements. Unit of analysis refersto the core subject around which the research is focused anddraws the boundary for data collection. The choice of unit ofanalysis is determined by the research questions (Yin, 1989). Awell-defined unit of analysis helps to impose boundaries on datacollection (Miles and Huberman, 1994). Several ESCOs wereincluded as they offer interesting insights from a solution provi-der’s perspective.

5. Results

A summary of the barriers identified from the interviews ispresented in Table 4 where A1–A5 denote the five barriers—in noorder of significance—reported by company A, B1 and B2 werebarriers by company B, etc. It can be seen that, by and large, the

barriers identified from the interviews are similar to thosereported in the literature, though the significance of differentbarriers differed in different organizations. For example, werecorded from the interviews with the local small-mediumenterprises that smaller organizations like them tend to facemore technical and financial barriers than larger organizations.

In addition to the list of barriers, a few interesting observationswere made and worth reporting:

�

There is a varying degree of commitment or motivation (andmaturity) to energy efficiency among the organizations. Dri-vers or motivations for energy efficiency are stronger forcompanies where energy cost is a substantial part of itsoperating cost (e.g. petrochemical companies), and those witha stronger sense of corporate social responsibility. In general,the motivation factors can be categorized as either economic(e.g. to reduce operating costs) or environmental (e.g. to be agood corporate citizen). � Larger organizations have more resources (time, staff, and

financial resources) and technical ability for energy efficiencyinvestments. Larger organizations enjoy wider internationalnetworks and, as reported by some foreign multi-nationalcompanies (MNCs), they were able to perform internal bench-marking with their factories in other locations. The samereason that larger organizations are faster and more successfulin adopting new technologies may be used to explain thisobservation (Rogers, 1995). Nevertheless, some smaller orga-nizations are still able to overcome this disadvantage byseeking ESCOs consultancy services, such as installation ofenergy monitoring and control systems.

� Many energy efficiency investments are not implemented due

to fear of disrupting production. Plant managers and ESCOsrevealed that the cost of loss in production tends to be greaterthan the savings projected from energy efficiency improve-ments. In addition, energy is a factor of production in theindustrial sector and, therefore, efficiency levels may bestructurally based or merely an artefact of initial installationand construction specifications. Also, given that much produc-tion runs 24 h a day, the window to modify the productionprocess for energy efficiency reasons is few and far between.

� There is a lack of data showing positive returns of energy

efficiency adoption in industrial organizations and this poses abig barrier to sustaining energy efficiency efforts. In mostindustrial organizations, the existing energy monitoring andcontrol systems were not designed to capture energy effi-ciency improvements. Traditionally, the energy consumptiondata has been used for deciding products’ costs and prices.With the increased pressure on being more energy efficient,some organizations started using their existing systems tomonitor energy efficiency. While these systems can measuresystem level energy efficiency improvements, it is difficult tocapture component level efficiency improvements. Componentlevel improvements are easily offset by other changes occur-ring in the production system such as changes in productionmix, volume, operating conditions, etc. Thus, although there isa general agreement that energy efficiency is important, it isstill difficult to convince top management about the benefits ofenergy efficiency as savings are often not ‘‘visible’’. In fact, thelack of appropriate energy efficiency metrics is a gap betweenindustrial needs and scientific literature that has been identi-fied by Bunse et al. (2010), though their main argument for theneed for appropriate energy efficiency metrics was for bench-marking purposes.

One objective of this analysis is to recognize possible relation-ships among barriers. We observed that some barriers were more

Table 3Details of data collected from industrial organizations. All interviews were conducted in 2010.

Company Industry/notes Interviews

(primary source)

Secondary

sources

Status on

energy efficiency

A � Petrochemical

� Multinational

� Annual revenue over USD250 billion

� One of the largest players in the world

� Technology/

Development Manager

� General Manager

(External Affairs and

Communications)

� Corporate

website

� Project

documents

� Annual report

� Hydrocarbon Asia

� Energy efficiency is taken as a continuous

improvement. Generally, energy efficiency practices

have been quite successful

B � Petrochemical

� Multinational

� Annual revenue over USD250 billion

� One of the largest players in the world

� Manager (Public and

Government Affairs)

� Advisor (Public and

Government and Media

Relations and

Communications)

� Corporate

website

� Project

documents

� Annual report

� Energy dialog

� Energy efficiency is taken as a continuous

improvement. Generally, energy efficiency practices

have been quite successful

C � Petrochemical

� Joint-venture between a local and

foreign multinational company

� Manager (Business

Development) and

Planning-cum-Energy

Manager

� Manager (Process and

Operation Technology)

� Corporate

website

� Energy efficiency is taken as a continuous

improvement. Generally, energy efficiency practices

have been quite successful but noted that ‘‘low

hanging fruits have been exhausted’’

D � Pharmaceutical

� Multinational

� Director (Engineering

Solutions)

� Corporate

website CSR

report company

posters

� Energy efficiency is taken as a continuous

improvement. Generally, energy efficiency practices

have been quite successful

E � Pharmaceutical and healthcare

� Multinational

� Annual revenue over USD50 billion

� Engineering Service

Director/Team Leader

Mechanical Engineering

Manager

� Corporate

website project

documents

annual report

� Energy efficiency is taken as a continuous

improvement. Generally, energy efficiency practices

have been quite successful

F � Petrochemical

� A small subsidiary of a multinational

company

� Plant Manager

� Engineering Manager

� Corporate

website

� Limited energy efficiency measures due to small scale

operation. Take cues from parent company.

G � Petrochemical

� Multinational

� Plant Manager � Corporate

website

� Energy efficiency is taken as a continuous

improvement. Generally, energy efficiency practices

have been quite successful

H � ESCOs

� Multinational

� A global leader in heating, ventilating

and air-conditioning (HVAC) system

� Annual revenue over US10 billion

� Technical Director � Corporate

website project

documents

� Energy efficiency is taken as a continuous

improvement. Generally, energy efficiency practices

have been quite successful

I � ESCOs

� Multinational

� Director (Urban

Solutions) Director

(Future Clean

Technology)

� Corporate

website

brochures project

documents

� Not applicable

J � ESCOs

� Multinational

� Director (Energy

Efficiency)

� Corporate

website

� Not applicable

K � ESCOs

� Local, Small medium enterprise

� Managing Director � Corporate

website

� Not applicable

L � ESCOs

� Multinational, HQ in US

� Regional Marketing

Director (Building

Solutions) Program

Manager (Building

Solutions)

� Corporate

website

Brochures

� Not applicable

K.-H. Chai, C. Yeo / Energy Policy 46 (2012) 460–472466

Table 3 (continued )

Company Industry/notes Interviews

(primary source)

Secondary

sources

Status on

energy efficiency

M � Food Manufacturing

� National

� Local, small medium enterprise

� Executive Director and

CEO Group Project

Manager (Group

Technical Department)

Head (Electrical

Department)

� Corporate

website

� Limited energy efficiency measures due to lack of

knowledge. Consider themselves beginners

N � Engineering services

� Local, small medium enterprise

� Managing Director � Corporate

website

� Limited energy efficiency measures due to various

barriers like lack of technical expertise, physical

constraints

O � Petrochemical

� A small subsidiary of a multinational

� Research and

Technology Manager

� Corporate

website company

profile report

� Energy efficiency is a continuous improvement

process. Generally, energy efficiency measures have

been successful

P � Engineering services

� A small subsidiary of a multinational

� Corporate Facilities

Manager

� Corporate

website

� Energy efficiency is a continuous improvement

process. Generally measures have been relatively

successful but measures taken were not extensive.

Table 4Key barriers faced by industrial organizations interviewed.

Key barriers Industrial organizations

A B C D E F G H I J K L M N O P

1. Fear of technical risks/cost of production loss � � � � � � � � � �

2. Perceived high cost of energy investments � � � � � � � �

3. Other capital investments are more important � �

4. Uncertainty of future energy prices � �

5. Lack of experience in technology �

6. Lack of information in EE and energy saving technology � � � � � � �

7. Lack of staff awareness/trained manpower � � � � �

8. Lack of energy metering � � � � � �

9. ESCOs lacking in specialised knowledge (empirically recorded) �

10. Limited access to capital/budget � � � � �

11. Lack of government incentives �

12. Weak policies and legislations � � � �

13. Too many government stakeholders (empirically recorded) �

14. Resistance to change �

15. Legacy System (Efficiency levels may currently be structurally

based, or merely be an artefact of initial installation and

construction specifications)

� � � � � � �

16. Space Constraint (empirically recorded) � � �

K.-H. Chai, C. Yeo / Energy Policy 46 (2012) 460–472 467

commonly reported in the presence of barriers. For example,companies that reported high cost of energy investments as abarrier also reported technical risks/cost of production, lack ofinformation on energy efficiency and energy saving technology,limited access to capital/budget, legacy system. Companies whofound lack of information on energy efficiency also faced barrier oflack of staff awareness or trained power. Fear of technical risks wasalso commonly reported with lack of energy metering and lack ofinformation on energy efficiency. These observations indicated thatbarriers could form a process through how a company adoptsenergy efficiency measures which will be elucidated in Section 6.In the following paragraphs of a case study, we specifically discusshow a company overcomes barriers to energy efficiency which alsocontributes to the conceptualisation of the framework in Section 6.

To understand how companies overcome related barriers toenergy efficiency and the relationships between the barriers, weconducted an in-depth case study on Glaxo Wellcome Manufacturing(GWM) Pte Ltd. Singapore, which is one of the companies listedin Table 3. GWM Singapore is a wholly owned subsidiary of

GlaxoSmithKline (GSK), a leading global pharmaceutical based inthe UK. Pharmaceutical products are generally less energy intensivecompared to products from industrial sectors such as steel, cementand petrochemicals. That is, energy cost is only a small part of theiroverall operating expenses. Hence, it was particularly useful to drawlessons from GWM Singapore as they have pursued energy efficiencyimprovements despite not having a strong financial motivation, andable to achieve remarkable results. In the following paragraphs, wediscuss how GWM Singapore achieved energy efficiency by firstexamining the primary drivers and then identifying the criticalsuccess factors for their improvements in energy efficiency.

GWM Pte Ltd. Singapore has been active and successful inpursuing energy efficiency and conservation since 2002. It allstarted with a production transfer from their manufacturing plantin the United Kingdom to the plant in Singapore. As a result of theincrease in production, the energy consumption was forecasted toincrease by 40%. In order to maintain price competitiveness of theproducts, the top management decided to pursue energy efficiencyand conservation with the goal that the increase in production

K.-H. Chai, C. Yeo / Energy Policy 46 (2012) 460–472468

would not lead to increased energy costs, effectively lowering thecost of production. To meet this goal, the Director of EngineeringServices and his cross-functional teams began a series of projectsfocusing on energy efficiency. These projects successfully avoidedthe forecasted 40% increase in energy expenses even though theproduction volume increased substantially.

There are notable success factors for GWM Singapore’s energyefficiency drive. Clearly, there was a strong motivation displayedby top management. The first notable major success factor wastop management support. The top management was motivated topursue energy efficiency and conservation to reduce the energycost of production and therefore rendered ample support toenergy efficiency activities and projects. Top management sup-port has been commonly reported in the literature as one of thecritical success factors for overcoming common barriers to energyefficiency such as limited access to capital and lack of dedicatedstaff (for energy efficiency). In this case, the management helpedovercome barriers like high perceived cost of energy investmentsin the company by allowing a longer payback period for thoseenergy investments, i.e. more access to capital. However, it mustbe noted that GWM, being an established multinational compa-nies possessed a stronger financial capabilities to do so.

To facilitate implementation of energy efficiency, GWM Singa-pore was divided into several zones, each led by a senior managerresponsible for energy initiatives and performance. Because of theclear division of responsibilities, there is no ‘‘running away’’ fromreally pursuing energy efficiency. In addition, there is a real timemonitoring system that monitors the energy use in each zone.This system allows the team to verify the actual energy savings asa result of the improvements made.

More recently in 2008, the management established an annualenergy savings target of a 5% reduction in energy consumptionyear on year. Indeed, energy consumption is one of the plant’s topfive key performance indicators, prominently displayed at thecentral common area of the plant alongside safety and qualityindicators. This served to overcome resistance to change and infact fear of risk to production as energy is viewed as importantlyas quality. As a result of these factors, GWM Pte Ltd. Singaporehas enjoyed seven years of positive returns from their energyefficiency efforts since 2002.

It is notable that when the management has motivation, manyother barriers can be overcome as the management would put inmeasures to overcome other barriers. These actions by manage-ment alluded to the fact that there is a process for a company’sdecision on energy efficiency investments.

6. Discussions

Applying the principles of systems approach to the resultsattained, we are able to (1) identify points of nteractions, (2) inte-grate perspectives of various stakeholders and (3) conceptualize aframework addresses this multidisciplinary issue. Integrating the

Fig. 2. Motivation-capability-implemen

qualitative results of data collection with the following thoughtsled to the development of our generic framework in Section 6.1.

1.

tat

Viewing the industrial sector as a heterogeneous ‘‘system’’: Theindustrial sector comprises of a large number of organizationswith variation in their degree of energy intensiveness andcorporate social responsibility, number of employees andextent of socio-technical networks.

2.

Interplay between technological, organizational and behavioral

barriers to energy efficiency: Barriers to energy efficiency influ-ence each other. For instance, if the engineering department ofan industrial organization is perceived to have low technicalcapability, it is likely that the production operation will bereluctant to give a ‘‘window’’ to implement energy efficiencyimprovements for fear of disrupting production quality. Thisshows that the barriers are inter-related, i.e. a technologicalbarrier can affect an organizational behavior barrier.

3.

Interests and objectives of stakeholders (organizations and gov-

ernments): It is inevitable that tensions exist between theinterests and objectives of an organization and of a govern-ment. Organizations and governments both have short-termeconomic concerns and long-term sustainability concerns totake care of. Such conflicting interests result in trade-offs.Often the consideration for short-term gains takes precedenceover long-term benefits.

4.

Energy efficiency adoption as a change process: During thediscussions with the managers from the various organizations,it was apparent that achieving energy efficiency involveschanging existing practices and also involves adopting newand more energy efficient technologies (in terms of equipmentand processes). Therefore, from an organizational perspective,energy efficiency improvements are innovations which involvechanges that have to be managed properly.

6.1. Motivation–capability–implementation–results (MCIR)

framework

We propose a conceptual generic framework that is based on astage-wise process with feedback. This framework, as depicted inFig. 2, shows the adoption and implementation of energy effi-ciency practices as a process which comprises of four importantstages, namely, Motivation, Capability, Implementation andResults, with a feedback effect. For each stage, we pose questionsthat capture factors affecting energy efficiency adoption andreflect the interests and objectives of stakeholders.

The framework begins with ‘‘Motivation’’ as the first stage. Atthis stage, the primary concerns are the organizations’ interests inpursuing energy efficiency and their awareness of energy effi-ciency opportunities.

‘‘Capability’’ is identified as the second stage of energy effi-ciency adoption. At this stage, organizations are now concernedwith their capability to pursue and implement energy efficiencycompetently, having been made aware of the opportunities in the

ion-results (MCIR) framework.

K.-H. Chai, C. Yeo / Energy Policy 46 (2012) 460–472 469

earlier stage, Motivation. Organizations will be interested inwhere and how they can access the capabilities needed.

The third stage of energy efficiency adoption, ‘‘Implementation’’, isthe stage where organizations actually implement energy efficiencyprojects. Here, the concern is whether the capabilities acquired in theprevious stage can result in successful energy efficiency projects.

‘‘Results’’ is the final stage of the process of energy efficiencyadoption. It refers to the outcomes of implementing energyefficiency projects. Top management will now ask if the effortsto implement energy efficiency were worthwhile. Given thatwhat you can measure you can manage, it is necessary to be ableto quantitatively demonstrate the returns on such efforts.

The outcome of the ‘‘Results’’ stage is the feedback to the‘‘Motivation’’ stage. Positive outcomes in terms of financial andeconomic gains are likely to ensure the continued adoption ofenergy efficiency programs. That is, positive and convincingresults from energy efficiency projects will have a positive feed-back effect, motivating top management to further invest inenergy efficiency. As the saying goes, success breeds success. Thispositive feedback needs to be emphasized as it is often over-looked despite reports of its importance. It was reported in a 2006UNEP report that the management of a Vietnamese fertilizercompany supported the implementation of additional (energyefficiency) options due to the validation of savings from projectsimplemented earlier (UNEP, 2006).

In the sections that follow, we will use the framework toanalyze barriers to energy efficiency, GWM’s successes in energyefficiency efforts and stakeholders’ roles in energy efficiencyadoption. Finally, we illustrate how the framework can be usedas a theoretical guiding framework for energy efficiency policies.

6.2. Mapping barriers into MCIR framework

In this section, we attempt to map barriers identified in thisstudy and the literature into ‘‘Motivation’’, ‘‘Capability’’, ‘‘Imple-mentation’’ and ‘‘Results’’ as shown in Fig. 3. Though one mayargue that the approach of categorizing barriers is similar to theliterature, we considered the interactions of barriers and the

Fig. 3. Mapping barriers int

possible sequence in which they may occur. By doing so, theMCIR framework can identify chokepoints of energy efficiency.

‘‘Motivation’’ barriers to energy efficiency are those barriers whichlower management’s interest in pursuing energy efficiency. Thesebarriers can be a lack of financial incentives (e.g. if energy expensesare only a small fraction of overall operating cost, lack of capital topursue capital-intensive technology), split incentives (Brown, 2001)or simply a lack of awareness of energy efficiency opportunities.

‘‘Capability’’ barriers can be broadly classified into technicaland financial barriers, as shown in Fig. 3. Typical barriers at thisstage are a lack of information on energy efficiency technologies,a lack of trained manpower or a lack of financial resources.

‘‘Implementation’’ barriers are barriers that inhibit the imple-mentation of the energy efficiency projects. Common ‘‘Implemen-tation’’ barriers include resistance to change and short windowsof opportunity for engineering changes given that many manu-facturing organizations operate on a 24/7 basis and there is a fearof disrupting existing production processes.

Barriers in the ‘‘Results’’ stage are widely reported but oftenarticulated in different ways. Essentially, the biggest barrier is thelack of positive results from energy efficiency investments. To theorganizations, results can be interpreted as economic and financialgains.

Companies expressed that there are often little or insignificantenergy savings from energy efficiency efforts, failing to recognizethe fact that energy costs often do not constitute a large portion oftotal operating costs, and hence energy savings through energyefficiency adoption may be easily offset by other changes such asincreased manpower and production changes.

After the conceptual development of our framework, weapplied it to analyze GWM’s success with energy efficiencyefforts. Matching qualitative data from GSK’s interview to theframework reveals that efforts must follow through all the stagesof the MCIR framework to actualize energy efficiency improve-ments. Fig. 4 shows how possible barriers were prevented orreduced in each stage of the MCIR framework. This analysis showsthat the framework provides a sound reasoning for energyefficiency adoption by organizations. More importantly, it implies

o the MCIR framework.

Fig. 4. Analyzing energy efficiency in GWM using the MCIR framework.

K.-H. Chai, C. Yeo / Energy Policy 46 (2012) 460–472470

that the level of energy efficiency adoption in an organization isonly as strong as the weakest link.

6.3. Understanding the roles of stakeholders from the MCIR

perspective

Having explained the rationale of the framework and barriersassociated at each stage, the MCIR framework can be applied tounderstand the roles of the major stakeholders in improving energyefficiency. This step is taken following the ‘‘advice’’ of systemsthinking on the need to see the bigger picture. This big pictureenables us to see the complex dynamics between the variousstakeholders in driving energy efficiency. In this study, we identifiedthe stakeholders as those who have a more direct influence onenergy efficiency actions. They include governments, the organiza-tions themselves, energy service companies (ESCOs) and customers.

Government: The nature of energy efficiency is such thatindividual projects often fetch little energy savings for the com-pany, but collectively they can save a substantial amount ofenergy for a large corporation or a nation. Energy efficiency isespecially important for energy importing nations as it is almostalways the most powerful tool for combating climate change andachieving energy security. Therefore, governments should bemajor stakeholders in realizing the potential of energy efficiencyin the industrial sector. Countries like Netherlands, Japan andKorea have shown how voluntary agreements are effective forenergy efficiency in the industrial sector without affecting indus-try competitiveness. Voluntary agreements are examples of moti-vation for the industrial sector to pursue energy efficiency as theycan provide win–win situations. In these countries, the govern-ment also provides capabilities for the industrial sector such as theprovision of energy manager training and financial incentives.Governments can also help to overcome ‘‘Implementation’’ and‘‘Results’’ barriers through target setting and establishing a stan-dard protocol for energy reporting respectively (refer to Table 5).

Industrial Organizations: Industrial organizations are oftenmotivated to pursue energy efficiency to reduce costs and displaycorporate social responsibility. Top management can induce anenergy efficiency culture to promote energy efficiency adoption intheir organization (such as GWM). Seeking help from technicalconsultants and appointing energy managers are ways to reduce‘‘Capability’’ barriers in organizations. In the case of GWM Singa-pore, cross-functional teams were formed which helped in imple-menting energy efficiency across the organization. For sustainedefforts in energy efficiency, organizations should collect relevantand accurate data on energy savings and energy efficiencyimprovements. Such data can also be used for benchmarking.

ESCOs: The role of ESCOs in the framework is mostly recog-nized for reducing ‘‘Capability’’ barriers, in particular the technicalcapability barriers. They do so by performing energy audits and

recommending energy efficiency improvement plans. Experi-enced ESCOs also provide a source of information for industrybest practices and benchmarks.

Customers: Customers are the reason for a company’s existence.Their demands will direct the company’s market and developmentalpolicies. Therefore, as the number of ‘‘green’’ customers increases,motivation for energy efficiency is expected to increase.

While research and academic institutes also play a part in theenergy efficiency landscape, we have excluded them for tworeasons: (1) in this study, we analyze the adoption of energyefficient technologies and practices to improve energy efficiencyrather than developing new energy efficiency products, which weassume are available in the market, and (2) their activities may bedependent on governments’ and private organizations’ funds anddirections, and the latter two are already identified as importantstakeholders.

Table 5 provides a holistic picture (but not exhaustive) of thecontributions of various stakeholders—in reducing or removingcertain barriers across the stages—that can help to facilitate asmooth process transition to energy efficiency adoption. In otherwords, stakeholders help to strengthen the link between stages.

6.4. A possible contribution to policymakers—using MCIR to assess

policy effectiveness

The systems thinking approach has helped to develop the MCIRframework. Although the MCIR framework reflects the process ofadoption of energy efficient technologies by organizations, givenits generic nature, policymakers can also use it to analyze energyefficiency shortfalls in the industrial sector of their countries.

Depending on the prevalence of the type of barriers, the countrycould be facing ‘‘Motivational’’, ‘‘Capabilities’’, ‘‘Implementation’’ or‘‘Results’’ barriers, or it could also be a combination of two or morecategories. Such an analysis gives clues to the weakest link in theframework, which then aids governments to determine the type ofpolicies to introduce. The following simplified scenario depicted byFig. 5 illustrates how the framework may help policymakers. Thevertical axis shows the number of organizations involved in eachstage of the adoption. It is implied that the higher the level of thestage, the greater the number of organizations are involved, thusthe fewer the barriers faced in that stage and higher adoption ofenergy efficient technologies and practices.

Fig. 5 depicts a situation where a large number of organizationsare motivated to pursue energy efficiency, as indicated by the highvertical in the ‘‘Motivation’’ column. However, these organizationslack technical capabilities among these organizations, as shown bythe partially shaded column. To strengthen the link from one stageto another, the government should first try to raise the capabilitiesof the organizations for energy efficiency, such as by promotingESCOs or providing energy efficiency and management training. If

Motivation Capability Implementation Results

1 2 3

No.

of

orga

nisa

tions

Fig. 5. Big gap between the number of organizations which are motivated and that

with EE capabilities. Possible solutions: (1) build the ESCOs industry, provide financial

grants and incentives, (2) enforce implementation, (3) monitor and track returns.

Table 5The roles of stakeholders.

Motivation Capability Implementation Results

Government� Voluntary agreements

� Education and awareness

� Regulations and legislations

� Financial grants and incentives

� Provision of energy manager

training

� Target setting

� Benchmarking

� Provision of network platforms

� R&D of energy efficient

technologies

� Standard reporting protocol to account for

economic benefits of EE improvements

Industrial organizations

� Corporate Social

Responsibility (CSR)

� Meeting employees’

expectations

� Energy audits

� Engage consultancy

� Overcome resistance to change/

alignment to values

� Target setting

� ISO 50000

� Outsourcing

� R&D of energy efficient

technologies

� Energy data collection and monitoring

ESCOs

� Energy audits and

improvement

recommendations

� Sharing of best practices

� Follow up sessions

� ISO 50000

� Benchmarking

� Lean and Six Sigma

� Techniques or tools to measure and quantify

benefits of EE

Customers

� Demand for ‘‘green’’ product

(lower carbon footprint)

K.-H. Chai, C. Yeo / Energy Policy 46 (2012) 460–472 471

‘‘Implementation’’ barriers exist after building capabilities, thegovernment can enforce implementation energy efficiency actions.Table 5, presented earlier, lists a few examples of governmentprograms in each stage of the MCIR framework.

7. Conclusion

This paper reviewed the various classifications of barriers toenergy efficiency in the literature and proposed a systems thinkingperspective to barrier analysis by considering interactions betweenthe barriers and different categories of barriers. Other elementssuch as stakeholders and government policies were also taken intoperspective, resulting in a process-oriented, sequential, closed-loopframework that was introduced to increase energy efficiencyadoption. The framework, which we termed the MCIR framework,consists of four stages connected in series: ‘‘Motivation’’, ‘‘Capabil-ity’’, ‘‘Implementation’’ and ‘‘Results’’. The outputs from ‘‘Results’’form a feedback loop into ‘‘Motivation’’ where positive results(demonstration of energy savings) sustain energy efficiency adop-tion. The framework also reveals that the level of energy efficiencyis only as strong as the weakest link (between the stages).

Our paper makes three important contributions. Firstly, byadopting a systems perspective, our proposed holistic framework

takes into account the relationship between the barriers based onthe process of energy efficiency implementation. The feedbackeffect that existing implementation has on future energy effi-ciency is explicitly recognized. This is different from previousstudies which traditionally treated barriers in an isolated andpiecemeal manner. Secondly, our framework can be used a policyguiding tool to analyze stages which need improvements, asshown in Section 6.4. This is an important contribution because,to the best of our knowledge, it is the first systematic method foranalyzing shortcomings in energy efficiency policy. Thirdly, ourframework, when extended to include the stakeholders, showsthe roles and responsibilities of the stakeholders involved inimplementing energy efficiency. This big picture view allowspolicymakers to formulate policies and actions which can helpto establish the necessary stakeholders so that they can contri-bute to the specific stage of energy efficiency.

8. Limitations and future research

The conceptual framework proposed in this paper presents anovel perspective on the relationships between the energyefficiency barriers in the industrial sectors. However, as it is builtupon inductive research approach through literature review andretrospective cases, the framework needs to be tested using theconventional hypothesis testing methodology (e.g. a large scalesurvey) in different industries. Future research in this directionwill be needed in other to advance and validate the framework.In addition, refinement may be needed for other sectors (e.g.residential) because of the different dynamics between thevarious stakeholders involved.

Acknowledgments

The authors would like to acknowledge and thank colleaguesB.W. Ang, K.G. Neoh, M. Quah, Elspeth Thompson, Neil Sebastiand’Souza and W.H. Chua for their useful comments and

K.-H. Chai, C. Yeo / Energy Policy 46 (2012) 460–472472

contributions to the completion of this work. We are grateful tothe Energy Studies Institute for funding this project.

References

Adamides, E.D., Mouzakitis, Y., 2009. Industrial ecosystems as technologicalniches. Journal of Cleaner Production 17 (2), 172–180.

Adler, P., 1989. When knowledge is the critical resources, knowledge managementis the critical task. IEEE Transcations on Engineering Management 36 (2),87–94.

Anderson, S.T., Newell, R.G., 2004. Information programs for technology adoption: thecase of energy-efficiency audits. Resource and Energy Economics 26 (1), 27–50.

Anh, N.S., Samuel, N., et al., 2009. Competition in the shipbuilding industry,sustaining competitive advantage in the global economy. SolBridge Interna-tional School of Business.

Bawden, R.J., Ison, R.L., Macadam, R.D., Packham, R.G., Valentine, 1985. A researchparadigm for systems agriculture. In: Remenyi, J.V. (Ed.), Agricultural SystemsResearch for Developing Countries. ACIAR, Canberra, pp. 31–42.

Bennett, S.J., Pearson, P.J.G., 2009. From petrochemical complexes to biorefineries?The past and prospective co-evolution of liquid fuels and chemicals produc-tion in UK. Chemical Engineering Research and Design 87 (9), 1120–1139.

Boulding, K., 1956. General systems theory: the skeleton of science. ManagementScience 2 (3), 197–208.

Brown, M.A., 2001. Market failures and barriers as a basis for clean energy policies.Energy Policy 29 (14), 1197–1207.

Bunse, K., Vodicka, M., et al., 2010. Integrating energy efficiency performance inproduction management—gap analysis between industrial needs and scien-tific literature. Journal of Cleaner Production 19 (6–7), 667–679.

Checkland, P.B., 1981. Systems Thinking, Systems Practices. Wiley, Chichester, UK.Chi, K., Nuttall, W.J., et al., 2009. Dynamics of the UK natural gas industry: system

dynamics modelling and long-term energy policy analysis. TechnologicalForecasting and Social Change 76 (3), 339–357.

Driscoll, P.J., 2008. Systems Thinking. Wiley, Hoboken, NJ.Egenhofer, C., 2007. The making of the EU emissions trading scheme: status,

prospects and implications for business. European Management Journal 25 (6),453–463.

Eisenhardt, K., 1989. Building theories from case study research. Academy ofManagement Review 14 (4), 532–550.

Energetics, Incorporated and E3M, Incorporated, 2004. Energy Use, Loss andOpportunities Analysis: U.S. Manufacturing and Mining, for U.S.. Departmentof Energy Energy Efficiency and Renewable Energy Industrial TechnologiesProgram.

Freeman, R., Tryfonas, T., 2011. Application of systems thinking to energy demandreduction. In: Proceedings of the 6th International Conference on System ofSystems Engineering, Albuquerque, New Mexico, USA.

Geller, H., Harrington, P., et al., 2006. Polices for increasing energy efficiency:thirty years of experience in OECD countries. Energy Policy 34 (5), 556–573.

Gielen, D.J., de Feber, M.A.P.C., et al., 2001. Biomass for energy or materials? AWestern European systems engineering perspective. Energy Policy 29 (4),291–302.

Gill, J., Johnson, P., 1991. Research Methods for Managers. Paul Chapman Publish-ing, London.

Gillingham, K., Newell, R.G., et al., 2009. Energy Efficiency Economics and Policy.Hasanbeigi, A., Menke, C., et al., 2009. Barriers to energy efficiency improvement

and decision-making behavior in Thai industry. Energy Efficiency, 33–52.Hendel-Blackford, S., Angelini, T., et al., 2007. Final Report Energy Efficiency in

Lifestyles: Europe and Japan.Howarth, R.B., Andersson, B., 1993. Market barriers to energy efficiency. Energy

Economics 15 (4), 262–272.Jaffe, A.B., Stavins, R.N., 1994. The energy-efficiency gap. What does it mean?

Energy Policy 22 (10), 804–810.Kounetas, K., D., Skuras, et al., 2010. Promoting energy efficiency policies over the

information barrier. Information Economics and Policy, 23, 72-84.Lin, J., Ng, T.S., 2009. A systems approach for managing concurrent product

development projects. In: Proceedings of the 3rd Asia-Pacific Conference onSystems Engineering, Singapore.

Matthews, R.B., Kropff, M.J., Horie, T., Bachelet, D., 1997. Simulating the impact ofclimate chnahe on rice production in Asia and evaluating options for adoption.Agricultural Systems 54, 399–426.

Miles, M.B., Huberman, A.M., 1994. Qualitative Data Analysis. Sage, Thousand Oaks.Mingers, J., White, L., 2010. A review of the recent contribution of systems thinking

to operational research and management science. European Journal of Opera-tional Research 207 (3), 1147–1161.

Nagesha, N., Balachandra, P., 2006. Barriers to energy efficiency in small industryclusters: multi-criteria-based prioritization using the analytic hierarchy pro-cess. Energy 31 (12), 1969–1983.

Ngai, E.W.T., et al., 2011. Development of the conceptual model of energyand utility management in textile processing: soft systems approach.International Journal of Production Economics http://dx.doi.org/10.1016/j.ijpe.2011.05.016.

Onut, S., Soner, S., 2007. Analysis of energy use and efficiency in Turkishmanufacturing sector SMEs. Energy Conversion and Management 48 (2),384–394.

Owens, S., Driffill, L., 2008. How to change attitudes and behaviors in the contextof energy. Energy Policy 36 (12), 4412–4418.

Palm, J., Thollander, P., 2010. An interdisciplinary perspective on industrial energyefficiency. Applied Energy 87 (10), 3255–3261.

Perry, C.J., 1996. The IIMI Water Balance Framework: A Model for Project LevelAnalysis. Research Report 5. International Irrigation Management Institute(IIMI), Colombo, Sri Lanka.

Pires, A., Martinho, G., Chang, N., 2011. Solid waste management in Europeancountries: a review of systems analysis techniques. Journal of EnvironmentalManagement 92, 1033–1050.

PNNL, 2006. Opportunity Analysis for Recovering Energy from Industrial WasteHeat and Emissions. Pacific Northwest National Laboratory (PNNL),Washington.

Qudrat-Ullah, H., Seong, B.S., 2010. How to do structural validity of a systemdynamics type simulation model: the case of an energy policy model. EnergyPolicy 38 (5), 2216–2224.

Ren, T., 2009. Barriers and drivers for process innovation in the petrochemicalindustry: A case study. Journal of Engineering and Technology Management26, 285–304.

Rogers, E., 1995. Diffusion of Innovation. The Free Press, New York.Rietbergen, M.G., Farla, J.C.M., et al., 2002. Do agreements enhance energy

efficiency improvement? Analysing the actual outcome of long-term agree-ments on industrial energy efficiency improvement in The Netherlands.Journal of Cleaner Production 10 (2), 153–163.

Rohdin, P., Thollander, P., 2006. Barriers to and driving forces for energy efficiencyin the non-energy intensive manufacturing industry in Sweden. Energy31 (12), 1836–1844.

Rohdin, P., Thollander, P., et al., 2007. Barriers to and drivers for energy efficiencyin the Swedish foundry industry. Energy Policy 35 (1), 672–677.

Rountree, J.H., 1977. Systems thinking—some fundamental aspects. AgriculturalSystems 2, 247–254.

Sardinou, E., 2008. Estimating space heating determinants: An analysis of Greekhouseholds. Energy and Buildings 40 (10), 1084–1093.

Senge, P.M., 1990. The Fifth Discipline: The Art and Practice of the LearningOrganization. Century Books, London.

Smith, A., Voß, J.-P., et al., 2010. Innovation studies and sustainability transitions:the allure of the multi-level perspective and its challenges. Research Policy39 (4), 435–448.

Sorrell, S., Schleich, J., Scott, S., O’Malley, E., Trace, F., Boede, U., Ostertag, K.,Radgen, P., 2000. Barriers to Energy Efficiency in Public and Private Organiza-tions. Final Report to DG Research Under the Project ‘Barriers to EnergyEfficiency in Public and Private Organizations’. SPRU—Science and TechnologyPolicy Research, University of Sussex, Brighton.

Stephens, W., Hess, T., 1999. Systems approaches to water management research.Agricultural Water Management 40, 3–13.

Stephenson, J., Barton, B., et al., 2010. Energy cultures: a framework for under-standing energy behaviors. Energy Policy 38 (10), 6120–6129.

Strauss, A., Corbin, J., 1998. Basics of Qualitative Research. Sage Publications, Inc.Thollander, P., Danestig, M., et al., 2007. Energy policies for increased industrial

energy efficiency: evaluation of a local energy program for manufacturingSMEs. Energy Policy 35 (11), 5774–5783.

Thollander, P., Ottosson, M., 2008. An energy efficient Swedish pulp and paperindustry—exploring barriers to and driving forces for cost-effective energyefficiency investments. Energy Efficiency 1, 21–34.

Thollander, P., Dotzauer, E., 2010. An energy efficiency program for Swedishindustrial small- and medium-sized enterprises. Journal of Cleaner Publication18, 1339–1346.

UNEP, 2006. Barriers to Energy Efficieny in Industry in Asia. United NationsEnvironment Program.

Uphoff, N., 1996. Understandng the world as a hetergeneous whole: insights intosystems from work or irrigation. Systems Research 13, 3–12.

von Bertalanffy, L., 1950. The theory of open systems in physics and biology.Science 111, 23–29.

Wang, G., Wang, Y., et al., 2008. Analysis of interactions among the barriers toenergy saving in China. Energy Policy 36 (6), 1879–1889.

Weber, L., 1997. Some Reflections on Barriers to the Efficient Use of Energy. EnergyPolicy 25 (10), 833–835.

Worrell, E., Bernstein, L., Roy, J., Price, L., Harnisch, J., 2009. Industrial energyefficiency and climate change mitigation. Energy Efficiency 2, 109–123.

Yin, R., 1989. Case Study Research: Design and Methods. Sage Publishing, BeveryHills, CA.