Journal of Cleaner Production 69 (2014) 133e142

Contents lists available at ScienceDirect

Journal of Cleaner Production

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

Triggers and barriers to energy efciency measures in the ceramic,

cement and lime sectors
Frank Venmans a, b, *
a
b

Waroqu Faculty of Economics and Management, UMons, 17 Place Waroqu, 7000 Mons, Belgium
Grantham Institute, London School of Economics and Political Science (LSE), Tower 3, Clements Inn Passage, London, UK

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 5 May 2013
Received in revised form
23 January 2014
Accepted 23 January 2014
Available online 3 February 2014

Evaluating and understanding the interplay of barriers to the diffusion of energy efciency measures is
highly relevant because, if policies are effective in overcoming these barriers, CO2 emissions can be
decreased at low cost. The study aims to understand how managers make decisions to invest in energy
efciency, how perceived barriers affect these decisions and how policy can overcome these barriers. We
apply neo-classical economic theory as well as insights from transaction cost economics and behavioural
economics to understand why hurdle rates, even when omitted costs and risk are taken into account, are
higher than the weighted average cost of capital. We nd that internal capital budgeting rules and the
effort of studying technical feasibility and protability are relevant to understanding the efciency gap.
The results indicate that the voluntary agreement and the emission trading scheme are complementary,
addressing different barriers in different contexts.
2014 Elsevier Ltd. All rights reserved.

Keywords:
Barriers
Efciency gap
Energy efciency
Emission trading
Hurdle rate
Voluntary agreement

1. The energy efciency paradox and the denition of barriers

Since the oil crises of 1973 and 1979, there has been an intense
academic debate around the existence of non-implemented costeffective energy efciency measures, known as the energy efciency gap. The International Energy Agencys World Energy
Outlook (IEA, 2012) estimates that economically viable energy efciency measures have the potential to halve the increase in world
primary energy demand by 2035 compared to the present New
Policies Scenario which already includes the current energy efciency policies. Note that estimating the macroeconomic effect of
energy efciency is difcult because it is partly neutralised by the
feedback of an increased purchasing power, known as the rebound
effect (Antal & van den Bergh, 2013). In addition to opportunities
available to families and public utilities, many studies nd important unrealised cost-effective investment opportunities in energyefciency within rms (DeCanio, 1998; Schleich, 2009; Sola and
Xavier, 2007).

Abbreviations: EU ETS, European Union Emission Trading Scheme; IRR, Internal

Rate of Return; MNC, Multinational Corporation; NPV, Net Present Value; WACC,
Weighted Average Cost of Capital.
* Waroqu Faculty of Economics and Management, UMons, 17 Place Waroqu,
7000 Mons, Belgium. Tel.: 32 65 37 32 13.
E-mail addresses: f.venmans1@lse.ac.uk, Frank.venmans@umons.ac.be.
0959-6526/$ e see front matter 2014 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jclepro.2014.01.076

Understanding barriers to energy efciency measures within

rms is important because addressing these barriers may yield
emission reductions at a very low cost. However, since the investment strategies of rms are typically the outcome of a complex
decisional process, a better understanding of barriers to energy
efciency is critical to the successful implementation of regulatory
policies and their evaluation.
There is considerable consensus on the existence of neoclassical barriers to energy efciency such as externalities, asymmetric information and incompleteness of contracts. However,
other barriers, such as organisational barriers and capital availability, and with them the size of the energy efciency gap, has
been a subject of polemical academic debate.
Neo-classical economists argue that important hidden costs
(hidden to the analyst, but not to the rms), such as overall energy
management costs, the cost of studying investment opportunities,
foregone option values, explain a large proportion of an apparent
energy efciency gap (Stavins et al., 2007; Sutherland, 1996; Jaffe
et al., 2004). Also, uncertainty is seen as inherent to the adoption
of new technology and may justify a higher risk-adjusted discount
rate than is typically used in calculations that suggest the existence
of an efciency gap. Limited access to capital is not seen as a barrier
but as an efcient market outcome. Capital markets certainly
constrain the allocation of capital, particularly to more risky borrowers, but this allocation is a source of efciency, not inefciency.

134

F. Venmans / Journal of Cleaner Production 69 (2014) 133e142

Furthermore, there is no reason to believe that capital markets systematically misallocate capital in ways that discriminate against energy conservation investments. (Sutherland, 1996) They conclude
that policy intervention is only justied when there are neoclassical market failures, such as negative externalities and asymmetric information.
The neo-classical stance has claried the debate by arguing that
indirect but unavoidable costs or risks related to the diffusion of
technology cannot be seen as economic barriers to energy efciency. However, following Sorrell et al. (2004), we consider costeffectiveness from different perspectives, including insights from
transaction cost economics and behavioural economics.
Transaction cost economics stresses the importance of routines,
rules of thumb and satiscing heuristics rather than maximising
heuristics in order to reduce the cost of making complex decisions
(Williamson, 1988, 2002). Transaction cost economics also emphasises that policy intervention and different institutional structures may lower transaction costs. Corporate cultures exhibit a rich
diversity of forms and show large variations in performance.
Managers struggle with the complicated challenge of getting the
most out of the resources they deploy. Dealing with problems of
agency, moral hazard, imperfect information and design of incentives is at the heart of modern management. These difculties
affect all aspects of the rm including capital structure, investment
budgeting, operational control, strategic positioning as well as investment decisions related to energy efciency. This contrasts with
the assumption of the neo-classical stance that under competitive
market conditions rms can be expected to rationally maximise
their value and nd a governance structure such that organisational
barriers to protable investments are negligible. DeCanio (1998)
nds statistical evidence of organisational and bureaucratic barriers among 1400 US rms participating in an efcient lighting
programme. Several other qualitative studies focused on organisational barriers to energy efciency in rms (DeCanio, 1993; Sola
and Xavier, 2007; Zilahy, 2004).
Behavioural economics departs from the assumption of rational
agents, pointing at systematic biases in human decision-making
(Kahneman and Tversky, 2000). Behavioural biases further increase the scope of barriers to energy efciency and highlight
barriers that could not be apparent in a neo-classical approach
based on a rational actor model (Sorrell et al. 2004; Weber, 1997).
Barriers are discussed according to the taxonomy developed by
Sorrell et al. (2004) (see also Fleiter et al. 2011; Schleich, 2009;
Schleich and Gruber, 2006). This taxonomy has been chosen
because it is particularly appropriate for our approach to different
theoretical frameworks, since each barrier can be considered according to neo-classical economics, transaction cost economics and
behavioural economics.
1. Hidden costs: companies may not invest in energy efciency
because there are costs that are hidden to the researcher, but not
hidden to the company. Managers may not include these costs in
protability calculations because they are difcult to quantify.
Certain hidden costs such as search costs, purchasing and procurement costs can be considered as barriers in the sense that
they may be avoidable under an alternative organisational
structure or policy. For example, the cost of information gathering, which may be seen as a hidden cost justifying the
apparent efciency gap in a neo-classical economic framework
(Jaffe et al. 2004 p.84), may be reduced by subcontracting with a
specialised audit bureau or by adopting a policy imposing energy efciency labelling.
2. Risk and uncertainty: a high discount rate in protability calculations may be a rational way of compensating for technical
risk, regulatory uncertainty or energy price uncertainty.

3.

4.

5.

6.

However, market design, contract structure, organisational

structure, policy etc. affect the risk prole of energy efciency
investments. Or if actors treat uncertainty in investment decisions in a biased way, as highlighted by behavioural economics, then the risky nature of a project may become an
economic barrier to a cost-effective outcome (Greene, 2011).
Imperfect information: when the technical feasibility or the
protability of an investment is not studied, cost-effective investments opportunities may remain non-implemented.
Organisational barriers such as lack of managerial time may
aggravate the role of imperfect information.
Split incentives: if the energy performance of equipment
installed by a subcontractor is unobservable or difcult to
enforce legally, this creates an incentive for the subcontractor to
build cheaper equipment with poorer energy performance. In
large organisations, different people and divisions may not be
accountable for their energy consumption or savings, impeding
incentives for energy-saving measures.
Capital budgeting: the transaction costs related to proving a
rms creditworthiness may limit access to extra loans for costefcient investments. Agency problems between managers and
shareholders may lead to internal capital budgeting rules
reducing the investment budget available to managers.
Bounded rationality: in contrast to orthodox neo-classical theory, transaction cost economics studies how people economise
on cognitive efforts of processing information leading to satiscing rather than maximising decision heuristics. Behavioural
economics stresses systematic biases in human decisionemaking, such as reference dependence, status quo bias and timeinconsistent discounting.

Note that these barriers may overlap, co-exist and interact, so

that a phenomenon may fall under more than one barrier category.
We study barriers and motivations in the ceramic, cement and
lime industries which are very energy-intensive and carbonintensive industries. To the best of our knowledge, this is the rst
study of energy efciency barriers in these sectors. Moreover, there
are only a few papers on barriers to energy efciency in energyintensive industries (de Groot et al. 2001; Sardianou, 2008;
Zilahy, 2004). Energy-intensive industries are of particular interest, not only because potential efciency gains can be considerable,
but also because many studies explain barriers by the fact that
energy is not part of their core business (Sorrell et al. 2004; Schleich
and Gruber, 2006), which is not the case for our sample. To the best
of our knowledge, this is the rst energy efciency study covering
an entire (over 95%) country-wide sector. Moreover, the bricks,
cement and lime sectors are important export sectors in the Belgian
economy. The companies investigated emit around 15% of Belgian
emissions that are covered by the European Union Emission
Trading Scheme (EU ETS).
In part 2 the methodology and data are discussed. Part 3 describes the results, structured according to the above-mentioned
barriers. Next we evaluate the impact of the EU ETS and the
voluntary agreement on efciency investments.
2. Methodology and data
The study focuses on the decision-making process leading to
investments in energy efciency to address the following research
questions: First, the study aims to evaluate the relative importance
of different barriers and to understand why certain barriers are
more important than others. Given our previous denition of barriers, it also questions to what extent insights from transaction cost
economics and behavioural economics are relevant for understanding the energy efciency gap. Next, the study investigates

F. Venmans / Journal of Cleaner Production 69 (2014) 133e142

how barriers relate to each other and the relationship between

different types of barriers and environmental policies. This is done
by a descriptive approach as well by a more formal approach, using
principle component analysis.
The main objective is not to quantify the potential of costeffective investments, but to understand why certain costeffective investments are postponed (or not implemented at all)
and why at a given moment cost-effective investments get a green
light. Since investment decisions are based on managers perceptions, the study is based on interviews and thus on perceptions of
managers. We do not focus on whether perceptions are in accordance with reality on the shop oor, but we focus on the consistency of perceptions and their departures from assumptions in
standard economic models. This is in line with the above denition
of barriers, encompassing insights from transaction cost economics
and behavioural economics. The role of satiscing heuristics, rules
of thumb, organisational structure and behavioural considerations
are intrinsically linked to perceptions of managers. Therefore the
choice was made for a multiple case study of 16 companies based
on a qualitative methodology. Economists have a good understanding of different theoretical concepts to estimate the value of
projects, including very complex option value models. Much less is
known about the way investment decisions are made in practice,
the decisional process that explains why certain companies make
certain investments or apply certain technologies and others do not
(Brealey et al. 2010). This also motivated the choice for a qualitative
methodology analysing managers perceptions.
Bewley (2002) advocates the use of loosely structured interviews instead of questionnaires, particularly in areas where
economic theory falls short of explaining observed behaviour. So an
important part of the interview was based on semi-structured
questions on organisational structure, risk, reliable information,
capital availability, contracts with suppliers of technology, the effect of policy etc. Next, in order to make the discussion as concrete
as possible, 13 motivations and 16 barriers were discussed for the
most relevant past or future potential investments in energy/carbon efciency. This helped to focus the interviews on empirical
experience, to go beyond mere intentions and to validate and
deepen the correct understanding of the rst part of the interview.
Interviews are processed both with a descriptive approach and
with a more formal approach based on scores. Managers scored
motivations and barriers on the following scale: very
important 3; important 2; moderately important 1; not
important 0. By reporting means and correlations, these values
are interpreted in a cardinal way which is a common way of
interpreting studies on barriers to investments (de Groot et al.
2001; Sola and Xavier, 2007; Sorrell et al. 2004). The interaction
between policies and barriers is analysed with a principal component analysis. The principal component analysis is applied as a tool
to observe clusters of variables and observation points, based on
correlations, without imposing hypotheses on the structure of
causal relationships.
In total, 53 energy efciency projects (34 completed projects
and 19 potential future projects) were discussed in detail. The selection of projects was based on managers perceptions of the most
important investment opportunities. Participating companies lled
in a preliminary questionnaire on their energy performance. This
questionnaire included an evaluation of the above-mentioned
motivations and barriers for the company in general, which resulted in company-level data for 11 companies. The preliminary
questionnaire helped to prepare the interviews and to ask for
clarications, should there have been any apparent contractions
between answers during the interviews.
In three out of the 16 companies investigated, two managers
were present during the interview. Both managers participated in

135

one single interview, to allow for discussion in case of divergent

opinions and to avoid reluctance to cross checking. In addition to 19
plant managers or environmental managers, all of whom were
responsible for investments with energy efciency gains of several
million EUR, four representatives from the professional federations
and one person from an equipment supplier were interviewed to
allow for cross examination. Interviews lasted between 1 h 150 and
2 h and were conducted between May and November 2012. Texts
were transcribed and structured with the software program NVivo.
Information from Belrst (Bureau Van Dijks accounting data) and
annual reports, when available, were also included to allow for
further triangulation.
All but 2 groups of the Belgian construction materials sector are
included: 9 Belgian brick producing companies or groups, three
companies producing other baked construction materials, two
cement and two lime producing groups. The companies are of
diverse sizes. Value added varied between 7 and 50 million EUR per
plant (plus one extra small company achieving a value added of 1.5
million EUR). Ceramic plants are typically smaller than cement and
lime plants. Moreover, nine plants investigated are part of a multinational corporation (MNC) with a worldwide turnover above one
billion EUR (Table 1).
Various studies describe the potential for efciency measures
and best available techniques in the ceramic sector (BREF, 2007;
Zakharov et al. 2009), in the cement sector (BREF, 2010; Madlool
et al. 2011; Pardo et al. 2011; Moya et al. 2011) as well as in the
lime sector (BREF, 2010). These studies helped to direct questions
onto the completed and potential efciency measures.
3. Results and discussion
Tables 2and 3 show the relative importance of motivations and
barriers for the companies in general, for their completed projects
and their potential future projects.
According to standard economic theory, a project adds value to
the company if it has a positive net present value (NPV). The NPV is
the sum of all cash ows of a project, discounted at a risk-adjusted
discount rate (the discount rate may vary over time). If the risk of a
project is constant from one period to another and corresponds to
the risk of the entire company, then the risk-adjusted discount rate
equals the weighted average cost of capital (WACC), i.e. the expected return of all equity and liabilities. The internal rate of return
(IRR) is the discount rate for which the NPV of a project equals zero.
So the observation of an efciency gap boils down to the presence
of non-implemented investment opportunities with a positive NPV
or an IRR above the risk-adjusted discount rate. In practice, all
companies used the payback time as the most important criterion
for project evaluations, which is simpler to calculate but theoretically less appealing. A payback time is unaffected by the cash ows
after the payback and indifferent to the exact moment at which
cash ows arrive before the payback.
3.1. Hidden costs
Some costs may be very difcult to estimate even if they are
known to exist. If protability calculations are upward biased
because such costs are omitted, then the efciency gap may be only
apparent and the outcome may be economically efcient. Therefore, the evaluation of hidden costs is important for giving a sense
to other potential barriers. Also, certain hidden costs are barriers to
a cost-efcient outcome because they could be avoided in a
different organisational or regulatory context.
All managers were asked to describe typically omitted costs in
their payback time or IRR calculations. A minority of the managers
estimates that they include almost all costs, while a majority

136

F. Venmans / Journal of Cleaner Production 69 (2014) 133e142

Table 1
Characterisation of plants interviewed according to international scope, turnover and emission allocation.
Value added per Range of yearly
Fuel
Part of MNC Not part of
allocated emissions
MNC (number plant 2011
(number
per plant (t CO2)
(million V)
of plants)
of plants)

% CO2 emissions
from decalcination

7e25
One case of 1.5

10,000e80,000

2
Other
construction
materials
Cement
4
& Lime

7e30

30,000e70,000

Generally below 5% 16%

Exceptionally
up to 35%
Between 5%
n.a.
and 50%

25e50

Bricks

a
b
c

Natural gas

2 Natural gas
1 petroleum
coke
400,000- 1,500,000 Coal/Lignite/
Above 60%
waste/biomass

Belgian production
Belgian net
2009e2011 compared exports 2011 as %
to 2006e2008a
of productiona

Cement
10%
Lime
10%b

15%

n.a.

Cement 5%
Lime positivec

Annual Report 2011 Fdration Belge de la Brique, Febelcem, Fediex.

Based on limestone extraction.
Exact numbers are not available.

estimates that some costs such as overhead energy management

costs, costs of production disruption, costs of tendering are not
included in payback times. But these costs were perceived to be of
minor importance.
Several advantages from investments were not included
either. A typical example was that no single brick producer
included the advantage of carbon credits from the EU ETS in its
payback or IRR calculations, although between 2009 and 2011,
with a carbon price oscillating around 15 EUR/ton and an industrial methane price around 25 EUR/MWh, the advantage of extra
allocations increased the protability of gas energy efciency by
12%. Most investments -such as the upgrading of kilns and dryersfacilitate energy management, lower maintenance costs and increase product quality. These advantages are typically strong
motivations, but are not formally included in payback times
either. So, in general, the perception of interviewees is that hidden costs are very limited when they calculate the protability of
energy efciency projects and in many cases are compensated by
omitted advantages. Note that this study is not designed to verify
the correctness of these perceptions. The aim of the paper is to
understand the decisional process leading to investments, which
is based on perceptions, rather than the unbiased reality behind
those perceptions.

3.2. Risk and uncertainty

In theory, protability estimates are based on expected values of
cash ows, i.e. uncertain but unbiased estimates of expenditures
and revenues. If one assumes efcient capital markets where investors possess a diversied portfolio of equity, then only uncertainty (volatility) correlated with the market return is relevant. If
investors do not have a diversied portfolio, (which is the case for
several family-owned businesses in our sample), the notion of risk
of a project is the extent to which the project affects the volatility of
the prot of the entire company. This is in line with the fact that the
risk that was perceived to be the most important one was Low
demand risk: investments entail xed costs that may be costinefcient when there is overcapacity during economic downturns. This barrier ranked second, obtaining an overall score of
1.20 (between moderately important and important). However, low
demand risk generally does not cause a higher risk for the project
compared to the risk to the company as a whole, hence it does not
justify a required IRR (hurdle rate) higher than the WACC.
Risk of production disruptions, hassle, inconvenience was
perceived as a barrier with a low score of 0.62, between unimportant and moderately important (ranking 8th). Only three of the
34 completed projects had setbacks creating losses instead of gains.

Table 2
Relative importance of motivations that triggered (or will trigger) energy efciency measures. Effects are obtained by OLS of Y a b1Dummygas only heat source b2Dummyfuture
project b3Dummycompany-wide obs.*,**,*** denote signicance at the 10%, 5% and 1% level. Observations include 31 completed projects, 19 potential future projects and 11
company-wide observations (N 61). The differences of motivations between these categories are captured by b2 and b3.

Economic reasons
Increasing energy prices
Result of in-house R&D
Replace obsolete equipment/production expansion
Technology applied has become cheaper
Policy
Voluntary agreements
Expected gains or cost savings from Emission Trading Scheme
Avoid risk induced by future ETS (unknown prices, cap,
proportion of free allocations)
Mentality change induced by Emission Trading Scheme
Comply with legal obligations (e.g. environmental licence)
Management
Commitment by top management to an environmental policy
Environmental image building towards clients
Implementation of environmental management system
(targets, procedures, evaluation...)
External pressures (from NGO, media, local community...)

Rank

Mean
score (Y)

Standard dev.
Score (sY)

Effect only
gas vs mixed
fuels (b1)

Effect future
projects vs completed
projects (b2)

1
5
9
12

2.5
1.3
0.8
0.5

(0.7)
(1.3)
(1.2)
(0.8)

0.4**
0.3
0.1
0.0

0.2
0.2
0.4
0.0

4
6
7

1.4
1.1
0.9

(1.2)
(1.1)
(1.0)

0.3
1.1***
0.7***

0.6*
0.6**
0.6*

8
13

0.8
0.5

(0.8)
(0.8)

0.7***
0.3

0.3
0.4*

2
3
9

1.6
1.5
0.8

(1.0)
(1.0)
(0.9)

0.7***
0.4
0.4

11

0.5

(0.9)

0.3

0.3
0.2
0.2
0.0

Effect company-wide
obs vs completed
project (b3)
0.3
0.5
0.8**
1.1***
0.6
1.0***
0.8**
0.6**
1.3***
0.1
0.3
0.3
0.6*

F. Venmans / Journal of Cleaner Production 69 (2014) 133e142

137

Solar panels lie on the roof and you are virtually sure about it: at
that time it will be paid back... Even if you throw everything out of
that hall, the solar panels will stay on the roof and continue to
produce electricity and make money. Nobody will be bothered by
them. But when you make a new investment in efciency and X
years later you want to introduce a big modernisation, then youve
lost your heat recycling system... And then you get to what most of
the companies say: if you make the investment, we prefer a
payback time of 2 or 3 years, because thats the horizon that I can
foresee. But if you then think about 5, 6 or 7 years payback time,
nobody does that... The market changes so fast that we dont know
what we will be doing in 5 years time.

So even if this type of risk is sometimes a hidden cost (certain rms

do not include technical risk in protability calculations) the
managers perception is that it is of minor importance. Indeed
many investments are modernisations entailing higher product
quality and reducing the risk of hassle which typically increases
with the age of installations. Also, since technical risk is uncorrelated to the market return, it is diversiable and thus irrelevant for
investors with diversied portfolios.
Protability measures are typically calculated using a static
approach, which is appropriate for investments that cannot be
postponed such as replacement investments. However, if the expected NPV of the investment opportunity changes over time, as in
the case of rapid technology improvement, rms may gain by
waiting, justifying the inclusion of an option value in the NPV
calculation (Ansar and Sparks, 2009; Abadie et al. 2013; Van Soest
and Bulte, 2001). However, this was perceived to be of low
importance, because most measures concerned relatively mature
technologies. This explains why the barrier We have waited (are
waiting) to see if the application of the technology by other companies/plants turned (turns) out to be reliable and protable
ranked only 11th (score 0.6). Also the motivation the technology
has become cheaper was the next to the last motivation (score
0.6). Only one company gave an example of an advantage for
waiting, i.e. the case of software packages that come out every few
years.
Unknown future decisions may create a gain for waiting because
postponing the investment maintains the option not to invest until
future decisions are known. For example, if a plant is completely
rebuilt or shuts down, the past retrot investment is lost. This was
mentioned by one interviewee.

Note that if it is correct that uncertainty about future decisions

justies a tighter investment criterion, the idea of a payback time
that is smaller than the foreseeable future is far from what economic theory advocates.
In two companies where a future plant closure was an option,
investments were restricted to those that were strictly necessary to
continue operations. In the rst case, the (former) plant manager
did not know if there would be a successor. In the second case,
losses during the preceding years, as well as a repositioning of the
mother company, opened up the option to sell the division. Investment opportunities with payback times of 1 or 2 years were not
carried out because of this uncertainty, even though the option of
selling or stopping was likely not to occur in the short run, not even
during the payback time, which indicates that the option value was
largely overestimated.
Energy efciency investments may also reduce risk. Since energy efciency investments induce a learning process, they increase

Table 3
Relative importance of barriers that explain why the energy efciency measures were not implemented earlier or are not yet implemented. Effects are obtained by OLS of
Y a b1Dummygas only heat source b2Dummyfuture project b3Dummycompany-wide obs.*,**,*** denote signicance at the 10%, 5% and 1% level. Observations include 31 completed
projects, 19 potential future projects and 11 company-wide observations (N 61). The differences of barriers between these categories are captured by b2 and b3.
Rank

Information
Technical feasibility wasnt studied before
Protability wasnt studied before
Lack of reliable information about the measure
Capital availability
Other (non-energy-related) investments receive (d) prior nancing
Difculties in nding (internal or external) nancing for investments
During former investments, strict capital budgeting for a given project
squeezed out the energy-efcient option.
Risk and uncertainty
Low demand risk: efciency investments entail xed costs that may
be cost-inefcient when there is overcapacity during economic
downturns.
Risk of production disruptions, hassle, inconvenience
We waited (are waiting) to see if the application of the technology
by other companies/plants turned (turns) out to be reliable
and protable
Incentives
Individuals/departments were (are) not accountable for
energy/emissions (or could (can) not benet from
energy/emission savings).
Conicts of interests within the company
Organisation
Lack of management time
Need for training of employees or engagement of specically skilled
employees (e.g. energy auditor)
Insufcient integration of energy/emissions into operating,
maintenance or purchasing procedures
Energy manager lacks inuence
Technology lock-in
Technique was/is incompatible with other elements of the plant

Effect only
gas vs mixed
fuels (b1)

Effect future
projects vs completed
projects (b2)

Effect company-wide
obs vs completed
project (b3)

Mean
all obs (Y)

Standard
dev. All
obs (sY)

3
5
6

1.2
0.9
0.8

(1.2)
(1.0)
(0.9)

0.0
0.0
0.1

0.9**
0.2
0.1

1
4
9

1.5
1.2
0.6

(1.1)
(1.2)
(0.9)

0.0
0.2*
0.5**

0.8***
0.6*
0.6**

1.1***
0.7
0.1

1.2

(1.2)

0.5

1.2***

1.1***

8
11

0.7
0.6

(1.0)
(0.9)

0.5**
0.1

0.0
0.1

0.1
1.1***

15

0.2

(0.6)

0.0

0.0

0.7***

14

0.3

(0.6)

0.0

0.1

7
12

0.8
0.6

(0.9)
(0.9)

0.0
0.2

13

0.4

(0.6)

0.3*

0.2

0.6***

16

0.1

(0.4)

0.2**

0.1

0.1

10

0.6

(1.0)

0.3

0.3
0.5*

0.5

0.0
1.3***
1***

0.3
0.9***
0.5

0.6*

138

F. Venmans / Journal of Cleaner Production 69 (2014) 133e142

the exibility of future technological options. For example, several

companies experimented with lower moisture levels of raw materials. This creates the capability to develop more advanced drying
methods in the future. One company worked on a new patented
drying process. In this way, the energy efciency learning process
creates future exibility and opens up new strategic options, for
example if there were to be a steep increase in energy prices. This
exibility adds positive option value to an investment. Also, many
energy efciency investments increase product quality and thus
again reduce the risk of not keeping up with market trends towards
higher quality products.
To conclude, risk from unforeseeable demand was perceived by
the managers as a major barrier to investments, but this does not
justify a hurdle rate above the WACC. The technical risk of energy
efciency investments was generally perceived to be low, because
most of the investments concerned known technologies and when
R&D and learning were required, the knowledge obtained was
mostly considered as a valuable asset for future strategic positioning. In a few cases, there was a gain from waiting driven by
reduced exibility (lost option value) in the case of an investment,
which may justify a higher hurdle rate or lower payback time.

Why?
Because these are very small gains. And because our kilns
consume 80% of our energy, while their consumption over their
lifetime is not linear. Their specic energy consumption increases
over time.
Imperfect information increases the uncertainty over future
gains and thus the risk prole of efciency projects. However, Lack
of reliable information to be able to estimate feasibility/protability (score 0.84) ranked 6th, lower than the fact that the option
simply had not been studied before. The technological feasibility
hasnt been studied before ranked 3rd (score 1.2) and the profitability hadnt been studied before ranked 5th (score 0.85). This
indicates that organisational barriers are at play.
One explanation for not studying investment opportunities may
be Lack of management time which was perceived as a moderately important barrier (rank 7, score 0.78).
Of course you need people that look after it, who pull the project
forward and who foster the project, defend it and go for it. So why
[didnt we invest] earlier. At the end of the day, it is capacity .
Lack of managerial time?

3.3. Imperfect information

Concerning information on energy ows, interviewees said they
had a good view over the energy consumption of different components of the plants. Energy sub-metring is widely applied, with
the exception of two plants that had only one gas metre. But in
general, sub-metring has improved greatly during the last decade
and is recognised as a very useful tool for nding energy efciency
opportunities and assessing the energy efciency of new equipment ex post.
Certain investments, such as variable speed drivers, have a
transparent certied energy efciency, even though the efciency
decreases with the age of the motor. For other investments, for
example when the production process is modied, ex-ante information on the energy efciency of new equipment is difcult to
estimate. Lack of reliable information to be able to estimate
feasibility/protability ranked as the 6th barrier (score 0.84). This
is especially the case when new plant-specic equipment needs to
be designed, such as upgrades of kilns and dryers or decreased
moisture content.
So we said, well make it one channel, we close all leaks, and that
brought us to where we stand now, with 20% less electricity consumption. But that never was the initial aim, that we would obtain
20%. We did know that we would be obtaining something, but how
much, we didnt know at all.
Ex-post evaluation of the energy efciency of new equipment
and measures has increased by sub-metring which is in many cases
followed up continuously. Even so, variable production conditions
make even ex-post efciency assessment difcult.
In my opinion it is unveriable because it depends so much on the
production volume of the plant, which depends on the economic
context. You can see it at the sectorial level: when we are in an
economic downturn the energy indices deteriorate and vice versa
The voluntary agreement has revealed the complexity of the
protability of an investment. When we take the initial list, one
investment will deliver, for example, 1% energy savings, another 2%
energy savings ... We made dozens of investments and overall, our
energy consumption didnt decrease. It doesnt decrease at all, you
cannot see it.

Thats a permanent problem. We save costs, and what costs do you

save? Staff. And in the long run everybody has to choose what he
can take up.
The fact that not studying an opportunity is a barrier, also points
to the role of routines, as is clear from the following quote.
So before the voluntary agreement, why were potential cost savings not observed?
I think, these companies have made gigantic prots in the past and
so there was no trigger to change. When everything goes very well
and the prots come in, why make big changes? An extra person
involved in the production didnt bother anybody either.

3.4. Split incentives

Asymmetric information on the quality of an article between a
buyer and seller may lead to adverse selection, as the seminal paper
by Akerlof (1970) described for the lemon market. Since the energy
efciency of new equipment may be difcult to observe for a purchasing company, this may impede technology suppliers from
selling the most efcient equipment and may even drive the most
efcient equipment out of the market (Sorrell et al. 2004).
When the equipment is standardised and energy efciency is
labelled, such as for new motors or certain grinders, energy efciency is transparent and room for split incentives is limited. In
most cases, however, equipment is plant-specic and built on-site
by subcontractors. We encountered different types of subcontracting with increasing potential for split incentives.
In the rst type of contract, the company does the design and
the subcontractor builds the equipment following a highly specied plan, as for a new channel for hot air or installations that are
highly dependent on the rms R&D. As design is determined by the
principal (purchasing company), there is no agency problem.
In the second case, which was encountered most often, the
subcontractor does the designing and energy efciency is part of
the contract under standardised theoretical conditions that are,
however, difcult to enforce. This is generally the case for new
kilns, cogeneration installations, entirely new designs for heat
transfer from kiln to dryers. However, even if energy efciency is

F. Venmans / Journal of Cleaner Production 69 (2014) 133e142

not enforceable, its transparency is increased by an intense interaction between the staff of the purchasing company and the
subcontractor.
In principle you go to different kiln builders, you start tendering
according to capacity, temperature etc., and then you see specic
energy consumption in these offers. And then you see big differences. So you start talking: how do you get this result, how do you
get that result? And then you see that one of them has a more
advanced technological concept.
Many of the managers interviewed had high levels of technical
knowledge. So contractual energy performance, combined with
theoretical insights from the design, limits the scope for unobservable efciency delivered by the subcontractor. However, in one
case, that of an entirely new plant, energy efciency was unclear.
Do you know how much energy your new bricks will need?
No, we know our consumption in the past, but what it will be, we
dont know at all. There are certain agreements, certain guarantees
that the deliverer needs to honour, but it ends there as a matter of
fact.
In the third case, the subcontractor does the designing, and
energy efciency is not contractually xed. In this case, energy efciency can be deduced indirectly from technical specications
and/or be based on condence.
Split incentives within the company, between different departments or persons, entail barriers that are perceived as unimportant by managers. Conicts of interests within the company,
Individuals/departments are not accountable for energy and
energy manager lacks inuence were the 3 barriers with the
lowest importance among all barriers in the questionnaire. Since
energy is part of the core business of the plant, plant managers
were highly involved in studying energy-saving opportunities.
3.5. Capital budgeting
For all projects where a payback time or IRR was calculated, 65%
had a payback time of 3 years or lower. The 3-year payback time to
get the green light for an investment is recognised as a typical rule
of thumb for non-compulsory investments by all the companies.
These very low payback times are also reported by other authors.
DeCanio (1998) nds a median payback time of 3.1 years for
lighting-efciency investments in 1400 US rms. Anderson and
Newell (2004) nd a typical payback time threshold of one to
two years for 39,920 audited energy efciency projects in 9034
small and medium US enterprises. The managers of 800 European
manufacturing plants reported a mean payback time for projects
they considered but failed to adopt of 3 years and 10 months (3
years in ceramics plants and 4 years for cement plants) (Anderson
et al., 2011). A payback time of 3 years for an investment with
constant yearly maintenance costs and energy gains lasting 10
years corresponds to an IRR of 31% (or 33% for an installation lasting
20 years, the tendency to underestimate gains in the long run is a
disadvantage of IRR compared to NPV). This is much higher than
the real WACC.
A rst candidate for such stringent payback times would be
hidden costs or risk (costs and risks that are hidden to the
researcher but not to the manager). However, as argued above,
hidden costs and risk are unlikely to justify a doubling or tripling of
the payback time, which would be a cost-effective payback time. No
single interviewee mentioned hidden costs or risk as a justication
for low payback times.

139

Former studies (Sorrell et al. 2004; DeCanio, 1998) suggested

that the required protability for energy saving projects was much
more severe because energy was not considered to be part of the
core business of the company. In our study no single rm recognised that the required protability for energy projects was higher
than the required protability of other non-compulsory projects.
With energy accounting for around 30% of total costs, energy efciency is understandably considered as part of the core business.
Moreover, two groups had special budgets for energy efciency and
innovation which made investments in energy efciency slightly
easier than other investments.
All interviewees explained the low payback times or high hurdle
rates by capital budgeting. The importance of capital budgeting is
conrmed by the fact that Other (non-energy-related) investments receive(d) prior nancing was the most important
reason invoked for delaying energy investments (score 1.46).
During the discussions of the voluntary agreement, they [government negotiators] say You have a payback time of 2 or 3 years, why
dont you do it? And it is something they dont want to hear, and we
repeat time after time that you need to see the investment capacity
in its overall aspect. In groups such as ours we have closed investment budgets. And inside these budgets we have to cope with new
installations, stay in business to keep our installations running and
only then do we look at what our priorities are. We have quality
investments, investments for the security of our personnel etc.
At the same time, this group explains that there is no difculty
in obtaining extra capital.
Generally, the group has easy access to capital. It is more a
constraint that it imposes itself... In 2008 we have tightened
operational budgets, but the family has reinvested in the company.
They have found the necessary nancing... so we got through the
period without serious problems.
Only one company mentioned difculties in obtaining extra
bank loans for investment projects, because they had suffered
losses over the previous years. The other companies had no problem in obtaining new loans at normal market rates. There are
several theories in corporate nance which explain such internal
capital budgeting rules. Capital budget discipline is a way of
reducing agency costs caused by asymmetric information between
managers and shareholders (Jensen and Meckling, 1976). Capital is
available only for those projects with the highest protability so
that managers cannot misuse capital to enlarge the company at the
expense of lower protability. But it leads to the refusal of investment projects with genuine positive NPV and a possible suboptimal investment level.
A second theory explaining strict internal capital budgeting
rules is the pecking order theory (Myers and Majluf, 1984), arguing
that companies, especially family-owned groups, prefer to use
retained earnings instead of raising external nance, because the
latter creates the signal that the company might be overvalued.
We are a family-owned business. The family does not want to
contract long term loans...There is another department where an
important investment has been in the pipeline for a few years. And I
think that we are now starting to put money aside to be able to
cope with that future investment. So investments that used to take
place in other departments will have to present a much better
payback time before these investments will be accepted.
Thirdly, the preference for retained earnings over debt may also
be explained by the transaction cost theory (Williamson, 1988),

140

F. Venmans / Journal of Cleaner Production 69 (2014) 133e142

which states that for plant-specic investments such as energy

efciency measures, retained earnings have lower transaction costs
than debt. Because debt is plagued with the difculty of aligning
the interests of debt holders and managers.
Difculties in nding (internal or external) nancing for investments was the fourth most important perceived barrier (score
1.18). Since only one company had had difculties in obtaining
external nancing for investments, even during the crisis when
banks were deleveraging, this barrier mainly concerns internal
nancing i.e. self-imposed capital budgeting rules that limit the
access to capital. Even if these capital budgeting rules may have
been second best solutions to problems of asymmetric information,
changes in organisational structures may overcome these barriers.
The short-sightedness embodied in a 2-year payback or 30%
hurdle rate may be not so much a result of defective vision as it is
the unintended side-effect of corporate policies put in place for
entirely different reasons-to simplify control, create appropriate
incentives, or overcome moral hazard... Measures adopted to deal
with them may in the past have been the only feasible (second best)
solutions, but give the appearance of a failure to fully optimise the
productive resources available to the rm. If this is so, innovative
management should be able to improve on old methods and move
the rm closer to its protability frontier (DeCanio 1993).
So even though energy efciency is considered as an important
question in the companies investigated, the fact that energy efciency investments are in most cases optional (non-compulsory),
creates a high hurdle rate for energy projects, driven by internal
capital budgeting rules. This indicates that if policy can increase the
credibility of investment proposals for shareholders (or the international mother company), for example by involving independent
energy auditors in a voluntary approach, this could be helpful to
reduce the barrier of capital availability.
3.6. Bounded rationality
For a minority of the projects completed (18%), no measure of
protability was calculated, including a multi-million euro investment in a completely new plant:
Do you have any idea of a payback time, IRR, NPV, or is it more
based on feeling?
How did I read it again -being somewhat philosophical- if you
want to swim across the sea, you need to jump into the water, and
thats what we did and we will swim. If you only look at the water
you wont cross it. Thats somehow how it is.
Did you calculate an IRR or another measure of protability, or a
multiple of turnover. do you have a calculation of 2 pages, with a
conclusion that you will recover your costs?
I dont think thats so obvious. And we hope that we can cover our
costs.
For most projects however, a measure of protability was
calculated even if not studying the feasibility/protability of a
measure was a main barrier to earlier implementation.
No rm calculated net present values, which is, however theoretically, the most appealing approach. As a rule of thumb, all
companies used payback times, sometimes complemented by IRR
calculations. A payback time does not give information on the
optimal level of investment because there is no link with the WACC.
The insight that a payback time of 3 years corresponds to a very
high IRR was surprising for most managers. A manager of a very big

company assured us that no project was discussed with an IRR

above 20%, while he had said just before that 3 out of the 4 projects
had a payback time below 3 years. Yet a payback time of 3 years
with a lifetime of 10 years corresponds to an IRR of 31%.
When required payback times were converted into IRR hurdle
rates and compared to their much lower WACC, not a single
manager made the link with a potential long run suboptimal level
of investments. A popular justication among managers for a
high hurdle rate is that it is logical to require a high protability
from optional (additional) investments to compensate for investments with low protability that are necessary anyway. This
contrasts with the economic insight that any marginal project
with an expected protability higher than the risk-adjusted discount rate (approximated by the WACC) increases the protability of the rm. Most managers focus on project ranking and
therefore they do not question the link between the hurdle rate
and the optimal level of investment. In this sense, the way IRR
and payback times are generally interpreted underestimates the
role of capital budgeting rules as a barrier to cost-effective energy
investments.
4. Effect of policy on energy efciency
The ceramic, cement and lime sectors are regulated by the EU
ETS. All companies investigated are part of the EU ETS and all had
freely allocated emission allowances above veried emissions. In
addition to the EU ETS, energy efciency is directly targeted by a
voluntary agreement with the Walloon and Flemish regional governments. These voluntary agreements do not lead to additional
emission savings since, given the EU ETS cap, any reduction by one
company allows an extra emission by another company. The
objective of the voluntary agreements is to address cost-effective
abatement opportunities that are impeded by barriers to energy
efciency. In this sense they underpin the efciency of the EU ETS,
contributing to equal marginal abatement costs among companies
(Henriksson and Sderholm, 2009). In Wallonia, subsidised energy
audits had been conducted between 2003 and 2005 in all energyintensive plants. Participating companies agreed to carry out all
investments with a certain feasibility and a payback time below 5
years and all investments with an uncertain feasibility and a
payback time below 2 years. In Flanders, a benchmarking commission estimated the energy performance of the worlds top 10%
most energy efcient installations in 2003. Participating companies
committed themselves to equalling this world top 10% by starting
investments with an IRR above 15% before 2006. If this was not
enough, they undertook to make extra investments with an IRR of
at least the return on government bonds. Only one local brick plant
was not part of the voluntary agreement.
Among motivations to invest in energy/carbon efciency, Expected gains or cost savings from Emission Trading Scheme;
Avoid uncertainty induced by future ETS and Mentality change
induced by Emission Trading Scheme had an averagely important
score of 1.1, 0.9 and 0.8 respectively (Table 2). The importance of
voluntary agreements as a motivation for triggering past and future
projects obtained a higher score of 1.4 (Table 2).
To investigate the interactions between the two policies and
with the 7 most important barriers, we report a principal component analysis (Fig. 1) of the scores on the policies, barriers, a dummy
variable for companies with gas as the only heat source and a
dummy variable for future projects. The correlogram can be sent on
demand.
The most striking observation is that the voluntary agreement
and the ETS are complementary in the sense that each policy addresses different projects, linked with different barriers. In the
principal component analysis, the cluster of the 3 ETS variables and

F. Venmans / Journal of Cleaner Production 69 (2014) 133e142

COVENANT

B3_TECHNIC_NOT_STUDIED

Component 2 (20.4%)

B7_TIME_MANAGEMENT
B5_PROFIT_NOT_STUDIED
B6_RELIABLE_INFO
GAS_ONLY

FUTURE

B4_DIFFICULT_RAISE_FUNDS

ETS_MENTALITY_CHANGE

B1_OTHER_INVEST_PRIORITY
B2_DEMAND_RISK

ETS_UNCERTAINTY

ETS_GAIN_LOSS

Component 1 (25.2%)

Fig. 1. Principal Component Analysis of scores on voluntary agreement (covenant), ETS

(expected gain/loss; uncertainty; mentality change), 7 highest ranked barriers (B1 to
B7), dummy for companies with gas as sole heat source and dummy for future projects.
Based on covariances.

the voluntary agreement are opposite and constitute the dominant

dimension of the 2nd principal component. The voluntary agreement was perceived as more effective for projects and companies
where information barriers were high: lack of management time;
protability wasnt studied before; lack of reliable information and
technical feasibility wasnt studied before (with positive loadings
on the second principal component). The ETS was perceived as
more effective for projects and companies where the following
barriers were important: other investments received prior
nancing; low demand risk (with negative loadings on second
principal component).
The voluntary agreement was perceived to be important for both
companies using gas exclusively and others. The effect of the type of
fuel on the importance of the voluntary agreement is zero, conditional on the type of observation (completed project, future project,
company-wide observation), as shown in Table 2. By contrast, the EU
ETS was perceived to be much more important for those companies
with more carbon intensive energy sources which, in addition, have
much higher process emissions (mean score is 1.1 higher, Table 2).
This is also observable from the positive loadings of the gas only
factor on the second principal component.
The voluntary agreement was a lower motivation for potential
future projects than for completed projects (mean score is 0.6
lower, signicant). On the other hand, the ETS was perceived as
more important for future projects compared to completed projects
(mean score is 0.6 higher, signicant). Many managers estimate
that the future potential of the voluntary agreement is limited
because there is no price mechanism and many low hanging fruits
are captured by the rst voluntary agreement already. This is again
conrmed by the negative loading of the factor future project on
the second principal component.
5. Conclusions
This study questioned 13 motivations and 16 barriers to energy
efciency in energy intensive companies. The most important

141

barriers are perceived to be related to capital availability (barriers

ranked rst and fourth). Companies postpone cost-effective measures not because energy related investments are seen as less
important (energy counts for 30% of production costs), but because
strict internal capital budgeting rules entail high hurdle rates for
optional projects. Yet many energy related investments are
optional. Several theories of corporate nance describe strict capital budgeting rules as a way of dealing with asymmetric information between managers and shareholders, or between managers
and creditors, but they are costly to the extent that they impede
genuine protability-increasing investments.
The problem of a higher than cost-effective hurdle rate may be
exacerbated by the fact that project selection is typically based on
payback times, where the contradiction between the implied IRR
and the WACC is not apparent and not acknowledged by the
managers interviewed.
Barriers related to information and knowledge are the second
most important explanation for postponed investments. Energy
efciency investments are in many cases related to the diffusion of
well-established but nevertheless modern technologies and thus
require a thorough technical feasibility and protability analysis,
which in many cases impedes or delays the investments. The fact
that these barriers related to the absence of a feasibility or protability analysis are more important than the barrier Lack of reliable
information indicates that organisational barriers are at play, such
as lack of management time.
The voluntary agreement and the ETS were found to be complementary. The voluntary agreement was perceived as more
important 1) in gas-fuelled companies, 2) for companies and/or
projects where information barriers are important and 3) more
important for past projects, compared to future projects. The ETS
had very little impact on efciency investments in companies with
gas as the only heat source. None of these companies included the
carbon price in payback time calculations. However, the ETS was
perceived as more important for companies 1) with mixed fuels 2)
where capital budgeting was an important barrier and 3) for future
projects.
Acknowledgements
Remarks from the Interdisciplinary Symposium on Sustainable
Development (Namur) and the Belgian Environmental Economics
Day (Leuven) are gratefully acknowledged. I also thank Kevin
Marchal and Walter Hecq for their valuable feedback.
References
Abadie, L.M., Chamorro, J.M., Gonzlez-Eguino, M., 2013. Valuing uncertain cash
ows from investments that enhance energy efciency. J. Environ. Manage 116,
113e124.
Anderson, B., Leib, J., Martin, R., McGuigan, M., Muls, M., de Preux, L., Wagner, U.,
2011. Climate Change Policy and Business in Europe. Evidence from Interviewing Managers. Centre for Economic Performance, London School of Economics, London, UK.
Antal, M., van den Bergh, J., 2013. Re-spending rebound: a macro-level assessment
for OECD countries and emerging economies. Energy Pol. (in press).
Akerlof, G., 1970. The market for lemons. Q. J. Econ. 84 (3), 488e500.
Anderson, S.T., Newell, R.G., 2004. Information programs for technology adoption,
the case of energy-efciency audits. Resour. Energy Econ. 26, 27e50.
Ansar, J., Sparks, R., 2009. The experience curve, option value, and the energy
paradox. Energy Policy 37, 1012e1020.
Bewley, T.F., 2002. Interviews as a valid empirical tool in economics. J. Socio Econ. 31
(4), 343e353.
Brealey, R.A., Myers, S.C., Allen, F., 2010. Principles of Corporate Finance. McGrawHill, Columbus, USA.
BREF, 2007. Reference Document on Best Available Techniques in the Ceramic
Manufacturing Industry. European Commission, Brussels, Belgium.
BREF, 2010. Reference Document on Best Available Techniques in the Cement, Lime
and Magnesium Oxide Industries. European Commission, Brussels, Belgium.

142

F. Venmans / Journal of Cleaner Production 69 (2014) 133e142

de Groot, H.L.F., Verhoef, E.T., Nijkamp, P., 2001. Energy savings by rms, decisionmaking, barriers and policies. Energy Econ. 23, 717e740.
DeCanio, S.J., 1998. The efciency paradox, bureaucratic and organizational barriers
to protable energy-saving investments. Energy Policy 26 (5), 441e454.
DeCanio, S.J., 1993. Barriers within rms to energy efcient investments. Energy
Policy 21 (9), 906e914.
Fleiter, T., Worrell, E., Eichhammer, W., 2011. Barriers to energy efciency in industrial bottom-up energy demand models e a review. Renew. Sustain Energy
Rev. 15, 3099e3111.
Greene, D.L., 2011. Uncertainty, loss aversion and markets for energy efciency.
Energy Econ. 33, 608e616.
Henriksson, E., Sderholm, P., 2009. The cost-effectiveness of voluntary energy efciency programs. Energy Sustain Dev. 13, 235e243.
IEA, 2012. World Energy Outlook. International Energy Agency, Paris, France.
Jaffe, A.B., Newell, R.G., Stavins, R.N., 2004. Economics of Energy efciency. In:
Cleveland (Ed.), Encyclopedia of Energy, vol. 2. Elsevier Science, Oxford, UK,
pp. 79e90.
Jensen, M.C., Meckling, W.H., 1976. Theory of the rm, managerial behavior, agency
costs and ownership structure. J. Financ. Econ. 3, 306e360.
Kahneman, D., Tversky, A. (Eds.), 2000. Choices, Values and Frames. Cambridge
University Press, New York, USA.
Madlool, N.A., Saidur, R., Hossain, M.S., Rahim, N.A., 2011. A critical review on energy use and savings in the cement industries. Renew. Sustain Energy Rev. 15,
2042e2060.
Myers, S.C., Majluf, N.S., 1984. Corporate nancing and investment decisions when
rms have information that investors do not have. J. Financ. Econ. 13, 187e221.
Moya, J.A., Pardo, N., Mercier, A., 2011. The potential for improvements in energy
efciency and CO2 emissions in the EU27 cement industry and the relationship
with the capital budgeting decision criteria. J. Clean. Prod. 19, 1207e1215.
Pardo, N., Moya, J.A., Mercier, A., 2011. Prospective on the energy efciency and CO2
emissions in the EU cement industry. Energy 36, 3244e3254.

Sardianou, E., 2008. Barriers to industrial energy efciency investments in Greece.

J. Clean. Prod. 16, 1416e1423.
Schleich, J., 2009. Barriers to energy efciency, A comparison across the German
commercial and services sector. Ecol. Econ. 68, 2150e2159.
Schleich, J., Gruber, E., 2006. Beyond case studies, Barriers to energy efciency in
commerce and the service sector. Energy Econ. 30, 449e464.
Sola, A.V.H., Xavier, A.A.P., 2007. Organizational human factors as barriers to energy
efciency in electrical motors systems in industry. Energy Policy 35, 5784e
5794.
Sorrell, S., OMalley, E., Schleich, J., Scott, S., 2004. The Economics of Energy Efciency. Edward Elgar Publishing, Cheltenham, UK.
Stavins, R., Jaffe, J., Schatzki, T., 2007. Too Good to Be True? An Examination of the
Economic Assessments of California Climate Change Policy. NBER working paper 13587.
Sutherland, R.J., 1996. The economics of energy conservation policy. Energy Policy
24, 361e370.
Van Soest, D., Bulte, E., 2001. Does the energy-efciency paradox exist? Technological progress and uncertainty. Environ. Resour. Econ. 18, 101e112.
Weber, L., 1997. Some reections on barriers to the efcient use of energy. Energy
Policy 25, 833e835.
Williamson, O.E., 1988. Corporate nance and corporate governance. J. Finance 43,
567e590.
Williamson, O.E., 2002. The theory of the rm as governance structure, from choice
to contract. J. Econ. Perspect. 16 (3), 171e195.
Zakharov, A.I., Begak, M.V., Guseva, T.V., Vartanyan, M.A., 2009. Prospects for
increasing the energy and ecological efciency of the production of ceramic
articles. Glass Ceram. 66, 356e362.
Zilahy, G., 2004. Organisational factors determining the implementation of cleaner
production measures in the corporate sector. J. Clean. Prod. 12, 311e319.