Journal of Cleaner Production 130 (2016) 25e34

Contents lists available at ScienceDirect

Journal of Cleaner Production

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

Effects of customized consumption feedback on energy efcient

behaviour in low-income households
Ales Podgornik a, *, Boris Sucic a, Bostjan Blazic b
a
Jozef Stefan Institute, Energy Efciency Centre, Jamova 39, 1000, Ljubljana, Slovenia
b
Faculty of Electrical Engineering, University of Ljubljana, Trzaska 25, 1000, Ljubljana, Slovenia

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

Article history: Smart metering and various consumption-feedback systems can be used as applicable technology to
Received 19 March 2015 encourage end-use energy efciency in the residential sector. Many studies have demonstrated the
Received in revised form positive inuence of socially contextualized feedback on the energy consumption of households, but not
29 January 2016
many such studies dealt with the specics of low-income households. This paper evaluates the effect of
Accepted 2 February 2016
Available online 6 February 2016
customized consumption feedback and other information interactions on energy-behaviour patterns and
energy savings in low-income households. The experimentation process was accompanied by an inter-
active awareness campaign, with the emphasis being on a proper understanding of the consumption
Keywords:
Consumption feedback
feedback and other complementary energy services provided to low-income households. The feedback
Energy behaviour actions were customized to tackle the recognised needs of each specic target household, considering
Low-income households both the social aspects and the typology of the dwelling. The results clearly conrmed the importance of
Smart meters customized information and efciency indicators for specic household groups, with a potential to in-
Energy awareness crease knowledge and develop awareness with respect to established habits and their relevance to
energy behaviour.
2016 Elsevier Ltd. All rights reserved.

1. Introduction Vorsatz et al. (2009), different non-technological determinants

Energy efciency has been recognised as an essential element of
the European Union's (EU) energy and climate change mitigation
policies. The EU has identied buildings as being the most prom-
ising target for improving energy efciency and has quantied a
signicant energy-savings potential associated with infrastructure
and equipment investments (27% in the residential sector and 30%
in the services sector) (EC, 2006). As households and buildings are
responsible for over 40% of energy usage, there is a need to develop
new approaches that examine the energy use in homes and to
encourage energy efcient behaviour, in addition to using energy
efcient devices in households (Chen et al., 2013). Also, the soft
organisational potentials for energy efciency improvements are
often neglected in policies both at the national and at the local level
(Bukarica and Robi
c, 2013). Improvement of energy efciency re-
quires informed consumers and awareness among all segments of
society as well as tailored information, education and training for
selected stakeholders (Gynther et al., 2012). According to rge-
* Corresponding author. Tel.: 386 1 5885 300.
E-mail address: ales.podgornik@ijs.si (A. Podgornik).
lead to different levels of energy consumption even within the
same technological setting. Jonsson et al. (2011) have concluded
that non-technological drivers can be as important, or even more
important, than the technologies used in determining the energy
consumption in buildings. Research conducted by Ouyang and
Hokao (2009) has shown that about 14% of the energy savings in
households could be achieved simply by improving the residents'
energy behaviour. According to Wood and Newborough (2003), it is
worth to invest additional efforts in optimising the design and
assessing the use of energy-consumption indicators in households,
in order to maximise the associated energy-saving potential. Darby
(2006) has identied several behavioural-change programmes that
have indicated that energy savings of up to 20% could be achieved
by applying different information strategies or by using improved
feedback. However, even though the behaviour of residential end-
users is one of major determinants with respect to energy use in
buildings, the energy-savings potential based on behaviour re-
mains neglected (Levine et al., 2007). To effectively plan and
execute energy-efciency programs, a sound understanding of the
determinants factors that drive household energy consumption
(such as the type, number and use of appliances, oor area, average

http://dx.doi.org/10.1016/j.jclepro.2016.02.009
0959-6526/ 2016 Elsevier Ltd. All rights reserved.
26 A. Podgornik et al. / Journal of Cleaner Production 130 (2016) 25e34
outside temperature, and number of occupants) is needed (Firth
et al., 2008). A more individualized social understanding and the
rationalities behind the households' energy consumption are
needed in order to specify the different consumption patterns. It is
important to be aware that a necessary precondition for informa-
tion about the relationship between the end-user's behaviour and
the corresponding energy use is the specication of the technical
features of the appliances for a given household, in combination
with individual monitoring (Vassileva et al., 2012). Furthermore, to
take full advantage of energy-efciency programs, they have to be
adapted to the various segments of residential customers. Low-
income households (LIHs) are one of those segments that have to
be specially targeted, due to their social constraints, and are
considered as far to reach through traditional public policies (Elih-
MED, 2010). Understanding the ne-grained differences within a
targeted income group can direct utility and municipal communi-
cation efforts with greater effect (Mazur-Stommen et al., 2013). This
understanding has to be used when adapting local energy-
efciency policies to encourage energy-efciency improvements
and striving for a more sustainable society. A sustainable society
can be assumed to be a society which consists of a sustainable
production system, a sustainable community, a sustainable
ecosystem, and a sustainable government (Shin et al., 2008). With
this in mind, utilities can offer individualized consumption feed-
back to encourage people in LIHs to become more aware of their
energy consumption and stimulate them to change their energy-
related behaviour. Utility companies and municipal authorities
should include an individual behavioural-assessment approach to
the energy consumption of households when developing energy-
savings strategies and measures (Vassileva et al., 2012). Local
climate conditions also represent a signicant factor for residential
energy consumption. According to Vassileva et al. (2012), a
household's income is one of the most important determinants
with respect to consumer behaviour for northern European climate
conditions, and it has been conrmed that a higher income results
in greater electricity consumption. On the other hand, Kavousian
et al. (2013) have recently shown that the residential electricity
consumption in United States is primarily determined through the
way households use electricity, and less by the way that they value
energy efciency and that there is no signicant correlation be-
tween electricity consumption and income level. Research con-
ducted by Beunder and Groot (2015) on a large sample of Dutch
households, have conrmed that household's cultural background
signicantly categorized the level of electricity consumption. The
general conclusion is that when analysing energy consumption,
particular attention should be paid to the individual characteristics
of the inhabitants, i.e., households should be treated individually.
The consumption feedback and recommendations to reduce
household energy consumption should be represented by different
determinants that are important for the household, while leaving
behind the unimportant, because those can have an adverse effect
(Desmedt et al., 2009).
This paper evaluates the impact of customized and adaptive
consumption feedback on energy-behaviour patterns and energy
savings in LIHs in the Mediterranean region. In this context, LIHs
are recognised as households that are most vulnerable to fuel
poverty. The experiments and data presented in this paper are re-
sults of the EU-funded project ELIH-Med (Energy efciency in low-
income housing in Mediterranean). ELIH-Med project concluded in
2014 and has been identied as a strategic project within the MED
Programme with the objective to identify innovative energy ef-
ciency measures and nancial instruments for LIHs in Mediterra-
nean. Implemented feedback actions were customized to tackle the
recognised needs of each specic target household, considering
both social aspects and dwelling typology. The described research
work was inspired by the recommendations proposed by Caeiro
et al. (2012), who suggested that future research should focus on
the need to communicate effectively, to engage stakeholders and to
address the complexity involved in the measurement and assess-
ment of sustainable household energy consumption. The general
objective was to check whether additional information about en-
ergy use provided to LIHs can stimulate them to reach measurable
energy savings without any additional investments in new equip-
ment. Selected households provided a real testing environment,
with the full support of the occupants, and open access to all
requested data for the validation of the customized feedback and
for the promotion of electricity-consumption awareness in LIHs.
According to Eisenhardt and Graebner (2007), this represents an
opportunity with unusual levels of research access. It has to be
aware the open access experimentation typically engages a much
smaller sample than would normally be considered adequate for
inductive hypothesis testing. However, as it was shown by Langevin
et al. (2013), open access provides a rich foundation on which to
base a follow up efforts, and identify key pieces of contextual in-
formation that would otherwise be missed by a purely quantitative
method for behavioural data collection.
2. Energy efcient behaviour
When referring to energy-related behaviours, there are always
two implicit dimensions, i.e., the behaviour in itself and the asso-
ciated energy consumption, in which the second is a consequence
of the rst, and quanties it (Lopes et al., 2012). While the adoption
of new energy efcient technologies relates to investments, the
regular use of appliances is related to habitual behaviour. This
habitual behaviour is automatic and routine, and is the way in
which individuals repeat and do things automatically without
consciously weighing up the pros and cons, usually delivering
suboptimal results in terms of energy efciency (Fischer, 2008). In
this context, the household can be seen as a system where system
efciencies rest on a combination of technology and behavioural
intelligence that reacts dynamically to its surroundings (Elliott
et al., 2012).
2.1. Impact of consumption feedback on behaviour
Strategies for promoting more efcient energy behaviour
have been categorised as antecedent or consequence strategies
(Abrahamse et al., 2005). Antecedent interventions are designed
to change factors that precede behaviour and usually include a
commitment and information to increase awareness. According
to Ueno et al. (2006), the installation of an on-line Energy-
Consumption Information System has an inuence on the
energy-saving awareness of consumers. Consequence in-
terventions focus on changing the consequences following
behaviour and usually include energy-consumption feedback
and different rate systems (in the form of nancial rewards).
The role of consumption feedback is to provide information
about energy consumption and to make it visible. According to
Fischer (2008), successful feedback has to capture the con-
sumers' attention, draw a close link between specic actions
and their effects, and activate various motives that may appeal
to different consumer groups (such as cost savings, resource
conservation, emissions reduction, competition, and others).
Consumption feedback has to activate a process called norm
activation for new norms and considerations to enter the
conscious decision-making process (Matthies, 2005). This
means that inefcient habits must be broken and a person must
realize that there are various options to choose from, and that
norms and criteria are needed for evaluating these options
 A. Podgornik et al. / Journal of Cleaner Production 130 (2016) 25e34
(Fischer, 2008). Through a norm activation process, end-users
become aware of the relevance of their behaviour and its rela-
tion to energy consumption, the possibilities to sustainably in-
uence personal behaviour and the environmental issues
associated with energy consumption (Fig. 1). According to
Rodrguez-Barreiro et al. (2013), environmental behaviour is
considered to be a set of effective, deliberate, and anticipated
actions aimed at accepting responsibility for the conservation
and preservation of nature and natural resources. Research
conducted by Lopes et al. (2012) has shown that there is no
single and perfect form of feedback for everybody and that
feedback should be tailored according to the specics of each
customer group and the various regional differences.
Tailored and continuous consumption feedback is also
important to minimize the danger of the fall-back effect. The fall-
back effect is dened as the phenomenon in which the newness
of a change causes people to react, but then that reaction di-
minishes as the newness wears off (Hayes and Cone, 1977). There
is also a danger of the rebound effect, when energy-efciency
measures reduce the operational costs of dwellings. The
rebound effect is a postulate that people increase their use of
products and facilities as a result of this reduction in operating
costs, thereby reducing the energy savings achieved by imple-
mented energy-efciency measures (Nadel, 2012). In spite of
considerable data restraints and research gaps, there are some
indications that the most successful feedback for households
combines the following features: appliance-specic consumption
breakdown, historical comparison, presentation in a clear and
appealing way, and the use of computerized and interactive tools
(Fischer, 2008).
2.2. Breaking the habits
The premise behind different time-of-use (TOU) tariff systems is
that consumers will behave rationally, in the sense that they
maximise a utility given incentives and that their preferences are
well organised, recognised, stable and consistent. Faruqui et al.
(2010) have shown that the real-time monitoring, enabled by in-
home displays (IHD) can reduce electricity consumption by 7.6%
when combined with TOU tariffs, whereas 6.7% savings were ach-
ieved when only IHDs were used and 3.3% savings were achieved
when only TOU pricing was used. However, there are evidences
that consumers do not always make consistent and rational de-
cisions, even if they are given perfect information (Lopes et al.,
2012). In this context, behavioural failures are considered to
represent consumer behaviour that is inconsistent with utility
maximization or with energy-service cost minimization
(Gillingham et al., 2009). According to Wilson and Dowlatabadi
Fig. 1. Norm activation.
27
(2007), inconsistencies in energy related behaviour in households
can be explained by:
- Time inconsistency: End-users do not make decisions in a time-
consistent manner using constant discount rates.
- Bounded rationality and heuristic decision making: End-users
are rational but face cognitive constraints in processing in-
formation; therefore, they tend to use a wide range of heu-
ristics (simple decision rules) to reduce the cognitive
requirements.
- Framing and reference dependence: End-users' decisions are
inuenced by the way information is presented to them e
different results may be obtained simply by framing the deci-
sion as a choice between losses or gains.
One of the reasons for such inconsistencies and irrational
choices is that human behaviour is often guided by habits, i.e.,
not fully conscious forms of behaviour. This may be one of the
reasons why energy consumption continues to increase despite
the rising environmental awareness among the end-users
(Mare
chal, 2010). Habits do mediate the intentionebehaviour
relationship in the eld of residential energy consumption and
this tends to conrm that the presence of strong habits can
explain the low effectiveness of traditional measures, such as
incentives (Mare
chal, 2009). Habits should therefore be fully
acknowledged and accounted for prior to designing energy-
efciency measures in the residential sector. Also, there is evi-
dence that households will return to their old consumption
habits if they cannot see the impact of their behavioural changes
in their electricity bill or if the savings are not considered to be
worth the bother (Annala et al., 2014).
As habits are context dependent, the change of the context and
circumstances in which habits were developed can offer a way for
changing them (Wood et al., 2005). Also, successful interventions
need to target not only change of old, unwanted behaviours but also
the maintenance of new, more desirable responses (Verplanken
and Wood, 2006). However, changes of context do not make
habits change, either automatically or directly, but they are better
viewed as windows of opportunity (Mare
chal, 2010). This win-
dow could be opened by implementing different energy-efciency
measures (i.e., retrotting a building) and joined by individualized
consumption-feedback and awareness campaigns to break the
consumers' habits.
3. Methods
In the addressed Mediterranean regions the need for heating is
relatively low, at least compared to the northern countries, and only
the electricity-consumption feedback was targeted in the experi-
ment due to there being no gas and district-heating infrastructure
in the pilot dwellings. Three types of electricity-consumption
feedback presentation were used in the selected dwellings to
target the desired actions of the consumer:
- Presentation as a consumption problem. The electricity con-
sumption was presented relative to the number of people living
in the household, the appliances used (appliance-specic
monitoring was required) and the base load of the household
was calculated. The base load of the household was dened as
the consumption state when only always-on appliances are
operating (i.e., refrigerator, freezer, PC in stand-by, etc.).
- Presentation as an environment problem. The annual household
CO2 emissions were calculated from the electricity consumption
using country specic CO2 emission factors. The CO2 indicators
were then presented relative to how many trees are needed to
28 A. Podgornik et al. / Journal of Cleaner Production 130 (2016) 25e34
absorb produced CO2 per each household. It was assumed that
the average yearly CO2 absorption is around 10 kg CO2 per tree.
- Presentation as a family-budget problem. In this type of pre-
sentation, electricity consumption was indicated in direct costs
relative to the number of people living in the household and
the appliances used. It was assumed that these indicators could
be particularly important for LIHs, as even a small indication of
the energy-costs savings can have a signicant impact on the
household budget.
Also, two different use cases for the consumption-feedback
services were used in the experiment:
1. Appliance-specic consumption breakdown with a simple kWh,
CO2 and costs presentation of the electricity consumption on
IHDs in 100 selected households. Web access with additional
consumption breakdown was also made available. Individual-
ized periodic reports (efciency indicators) were distributed to
the users.
2. Utility smart meters were used as the main focus point of the
consumption monitoring in 25 pilot households. Informative
billing was used as the main consumptionefeedback medium
and individualized periodic reports (efciency indicators) were
distributed to the users. These pilots are further involved in
complementary energy services utilizing Mediterranean-
specic climate conditions (installation of photovoltaic system
based on Net metering scheme).
The selection of the pilot households that were participating
in the consumption-feedback experimentation was based on the
low-income criteria (household members do not have sufcient
resources or access to credit in order to invest in energy ef-
ciency), the age of the building (20 years or older) and the
location (Mediterranean area). The households were selected
with tenders and participated in an experimentation process on a
voluntary basis. The 100 pilot dwellings selected for rst use case
were located in Malaga (44 in Andalusia, Spain), Valencia (16 in
Valencia, Spain), Mas Thibert (25 in d'Arles, France) and Malta
(15 dwellings). Unfortunately, in 14 households (12 in Malaga
and 2 in Mas Thibert) key elements (ownership, number and
education of occupants) changed during the experimentation
phase and that was a reason why they were eliminated from the
experiment. The 25 pilot dwellings for second use case were
located in Cyprus. It is important to emphasise that selected
buildings and households previously have never participated in
any other energy efciency campaign and no investments in
efcient household equipment have been made during the
observed baseline period.
An important milestone in the presented research was the
design of the consumption-feedback experimentation plan. Ac-
cording to this plan, three distinct phases of interactions with
selected LIH pilot dwellings were recognised:
1. The pre-installation.
2. The installation of the consumption-monitoring equipment and
the feedback medium.
3. The monitoring and living-lab activities.
The interaction phases were designed as an interactive pro-
gramme to inuence the energy awareness and related behaviour at
all stages of the experimentation (Fig. 2). This approach incorporates
the main objectives of the Household Sustainable Consumption (HSC)
assessment metrics and indicator system (Caeiro et al., 2012), with the
emphasis on providing easily comprehensible and meaningful infor-
mation to the users, using an integrative approach among the
different domains, providing guidance via illustrative best practices
and providing nancial and non-nancial incentives.
3.1. Pre-installation
An awareness campaign has been launched (posters, brochures,
yers, information points, informative meetings, web platform) in
the pre-installation phase to motivate the households participating
in the experiment as part of the norm-activation process. In case of
multi-apartment buildings information campaigns were designed
to target all and not just selected pilot households. Also, designated
local representatives were selected to coordinate the building's
activities (organization of education and awareness sessions, co-
ordination of the installation process between the installers and the
users, etc.). The general idea behind this approach was to provoke a
positive it can be done attitude among all building occupants and
it was estimated that properly trained local representatives can
have a positive role in the motivation process. The awareness
campaign was followed by a survey in selected pilot households.
This question-based survey was used to determine the economic
and social status of the users, the dwelling typology and other
technical characteristics of the household (i.e., the appliances used
in the household). The survey was also used to establish an initial
understanding regarding the users' habits with respect to appliance
usage.
In each participating household a walk-through energy audit
(half a day) was conducted. Also, this phase was used to establish
the baseline electricity consumption of the selected households.
The data was obtained (with the users' consent) from electricity
utilities or building managers (in the case of social housing) for the
previous three years (from 2010 to 2012) and was normalised to
average climatic conditions in respective regions. Vital output from
this phase was an identication of individual and group level bar-
riers and constraints on energy efcient actions at the household
level and preparation of plan how to cope and overcome them.
3.2. Installation of the consumption-monitoring equipment and the
feedback medium
The equipment-installation phase was used to install the
monitoring equipment (smart meters and IHDs) and educate the
participating households about equipment usage. The end-users
became familiar with the installed smart-metering equipment
and understood how they can use the feedback information to
manage their electricity consumption effectively. Also, during this
phase the main ndings from the walk-through energy audit were
presented to the household occupants. In all addressed dwellings
special attention was given to the appliance-specic consumption
and end-users were able to decide which appliances would be
monitored in their dwellings. Their decisions were based on the
information provided during the awareness campaign in the pre-
installation phase. The most common appliances selected by the
users were air-conditioning units, electric heaters (heating and
domestic hot-water preparation), refrigerators, microwave ovens,
personal computers and televisions.
Commercially available IHDs were used to provide individual-
ized, appliance level and real-time feedback in the targeted
dwellings where the electrical infrastructure was not optimal for
sophisticated and centralized approaches to monitoring. Installed
system included a numerical IHD that was proven to be the best at
providing information about energy-use changes compared to the
analogue and ambient designs (Chiang et al., 2012). Commercially
available utility smart meters and data loggers were used in the
second use case.
 A. Podgornik et al. / Journal of Cleaner Production 130 (2016) 25e34 29

Fig. 2. Interaction phases.

3.3. Monitoring and living-lab activities
This phase was used to monitor the impact of the implemented
consumption-feedback services on the LIHs' behaviour. The mea-
sures were adjusted according to the responses from the house-
holds, facilitating the living-lab approach to ne-tune the different
smart-metering services to the different households (i.e., the
impact of social, cultural and regional differences). The results of
the pre-installation phase survey were used to prepare and issue
individual (building-specic) bulletins (Fig. 3). Several energy-
efciency indicators were presented in these bulletins (i.e., elec-
tricity consumption in observed period and by appliance, bench-
mark with neighbours) and individualized energy-efciency and
energy-costs saving tips were suggested (i.e., how to use timers/
automation for household-specic appliances, presentation of
different electricity suppliers, etc.). Based on the initial response to
the individualized bulletins, an additional adjustment of the con-
sumption feedback was made during this phase (the adjustment of
the presented consumption indicators and the rearrangement of
the monitored household appliances). The monitoring was per-
formed from the beginning of 2013 until the end of 2014.
3.4. Window of opportunity
In parallel with the consumption feedback experimentation
selected pilot dwellings on Cyprus were invited to participate in
complementary measures like installation of photovoltaic system
based on Net metering scheme. These additional activities were
used in the consumption-feedback experiment as a window of
opportunity to change the household context, i.e., to break existing

Fig. 3. Information bulletin examples.

30 A. Podgornik et al. / Journal of Cleaner Production 130 (2016) 25e34
Fig. 4. Window of opportunity.
habits (Fig. 4). The awareness campaign also included information
on available nancing programs for some other energy-efciency
measures which can be implemented by LIH (i.e., nancial in-
centives for LIH to purchase new and efcient refrigerators).
4. Results
4.1. Households survey
Information about the dwellings' typology, living conditions,
residents' characteristics, and habits were collected through a
questionnaire-based survey. The questionnaire covered the
following topics:
- Economic status of the household members.
- Social characteristics of the residents and the typology of the
household (household members, age, education and employ-
ment, type of property, number of rooms, size of dwelling, type
of ownership, year of construction, internet access) including
energy-related habits (cooking and temperature comfort).
- Technical properties of the dwelling (type of heating/cooling
and domestic hot water, audit of household appliances and their
usage, type of lightning).
One of the purposes for the careful household categorization
was to enable long-term comparisons with an average, normalized
or benchmarked consumer in the same user category. The need for
this comparison is specied in the Energy Efciency Directive
(Directive, 2012/27/EU).
4.1.1. Income type
The income diversity of the participating LIHs is presented in
Fig. 5. All pilot households were classied as low-income, even in
the cases of full employment.
4.1.2. Household structure
The structure of the participating LIHs is presented in Fig. 6. The
category Family with children indicates underage children in the
Fig. 5. Household income.
Fig. 6. Household structure.
household (including single-parent families), whereas Family
indicates more than two adults living in the same household. The
category Couple indicates two adults living together.
4.1.3. Dwelling characteristic
The dwelling typology and the ownership of participating LIHs
are presented in Fig. 7. This gure clearly shows the high percent-
age of ownership in the Mediterranean area even in the low-
income category. However, due to the low income of the tenants,
these dwellings are frequently neglected and provide poor living
conditions.
The use of air conditioning and heating sources is presented in
Fig. 8. The climatic conditions of the Mediterranean are the reason
for the higher use of air-conditioning units, which are present in
60% of the participating LIHs.
4.2. Households electricity consumption
Table 1 shows the household electricity consumption for the
LIHs participating in the experiment and the national averages
(Spain, Cyprus, France and Malta) obtained from the Odyssee-Mure
project database (Lapillonne et al., 2013).
To better understand the consumption habits of the LIHs and
electricity-savings potential, electricity consumption per house-
hold member was calculated. Fig. 9 shows the distribution of
electricity consumption per household member relative to the size
of the household.
4.3. Savings achieved
In the rst use case awareness-raising activities about energy
efciency were performed in parallel with the consumption-
feedback experimentation. In case of Malaga (Spain), where the
pilot households were located in multi-apartment buildings, in-
formation campaigns were designed to target all and not just
Fig. 7. Dwelling typology and ownership.
 A. Podgornik et al. / Journal of Cleaner Production 130 (2016) 25e34
Fig. 8. Use of air conditioning and heating sources.
selected pilot households. As a result of the awareness campaigns
which started in 2013, there was a 12.1% decrease in the overall
electricity consumption for the whole building (140 dwellings) in
2013 and 13.3% in 2014 (Fig. 10). Selected pilot dwellings in this
building were equipped with IHDs in 2013, and these were
receiving an individualized bi-monthly efciency bulletin. These
dwellings achieved a 13.6% decrease in electricity consumption in
2013 and a total decrease of 22.4% was achieved in 2014 (Fig. 11).
Local representatives acknowledged that the occupants highly
appreciated comparative feedback enriched with efciency in-
dicators and graphic representation of billing information
compared to consumption in comparable past periods, to average
consumption in the past year, and correlated to the inside and
outside temperature difference. It is clear that customised con-
sumption feedback provided occupants with new knowledge and
empowered them to reach additional savings.
Impact of the age structure has been evaluated in a building
with predominantly retired residents in Valencia (Spain). Similar to
the case above, all the dwellings (56 dwellings in total) in this
particular building were targeted by awareness activities and 16
pilot dwellings with retired residents were equipped with IHDs and
were receiving an individualized bi-monthly efciency bulletin.
These pilots exposed the inefciency of the IHDs in this social
segment (elderly people with low incomes). As a result of the
awareness campaigns, there was a 10.2% decrease in the electricity
consumption in 2014 on average for the whole building, but no
additional savings were observed for 16 pilot households with
IHDs.
In the second use case the inuence of the informative billing
was evaluated in 25 LIHs in Cyprus where 11 of these dwellings
were additionally equipped with a photovoltaic system at the end
of 2013. These 11 pilots were involved in a utility Net metering
scheme. This scheme allows residential customers who generate
their own electricity from solar power to feed the electricity that
they do not use back into the grid and the meter offsets the
production-consumption of electricity. As a result, these customers
pay only the net amount for both transactions. Awareness-raising
activities on energy efciency were performed in parallel with
the consumption-feedback experimentation. The total consump-
tion for the household involved in the Net metering scheme was
Table 1
Average electricity consumption per dwelling.
[kWh/year] Spain (n 48)
LIH pilots 3,078
Odyssee-Mure average 3,800
Difference [%] 19.0%
31
Fig. 9. LIH electricity consumption per person.
Fig. 10. Yearly electricity consumption e reference with the awareness campaign.
used in the calculations (the photovoltaic production was not
subtracted from the household's electricity usage). During the
awareness campaigns, the local experts trained end-users to
properly understand their load diagrams and connect usage of
different appliances' with their behaviour. At the beginning of this
process it was noticed that all pilot households were completely
unaware of opportunities for energy savings in their homes. This
unawareness of opportunities was the main challenge for the local
experts who were supporting selected pilot households during the
experimentation process. However, it is important to mention that
all pilot households provided strong support and extra good will
during the entire experimentation process. The pilot households
provided only with informative billing (14 households in total)
achieved a 24.2% decrease in electricity consumption in 2013. A
decrease of 27.4% was achieved in 2014 (Fig. 12).
Local experts who were working with pilot households involved
in the Net metering scheme (11 households in total) acknowledged
that the households' occupants after installation of photovoltaic
system were more active in trying to make sense of consumption
data and provided information in order to develop their own un-
derstanding of inuencing parameters that have resulted in an
observed data pattern or in an unexpected operational event. It was
recognised that proper training and motivation were necessary to
Cyprus (n 25) France (n 23) Malta (n 15)
4,928 3,972 2,788
5,500 5,650 4,850
10.4% 29.7 42.5%
32 A. Podgornik et al. / Journal of Cleaner Production 130 (2016) 25e34
Fig. 11. Yearly electricity consumption e pilots with IHDs.
trigger the change in people attitude towards energy use in order to
cut energy costs through more efcient and responsible operation
of household appliances, especially during summer months when
the majority of electricity is consumed for cooling. At the end of the
experimentation process, it was recorded that the pilot households
involved in the Net metering scheme achieved a greater con-
sumption decrease for both years, almost 30% in 2013 and 36.4% in
2014 (Fig. 13). However, it has to be aware that before drawing any
other conclusion the approach presented and additional analysis
should be carried out for a large group of similar buildings.
5. Discussion
The LIHs participating in the consumption-feedback experi-
mentation have shown a great interest in energy-efciency im-
provements. All proposed actions were explained in the pre-
installation phase and the users' reactions were positive. Equally
important was the community involvement. Including represen-
tatives of the local community in the preparations and awareness
campaigns has proved to be a crucial point in the transfer of sus-
tainable ideas. Initial scepticism and fear of government control,
which was particularly present in LIHs, has turned into enthusiasm
and a desire to cooperate. Building a foundation of trust is needed
in this social segment and awareness campaigns should be used
when tackling distrust issues.
The three main recognizable benets for the participating LIHs
have been experienced during the living-lab activities:
- Users have become aware of the consumption of individual
appliances and expressed surprise about their impact on the
monthly electricity bill (i.e., users started to turn down the
Fig. 12. Yearly electricity consumption e pilots with informative billing.
Fig. 13. Yearly electricity consumption e pilots with informative billing and Net
metering.
heating/cooling when excess consumption was indicated on the
IHD).
- Awareness of current consumption and the households' envi-
ronmental footprint has instigated behavioural changes and a
clear interest in additional energy-efciency measures to be
implemented in the households.
- Multiplying the effect of complementary energy-efciency
measures. When an additional energy-efciency measure is
implemented (i.e., the installation of a photovoltaic system), the
consumption feedback additionally motivates the users to adapt
their behaviour to the implemented technical updates and
modications.
Although the interest in using an IHD with real-time con-
sumption and appliance-specic consumption has reduced with
time, the appliance-specic metering was well received when
presented in the periodic efciency bulletin with the appliance-
specic consumption breakdown and other efciency tips. The
residents of the pilot households rated the periodic bulletin on
energy efciency and electricity consumption as the best
consumption-feedback activity of the consumption-feedback
experimentation. The breakdown of the electricity-consumption
costs per appliance was rated as the most valuable feedback in-
formation by the users, whereas CO2-related indicators were
mostly disregarded. Therefore, the use of simple efciency in-
dicators is recommended in LIHs, concentrating on the costs of
energy consumption. Also, early achievement of certain visible re-
sults was proved to be very important and boosted the condence
of household occupants.
Cultural background had a major impact on living habits.
Traditional habits are hard to break and represent one of the
greatest obstacles when it comes to changing energy behaviour.
This is evident in the LIH pilots with retired residents, who
refused to accept new technologies into their daily life. Therefore,
the utilization of the so-called window of opportunity by
merging different energy-efciency measures must be combined
with proper understating of the implementation context. The
optimum results for breaking habits can be achieved by
combining individualized consumption feedback (informative
billing and/or the periodic energy-efciency bulletin) and the
targeted awareness campaign on energy efciency. These cam-
paigns must target the LIHs specically (the economic con-
straints), and take cultural specics into account. It is important
to be aware that LIHs tend to see monitoring as an intrusion of
their privacy. Therefore, a further sociological research/evalua-
tion of this social group is needed. It has to be combined with
awareness raising and providing individualized consumption
 A. Podgornik et al. / Journal of Cleaner Production 130 (2016) 25e34
feedback in a form that will transfer their us versus them
attitude to an it can be done attitude for all the targeted
households. This further research should not be focused on the
devices used in monitoring (IHDs, utility meters etc.) but on how
the attitude barriers can be overcome. Also, it is clear that
additional research work must include experts from social sci-
ences to deal with people's values and attitudes toward pro-
environmental behaviour.
6. Conclusion
Current EU policies and instruments for achieving an energy-
savings potential are mainly focused on technical systems
(building envelope, heating systems, electrical appliances), while
behavioural changes are generally neglected. Energy efciency is
one of essential elements in energy and climate change mitiga-
tion policies, where buildings are the most promising target to
improve the situation. Substantial energy savings can be ach-
ieved in the residential sector with infrastructural and equip-
ment investments, but one of the major factors for energy use in
buildings is the behaviour of the residential end-users. This
behaviour can be inuenced by smart metering, consumption-
feedback systems and awareness campaigns. Low-income
households represent a special residential group, with partic-
ular nancial and social constraints, that needs to be specially
targeted by these actions. This research has shown electricity
savings of between 22% and 27% when individualized and
household-specic awareness campaigns and contextualized
consumption feedback are introduced to the targeted house-
holds. The achieved savings reached the upper boundaries indi-
cated in previous consumption-feedback studies. The savings
reached as high as 36% when efcient energy behaviour was
additionally stimulated by complementary measures. This clearly
shows the potential for behaviour-induced energy savings in
low-income households, when household specics are taken into
account. Also, this further emphasizes the importance of
knowledge when energy behaviour is concerned. Due to the
limited assets of the targeted social group, the use of awareness-
raising features of feedback is even more important. The ob-
tained results clearly conrmed the importance of customized
information and efciency indicators for specic household
groups with a potential to activate the awareness of established
habits and their relevance to energy behaviour. Awareness of
energy consumption and the households' environmental foot-
print also raised the users' interest in additional energy-
efciency measures and to adapt their behaviour to these
changes. Therefore, energy efcient behaviour should not be
neglected in national energy policies and programs. However,
having in mind the recommendations from Caeiro et al. (2012)
and Kang et al. (2012), additional research work has to be con-
ducted for a larger group of LIHs to investigate whether the
described customized feedback can be upgraded with additional
set of consumption indicators and practices of collective energy-
saving activities in the apartment complexes to provide even
greater savings.
Acknowledgement
The authors would like to thank the European Regional Devel-
opment Fund (ERDF) and the partners of the MED Programme
project ELIH-Med e Energy efciency in low-income housing in
Mediterranean (http://www.elih-med.eu/, 1S-MED 10-029) for
their support.
33
References
Abrahamse, W., Steg, L., Vlek, C., Rothengatter, T., 2005. A review of intervention
studies aimed at household energy conservation. J. Environ. Psychol. 25 (3),
273e291. http://dx.doi.org/10.1016/j.jenvp.2005.08.002.
Annala, S., Viljainen, S., Tuunanen, J., Honkapuro, S., 2014. Does knowledge
contribute to the acceptance of demand response? J. Sustain. Dev. Energy Water
Environ. Syst. 2 (1), 51e60. http://dx.doi.org/10.13044/j.sdewes.2014.02.0005.
Beunder, A., Groot, L., 2015. Energy consumption, cultural background and payment
structure. J. Clean. Prod. 94, 137e143. http://dx.doi.org/10.1016/j.jclepro.2015.
01.083.
Bukarica, V., Robi
c, S., 2013. Implementing energy efciency policy in Croatia:
stakeholder interactions for closing the gap. Energy Policy 61, 414e422. http://
dx.doi.org/10.1016/j.enpol.2013.06.052.
Caeiro, S., Ramos, T.B., Huisingh, D., 2012. Procedures and criteria to develop and
evaluate household sustainable consumption indicators. J. Clean. Prod. 27,
72e91. http://dx.doi.org/10.1016/j.jclepro.2011.12.026.
Chiang, T., Natarajan, S., Walker, I., 2012. A laboratory test of the efcacy of energy
display interface design. Energy Build. 55, 471e480. http://dx.doi.org/10.1016/j.
enbuild.2012.07.026.
Chen, C., Cook, D.J., Crandall, A.S., 2013. The user side of sustainability: modeling
behavior and energy usage in the home. Pervasive Mob. Comput. 9 (1), 161e175.
http://dx.doi.org/10.1016/j.pmcj.2012.10.004.
Darby, S., 2006. The Effectiveness of Feedback on Energy Consumption, a Review for
DEFRA of the Literature on Metering, Billing and Direct Displays. Oxford. [On-
line]. Available: http://www.eci.ox.ac.uk/research/energy/downloads/smart-
metering-report.pdf (accessed 16.11.13.).
Desmedt, J., Vekemans, G., Maes, D., 2009. Ensuring effectiveness of information to
inuence household behaviour. J. Clean. Prod. 17 (4), 455e462. http://dx.doi.
org/10.1016/j.jclepro.2008.08.017.
Directive 2012/27/EU of the European Parliament and of the Council of 25 October
2012 on energy efciency, OJ L 315:1e56.
EC, 2006, 2006. European Union: European Commission, Communication from the
Commission e Action Plan for Energy Efciency: Realising the Potential. COM,
545 Final.
Eisenhardt, K.M., Graebner, M.E., 2007. Theory building from cases: opportunities
and challenges. Acad. Manag. J. 50 (1), 25e32. http://dx.doi.org/10.5465/AMJ.
2007.24160888.
ELIH-Med, 2010. Energy Efciency in Low-income Housing in the Mediterranean,
MED (2007-2013). Ref: 3677, Application form, STC Programme MED, [Online].
Available: http://www.elih-med.eu (accessed 16.01.12.).
Elliott, N., Molina, M., Trombley, D., 2012. A Dening Framework for Intelligent
Efciency. American Council for an Energy-Efcient Economy [Online]. Avail-
able: http://aceee.org/research-report/e125 (accessed 16.01.14.).
Faruqui, A., Sergici, S., Sharif, A., 2010. The impact of informational feedback on
energy consumptiondA survey of the experimental evidence. Energy 35 (4),
1598e1608. http://dx.doi.org/10.1016/j.energy.2009.07.042.
Firth, S., Lomas, K., Wright, A., Wall, R., 2008. Identifying trends in the use of do-
mestic appliances from household electricity consumption measurements.
Energy Build. 40 (5), 926e936. http://dx.doi.org/10.1016/j.enbuild.2007.07.005.
Fischer, C., 2008. Feedback on household electricity consumption: a tool for saving
energy? Energy Efc. 1 (1), 79e104. http://dx.doi.org/10.1007/s12053-008-
9009-7.
Gillingham, K., Newell, R.G., Palmer, K.L., 2009. Energy Efciency Economics and
Policy. NBER Working Paper Series, Cambridge. http://dx.doi.org/10.3386/
w15031.
Gynther, L., Mikkonen, I., Smits, A., 2012. Evaluation of European energy behav-
ioural change programmes. Energy Efc. 5 (1), 67e82. http://dx.doi.org/10.
1007/s12053-011-9115-9.
Hayes, S.C., Cone, J.D., 1977. Reducing residential electrical energy use: payments,
information, and feedback. J. Appl. Behav. Anal. 10 (3), 425e435. http://dx.doi.
org/10.1901/jaba.1977.10-425.
Jonsson, D.K., Gustafsson, S., Wangel, J., Ho 2011.
jer, M., Lundqvist, P., Svane, O.,
Energy at your service: highlighting energy usage systems in the context of
energy efciency analysis. Energy Efc. 4 (3), 355e369. http://dx.doi.org/10.
1007/s12053-010-9103-5.
Kang, N.N., Cho, S.H., Kim, J.T., 2012. The energy-saving effects of apartment resi-
dents' awareness and behavior. Energy Build. 46, 112e122. http://dx.doi.org/10.
1016/j.enbuild.2011.10.039.
Kavousian, A., Rajagopal, R., Fischer, M., 2013. Determinants of residential electricity
consumption: using smart meter data to examine the effect of climate, building
characteristics, appliance stock, and occupants' behavior. Energy 55 (15),
184e194. http://dx.doi.org/10.1016/j.energy.2013.03.086.
Langevin, J., Gurian, P.L., Wen, J., 2013. Reducing energy consumption in low income
public housing: interviewing residents about energy behaviors. Appl. Energy
102, 1358e1370. http://dx.doi.org/10.1016/j.apenergy.2012.07.003.
Lapillonne, B., Pollier, K., Sebi, C., 2013. Lessons from Odyssee-Mure project, Energy
Efciency Trends in the EU [Online]. Available: www.odyssee-mure.eu/
publications/br/energy-efciency-trends-in-Europe.html (accessed 13.03.14.).
Levine, M., rge-Vorsatz, D., Blok, K., Geng, L., Harvey, D., Lang, S., Levermore, G.,
Mongameli Mehlwana, A., Mirasgedis, S., Novikova, A., Rilling, J., Yoshino, H.,
2007. Residential and commercial buildings. In: Metz, B., Davidson, O.R.,
Bosch, P.R., Dave, R., Meyer, L.A. (Eds.), Climate Change 2007: Mitigation.
Contribution of Working Group III to the Fourth Assessment Report of the
34 A. Podgornik et al. / Journal of Cleaner Production 130 (2016) 25e34
Intergovernmental Panel on Climate Change. Cambridge University Press,
Cambridge, United Kingdom and New York [Online]. Available: http://www.
ipcc.ch/pdf/assessment-report/ar4/wg3/ar4-wg3-chapter6.pdf (accessed
19.11.13.).
Lopes, M.A.R., Antunes, C.H., Martins, N., 2012. Energy behaviours as promoters of
energy efciency: a 21st century review. Renew. Sustain. Energy Rev. 16 (6),
4095e4104. http://dx.doi.org/10.1016/j.rser.2012.03.034.

chal, K., 2009. An evolutionary perspective on the economics of energy con-
Mare
sumption: the crucial role of habits. J. Econ. Issues 43 (1), 69e88. http://dx.doi.
org/10.2753/JEI0021-3624430104.

chal, K., 2010. Not irrational but habitual: the importance of behavioural lock-
Mare
in in energy consumption. Ecol. Econ. 69 (5), 1104e1114. http://dx.doi.org/10.
1016/j.ecolecon.2009.12.004.
Matthies, E., 2005. Wie ko nnen PsychologInnen ihr Wissen besser an die Prakti-
kerIn bringen? Vorschlag eines neuen, integrativen Einussschemas umwelt-
gerechten Alltagshandelns (How can psychologists better put across their
knowledge to practitioners? Suggesting a new, integrative inuence model of
pro-environmental behavior). Umweltpsychologie 9 (1), 62e81.
Mazur-Stommen, S., Vigen, M., Farley, K., el-Hattab, S., Huff, P., 2013. Trusted Part-
ners: Everyday Energy Efciency across the South. American Council for an
Energy-Efcient Economy [Online]. Available: http://aceee.org/trusted-
partners-everyday-energy-efciency-across-south (accessed 16.01.14.).
Nadel, S., 2012. The Rebound Effect: Large or Small? American Council for an
Energy-Efcient Economy [Online]. Available: http://aceee.org/white-paper/
rebound-effect-large-or-small (accessed 16.01.14.).
Ouyang, J., Hokao, K., 2009. Energy-saving potential by improving occupants'
behavior in urban residential sector in Hangzhou City, China. Energy Build. 41
(7), 711e720. http://dx.doi.org/10.1016/j.enbuild.2009.02.003.
Rodrguez-Barreiro, L.M., Ferna
ndez-Manzanal, R., Serra, L.M., Carrasquer, J.,
Murillo, M.B., Morales, M.J., Calvo, J.M., del Valle, J., 2013. Approach to a causal
model between attitudes and environmental behaviour. A graduate case study.
J. Clean. Prod. 48, 116e125. http://dx.doi.org/10.1016/j.jclepro.2012.09.029.
Shin, D., Curtis, M., Huisingh, D., Zwetsloot, G.I., 2008. Development of a sustain-
ability policy model for promoting cleaner production: a knowledge integration
approach. J. Clean. Prod. 16 (17), 1823e1837. http://dx.doi.org/10.1016/j.jclepro.
2008.06.006.
Ueno, T., Sano, F., Saeki, O., Tsuji, K., 2006. Effectiveness of an energy-consumption
information system on energy savings in residential houses based on moni-
tored data. Appl. Energy 83 (2), 166e183. http://dx.doi.org/10.1016/j.apenergy.
2005.02.002.
rge-Vorsatz, D., Novikova, A., Ko ppel, S., Boza-Kiss, B., 2009. Bottom-up assess-
ment of potentials and costs of CO2 emission mitigation in the buildings sector:
insights into the missing elements. Energy Efc. 2 (4), 293e316. http://dx.doi.
org/10.1007/s12053-009-9051-0.
Vassileva, I., Wallin, F., Dahlquist, E., 2012. Analytical comparison between elec-
tricity consumption and behavioral characteristics of Swedish households in
rented apartments. Appl. Energy 90 (1), 182e188. http://dx.doi.org/10.1016/j.
apenergy.2011.05.031.
Verplanken, B., Wood, W., 2006. Interventions to break and create consumer habits.
J. Public Policy Mark. 25 (1), 90e103. http://dx.doi.org/10.1509/jppm.25.1.90.
Wilson, C., Dowlatabadi, H., 2007. Models of decision making and residential energy
use. Annu. Rev. Environ. Resour. 32, 169e203. http://dx.doi.org/10.1146/
annurev.energy.32.053006.141137.
Wood, G., Newborough, M., 2003. Dynamic energy-consumption indicators for
domestic appliances: environment, behaviour and design. Energy Build. 35 (8),
821e841. http://dx.doi.org/10.1016/S0378-7788(02)00241-4.
Wood, W., Tam, L., Witt, G.M., 2005. Changing circumstances, disrupting habits.
J. Personal. Soc. Psychol. 88 (6), 918e933. http://dx.doi.org/10.1037/0022-3514.
88.6.918.