Examensarbete i Hållbar Utveckling 138
Consumer Behaviour in Renewable Electricity
Can identity signaling increase demand for renewable electricity?
Consumer Behaviour in Renewable Electricity
Raphael Hanimann
Raphael Hanimann
Uppsala University, Department of Earth Sciences Master Thesis E, in Sustainable Development, 30 credits Printed at Department of Earth Sciences,
Geotryckeriet, Uppsala University, Uppsala, 2013.
Master’s Thesis E, 30 credits
Can identity signaling increase demand for renewable electricity?
Supervisor: Johan Vinterbäck Evaluator: Cecilia Mark-Herbert
Examensarbete i Hållbar Utveckling 138
Consumer Behaviour in Renewable Electricity
Can identity signaling increase demand for renewable electricity?
Raphael Hanimann
Content
1 Introduction ... 1
2 Goal and objectives ... 4
3 Background ... 5
3.1 Research within renewable electricity marketing ... 5
3.2 The Swedish (renewable) electricity market ... 5
3.2.1 Development since the liberalization 1996 ... 5
3.2.2 Supply sided characteristics: Green alternatives widely offered ... 6
3.2.3 Demand sided characteristics: Low involvement ... 6
3.2.4 Market potential ... 7
3.3 The interest-action gap for renewable electricity consumption in Sweden ... 7
3.4 Experience from other green products: The Toyota Prius ... 8
4 Theoretical framework... 9
4.1 Identity signaling ... 9
4.2 The role of self-identity in consumer behaviour theory ... 9
4.2.1 The theories of ‘Reasoned Action’ and of ‘Planned Behaviour’ ... 9
4.2.2 TRA and TBA in consumer behaviour research ... 9
4.2.3 The role of identity signaling within consumer behaviour research ... 10
4.2.4 Market-based consumer behaviour research on identity signaling ... 10
4.2.5 Overview of theoretical framework... 11
4.3 Enabling the impact of identity signaling ... 11
4.3.1 The moderating role of public versus private consumption (visibility) ... 12
4.3.2 The moderating role of product involvement ... 12
4.4 Conceptual framework and research question ... 13
4.5 Relevance of research question ... 13
5 What Values does renewable electricity consumption signal?... 15
6 Scope and delimitations... 16
6.1 Other influences on consumer choices ... 16
6.1.1 Product characteristics ... 16
6.1.2 Reference groups ... 16
6.2 Market-related delimitations ... 16
6.2.1 Characteristics of a young market ... 16
6.2.2 Complexity of electricity markets ... 16
6.3 Generalization of identity signaling ... 17
6.4 Survey related delimitations ... 17
6.4.1 Unrealistically high price premiums... 17
6.4.2 Implications of an unprobability sample ... 17
6.4.3 Involvement in survey implies an above average interest in electricity ... 17
7 Research methodology ... 18
7.1 Approach ... 18
7.2 Method ... 18
7.2.1 Unit of analysis ... 18
7.2.2 Data gathering ... 18
7.2.3 What kind of analysis is possible ... 21
8 Results from the preliminary survey ... 22
8.1 Survey statistics: Limited relevance of company size ... 22
8.2 Little experience with symbols for identity signaling ... 23
8.3 Relevance of identity signaling for electricity suppliers in Sweden ... 23
9 Results from the main questionnaire ... 25
9.1 Survey statistics ... 25
9.2 Choice of electricity contract ... 25
9.3 Limited desire for a visible symbol ... 28
9.4 Few differences between more involved and average households ... 29
9.5 Strong influence of trust in renewable electricity schemes... 29
10 Discussion ... 31
10.1 Crucial factors for a successful enabling of identity signaling ... 31
10.2 Insights for policy makers ... 32
10.3 Insights for marketers of renewable electricity ... 32
10.4 Matters for future research ... 33
11 Conclusions ... 34
Acknowledgements... 35
References ... 36
Appendix I – Statistical Evidence ... 39
Appendix II – Preliminary Survey ... 43
Appendix III – Survey: Control Scenario ... 47
Appendix IV – Survey: Symbol Scenario A ... 50
Appendix V – Survey: Symbol Scenario B ... 53
Consumer Behaviour in Renewable Electricity
RAPHAEL HANIMANN
Hanimann, R., 2013: Consumer Behaviour in Renewable Electricity. Master thesis in Sustainable Development at Uppsala University, No. 138, 55 pp, 30 ECTS/hp
Abstract: A higher percentage of energy from renewable resources is an important goal on many sustainable development agendas. In liberalized electricity markets, an increase in demand for renewable electricity can stimulate further investments in it, yet demand for renewable electricity has developed much slower than demand for other green products. So far, research has mainly examined the willingness to pay for renewable electricity, but limited research has been conducted to the motivations behind it. The concept of identity signaling has been proved to play a significant role in consumer behaviour for green products. However, (renewable) electricity in the Swedish residential market typically lacks two important drivers for idenity signaling: visibility and product involvement.
A survey among Swedish households conducted for this study compares choices for renewable electricity under two scenarios, whereof one simulates a higher visibility of and involvement in electricity contracts. The results show that identity signaling has a positive effect on demand for renewable electricity. The results lead to implications for policy-makers, electricity suppliers and researchers.
Keywords: Renewable Electricity, Identity Signaling, Consumer Behaviour, Sustainable Development, Marketing
Raphael Hanimann, Department of Earth Sciences, Uppsala University, Villavägen 16, SE-752 36 Uppsala, Sweden
Consumer Behaviour in Renewable Electricity
RAPHAEL HANIMANN
Hanimann, R., 2013: Consumer Behaviour in Renewable Electricity. Master thesis in Sustainable Development at Uppsala University, No. 138, 55 pp, 30 ECTS/hp
Summary: Energy production is one of the biggest contributor to man-made climate change and has severe environmental impacts. Sustainable development goals most often include plans to reduce the environmental impact of energy production by increasing the use of renewable energy. In Sweden the electricity market is liberalized between 1996, and since that year consumers can choose to purchase renewable electricity.
According to market dynamics, a significant increase in demand for renewable electricity would nudge suppliers towards more investments in renewable electricity production capacities.
Green consumerism experiences constant growth in Sweden as in other European countries, however, demand for renewable electricity develops much slower. So far, there has been considerable research about the willingness to pay for renewble electricity. Consistently it was found that people in general are willing to pay a surplus for renewable electricity. Nevertheless, there has been only limited reasearch behind the motivation to pay extra and on consumer behaviour for electricity in particular.
The concept of identity signaling has a significant influence on consumer choices. Identity signaling suggests that consumers choose products not solely in accordance of their utility, but a product or brand should also match the purchasers own values and its identity. Identity signaling has particularly been proven to motivate green consumerism. However, visibility of the products or the act of consumption and the involvement of the consumer in a certain product category are important drivers behind identity signaling. Both are lacking in the Swedish renewable electricity market.
This study uses a survey to examine if a higher visibility of each households electricity contract and a higher involvement of the Swedish society with the subject electricity could enable identity signaling. A group of respondents received a survey where we simulated a higher visibility and more involvement, another group answered the survey under normal circumstances.
The results show that respondents that answered under the scenario with a simulated higher visibility and increased involvement chose more often renewable electricity. Thus, identity signaling has a positive effect on the demand for renewable electricity. This yields implications for policy-makers, electricity suppliers and furture research in this field.
Keywords: Renewable Electricity, Identity Signaling, Consumer Behaviour, Sustainable Development, Marketing
Raphael Hanimann, Department of Earth Sciences, Uppsala University, Villavägen 16, SE-752 36 Uppsala, Sweden
1
1 Introduction
Energy in the wider context of sustainable development goals
One of the biggest challenges to humanity is to transfer the lifestyle of its vastly growing societies to a sustainable one. Thus, to create societies that successfully enable humanity to comfortably live on earth and from its resources for centuries to come. The number of high-level political conferences that are devoted to finding solutions for more sustainable societies, such as those that recently took place in Copenhagen (2009), Cancun (2010), Durban (2011), Rio de Janeiro (2012) and Doha (2012) showcase the popularity and pressing importance of the issue. The difficulty to achieve effective agreements proves the intricacy involved if short-term interests ought to be tackled in order to achieve conflicting long-term sustainable solutions.
The task to develop today’s societies in a way that they can endure in the long term demands not solely conferences at the global level, but requires action all the way to very local societies. In any case, energy is one of the most pressing, yet one of the most demanding challenges to tackle: On the one hand, energy is indispensable for our societies and economies to function safely and provide nourishment, health and safety. On the other hand, it is the main contributor to man-made climate change and pollution of the earth. Hence, energy is a janus face for those thriving to achieve sustainable development in consideration of the tripple bottom line1. Increasing importance of electricity in the Swedish energy supply
Therefore, the transformation of our current energy system towards one that is based on renewable energies is a major milestone towards various sustainability goals (Sveriges Energimål 2020, EU 2020, UN 2050), but also an eminent challenge. Electricity in particular plays a key role in that transformation. Firstly, because electricity is at the backbone of a highly IT-dependent society, such as Sweden. An immediate break-down of the electricity supply would lead to nationwide chaos, with consequences reaching internationally. With a yearly per capita consumption of 16’000 kWh Sweden’s energy supply is much more electricity reliant than the energy mixes of most other European or OECD countries (Statens energimyndighet, 2010). The useage of electricity in relation to the total energy consumption can even be expected to further grow, for instance with the increased use of personal IT-appliances or of electric vehicles for transport.
If sourced by renewable electricity, the interconnected electric grid – arguably the worlds biggest machine - can be an important part of a solution for a more sustainable, yet reliable energy system, satisfying the demands of all three pillars of the triple bottom line. To achieve this, the economic and technical structures and user behaviour need to change fundamentally to allow the integration of intermittent electricity production from renewable resources.
Influencing the liberalized electricity market requires understanding of market dynamics and consumer behaviour
A first step has been undertaken in 1996, when the Swedish electricity market was liberalized. Because of the liberalized market the end-consumers on the demand side play an increasingly important role in the development of the future electricity system, for instance if their demand for renewable electricity rises significantly.
Thus, the understanding of consumer behaviour for electricity is essential for predicting and influencing the electricity market, and a critical success factor for the development of a growing renewable electricity market in particular.
1 The tripple bottom line is a concept used by the United Nations and corporate social responsibility initiatives (Elkington, 1999) to describe the necessary inclusiveness of economic, social and environmental goals in order to achieve sustainable development. According to the triple bottom line concept, sustainability can only be achieved if economic, social and environmental goals are obtained simultaneously and at equal value.
2
Concretely, the understanding of consumer behaviour within electricity is crucial for
a) policy makers who want to implement effective instruments in order to raise the consumption of electricity from renewable sources, and
b) electricity companies (former monopolies) that want to sustain and grow their customer base in a competitive market.
Identity-signaling as a concept of consumer behaviour can be used to nudge consumers to choose the green option
In order to change consumtion patterns towards more green electricity choices, behavioural changes are advantageous in the way that they are very effective but comparably inexpensive (Loock et al., 2011). Although difficult to control, behavioural changes can lead to results faster than the implementation of policy instruments which have to clear many bureaucratic and political hurdles. Behavioural changes could, thus, also shift consumer patterns towards more renewable electricity contracts.
To understand, predict and influence consumer behaviour, the concept of identity signaling, is a determinant that has to be considered. Which products we choose does not solely depend on how well they satisfy a materialistic need, but also on how we see ourselves and how we want to be seen by others. Behavioural changes have been suggested as an instrument to change consumer behaviour and nudge consumer choices towards healthier foods or towards less risky behaviour in transport (Berger & Rand, 2008). Thus, the choice of a certain product can be used to underline our identity and to signal our values and characteristics to ourselves and our peers.
Thus, consumer choices are partly used as instruments to signal a certain set of values or a desired identity. This is especially the case for products that effectively symbol a certain identity, such as traditional status symbols like fancy cars, but more often applies to green products as well. How we construct our own identity what values how to signal this to our surroundings has been subject to extensive research in the psychological and sociological disciplines and resulted in a number of definitions. To generalize, the kind of person as which we see ourselves and want to be seen by others can be described as our self-concept or self-identity, whereas idenity signaling is the act of communicating the values and characteristics of that person to others.
The self-concept or self-identity is an important driver of consumer behaviour (Paladino & Pandit, 2012) and identity signaling plays an important role for consumer choices. It has effectively been used to reduce risky health behaviours, such as binge drinking or fast food consumption in undergraduate student dormatories (Berger & Rand, 2008). But it also has proven to be an effective force to nudge consumer choices towards more environmental friendly alternatives, for example in the market for hybrid cars (see 3.4). Such examples suggest that a successful strategy to market green products, such as renewable electricity contracts, could be, to make the products visible and effective tools to symbolyze it’s purchasers progressive and green identities.
Market volumes for green products are growing
Although environmental friendly products, such as hybrid cars or the consumption of renewable electricity, are often perceived with either higher costs or less comfort, markets for green products are growing. In the cases in which this perception holds true, the disadvantages of higher costs can be compensated by other benefits, for example status or the signaling of one's identity as a progressive and environmentally friendly person (Loock et al., 2011; Leonidou et al., 2010).
Electricity from renewable resources is also a product that can be categorized as green, therefore, its market share can be expected to grow. However, so far there has only been little growth in the demand for green electricity, eventhough three quarters of Swedish households say they could seriously imagine to purchase electricity from renewable resources (see section 3.2.4).
Limited research in green electricity marketing
Yet, the significant market potential of green electricity, faces an actual low demand. On the one side this can be explained by the young age of the electricity market and the lack of experience from both, marketers of green electricity in selling their products, and consumers to understand the concept of electricity as a range of differentiated products they can actively choose from.
On the other side, only little research has been conducted on consumer behaviour or marketing strategies for electricity and green electricity in particular (Paladino & Pandit, 2012; Rundle-Thiele et al., 2008). However, a preliminary study among Swedish electricity marketers (see section 8) which was conducted as a premeasure for
3 this master thesis report, proved that, on the one hand, the topic of consumer behaviour and identity signaling is perceived as relevant to electricity marketers in Sweden. On the other hand, it showed as well that there has been little experience collected since the market opened in 1996.
The strength of the effect identity signaling has on consumer behaviour depends largely on the visibility of the product category and consumer involvement. Compared to, for instance, a car, an electricity contract has a comparingly low visibility and the choice of an electricity contract usually draws comparingly less involvement.
Thus, concerning electricity, the effect of identity signaling on consumer choices has always to be considered together with the currently low degree of involvement and visibility. It can be suggested that there is a potential underlying demand for renewable electricity products, but there has been a lack of consumer involvement to develop such a market. Consequently, higher consumer involvement and better visibility would be suggested to enable the effect of identity signaling on consumer choices and increase the market volumes for renewable electricity.
The goal of this report is to contribute to a wider understanding of consumer behaviour for green electricity markets. As identity signaling is understood as a potential influence on consumer choices, it is here suggested to be a potential driver for the demand for renewable energy, concretly renewable electricity. A better understanding of identity signaling in connection with electricity contracts should enable to boost demand for renewable electricity, develop successful marketing strategies and, thus, contribute to a faster development of green electricity markets and renewable energy systems.
For this report, the suggestion of using identity signaling to nudge more consumer choices towards green electricity contracts was tested by a survey among households in Sweden. We tried to examine the effect of identity signaling by an experiment with two surveys. Identical sets of questions were sent out to Swedish households, whereas the setting of the surveys were different: a group of respondents received the questions in a basic survey (control scenario), while two other groups of respondents received the questions in an enhanced setting that simulated possibilities for identity signaling (symbol scenarios).
4
2 Goal and objectives
This paper examines the effect of identity signaling on the demand for electricity products, and renewable electricity products in particular. The aim is to identify strategies connected to identity signaling that could successfully change consumer behaviour and raise the degree of involvement for electricity products. The availability for customers to use electricity for identity signaling hypothetically leads to higher demand for renewable electricity products, strengthens long-term customer relationships and enables suppliers to strengthen their brands.
Hypothesis A: We suggest that the percentage of renewable electricity choices will be higher for those survey scenarios that allow identity signaling (symbol scenario A and B), than for the control scenario that does not provide possibilities for identity signaling.
H0: pcontrol≥ psymbol px = percentage of subtractors to
renewable electricity contracts for the control scenario pcontrol
Ha: pcontrol < psymbol and the symbol scenarios psymbol
Hypothesis B: We suggest that there is a significant correlation between the environmental friendliness of the electricity choice and the control and symbol scenarios, respectively.
Since this project should deliver first results on the effect of identity signaling for electricity and potentially trigger further research, the significance level is chosen at α=10%.
5
3 Background
3.1 Research within renewable electricity marketing
In the recent years there has been significant growth in markets for products and services that are considered as
“green” or sustainable, and these markets are expected to grow further in the future (Kotchen, 2005). As a result, the field of Sustainability Marketing as a research discipline has emerged and sustainability has become an important branding topic (Belz & Peattie, 2012). Today, the simple fact of being green can create value for a brand and lead to significant marketing benefits (Porter & Linde, 1995).
However, this growth has not taken place in the renewable electricity sector, with a typical marketshare of renewable electricity2 on the total global electricity sales of about 1% in 2002 (Bird et al., 2002).
Due to – or because of – that, research within renewable electricity marketing has been limited. Most of the so far carried out research on consumer behaviour of electricity is conducted to the willingness to pay for renewable electricity (Kaenzig et al., 2013; Rowlands et al., 2003; Salmela & Varho, 2006). Generally, the price premium3 charged on renewable electricity has not proved to be the most relevant restraint for consumers (Kaenzig et al., 2013; Paladino & Pandit, 2012). Thus, the vast majority of studies conclude a willingness to pay a certain price premium for renewable electricity, while most recent results from Germany show an average willingness to pay a price premium for renewable electricity of about 16% (Kaenzig et al., 2013).
Hence, expanded research has been conducted on the willingness to pay for renewable electricity and green products, whereof a general willingness to pay price premiums for renewable electricity can be concluded.
However, little is known about what motivates people to pay a price premium for renewable electricity. Only few studies investigated the motivational reasons behind that willingness to pay (Rowlands et al., 2003), and behind the fact that despite the rather modest price premiums4, actual demand is still very low.
3.2 The Swedish (renewable) electricity market
3.2.1 Development since the liberalization 1996
For decades electricity has been a regulated utility service. Concretely, electricity was supplied by natural, vertically integrated monopolies at prices regulated by governmental energy agencies. The service was comparable with tap water, thus, a competitive market was inexistent. Followingly, consumers did not have the choice between different suppliers and services.
As of 1996 the electricity market in Sweden is liberalized and suppliers can freely enter and exit markets as consumers can choose whom to buy their electricity from. Since 1999 households could furthermore switch electricity suppliers without transition costs, so that the market reached full competition (Bird et al., 2002).
With the opening of European electricity markets coal and nuclear electricity from Poland, Germany and Denmark also became blended in the standard Swedish electricity mix (Kåberger, 2003).
2 Here, the marketshare of renewable electricity resembles the amount of electricity that is sold under contracts that guarantee an equivalent of the customers consumption to be produced from renewable resources. It does not resemble the amount of renewable electricity delivered to the grid. In Sweden, about 40% of electricity in the grid is produced by renewable resources. Thus, suppliers are only forced to build new renewable production capacities if the demand for guaranteed renewable electricity contracts reaches the already installed production capacity of about 51% (Statens energimyndighet, 2010).
3 The price premium of a product the surcharge or deduction of its selling price as compared to a market average benchmark. E.g. a sweatshirt of a certain brand might sell at a price premium of 20% as compared to the average market price of sweatshirts within the same quality category. Electricity from renewable sources are usually charged with a price premium as compared to electricity from unidentified sources.
4 The price premiums for windenergy among the three biggest Swedish suppliers range between c.a. 2-5%
(Sjöström, 2012).
6
Renewable electricity products have been made available to Swedish households since the opening of the market in 1996. The percentage of renewable electricity contracts varies among different studies, years and definition in use. For instance, some studies include all electricity that is sold with a RECS5 certificate while others only accont for electricity that has been sold under a more strict label, such as the “Bra miljöval” label issued by the Swedish Society for Nature Conservation6. In general are the borderlines between renewable, new renewable and green electricity not exactly clear and vary especially between declarations of different market actors.
According to Ek and Söderholm (2008), 1% of Swedish housholds purchased electricity from renewable sources in 1999; yet Kåberger (2003) writes that green electricity contracts stood for 7% of the total market in 2000, while it stood for 6% of the market share in 2002 according to Bird et al. (2002), respectively 5% in 2009 according to Flood (2009). Thus, the percentage of contracts with an active choice for renewable electricity can be expected below 10% in 2013. Moreover, most of the electricity labelled as renewable or green was not consumed by households, but by companies and institutions (Bird et al., 2002), such as the government-owned railway operator SJ (Ek & Söderholm, 2008).
3.2.2 Supply sided characteristics: Green alternatives widely offered
With the liberalization of the electricity market for residential customers in 1996 the Swedish government laid the fundamentals for a free and competitive market. On the supply side, the market regulations have been further adjusted over the years, e.g. with the ban of a general switching fee between suppliers. As of now there are no significant supply-sided barriers that would hinder the development of a competitive market dynamics.
The three major electricity suppliers are Vattenfall, E.On and Fortum. These three account for 75% of the market volume (Sjöström, 2012; Konsumentverket, 2002). However, the Swedish household can choose among more than 120 suppliers.
All three of the major suppliers offer contracts for renewable electricity, whereas the vast majority of the smaller suppliers also offers such options (Kåberger, 2003).
Since 1999 the market regulations do not allow a general switching fee between suppliers. Such a fee might only apply if consumers want to abort an ongoing contract that runs over a fixed period of time (Bird et al., 2002).
Swedish Households can compare the contract offers valid for their residence swiftly and free of charge on comparison sites online. Services such as www.elpriskollen.se from the Swedish Energy Markets Inspectorate are easy to use and offer a transparent comparison of suppliers. Another decorated online service based on a private initiative is www.elskling.se. Especially with the help of these comparison services, the actual change of a contract does not imply significant administrative barriers, but can be completed within a time span of several minutes online or by telephone.
Thus, there is no shortage of supply or a lack of competition due to the structure of the supply side. Also, the the transfer costs are low due to readily accessible comparison sites, no general switching costs and low bureaucratic hurdles. The so far slow market dynamics are due to characteristics on the demand side.
3.2.3 Demand sided characteristics: Low involvement
As the market became deregulated, an effectual marketing strategy and the understanding of consumer behaviour could be expected to turn into relevant success factors for marketers in order to successfully sell their electricity. However, so far the switching behaviour of Swedish residential customers has prooved to be rather low (Ek & Söderholm, 2006; Ek & Söderholm, 2008) and due to the inactive demand side no competitive market dynamics could develop so far.
5 In Europe the Renewable Energy Certificate System (RECS) enables electricity producers to get a certificate for every megawatt-hour of electricity which is produced by renewable sources. Such a certificate can be traded and guarantee the purchaser that the eqvivalent of the certificate of one megawatt-hour has been produced from renewable sources and been fed into the grid.
6 The “Bra miljöval” label requires not solely production from renewable sources, but sets more specific standards for environmental friendly production. For instance qualify only hydropower plants that guarantee a certain amount and natural residual water flow or biomass plants that use FSC certified raw material.
7 This can partly be explained because the idea of electricity as a product itself was completly new – for marketers to sell it as well as for consumers, who were used to electricity as a utility service.
3.2.4 Market potential
In general, green market segments are growing globally (Emery, 2012). A prominent market segment for green products is the one for LOHAS consumers. LOHAS stands for a lifestyle of health and sustainability. Consumers that identify themselves with that group also desire to buy green or sustainable products in order to signalize their belonging. Today about 20% of American consumers are estimated to belong to that market segment, while this number is still expected to grow (Belz & Peattie, 2012). Although the LOHAS consumer category is originating from America, the demand for green products is growing in Sweden as well. Remarkable examples in Sweden are the market volumes for organic food and biofuels, which both multiplied during the last decades (see 3.2.4; Statens energimyndighet, 2010).
Thus, Swedish consumers show a high sensibility to environmental friendly consumption. At least, this has been proven in terms of food: Consumers from Sweden, along with those from their Scandinavian neigbouring and the Alpine countries spend the most per capita on organic food. In these countries organic food also has the highest market share of total food sales (Willer & Kilcher, 2009).
Obviously, countries differ in the rank-order of which issues have the highest priority on the environmental protection agenda. Furthermore, a high sensitivity for environmentally friendly food consumption, or even a general proenvironmental attitude, does not necessarily imply a high sensitivity for environmental friendly produced electricity (Belz & Peattie, 2012; Salmela & Varho, 2006). However, it can be assumed that there is a general sensitivity for environmental friendly made products among Swedish consumers and, therefore, an underlying market potential for renewable electricity (Sjöström, 2012).
A study by Swedenergy, the umbrella organization for the Swedish energy industry, strengthens that assumption, as it showed that 75% of the responding households could seriously imagine to buy green labelled electricity, while 40% answered that they could even consider being green labelled electricity if they have to pay a price premium for it (Kåberger, 2003).
3.3 The interest-action gap for renewable electricity consumption in Sweden
Although we concluded a significant market potential for green electricity, the numbers of actual purchases are inconsistent with the market potential and especially with the stated interest in the mentioned study by Swedenergy. In that same study, only 1% of the respondents actually bought green electricity, while 40% said they could seriously consider buying green labelled electricity even at a price premium (Kåberger, 2003).
Obviously, there is a significant gap between potentital buyers of green labelled electricity and the actual consumption.
This observation can be further underlined by the results of a study conducted by Kåberger (2003) about environmental labelling in Sweden. He stated that demand for electricity that sells under the “Bra miljöval”7 label from the Swedish Society for Nature Conservation [SSNC], is growing much slower than demand for any other “Bra miljöval” labeled products.
Thus, it can be concluded that there is a potential market for green labelled electricity of significant size, but so far suppliers were not able to skim that potential. The reasons for the slow growth of the Swedish green electricity market assumingly are (partly) to be found in a lack of or unsuccessful marketing efforts. In this paper we suggest, that the failure of marketing attempts to include identitiy signaling theory and adress consumers self- concept is partly responsible for the low demand for green labelled electricity.
7 “Bra miljöval” translates to “Good Environmental Choice” in English. “Bra miljöval” is one of the most recognized among eco-labels in Sweden and applies tough standards that get stricter over time. At times 709 products and services are labelled with “Bra miljöval” in Sweden (Naturskyddsföreningen, 2013).
8
3.4 Experience from other green products: The Toyota Prius
A benchmark for successful implementation of identity signaling in marketing for green products is the Toyota Prius. The Prius was one of the first hybrid cars available on the market, but by far the most successful one, with demand exceeding supply in many markets (Rundle-Thiele et al., 2008). For this study about identity signaling and renewable electricity it is not the mere success of Toyota’s hybrid car which makes this example a significant reference scenario. The example of hybrid cars is relevant because one model (the Prius) produced record high sales statistics compared to the only moderate success of other hybrids (Maynard, 2007). It is a legitimate question to ask wheter exactly this model is so much more successful than other comparable alternatives, e.g. from Honda or Chevrolet.
The significant difference is that the Prius is solely a hybrid model, thus, the car can be identified on first sight as a hybrid. Other hybrids are based on standard models and can only be identified as a hybrid if one spots the sticker on the back of the car. In a study of CNW Marketing Research among Prius owners, environmental concern ranked low on the reasons for purchase (Griskevicius et al., 2010). However, for more than 50% of the respondents the primary reason for chosing the Prius was because it would make a statement about them (Maynard, 2007). The Prius, therefore, is a highly effective and desired symbol to provide its owner with a reputation of environmental friendliness.
Thus, the benefits of the Prius for identity signaling are an important factor in making it more successful than any other hybrid vehicle and giving Toyota the reputation of an innovative and sustainable brand. Similar effects can be expected for renewable electricity and its supplier brands.
9
4 Theoretical framework
4.1 Identity signaling
Within consumer behaviour, identity signaling is since a long time an acknowledged research field. It builds on the observation that people do not only buy products for their direct utility, but also for what those products symbolize. Already in 1959 Levy wrote about "how people behave in consistency with their self-concepts" and that
"choices are made [...] because one object is symbolically more harmonious with our goals, feelings, and self- definitions than another" (p. 120).
Thus, the symbolic meaning of brands and products are used in alignment of one’s self-concept of identity. And products - above all their symbolic meaning - are a medium to communicate one’s values and identitiy to others.
For instance, regular coffee might serve our needs for taste and coffeine as well as fair-traded coffee, but one might choose fair-traded coffee because one thinks of itself as a concerned and careing person and also wants to communicate these characteristics to others. The choice of coffee can symbolize the values whith which that person wants to be associated with.
4.2 The role of self-identity in consumer behaviour theory
4.2.1 The theories of ‘Reasoned Action’ and of ‘Planned Behaviour’
Consumer behaviour is an interdisciplinary research field that employs mainly psychological, sociological and economic theories to assess and predict consumer choices. One of the widely used underlying theories in consumer behaviour research is the theory of reasoned action [TRA] by Fishbein and Ajzen (1975). The model was developed in the psychological and sociological disciplines as a tool to predict and influence behaviour.
Fishbein and Ajzen explain behaviour mainly by attitudes and suggest a causal relationship whereas attitudes are the deciding influence on behaviour. Later the model was extended by Ajzen to the theory of planned behaviour [TBA] (Ajzen, 2012).
The model has been tested in a row of research projects and substantial support has been found for it.
Nevertheless, a growing number of researchers in behavioural studies claim their model to be valid, though incomplete, partly because it lacks inclusion of the theory of identity and self-concept (e.g. Gray, 1985 cit. in Ölander & Thøgersen, 1995; Biddle et al., 1987).
Because behaviour can also include consumer choices, the model’s approach to predict consumer choices by psychological and sociological factors made it very popular in consumer behaviour research (Ölander &
Thøgersen, 1995).
4.2.2 TRA and TBA in consumer behaviour research
As mentioned, TRA, and later TBA, experienced widespread use in consumer behaviour theories in order to underline the influence of norms and attitudes on consumer choices. TBA is frequently used in academic research on consumer behaviour and is a widely acknowledged model to explain consumer behaviour. The model is refered to in major publications about consumer behaviour, from academic articles (e.g. Sparks &
Shepherd, 1992; Terry et al., 1999) to mainstream university textbooks (e.g. Evans et al., 2009; East, 1990;
Kotler, 1991) and to newer literature about sustainable consumer behaviour (e.g. Emery, 2012).
Hence, TRA and TBA rank probably among the most influencial models for consumer behaviour research, although they can only be applied with major delimitations and are occasionally criticized for their incompleteness. This is, because consumer choices depend on many more factors than identity signaling, for instance, availability, product characteristics or purchase habits.
Because of its rather narrow and specific approach, TBA is mainly used in academic research, but has less relevance in practical market-based research (East, 1990). Yet, its widespread use in academia still makes it an appropriate theoretical framework to embrace research within the field (see an overview in section 4.2.5).
10
4.2.3 The role of identity signaling within consumer behaviour research
The aim of this project is to research a potential influence of identity signaling on consumer choices for electricity contracts. However, within both, TRA and the later extended version TBA, self-identity has not a direct causal influence on consumer choices, but only an indirect one. TRA and TBA suggest that identity signaling, along with other impacts, influences our norms and attitues. Our norms and attitudes influence our (consumer) behaviour (Ölander & Thøgersen, 1995). Thus, all behavioural change due to identity signaling happens through adaption of either attitude or norms, whereas identity signaling is not suggested to have a direct effect on consumer choices.
The role of identity signaling, as specifically seen by Ajzen and Fishbein, is neither supported by all research on neither consumer behaviour (East, 1990, p. 108), nor by all research on self-identity within psychological and sociological contexts (Sparks & Shepherd, 1992 ; Biddle et al., 1987).
Still, self-identity (also refered to as self-concept) theory, has its roots in psychological and sociological sciences and is acknowledged to have significant influence within these disciplines, and on behavioural studies in particular (Sparks & Shepherd, 1992). But not all research supports Fishbein and Ajzen’s suggestion of an indirect impact.
Biddle et al. (1987) suggested that “self-referent identity labeling” (or identity signaling) has a direct and independent influence on consumer choices (p. 322), while empirical research supported that hypothesis (p.
329).
4.2.4 Market-based consumer behaviour research on identity signaling
Market based research tends to use more inclusive, but less specific approaches than academic research (East, 1990). Still, psychologic or sociologic approaches are often included in market-based research, especially when theories such as the homo oeconomicus fall short to explain observed consumer behaviour.
The role self-identity or self-concept plays for consumer choices has already been observed by Levy (1959) and has gained further support by empirical research (e.g. Berger & Heath, 2007; Sirgy, 1982).
This has also been the case in relation to environmentally related consumer choices. Sparks and Shepherd (1992) researched motivations for the choice of organically grown food and found self-identity to be a significant factor.
A more recent study by Loock et al. (2011) found self-identity as a significant determinant of energy-efficient behaviour in terms of warm water consumption in the shower. The market-study (see section 3.4) among buyers of hybrid cars has shown that the concept of self-identity has been a major contribution to the success of the Toyota Prius. Finally, Paladino & Pandit (2012) suggested a significant positive influence of self-concept on consumer choices for green electricity.
11
4.2.5 Overview of theoretical framework
Generally, the effect of identity signaling on behaviour and consumer choices finds support in a wide range of research project within psychological, sociological and consumer behaviour disciplines. Graph 1 shows an overview on conducted studies and results that suggest the effect of idenity signaling.
Graph 1: Theory of Planned Behaviour based on Ajzen (2006), extended by the author. Involvement and visibility strengthen the causal influence.
Thus, the importance of identity signaling (or self-referent identity labelling) for behaviour and consumer choices has been supported by behavioural research within psychology and sociology, as well as by research on consumer behaviour in particular. Based on the research presented in graph 1 we assume that there is a causal relationship between identity signaling and consumer choices.
However, whether the effect identity signaling has on products choices (among other factors) is strong enough to nudge significantly more consumers towards green electricity contracts depends largely on product involvement and the visibility of the consumption process. Not all products are equally effective for identity signaling. Thus, a more precise assumption is: there is a causal relationship between identity signaling and consumer choices, if the product in question enables its purchaser to pursue identity signaling with that product.
4.3 Enabling the impact of identity signaling
Thus, the symbolic meaning of brands and products are used in alignement of our self-concept of identity. And products - above all their symbolic meaning - are a medium to communicate our values and identities to others.
However, not all products are equally useful to signalize one's identity. Research suggests that some categories are more relevant for identity signaling than others. Typical identity relevant categories are clothes, favourite artists or haircuts (e.g. Berger & Heath, 2007).
Less identity-relevant domains are, for example, shovels or dishwasher detergent (Suh & Yi, 2006). Also electricity and other commodities (products that bear no perceived differentating features) are generally seen as
12
not identity-relevant goods (Paladino & Pandit, 2012). The two most important determinants of identity relevance are visibility and product involvement8.
4.3.1 The moderating role of public versus private consumption (visibility)
It lies in the nature of identity signaling – or self-referent identity labeling – that certain consumer choices are made in order to signal a set of distinct values that are associated with the product. Thus, the product becomes a symbol to communicate someone’s personal values (Levy, 1959). Followingly, a product needs a certain visibility in order to function as a symbol to signal certain values.
Ratner and Kahn (2002) proved that people adapt their travel plans depending on whether they expect others to hear about their vacation afterwards or not. Concretely, the participants of the study chose a travel plan with much more variety and activity if they knew their vacation would become public to others – in order to appear as an open-minded and interesting personality rather than narrow-minded. This phenomena is, however, significantly weaker when the vacation is expected to be private.
Griskevicius et al. (2010) assume the Toyota Prius to be a very effective way to signal one’s proenvironmental values, because it is highly visible and easily to identity. In other words, the Toyota Prius is suggested to be a very effective product for identity signaling. In their research project where consumer choices for green versus nongreen options were observed, participants switching to the green option chose a green car rather than green soap and they chose green soap rather than a green dishwasher. This implies that choosing green is mostly common for visible product categories, such as cars, while less relevance is put on almost non-visible products such as dishwashers. In a further experiment Griskevicius et al. (2010) observed that participants were more likely to prefer the green option of a product when they were shopping publicly in a shopping mall as compared to online shopping. Accordingly, they conclude that people are “especially sensitive to what their behaviours might signal to others when such behaviour is observable” (p. 396). Hence, the visibility of the consumption process or the product itself increses the likelihood that consumers choose the green option (Belz & Peattie, 2012)
These observations are not suprising, since products can only communicate values and information about one’s identity if they are visible to others. Sirgy (1982) even writes that a product “has to be purchased and/or consumed conspicuously or visibly” in order to have “personality associations” (p. 288). Thus, the effect identity signaling has on product choice can only be expected to be significant if the consumers choice becomes to some degree visible.
Electricity, however, is consumed almost invisibly today, with the only visible act of the consumption process being the bill received by the consumer. As compared to a car, which is driven in public or parked in front of a house, the electricity bill is usually invisible to the customers surroundings. As of today, the lack of visibility does not enable identity signaling for electricity. Part of the research conducted in this project is dedicated to the question wheter incresed visibility enables identity signaling effects. Since electricity as we use it is not visible, other symbols have to be used to signalize brands and types. Paladino and Pandit (2008) suggest “promotional materials such as stickers and fridge magnets which could help a consumer display their use of green power and increase the visibility of their consumption" (p. 380).
4.3.2 The moderating role of product involvement
Wheter a product is publicly or privatly consumed has a significant influence on its benefits for identity signaling. However, the visibility of a product is not the only factor that makes it more beneficial for identity signaling. It has a direct influence on consumer choices, but it is also an important determinant for product involvement. Product involvement is an ultimate determinant of a product’s identity relevance, thus, how beneficial it is to signalize identity values. Thus, the effect of identity signaling on a consumer decision increases with the level of product involvment (Berger & Heath, 2007).
8 Product involvement can be summarized as the general level of interest one has in a product, the relevance someone gives to make the right choice or simply the time spent on a decision (Suh & Yi, 2006).
13 Product involvement and identity relevance of a product category vary among people. For instance, the brand of a PC might be much more identity relevant for an IT specialist than for other professionals. Nevertheless, there are product categories that are generally more identity relevant than others.
In general, hedonic products generate higher product involement than purely functional or utalitarian products do (Berger & Heath, 2007). For example, ones favorite music album or clothing habits are better symbols to signalize identity values or the belonging to a certain clique than the use of a certain dishwasher detergent or broom. This is because hedonic products are by nature intended to serve as symbols for identities or certain values, but do not have a functional purpose. A new haircut, for instance, signalizes that one cares about his or her appearance, while a certain style might further symbolize other values or clique-belongings. Yet, a new haircut is usually not a practical necessity. Dishwasher detergent, on the other hand, serves mainly the purely functional purpose to effectively clean dishes, but carries hardly any deeper meaning about the consumers values on a meta-level.
Electricity is a typical low-involvment good, which is perceived more as a utility service with no differentiating features than an actual product (Paladino & Pandit, 2012). The hypothesis of this study is, to make it more identity relevant in simulating different scenarios (see section 2).
It can be concluded that product categories with a higher level of product involvement are more identity relevant (Stokburger-Sauer et al., 2012). Electricity, so far, is a typical utilitarian, privately consumed and, thus, low involvement product with no or weak linkages to consumer identities. The impact of identity signaling on product choice, as shown in graph 1, is assumed to be rather low. According to our hypothesis, a stronger product involvement and higher visibility of electricity consumption will increase the impact of identity signaling, also for electricity choices.
4.4 Conceptual framework and research question
The presented research on self-concept and self-identity within consumer behaviour and practical examples such as the Toyota Prius provide sufficient evidence to assume a significant effect of identity signaling on the success of green electricity on the Swedish market.
Identity signaling theory has been applied in research for energy related consumer behavior firstly by Loock et al. (2011) in concern of energy-efficient behavior. Concerning the choice of green electricity Paladino and Pandit (2012) as well as Ek and Söderholm (2008) suggested the importance of identity signaling, however, without empirically testing it. According to the literature research undertaken for this study, there is no published study this far that would provide empirical evidence for the importance of identity signaling on consumer choices for electricity contracts.
We expect a higher demand for green products, in this case electricity, if we manage to simulate higher visibility and product inolvement. Firstly, because it was observed that people are more likely to choose the green option if product involvement and visibility are higher and, secondly, because people get the possibility to use green electricity to signalize proenvironmentalism or other values connected to green electricity, both are not the case under the current circumstance of anonymous consumption. This project should be a first attempt to provide empirical results of the suggested importance of identity signaling on consumer choices for electricity.
The goal of this study is to examine how visibility and product involvement could be increased and if that could strengthen the effect of identity signaling on the demand for green electricity. If so, green electricity could become a more identity-relevant product. Therefore it would become more beneficial for identity signaling and demand would rise.
4.5 Relevance of research question
The research conducted is relevant for policy makers, business strategists and academic researchers.
Policy makers, especially bodies of the Swedish government, should be interested in increasing involvement and visibility of electricity consumption
a) if this raises demand for green electricity, in order to reach set energy goals, and b) to generate market dynamics in order to reach a functioning market.
14
Electricity companies that actively want to increase sales volumes of its green electricity portfolios, develop their residential customer base and secure long-term customer relationships by putting forward a distincitve branding strategy.
Academic research within marketing for electricity is still limited. However, with the rather newly liberalized markets growing up more research in this area has become necessary.
15
5 What Values does renewable electricity consumption signal?
Green consumerism has already existed in later decades of the 20th century (Belz & Peattie, 2012) and is associated to different attitudes. Since we research the effect identity signaling has on the demand for green electricity, it is highly relevant to know what values green electricity is associated to. In order to generate a positive effect of identity signaling on demand for renewable electricity, the values it is associated with need to be desirable. Evidently, the effect of identity signaling on the demand for renewable electricity would be negative if the consumers in question associate it with negative connotations.
The values that green consumerism symbolize vary greatly among cultures, sub-cultures, the product category, the marketing efforts and so forth. Thus, it is difficult to define a set values which green consumption yields. In the case of green electricity the definition of values it would symbolize is even more difficult, since the product itself is rather new and the mainstream consumer is not very concious about the meaning and implications of switching to a green electricity contract (Ek & Söderholm, 2006).
An appropriate approach to somewhat generalize the values of green consumerism and green electricity is the LOHAS lifestyle [Lifestyle of Health and Sustainability]. Recent research suggests that about one fifth of Americans can be categorized as LOHAS consumers, whereas the phenomena has spread over the globe (Belz &
Peattie, 2012). LOHAS are usually seen as early-adopters and trendsetters living a modern and urban lifestyle.
We assume that green electricity would rather be connotated with modern and progressive lifestyles, than with extreme environmentalism and anti-consumerism. Hence, we expect renewable electricity to yield values that can be assumed to be desirable for the mainstream public.
Nevertheless, it has to be considered that green consumerism can for some people signalize undesired identities with a negative connotation such as environment geek [miljönörd] or health freek [hälsofreak]. For others renewable electricity might have negative connotations because they oppose the shareholder-value driven producers behind it or large-scale industries in general. However, to achieve a significant higher demand for renewable electricity it is relevant that the mainstream consumer connotates renewable electricity with positive values.
Summing up, academic research suggests that green consumerism has in general positive associations and is even connotated to high status, altruistic attitudes and moral responsibility (Paladino & Pandit, 2012; Salmela &
Varho, 2006; Griskevicius et al., 2010). Concerning Sweden in particular, we assume the same generally positive associations with green consumerism, especially in accordance to its high level of organic food sales and the Ek and Söderholm’s (2008) suggested positive self-image of green consumption for their research among Swedish households.
16
6 Scope and delimitations
6.1 Other influences on consumer choices
6.1.1 Product characteristics
Identity signaling is an important aspect of consumer behaviour, but consumer choices are influenced by other factors as well. In the case of electricity contracts such other factors can include, but are not limited to the price, customer service quality, availability of options, satisfaction with personal encounters, advertisments or sponsoring, individual affection to a certain supplier and so forth. All these factors are not accounted for in this study. Instead, only one variable, the possibility for identity signaling, should be changed to observe it's impact on results. All other factors are assumed to stay unchanged (ceteris paribus).
Kaenzig et al.(2013) suggest also a possible impact of default options on the choice of electricity contracts, thus a basic product choice that is automatically selected, if the customer does not actively make another choice.
However, in our survey there are no default options, thus, there is no preselected option the respondent would get if no active choice is made. By doing so, a potential influence of such default choices is avoided.
6.1.2 Reference groups
People that desire to signal a clearly more than average green identity might not want to be associated with green large-scale industries. The same applies to LOVOS consumers [Lifestyle of voluntary simplicity], who are rather expected to positively connotate with particularly less, as well as more local, consumption (Belz & Peattie, 2012), but see large-scale renewable electricity grids as non-desirable. Furthermore, there might be out-groups that distinctively want to distance themselves from new green trends or ”environment geeks”. These groups might have negtive associations to renewable energy and therefore desire to signal an identity with a particular non-green image. For members of these groups, the visible consumption of e.g. windenergy could yield negatively associated values to their peers, therefore the effect of identity signaling has an expected negative influence on the demand for renewable electricity. However, it has to be expected that members of those groups are included in our sample. An example of such a group relevant for this study is the Swedish Society for Landscape Conservation (Föreningen Svenskt Landskapsskydd), which has a strong environmental-friendly perspective, but wants to prevent the landscape from being disturbed by large-scale electricity production sites, such as wind parks or hydrodams.
6.2 Market-related delimitations
6.2.1 Characteristics of a young market
The electricity market for residential customers is rather young, it exists in Sweden since 1996. The idea of electricity as a range of differentiated products, from which customers can choose, is a new one. The passivity of the households and the low switching behaviour towards greener choices, can, partly be explained by the young age of the market. Experiences from other market liberalizations, e.g. the one of the telecommunications sector, show that switching increases over time as, markets mature (Bird et al., 2002).
6.2.2 Complexity of electricity markets
The functioning of electricity markets, especially the idea of green electricity contracts, is very abstract and can be difficult to understand. The electricity bill consists of a row of items such as the price for the electricity itself, tax, transmission fee, distribution fee and so forth. For consumers it is especially challenging to see why they should pay more for green electricity if it is sent through the same grid like their neighbours’, which might not pay the premium. The distincition between the economic and the physical path of electricity consumption needs to be more clearly explained.
17
6.3 Generalization of identity signaling
Psychological and sociological research sometimes examines behaviour choices in terms of, either internal or external identity signaling (Berger & Rand, 2008). Strictly speaking, there are two different reasonings for internal and external identity signaling. Internal identity signaling refers to "how I see myself", while external identity signaling refers more to "how I want to be seen by others" (Biddle et al. 1987). Thus, one could think "I am more this type of person, but I would like to be perceived as another kind"; in such a case internal and external identity signaling lead to different behaviours or consumer choices. In our case, internal identity signaling could mean that one purchases renewable electricity because one likes to see oneself as an environmental friendly person, without considering how peers might perceive this. External identity signaling would mean that one purchases renewable electricity in order to signalize a green identity to the others, without considering one's own opinion on renewable electricity.
Most often the two mechanisms are rather converging than diverging, thus, one could say "I see myself as an environmental friendly person, and I want my peers to perceive me in that way". In mainstream consumer behaviour research there is hardly a distinction between internal and external identity signaling, thus, they are assumed to converge. In this project, we assume the two mechanisms to be converging and summarize them simply as identity signaling.
6.4 Survey related delimitations
6.4.1 Unrealistically high price premiums
The price premiums on electricity from renewable resources were significantly higher in the survey than they are on the market. In the survey the price premium for windenergy was 10%, whereas the price premiums on the market are between 2-5%. The unrealistic high price premium was intended to reduce the usual gap between what people state they are willing to pay in surveys, and their actual willingness to pay in real purchase situations.
6.4.2 Implications of an unprobability sample
Since an unprobability sample is used for the survey, the results cannot, in a statistically correct manner, be generalized for the population of private Swedish households (Ritter & Sue, 2007). Rather, the survey results serve as first insights on the effect of identity signaling on consumer behaviour for electricity. For statistically relevant results, that can be generalized to the Swedish households, extended research that exceeds the frame and possibilities of a master thesis is necessary.
6.4.3 Involvement in survey implies an above average interest in electricity
The fact that repondents took time to answer the survey implies, that they are likely to have a certain interest in the subject of (renewable) electricity. Less interested households are more likely to have ignored the invitation to this survey. Therefore, it must be assumed that the sample resembles respondents that are on average more interested and involved in the topic than an average household.
This could be one factor that explains why a relatively high rate of the respondents of this survey said they buy renewable electricity, which is much more than the Swedish average (see section 3.2.4).
18
7 Research methodology
7.1 Approach
The problem is approached through the analysis of primary and secondary data. Firstly, secondary data was used for a literature review in order to assess the current status of research on consumer behaviour for renewable electricity. Also, a theoretical background on identity signaling theory was concluded from secondary sources.
Secondly, the relevance and perception of identity signaling throughout the industry was assessed by qualitative as well as quantitative research, concretely with phone call interviews and a preliminary online survey among Swedish electricity suppliers. An additional aim of that stage was to assess wheter first measurements have already been tested in practice, and – if so – what experiences have been concluded.
Thirdly, the main research body consists of quantitative empirical research, namely of an online survey sent out to households in Sweden.
Online surveys have a row of advantages as compared to paper surveys and interviews: Depending on the availability of software tools and the conductors personal computer skills, online surveys can be considerably cheaper than paper surveys, especially as compared to surveys that are subject to postage. Also, the return of the data follows immediately as respondents answer the questions, so that no time is lost for mailing (Gaddis, 1998).
The number of error can also be improved, since paper does allow respondents to send back questionnaires that are incomplete or answered in an incoherent way (Ritter & Sue, 2007). Another advantage of online surveys is, that the data is readily available to be processed in statistics programs, while paper surveys first have to be digitalized.
According to the financial limitations of the master thesis and the restricted time frame, the online survey was an appropriate method to use. A disadvantage of online surveys is the difficulty to find and select a significant number of respondents. Unlike post mail adresses, email adresses are difficult to obtain (Poynter, 2010).
Institutions such as universities or companies usually have email-registries of their employees or members (Ritter & Sue, 2007), however, to choose the employees of a university as a sample can easily biase the results, so that they might not be valid for a general household.
To build a database of respondents that can be reached by email is time consuming and needs to be registered with the Swedish Data Inspection Board [Datainspektionen].
7.2 Method
7.2.1 Unit of analysis
Although electricity companies are involved in preliminary research, the insights about the preliminary research are solely intended to better understand the practical implementation of identity signaling in current marketing strategies and, to improve the main survey among households.
The level of analysis for the main survey, that will help us to test our hypothesis, are the residential endconsumers of electricity, thus, the private households in Sweden. The number of households is constantly changing, and there is no finite list about the number of households that can be reached by email. Such moving populations are called open populations (Ritter & Sue, 2007).
7.2.2 Data gathering
Preliminary survey among electricity companies
The preliminary survey was sent to 91 electricity suppliers in Sweden with the goal to examine energy suppliers’
interest in identity signaling research and the steps they have undertaken or plan to undertake in order to enable identity signaling for their customers. The electricity suppliers were picked from a member list of the trade association Swedish Energy. All members with electricity distribution to end-customers were adressed.
The answers of suppliers were used
