1. Introduction
Over the past several years, many countries have been liberalising the electricity market, enabling consumers to choose their supplier freely. One of the first countries to reorganise the energy sector in 1987 was Chile. A typical enterprise strategy in this sector was vertical integration, where several large companies often had controlled generation, transmission, distribution and retailing of energy on domestic markets [1]. It was assumed that breaking up the vertical integration would allow enterprises to be competitive in the area of generation and retailing of energy, while transmission would retain its monopoly status. However, the reason for the reform implementation in each country was different. In the UK, it was decided to privatise the state-owned electricity utility [2]. In Central and Eastern European countries, it was the decentralisation of government control of the energy sector and partial privatisation of the industries, while in the US, increased competition between private suppliers and reduced regulation [3] was the main driving force. The Electricity Market Closeness Index calculated for a total of 55 countries shows significant progress in the liberalisation of the energy market in the most developed and developing countries in 1989–2007 [2]. The countries with the highest energy market liberalisation are the UK, Germany and Spain. As [2] confirms, the energy market liberalisation process is influenced by the industry sector, foreign financial support, and government ideology. Undoubtedly, one of the effects of energy sector reforms is the third-party access (TPA) to the electricity transmission grid. This allows each end customer to choose an electricity supplier.
The article aims to identify factors influencing the intention to change an energy supplier among customers of incumbent suppliers. The paper proceeds as follows: the next section provides a theoretical background with four research hypotheses. The sections that follow present research methodology and study results. The paper ends with the conclusions with some managerial implications and proposals for future research.
2. Theoretical Background
The most common indicators of the electricity market competitiveness refer to a change of the supplier or a change of the contract with the current (often incumbent) supplier. Unfortunately, a relatively small percentage of households decide to change their energy suppliers and, consequently, in many markets, current suppliers have a dominant market share. The data at the end of 2006 showed a relatively low tendency to switch suppliers in most European countries [4]. Since the market opening, suppliers have been changed by a few to several per cent of customers. A larger percentage of supplier changes was recorded in Sweden and Norway (around 30%) and Great Britain—47%. The situation in ten of the US states where the energy market was liberalised was similar to Europe. A supplier has been changed by approximately 12% of customers [4]. A much higher percentage of supplier changes was recorded in several Australian States—it averaged around 40%. The data from the 2016 EC Report [5] indicate that, on average, the supplier change rate among the 28 European Union countries is 14%. Among these countries, the largest percentage of inhabitants, that is 28, who changed energy suppliers was in the Netherlands, Great Britain and Ireland. The average ratio of the supplier change achieved in Portugal and Italy was 14% and 13%, respectively. Bulgaria and Romania, in turn, have the lowest percentage of people who changed the energy supplier and that is below 1%. Among EU citizens who switched suppliers, 11% did so because of the change of residence. If we exclude this part of the population, it turns out that the percentage of people voluntarily switching energy suppliers decreases to 12% (Table 1).
Additionally, the percentage of consumers who have changed the tariff of the current supplier in the last three years has been analysed. On average, in each of the 28 EU states, 13% of consumers made such a change, with the largest percentage of those recorded in the United Kingdom—28%, followed by Portugal—19% and the Netherlands—18%. The least frequent tariffs changes were performed by the inhabitants of Bulgaria—1%, Iceland—1% and Hungary—2%. As shown in the data presented in Table 1, some countries have both a high percentage of energy supplier change and tariff change (e.g., Great Britain, Netherlands, Belgium). In some countries, a relatively low percentage of energy supplier change was compensated by a high percentage of consumers changing tariffs (e.g., Estonia, the Czech Republic, Spain). There is also a clearly indicated group of countries where residents neither change their energy suppliers nor tariffs (e.g., Bulgaria, Hungary, Romania).
Customer retention is a key issue for many industries. This refers to the continuation of the transaction with the firm [6]. The research presents the service industry as being particularly sensitive to customer retention [6,7,8]. There is no doubt that such companies are exposed to high losses due to customer churn. Churn does not necessarily mean terminating the company’s services; some authors also include limiting the purchase amount [9]. The customer churn directly affects the length of the customer’s relationship with the company and the customer lifetime value (CLV) [8]. As researchers point out, churn does not always have to be a negative phenomenon. Although it is most often associated with consumer dissatisfaction with the services provided, researchers also indicate positive churn [10]. Positive churn occurs when the consumer resigns from the services provided, but the customer’s predefined objective is satisfied, or the service is obsolete [10].
In the literature, a supplier change is compared to a change of the sales channel; in both cases, there are switching barriers and opportunity costs [11]. A change of purchasing channel may or may not lead to a change of services provider. The consumer may change the channel of the same supplier during the purchase process, change the channel, and change the supplier (cross channel free-riding) [12]. Customer migration can be caused by push effects, pull effects, or mooring effects—PPM model switching [13]. An example of a push factor may be the low level of satisfaction; pull factor—the attractiveness of alternatives; and mooring effects—the high cost of change. The same three factors—satisfaction, cost of change, and alternative attractiveness—were most often analysed in the service industry research [14]. The effect of the first two factors on changing suppliers is much greater than the attractiveness of alternatives.
Poland is one of these countries where the coal market is particularly relevant as coal is the first choice to meet energy demand in this country. From 2019–2020 the percentage change of hard coal and lignite generation (bars) production was −8.0%, while the average decline for the EU was −18.0. In 2020, in Poland, 83% of electricity came from fossil fuels [15]. In Poland, the greatest share in electricity sales to final customers is still held by incumbent suppliers, which are default suppliers to households that have not selected a new supplier. In 2019, five incumbents and over 136 alternative trading companies were active in the electricity supply to final customers, including suppliers operating on the household market [16,17].
The authors of [18] pointed to the critical importance of the behaviour of incumbent energy companies for a transition to a sustainable energy system. The goal of the competitive reform of the European electricity industry was that all customers will be able to choose freely their electricity supplier [19]. There is a vast number of studies identifying the barriers for consumer switching behaviour in the energy market—for some of these research the rational choice theory is useful—while the other part of research has its roots in behavioural economics [20].
As a result of the liberalisation of the Polish energy market, competition has intensified, and one of the big concerns for incumbent suppliers is customers switching to competitors. Consumer switching can generate lower profits and higher costs [21]. Overall, in today’s highly competitive market the level of customer loyalty is declining [22]. Moreover, it costs several times more to acquire new customers than retain current ones. For this reason, research on the factors influencing the tendency to abandon an energy supplier seems justified. Loyalty program, company image and price transparency have been studied extensively on their own. However, to the best of our knowledge, no prior research integrates these domains to establish a set of factors as predictor of the intention to change the incumbent energy supplier. This is what this study seeks to do. We believe that our article follows the trend of describing and analysing the situation of incumbent energy suppliers in various European countries after energy market liberalisation [20,23,24].
According to researchers, transparency is one of the most prominent demands of consumers today [25]. Price uncertainty is a permanent element of consumers’ choice decisions and companies tend to reduce this by price communications activity [26]. Price transparency as well as other information are useful in defining a value of the offer and enables for a conscious decision making [27]. Provision of price information creates the possibility to choose where and how the customers obtain the offer they look for [28]. Price transparency is defined as “the degree to which market participants know the prevailing prices and characteristics or attributes of goods or services on offer” [29]. The issue of price transparency has been commonly analysed in the e-commerce context [29,30], transactions in the stock market [31] or health care sector [32,33,34]. The popularity of this phenomenon is strongly connected with the growing popularity of multi-channel sales when online channel plays a crucial role. As [26] stated, “the Internet represents the ideal platform for enabling price transparency to flourish”. Now consumers are armed with information provided by the Internet and in effect the information asymmetry diminishes [35]. One of vital advantages for potential buyers is the possibility to make clear price comparisons [36]. Researchers rightly noted that higher price transparency creates benefits for buyers at the expense of sellers [30,37]. The higher the perceived pricing transparency the easier the offer and price structure evaluation [38]. The authors of [39] indicated the moderating effect of the price transparency in the relationship between the purchase intention and the purchase behaviour. Price transparency is identified as one of factors that affects behavioural intention. According to [40], price transparency may decrease consumers’ intention to purchase a product in the context of green buying decision. However, there is an important element of the meaning of price transparency—when the consumer notices a price increase, price transparency can be helpful in understanding the reasons behind such a price change [41].
Therefore, we proposed the H1 hypothesis:
Price transparency positively impacts the intention to change an energy supplier.
Today’s competitive market requires loyalty programs [42] that are seen as the state-of-the-art marketing tools [43]. Loyalty programs are a growing marketing tool for managing retention and creating switching costs [44]. According to [45], loyalty program is “any institutionalised incentive system that attempts to enhance consumers’ consumption behaviour over time.” Loyalty programs are an important marketing and strategic tool used in order to impact repurchase intention and create stronger customer relationships [46,47]. The main goal of any loyalty program is to foster and reward customer loyalty [48,49]. Technological progress creates new ways of interacting with customers and collecting rich customer data [33] as well as higher level of consumer engagement [50] that is of high interest due to considering the high costs of customer acquisition [51]. The authors of [52] emphasise the growing role of loyalty programs in shaping the overall customer experience; for this reason loyalty programs should become less transactional and provide new interaction opportunities in the customer journey. Despite the constant popularity of loyalty programs, the question of the impact on the quality of the customer’s relationship with the company is still valid [53]. Many empirical studies suggest that a loyalty program has a positive impact on customer loyalty [54,55,56], and it also alters purchasing that is customer purchase volume/frequency and share-of-wallet [48,57,58]. However, some research publications are devoted to the negative aspects of loyalty programs [59,60]. These publications noted the dark side of loyalty programs connected with negative experiences among non-member customers. According to [54], the customers’ feeling of under-rewarding can cause weaker repurchase intention due to perceptions of injustice. It means that the construction of a loyalty program requires great caution so that it does not turn out to be a harmful marketing tool.
Therefore, we proposed the H2 hypothesis:
Loyalty programs negatively impacts the intention to change an energy supplier.
Corporate image may be defined as social or public knowledge or perceptions about a given company [61]. Earlier definitions of corporate image referred to the overall impression of the firm held by the public [62] or associations and meanings connected with an organisation [63]. An interesting review of corporate image definitions has been presented by [64]. Crucial resources (money, time, people, etc.) are used to build a strong corporate image [65]. Researchers have seen the corporate image as the decisive factor in the process of influencing consumers’ subjective perception and their subsequent behaviour. Corporate image along with customer satisfaction, and perceived value are the significant predictors of the customer behaviour intention [66]. Corporate image may be influenced by the actions of contact personnel [67]. Brand image is seen as the precondition both for value and loyalty [68,69]. As a result of establishing a strong corporate image the following phenomena may be generated: consumer preference and loyalty, company credibility and its sustainable competitive advantage [70]. Although [71] noted a “marginally significant” influence of corporate image on purchase intention and [72] identified as an insignificant impact of green image on behavioural intention, most studies indicated a positive impact of corporate image on customers’ behavioural intention [73,74].
Therefore, we proposed the H3 hypothesis:
Corporate image negatively impacts the intention to change an energy supplier.
By analogy to corporate image, in our study the green image was also taken into consideration in our study. A green image refers to the subjective perception of clients regarding the image of the company about the environment [75]. It is the consumer’s perception of the corporate brand to be environmentally sustainable and eco-friendly [76]. According to [77], green image can be understood as the consumers perception consumers of firm’s commitment to environmental protection. The authors of [78] indicated two challenges for the organisations—to produce eco-friendly products that match if not outperform the functional performance of conventional products. The other one is to generate positive perception towards eco-friendly branding activities. According to [79], the effect of the green innovation behaviours of enterprises can be consumers’ green purchase intention. There are other studies that noted the positive impact of green image on customer re-visit intention [75,80,81]. We believe that, taking into account the industry uniqueness, it is reasonable to distinct the consideration of the environmental image of energy supplier in the research model.
Therefore, we proposed the H4 hypothesis:
Company green image negatively impacts the intention to change an energy supplier.
3. Materials and Methods
Methodologically, our study falls under industry survey-based empirical study. As it is indicated by [82], such studies have often utilised statistical methods such as structural equation modelling. The research results were used to a multigroup SEM (Structural Equation Modelling) analysis using AMOS 26 to analyse the hypothesised relationships in the research model. Researchers indicated several reasons for choosing the structural equation modelling. In our model, where there is no indirect effect, the strongest argument relates to the fact that SEM enables evaluation of the overall construct in its entirety and examines the interrelated relationships among a set of posited constructs simultaneously, with each construct measured by one or more observed items (measures). In effect SEM makes it possible to verify our research hypotheses [82,83]. For the research purposes of this article, only a part of the research questionnaire, prepared as part of a larger project, was used. Except for the demographic part of the questionnaire, all questions were based on a 10-point Likert scale (1-strongly disagree; 10-strongly agree). In the SEM research model latent variables are expressed by several ‘manifested’ items in the questionnaire [84]. Number of points on the scale is subjected to a lot of debates and controversies [84]. The author of [85] indicated that 10-point Likert scale is comparable to 5-or 7-point scale for analytical tools such as equation models. However, research made by [86] showed that in path modelling, the 1-10 scale gives clearly better measures of model fit (in terms of its internal and external stability) compared to the 1-5 scale.
In our research model (Figure 1) four factors have been taken into consideration: perceived price transparency, loyalty program, company general image and company green image. What is vital for the research idea is that all of these factors are strongly connected with the energy supplier.
All latent variables were adapted from the literature. The behavioural intention in the research model is the customer intention to change the actual energy supplier. Behavioural intention (INT): INT1: I was considering changing my current energy supplier; INT2: I am planning to change my current energy supplier; INT3: There is a good chance that I will change the energy supplier. Perceived price transparency (PT) PT1: I know what I pay for in my energy bill; PT2: The content of energy bill is clear to me; PT3: I know how much I pay per kWh. Loyalty program (LP): LP1: I use loyalty programs from various sectors; LP2: I use my current supplier’s loyalty program; LP3: I find the rewards of the energy provider loyalty programs attractive. Company image (CI): CI1: My energy supplier cares about its image; CI2: My energy supplier is a famous brand; CI3: My energy supplier runs their business ethically; CI4: It is important to me that my energy supplier represents domestic capital. Green image (GI): GI1: I would like to use green energy. GI2: I have a positive opinion about my supplier’s environmental activity; GI3: I believe my supplier cares about the environment.
4. Results
4.1. Research Sample Characteristics
The survey was carried out at the beginning of 2020, before the SARS-CoV-2 pandemic in Poland, and the research sample consisted of 1216 adults (Table 2). The respondents were customers of one of the largest energy suppliers in Poland, an incumbent supplier. This incumbent supplier owns approximately one-third share of the domestic market (energy sold to the end users). This company serves about 5.5 million households. We used quota sampling method—the household size was an important criterion for the research sample structure. Before the formal data collection, we provided the paper questionnaires to a group of potential respondents in the two customer service offices, and then invited these people for the pretest. After some small revisions, we conducted formal data collection. Data collection was done in Poland. The formal research was carried out using a paper questionnaire, which was distributed to the respondents in customer service offices of one of the largest energy company in Poland. These customer service offices were located in 41 cities, in 8 out of 16 voivodeships in Poland. During the questionnaire distribution, it was assumed that every 10th client will be invited to the survey. There were a small number of refusals during the study. After the research, the interviewers checked the quality and completeness of each questionnaire. Incomplete questionnaires were not included in the research sample.
4.2. Study Results
4.2.1. Research Model for the Total Sample
Our research goal was to verify the research model not only for the research group as a whole but also for both of two customer groups distinguished on the basis of the amount of the monthly bill paid for energy (low vs. high (Low energy bill (L group)—below 45 euro per month; high energy bill (H group)—above 45 euro per month.)). For this purpose, a multi group SEM (Structural Equation Modelling) analysis using AMOS 26 was conducted. The CFA models were performed using a Maximum likelihood (ML) estimation. The ML estimation method has been described as being well suited to theory testing and development [83]. The estimates presented relate to the standardised regression weights. Table 3 presents convergent validity and discriminant validity. The model fit for our research model is as follows:—CMIN/DF 2.765, GFI 0.963, AGFI 0.943, RMSEA 0.027 (LO 90 0.025—HI 90 0.030), PCLOSE 1.00.
In this study, AMOS 26.0 was used to test the goodness of fit of the model. It was generally considered that the model with root mean square error of approximation (RMSEA) less than .08 and goodness-of-fit index (GFI), and adjusted goodness-of-fit index (AGFI) greater than .9 had the better goodness of fit. All the measured variables met the requirements.
The first analysis concerns the total research sample—consumers as a whole. Both Figure 2 and Table 4 refer to the total research sample. The featured model (Figure 2) explains 9.0 per cent of the dependent variable (the intention to change an energy supplier).
The results revealed that the Cronbach’s alpha coefficients ranged from .71 to .91. The minimum acceptable value for Cronbach’s alpha considering .70 as the threshold value for already developed constructs and .60 for newly developed constructs [82]. The degree of interrelation for the measurement items of each construct was measured through convergent validity and was estimated by calculating the average variance extracted (AVE) and composite reliability (CR). A strong interrelation of items was considered based on the reference value (AVE ≥ .50 and CR ≥ .70) [80]. According to the tests of reliability and validity, it seems that adequate reliability and validity exist in this study.
4.2.2. Research Models for Two Groups
The second analysis concerns the group paying relatively low energy bills. Both Figure 3 and Table 5 refer to the L group.
The last analysis concerns the group paying relatively high energy bills. Both Figure 4 and Table 6 refer to the H group.
5. Discussion
The overall purpose of this research study was to explore factors affecting the intention to change an energy supplier among customers of incumbent supplier in Poland. We believe that the changes taking place on the energy market in Poland justify the analysis of this market. Moreover, the presented conclusions may be valuable for other countries in Europe and beyond. The presented conclusions may be particularly important for countries where, like in Poland, the incumbent suppliers still have a high share in the energy market, e.g., Lithuania, Hungary and Romania.
The presented results allow for a formulation of several important conclusions and, consequently, suggestion of some practical recommendations. First, when analysing consumers in general, it is possible to overlook important relationships. However, some relationships are so strong and common that they are visible in both general and detailed analyses (L group vs. H group). A manifestation of such dependencies is the most significant and negative impact of the general image of an energy supplier on the clients’ intentions to change same. This shows an undoubtedly stronger position of these enterprises with a strong image on the market regarding retaining their customers. In this respect, incumbent suppliers naturally have an advantage over potential new energy suppliers. Thus, the position of existing suppliers on the market creates a barrier to market entry. Expenditures on strengthening and presenting the brand image create an entry barrier for potential new suppliers. Our finding on the impact of the brand image finds a strong confirmation in the literature [88,89]. Brand perception is helpful in understanding the competitive landscape and creating effective marketing communications [90]. Brand image is one of the components of the brand knowledge alongside brand awareness, brand recall and brand associations [91], thus it is highly recommended investing some company resources in improving not only brand image but also related elements. The brand image is linked with the associations kept in consumer’ mind [92]. Some researchers pay attention to the lower impact of brand image on consumer loyalty due to the strong competition in the market, as a result investment in corporate social responsibility (CSR) is advisable [93]. The fact that the concept of CSR includes not only environmental efforts, but also economic, legal and philanthropic efforts seem essential for understanding the next conclusion from our study.
The second important conclusion is the lack of importance of the ecological image on the intention of customers to change the energy supplier. To conclude, the influence of green image on customer behavioural intentions was found to be non-significant. Such a conclusion is consistent with the results presented by [72], who noticed that the fact of positively perceiving a firm’s green image can be a motivation for consumers to make a positive evaluation of this service provider in terms of cognitive attributes, but it does not necessarily determine their purchase decisions. The authors of [94] also indicated that consumers’ general opinions are not strong enough to pro-social acting—this is called the value-action gap [94]. In our opinion, the lack of influence of environmental issues on decisions to change the energy supplier proves the low level of ecological maturity of Polish society, for whom the ecological commitment of the energy supplier seems irrelevant to their decision to continue cooperation with this particular company. In effect, there is the lack of positive relation between consumers’ environmental beliefs and their environmental behaviours. The research was carried out in Poland where a significant part of energy is still generated from coal and only the political pressure from the European Union (EU climate policy) prompts the country to decarbonise the energy system. The awareness of the negative impact of this non-renewable energy sources on environment pollution and the related health problems for the society is still low. The more general examination of no influence of green image on customer behavioural intentions can be found in the concept of the ‘selfish altruism’ proposed by [95]. According to this concept, customers’ green purchase intentions depend on their assessment of the perceived potential beneficiaries of eco-friendly offer. Thus, there is a greater likelihood of purchasing an ecological offer if it is beneficial not only for the planet but also for that particular buyer (in reference to those that exclusively benefit the planet) [72]. This implies that individual households tend to make a specific pro-environmental purchase decision when they believe that this conduct provides a direct benefit to them.
In the general model, there are no visible differences in the effect of perceived price transparency, as indicated by the division into two groups according to the amount of the monthly electricity bill. The perceived price transparency has a significant positive impact on the intention to change the energy supplier among customers with high energy costs. What is important to them is the legibility of the electricity bill and the individual components of the total amount payable. Therefore, it can be seen that the more a user pays for something, the more he/she is interested in the details of what he/she pays for. As a consequence, the price transparency of the bill for people paying a high price for energy positively influences their willingness to change an energy supplier. Price transparency allows such people to accurately estimate possible savings / or no savings when switching a supplier. For those with low electricity bills, price transparency has no bearing on the intention to switch supplier. The intrinsically complicated energy bills attract attention and inspire the analysis of those paying larger amounts of money.
Interestingly, contrary to the authors’ expectations, for people paying low bills, the loyalty program (assessment of its attractiveness and participation in it) may increase the intention to change the energy supplier. Such conclusion is consistent with the results presented by [21], who showed that loyalty programs do not universally prevent switching behaviour. Additionally, as [96] indicated, negative consequences of loyalty programs such as customer’s expectation about greater discounts and extra privileges. Offered benefits/incentives in the loyalty program make such people aware of the possibility to take advantage of an even better offer, which prompts them to seek incentives from other energy suppliers. The authors of [97] noted that in case of hierarchical loyalty programs the effect of status demotion causes members to some negative experience feelings (frustration and social discomfort) can lead them to switch service providers. Thus, by analogy, in our study consumer in L group may perceive themselves as less important to the company and hence there is a tendency to switch to another supplier. A low bill amount does not ensure a sufficiently strong bond between the customer and the energy supplier, hence the tendency to switch current supplier. A low energy bill is a symptom of “a little investment” in the customer-energy supplier relationship made by the client. There are weak psychological bonds between individuals and energy supplier brand that creates a tendency to brand switching [98]. Thus, financial incentives under a loyalty program can generate the opposite effect of the intended—instead of consolidating the relationship with the client—they open their eyes to the possibility of obtaining even greater benefits, perhaps from another energy supplier.
Despite the apparent contradiction, a similar pattern appeared on the telecommunications services market, when mobile service providers offered their actual customers better terms during a contract extension. By such actions, service provider triggered the client’s decision-making process that often prompted the change of supplier. Similarly, clients with low energy bills participating in and gaining from the benefits of energy companies’ loyalty programs may realise the value of additional incentives and analyse them with alternative suppliers. This effect is absent in case of people paying high energy bills, when the relationship between loyalty program and tendency to change an energy supplier is insignificant.
6. Conclusions
Our proposed research model addressed the research gap by considering set of factors affecting the intention to switch and incumbent energy supplier. Provided practical implications can be useful for both the incumbent energy suppliers and the latecomers in many countries. The conclusions may be particularly useful for EU member states that have a similar structure of the energy market in terms of energy suppliers (incumbent vs. latecomers). For energy suppliers in general, the practical implications could be as follows. First, the strong position of incumbent suppliers in the markets with low consumers’ propensity to switch is due to their exceptionally strong image on these markets. Expenditure on its maintenance is the strongest barrier for new suppliers The method of obtaining energy for most recipients (clients) is not even important. New energy suppliers entering the market must perform several actions to enter the business. Firstly, they should look for differentiators and position themselves as companies offering transparent terms of cooperation (energy bills). Additionally, they should look for groups of recipients with specific needs (also those looking for green energy). New suppliers should also put emphasis on loyalty programs, which in their case may lead to the choice of a new supplier.
Although important issues emerged from this work, there are some limitations which should be taken into account as they suggest directions for further research. The first limitation concerns the inclusion of customers from one country in the survey. The subject of the study are customers from one company—a very well-known energy supplier in Poland. The second issue is connected with ignoring a potential impact of the country’s energy policy, which may affect the intention to change the supplier, e.g., by regulating prices for individual consumers. In the presented model we consciously resigned from considering the price-aspects in our analysis, due to some research indicating that price is not as significant determinant of consumer switching in deregulated electricity markets as non-price attributes [99]. It is advisable to investigate the pricing issues, however, with a different research approach. Many researchers indicate the usefulness of the game theory [100,101]. Finally, future research could also extend empirical testing to these dependencies (the impact of loyalty program and green image), which proved to be quite surprising and are not fully confirmed by the literature.
Author Contributions
Conceptualization, D.D. and M.L.; Methodology, M.L.; Software, M.L.; Validation, M.L. and I.B.; Formal Analysis M.L. and I.B.; Investigation, D.D.; Resources, I.B.; Data Curation, M.L.; Writing—Original Draft Preparation, M.L., I.B. and D.D.; Writing—Review and Editing, M.L. and I.B.; Visualization, M.L. and I.B.; Supervision, M.L.; Project Administration, D.D. and M.L. All authors have read and agreed to the published version of the manuscript.
Funding
This research received no external funding.
Institutional Review Board Statement
Not applicable.
Informed Consent Statement
Not applicable.
Data Availability Statement
Data will be available on request.
Conflicts of Interest
The authors declare no conflict of interest.
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Figures and Tables
Enlarge this image.Figure 1. Conceptual research model. Note: (+) positive impact, (−) negative impact.
Enlarge this image.Figure 2. Summary of the research results (total research group). Note: ** p < .001; * p < .05, ns—nonsignificant.
Enlarge this image.Figure 3. Summary of the research results (L group). Note: ** p < .001; * p < .05 ns—nonsignificant.
Enlarge this image.Figure 4. Summary of the research results (H group). Note: ** p < .001; * p < .05, ns—nonsignificant.
Percentage of people changing energy suppliers or tariffs among citizens of selected EU countries in 2013–2015.
| Selected EU Countries | Level of Provider Switching | Level of Provider Switching without Those Who Did It Because Moved Home | Level of Tariff Switching | Provider and Tariff Switching—Excluding Moved Home |
|---|---|---|---|---|
| Belgium BE | 23 | 20 | 18 | 38 |
| Bulgaria BG | 0.2 | 0.1 | 1.0 | 1.1 |
| Czech Republic | 8 | 7 | 12 | 19 |
| Denmark DK | 19 | 16 | 11 | 27 |
| Germany DE | 25 | 21 | 14 | 35 |
| Estonia EE | 7 | 7 | 13 | 20 |
| Croatia HR | 7 | 7 | 4 | 11 |
| Ireland IE | 28 | 25 | 11 | 36 |
| Greece EL | 1 | 1 | 15 | 16 |
| Spain ES | 7 | 7 | 13 | 20 |
| France FR | 2 | 2 | 5 | 7 |
| Italy IT | 13 | 12 | 12 | 24 |
| Lithuania LT | 1 | 1 | 6 | 7 |
| Hungary HU | 1 | 1 | 2 | 3 |
| Netherlands NL | 28 | 26 | 18 | 44 |
| Austria AT | 17 | 15 | 9 | 24 |
| Poland PL | 4 | 4 | 6 | 10 |
| Portugal PT | 14 | 13 | 19 | 32 |
| Romania RO | 0.2 | 0.2 | 6 | 6.2 |
| Slovenia SI | 18 | 17 | 11 | 28 |
| Slovakia SK | 7 | 7 | 11 | 18 |
| Finland FI | 20 | 17 | 5 | 22 |
| Sweden SE | 18 | 15 | 11 | 26 |
| United Kingdom | 28 | 23 | 28 | 51 |
| Norway | 19 | 15 | 11 | 26 |
| EU 28 | 14 | 12 | 13 | 25 |
Research sample characteristics.
| Characteristics | Number of Respondents | Percentage of Sample | |
|---|---|---|---|
| Gender | Female | 610 | 50.2 |
| Male | 606 | 49.8 | |
| Age (years) | 18–19 | 158 | 13 |
| 30–39 | 300 | 24.7 | |
| 40–49 | 360 | 29.6 | |
| 50–59 | 257 | 21.1 | |
| Over 60 | 141 | 11.6 | |
| Average monthly energy bills (euro) | Below 22.0 | 235 | 19.3 |
| 22.5–45.0 | 421 | 34.6 | |
| 45.5–67.0 | 310 | 25.5 | |
| 67.5–90.0 | 142 | 11.7 | |
| 90.5–112.0 | 52 | 4.3 | |
| Over 112.0 | 56 | 4.6 | |
| Household size (number of person) | 1 | 97 | 8 |
| 2 | 309 | 25.4 | |
| 3 | 335 | 27.5 | |
| 4 | 327 | 26.9 | |
| 5 or more | 148 | 12.2 | |
| Number of energy collection points | 1 | 988 | 81.3 |
| 2 | 191 | 15.7 | |
| 3 | 31 | 2.5 | |
| 4 or more | 6 | .5 | |
Selected measures of contracts’ reliability and validity.
| Dimension | Criterion of Fornell-Larcker | Cronbach’s Alfa | AVE | CR | |||||
|---|---|---|---|---|---|---|---|---|---|
| GI | CI | LP | PPT | INT | |||||
| Green image (GI) | GI | .854 | .85 | .73 | .85 | ||||
| Company image (CI) | CI | .59 | .755 | .78 | .57 | .84 | |||
| Loyalty programe (LM) | LP | .44 | .32 | .774 | .71 | .60 | .82 | ||
| Perceived price transparency (PPT) | PPT | .40 | .53 | .33 | .825 | .80 | .68 | .86 | |
| Intention to change energy supplier (INT) | INT | −.05 | −.24 | .08 | −.03 | .911 | .91 | .83 | .94 |
Note: The diagonal matrix contains square root values with AVE for individual constructs, while numbers beyond the diagonal are the values of the correlation coefficients between the respective constructs [87].
Table 4Verification of research hypotheses (total research group).
| Hypothesis. | p-Value | Estimates | Acceptance or Rejection | |
|---|---|---|---|---|
| H1 | Price transparency→ behavioural intention | .029 | .093 | Acceptance |
| H2 | Loyalty program → behavioural intention | .000 | .141 | Rejection |
| H3 | Company image → behavioural intention | .000 | −.365 | Acceptance |
| H4 | Green image → behavioural intention | .091 | 0.062 | Rejection |
Verification of research hypotheses (L group).
| Hypothesis | p-Value | Estimates | Acceptance or Rejection | |
|---|---|---|---|---|
| H1L | Price transparency→ behavioural intention | .224 | .069 | Rejection |
| H2L | Loyalty program → behavioural intention | .023 | .119 | Rejection |
| H3L | Company image → behavioural intention | .000 | −.370 | Acceptance |
| H4L | Green image → behavioural intention | .183 | .086 | Rejection |
Verification of research hypotheses (H group).
| Hypothesis | p-Value | Estimates | Acceptance or Rejection | |
|---|---|---|---|---|
| H1H | Price transparency→ behavioural intention | .047 | .129 | Acceptance |
| H2H | Loyalty program → behavioural intention | .073 | .104 | Rejection |
| H3H | Company image → behavioural intention | .000 | −.369 | Acceptance |
| H4H | Green image → behavioural intention | .316 | .073 | Rejection |
© 2021 by the authors.