Why continue sharing: determinants of behavior in collaborative economy services

WHY CONTINUE SHARING: DETERMINANTS OF BEHAVIOR IN COLLABORATIVE ECONOMY SERVICES Arteaga Sánchez Rocío; Belda Ruiz María; Ros Gálvez Alejandro; Rosa García Alfonso Universidad Católica de Murcia RESUMEN La economía colaborativa está revolucionando la forma en que los consumidores utilizan los bienes y servicios. En nuestro estudio modificamos y ampliamos el modelo de confirmación de expectativas para determinar los factores motivacionales que impulsan la satisfacción y la intención de continuar usando los servicios de viajes colaborativos. Más importante aún, agregamos el valor social como un factor adicional. En este estudio fueron encuestados usuarios españoles experimentados de BlaBlaCar. La calidad del servicio, la utilidad percibida, la confianza y el valor social son determinantes de la satisfacción de los usuarios experimentados y, a través de ella, de la intención de continuar usando; mientras que no es el caso para el impacto ambiental ni para los beneficios económicos. Además, la confianza afecta directamente a la intención de continuar. Estos resultados tienen implicaciones gerenciales relevantes, mostrando que los usuarios de algunos servicios colaborativos están motivados por otros factores además de los económicos. Palabras clave: Economía colaborativa, viajes colaborativos, valor social, modelo de confirmación de expectativas. ABSTRACT The sharing economy is revolutionizing the way consumers use goods and services. In our study we modify and extend the expectation confirmation model to determine the motivational factors which drive the satisfaction and continue intention to use ridesharing services. Most importantly, we add social value as an additional factor to those previously studied in the literature. We apply our model in a survey among experienced Spanish users of BlaBlaCar. Service quality, perceived usefulness, trust and social value are determinants of satisfaction of experienced users and through it, of intention to continuance; while it is the case neither for environmental impact nor for economic benefits. Additionally, trust affects directly continuance. These results have relevant managerial implications, showing that users of some sharing services are motivated by other factors than purely economic. Keywords: Sharing economy, ridesharing, social value, expectation confirmation model. XXIX CONGRESO DE MARKETING AEMARK 2017 735 de 1617 ÍNDICE TRABAJOS PONENCIAS MM 1. Introduction Sharing economy is revolutionizing the way consumers use goods and services. Worldwide companies like Airbnb, Uber or BlaBlaCar are at the top of their respective sectors in several markets. Despite some problems encountered with the appearance of some collaborative platforms, experts predict that sharing economy will continue to grow. According to the consulting firm PwC (2015) sharing economy will have a global market with a potential of 335,000 million dollars (296,000 million euros) in 2025 and it is expected that 64% of adults use collaborative consumption platforms both to save and to obtain a return on their stuff. The same study shows that Spain is third in the European ranking, with more than 500 companies in that sector and begins to celebrate that Barcelona is one of the leading cities in this kind of economy. Within sharing economy, BlaBlaCar is among the favorite service of all users who want to make long car trips by sharing costs and saving money. BlaBlaCar was created in 2006, as a startup expected to build a structure based on connecting people who want to share a car on long trips in exchange for sharing expenditures (BlaBlaCar, 2016). It has between 2.5 and 3 million of users in Spain, figures that make up the success of their product in this country (BlaBlaCar, 2016). Some researchers point out that sharing economy is an appealing alternative for consumers due to its economic benefits, which was considered important after the global economic crisis (Bardhi and Eckhardt, 2012; Tussyadiah, 2015). However, Botsman and Rogers (2011) argue that collaborative consumption is driven by motivations that extend beyond cost-savings such as ecological and social. For some time, we have seen more and more how these exchange-based business models infiltrate national economies, but what are the reasons or motivations for consumers to initiate collaborative consumer schemes? and what are the motivational factors that cause users to engage in collaborative consumption on a repetitive basis? The fact of knowing the motivations of the collaborative consumption would help companies to take advantage of the resources that are in disuse changing the habits of consumption, all in favor of a greater efficiency and sustainability. In this context, despite a growing practical importance of knowing consumer motivations and satisfaction towards collaborative activities, there is a lack of quantitative studies which focus on explaining these motivational factors that affect consumer intentions to use. As Heinrichs and Grunenberg (2013) say, existing studies on sharing economy are theoretical, conceptual and normative. In order to address the identified research gap, the aim of this paper is to determine the motivational factors that drive the satisfaction and continue intention to use ridesharing services among Spanish experienced users of BlaBlaCar services in Spain. We also contribute to the literature by adding social value, in accordance with Chudzian (2015), as a new driver of satisfaction and continuance intention in this context. The rest of the paper is structured as follows. Section 2 presents the theoretical framework. Section 3 proposes the research model and hypotheses. Section 4 describes the research method. Section 5 includes data analysis. Finally, in section 6, discussion and conclusion are presented. 2. Theoretical framework Sharing economy has been attracting a great deal of attention in recent years. Platforms such as BlaBlaCar and Airbnb are experiencing an explosive growth, which, in turn, has led to regulatory and political battles. The Expectation Confirmation Theory (ECT) has dominated the literature focused on consumer satisfaction from its beginning in the early years of the seventies (Vroom, 1964; Porter and Lawler, 1968; Oliver, 1980). This theory conceives satisfaction as the result of a comparison between reality perceived by the individual and expectations. XXIX CONGRESO DE MARKETING AEMARK 2017 736 de 1617 ÍNDICE TRABAJOS Based on ECT, Bhattacherjee (2001) conceptualizes and tests a model that distinguish acceptance and continuance behavior in the information system context. In this model, the Expectation Confirmation Model of Information System Continuance (ECM), the viability of an information system depends on its continued use. Information system continuance intention is determined by user satisfaction and perceived usefulness. User satisfaction is determined by confirmation of expectations and perceived usefulness (Hadji and Degoulet, 2016). The ECM model and its variations have been applied to a variety of contexts. Oghuma et al. (2016) analyze the impact of perceived usability, perceived security, perceived service quality, and confirmation on user continuance intention to use the mobile instant messaging. Their results showed that perceived service quality and perceived usability significantly affect user satisfaction and continuance intention to use mobile instant messaging. Similarly, Hsu and Lin (2015) modify the ECM by incorporating app rating, free alternatives to paid apps and habit as belief-related constructs to predict user behavior. They find confirmation was positively related to perceived value and satisfaction. Although continuance intention has been widely studied in the literature, it has been scarcely studied in the sharing economy context. To our best knowledge Belk (2014), Hamari et al., (2016), Mohlmann (2015), Schiel (2015), Chudzian (2015) are the few recent studies explaining the importance and value of collaborative consumption in other contexts. Mohlmann (2015) develops and tests a framework on the determinants of choosing a sharing option with two quantitative studies. The results reveal the satisfaction and the likelihood of choosing a sharing option again to be predominantly explained by determinants serving users’ self-benefit. Utility, trust, cost savings, and familiarity were found to be essential, and also service quality and community belonging. Hamari et al. (2016) show that participation in collaborative consumption is motivated by many factors such as its sustainability, enjoyment of the activity as well as economic gains. Furthermore, the results suggest that in collaborative consumption an attitude behavior gap might exist; people perceive the activity positively and say good things about it, but this good attitude does not necessary translate into action. Additionally, Schiel (2015) clarifies the impact of different types of motivations onto attitude towards and participation in co-consumption models. Findings suggest that the majority of respondents have been in touch with alternative modes of use and consumption. Across sharing categories, participants are driven by a triad of economic, ecological and social motivations. At the same time, respondents with no sharing history differ significantly in demographic attributes and personal values. Finally, Chudzian (2015) points out the awareness and activity of young consumers in the area of collaborative activities and indicate what factors condition such attitudes and behaviors. Results show clearly that people who do not use this form of collaborative consumption consider economic aspects more important. Active users, on the other hand, value higher ecological, social and psychological benefits. 3. Research model and hypotheses Using the expectation-confirmation framework, we develop an adapted model of continuance intention to use BlaBlaCar services (see Figure 1). We propose that economic benefits, service quality, perceived usefulness, trust, environmental impact and social value have a positive impact on the variable satisfaction with BlaBlaCar which itself has a positive impact on the continue intention to use it. Economic Benefits Owyang (2013) and Tussyadiah (2015), suggest three drivers of collaborative consumption: societal, economic and technology. As a consequence of the financial crisis, consumers are more concerned about resources they use and how they spend their money (Chudzian, 2015; Gansky, 2010; Tussyadiah, 2015). A new way of consumption where resources are emerges as a result of these changes XXIX CONGRESO DE MARKETING AEMARK 2017 737 de 1617 ÍNDICE TRABAJOS (Botsman and Rogers, 2011). This economic motivation of sharing is by far the most widely identified (Barnes and Mattsson, 2016; Hamari et al., 2016; Mohlmann, 2015; Schiel, 2015). Moreover, Barnes and Mattsson (2016) point out in their Delphi study that the largest current drivers of collaborative consumption are economic, underpinned by economic problems and a need to economize. Therefore, we present the following hypotheses:  H1: Economic Benefits have a positive impact on the Satisfaction with BlaBlaCar service.  H2: Economic Benefits have a positive impact on the Continuance Intention to use BlaBlaCar service. Figure 1: Measurement model Service Quality Service quality can be defined as the discrepancy between customers’ expectations and perceptions of the service (Grönroos, 1984; Parasuraman et al., 1988). It can also be conceptualized as “the consumers’ overall impression of the relative inferiority/superiority of the organization and its services” (Bitner and Hubbert, 1994). There is growing support related to the positive impact of service quality in many research studies among sharing economy option (Mohlmann, 2015), continuance intention among mobile data services users (Boskye, 2015), information-exchange virtual communities (Zheng et al., 2013) and organizations (Gorla et al., 2010). In our context of sharing economy, users of BlaBlaCar services might be more likely to use the service again after having positive experience with the customer service. Therefore, we present the following hypotheses:  H3: Service Quality has a positive impact on the Satisfaction with BlaBlaCar service.  H4: Service Quality has a positive impact on the Continuance Intention to use BlaBlaCar service. Perceived Usefulness Perceived usefulness can be defined as “the degree to which an individual believes that using a particular system would enhance his/her job performance” (Davis, 1989, p. 320). In this context perceived usefulness refers to the degree to which BlaBlaCar users think that by using BlaBlaCar services their trip XXIX CONGRESO DE MARKETING AEMARK 2017 738 de 1617 ÍNDICE TRABAJOS could be easier and more efficient. Studies indicate that perceived usefulness is positively associated with continuance intention in the context of Web-based learning (Chiu and Wang, 2008), social networking sites (Yin et al., 2015) and mobile instant messaging (Oghuma et al., 2016). Therefore, we present the following hypotheses:  H5: Perceived Usefulness has a positive impact on the Satisfaction with BlaBlaCar service.  H6: Perceived Usefulness has a positive impact on the Continuance Intention to use BlaBlaCar service. Trust Trust is a subjective feeling that the trustee will behave in a certain way according to an implicit or explicit promise she/he makes (Ert et al., 2016). Trust has been regarded as one of the important factors in the context of information system usage (Pavlou, 2003). According to Mohlmann (2015) the most important determinants to explain satisfaction with a sharing option and the likelihood of choosing again the same sharing option are utility, trust, cost savings, and familiarity. Barnes and Mattsson (2016) point out that among the main inhibitors to collaborative consumption are “Establishing trust”. In this context, we believe that if users find BlaBlaCar services trustworthy, they may engage with these services. In addition, the relationship may continue if the user has a positive experience with the BlaBlaCar services. Therefore, we present the following hypotheses:  H7: Trust has a positive impact on the Satisfaction with BlaBlaCar service.  H8: Trust has a positive impact on the Continuance Intention to use BlaBlaCar service. Environmental Impact Ridesharing reduces the vehicle ownership (Efthymiou et al., 2013). Millard-Ball et al. (2006) found that many people canceled a car purchase or sold their cars after joining a car-sharing vehicle scheme. An increasing awareness of environmental pressure drives people to find ways to use resources more efficiently in order to have a more sustainable society (Gansky, 2010). Collaborative consumption is supposed to reduce the negative impacts on the environment since it reduces the development of new products and the consumption of raw materials (Botsman and Rogers, 2011; Chudzian, 2015; Tussyadiah, 2015). Therefore, we present the following hypotheses:  H9: Environmental Impact has a positive impact on the satisfaction with BlaBlaCar service.  H10: Environmental Impact has a positive impact on the Continuance Intention to use BlaBlaCar service. Social Value The social aspects of collaborative consumption are important to people. Chudzian (2015) discovers that meeting new people, helping other people and the care for a common natural environment are among the main advantages in collaborative consumption. Collaborative consumption activities strengthen social cohesion (Owyang et al., 2014; van de Glind, 2013). The attached social benefits come with sharing: participation brings joy, recognition and thus, self-confidence and satisfaction (Hamari et al., 2016; Schiel, 2015; Owyang et al., 2014). Surprisingly, Chudzian (2015) realized that individual factors seem to be less relevant, what is interesting enough to be a premise for further research in the area of individual features of collaborative consumption users. In this way we add this new variable to our model. We expect that social value will have a positive and significant influence on satisfaction with BlaBlaCar service and on the continuance intention to use BlaBlaCar service. Therefore, we present the following hypotheses: XXIX CONGRESO DE MARKETING AEMARK 2017 739 de 1617 ÍNDICE TRABAJOS  H11: Social Value has a positive impact on the Satisfaction with BlaBlaCar service.  H12: Social Value has a positive impact on the Continuance Intention to use BlaBlaCar service. Satisfaction Consumer satisfaction is essential to the longevity of any business and it is in fact one of the most researched topics in marketing (Moriuchi and Takahashi, 2016). Consumer satisfaction is defined as the overall evaluation of a consumer’s total purchasing and consumption experience with products or services over a period of time (Anderson et al., 2004). The relationship between user’s satisfaction and continuance intention is well supported by previous research (Hsiao et al., 2016; Kaewkitipong et al., 2016; Tan et al., 2015). Therefore, we present the following hypothesis:  H13: Satisfaction with BlaBlaCar service has a positive impact on the Continuance Intention to use it. Continuance Intention Bhattacherjee (2001) indicates a positive relation between continuance intentions to satisfaction. Literature indicates that satisfaction is essential to build and retain a loyal base of long-term consumers (Kumar et al., 2013). When developing this variable we use a combined measure to know the likeliness to recommend BlaBlaCar service to others. In doing so, we have extended continuance intention variable with loyalty concept. Therefore, in our study, continuance intention is defined based on his or her likelihood to continue using the service in the future and to recommend it to others (van Lierop and El- Geneidy, 2016). Literature suggests that a satisfied consumer may be motivated to be loyal to the other party of the relationship (i.e. Gracia et al., 2015). Finally, continuance intention is the dependent variable and the one we wish to measure, which captures the degree to which BlaBlaCar users will want to continue using BlaBlaCar service. 4. Research method 4.1. Sample In order to identify individuals interested in BlaBlaCar, we collected all the users that participated in Forocoches, the biggest Spanish internet forum, in discussion threads that included BlaBlaCar in the title. We obtained a total of 258 valid surveys of discussants that had experience as users of BlaBlaCar. Most of our respondents were males (93%), individuals not married (93%), with university level studies (65%) and between 25 and 34 years old (57.4%), in line with characteristics of participants of the forum we used. Among the respondents, 8.9% had less than 1 year of experience using BlaBlaCar, 31.8% between 1 and 2 years, 31.4% between 2-3 years and 27.9% during more than 3 years. 4.2. Instrument We developed a questionnaire to measure the motivational factors that could lead to BlaBlaCar users to be satisfied with the service provided and to take positive intentions to the company. The Economic Benefits (EB) construct measurements were borrowed from Bock et al. (2005) (i.e., “BlaBlaCar XXIX CONGRESO DE MARKETING AEMARK 2017 740 de 1617 ÍNDICE TRABAJOS allows me to save money”). The measurements for Service Quality (SQ) were adapted from Parasuraman et al. (1985, 1988b) and Seiders et al. (2007) (i.e., “BlaBlaCar mobile application is appealing to me”). The measurements for Perceived Usefulness (PU) were adapted from Davis (1993) and DeLone and McLean (2003) (i.e., “BlaBlaCar helps me to travel more efficiently”). The Trust (TR) construct measurements were borrowed from Bhattacherjee (2002) and Chai et al. (2012) (i.e., “BlaBlaCar users are truthful in dealing with others”). The measurements for Environmental Impact (EI) were obtained from Lamberton and Rose (2012), Hamari et al. (2016) and Moeller and Wittkowski (2010) (i.e., “BlaBlaCar helps to save natural resources”). We contribute to the literature in this study by introducing Social Value (SV) construct measurements (i.e., “BlaBlaCar allows me to meet future good friends”). The Satisfaction (SA) construct measurements were borrowed from DeLone and McLean (2003); Fornell et al. (1996) and Wu et al. (2010) (i.e., “Overall, I am satisfied with BlaBlaCar”). Finally, to measure Continuance intention we follow Bhattacherjee (2001), Johnson et al. (2006) and Vogel et al. (2008) (i.e., “It is likely that I use BlaBlaCar in the near future”). All scales are evaluated on a seven-point Likert-type scale, from (1) “Strongly disagree” to (7) “Strongly agree”. A pilot test was run with twenty respondents who had used BlaBlaCar service previously. The comments received were taken into account to modify some items and to guarantee the content validity of the instrument. In Appendix 1 we present the measurements used in our questionnaire. 5. Data Analysis We use Structural Equation Modeling (SEM) to examine the relationships among motivational factors, Satisfaction and Continuance Intention, and their measurement variables. We test our model through Partial Least Squares (PLS) technique because of its robustness in small samples and the weak assumptions required for the observed variables. PLS has been shown to have a number of advantages over other techniques (Chen and Lin, 2015) and is being increasingly used for analyzing SEM models. First, we assess the reliability and the validity of the measurement model, and second, we assess the structural model (Hulland, 1999). 5.1. Reliability and validity of the measurement model We perform a Confirmatory Factor Analysis (CFA) with SmartPLS 3.0 (Ringle et al., 2015) to check the reliability and validity of the latent variables in our model. PLS algorithm and bootstrapping procedure reveal that all the items have loadings higher than 0.707 (Hair et al., 2006), except items for Economic Benefits (EB2, EB4 and EB6), Perceived Usefulness (PU2 and PU3) and Trust (TR4). These items were dropped from the model. Composite reliability (CR) shows values higher than 0.700, confirming the internal consistency of the measurement model (Peterson, 1994). The reliability of the measurement model is confirmed. Construct validity is assessed by analyzing convergent and discriminant validity. Convergent validity is guaranteed, being the average variance extracted (AVE) and composite reliability (CR) higher than 0.500 and 0.700, respectively (Fornell and Larcker, 1981). Discriminant validity is assessed by the application of Heterotrait-Monotrait Ratio of Correlations (HTMT), an alternative approach that is recommended in the literature to be superior to the examination of cross-loadings and the Fornell-Larcker criterion (Henseler et al., 2015; Hair et al., 2017). Discriminant validity between two reflective constructs is guaranteed if the HTMT value is below 0.900, being fulfilled in the model for all the latent variables, except for Trust and Satisfaction. After removing some problematic items from both constructs (TR5 and XXIX CONGRESO DE MARKETING AEMARK 2017 741 de 1617 ÍNDICE TRABAJOS TR6 for Trust and SA1 for Satisfaction), discriminant validity was guaranteed. Bootstrapping procedure (5,000 samples) also confirmed that none of the confidence intervals includes the value 1. Consequently, discriminant validity of the latent variables in the structural model is established. 5.2. Results of structural model testing We check the structural model for collinearity issues by examining the VIF values of all sets of predictor constructs in the structural model. All VIF values are clearly below the recommended threshold of 3.3 (Kock and Lynn, 2012). Therefore, there is not collinearity among the predictor constructs in the structural model. Figure 2: Research model A bootstrapping procedure (5,000 samples) was used to assess the significance of path coefficients and to find the relative importance of the exogenous driver constructs for the endogenous constructs (Figure 2). We find that Trust (H7; β = 0.275; p < 0.01) is the most important key driver of Satisfaction, followed by Service Quality (H3; β = 0.224; p < 0.01), Social Value (H11; β = 0.206; p < 0.01), Perceived Usefulness (H5; β = 0.159; p < 0.01) and Environmental Impact (H9; β = 0.128; p < 0.05). However, Economic Benefits is not a key construct explaining Satisfaction (H1; β = 0.128; p > 0.10). With respect to Continuance Intention, relatively important key drivers are Satisfaction (H13; β = 0.560; p < 0.01), Perceived Usefulness (H5; β = 0.155; p < 0.01) and Economic Benefits (H2; β = 0.107; p < 0.10). However, neither Service Quality (H4; β = -0.086; t < 1.65), Trust (H8; β = -0.086; p > 0.10), Environmental Impact (H10; β = 0.071; p > 0.10) nor Social Value (H12; β = 0.079; p > 0.10) directly affect Continuance Intention (See Table 2). Table 2: Hypotheses testing. Direct, indirect and total effects. Hypotheses Path Direct Indirect Total Supported H1 EB → SA 0.055 0.055 No H2 EB → CI 0.107* 0.032 0.138** Yes XXIX CONGRESO DE MARKETING AEMARK 2017 742 de 1617 ÍNDICE TRABAJOS H3 SQ → SA 0.224*** 0.224*** Yes H4 SQ → CI -0.086 0.128*** 0.042 No H5 PU → SA 0.159*** 0.159*** Yes H6 PU → CI 0.155*** 0.091*** 0.246*** Yes H7 TR → SA 0.275*** 0.275*** Yes H8 TR → CI -0.002 0.158*** 0.156** No H9 EI → SA 0.128** 0.128** Yes H10 EI → CI 0.071 0.073** 0.144** No H11 SV → SA 0.206*** 0.206*** Yes H12 SV → CI 0.079 0.118*** 0.197*** No H13 SA → CI 0.573*** 0.573*** Yes Note: *** p-value < 0.01; ** p-value < 0.05; * p-value < 0.10. We study also the total effects of exogenous constructs on Continuance Intention via the mediating construct Satisfaction. We find empirical support for significant indirect effects of Service Quality (β = 0.128; p < 0.01), Perceived Usefulness (β = 0.091; p < 0.01), Trust (β = 0.158; p < 0.01), Environmental Impact (β = 0.073; p < 0.05) and Social Value (β = 0.118; p < 0.01). Conversely, Economic Benefits does not indirectly affect Continuance Intention (β = 0.032; p > 0.10). With regards to total effects, among the six exogenous driver constructs, Perceived Usefulness (β = 0.246; p < 0.01) has the strongest effect on Continuance Intention, followed by Social Value (β = 0.197; p < 0.01), Trust (β = 0.156; p < 0.10), Environmental Impact (β = 0.144; p < 0.10) and Economic Benefits (β = 0.138; p < 0.10). Service Quality (β = 0.042; p > 0.10), for its part, has no significant total effect. These results highlight the importance of Satisfaction when explaining Continuance Intention of the users of BlaBlaCar service. The coefficient of determination (R2) of Satisfaction is 71.6% and 63.4% for Continuance Intention, showing a moderate value (Hair et al., 2011). We examine the size effects (f2) for all combinations of endogenous constructs and corresponding predictors. Regarding to Satisfaction, Environmental Impact (f2 = 0.031), Perceived Usefulness (f2 = 0.044), Service Quality (f2 = 0.088), Social Value (f2 = 0.082) and Trust (f2 = 0.105) have small effects on it. Economic Benefits has no effect on Satisfaction. As to Continuance Intention, only Perceived Usefulness (f2 = 0.031) and Satisfaction (f2 = 0.255) have, respectively, small and larger effects on it. The other predictors have no effect on Continuance Intention. We also assess the predictive relevance of the structural model by examining the Stone-Geisser’s Q2 value (Geisser, 1974; Stone, 1974). The Q2 values for both endogenous construct are considerably above zero. Specifically, with a Q2 value of 0.542 for Satisfaction and 0.459 for Continuance Intention, the results provide clear support for the model’s predictive relevance. XXIX CONGRESO DE MARKETING AEMARK 2017 743 de 1617 ÍNDICE TRABAJOS