Pro-environmental behavioural intentions of sharing economy users

Technological advancements, globalisation and economic growth have sparked an increase in consumption, which in turn creates numerous environmental problems including pollution, resource depletion and excess waste (Udall et al., 2020). These challenges have strengthened calls for more sustainable consumption (World Economic Forum, 2019), with consumer engagement in environmental sustainability being particularly emphasised. Specifically, consumers’ pro-environmental behaviour (PEB), defined as “behaviour that harms the environment as little as possible, or even benefits the environment” (Steg and Vlek, 2009, p. 309), has been recognised as a crucial component of pro-social consumption practices; thus, contributing positively to the environment and the society as a whole.

Although PEB has been widely examined in traditional consumption contexts, it remains under-explored within the rapidly growing sharing economy. Driven by digitalisation and peer-to-peer exchange, the sharing economy is expanding at an estimated rate of 32% and is valued at over US$113bn (Proficient Market Insights, 2022). Its popularity is attributed not only to the socio-economic benefits enjoyed by consumers such as social equity, economic affordability and income-generation opportunities (Cheng, 2016) but also to its potential to enhance sustainability through the more efficient use of under-used resources. Indeed, the sharing economy enables the redistribution of under-used resources, which may contribute to waste reduction and by extent environmental sustainability (De Las Heras et al., 2021).

Yet, the benefits of the sharing economy have been questioned (Buhalis et al., 2020; Cheng et al., 2020) as its rapid expansion was found to exacerbate socio-economic problems and contribute to environmental harm (Mosaad et al., 2023). Studies indicate that sharing economy sectors such as ride sharing and shared accommodation may contribute to social and environmental problems including overcrowding (Stergiou and Farmaki, 2020) and pollution through increased emissions (Wang et al., 2022). The sharing economy may have rebound effects where cost savings lead to increased consumption, requiring additional resources and energy use. In addition, the popularity of the sharing economy has led to increased professionalisation of the service offered, blurring the distinction between traditional and sharing economy marketplaces (Farmaki and Kaniadakis, 2020). As a result, concerns have been raised about the sharing economy’s true sustainability, even referring to it as “a nightmarish form of neoliberal capitalism” (Martin, 2016, p. 149).

Evidently, the topic of sustainability in the sharing economy needs further investigation. While sustainability issues have attracted attention in sharing economy research, the literature remains limited in at least three key aspects. Firstly, prior studies have primarily examined economic and social sustainability (Geissinger et al., 2019), leaving environmental dimensions under-explored. Recent studies provide evidence that sharing economy platforms’ green marketing may encourage consumer engagement in PEB (e.g. Huang et al., 2025). Given the emphasis placed on sustainability in the sharing economy post-COVID19 (Faraji et al., 2024), additional research is required on the pro-environmental behaviours of sharing economy users. Secondly, research on consumer engagement with sustainability in the sharing economy has largely focused on pre-consumption decisions such as booking green shared accommodation (e.g. Shin et al., 2023). While sustainability may act as a motivator for sharing economy consumption (Garrod et al., 2023), in sectors like accommodation consumption is largely based on pleasure-seeking (Dolnicar et al., 2019). As such, environmental concerns at the early stages of the decision-making process may not necessarily translate into PEB intentions at later stages reinforcing the need to examine post-booking PEB intentions. Thirdly, past studies on PEB relied on linear analytical models in their investigation which fail to capture the complex interactions shaping behaviour in a sustainability context (El Degheidy and Darrag, 2025; Olya et al., 2019). This study attempts to fill in these research gaps and gain insights into the PEB intentions of sharing economy users. In particular, the study attempts to answer the following research questions (RQ):

To address these RQs, we use complexity theory and apply fuzzy-set Qualitative Comparative Analysis (fsQCA) and Necessary Condition Analysis (NCA). Drawing from a shared accommodation context, and specifically from Airbnb guests, we identify multiple pathways that can lead to the PEB intentions of users. Our focus on Airbnb guests is not coincidental. Airbnb represents the largest shared accommodation platform worldwide (Kim and Yeon, 2025). Also, in recent years, the platform attempts to promote green practices among hosts with the aim of improving its sustainability impact (Farmaki et al., 2025). Hence, an examination of Airbnb guests’ PEB intentions is timely and valuable for understanding users’ PEB. We specifically focus on Airbnb guests staying in Athens (Greece), where a proliferation of Airbnbs is noticed in the past decade. The increase in Airbnb supply caused several socio-economic and environmental problems on the local community (Stergiou and Farmaki, 2020, 2025). As such, the Athenian context presents an interesting and meaningful study setting for the examination of Airbnb guests’ PEB intentions.

Overall, the study makes significant theoretical and practical contributions. Firstly, it expands research on environmental sustainability within the sharing economy by highlighting the role of individual, societal and contextual factors stimulating consumer engagement in pro-environmental practices. Individual consumer characteristics, societal influences as well as factors driven by service providers can promote environmental sustainability in the sharing economy context during the consumption stage. Secondly, by applying a non-linear analytical approach we better capture the complexity underpinning PEB intentions. By offering three possible pathways leading to the PEB intentions of sharing economy users, our study informs extant literature on environmental sustainability in relation to the sharing economy. Practically, our study informs sharing economy actors on promoting environmental sustainability through targeted interventions such as platform-driven informational material and marketing campaigns, consumer-focused service provider practices and appropriate resource provision to users.

The sharing economy, also known as the collaborative economy or access economy, refers to a digital phenomenon that was popularised following digital revolution. The term describes various platforms that connect users with owners of assets, enabling the sharing of goods and services for free or in exchange for a fee (Belk, 2014). The sharing economy thus represents not only a technological advancement but a socio-economic system (Khalek and Chakrabotry, 2023) that grants consumers temporary access to a good or service. As sharing involves under-used assets (Etter et al., 2019), arguments have been put forth that the sharing economy contributes to socio-economic sustainability (Rathnayake et al., 2024; Rojanakit et al., 2022). Indeed, when the sharing economy model emerged, it was described as a form of decentralised, equitable and sustainable economy that could reduce environmental impact by decreasing the use of resources (De Las Heras et al., 2021). According to Huang et al. (2023), the resource redistribution effected by the sharing of goods and services improves resource utilisation and reduces access costs, therefore offering economic, social and environmental benefits.

Despite these benefits, over the years questions have been raised over what can be labelled as “true sharing”. The marketisation of popular sharing economy companies such as Airbnb and Uber (Xiang et al., 2023) blurred the line between traditional and sharing marketplaces (Farmaki and Kaniadakis, 2020). According to Mosaad et al. (2023), the rapid growth and professionalisation of sharing economy platforms implies that the concept departed from its initial idea of peer-to-peer transactions. Given the indications of the sharing economy’s negative effects, recently researchers began to investigate the sustainability angle of the phenomenon (e.g. De Las Heras et al., 2021; Geissinger et al., 2019; Lecuyer et al., 2023). This stream of research, though, is limited in at least three aspects. Firstly, past research focused mostly on the economic sustainability of the sharing economy followed by a consideration of social sustainability, yet minimal attention was paid on environmental aspects (Lecuyer et al., 2023). This is surprising given evidence that sharing economy sectors have adverse effects on the environment. For instance, ride sharing was found to contribute to environmental pollution (Wang et al., 2022) whereas the growth of shared accommodation platforms was accused of creating excess waste and exacerbating pollution (Álvarez-Herranz and Macedo-Ruíz, 2021; Cheng et al., 2020).

A foray into pertinent literature reveals that there are several studies examining sustainable consumer behaviours in a sharing economy context (e.g. Huang et al., 2023; Toni et al., 2018; Wang et al., 2019). However, these focus on general conceptualizations of sustainable and/or socially responsible behaviours rather than environmentally friendly behaviours. Sustainable consumption behaviours are understood to encompass a range of pro-environmental, pro-social and equitable behaviours that aim at environmental, social and economic sustainability (Toni et al., 2018). The scope, thus, of sustainable consumption behaviour is broader (Wang et al., 2019) than environmentally oriented behaviours. Research informs us that consumers view sharing economy platforms as part of a wider social change that may improve sustainability and societal well-being (Faraji et al., 2024; Huang et al., 2025). However, they may also yield environmental benefits through users’ pro-environmental practices. Eliciting environmentally oriented behaviours among consumers is thus imperative to decrease environmental problems (van Valkengoed et al., 2022). Research is needed on the PEB of sharing economy users if their role in environmental sustainability is to be understood.

Second, a handful of studies exist that draw from a shared accommodation context and focus on intentions to book green properties (e.g. Agag, 2019; Shin and Kang, 2021; Shin et al., 2023). Yet, these studies place emphasis on the initial stages of the decision-making process. While sustainability concerns can act as a motivator for booking a green accommodation (Garrod et al., 2023), individuals may not necessarily act pro-environmentally whilst at the property. Prevalent sectors of the sharing economy (i.e. accommodation and transportation) are part of the tourism industry, which is characterised by hedonism. Dolnicar et al. (2019) concluded that the hedonism underlying the tourism activity often leads to unwillingness to enact PEB whilst on holiday. Indeed, research indicates that the possibility of behaving environmentally friendly in a holiday setting is less than that of undertaking such behaviours at home (Holmes et al., 2021; Xu et al., 2020) where utilitarian benefits such as cost-saving may be gained (Foroughi et al., 2022). Therefore, the presence of pro-environmental concerns or motives that shape willingness to book green shared accommodation does not guarantee PEB intentions once at the property.

This brings us to the third limitation. Relevant studies on PEB (e.g. Aviste and Niemiec, 2023; Hogg et al., 2024; Shanmugavel and Balakrishnan, 2023) used linear analytical approaches using theoretical frameworks such as Ajzen’s (1991) Theory of Planned Behaviour (TPB), Schwartz’s (1977) Norm Activation Model and Stern’s (2000) Value-Beliefs-Norm theory. These approaches, however, largely fail to capture the complexity underpinning PEB. As Olya et al. (2019) posited, linear analyses are insufficient in fully encapsulating the complexity driving consumer pro-environmental decision-making, which relies on dynamic processes that incorporate multiple antecedents with varying deterministic influences. This is because linear models (e.g. regression and structural equation modelling) assume symmetrical relationships between antecedents and behavioural intentions, implying that a given factor (such as attitudes) exerts a uniform and additive effect across all individuals. Yet, as PEB intentions are shaped by multiple interdependent conditions (Olya et al., 2019), different combinations of antecedents may predict an outcome reflecting equifinality (i.e. individuals may reach the same behavioural outcome through different contextual routes). Hence, alternative analytical models such as fsQCA may be more suitable in examining PEB intentions, as they can capture causal asymmetry. Causal asymmetry suggests that the presence of a factor may promote PEB in some configurations but its absence does not necessarily inhibit PEB in others. This allows for a more nuanced, realistic understanding of behaviour than linear models.

To this end, we use complexity theory and, more specifically, fsQCA to examine the PEB intentions of individuals staying in shared accommodation properties booked through the Airbnb platform. The study also uses NCA to identify the necessary factors that influence sharing economy users’ PEB intentions.

Based on chaos theory, complexity theory prescribes that certain phenomena cannot be adequately explained by cause-and-effect relationships that underpin linear analyses. Complex phenomena consist of multiple components connected by non-linear relationships, which may evolve and change over time. As small differences can lead to widely divergent outcomes, long-term behavioural patterns are inherently unpredictable (Kellert, 1994). Complexity theory assumes that non-linear examinations of the dynamic interaction of different factors are more suited to explain complex phenomena. Specifically, complexity theory proposes that different combinations of antecedents can lead to multiple pathways for the same outcome (Woodside, 2017). Accordingly, complex systems may be considered chaordic – emerging from dynamic interactions that generate new structures characterised by both unpredictability and pattern, or chaos and order (Olmedo, 2011).

Complexity theory has become a popular theoretical lens in the social sciences, particularly for explaining complex phenomena such as human behaviour (Olya and Al-Ansi, 2018). In business research, complexity theory enables the explanation of consumers’ behavioural aspects which include multiple complex characteristics (Pappas and Woodside, 2021). These multi-element configurational patterns can describe the derived behavioural complexity and identify the necessary conditions for the heterogeneous phenomena being examined. Given the above, complexity theory is well-suited for examining PEB intentions. As Olya et al. (2019) argued, traditional linear approaches are insufficient for capturing the configurational complexity that characterises pro-environmental decision-making. Because behavioural intentions constitute a critical stage in this process (Esfandiar et al., 2022), adopting a complexity-based perspective allows for a deeper understanding of how such intentions are shaped. Moreover, the sharing economy displays high levels of behavioural complexity. This is evident in consumers’ environmental behaviour (Dabbous et al., 2025), the criteria guiding the use of shared accommodation (De Canio et al., 2020), the decision-making underpinning purchasing intentions (Pappas, 2017) and ultimately their experience formulation (Pappas, 2019).

Drawing on complexity theory, this study proposes a research framework to examine the PEB intentions of sharing economy users. Specifically, we explore how different configurations of antecedents shape PEB intentions by combining key constructs from the widely applied TPB model (Yeh et al., 2021) with additional simple conditions, namely, green self-identity and host green attributes. This approach is consistent with recent studies that apply complexity theory and fsQCA to examine how TPB components (attitudes, subjective norms and perceived behavioural control), in combination with other variables, form multiple causal pathways towards PEB (e.g. Olya et al., 2019; Olya and Akhshik, 2019; Kumar et al., 2026).

Attitudes refer to how a person feels about a certain phenomenon and are based on a person’s beliefs that acting in a certain way is beneficial (Ajzen, 1985). If a person’s attitudes towards the environment are more positive, then he/she will be more likely to behave environmentally friendly. Subjective norms refer to the opinions of others close to an individual (Ajzen and Fishbein, 1980). In the context of PEBs, the greater the subjective norms the greater is the likelihood of an individual to act environmentally friendly. Perceived behavioural control suggests that a person may be limited in his/her capacity to behave accordingly due to lack of time, money or opportunities (Ajzen, 1991). Hence, the more time, resources or opportunities people have, the greater are the chances that they will behave environmentally friendly. In general, the higher the attitudes, subjective norms and perceived behavioural control of a person, the more likely that person will engage in a specific behaviour and vice versa.

In addition, we include in our examination the simple conditions of “green self-identity” and “host green attributes”. These antecedents are deemed suitable as the first reflects a personal guest characteristic whereas the latter refers to the green attributes and cues used by sharing economy service providers (i.e. hosts) to encourage PEB among guests. Their inclusion ensures that both personal factors and property green characteristics are considered in the examination of PEB. Hence, out study responds to suggestions that PEB is determined by consumer characteristics and green accommodation attributes and interventions (e.g. Dharmesti et al., 2020; Foroughi et al., 2022). In particular, green self-identity is argued to be an important influencer of behavioural intentions (Becerra et al., 2023) as it impacts the way people feel about acting pro-environmentally (Bradley et al., 2020). Green self-identity is defined as:

It positively impacts perceived value and encourages environmental-friendly behaviour (Confente et al., 2020). Research into green consumption reveals that self-identity is a key triggering factor eliciting PEB (Udall et al., 2020). Indeed, how a person sees himself/herself is critical to how this person will behave as self-identity reflects personal values, habits and motives driving a behaviour (Dermody et al., 2015).

Finally, we consider host green attributes which refer to green property features and green practices communicated to consumers by accommodation providers (i.e. recycling and water saving) with the aim of encouraging sharing economy users’ PEBs. These may include environmentally friendly features in properties such as energy-saving appliances, whereas interventions in the form of signage, verbal communication and/or marketing strategies are often used to warn consumers about environmental harm and in turn prompt environmentally friendly actions among consumers (Huang et al., 2025; Tkaczynski et al., 2020; Qin and Hsu, 2022). Extant literature reports that the communication of green attributes and informational interventions may exert a positive influence on guest pro-environmental attitudes (Trang et al., 2019) and actual PEB (Dharmesti et al., 2020; Mair and Bergin-Seers, 2010) as they reinforce behavioural intentions by reminding them to act environmentally friendly. Likewise, green attributes in shared accommodation establishments are likely to positively influence pro-environmental attitudes and intentions.

Based on complexity theory, this study examines the PEB intentions of sharing economy users through the proposed model depicted in Figure 1. In doing so, it assesses the complex causal relationships of attitudes, subjective norms, perceived behavioural control, green self-identity and host green attributes, also including the socio-demographics of age, gender and level of education.

Contrary to linear analyses where hypotheses are tested, non-linear analytical approaches use tenets. These are testable principles related to the identification of complex conditions (Papatheodorou and Pappas, 2017). To examine such conditions, it is necessary to understand the adequacy of complex configurations, whereby the same set of causal factors may lead to different outcomes (Ordanini et al., 2014). Extant literature includes several sustainability studies that have used tenets when investigating complexity-related phenomena. For instance, tenets have been applied to examine:

• the effect of information and communication technologies on sustainable development (Tekic and Tekic, 2024);

• the drivers of social change (Küçükergin et al., 2024);

• sustainable firm performance (Acquah, 2024); and

• pro-environmental buying behaviours (De Canio et al., 2024).

In addition, sustainability studies within the sharing economy have used tenets to explore issues such as peer-to-peer accommodation (De Canio et al., 2020) and the configurations that contribute to sustainable competitiveness (Dabbous et al., 2025).

In this study, the presence or absence of configurations of binary-state combinations is tested in relation to sharing economy users’ PEB intentions. The following tenets are accordingly formulated:

• _ T1: The impact of a single attribute on sharing economy users’ PEB intentions varies depending on its interaction with other attributes.

• _ T2: Complex pathways involving at least two antecedents can lead to consistently high PEB intention scores.

• _ T3: Complex configurations are able to affect sharing economy users’ PEB intentions.

• _ T4: Different combinations of simple conditions can positively or negatively influence sharing economy users’ PEB intentions.

• _ T5: High scores in PEB intentions are not always required for significant behavioural effects.

• _ T6: High PEB intention scores do not ensure uniform effectiveness of a particular strategy across all cases.

Our study draws from one of the most popular sharing economy platforms, Airbnb, and specifically from its guest community. In light of concerns over the negative impacts of its rapid expansion to the environment, the platform recently launched a sustainable hosting initiative as part of its commitment to environmental sustainability (Airbnb, 2023). Specifically, it issued recommendations to guide its host community into offering “greener” properties including waste management, and energy and water conservation (i.e. use energy-efficient appliances, install low-flow water fixtures). Hosts must also remind guests to unplug unused devices and place cues in their properties to encourage PEBs such as recycling. Evidently, guests’ PEB intentions emerge as critical in the successful application of pro-environmental practices in the sharing economy.

Data collection took place in Athens (Greece) between November 2024 and January 2025 among adults staying in Airbnb properties in the areas of Koukaki, Neos Kosmos and Plaka. These adjacent neighbourhoods serve as significant hubs for Airbnb activity in Athens (Stergiou and Farmaki, 2020), hosting a large share of the city’s nearly 12,000 listings (Souki, 2022). Therefore, their popularity among visitors ensured access to a diverse and representative sample of guests. To target guests, the help of Airbnb hosts was sought in various Greek host groups and forums on social media; an approach that has proved useful in prior relevant studies (e.g. Farmaki and Kladou, 2020). Specifically, after gaining permission from the group administrators, an open call was made to invite hosts with listings in the above-mentioned areas to support the study by placing self-assessed Greek and English-written questionnaires in their Airbnb establishments. Hosts were asked to sign a consent form prior to the data collection which outlined the purpose of the study, explaining that data was to remain confidential and used for academic purposes only. A total of 29 hosts took part in the study, some of which had multiple listings in the designated areas.

Questionnaires were provided to hosts in resealable envelopes and left in participating Airbnb properties before guests’ arrival, with instructions for completion and a request to leave them in a designated spot, as specified by the host, before checkout. The use of resealable envelopes safeguarded guest privacy and anonymity, as hosts could not view completed responses. To minimise the likelihood of duplicate responses, instructions emphasised that only one questionnaire should be completed per guest. The research team screened all data for possible duplicates, with no duplicate questionnaires identified. Notwithstanding, our chosen data collection method may inherently invite some degree of self-selection bias, potentially favouring participation from more engaged or environmentally conscious guests, as is common in voluntary participation methods including self-administered surveys (Dillman et al., 2014). However, it was chosen because it safeguards anonymity, ensures a reasonably high response rate and allows for the collection of a large sample within a short time frame (Pappas, 2017). To minimise interference with guests’ leisure time, the questionnaire was designed to be straightforward and easy to complete. In addition, to encourage broader participation and reduce the potential for self-selection bias, care was taken to design the survey with neutral questions to ensure accessibility and appeal to a broad range of guests (Fowler, 2013).

Each host initially received 25 questionnaires, but adjustments were made for some hosts who requested additional questionnaires or returned unused ones, resulting in a total of 764 distributed questionnaires. Of the total 764 questionnaires given, 420 were fully completed yielding a response rate of 55%, which aligns reasonably with the benchmark of 53.54% recommended by Ali et al. (2021) for surveys conducted in accommodation settings. Partially completed questionnaires were excluded from the data set using list-wise deletion, an approach widely regarded as the most appropriate for handling missing data (Raghunathan, 2020).

As suggested by Akis et al. (1996), an acceptable sample size should have a minimum of 95% level of confidence, with a subsequent maximum statistical error of 5%. When respondent perceptions are unknown, as in our case, the 50/50 conservative response format is recommended, assuming half the responses are positive and half negative (Akis et al., 1996), with the cumulative probability (Z) set to 1.96 (Sekaran and Bougie, 2020). The research has finally collected 420 fully completed questionnaires, yielding a confidence level of 95.22% and a statistical error of 4.78%.

The questionnaire comprised 36 Likert scale items, measured on a five-point scale from 1 (strongly disagree) to 5 (strongly agree). All items were derived from previous studies (Attitudes [Han, 2015; Olya et al., 2019]; Subjective norms [Farmaki et al., 2022; Nimri et al., 2020]; Perceived behavioural control [Ajzen, 1991]; Green self-identity [Becera et al., 2023; Bradley et al., 2020]; Host green attributes [Dharmesti et al., 2020; Trang et al., 2019]; Pro-environmental behavioural intentions [Dharmesti et al., 2020; Dolnicar and Leisch, 2008]), whilst three socio-demographics (Age [18–35; 36–50; Over 50]; Gender [Male; Female]; Level of education [Secondary; Diploma; Bachelor; Postgraduate]) were also included in the questionnaire. The items were adapted from validated instruments in prior research and minimally adjusted to fit the study context. For example, the attitude item “Acting environmentally friendly during my stay is beneficial” was adapted from Han (2015). All items were then assessed for reliability and validity using factor loadings, Cronbach’s alpha, Average Variance Extracted (AVE) and Composite Reliability (CR).

As indicated by Olya and Al-Ansi (2018), the most versatile method for examining configurational complexity is fsQCA, hence its adoption in the current research. Following the suggestions of Woodside and Zhang (2013), the negated sets’ evaluation (potential inclusion of an examined simple condition) was also used, indicating the absence of an antecedent with “∼”. The progression to non-linear analysis was justified by the presence of general asymmetry, as shown in the low correlations among antecedents (all values < 0.6; see Table 1) (Skarmeas et al., 2014). In such cases a non-parametric analysis should be used (Farmaki et al., 2024). Moreover, a complex pathway is acceptable when raw coverage ranges between 0.25 and 0.75, and consistency exceeds 0.74 (Skarmeas et al., 2014). The study examines the PEB intentions of sharing economy users by evaluating the antecedents of: attitudes, subjective norms, perceived behavioural control, green self-identity and host green attributes.

The necessity measurement is considered essential when using fsQCA (Ragin, 2008). This study uses NCA for this purpose, as it can identify the crucial factors for achieving the desired outcome (PEB intentions) and offers greater precision in necessity assessment than fsQCA alone (Dul, 2016).

The study used 37 cases for calibration, with membership for each causal condition defined between zero (non-membership at 5%) and one (full membership at 95%) (Ragin, 2008), and the crossover point set at 0.5. Thresholds were established at two, three and four to reflect the five-point Likert scale used (Pappas and Woodside, 2021). Calibration also included three socio-demographic variables: age, gender and level of education. Information loss was prevented through the calibration of the fuzzy sets (referring to the collection of elements, each with a gradual degree of membership to the set) (Wagemann et al., 2016). The robustness was set by the addition and extraction of 0.25 in the non and full membership, as well as the crossover point (Xie and Wang, 2020). The absence of significant differences in the final solution confirmed the robustness of the results. In fsQCA, the algorithms account for the degree of inconsistency and prioritise cases with strong set membership, as these are considered most relevant for identifying sufficient conditions. Therefore, the process uses scores of fuzzy membership to evaluate each case’s relevance. The respondents’ PEB intentions “f_pbi” were evaluated by using the fuzzy sets of gender “f_g”, age “f_ag”, education “f_e”, attitudes “f_a”, subjective norms “f_sn”, perceived behavioural control “f_pbc”, green self-identity “f_gs” and host green attributes “f_hga”.

Respondents’ socio-demographics are provided in Table 2, and the full items used (per simple condition) including the study’s descriptive statistics are illustrated in Table 3.

As previously noted, all items were derived from established studies. Hence, Confirmatory Factor Analysis was conducted (Table 4), with all items loading above the minimum acceptable threshold of 0.4 (Norman and Streiner, 2008). Reliability for all simple conditions was acceptable, as Cronbach’s alpha values exceeded 0.7 (Nunnally, 1978). Validity was also supported, with the AVE exceeding 0.5 (Kim, 2014) and remaining below the CR (Huang et al., 2013).

The complexity analysis (fsQCA) has generated four sufficient pathways associated with PEB intentions among sharing economy users (Table 5). Each of the sufficient complex configurations is based on the combination of the examined socio-demographics (age, gender and education) and simple conditions (attitudes, subjective norms, perceived behavioural control, green self-identity and host green attributes) whilst those that do not generate high outcome scores are labelled with the symbol “∼”. When an antecedent does not include high outcome scores, it means that it does not affect the generated solution. Thus, only those with high outcome scores are considered in the sufficient pathway.

The first configuration combines subjective norms and perceived behavioural control with the socio-demographic of age. This pathway has the highest consistency ^[1] among the generated solutions. The second configuration combines consumer attitudes and host green attributes with the socio-demographics of gender and education. This solution also generates the highest raw coverage ^[2]. The third configuration combines subjective norms and green self-identity with age, shaping PEB intentions. The fourth configuration involves perceived behavioural control, green self-identity and host green attributes, in combination with age and education. This is the only pathway that includes three antecedents and also produces the highest unique coverage ^[3], referring to outcome memberships not explained by other configurations.

NCA was used to address the first RQ and examine the necessity of the simple conditions. Although fsQCA can generate several sufficient pathways that can lead to the same outcome, they can only be considered acceptable when at least one of the simple conditions included in the solution is considered necessary. Therefore, the implementation of necessity analysis is vital to consider a generated complex configuration as necessary. Despite the fact that the necessity evaluation of the examined simple conditions can also be estimated through fsQCA, NCA is considered as a more precise method (Bol and Luppi, 2013), as it also provides size effects for the evaluated simple conditions (Dul, 2016). The size effects and the p-value for every examined antecedent are presented in Table 6, whilst Figure 2 illustrates the NCA plots. In line with Dul (2020), an antecedent is deemed necessary when it generates a size effect and its p-value is below 0.5. Subsequently, a complex configuration is considered acceptable when at least one of its conditions is identified as necessary (Mariani et al., 2024). With the exception of green self-identity, all other antecedents generated size effects. However, aside from green self-identity, subjective norms also failed to meet the required significance level (p < 0.5) to be classified as necessary. As a result, only three out of five simple conditions (attitudes, perceived behavioural control and host green attributes) are considered necessary. Thus, following Mariani et al. (2024), the third configuration – which includes the simple conditions of subjective norms and green self-identity – cannot be accepted, as an acceptable solution must include at least one necessary condition. Therefore, only the first, second and fourth sufficient configurations are considered acceptable.

Following Pappas (2021), the confirmation of the tenets should be based on the pathways derived from the asymmetric analysis (fsQCA), as this method is not related with the identification of necessary conditions (NCA). The fsQCA results confirm all six previously presented tenets. Firstly, each examined antecedent appears in at least one sufficient configuration, confirming the first tenet (T1). Secondly, all generated pathways include at least two simple conditions – for instance, the first and second solutions include two each, while the fourth includes three – thus confirming the second tenet (T2). Thirdly, all four pathways lead to the same outcome, the PEB intentions of sharing economy users, confirming the third tenet (T3). Moreover, the antecedents are combined differently in each complex configuration, yet they consistently influence sharing economy users’ PEB intentions, confirming the fourth tenet (T4). This is consistent with the principle of equifinality (Woodside, 2014), which posits that different pathways can lead to the same outcome. Because the outcome score for participants’ PEB intentions is not high, as shown in Table 5, the fifth tenet (T5) is confirmed. Finally, the coverage values are below one in all cases, indicating that no single configuration encompasses all examined cases (Olya and Altinay, 2016), thereby confirming the sixth tenet (T6).

As it was presented in the theoretical framework, the generated solutions from the fsQCA are in line with the three key concepts of the complexity theory:

1. provision of asymmetrical association;

2. generation of multiple alternatives leading to the same outcome; and

3. effect of simple conditions’ combination.

Concerning the second research question, the necessity analysis (NCA) indicated that out of the four sufficient complex configurations generated, only three can be finally accepted as the third one does not include a necessary antecedent. Specifically, the first acceptable pathway focuses on the behavioural aspects of the sharing economy users, particularly the role of subjective norms and perceived behavioural control. This pathway also indicates that age may influence the PEB intentions as it is included in the configuration. Previous studies have highlighted the effect of age on PEBs, suggesting that younger generations are more environmentally conscious (e.g. Abbas and Iftikhar, 2025; D’Arco et al., 2025). Equally, existing literature suggests that a person’s behavioural patterns depend on subjective norms and their perceived behavioural control (Ajzen, 1991), as both variables are recognised as key drivers of behaviour (Dermody et al., 2015). What distinguishes this configuration then is its emphasis on how these behavioural factors, in combination, can lead to an approach that may stimulate consumer engagement in pro-environmental practices.

The second acceptable configuration involves the simple conditions of attitudes and host green attributes, which are influenced by the socio-demographics of gender and education. As past studies show (e.g. Patel et al., 2017; Tan et al., 2022), gender and education can impact the PEB patterns of consumers, since women and highly educated individuals are considered more environmentally aware. Unlike the first configuration, which focuses on behaviour, this configuration suggests that host attributes and attitudes of consumers staying in shared accommodation establishments can affect their PEB intentions. Specifically, consumer attitudes (Foroughi et al., 2022) are further strengthened by host green attributes (Trang et al., 2019) and behaviour (Dharmesti et al., 2020), which act as a reinforcement for acting environmentally friendly during their stay. This configuration thus extends our understanding of how attitudinal factors and host cues interact to influence PEB intentions in the sharing economy setting, offering deeper insights into consumer engagement with environmental sustainability.

The last acceptable configuration includes the antecedents of perceived behavioural control, green self-identity and host green attributes and actually focuses on the pro-environmental Airbnb. In this case, the age and education of the consumers affect their propensity to adopt environmentally friendly practices, with prior work showing that younger and more highly educated individuals tend to demonstrate greater sensitivity to environmental issues (Casalegno et al., 2022; Suárez-Perales et al., 2021). The pro-environmental practices of an accommodation establishment such as effective resource management (e.g. energy conservation, water usage and recycling) (Han et al., 2010) can influence sharing economy users’ behaviours. Yet, this configuration underscores the role of green self-identity alongside perceived behavioural control and host attributes. This highlights how a consumer’s self-identification with pro-environmental values can significantly drive their intentions to engage in environmentally friendly behaviours when aligned with their sense of control and supported by the host’s green practices. This configuration underscores the importance of both individual and systemic factors in creating a space that we can refer to as a “pro-environmental Airbnb”, where sustainable practices are reinforced through the interaction of personal characteristics and environmental cues.

In sum, the accepted configurations illustrate the importance of various combinations of platform, host and user related factors in shaping sharing economy users’ PEB intentions. Given that these factors operate jointly, effective interventions should mirror this interdependence and be designed as complementary mechanisms rather than as isolated actions. At the platform level, interventions can create an overarching normative framework that legitimises and incentivises environmentally responsible behaviour in the sharing economy. These initiatives set the stage for host-level actions, where service providers can communicate their green attributes to guests. By creating direct situational cues, guests’ PEB intentions can be reinforced especially those of environmentally conscious individuals such as women, younger and more educated guests. At the user level, the proposed configurations highlight the importance of subjective norms, attitudes and perceived behavioural control. These underline the need for more personalised interventions that can strengthen perceived control and self-efficacy. Overall, our study illustrates how platform-driven normative framing, host-mediated situational reinforcement and user-level engagement can promote PEBs throughout the consumption experience through their interdependence.

Despite the rapid expansion of the sharing economy, users’ PEB intentions have not received adequate academic attention. This omission is surprising as the sharing economy’s growth has been associated with a number of environmental problems including increased waste, pollution and higher carbon footprint (e.g. Álvarez-Herranz and Macedo-Ruíz, 2021; Cheng et al., 2020). The PEB intentions of sharing economy users are thus imperative for the environmental sustainability of the sector. Accordingly, the aim of this study was to investigate sharing economy users’ PEB intentions. In so doing, we used complexity theory, with fsQCA and NCA as methods, which is appropriate in the investigation of complex phenomena such as human behaviour. Overall, three sufficient solutions can lead to sharing economy users’ PEB intentions: behavioural aspects, attitudes and attributes and pro-environmental Airbnb. The results indicate that sharing economy users’ PEB intentions can be encouraged by different combinations of antecedents (attitudes, subjective norms, perceived behavioural control, host green attributes, green self-identity and socio-demographics). Important theoretical and practical implications are thus derived for consumer engagement in environmental sustainability.

This study contributes to the emerging yet limited stream of research on the PEB intentions of sharing economy users, making the following contributions to extant literature. Firstly, our study adds to existing research on sustainability in the sharing economy which has primarily focused on economic and/or social sustainability aspects (e.g. De Las Heras et al., 2021; Geissinger et al., 2019). Despite evidence that sharing economy sectors have adverse effects on the environment (Cheng et al., 2020; Wang et al., 2022), the environmental sustainability of the sector remains under-researched. Pertinent studies focused mostly on broader sustainable behaviours (Toni et al., 2018; Wang et al., 2019) rather than specific environmentally oriented behaviours. PEB intentions play an important role in environmental sustainability as they signal consumer engagement in environmentally oriented actions. By focusing on the PEB intentions of sharing economy users, our study advances extant literature and informs researchers of how behavioural intentions of consumers may contribute to the environmental sustainability of the sharing economy. Environmental sustainability has become important in the sharing economy due to its rapid expansion creating environmental pressures on local communities (Stergiou and Farmaki, 2020). In particular, post-COVID-19, there have been calls for the sharing economy to align more closely with sustainability initiatives and environmentally responsible behaviours (Faraji et al., 2024).

Second, our study examines sharing economy users’ PEB intentions at later stages of decision-making therefore contributing to PEB literature. Contrary to past studies examining intentions to stay in green shared accommodation (e.g. Agag, 2019; Shin et al., 2023) or recognising sustainability concerns as a motive to shared accommodation selection (Garrod et al., 2023), our study investigates users’ PEB intentions while at the property. By shifting the focus from pre-consumption to post-booking and consumption stages, we better capture the behavioural intentions of consumers as the presence of environmental concerns or motives before a purchase does not necessarily translate into PEB actions during consumption. This distinction is important in the accommodation sector, a key area of growth for the sharing economy, where pleasure-seeking and hedonism characterise motivations and behaviours (Dolnicar et al., 2017). This is particularly true of environmentally friendly behaviours, as in the context of leisure activities individuals have been found to behave less environmentally while on vacation than they do at home (Holmes et al., 2021; Xu et al., 2020).

Thirdly, sharing economy users’ PEB intentions are examined in this study through complexity theory, operationalised via fsQCA and NCA. By adopting a non-linear analytical approach, this study allows untangling the complexity characterising PEB decision-making (Olya et al., 2019), which has been insufficiently addressed in prior linear studies. Specifically, our study illustrates how selected core components – social norms, perceived behavioural control, attitudes and socio-demographics – interact not only with one another but also with environmental cues and self-identity to shape PEB intentions. This interaction highlights the complex and non-linear nature of these intentions within the sharing economy context. Decisions related to environmental issues tend to be underpinned by dynamic processes involving multiple influencing factors that may yield various effects on an outcome (Lucas et al., 2008). In this respect, the use of complexity theory through fsQCA and NCA offers an appropriate lens for investigating these intricacies and allows the identification of multiple configurations (i.e. combinations of antecedents) that may encourage sustainability implementation within the sharing economy. By revealing how user-level cognitions interact with host and platform cues, our study offers a theoretical advancement towards a systems-based, interactional perspective on sustainability behaviour.

In practical terms, our study provides implications for sharing economy actors regarding environmental sustainability. For the first configuration (including subjective norms and perceived behavioural control), sharing economy platforms could develop informational materials and marketing campaigns at the community level that reinforce social norms around environmental responsibility and enhance consumers’ confidence in behaving pro-environmentally. For example, providing digital guides with practical tips on energy conservation and waste reduction can empower consumers with knowledge, enhancing their perceived behavioural control and encouraging environmentally friendly actions. In addition, launching campaigns featuring stories of service providers and consumers who engage in pro-environmental practices can foster a culture of environmental responsibility in the sharing economy. These stories may act as nudges, encouraging reflection and emulation of environmentally responsible behaviours by highlighting relevant past actions within the community (Majid et al., 2025). Feedback loops can also be created to improve guidelines and targeted initiatives. Such efforts could consider the effects of the socio-demographic factors of age and gender (Casalegno et al., 2022; Patel et al., 2017) and feature information targeted to women and/or younger consumers. Policymakers can support such efforts by promoting environmental education and public awareness campaigns that reinforce sustainable consumption practices. Accordingly, service providers in shared accommodation settings could implement simple measures such as clear recycling instructions and energy-efficient appliances, to promote stays with minimal environmental impact.

The second acceptable configuration, focusing on attitudes and host green attributes, underlines the importance of aligning service providers practices with consumer factors. For service providers in a shared accommodation setting, this suggests a need to clearly market their green attributes and practices. For instance, they can prominently feature eco-friendly certifications on their property listings to attract environmentally conscious consumers. Including detailed descriptions of pro-environmental features can further demonstrate their commitment to environmentally oriented actions. Platforms can facilitate this by including features such as sustainability badges or filters that allow consumers to identify listings based on specific environmental characteristics. These could include filters for energy-saving and recycling, which would cater to environmentally conscious consumers and motivate service providers to adopt and promote PEB.

Finally, perceived behavioural control, host green attributes and green self-identity jointly constitute the necessary conditions identified in the third configuration. This entails that sharing economy users must be provided with the necessary resources to act on their pro-environmental intentions whilst aligning service providers’ practices with guest expectations and needs. For instance, personalised messaging that affirms users’ pro-environmental values while enabling practical action through property-level cues and digital reminders may be used. Insights from prior research on demographic influences – such as the generally higher environmental awareness and stronger PEB intentions among younger and more educated consumers (Suárez-Perales et al., 2021; Tan et al., 2022) – may further help platforms refine such initiatives. Local policymakers can also encourage green practices among sharing economy service providers through incentives such as tax reductions for certified properties or setting sustainability standards, thereby motivating them to implement and communicate their green attributes and foster environmentally responsible consumer behaviour.

Overall, the findings highlight that effective promotion of PEB in the sharing economy requires coordinated action from platforms, service providers and policymakers. For Airbnb, this means strengthening sustainability-oriented design features such as clearer green labels, nudges embedded in the booking interface and host support tools that make environmental cues more visible and intuitive for guests. Hosts, in turn, can reinforce these efforts by integrating practical, easy-to-follow prompts within their properties and by communicating their green attributes more explicitly to guests. Policymakers can complement these initiatives by establishing or incentivising sustainability standards for short-term rentals, ensuring that environmentally responsible practices become a baseline expectation rather than a voluntary choice. Together, such aligned actions can create an ecosystem in which environmentally friendly behaviour becomes not only accessible but also normative and rewarding for sharing economy users.

This study, like any other study, is not free from limitations. Firstly, the use of fsQCA and NCA implies that further research is required to better explain their potential as analytical methods within the context of PEB, as these represent novel methods of analysis. Given that this study identifies multiple configurations leading to PEB intentions, future research could explore how these configurations evolve over time or under changing situational contexts (e.g. different trip purposes or lengths of stay). Longitudinal or experimental designs could help establish causal sequencing among various antecedents, offering deeper insights into the temporal dynamics of PEB formation. Secondly, our study focused on sharing economy users staying in shared accommodation in Greece. This focus raises questions about the generalisability of the results, as cultural and regional differences can play a significant role in shaping PEB intentions and behaviours. Research in other geographical contexts is necessary to validate and extend the findings across different cultural settings, including non-western countries. Cross-country replication could also be considered in examining PEB intentions of sharing economy users.

In addition, the effects of self-selection bias cannot be ruled out. While leaving questionnaires in the properties allowed for efficient data collection, it may have attracted more engaged individuals, potentially limiting the representativeness of the sample. Future research could examine specific characteristics of sharing economy users in relation to PEB, focusing on how socio-demographic factors (e.g. age, education and income) intersect with pro-environmental behaviours, potentially altering outcomes and influencing consumer engagement. As in a sharing economy context peer-to-peer exchanges between service providers and consumers may lead to the co-creation of value (Buhalis et al., 2020), the pro-environmental intentions and behaviours of service providers may be examined, compared and contrasted with those of consumers. Comparative studies can also be undertaken to evaluate sharing economy users’ intended and actual PEB to enable for a more holistic view of consumer engagement in environmental sustainability in the sharing economy. Finally, future research could extend the configurational approach beyond shared accommodation to other sharing economy sectors such as mobility or product-sharing platforms.