When access leads to ownership: Explaining ownership rebound in access-based services

While ownership entails the permanent acquisition of a product and full control over its use (Philip et al., 2015), access-based services (ABS) allow customers to enjoy product benefits without the risks and responsibilities associated with product ownership (Hazée et al., 2017). By enabling temporary access, ABS are commonly understood as substitutes for ownership (Fritze et al., 2020). However, emerging evidence from the mobility sector suggests a counterintuitive pattern: consumers who initially engage with ABS may ultimately transition to product ownership. This raises a fundamental question: why would satisfied ABS users return to the very risks and responsibilities that ABS are designed to eliminate?

This paradox is particularly visible in emerging mobility markets such as e-scooters, where rental services have significantly lowered adoption barriers (Christoforou et al., 2021; Nikiforiadis et al., 2021) and often serve as consumers’ first point of engagement with the category (Lehr et al., 2020). Yet, despite the convenience and flexibility associated with access-based consumption (Lamberton and Rose, 2012), repeated use does not always remain access-based. Instead, some users move from renting to owning, suggesting a reversal of the traditionally assumed consumption trajectory. Rather than replacing ownership, access may serve as an entry point to facilitate it.

Existing research on ABS has largely examined contexts in which product ownership is the primary mode of consumption, with access serving as a substitute (e.g. Morewedge et al., 2021; Nansubuga and Kowalkowski, 2021; Stough and Carter, 2023). Within this stream, ABS are typically theorized as mechanisms that reduce the need for ownership by alleviating its associated burdens (Moeller and Wittkowski, 2010). This creates an inconsistency: if ABS derive their value from reducing ownership burdens, continued positive use should diminish the desire to own. Yet in emerging contexts, this expectation does not always hold. Observed transitions from satisfied ABS use to ownership expose a contradiction that substitution-based ABS logic struggles to explain. This inconsistency points to a broader limitation in existing ABS research, which has largely conceptualized access and ownership as competing modes of consumption and has paid limited attention to how they may dynamically interact within the same consumption trajectory.

To address this tension, we conceptualize this observed phenomenon as what we term “ownership rebound” (c.f., Ackermann and Tunn, 2024; Munten et al., 2024), a process in which repeated engagement with access-based consumption increases consumers’ propensity to purchase and own the accessed product, such that access functions as a pathway to ownership rather than a substitute for it. This challenges the core servitization assumptions that value creation shifts from ownership to access (e.g. Kowalkowski et al., 2017; Vandermerwe and Erixon, 2023) by suggesting that access may instead serve as a precursor to ownership. Instead of viewing access and ownership as mutually exclusive consumption modes (Morewedge et al., 2021; Stough and Carter, 2023), this perspective highlights their dynamic interplay. It also raises important implications for service providers: if ABS can inadvertently stimulate ownership, they may undermine their own long-term demand.

Against this backdrop, this study investigates why and how ABS users move toward product ownership in contexts where ABS often constitute consumers’ initial point of entry. Specifically, we develop and test a conceptual model explaining how ownership rebound emerges from repeated access-based consumption. The model proposes that repeated ABS engagement, captured through usage frequency and service loyalty, fosters familiarity and value realization, which may strengthen positive attitudes toward ownership and subsequent purchase intentions. In doing so, the model provides a mechanism for explaining the observed transition from access to ownership. The model is tested using survey data from 430 e-scooter rental users in urban environments where ABS is widely adopted.

This study makes three key contributions. First, it theorizes ownership rebound as a process through which access-based consumption may facilitate, rather than substitute for, ownership adoption. Second, it advances emerging debates on ABS rebound effects (e.g. Ackermann and Tunn, 2024; Munten et al., 2024) by identifying consumer-level mechanisms through which ABS may reinforce, rather than reduce, material acquisition. Third, it extends prior research by examining a reversed consumption trajectory, where access precedes ownership rather than the other way around.

The remainder of this paper is structured as follows. First, we review relevant ABS and ownership literature. Next, we develop the conceptual model and hypotheses. We then present the methodology and results, before concluding with a discussion of theoretical and managerial implications.

Access-based services have gained increasing scholarly attention over the past decade, reflecting a growing interest in alternative consumption models that challenge traditional notions of product ownership. This surge in research has led to deeper insights into consumer behavior dynamics, the psychological and functional trade-offs between access and ownership, and the implications for service providers.

Early work in ABS research was instrumental in defining the conceptual boundaries of the field. Scholars such as Bardhi and Eckhardt (2012) and Lamberton and Rose (2012) differentiated ABS from other non-ownership models, including collaborative consumption and peer-to-peer sharing. Their contributions clarified the specific characteristics of ABS, such as market mediation, anonymity, and absence of product identification, which set it apart from more socially embedded sharing practices.

While these conceptualizations have been instrumental in distinguishing ABS from ownership-based consumption, they largely adopt a static perspective in which access and ownership are treated as distinct and often opposing modes of consumption. This perspective offers limited insight into how consumers may transition between access and ownership over time, particularly in contexts where access precedes ownership.

To better understand consumer behavior in ABS contexts, researchers have drawn on broader consumption frameworks. A particularly influential one is the distinction between liquid and solid consumption (Atanasova and Eckhardt, 2021). Liquid consumption refers to transient, dematerialized, and access-based engagements with products or services. Consumers in this mode value situational utility, impermanence, and versatility, often viewing material possessions as burdens rather than extensions of identity.

Conversely, solid consumption is rooted in enduring product ownership, material attachment, and identity formation. Solid consumers tend to derive meaning from possessions, associating them with stability, social status, and emotional significance (Atanasova and Eckhardt, 2021; Bardhi and Eckhardt, 2017). While the distinction between liquid and solid consumption provides a useful lens for understanding differences between access and ownership, it reinforces a binary view in which consumers are positioned within one mode or the other. Such a perspective offers limited insight into how consumers may oscillate between these modes over time or how engagement in one mode may reshape preferences for the other.

Bardhi and Eckhardt (2012) further shaped the field by introducing a multidimensional framework to characterize ABS, highlighting features such as temporality, market mediation, and a lack of product identification. Complementing this, Lamberton and Rose (2012) developed an early quantitative model that examined consumer decision-making in access scenarios, identifying cost efficiency and flexibility as key drivers. Subsequent studies have built on this framework: for example, Akbar (2019) found that perceived product scarcity risks significantly shape ABS adoption, while Oyedele and Simpson (2018) identified utility flexibility as a central motivator.

In addition to these structural perspectives, prior research has begun to examine the role of psychological ownership in access-based contexts. Psychological ownership refers to the feeling that a product is “mine,” even in the absence of legal ownership (Kleinaltenkamp et al., 2018; Morewedge et al., 2021). Within ABS, such feelings may emerge through repeated interaction, control over usage, and growing familiarity with the product. On the one hand, this research suggests that access can partially substitute for ownership by allowing consumers to experience some of the psychological benefits of possession without actual ownership (Fritze et al., 2020). On the other hand, it also raises an unresolved question: if access can generate ownership-like feelings, repeated use may not only substitute for ownership but may also reinforce attachment to the product itself. This possibility remains underexplored, particularly in contexts where access precedes ownership, and consumers accumulate experience over time.

Collectively, these studies form the theoretical backbone for understanding how ABS challenges traditional consumption norms and reshapes customer-service relationships. However, while prior research has been instrumental in defining the conceptual boundaries of ABS, this literature has largely treated access and ownership as opposing consumption modes and has paid limited attention to how they may dynamically interact within the same consumption trajectory.

Research has consistently identified a set of motivations that drive consumers toward ABS. A common theme is the desire to avoid the burdens of ownership. Moeller and Wittkowski (2010) highlighted how consumers perceive product ownership as onerous due to factors like underutilization, maintenance responsibilities, and upfront costs. Other scholars have examined the risk-avoidance aspect of ABS. For instance, Lawson et al. (2021) and Schaefers et al. (2016a) emphasized that customers often adopt ABS to mitigate performance risks and financial risks. Access thus enables consumers to “test before they buy” or avoid ownership altogether. Another key motivator is cost efficiency, especially in the short term (Akbar, 2019; Milanova and Maas, 2017; Möhlmann, 2015). These studies portray ABS adoption as a rational, utility-driven choice.

However, these explanations predominantly capture motivations at the point of initial adoption and portray access-based consumption as a static, utility-driven decision. In practice, consumer motivations evolve over time (Grönroos, 2008; Sheth et al., 1991). As usage becomes more frequent, initial drivers such as cost savings or risk avoidance may give way to considerations related to convenience, control, and cumulative value. Repeated interaction with a product through access can reduce uncertainty, increase familiarity, and enhance perceived utility (Alba and Hutchinson, 1987). In such cases, access may shift from an alternative to ownership to a pathway to ownership (Bardhi and Eckhardt, 2017; Lehr et al., 2020), particularly when continued use highlights the limitations of temporary access. This temporal evolution of motivations remains insufficiently theorized in the ABS literature.

Despite the benefits of ABS, several challenges may discourage continuous use. Hazée et al. (2017) conceptualize these as burdens of access, including complexity, reliability concerns, contamination, responsibility, compatibility, and image-related issues. These barriers have been shown to reduce adoption and continued usage of ABS (Hazée et al., 2019).

Complexity refers to the perceived difficulty in understanding how to use the service, such as navigating a car-sharing app for the first time. Reliability concerns arise when users question the product’s performance or the behavior of other users, as seen in cases where rented items are returned damaged or in poor condition. Contamination relates to discomfort with shared usage and is especially salient in ABS contexts where consumption entails intimate bodily proximity to the product or elevated hygiene concerns, for example, in clothing rentals or carsharing (Hazée et al., 2019). Contamination concerns tend to be less central in low-intimacy access contexts, such as e-scooters. Responsibility is another common barrier, as consumers may fear being held liable for damage, such as minor scratches to rented electronics. Compatibility issues emerge when ABS offerings do not align with a consumer’s lifestyle, such as a frequent skier preferring to own sports equipment rather than rent it. Lastly, image barriers stem from a perceived misalignment between the service’s values and an individual’s self-perception, as seen when consumers hesitate to engage with sharing platforms due to skepticism about their sustainability claims.

While prior research has primarily examined these burdens as inhibitors of ABS adoption, it does not fully examine how such burdens may actively motivate consumers to consider purchasing a product and transition toward ownership. Importantly, as usage intensity increases, these burdens may not only discourage continued use but may also reframe ownership as a more attractive alternative as consumers accumulate experience and compare the relative benefits of access and ownership over time (Alba and Hutchinson, 1987). This suggests that barriers to access can serve not only as inhibitors of adoption but also as drivers of ownership transitions, reinforcing the need for a more dynamic understanding of the interplay between access and ownership. While ABS offers financial and practical advantages, the ownership-access dichotomy is not absolute. Many consumers remain attached to material possessions (Moeller and Wittkowski, 2010). Prior studies (e.g. Lehr et al., 2020; Rosenberg et al., 2023) show that consumers often oscillate between access and ownership over time, depending on life stage, usage experience, and product type. This fluidity underscores the need for more dynamic theoretical models that account for shifting consumer preferences.

Building on the tension identified in the introduction, prior research has generated valuable insights into why consumers adopt or resist ABS. However, as illustrated in Table 1, several theoretical limitations remain. First, most prior research adopts a substitution logic, conceptualizing access and ownership as competing modes of consumption. Empirical studies in this stream primarily focus on the drivers of access adoption or of ownership reduction (e.g. Moeller and Wittkowski, 2010; Möhlmann, 2015; Schaefers et al., 2016a). Consequently, access-based consumption is predominantly theorized as an alternative to ownership rather than as part of an evolving and potentially reversible consumption trajectory.

Second, although psychological constructs have received increasing attention, these studies mainly examine how access differs from ownership and how it attenuates ownership-related outcomes (e.g. Fritze et al., 2020; Guo et al., 2024; Kleinaltenkamp et al., 2018; Liu et al., 2024b). However, these perspectives provide limited theoretical insight into whether and how repeated use of ABS may cultivate ownership-related attitudes or purchase intentions over time. Third, overall, empirical evidence on access–ownership coexistence (Khalek and Chakraborty, 2025; Lehr et al., 2020; Rosenberg et al., 2023) remains fragmented and lacks integration into a generalizable framework. Finally, Table 1 highlights a contextual concentration in ownership-dominant markets, with comparatively limited theorization of the interplay between access and ownership (e.g. Christoforou et al., 2021; Nikiforiadis et al., 2021).

Addressing this gap, the present study draws on psychological ownership and service loyalty to theorize how repeated use of ABS may foster ownership-related attitudes and purchase intentions rather than merely substitute for ownership. This process aligns with emerging discussions on rebound effects in ABS (Munten et al., 2024) and is conceptualized in this study as ownership rebound, whereby access-based consumption may reinforce, rather than reduce, material acquisition. By explaining how service loyalty and ownership-related evaluations jointly shape consumers’ transition from access to ownership, this study advances a more process-oriented and theoretically integrated understanding of consumer behavior in access-based markets. This synthesis informs the hypotheses developed in the following section.

Our framework builds on past studies that highlight the barriers to ABS (e.g. Hazée et al., 2017) and the argument that materialistic consumers still prefer product ownership to access (e.g. Moeller and Wittkowski, 2010). We anticipate that customers’ satisfaction with their experience of ABS and the consequent expression of loyalty intentions may not completely diminish their intentions to purchase the products they have rented. Despite loyalty, we suggest that this intention to purchase is influenced by the level of compatibility between ABS and customers’ lifestyles, the frequency of product usage, and customers’ attitudes toward the advantages of product ownership. We briefly explain each construct in the following subsection and formulate the hypotheses of our conceptual model (see Figure 1).

Service loyalty refers to customers’ favorable attitudes toward a service, often demonstrated through their intentions to repeatedly use or recommend it (Chen, 2012; De Ruyter et al., 1998). In line with prior research, we conceptualize service loyalty as comprising both attitudinal and behavioral components, such as positive word-of-mouth and repeat patronage (Bloemer et al., 1999). A key antecedent of service loyalty is customer satisfaction, defined as an overall evaluation of a customer’s experiences with a service provider (Han et al., 2008; Kamath et al., 2020). Consistent with previous service research that has demonstrated a direct positive link between satisfaction and loyalty, where higher satisfaction enhances customer retention and increases service usage (e.g. Bloemer et al., 1999; Bolton, 1998; Caruana, 2002; Choi and Kim, 2020), we include satisfaction as a control variable in our model to isolate the unique effects of loyalty.

While satisfaction and loyalty are often linked to continued service engagement, their connection with product purchase intention is less clear in the context of ABS. The use of ABS is frequently influenced by situational factors such as availability, proximity, and convenience rather than solely by stable, favorable attitudes (Trujillo-Torres et al., 2024). Nevertheless, we predict that attitudinal loyalty may still increase the likelihood that customers select an ABS when situational conditions permit use. This is because loyal customers are more inclined to trust the provider, view the service as a preferred option, and remain open to repeated engagement. Therefore, while attitudinal loyalty may not fully determine usage behavior, it is expected to positively influence usage frequency.

While satisfaction and loyalty are typically linked to continued service use, their relationship with product purchase intention remains ambiguous in the context of ABS. Prior research indicates that repeated exposure to products through ABS may enhance the likelihood of eventual product purchase, particularly after a period of trial-based consumption (Lehr et al., 2020). This challenges the conventional belief that loyalty to an accessed service diminishes the appeal of product ownership. Although ABS has gained popularity by providing flexible, low-commitment alternatives to ownership, many consumers still prioritize product ownership for reasons such as long-term convenience, control, and symbolic status (Bardhi and Eckhardt, 2012; Moeller and Wittkowski, 2010). Therefore, while attitudinal loyalty to ABS may suggest satisfaction and repeated usage, it does not necessarily exclude the development of product purchase intentions. Rather, we expect that loyalty to the service and a desire for product ownership can coexist—especially when ABS acts as a conduit for product familiarity and trust.

In statistical terms, frequency denotes the number of times a particular event occurs within a defined period (Carr et al., 2018). Applied to the context of ABS, usage frequency reflects how often customers engage with such services—for instance, renting an e-scooter or car via a shared mobility platform. While situational factors such as convenience, availability, and context are important drivers of sharing economy services (e.g. Guyader et al., 2023; Hamari et al., 2016; Lang et al., 2022; Moeller and Wittkowski, 2010), attitudinal loyalty captures a deeper predisposition to engage with a service, reflecting prior positive experiences and post-consumption affect (e.g. Choi and Kim, 2020). Loyal users are therefore more likely to overcome situational frictions and consistently access ABS, even when minor inconveniences occur. Repeated engagement by loyal users should translate into habitual use and stronger adoption of ABS.

However, while frequent use often signals strong engagement or loyalty, it may also expose customers to some of the less favorable aspects of ABS, including service responsibility, contamination, and inconvenience (Hazée et al., 2017). Over time, these accumulated burdens may erode the perceived advantages of access—such as convenience and flexibility—and lead users to reassess the value proposition of ABS. For example, frequent users may become increasingly frustrated with the need to locate, unlock, and return rental items, or with concerns over hygiene and damage caused by previous users.

At the same time, frequent use increases product familiarity and allows customers to better evaluate the fit between the product and their needs. Through repeated use and increased familiarity, consumers gain experiential knowledge that reduces uncertainty, increases perceived fit, and enhances confidence in product assessment. As a result, loyal ABS users may increasingly perceive ownership as a more attractive alternative to continued access, particularly when cumulative access costs or service limitations become salient (Lehr et al., 2020). Moreover, attitudinal loyalty may influence purchase intention both directly (i.e. loyal users are more likely to trust the provider and its offerings, which increases their willingness to purchase) and indirectly (i.e. frequent access to the service allows users to trial the product in real-world contexts, providing evaluative feedback that further supports ownership decisions). To capture these relationships, we develop the following hypotheses:

Taken together, these arguments suggest that usage frequency represents a key behavioural mechanism linking attitudinal loyalty to product purchase intention, building on H1a–H1c. While loyalty may increase customers’ openness to ownership, this effect is likely to occur indirectly through increased engagement with the service, which, in turn, generates experiential learning, increases familiarity, and enhances evaluative clarity regarding the accessed product, thereby facilitating the translation of positive attitudes into ownership decisions (Alba and Hutchinson, 1987; Hoeffler, 2003).

Increased usage frequency may explain the inclination to purchase the accessed product as a means of overcoming the burdens of access (Lehr et al., 2020). In addition, repeated users can better assess product fit and overcome situational barriers, including minor inconveniences or service limitations, further increasing the likelihood of purchase. Therefore, we expect usage frequency to mediate the relationship between attitudinal loyalty and product purchase intention, resulting in a positive indirect effect. Thus:

Compatibility is the degree to which an innovation is perceived to align with the sociocultural values, beliefs, previous experiences, and individual needs of its potential users (Chen, 2013). In the context of ABS, service compatibility specifically refers to how well an ABS offering fits into the everyday lives, routines, habits, and broader lifestyle preferences of its target customers (Hazée et al., 2017; Karahanna et al., 2006). This alignment is crucial because innovation adoption is often contingent upon how naturally a new service integrates into a consumer’s existing patterns of behavior, social expectations, and experiential background (Karahanna et al., 2006).

Consumers typically build routines around services or products that they find useful, and that fit well with their daily schedules and expectations. As Kleijnen et al. (2009) suggest, such routines can gain personal significance over time, especially through repeated and long-term use. When an ABS offering seamlessly supports these habits by being conveniently located, flexible, affordable, or aligning with environmental or mobility values, it is more likely to be used frequently. In contrast, when a service feels misaligned with the user’s lifestyle or values, this perceived incompatibility can act as a barrier to access and discourage repeated usage (Hazée et al., 2017).

Given this, we argue that service compatibility plays a dual role: not only does it directly promote more frequent service usage, but it also moderates the impact of attitudinal loyalty on usage frequency. While attitudinal loyalty reflects a customer’s favorable disposition toward an ABS provider, this alone may not guarantee high levels of engagement unless the service is perceived as well-suited to the customer’s lifestyle. In other words, even loyal users might not frequently engage with an ABS if they view it as inconvenient or incompatible with their daily routines. However, when the service is highly compatible, loyal customers face fewer situational barriers, allowing their positive attitudes to translate more fully into repeated usage. Conversely, when compatibility is low, even loyal customers may engage less frequently. Thus, service compatibility strengthens the relationship between attitudinal loyalty and usage frequency by reducing contextual frictions. Service compatibility strengthens the effect of loyalty on usage because when the service aligns with consumers’ routines and lifestyle preferences, loyal attitudes are more readily converted into repeated engagement, whereas low compatibility creates contextual frictions that inhibit this translation. Based on these arguments, we propose the following hypotheses:

While ABS have gained popularity for their flexibility, affordability, and reduced commitment, many studies have emphasized their value primarily in contrast to the limitations of traditional product ownership. Specifically, prior research highlights how product ownership can involve various risks—such as the possibility of selecting the wrong product, the burden of long-term maintenance, or the inflexibility to adapt to changing needs (Moeller and Wittkowski, 2010; Schaefers et al., 2016a). This may deter some consumers and make ABS a more attractive, low-risk alternative. In this light, ABS is often positioned as a convenient solution for those hesitant about product ownership.

However, despite these advantages, emerging literature suggests that product ownership remains significant to many consumers. Studies by Bardhi and Eckhardt (2012) and Wittkowski et al. (2013) argue that ownership continues to be seen as providing distinct benefits, including long-term cost efficiency, personal control, and a sense of security. These benefits become particularly noticeable for consumers who frequently engage with ABS and begin to evaluate the long-term implications of continuous access versus ownership. For example, a consumer who rents an e-scooter daily may begin to perceive ownership as a more cost-effective and convenient solution over time, especially as the cumulative costs and occasional drawbacks of renting—such as availability issues or service restrictions—start to outweigh the flexibility that ABS initially offers.

Therefore, we posit that consumers who perceive ownership to offer greater value, control, or security than access are more likely to develop intentions to purchase the products they currently access through ABS. Furthermore, we argue that these attitudes can moderate the relationship between ABS usage frequency and purchase intention. Attitude toward ownership amplifies the effect of usage frequency on purchase intention because users who already value ownership interpret repeated usage as both evidence of the desirability of ownership (cognitive justification) and as strengthening their personal attachment or identity alignment with owning the product, making the transition from access to purchase more likely. That is, while frequent usage alone may lead some users to consider product ownership, the strength of this effect is likely to be enhanced among those who already value ownership as a preferable mode of consumption. For these consumers, repeated access may increase attachment and provide cognitive justification for transitioning to ownership. Based on this reasoning, we propose the following hypotheses:

This section outlines the data collection process, sample characteristics, measurement instruments, and analytical procedures used to validate the research model and test the hypotheses.

To test the hypotheses, we conducted a survey targeting customers of e-scooter rental services. The survey was conducted according to General Data Protection Regulation (GDPR) guidelines. The survey attracted 449 individuals with e-scooter rental experience, intercepted on two of the largest university campuses in Sweden. Students make up most e-scooter rental customers in many cities (Nikiforiadis et al., 2021), and past studies have shown that they are key customers for ABS in general (e.g. Hazée et al., 2019; Oyedele and Simpson, 2018). From the total sample of 449, we excluded 18 respondents who already owned an e-scooter (as measures of perceived ownership advantage and purchase intentions were irrelevant for them). We also identified one instance of duplicate participation; hence, 19 responses were excluded, resulting in a final sample size of 430 participants. All survey questions were mandatory; hence, there are no missing values to report.

The theoretical constructs in our conceptual model were measured with Likert scales from 1 to 7 (see Appendix A) based on previous research (Bloemer et al., 1999; Chen et al., 2021; Fornell et al., 1996; Gustafsson et al., 2005; Hawlitschek et al., 2016; Hazée et al., 2017; Schaefers et al., 2016b). ABS usage frequency was measured on an ordinal scale: daily (n = 7, 1.6%), weekly (n = 26, 6%), monthly (n = 131, 30.5%), and yearly (n = 266, 61.9%). In addition, the age (average age was 22.4 years), gender, income, and the ABS access mode (e.g. monthly subscription with unlimited rides) of each participant were also collected. The survey was pretested by three marketing scholars with expertise in the access economy context and quantitative research to control for the face validity of the constructs. Overall, the feedback was positive with only minor concerns raised regarding the survey length and the use of forced responses. The sample characteristics are presented in Table 2.

We employed partial least squares structural equation modeling (PLS-SEM) to analyze the data and test our hypotheses. This method has been used in service management research for over a decade due to its suitability for complex models, prediction-oriented research, and exploratory studies. We selected PLS-SEM for two key reasons. First, our model includes constructs with fewer than three indicators, and unlike covariance-based SEM, PLS-SEM can estimate such models without identification issues (Hair et al., 2022). Second, PLS-SEM prioritizes explaining variance in the dependent variable, making it particularly appropriate for predicting purchase intentions among ABS customers (Hair et al., 2022). In our model, the maximum number of structural paths directed at a latent variable is six (including control variables), so the minimum sample size recommendation assuming a significance level of α = 1% and a minimum R² ≥ 0.25 is 66 (Hair et al., 2022). This makes our sample of 430 large enough to employ PLS-SEM.

The survey data were analyzed following the latest guidelines (Hair et al., 2022; Matthews et al., 2016; Shmueli et al., 2019) using SmartPLS v.4 (Ringle et al., 2022). Our results are presented in three steps. First, we report on our assessment of the reliability and validity of the measurement model with reflective indicators. We also performed robustness checks using the Gaussian copula approach and the finite mixture (FIMIX) PLS procedure. Second, we present the PLS-SEM estimates for the structural model’s relationships (i.e. hypotheses). The PLS algorithm (with a maximum of 300 iterations) was used to estimate the structural model’s path coefficients (β) and their effect sizes (f² values), and the dependent variables’ coefficients of determination (R² values). The PLS prediction procedure (with ten folds, ten repetitions, and pairwise deletion) was used to assess the out-of-sample predictive power of the model (i.e. its accuracy in predicting the outcome values), reported with Q²_predict values. The bootstrapping procedure (with 10,000 samples, pairwise deletion algorithm, bias-corrected and accelerated confidence interval estimation method, and two-tailed testing at the 0.05 level) was used to estimate statistical significance. Third, to go beyond standard path coefficient analysis, we report the results of the Importance-Performance Map Analysis (IPMA) to provide managerial insights into which constructs have the most impact on a target variable and how well they perform.

The constructs were assessed using Cronbach’s α, the composite coefficient ρ_C, and the approximately exact (or consistent) coefficient ρ_A: all above the threshold of 0.7 (see Table 3), thus establishing reliability (internal consistency). All average variance extracted (AVE) were above the threshold of 0.5 (see Table 3) and all outer loadings exceeded the threshold of 0.7 (see Appendix A), thus establishing convergent validity. The latent variable correlation matrix shows no correlation >0.6 (see Table 4), indicating no potential redundancy, multicollinearity issues, or common method bias, and the heterotrait–monotrait (HTMT) ratios are all below 0.7, thus establishing discriminant validity (see Appendix B). The VIF values are all well below the threshold value of 3.3, and this confirms that there are no multicollinearity issues to consider (see Appendix C). We also performed Harman’s single-factor test, which revealed an extracted variance of just 29.6% (i.e. below the threshold value), suggesting the absence of common method variance (Podsakoff et al., 2003).

In addition, we performed the following robustness checks. To assess potential endogeneity, we used the Gaussian copula approach. Endogeneity could have arisen when explanatory variables are correlated with the error term, leading to biased estimates and thus, latent constructs (e.g. Product Purchase Intention) may be influenced by omitted variables or simultaneity effects. We added Gaussian copula terms to the regression model on the endogenous variable Product Purchase Intention, and we tested all other combinations of Gaussian copula terms included in our model; none were significant (see Appendix D). This suggests that endogeneity is not a major concern in our study. This strengthens the validity of our estimated path coefficients.

Furthermore, we used the FIMIX-PLS procedure for assessing unobserved heterogeneity that occurs when subgroups in the data respond differently to model relationships, which can distort overall results if not accounted for. FIMIX-PLS is an established method for identifying latent segments within a dataset (Matthews et al., 2016), which we applied using default settings for the stop criterion (10⁻⁷), the maximum number of iterations (5,000), and the number of repetitions (10). Given that our sample size is 430, we followed standard recommendations (Hair et al., 2022; Matthews et al., 2016) to determine the maximum number of segments (six) and ran FIMIX-PLS for one to six segments (see Appendix E). The fit indices suggested different segment solutions (e.g. AIC3, AIC4, and HQ suggested five segments, while CAIC, BIC, and MDL5 suggested three). Examining the relative segment sizes shows that none of these solutions (i.e. five or three segments) met the minimum threshold required for segment-specific PLS-SEM analysis of 66 observations per segment (i.e. 15.35%). Consequently, we conclude that unobserved heterogeneity is not an issue. Altogether, these assessments support the robustness of the results obtained from analyzing the entire data set.

Estimates of the model’s structural relationships are reported in Figure 2 and Table 5. First, the percentage of explained variance (R² values) can be considered weak for ABS Loyalty (0.405), ABS Usage Frequency (0.279), and Product Purchase Intention (0.304) — according to Hair et al. (2022)’s thresholds for interpretation. Second, the Q²_predict values are all above 0, which confirms that the model outperformed the naïve benchmark. As the prediction-error distribution for Product Purchase Intention (i.e. the endogenous construct) visually follows a symmetrical bell-curved shape, we examined whether the root mean squared error (RMSE) scores in the PLS-SEM were lower than the linear regression model (LM) RMSE scores (see Appendix F). All indicators scored lower on the RMSE in the PLS-SEM than in the LM; our model has high predictive power (Shmueli et al., 2019). Third, as recommended by Henseler et al. (2016), we also report a goodness-of-fit of our PLS model as follows: the SRMR for the saturated model was 0.058, while the SRMR for the estimated model was 0.071, both suggesting an adequate structural model fit (Henseler et al., 2016). Fourth, all paths are statistically significant, which provides support for all our hypotheses.

The direct effect of ABS Loyalty on ABS Usage Frequency is significant and positive (β = 0.177, t = 3.406, p = 0.001, CI: 0.071–0.277) and is considered medium (f² = 0.029); thus, H1a is supported. The direct effect of Usage Frequency on Product Purchase Intention is significant and positive (β = 0.107, t = 2.138, p = 0.033, CI: 0.008–0.203) and is considered small (f² = 0.013); therefore, H1b is supported. The direct effect of ABS Loyalty on Product Purchase Intention is significant and positive (β = 0.206, t = 4.282, p < 0.001, CI: 0.112–0.300) and is considered large (f² = 0.050); hence, H1c is supported. Regarding mediation analysis, we followed the procedure to analyze each specific indirect effect using PLS-SEM. We first examined the specific indirect effect of ABS Loyalty on Product Purchase Intention via Usage Frequency (β = 0.019, t = 2.020, p = 0.043, CI: 0.004–0.042), which is statistically significant; meanwhile, the direct effect (i.e. direct path) from ABS Loyalty to Purchase Intention is also statistically significant (H1c). We can conclude that there is a complementary (partial) mediation of Usage Frequency on the relationship between ABS Loyalty and Product Purchase Intention; therefore, H1d is supported.

Next, the direct effect of ABS Compatibility on ABS Usage Frequency is significant and positive (β = 0.380, t = 7.697, p < 0.001, CI: 0.281–0.475) and is considered small (f² = 0.132); hence, H2a is supported. The effect of ABS Loyalty is hypothesized (H2b) to be moderated by ABS Compatibility. This moderating effect was assessed by creating an interaction term using a two-stage approach. This method operationalizes the interaction term while controlling for the direct impact of the moderator on the endogenous construct (H2a) to avoid inflating the moderating effect. It generates single-indicator interaction terms by multiplying the scores of the endogenous variable (ABS Loyalty) and the moderating variable (ABS Compatibility), which reduces collinearity, minimizes estimation bias, and enhances the interpretation of the moderating effects (Becker et al., 2018). The interaction term was estimated simultaneously with the main effects within the full structural model. We found that the interaction term significantly and positively affects Usage Frequency (β = 0.140, t = 3,175, p = 0.002, CI: 0.050–0.224), and this effect is considered medium (f² = 0.027). This moderating effect indicates that attitudinal loyalty translates into more frequent e-scooter rental service usage, particularly when the service is perceived as compatible with consumers’ needs and lifestyles. This indicates that increasing ABS Compatibility by one standard deviation (SD) would enhance the relationship between ABS Loyalty and Usage Frequency from 0.178 to 0.318 (see Appendix G). In other words, ABS Compatibility acts as an enabling condition that activates the behavioral consequences of loyalty, such as at higher levels of ABS Compatibility, ABS Loyalty exerts a stronger effect on Usage Frequency; thus, H2b is supported.

The direct effect of A. toward Ownership on Product Purchase Intention is significant and positive (β = 0.453, t = 12.444, p < 0.001, CI: 0.377–0.519) and considered medium (f² = 0.287), hence, H3a is supported. In addition, the effect of Usage Frequency is also hypothesized (H3b) to be moderated by A. toward Ownership. Using the same analytical procedure as for H2b, we found that the interaction term significantly positively affects Product Purchase Intention (β = 0.083, t = 2.083, p = 0.037, CI: 0.003–0.159) and that this effect can be considered small (f² = 0.010). This moderating effect suggests that frequent e-scooter rental service users who hold favorable attitudes toward ownership are more likely to buy their own private e-scooter. Specifically, one SD higher A. toward Ownership increases the relationship between Usage Frequency and Purchase Intention from 0.107 to 0.190 (see Appendix H). In other words, usage experience promotes ownership intentions primarily when ownership is perceived as a desirable mode of consumption, since at higher levels of favorable A. toward Ownership, Usage Frequency has a stronger effect on Purchase Intention; hence, H3b is supported.

Finally, the total effects of ABS Loyalty (f² = 0.225), Usage Frequency (f² = 0.107), ABS Compatibility (f² = 0.041), and A. toward Ownership (f² = 0.453) on Purchase Intention are all positive, and considered medium, small, small, and large, respectively (see Appendix I).

Regarding control variables, the direct effect of Satisfaction on ABS Loyalty is significant and positive (β = 0.636, t = 21.278, p < 0.001, CI: 0.573–0.690) and considered large (f² = 0.681), in accordance with previous research (Bloemer et al., 1999; Bolton, 1998; Caruana, 2002; Möhlmann, 2015; Stough and Carter, 2023). Furthermore, Gender (β = −0.151, t = 3.544, p < 0.001, CI: 0.234 to −0.068) has a medium effect (f² = 0.031) on Purchase Intention, so that women expressed a lower purchase intention. Age (β = −0.032, t = 0.774, p = 0.439, CI: 0.113 – 0.048) and Income (β = 0.001, t = 0.019, p = 0.985, CI: 0.088 – 0.094) do not have a significant influence.

This study offers novel insights into the relationship between attitudinal loyalty toward ABS and product purchase intentions. Our results show that attitudes toward product ownership not only directly influence purchase intentions but also moderate the relationship between rental frequency and purchase intentions. Additionally, perceived compatibility positively moderates the relationship between service loyalty and rental frequency. Unlike prior research, which often takes the shift from ownership to access as its focal assumption, this study explores the co-existence of ABS and product ownership. Our results show that loyalty to an ABS does not necessarily diminish customers’ intentions to purchase the products they accessed when they perceive meaningful benefits in ownership. Indeed, even customers with strong loyalty to an ABS may still wish to purchase the products they rent. In the following section, we elaborate on the key theoretical and managerial implications of these results. We discuss how they contribute to our understanding of ABS, which are conceptually distinct from the peer-to-peer contexts of the sharing or gig economy phenomena (e.g. Breidbach and Brodie, 2017; Davidson et al., 2023; Lang et al., 2022). We argue that the choice between ABS and product ownership is influenced by a balance of motivations and barriers. While consumers seek to avoid the burdens of product ownership, they may sometimes resist ABS due to concerns about control, reliability, and social status.

In the following, we discuss this study’s contributions to ABS theory through the lens of ownership rebound. We argue that the relationship between ABS and product ownership is not adequately captured by a simple substitution logic, which assumes that access reduces ownership. Instead, our findings suggest a more dynamic process in which consumers iteratively evaluate access and ownership over time, giving rise to ownership rebound, where continued ABS can reinforce rather than diminish ownership intentions.

Rather than viewing co-existence as a stable outcome, our findings suggest that it reflects an underlying dynamic consumption trajectory, in which consumers move between access and ownership over time. This study advances the theoretical discourse on ABS by demonstrating that product purchase intentions can coexist with continued use of ABS, particularly in contexts where access serves as an initial point of market entry. This finding challenges the traditional binary view that positions access and ownership as mutually exclusive alternatives (Khalek and Chakraborty, 2022; Lawson et al., 2021), and more fundamentally, the assumption that access and ownership operate as substitutes rather than as interrelated stages within a consumption trajectory.

Prior research has often assumed that ABS substitutes product ownership and thereby reduces material consumption (Ackermann and Tunn, 2024). In contrast, our results suggest a complementary relationship between the two modes. This co-existence can be theoretically explained through the lens of liquid and solid consumption (Bardhi and Eckhardt, 2017). Liquid consumption emphasizes flexibility, temporality, and situational relevance, whereas solid consumption is associated with stability, control, and enduring attachment. Rather than adhering strictly to either ownership (solid consumption) or access (liquid consumption), consumers strategically navigate between the two forms based on situational relevance, frequency of use, and personal attitudes toward ownership (Bardhi and Eckhardt, 2017; Rosenberg et al., 2023). In such contexts, ABS enable liquid consumption by allowing consumers to use products flexibly without long-term commitment, while ownership remains attractive for products that are frequently used or emotionally significant (cf. Ehret and Wirtz, 2018).

Our findings suggest that repeated access-based usage may help clarify when solid consumption becomes preferable. Consumers may rely on ABS for occasional or situational needs while simultaneously developing intentions to purchase products that play a more central role in their daily routines (Bardhi and Eckhardt, 2017). This reinforces recent arguments that liquid and solid consumption are not mutually exclusive but interdependent (Rosenberg et al., 2023). In this sense, liquid consumption can even reinforce the value of solid consumption by highlighting situations in which stability, personalization, or long-term efficiency is desirable (Eckhardt and Bardhi, 2020). Thus, our study extends the liquid-versus-solid consumption theory by demonstrating how access-based use can serve as a mechanism through which consumers evaluate and selectively transition toward ownership. From this perspective, coexistence is not merely a static condition, but an observable manifestation of ownership rebound, whereby repeated access-based use enables consumers to evaluate and, in some cases, transition toward ownership.

By examining a market with access as the primary entry point, this study also has implications for research on servitization and deservitization by showing how ownership rebound at the consumer level can give rise to reverse transitions from access to ownership. While most prior research on ABS and servitization more broadly assumes a product ownership-to-access transition, as seen in sectors like automotive leasing or music streaming (e.g. Vandermerwe and Erixon, 2023) and in industrial settings (e.g. Cao et al., 2026; Liu et al., 2024a), our findings reveal evidence of a reverse dynamic in which consumers move from access back to ownership. Rather than treating deservitization as a primary phenomenon, our findings suggest it is a consequence of ownership rebound, in which prolonged engagement with access-based services increases the likelihood of transitioning to ownership. Interestingly, this “deservitization” trend (Kowalkowski et al., 2017; Valtakoski, 2017) does not reflect a rejection of services but rather emerges from prolonged engagement with them, highlighting how service use itself can generate conditions for ownership rebound.

A key mechanism underlying this ownership rebound process is the role of unintended product trials. Consistent with prior research (Lawson et al., 2016; Lehr et al., 2020), our results suggest that ABS unintentionally provide consumers with extended opportunities to test and evaluate products through repeated usage. Rather than functioning as discrete trial episodes, these experiences accumulate over time, fostering familiarity, confidence, and clarity in evaluation, allowing consumers to assess whether ownership would offer greater value than continued access (Lehr et al., 2020). This mechanism aligns with the process-oriented perspective developed in the theoretical background, where repeated access-based use fosters familiarity, reduces uncertainty, and enables consumers to reassess the relative value of ownership (Grönroos, 2008).

Accordingly, the positive relationship between customer loyalty to ABS and product purchase intentions can be understood as an outcome of prolonged, usage-based evaluations, particularly among frequent users and those with favorable attitudes toward ownership.

While prior research conceptualizes deservitization primarily as a firm-level phenomenon related to performance recovery and strategy restoration (Gomes et al., 2021), our results illustrate how deservitization-like outcomes may emerge at the consumer level through different underlying mechanisms, with implications for sustainability. Specifically, access-based usage can generate rebound effects not only through intensified use but also through ownership rebound, whereby access-based consumption leads to subsequent product acquisition. Munten et al. (2024) demonstrate that ABS can encourage increased consumption rather than substitution, thereby challenging assumptions about their inherent sustainability. Our findings complement this perspective by showing that unintended product trials through ABS may contribute to rebound effects not only by increasing usage but also by stimulating ownership intentions. Repeated usage of ABS may thus lead to both intensified consumption and eventual product acquisition, reinforcing material consumption rather than reducing it. Importantly, the presence of rebound effects does not imply that all, or even a majority of, ABS users will transition to ownership. Rather, even partial transitions, when systematically enabled by access-based usage, may be sufficient to offset expected reductions in material consumption. From this perspective, our results highlight a previously underexplored rebound mechanism operating through ownership acquisition rather than solely through usage intensity.

This has important implications for service providers, as ownership rebound may undermine long-term demand for access-based services by converting loyal and high-usage customers into product owners. In this sense, the very mechanisms that drive service adoption and engagement may also erode future service revenues. Further, from this perspective, deservitization can be understood as an expression of ownership rebound, rather than as a deviation from service logic. ABS may initially promote flexibility and reduced commitment, but over time, unintended trials and usage-related evaluations can prompt consumers to reinterpret ownership as a more efficient, controlled, or desirable mode of consumption (Lehr et al., 2020). This extends the discourse on deservitization by highlighting how consumer-driven transitions from services to products can arise from access-based usage itself and by linking these transitions to broader discussions of rebound effects and sustainability challenges in ABS.

Within urban mobility, strategic collaborations are becoming increasingly vital. The following implications delve into how rental e-scooter companies and product manufacturers can leverage partnerships to navigate regulatory challenges, enhance customer experiences, and drive mutual growth.

The widespread availability of rental e-scooters highlights how regulatory concerns and public perceptions can strongly influence consumer behavior. Safety-related issues, such as reckless driving and sidewalk clutter, have prompted increased scrutiny by regulators (Gössling, 2020). As of 2023, over 700 million urban residents live in areas where rental e-scooters are either highly regulated or banned (Heineke et al., 2023). These interventions reveal the operational volatility that ABS providers face in sensitive urban settings. Importantly, such interventions can also affect product owners, especially in overlapping markets. To improve resilience and unlock new growth opportunities, service providers could pursue strategic partnerships with product retailers and manufacturers. Joint marketing efforts between rental companies (e.g. Bird) and producers (e.g. Xiaomi) could position rental services and product sales as complementary rather than substitutable. For instance, campaigns might frame rental as an “on-the-go” solution and ownership as the ideal for frequent riders.

Drawing on our IPMA (see Appendix J), firms can better allocate resources between cultivating ownership attitudes and building ABS loyalty. Since product ownership attitudes have a stronger influence on purchase intentions than ABS loyalty, we suggest several possible segmentation strategies: (1) for customers with a strong product ownership mindset, retailers should emphasize product sales, offering exclusive promotions or bundled deals with ABS providers to ease the transition to ownership; and (2) for customers with low product ownership inclination, ABS firms should highlight the convenience and flexibility of, possibly integrating loyalty incentives to enhance engagement. Such partnerships may also create synergies: retailers gain access to high-frequency riders, while ABS firms enhance brand value through trusted retail collaborations. For example, bike stores could integrate e-scooter maintenance and service workshops to cater to both renters and prospective buyers. By pooling expertise and resources, ABS providers and product retailers can co-create value, driving revenue diversification and fostering long-term customer relationships.

If a growing share of ABS users transitions to ownership, this may create opportunities for product ownership, which creates a growing market for service and repair workshops. As rental customers transition into product ownership, they increasingly seek reliable maintenance services to keep their e-scooters in optimal condition. This presents a significant opportunity for service and repair workshops to cater to an expanding customer base, moving beyond serving only rental firms and into direct consumer markets. Workshops could capitalize on this trend by building solid relationships with local e-scooter retailers to position themselves as authorized repair centers and offering comprehensive service packages, such as regular maintenance plans and warranty extensions. As customers with high attitudinal loyalty to ABS shift toward product ownership, workshops should focus on customer service excellence to build long-term loyalty among this growing user base by offering value-added services that differentiate them and capture a sizable share of the e-scooter maintenance market.

While this study sheds light on the nuanced relationship between customer satisfaction, loyalty, and purchase intentions in ABS, several avenues for further research emerge to better explain factors influencing the shift from rental to ownership. First, future work could employ switching path analysis to examine the conditions that prompt customers to abandon mixed consumption in favor of ownership. Such exploration may reveal underlying preferences, motivations, and situational factors shaping these consumption shifts. Second, broadening the scope by incorporating additional mediators and moderators between ABS loyalty and purchase intention could yield richer insights. Constructs such as perceived value, product quality, risk, and social influence may clarify why satisfied customers consider ownership, revealing the cognitive processes that drive their intention.

Third, while this study measures purchase intentions—a standard survey metric (e.g. Keiningham et al., 2015)—future research should strengthen contributions by examining actual behavior. Longitudinal designs that track decisions over time would better capture how satisfaction, loyalty, and attitudes toward ownership influence real-world purchases, thereby bridging the gap between intention and action. A further limitation of our study pertains to the estimation of the loyalty construct based on only two items, which capture the price sensitivity and word-of-mouth dimensions of service loyalty (Bloemer et al., 1999). Further research should also include a reliable measure for the complaining dimension of the construct.

Fourth, this study’s reliance on student participants imposes limits. Although students are heavy e-scooter users, they are relatively homogeneous in age, lifestyle, and socioeconomic status (Peterson, 2001). Prior research cautions against generalizing from student samples (Beltramini, 1983; Enis et al., 1972). While attitudes may remain stable, purchase intentions are more likely to vary across life stages. Replicating the study with more diverse samples, such as working adults, older consumers, and families, could test whether relationships between ABS loyalty, ownership attitudes, and purchase intention persist across groups. Finally, future research could examine hybrid consumption, in which customers engage in both rental and ownership, to reveal the motivations and perceived benefits underlying this evolving mode of access.

During the preparation of this work, the authors used Microsoft Copilot to proofread and copyedit sections of the text. After using this tool, the authors reviewed and edited the content as needed and take full responsibility for the content of the publication.

The supplementary material for this article can be found online