The role of trust in purchase intentions in collaborative consumption in emerging economies: A Colombian perspective

Abstract

This study explores the role of trust in purchase intentions for collaborative consumption services in Colombia. Data were collected through an online survey (n = 233) to test the proposed conceptual model. Results confirmed the direct relationship among disposition of trust, trust in the intermediary, and trust in the service provider and, consequently, the positive relationship among these variables in relation to purchase intentions. Additionally, the disposition of trust does not directly influence purchase intentions. Practical implications for the enhancement of trust are provided. This work advances the literature on the influence of trust on collaborative consumption in emerging economies.

Keywords:

• collaborative consumption
• trust
• purchase intentions
• emerging economies

1. Introduction

Collaborative consumption (CC) has transformed the way consumers and firms interact (Chen & Wang, 2019; Mittendorf, 2018; Netto & Tello-Gamarra, 2020). CC highlights the dual role of consumers as obtainers and providers engaging on a platform in the interchange of services or resources with other consumers (Ertz et al., 2016). It also provides incentives for sustainable consumption practices due to the circulation of resources and temporary access to services (Billows & McNeill, 2018). Other terms used in academic literature to designate CC include sharing economy or gig economy (Parente et al., 2018). CC significantly contributes to the global economy, by 2022, the global market size was valued at more than USD 149 billion (New Wired Media, 2023) and is expected to be valued at 600 billion by 2027 (Statista, 2023b).

Despite CC’s global importance, studies mainly focused on developed countries. Researchers agree that studies addressing online adoption behaviors in emerging markets such as Latin America are still scarce (Alzamora-Ruiz et al., 2020; Bianchi & Andrews, 2012; Dakduk et al., 2017). Furthermore, Latin American markets present certain particularities, such as slow online adoption, cultural barriers, lack of trust, and informality, which directly impact consumer behavior and adoption of online services (Nasco et al., 2008; Ventre & Kolbe, 2020).

In emerging markets, CC services encounter unique situations in terms of customers and firms, mainly derived from not being institutionally based compared with developed countries (Chen & Wang, 2019). In Colombia, CC faces challenges due to the lack of regulatory and policy frameworks. However, the contribution of the sharing economy to the economy and employment access of the population is significant (Giraldo, 2019). By 2023, penetration of the internet among Colombians was about 75%, with around 25 million digital buyers (Hootsuite, 2023). It was the third largest e-commerce market in the region and was expected to continue growing over the years (Statista, 2023a). In Latin America, Colombia occupied the fifth place in CC initiatives after Brazil, Argentina, México, and Perú (Quintero-Ramírez, 2018).

In CC, as interactions are given through platforms in an online context, trust is identified as one of the most influential factors for consumers to engage in CC activities (Mittendorf et al., 2019; Wagner-Mainardes et al., 2019). In a literature review of the effect of trust in CC, Ter Huurne et al. (2017) identified trust as the most important driver of C2C activities. There is considerable evidence indicating that trust is a necessary condition to participate in online interrelationships (e.g., Gefen, 2000; Kim et al., 2008; Lee & Lee, 2004). In emerging economies, lack of trust is identified as one of the main barriers to engaging in CC activities (Chen & Wang, 2019) and related online activities such as e-commerce (Hajli et al., 2017; McCole et al., 2010; Ventre & Kolbe, 2020).

Therefore, this study fills this gap by providing empirical evidence to a conceptual model that analyzes the role of trust disposition, trust in the intermediary, and trust in the service provider as antecedents of purchase intentions for CC in Colombia. This model is based on previous research conducted by Mittendorf (2018), which analyzes similar variables in developed countries (i.e., EEUU and Germany). However, our research establishes two main distinctions to advance in theoretical and practical contributions to the extant literature: First, while Mittendorf (2018) only focuses on the hospitality platform Airbnb, our study analyzes the CC integrating different services (e.g., Rappi, Uber, Airbnb, etc.) to understand CC as a general phenomenon. Second, this study clarifies the role of trust in purchase intentions on CC in emerging economies by exploring the perceptions of Colombian consumers. This article addresses the following research question: What is the role of trust in purchase intentions on platforms of CC for emerging markets? Therefore, our contribution is twofold: to analyze the Colombian consumer’s perception of trust and its influence on the purchase intentions of CC and explore the implications of conceptual models empirically tested in developed countries in different realities such as Colombia.

To accomplish these objectives, the remaining paper is structured as follows. First, we review the literature on CC and the role of trust to build the hypotheses for a conceptual model. Second, we present the methodology and results. Finally, we detail the theoretical and practical contributions and suggestions for future research.

2. Literature review

2.1. Trust in collaborative consumption

Belk (2014) considers that sharing involves the act and process of distributing what is ours to others for their use. The consumer shares by nature, i.e., a smoker does not deny a cigarette to another smoker, and a person does not deny another person information about some specific direction. These behavior patterns are a fundamental principle of CC.

CC has been in play since long. Examples include WiFi sharing with friends or customers, sharing cars with strangers (Uber pool), bike share systems offered by governments in cities to improve mobility, and worldwide accommodation (Airbnb) (Möhlmann, 2015).

However, CC is only the tip of the iceberg, and from this point on, several possibilities open up to understand and comprehend that society’s attitudes and behaviors are changing. This allows us to question whether traditional economic models are really efficient and sustainable (Cañigueral Bagó, 2014).

CC is strengthening through the intentions of human beings; it is influenced by personal norms and attitudes rather than by subjective norms. CC is influenced and conditioned by economic reasons (Roos & Hahn, 2017).

Today, internet technology and renewable energies are beginning to merge to create a powerful infrastructure for a Third Industrial Revolution that will change the way power is distributed (Rifkin, 2012).

User participation in collaborative systems is associated with sustainable behaviors that motivate their interaction with CC. The main drivers of CC are trust, environmental awareness, cost/benefit ratio, convenience, and resistance to overconsumption (Pizzol et al., 2017).

Trust is approached from different domains. In psychology, trust is defined as the level of dependency that an individual shows with respect to other people in multiple situations (Nyamekye et al., 2022). In addition, trust is the determining factor in any exchange relationship. In e-commerce, trust is considered the most influential factor in a user’s decision-making process. Different investigations have shown that most people participate in the sharing economy when there are high levels of trust between service providers and consumers (Möhlmann, 2015).

Some studies show evidence of how trust in millennials helps improve the attitude toward collaborative consumption, which increases participation and purchase intention (Bhalla, 2023); likewise, trust is considered a tool that helps reduce consumers’ perceived risk and transaction costs and allows for breaking paradigms in the adoption of new Internet-based technologies (Tumaku et al., 2023). However, building and maintaining trust in the sharing economy is quite difficult as the model differs from other types of e-commerce transactions (Yang et al., 2018).

Disposition to trust is understood as a personality trait that may vary among consumers and determines propensity to trust (Kim et al., 2008). Trust levels may vary according to the interactions with other human beings. For instance, when the parties interact for the first time, the level of trust is low. Some authors state that trust is built over time, as relevant knowledge is accumulated to establish trust with another person (Mcknight et al., 2007), and it reduces the uncertainty and risk associated with social interaction (Jones & George, 1998). Therefore, the following hypotheses are proposed:

H1:

Users’ disposition to trust has a positive effect on the level of trust in technology platforms.

H2:

Users’ disposition to trust has a positive effect on the level of trust in service providers.

H3:

Users’ disposition to trust has a positive effect on end-user purchase intention.

2.2. Trust in collaborative technology platforms

Information and communication technologies (ICTs) have created opportunities for new economic models, including the collaborative economy. Years ago, it was believed that technological tools would transform organizations and that intermediaries would disappear. On the contrary, ICTs create a greater volume of data and information for providers and users (Rabinovich & Knemeyer, 2012). The internet offers new opportunities as well as challenges for companies that want to develop greater integration between users and providers. Main opportunities include the ability to integrate independent business models, increase customer service levels worldwide, enhance the ability to track the status of each supplier activity, and reduce costs associated with response times and service (García-Dastugue & Lambert, 2003).

Collaboration and trust in new technologies influence the way people and governments act (Xiang et al., 2015). Creating content in these collaborative networks results in continuous growth in this type of economy, wherein sustainability of the model is the main objective, supported by economic, social, and environmental benefits (Barnes & Mattsson, 2016).

Moreover, some authors claim that digital technology threatens democracy, social trust, and human relationships, arguing that the main input in the era of capitalism is people’s data, which are used to predict and modify human behavior. Those who control the data flow, such as Google, monopolize absolute power, a power never seen before in the history of mankind (Zuboff, 2015). However, trust makes the development of activities among strangers much easier. Decades ago, it was unthinkable to share a car with strangers or allow them to stay in our homes, but nowadays, it is possible. How is this possible? Platforms have changed social norms and particularly the way confidence is built (Hawlitschek et al., 2020).

The new ICTs allowed users to gain access to new tools such as platforms to share experiences, but a significant drawback of the collaborative model is the lack of trust in interacting with online communities (Cheng et al., 2020).

Some studies focus on developing the concept of ethics on platforms (Nadeem & Al-Imamy, 2020) and show that unethical behaviors are higher in e-commerce compared with conventional commerce. Conversely, trust in platforms emerges from digital identity (Zloteanu et al., 2018).

Users develop more trust in a platform if it provides more information on profiles and photos of hosts and users (Mao et al., 2020), which increases the repurchase intention for products and/or services. However, sharing personal data online makes data control and dissemination vulnerable, which entails privacy consequences and concerns, but conducting transactions on the internet without giving away data is almost impossible. Thus, one of the great challenges experienced by technology platforms is to be able to control and provide security and trust on the treatment of data to users (Lutz et al., 2018).

From a rational perspective, trust focuses on self-interest; however, an increase in trust toward the platform will decrease transaction costs (Tussyadiah & Park, 2018). Therefore, the concept of trust from the perspective of digital economy shows that internet technology can serve as a substitute for trust for businesses that do not require human contact (Wentrup et al., 2019). Today, however, one of the main issues concerning platforms is the accumulation of questionable information, e.g., veracity of honest and trustworthy comments (Celata et al., 2017).

This is how digital platforms make investments in technological infrastructure: protecting the exchange developed in the collaborative economy through intermediation tools and institutional mechanisms that allow the construction of safe, credible, and reliable environments (Lu & Yi, 2023). Risk perceptions, such as uncertainty and perceived risks, are variables that inhibit, and prevent users from engaging in online transactional relationships. Therefore, trust is not independent of the risk situation; as in the collaborative economy, interactions with platforms imply greater concern about information privacy and security (Alrawad et al., 2023).

A breaking point for trust in collaborative platforms is the time of payment. Previous studies define trust as the willingness of consumers to make payments through a mobile network, for which users expect platforms to comply with all the conditions and obligations regardless of the number of transactions or customer capacity (Cao et al., 2018). On the other hand, some authors refer to social trust as a concept associated with the way people feel about society as being trustworthy, which works to maintain a cooperative social environment promoting collective action. Therefore, social trust could improve trust in technology since it facilitates sharing valuable information with various actors and plays an essential role in innovative processes and collective learning (Cha & Lee, 2022).

By establishing two different trust constructs, a possible trust transfer can be investigated, wherein the individual trusts a secure and reliable source (technology platform) and the trust can be transferred to an unsecure and unreliable source (service providers) (Gong et al., 2019). It is highly unlikely that if the trust is weak in the secure source (platforms), it can be transferred to the unreliable source (providers). Therefore, we propose the following:

H4:

Trust in technology platforms has a positive effect on trust in service providers.

H5:

Trust in technology platforms has a positive effect on end-user purchase intention.

2.3. Trust in service providers

People play a decisive role in the collaborative model; if there is no trust between human beings, it would be difficult to make the model sustainable. Service providers must not only interact with technology platforms, which generate demand, but also establish direct contact with service users (Li & Wang, 2020). Thus, a higher level of risk and uncertainty is generated because the insecurity involves not only the customer but also the person providing services. Various mechanisms have been developed to build trust in the CC environment and ensure security in a bidirectional manner, for example, reviews and rating of both parties and insurance coverage (Cheng et al., 2020), which frames that trust in people is created not only in the user; the provider must also build it.

However, the user is more exposed, for example, in ride-sharing, as he/she puts his/her life in the hands of a stranger, who may or may not be a safe driver. There are several reports of assaults by Uber and Lyft drivers on passengers. Likewise, users must assume that drivers will take the fastest and most efficient route and not cause delays or mistakes in relation to the destination (Etzioni, 2019). Other factors that may discourage the creation of user trust in providers is the sharing of personal information, financial information, or location information, which may raise particular concerns about security and privacy risk (Lee et al., 2018).

Interaction between service provider and customer becomes necessary and is composed of several aspects such as service duration, investment, control, and personalization (Mittendorf et al., 2019). In many cases, these encounters take place between two complete strangers (Belk, 2014; Dreyer et al., 2016; Hamari et al., 2016). Therefore, the transactions involved are derived from independent factors such as time, social interaction, spatial proximity, and trust.

As in any service, in collaborative economy, users must subscribe before experiencing the service. Under this premise, trust toward service providers is gradually transformed. The more service providers reveal information in the initial service, the greater the users’ trust in them (Tran et al., 2022). Li and Tsai (2022) found that the more ratings and comments service providers obtain, the higher the level of the users’ trust. Information stimulate user perceptions: the more quality information the platform offers about the service provider, the greater the users’ security in using the services of the collaborative economy (Bhalla, 2023).

Users’ trust in service providers affects repurchase intentions; being able to evaluate service providers through the platform allows trust to be transferred from one user to another since the platforms offer supplier evaluation systems, which positively influence purchase intention (Cha & Lee, 2022). Several studies have found that social enablers, such as the perception of value, social interaction, and previous experience, are essential to explaining trust in service providers since functional, emotional, and social values are determining factors that increase and influence the ratings of service providers (Kong et al., 2020).

There is a difference between consumers of services on collaborative platforms and those who consume services through conventional models. In other words, trust and satisfaction are critical and determining variables that encourage participation and purchase intention in the CC environment (Berg et al., 2020).

When people interact socially, the parties involved initiate a negotiation process; therefore, sharing information on platforms can be regarded as a social form of interaction between providers and consumers (Cho et al., 2019).

Hawlitschek et al. (2018) argue that increasing the level of trust in the service provider and the technology platform increases users’ purchase intention. Therefore, the following hypotheses focus on the purchase intention developed by collaborative platform users.

H6:

Trust in service providers has a positive effect on end-user purchase intention.

This study develops a research model based on the literature review to test the research hypotheses, which are formulated on the following variables: disposition to trust (DIT), trust in technology platforms (TTP), trust in service providers (TSP), and purchase intention (PI), Figure 1.

Figure 1. Research conceptual model.

3. Methodology

In a collaborative economy, various relationships are built and lead to the development of solid levels of trust between participants in a bidirectional way. There are few studies on building TTP and service providers; moreover, studies in Latin America are limited and scarce. Data were collected through an online questionnaire conducted in Colombia.

The present study was carried out with the purpose of studying the influence of trust in relation to collaborative technology platforms, service providers, and purchase intentions. Using a simple random sampling method, the data was collected from respondents who were users of collaborative consumption services such as Aribnb, Rappi, Uber, or Didi, among others. At the beginning of the questionnaire, a filter question was inserted, with the objective to validate whether the respondents used services through these platforms. Users who had not used any of these services were excluded from the study. Each item was evaluated using a 5-point scale (strongly disagree (1)—strongly agree) (5).

Although variations to our constructs and items had already been used in previous research (Gefen, 2000; Mcknight et al., 2007; Mittendorf, 2018; Mittendorf et al., 2019; Pavlou & Gefen, 2004), we conducted a pilot test on 136 respondents. The objective of this pilot test was to identify word ambiguity and misinterpretation of individual terms or concepts and verify the reliability and validity of each construct and item. For this set, the data was very satisfactory, and no adjustments were made to the instrument.

Afterward, an invitation with a link to an online questionnaire was sent to a database of undergraduate and graduate students at a major private university. All respondents were asked for their approval to participate in the study. Data processing, security, and confidentiality of the collected data were guaranteed.

3.1. Measurement of variables

The questionnaire included 14 items adapted from previous studies, which were reviewed, selected, and adapted in the context of collaborative economy using a 5-point Likert scale (5 = “strongly agree”; 1 = “strongly disagree”).

Five items were adapted from Gefen (2000) and Mittendorf (2018) to measure the disposition to trust in the collaborative economy environment, namely, “I generally trust other people,” “I am used to trusting other people,” “I generally have faith in humanity,” “I feel that people are generally trustworthy,” and “I generally trust other people unless they give me reasons not to.”

Three items were adapted from Gefen (2000) and Mittendorf (2018) to measure TTP (intermediaries), namely, “I trust technology platforms,” “I believe technology platforms are trustworthy,” and “I believe technology platforms are trustworthy.”

Three items were adapted from Mittendorf (2018) and Pavlou and Gefen (2004) to measure TSP, namely, “I trust service providers,” “I feel that service providers are honest,” and “I believe that service providers are trustworthy.”

Finally, three items were adapted from Mittendorf (2018) and Pavlou and Gefen (2004) to measure users’ PI, namely, “I am likely to purchase some kind of service through technology platforms in the future,” “I would not hesitate to purchase some service through technology platforms,” and “If I have the opportunity I intend to purchase services through technology platforms.”

3.2. Analysis of results and discussion

The sample comprised 123 women (52.7%) and 110 men (47.3%), Mage = 21.39, SD = 3.91. We used SPSS Statistics 26 for Windows and SPSS AMOS 26 to perform statistical analysis of the data collected. First, we conducted the reliability analysis, which yielded a Cronbach’s alpha value of 0.884, i.e., a great internal consistency reliability of the instrument. Next, an exploratory factor analysis was performed to evaluate the correlation between the items and, thus, group them into the different variables with significant correlations.

3.3. Exploratory factor analysis

First, to conduct an exploratory factor analysis (EFA), we performed the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy and Bartlett’s test of sphericity to validate whether it was appropriate to develop an EFA. Some authors recommend a KMO value above 0.60 as acceptable, and the result for the KMO data set was 0.852, which is considered good. Bartlett’s test is highly significant (p = 0.000); therefore, performing an EFA is relevant and appropriate (see Table 1).

Table 1. KMO and Bartlett’s test

Kaiser-Meyer-Olkin Measure of Sampling Adequacy		0,852
Bartlett’s Test of Sphericity	Approx. Chi-Square	1917,524
	df	91
	Sig.	0,000

Source: author´s own

Next, to evaluate the existing correlation between the items, the EFA grouped the data set into four variables, yielding a 75.061 percentage of the explained variance (see Table 2). In addition, we identified the items with factor cross loadings higher than 0.6 (Churchill, 1979) and reviewed the item correlation matrix wherein an oblique Promax rotation was applied, thus predetermining a correlation between the variables and confirming the relationship between the items that make up each variable (Ha & Stoel, 2009) (see Table 3).

Table 2. Total variance explained

Component	Initial Eigenvalues			Extraction Sums of Squared Loadings			Rotation Sums of Squared Loadings
	Total	% of Variance	Cumulative %	Total	% of Variance	Cumulative %	Total
1	5,742	41,011	41,011	5,742	41,011	41,011	4,066
2	2,287	16,338	57,349	2,287	16,338	57,349	4,274
3	1,397	9,977	67,327	1,397	9,977	67,327	3,794
4	1,083	7,734	75,061	1,083	7,734	75,061	3,193

Source: author´s own

Table 3. Results of the exploratory factor analysis

Item	Variable
	DISC	CPSS	CPTI	IDC
DISC04	0,852
DISC01	0,843
DISC02	0,807
DISC05	0,776
DISC03	0,707
CPSS02		0,948
CPSS03		0,921
CPSS01		0,800
CPTI01			0,940
CPTI02			0,876
CPTI03			0,692
IDC01				0,889
IDC03				0,880
IDC02				0,791
Extraction Method: Principal Component Analysis. Rotation Method: Promax with Kaiser Normalization^.a
a. Rotation converged in 5 iterations.

Source: author´s own

3.4. Common method bias

For this research, we collected information through a self-administered online questionnaire at a specific time. It is surprising how respondents try to maintain consistency in their answers to similar questions (consistency effect) (Salancik & Pfeffer, 1977). Thus, the consistency motif as a potential source of method variance may generate illusory correlations (Podsakoff et al., 2003).

Therefore, the analysis of results validated the common method bias (CMB) to prevent inconsistencies in the validity and reliability of the conclusions on the relationships between the measurements (Podsakoff et al., 2003). CMB is used to explain the observed correlation between variable measures, and any error due to CMB can bias the estimates of the effects between variables. This affects hypothesis testing, leads to incorrect views about the amount of variance attributed to the item of a variable, and reduces the discriminant validity of the scale (Jordan & Troth, 2020; Podsakoff et al., 2003). Harman’s single-factor test was used to control for CMB, and the component factor analysis shows that the first factor does not explain more than 50% of the total shared variance of the items (Fuller et al., 2016; Jordan & Troth, 2020; Podsakoff et al., 2003). Harman’s single-factor test for our data set was 41.01%, i.e., there is no single factor explaining the extended variance in the multi-dimensional model.

3.5. Evaluation of the structural model

This research determines the relationships between variables and evaluates the item factor structure. Likewise, we determined the causal relationships of the variables among themselves and develop a confirmatory factor analysis using the structural equation modeling (SEM) to evaluate and confirm the validity of each construct (Anderson & Gerbing, 1988).

To evaluate the SEM fit, we performed the following analyses: initially, we evaluated how well a priori model reproduces the sample data through the absolute fit indices and then measured the proportional improvement of the model over a null model through the incremental fit indices, and finally, the parsimony fit indices. All these evaluations were performed using the AMOS program (Hu & Bentler, 1999; Marsh et al., 2004).

First, we evaluated the absolute fit indices used to determine the goodness-of-fit index (GFI) of the model, and the data reproduction yielded the following information: (x2 = 144.14; df = 71; GFI = 0.918; RMSEA = 0.067; PCLOSE = 0.042). Next, the incremental fit indices (IFIs) were evaluated (NFI = 0.927; TLI = 0.950; CFI = 0.961), followed by the parsimony fit indices (PFIs) (AGFI = 0.879). The SEM shows an overall good fit of the data; the RMSEA shows a result close to 0.06, considered a very good fit value; the IFIs are close to 0.95; and the GFI and AGFI are above the recommended thresholds of 0.90 and 0.80 respectively, for a good model fit (Hu & Bentler, 1999).

The analysis shows that the data collected fit the research model very well (Hu & Bentler, 1999). Furthermore, the explanatory values for R-squared were as follows: 0.14 (trust in the intermediary), 0.44 (trust in the service provider), and 0.28 (PI). The SEM results are shown in Figure 2 and Table 4.

Figure 2. Analysis of the research model with standardized coefficient (AMOS).

Source: author’s own

Table 4. Final model results

Parameter			Estimate	P-Value	$R^2$
CONINT	<—	DISCON	0,368	0,006	0,14
COPROV	<—	DISCON	0,277	0,012	0,44
COPROV	<—	CONINT	0,508	0,009
INTCON	<—	DISCON	−0,045	0,547	0,28
INTCON	<—	CONINT	0,293	0,015
INTCON	<—	COPROV	0,319	0,026
DISC05	<—	DISCON	0,750	0,023	0,56
DISC04	<—	DISCON	0,819	0,007	0,67
DISC03	<—	DISCON	0,632	0,010	0,40
DISC02	<—	DISCON	0,699	0,006	0,49
DISC01	<—	DISCON	0,820	0,025	0,67
CPTI01	<—	CONINT	0,842	0,012	0,71
CPTI02	<—	CONINT	0,895	0,036	0,80
CPTI03	<—	CONINT	0,739	0,004	0,55
IDC01	<—	INTCON	0,738	0,011	0,55
IDC02	<—	INTCON	0,749	0,013	0,56
IDC03	<—	INTCON	0,904	0,012	0,90
CPSS03	<—	COPROV	0,928	0,021	0,86
CPSS02	<—	COPROV	0,918	0,014	0,84
CPSS01	<—	COPROV	0,788	0,020	0,62

Source: author´s own

The results show the positive influence of disposition to trust on trust in intermediaries (H1). Likewise, the influence of disposition to TSP is weaker than the previous one (H2), while the effect of disposition to trust on PI is negative (H3) (not significant). Furthermore, the effect of trust in intermediaries on TSP is the most significant (H4), thus supporting H2. As stated in H5 and H6, trust in intermediaries and TSP significantly impact PI. Thus, the results show a transfer of trust from the intermediary to the service providers that affect PI.

Finally, the data analysis answers the research question and shows the importance of trust in intermediaries and service providers as the constructs that exert the greatest influence on PI.

3.6. Theoretical and practical contributions

The main objective of this research is to contribute to marketing and collaborative consumption literature by analyzing the role of trust in purchase intentions on CC platforms for emerging markets, specifically by providing empirical evidence for a conceptual model that explores the role of disposition of trust on trust in the platform and trust in the service provider and its effects on purchase intentions in Colombia.

While previous research has already explored the role of trust as a foundation to strengthen the positive effects of the sharing economy among its users (e.g., Zarifis et al., 2019), “the significance and mechanisms of trust development within the sharing economy are largely unexplored” (Räisänen et al., 2021, p. 3). Specifically, the role of trust in purchase intention in emerging economies remains undetermined. The present study contributes to the literature by analyzing the antecedents of purchase intentions for CC, integrating different platforms to understand CC as a general phenomenon, and exploring the perceptions of Colombian consumers.

Few studies have analyzed the role of trust and CC in developing countries (e.g., Gaber et al., 2021; Ratilla et al., 2021). However, no previous study has been conducted in Latin America, where the CC is growing at a relevant rate. Therefore, the findings of this study provide important theoretical and practical contributions.

First, our results support that disposition of trust influences trust in the intermediaries and trust in the service providers. Suggesting that personal willingness to trust others predisposes individuals to rely on the services provided through the CC platforms. Moreover, the effect of disposition of trust is stronger for trust in the platform than for the service provider. This may be explained by the fact that many of the platforms are well known and are the first contact for consumers, thus being crucial for the acceptance of the behavior. This result is consistent with previous research that established the importance of trust disposition as an antecedent of trust for the development of online transactions by influencing behavioral outcomes (e.g., Kim et al., 2008; Lee & Lee, 2004).

Second, the results imply that higher trust in the platform leads to higher trust in the service provider. This result indicates that consumer’s reliance on the platform is key for engagement in CC activities. Platforms may reduce the risk perceptions of the service provider. In CC, service providers are evaluated by the intermediaries that put forth certain requirements for providing services. Therefore, consumers may perceive that platforms have a selection process in place to guarantee the security of the services offered. Therefore, service providers registered on the platform must be trustable. This finding adds to previous research that has to acknowledge that trust can be transferred among different sources (Hong & Cho, 2011; Möhlmann, 2015; Pavlou & Gefen, 2004). In e-commerce settings, there is evidence that trust can be transferred from the intermediary to the seller (Verhagen et al., 2006; Wei et al., 2014).

Third, another key finding of this study is the direct effect of trust in the platform and trust in the service provider on purchase intentions. Trust may be transferred from the intermediary to the service provider such that consumers may engage in the behavior. This finding highlights the importance of trust for the acquisition of CC services and adds to the extant literature on the decisive role of trust for a consumer to engage in online transactions (Achmad Hidayanto et al., 2014). Specifically, it is consistent with Latin American studies that found a direct relationship between trust and consumers’ online purchase intentions (Ventre & Kolbe, 2020).

Interestingly, the direct relationship between trust disposition and purchase intentions was not significant, and the results did not support Hypothesis 3. In a similar study, Bianchi and Andrews (2012) found that Chilean consumer propensity to trust did not have a significant effect on the intention of online purchases. The authors argued that this could be because their sample consisted of highly educated and experienced consumers (as in our study) who are more complex in the relation of their disposition to trust as a trait of personality. Similarly, Wu et al. (2010) found a nonsignificant effect of disposition to trust and initial trust online, arguing that variables such as perceived Web assurance or perceived interactivity that they included in their model may have displaced the disposition to trust. In our study, this finding may imply that consumers need to trust in the intermediary and service providers to be involved in CC activities. In particular, in Colombia and Latin America, lack of trust is one of the main barriers to engaging in online behaviors (Chen & Wang, 2019; Nasco et al., 2008).

Overall, our results contribute to the developing interest in the study of trust in CC purchase intentions. Furthermore, our findings extend current research by providing empirical evidence of these behaviors in emerging economies, which has been relatively unexplored (Alzamora-Ruiz et al., 2020). This study focuses on the case of customers in Colombia where CC is getting increasing attention.

Findings also indicate important practical implications for marketing managers and policymakers. Strategies should foster trust in the customer and potential customers since it is a positive antecedent on purchase intentions. Therefore, trust in the platform and service providers must be enhanced. First, literature explores the critical role of security, privacy measures, and electronic service quality on the enhancement of trust toward the platform (Lee & Lee, 2004; Zhang et al., 2014). Platforms should guarantee reduction in risk perceptions and provide high-quality design, variety of products, and emotional appeal to influence e-service quality and impact trust (Gregg & Walczak, 2010).

Second, to enhance the trust toward the service provider, reputation has been identified as one of the most influential factors for the influence of seller’s trust (Ertz et al., 2016; Wang et al., 2015). Reputation is evidenced by positive customer feedback, ratings, and referrals through online reviews (Pavlou & Dimoka, 2006). Therefore, managers should continuously monitor their online reputation and provide proper information and response to customers’ requests. Furthermore, it is crucial to managing the community and its inputs via forums and online review, since it also has a positive influence on trust (Ventre & Kolbe, 2020).

Third, our findings demonstrated that trust in the intermediary influences trusts in the service providers. Therefore, platforms must make providers’ registration filters more demanding to enhance customers’ trust. This is an important issue since trust toward the service provider has a higher influence on purchase intentions. Managers should focus on building convincing brands since it has positive effects on trust-building for the CC context (Möhlmann, 2015).

For academics, this model provides a foundation for understanding the role of trust in emerging economies, which is an unexplored relationship in previous literature. Trust perceptions become essential in online transactions in Latin America since Internet users demonstrate a higher degree of perceived risk when buying online (Sánchez-Torres et al., 2017). Risk perception influences product adoption and is fundamental to purchase intentions. Further models could include different mediation or moderation variables to test further effects.

Finally, trust provides opportunities and challenges for users, service providers, and technology platforms; offering secure transactions and managing relationships with users are some of the challenges platforms face. Moreover, service providers have the obligation to offer accurate information and humanize the service, making the difference with conventional systems. Likewise, end users now realize the value that CC provides, such as security, efficiency, customer service, and service valuation, are some of the most important factors that collaborative technology platforms offer. Therefore, increasing trust in the model generates the durability and sustainability of CC in Latin America.

3.7. Limitations and suggestions for future research

This study has some limitations that have to be acknowledged. First, we employed a convenience sample from online purchasers, which is not representative of Colombia’s general population. Future research may incorporate representative samples that allow the generalization of results or test differences in specific age groups such as millennials or centennials, which show differences in online behaviors (Godelnik, 2017; Jose & Senthilkumar, 2020).

Second, trust has been recognized as a multi-dimensional complex construct, determinant to engage in social and economic relationships (Gefen, 2000; Ter Huurne et al., 2017; Wagner-Mainardes et al., 2019). Therefore, future research may incorporate other dimensions related to trust (e.g., trust toward the product, security, privacy, reputation, etc.) that may enrich the present model and that has been incorporated in other studies addressing CC (Tussyadiah & Park, 018; Yang et al., 2019).

Finally, since our study only focuses on one Latin American country, the generalization and external validity of the findings may be limited. Evidence of trust in online contexts shows that results may differ across emerging markets (Ventre & Kolbe, 2020; Wagner-Mainardes et al., 2019). Therefore, subsequent studies may test the conceptual model in other Latin American countries or emerging economies to better understand the phenomenon. As suggested by Bianchi and Andrews (2012), subsequent studies may also incorporate non-purchasers in the sample to allow comparison among subjects.

Disclosure statement

No potential conflict of interest was reported by the authors.