Exploring trust determinants influencing the intention to use fintech via SEM approach: Evidence from Pakistan

Abstract

Many fintech consumers are hesitant to perform online transactions given the lack of trust in online companies. In response to this reality, we construct and evaluate a trust-theoretic user acceptability model for financial technologies. The research seeks to look at the influence of trust on the intention to use fintech. A sampling of potential users in Pakistan is used to validate the model experimentally. Smart PLS 3 has been used to robust the theorized configuration of constructs (structural equation modeling) based on 275 survey responses. Contrary to other scenarios, the findings indicate that "customer trust in fintech" is more important than other aspects in determining technology adoption. Pakistani consumers' intention is positively affected by trust. Trust facilitators influence consumers' trust; among them, trust propensity is the factor having high values, followed by perceived size, interaction with online customers, perceived benefit, third-party seal, perceived ease of use, and perceived reputation.

1. Introduction

The term fintech combines the words "finance" and "technology" and has gained commercial interest [1]. EY defines fintech as a fintech company that combines advanced technology and business concepts to improve financial services [2]. Scholars describe fintech as a fresh financial sector segment that employs technology to increase financial operations [3]. Fintech is a novel financial bloc that takes up technology to increase monetary activities. Fintech businesses provide financial programming solutions. The use of fintech has increased faster than expected in 2009 [2]. Fintech companies operate in 189 countries, according to World Bank research [4]. Cooperating with fintech firms reduces administrative costs, lowers client expenditures, and speeds up service, boosting market profitability and financial success. Despite these benefits, financial practitioners and academics have identified various concerns linked with fintech companies due to their tasks and activities reducing customer trust [5,6]. 79 % of fintech organizations are susceptible to network security concerns, reducing customer confidence [4].

Meanwhile, the worldwide level of fintech awareness is remarkable, as transfers through PC or smartphone account for 89 percent of all transactions. In comparison, non-bank money transfers and peer-to-peer payments account for 82 percent [2]. The fundamental causes why consumers don't shop over the internet are cyber security and policies, company reliabilities, and company website technology. Web-based trust is vital for websites to thrive worldwide and retain client connections [7]. A party's desire to remain watchful of the other's behavior with the expectancy that the other would carry out a specified action crucial to the trustor, despite the trustor's ability to oversee or control another party, is referred to as a concept of trust. High levels of trust encourage and satisfy clients' expectations for pleasurable electronic shopping. This lowers uncertainty, perceived hazards, and interdependencies.

Fintech consciousness is alarmingly widespread. The IOSCO Research Report on Financial Technologies (Fintech) identifies eight key categories, including Payments, Insurance, Planning, Loan and crowdfunding, Blockchain, Trading and investments, Data and analysis, and Security [8,9]. These categories, as outlined by Giglio, cater to basic client demands and focus on the fundamental functions of financial markets. In contrast, EY's 2019 report expands the classification to nine classes: asset management, insurance, financing, exchange services, regulatory technology, loyalty programs, risk management, and payment, as well as different regions, including education. These variations in classification underscore the dynamic nature of the fintech landscape and the diverse ways organizations conceptualize their services. This research, centered on consumer behavior at the individual level, specifically examines factors influencing individuals' intention to use fintech in Pakistan. The technological landscape in Pakistan, while on the rise, remains in a developmental phase, with limited familiarity among the population. Notably, only literate individuals or the younger generation tend to be acquainted with financial technology. In contrast to countries like China, where even beggars might use QR codes for transactions, the majority of literate individuals in Pakistan or even in emerging economies are not aware of such facilities. There is a significant gap in technological adoption between emerging economies like Pakistan and more developed countries. The study does not focus on specific products or services within fintech, such as online websites, ATMs, mobile banking, or shopping apps. Instead, it concentrates on the broader categories of fintech, specifically within the risk management and payment domain. It's essential to acknowledge the limitations of this research, particularly its focus on individual behavior and the selected categories. Despite these limitations, this approach provides valuable insights into the unique challenges and opportunities within the Pakistani context, shedding light on the current state of fintech adoption in the country.

Additionally, customers are more inclined to get a commodity from a brand type they trust, thus making enterprises retain their clientele. According to the speedily expanding usage of Internet technology in commerce, searching and purchasing online items and services has potentially enhanced its part in fintech [10]. Even though cyber commerce has become more prevalent over the years, fintech users still encounter obstacles while enjoying buying online or using fintech services. Barriers to online buying include leakage of sensitive data privacy, difficulties evaluating items, issues with returning and exchanging things, delivery expenses, and discomfort with vendor anonymity, which reduces consumers' trust. Additionally, some analysis has been conducted on the influence of online buying and banking. Empirical studies have examined the connections between customer views of transaction expenses and both boosters and deterrents of the usage of fintech. The impact of consumers' perceived trust upon their inclination to operate fintech to acquire goods or services has also received minimal research. Investigating the perceived underlying trust elements that affect Pakistani clients' perceptions of fintech is thus crucial.

The following sections of this research primarily constitute its scientific contributions: First, this study widens an adoption decision's focus by considering positive factors (perceived trust). Numerous elements that affect fintech usage have been recognized by more recent studies. There haven't been many studies on how trust affects Pakistani consumers utilizing fintech. Because of this, it's crucial to examine the causes of perceived trust as they affect Pakistani consumers' propensity to use fintech. Most fintech research currently published in Pakistan revolves around the technology acceptance model (TAM); however, the factors pertaining to perceived trust have not been adequately studied. We focused on the elements of perceived trust in Pakistan. Previous researchers have not examined the perceived trust in-depth in Pakistan. Agha and Saeed [11] employed TAM and a single social risk factor to determine if buyers would adopt the technology. A scholar, Saleem [12], used the theory of perceived risk and the theory of reasoned action to research how auditors limit fintech risk management. However, this study has been limited by selecting the target audience of auditors in the context of the financial industry who already have a high level of experience in working in the financial department; this research also didn't work on the adoption behavior by considering the normal population. Ashraf, Hafeez [13], and Saleem [12] researched perceived trust, although the factors influencing perceived trust have not been appropriately studied. Another scholar examined customer behavioral intentions when utilizing fintech services, combining TAM with brand and service trust [14]. Some other scholars didn't research proper adoption behavior but researched the financial inclusions, such as Noreen, Mia [15], who showed the role of government policies in fintech adoption and financial inclusion in Pakistan and that the government of Pakistan has successfully adopted financial inclusion policies and initiatives. Other scholars researched financial inclusion [[16], [17], [18]].

Secondly, the research may give experts a greater awareness of how consumers perceive trust, which may be applied to establish building confidence procedures to enhance and promote users' adoption of online commerce, particularly in the expanding area of electronic payments for acquiring products and amenities. Thirdly, expanding the research field of economic effects based on fintech usage requires regulating related trust concerns. In conclusion, research from this angle would enable us to understand better the crucial role trust factors play in embracing fintech among Pakistan's general populace.

The following are the purposes of this investigation:

• 1.How can online customers establish their trust in fintech?
• 2.Examining the causes of perceived trust along with the way they affect customer inclinations to use fintech.
• 3.To recognize the elements that significantly affect fintech use intent.
• 4.To determine if combining different theories (TAM, Innovation diffusion theory (IDT), Vested Interest theory, Social network theory) offers a solid theoretical framework for examining the adoption of fintech in the context of Pakistan.

The remaining portions of the essay are structured in the following manner: The second section discusses the pertinent studies in the conceptual framework and provides the hypothesis; the third section describes the method; Section 4 lists the findings; the fifth section talks about the discussion; and Section 6 concludes and includes implications, limitations, and possible future study.

2. Conceptual framework

2.1. Theoretical development

The intended consumers' acceptance of novel technical systems, such as m-payment systems, is a critical first step toward their acceptability and ultimate success [19]. Because of the rapid evolution of technological developments and changing settings, the fundamental acceptance question "What factors or circumstances cause consumers to embrace a new technology?" is universally applicable to future novel technologies. This fundamental research topic has been studied from various angles, including technological features, demographics, and trust. The adoption of novel technologies has often been explained by a number of theories, such as the innovation diffusion theory [20], the theory of reasoned action (TRA) [21], TAM [22], and the theory of planned behavior (TPB) [23]. Thus, we used TAM, IDT, Social network theory, and Vested interest theory for this research.

TAM is a modified version of TRA [21] and was developed initially to describe user acceptance of IT [22]. This model speculates that behavioral intentions of use straightforwardly dictate framework use. This is subsequently affected by consumers' attitudes in relation to using the framework and the apparent value of the framework or the perceived usefulness of the system.

The IDT is a comprehensive social psychology theory that tries to describe adopting trends, pinpoint processes, and assist in predicting if or when innovations will be successful. Innovation as a concept, action, or thing that an individual or another level of adoption considers unique. Diffusion is the procedure by which technology is shared over time by a large portion of the population via designated channels [24].

According to the vested interest theory, something is of significant invested concern if it can potentially have significant personal repercussions. The immediateness of these effects and an individual's confidence in carrying out essential activities are two examples of elements that have an impact on vested interest [25].

According to social network theory from literature related to marketing, trust may be passed from one person to another [26]. That is, others would have an impact on a person's initial degree of confidence in an institution. According to social network theory, informal communication channels convey market information when the services are complicated to appraise. As a result, via these unofficial routes of contact, the trust may be transferred.

This research uses the TAM as the starting point and integrates it with literary work on "trust" to propound a trust-hypothetical model for fintech adoption. TAM (perceived ease of use, perceived usefulness), IDT (perceived benefits), and the vested interest theory (perceived ubiquity) were all employed. The study model incorporates trust factors in the electronic network like propensity-to-trust, word-of-mouth (WOM) referrals, structural guarantees based on social network theory, and trust theory. Additional user elements that may impact the user's adoption decision may not be adequately explained by the two basic belief constructs of PU and PEOU in the TAM. Previous IS research has added constructs like "perceived playfulness" [27], "perceived enjoyment and user involvement” [28], "personal innovativeness" [29], and "perceived information quality" [30].

Researchers have examined "trust" as a separate notion in internet commerce and e-Government. Pavlou [31] studied the customer acceptance of e-commerce by incorporating "trust" and "perceived risk" into the acceptance model and taking into account online technology-related uncertainties, "consumer trust," which depicts the user's questions about the sustainability of technology-based transactions [32], maybe an essential aspect of m-payment systems. Featherman and Pavlou [31] looked at trust as a precursor of cognitive assumptions for e-commerce adoption. We relate the trust antecedents to perceived trust regarding the intent to use fintech in a situation and offer a trust-theoretic fintech use model based on Pavlou's study. Consumer behavior and intention to use fintech are heavily moved by consumer trust [33]. The study's hypothesis and research model are given in Fig. 1.

Fig. 1.

Proposed model.

2.2. Hypothesis development

2.2.1. Perceived ubiquity

Ubiquity is a term that combines the concepts of portability, availability, and reachability into a single concept. People may connect to networks anywhere and anytime, making them traceable from any location [34]. Okazaki demonstrated four phases of perceived ubiquity: continuity, portability, immediacy, and searchability [35]. Compared to traditional payment methods, a new fintech has liberated consumers from time and space constraints [36]. However, because consumers are continually on the go, fintech providers face a problem providing ubiquitous financial services. To ensure the dependability and accessibility of fintech payment processing, they must put in a lot of work and resources. Users' initial trust may be enhanced as a result of this. Users may believe that network operators lack the capacity and honesty to deliver excellent services if fintech payment services are unavailable and inaccessible for online transactions. This will erode their faith in you. The impact of ubiquity on consumer trust and perceived usefulness has been studied previously. Previous authors discovered that ubiquitous connectivity impacted mobile user trust and the perceived utility of mobile payment [[37], [38], [39]]. Hence, we come up with that:

H1

Perceived ubiquity positively affects perceived trust.

2.2.2. Perceived site quality

Because the merchant is anonymous online, the interface serves as the 'online shop,' where initial impressions are established. It goes to reasoning that if a customer views a company's website as excellent quality, they will have strong, trusting perceptions of the seller's expertise, honesty, and kindness and will be willing to rely on the supplier. The quality of site information and decent system architecture and design help build consumer confidence and perceived value [40,41]. Correspondence, brand equity, and attractiveness influence consumers' perception of site value, allowing them to trust a financial platform. Efficient website design and attractiveness engage visitors and capture their interest, making design aesthetics a vital tool for building trust [42]. A well-built website that is simple to use and delivers coherent and successful purchasing insights necessitates specific corporate competencies. Customers can extend those qualities to the firm's ability to serve customers successfully. Thus, we propound that:

H2

Perceived site quality positively affects perceived trust.

2.2.3. Perceived benefits

Comparative advantages are obtained when a novel and trendy service appraises clients more than persisting services regarding economic advantages, self-perception, efficiency, and contentment [43]. On the other hand, offline and conventional internet channels offer their benefits. Offline banking/purchasing is based on information-rich participant interactions, preventing indifference and privacy while providing more security than wireless channels. Even if they have no prior experience in the field, individuals may respond positively toward fintech if they discover that it provides benefits that outweigh other exchanges, such as perceived willingness to customize, enhanced comfort, cost reductions, savings in time, enhanced diversity of goods to choose from, etc. [44,45]. As a result, we propose the following:

H3

A consumer's perceived benefits positively affect a consumer's perceived trust.

2.2.4. Perceived reputation

Consumer trust typically increases when a firm is thought to have a strong reputation [46,47]. Reputation refers to how well people regard a financial service provider. An excellent and favorable reputation ensures a business's competency, honesty, and goodwill, continuing to develop confidence when customers have no prior experience with the organization [48]. Furthermore, even without past commercial interactions, the company website's image favors the organization's perceived competency and enhances a person's trust in the company [49]. Thus, we propose:

H4

Perceived reputation positively affects the consumer's perceived trust.

2.2.5. Perceived size

A business's competency influences users' online confidence and purchasing intent, including company size [50]. Its perceived size characterizes customers' perceptions of a company's size. In outdoor shopping, apparent size can boost confidence in a firm. Clients may also think a big firm is more equipped and prepared to reimburse them for an equipment breakdown, especially if it has committed more to its reputation [51]. According to previous research, when the internet firm was an airline travel agency, its size had a good connection with consumer trust. Still, it had no connection with confidence when the internet company was an electronic bookshop [51]. Thus, we propose the following:

H5

Perceived size positively affects consumers' perceived trust.

2.2.6. Third-party seal

A third-party seal (TPS) alludes to a third-party certification authority, including a bank, consumer union, accountant, or computer firm, assuring an Internet seller [52]. Even if an individual has never used the website before, seals recognized by certified organizations may help lower the perceived risk in the course of a transaction [53] because trusted third-party guarantors are regarded to have some coercion over the internet seller through the enactment and by making policies of specific rules [54]. Similarly, the exhibition of a third-party seal like Web-Trust signals to customers that the supplier would genuinely attempt to fulfill its contractual responsibilities, increasing the consumer's trust in the supplier. As a result, when Web users see the seals on a website, it enhances their level of trust and lowers their perceived risk. As a result, we suggest that:

H6

The presence of a third-party seal positively affects the consumer's perceived trust.

2.2.7. Referral

The degree to which a customer is anxious about a positive reference or suggestion (referral) from interpersonal and social resources is referred to as the perceived benefits of effective referral or recommendation (referral) (e.g., family, friends, professional review comments, etc.). Consumers seek suggestions or views about fintech from many other interpersonal and social channels to decrease the risk of purchasing. The evidence obtained through word-of-mouth (WOM) recommendations may affect consumers' unsure impressions about fintech [55]. If someone with individual connections receives good WOM recommendations on the net as a financial medium, he may develop greater initial confidence in the e-channel as an economical medium [56]. Thus, we propose:

H7

There will be a more substantial positive effect of the perceived importance of referral on consumers' perceived trust.

2.2.8. Perceived ease of use

Perceptions of the fintech webpage are based on three factors. Two of those beliefs come from the TAM, a reliable model for determining how consumers acquire attitudes about technology and when they select to employ it [22]. TAM has been shown to work in various empirical investigations, including user adoption of word processors [22], spreadsheets, email, voice mail, telemedicine technologies, and web-based commerce. The subjective evaluation of a user of the site's utility in their buying or utilizing task is known as perceived usefulness. The buyer's intuitive evaluation of the period and exertion needed to navigate and use the website is perceived as ease of use [57]. PEOU has been found to affect the intention of users to purchase on online platforms [58]. A fully designed and handy website that is simple to browse and offers a systematic and successful buying experience necessitates specific business skills. Customers may extend such skills to the company's more general ability to service its customers effectively.

H8

Perceived ease of use positively affects trust in using fintech.

2.2.9. Interaction with online customers

A company's capability influences customers' online trust and purchasing intent, including encounters with online consumers [50]. Interaction with customers on a financial platform may enhance fintech users' confidence [59]. As a result, we suggest the following:

H9

Interaction positively affects trust in using fintech.

2.2.10. Trust propensity

Trust propensity or consumer disposition to trust is a user antecedent of trust that indicates a customer's qualities that contribute to assumptions about trustworthiness. An individual's trust disposition is a broad desire to believe in humankind and trust others [44]. Consumers' innate inclination to trust varies due to their individual development, character traits, and diverse cultures. This propensity is the consequence of continuing lifelong encounters and socializing rather than interaction with or awareness of a single trusted person [44]. A user's inclination to believe others influences their confidence in a particular supplier favorably. In contrast, a user's low tendency to trust others can negatively impact their confidence in a specific selling party. As a result, it is proposed that:

H10

Trust propensity positively affects the consumer's perceived trust.

2.2.11. Perceived trust and intentions to use

Scholars have described trust as a response relying on one's perceptions of another's traits [60]. He also presented a model of relational trust in organizational interactions to validate the proposed, which incorporates trustee and trustor qualities that impact the establishment of trust. Ability, kindness, and integrity are the three traits the framework represents to reflect the trustee's perceived trustworthiness.

The reasoning behind this approach is that if a trustor believes a trustee's (e.g., a vendor's) competency, kindness, and honesty are adequate, the trustor will build trust (an intent to tolerate susceptibility) in the trustee about fintech or vice versa. In other words, trust is a vital determinant of behavior in a scenario with a perceived risk of a bad result. Several trust researchers identified a link between confidence and readiness to purchase online from Web merchants [61]. As a result, we anticipate that increased trust will directly and positively impact fintech adoption. We propose the following based on the arguments mentioned above:

H11

A consumer's trust positively affects the consumer's intention to use fintech.

3. Methodology

3.1. Measurement development

This research experiment was designed to include a two-part survey with Likert scales with five points in the second and first parts using nominative scales. As a consequence, the first portion is made up of basic information. This survey section was designed to collect participants' descriptive insight, including their gender, age, education level, work status, and fintech experience.

The questionnaire's second section was built on constructs around the ideas of perceived site quality, perceived structural assurance such as perceived reputation, perceived size, third party seal, and referral, perceptions about website or application such as perceived ease of use, interaction with online consumers, trust propensity and intention to use. The appraisal of perceived ubiquity was taken from Okazaki and Mendez [35], containing three items, perceived site quality from Wakefield, Stocks [62] including three items, perceived benefit from Kim, Ferrin [44] and Kim, Shin [63] containing three items, perceived structural assurance such as perceived reputation, and perceived size from Chen and Barnes [33] and Koufaris and Hampton-Sosa [50], third party seal from Kim, Ferrin [44] and Wakefield, Stocks [62], and referral from Kim and Prabhakar [56] containing three items for each. The assessment of Perceptions about websites or applications, such as PEOU from Khatri, Gupta [22], contains three items for each interaction with online consumers from Chen and Barnes [33] contain three items; trust propensity from Hampton-Sosa and Koufaris [64] contains three items; trust from Srivastava, Chandra [65] contains three items and IU from Kim, Mirusmonov [37] contain three items.

For the data analysis, this research explores the intricate relationships among multiple latent constructs and observed variables. SEM offers a comprehensive framework that aligns with this research objectives by allowing us to assess complex interdependencies and test hypotheses. Given the exploratory nature of our study and the need to capture potential indirect effects, SEM's capacity to handle intricate models is a key advantage. Furthermore, the partial least squares (PLS) approach within Smart PLS 3 is particularly suitable for the relatively smaller sample size and offers flexibility in addressing the research questions of this research.

3.2. Data collection and descriptive statistic

A virtual survey was conducted to collect data from respondents and Pakistani nationals living in different cities within the country. This article employed a convenient sample strategy for data collection. The questionnaire's electronic format restricted the sample to those with at least rudimentary internet knowledge, skill, and experience. Five hundred individuals from various backgrounds who used WhatsApp and email were emailed a link to a survey. Due to security concerns, only 350 surveys were filled out. Seventy-five replies were removed because they were incomplete. The survey lasted three months, between September 1 and November 30, 2022.

Concerning convenience sampling, in the context of Pakistan's evolving fintech landscape, characterized by a growing but still limited awareness, we employed a convenience sampling technique. This decision was driven by practical considerations, given the nascent stage of fintech adoption in the country. Our approach involved reaching out to individuals with varying levels of fintech experience and sharing the survey within groups to capture insights from those already familiar with fintech applications. While this method may not represent a highly targeted approach, it aimed to address the challenge of limited fintech awareness and ensure a more inclusive representation of participants in our study.

Table 1 displays respondents' personal demographics. All of the replies were applied to explore the proposed hypotheses. According to the results, men comprised 69.82 % of the respondents. Concerning age, 62.18 % of the population was between the ages of 20 and 30. Regarding education, 46.91 % of respondents had master's degrees, and 40 % had doctoral degrees. Furthermore, 64 % of respondents were students, the most common occupation. The study invited persons with a college or university degree who are likely to earn higher incomes; the pay levels of the vast masses were rather high. 37.09 % of people have used fintech for more than three years.

Table 1.

Responders' demographic profile.

Variable	Descriptions	Frequency	Percentage
Gender	Male	192	69.82 %
	Female	83	30.18 %
Age	20–30 years old	171	62.18 %
	31–40 years old	100	36.36 %
	41–50 years old	4	1.45 %
	More than 50 years old	0	0 %
Education	Bachelor	36	13.09 %
	Master	129	46.91 %
	Doctoral Degree	110	40 %
	Others	0	0 %
Occupation	Working	91	33.09 %
	Self-Employed	8	2.91 %
	Unemployed	0	0 %
	Housewife	0	0 %
	Pensioner/Retired	0	0 %
	Student	176	64 %
Income Level	Less than 25000	19	6.91 %
	25001–35000	33	12 %
	35001–50000	28	10.18 %
	More than 50000	118	42.91 %
	No income	77	28 %
Years of using fintech	Less than one year	72	26.18 %
	1–3 years	102	37.09 %
	More than 3 years	101	36.73 %

Note: Income level has been shown in Pakistani rupees.

3.3. Transparency and openness

We conformed to the journal's criteria and provided details about our sampling strategy and other study-related metrics. The corresponding author can access all data, analytic code, and study materials. Smart PLS, version 3, was practiced to check the data. Both the analysis and the design of this study were not preregistered.

4. Results

We employed Anderson and Gerbing [66] two-step approach for determining the data we gathered. We checked the measurement model for convergent and discriminant validity first. The structural model was then examined to analyze the direction and strength of the connections among the constructs. By looking at the structural model, it became possible to identify the strength and direction of the connections between the constructions. Structural equation modeling (SEM) is often utilized to examine such data. SEM enables researchers to develop a rapid and accurate examination of hypothetical relationships among theoretical constructs and the connections among the constructs and their observed indicators with an adequate sample size [67]. For the robustness of the analysis, this study applied measurement model analysis and structural model analysis, which included different tests and have standard values, such as Construct Reliability and Validity, Fornell-Larcker Criterion, Heterotrait-Monotrait Ratio (HTMT), Inner VIF, Path coefficients, and R square. The statistics in the tables are correct and satisfy the specifications.

4.1. Measurement model assessment

Hair Jr, Hult [68] described that factor loading should be considered together with the reliability and validity of each item. Reliability refers to a measure's consistency. A measure must deliver consistent findings under predictable conditions and have a loading value of at least (0.7) for each item to be considered trustworthy. Cronbach's Alpha values should likewise be at least as high as (0.7) (Fig. 2). For the composite reliability, values between 0.5 and 0.7 are acceptable, while values above 0.7 and below 9.5 are regarded as excellent. The items shown in Fig. 3 are all compliant. Additionally, validity is defined as the grand mean of the squared loadings of the construct-related items. The accepted metric for figuring out convergent validity is the AVE. That is the percentage of variation that a latent construct's indicators can account for. If the construct's AVE value is (0.5) or above, it articulates more than 50 % of the variation of its components [68]. As seen in Fig. 4, Cronbach's Alpha is within acceptable bounds, and AVE values exceeded 0.5. The convergent validity of constructs is demonstrated as a result.

Fig. 2.

Cronbach's alpha.

Fig. 3.

Composite reliability.

Fig. 4.

Average variance Extracted (AVE).

To evaluate discriminant validity, this study examined the cross-loadings and Fornell-Larcker criterion. The Fornell-Larcker criteria are used to compare the square root of the AVE value to latent variable correlations (see Table 2). To evaluate cross-loading, the loadings of each indicator should be larger than the loadings of the indicators of the associated variables. The cross-loading requirements are perfect (see Appendix B); every item has a value greater than (0.7), and its most important value is when compared to several other items.

Table 2.

Fornell-Larcker criterion.

	IOC	IU	PB	PEOU	PREP	PS	PSQ	PUB	R	T	TP	TPS
IOC	0.889
IU	0.431	0.905
PB	0.539	0.707	0.895
PEOU	0.383	0.689	0.661	0.886
PREP	0.535	0.561	0.75	0.611	0.865
PS	0.534	0.61	0.667	0.559	0.665	0.922
PSQ	0.687	0.497	0.667	0.573	0.679	0.69	0.866
PUB	0.368	0.59	0.686	0.618	0.569	0.521	0.663	0.866
R	0.58	0.483	0.635	0.561	0.744	0.559	0.59	0.432	0.845
T	0.621	0.707	0.642	0.6	0.711	0.673	0.592	0.486	0.652	0.837
TP	0.41	0.321	0.207	0.275	0.498	0.311	0.388	0.234	0.511	0.613	0.901
TPS	0.501	0.401	0.449	0.377	0.543	0.507	0.596	0.346	0.654	0.563	0.32	0.9

Note: Diagonal elements are the square root of AVE.

4.1.1. Internal consistency reliability and convergent validity

The findings indicate that the constructs and indicators satisfy the reflective measurement criteria, meaning all indicators have loadings above 0.7. The Average Variance Extracted (AVE), as shown in Fig. 4, exceeds the threshold of 0.5. Additionally, the Composite Reliability (Fig. 3) values exceed 0.70, and Cronbach's alpha values (Fig. 2) meet the standard. In conclusion, the results confirm that all the indicators are suitable, convergence validity is established, and internal data consistency is achieved.

4.1.2. Discriminant validity

Fornell and Larcker [69] contend that the model's construction loading should be higher than the remaining constructs to achieve discriminant validity. The constructions in Table 2 all satisfy this requirement. The discriminant analysis technique for comparing cross-loads across structures is shown in Appendix B.

The discriminant validity results are evaluated using the heterotrait-monotrait (HTMT) correlation criterion, as shown in Table 3. According to scholars, the findings meet the HTMT 1 (less than one) criterion, indicating that discriminant validity has been demonstrated [70].

Table 3.

Heterotrait-monotrait ratio (HTMT).

	IOC	IU	PB	PEOU	PREP	PS	PSQ	PUB	R	T	TP	TPS
IOC
IU	0.468
PB	0.623	0.783
PEOU	0.432	0.774	0.751
PREP	0.609	0.623	0.878	0.712
PS	0.595	0.666	0.747	0.625	0.752
PSQ	0.799	0.554	0.769	0.672	0.799	0.792
PUB	0.434	0.656	0.793	0.712	0.675	0.595	0.798
R	0.639	0.53	0.737	0.651	0.884	0.63	0.703	0.515
T	0.739	0.822	0.751	0.708	0.845	0.786	0.705	0.584	0.736
TP	0.426	0.34	0.225	0.309	0.571	0.331	0.437	0.257	0.575	0.703
TPS	0.561	0.436	0.51	0.424	0.624	0.567	0.684	0.413	0.754	0.666	0.358

4.1.3. Structural model assessment

To assess the structural model and validate the hypotheses, a bootstrapping procedure with 5000 sub-samples was used. The structural model is evaluated based on the estimations and test results for the causal relationships between the variables shown in the route diagrams (Fig. 6).

Fig. 6.

Structural equation model.

4.1.4. Collinearity

Construct correlations are strong, according to Table 2, ranging from 0.837 to 0.922. Regression analysis could potentially be used to evaluate the probability of multicollinearity explicitly. The variance inflation factor (VIF), a common measure of collinearity in regression research, measures the degree to which other predictor variables characterize a predictor variable [71]. According to Hair, Anderson, and colleagues (1998), a threshold VIF of less than or equal to 10 (i.e., tolerance >0.1) is often advised. The probability of common method bias (CMB) has always been associated with self-reported measurements as a common issue. We have methodically compared all the variables to the single factor variable to prevent such uncertainty. Smart PLS 3 was used in our study, and the AVE values were fewer than 10. The inner VIF values for each construct in the model are shown in Table 4 and range from 1.737 to 4.102. The structural model has no collinearity issues since all constructs' VIFs are less than five.

Table 4.

Inner VIF values.

	IOC	IU	PB	PEOU	PREP	PS	PSQ	PUB	R	T	TP	TPS
IOC										2.308
IU
PB										4.102
PEOU										2.172
PREP										3.954
PS										2.406
PSQ										3.952
PUB										2.557
R										3.421
T		1
TP										1.737
TPS										2.082

4.1.5. Assess path coefficient

Five thousand subsamples were used to assess the path coefficient in a bootstrapping process. The testing findings for the hypothesis are summarised in Table 5, which shows that hypotheses 3,4,5,6,8,9,10, and 11 are supported. P values fall below 0.05. The specified requirements are met, and the path coefficients (−1 to 1), Mean, standard deviation (±2), and T values (greater than 1.96) are all up to par with the supporting hypotheses. Meanwhile, the other hypotheses have greater p values and thus are not supported.

Table 5.

Hypothesis testing.

Hypothesis		Original Sample (O)/Path Coefficient	Sample Mean (M)	Standard Deviation (STDEV)	T Statistics (|O/STDEV|)	P Values
H1	PUB - > T	0.014	0.014	0.055	0.250	0.803*
H2	PSQ - > T	−0.292	−0.289	0.046	6.366	0.000 ***
H3	PB - > T	0.23	0.224	0.067	3.437	0.001***
H4	PREP - > T	0.09	0.093	0.046	1.976	0.048**
H5	PS - > T	0.258	0.259	0.055	4.688	0.000***
H6	TPS - > T	0.215	0.218	0.051	4.177	0.000***
H7	R - > T	−0.146	−0.144	0.064	2.275	0.023**
H8	PEOU - > T	0.203	0.207	0.055	3.708	0.000***
H9	IOC - > T	0.247	0.243	0.04	6.119	0.000***
H10	TP - > T	0.393	0.388	0.047	8.387	0.000***
H11	T - > IU	0.707	0.705	0.043	16.436	0.000***

Note: P-value statistics values (*p < 0.1, **p < 0.05, ***p < 0.01).

Mediation analysis assessed the mediating role of perceived trust on the linkage between independent variables and intention to use fintech. The results (see Table 6) revealed that the indirect effect of PB, PREP, PS, TPS, PEOU, IOC, and TP on IU was significant. PSQ and referral are substantial, but their beta values are negative, which shows that these two hypotheses are not as per the proposed hypothesis, whereas PUB is rejected due to its high P values. This indicates that the relationship between PB, PREP, PS, TPS, PEOU, IOC, and TP on IU was partially mediated.

Table 6.

Indirect effect.

	Original Sample (O)	Sample Mean (M)	Standard Deviation (STDEV)	T Statistics (|O/STDEV|)	P Values
PUB - > IU	0.010	0.010	0.039	0.250	0.803
PSQ - > IU	−0.206	−0.204	0.036	5.709	0.000
PB - > IU	0.162	0.158	0.048	3.378	0.001
PREP - > IU	0.064	0.065	0.032	1.973	0.049
PS - > IU	0.183	0.182	0.041	4.495	0.000
TPS - > IU	0.152	0.154	0.038	4.027	0.000
R - > IU	−0.103	−0.102	0.047	2.217	0.027
PEOU - > IU	0.143	0.146	0.041	3.463	0.001
IOC - > IU	0.175	0.172	0.032	5.540	0.000
TP - > IU	0.278	0.273	0.031	8.926	0.000
T - > IU	0.707	0.705	0.043	16.436	0.000

4.1.6. An explanatory model

As disclosed in Fig. 5, the goodness of fit of the given model is fine. The explanatory factors explain 0.764 concerning perceived trust and 0.5 in the intention to use fintech; thus, R2 indicates that the model's explanatory power is good.

Fig. 5.

R square.

4.2. Impact of trust-associated factors

As per the results, hypotheses 3,4,5,6,8,9,10,11 are approved and related to fintech use, whereas the remainder are not. Pakistani consumers' intention is positively affected by trust (β = 0.707). Some facets do not positively shape consumers' trust; thus, consumers lack trust in using fintech. Factors that influence consumers' trust among them trust propensity (β = 0.393) is the greatest of elements, then PS (β = 0.258), IOC (β = 0.247), PB (β = 0.230), TPS (β = 0.215), PEOU (β = 0.203), and PREP (β = 0.090). Whereas the P values of referral and PSQ are found to be significant, the hypothesis wasn't supported due to their negative beta values (β = −0.146) and (β = -0.292), respectively. PUB didn't support it and was found insignificant.

H1 determines that perceived ubiquity positively impacts trust; however, results (PUB - > T, β = 0.014, t = 0.250, p = 0.803) show no significant positive impact on consumer trust. H2 determines whether perceived site quality positively affects consumers' trust. According to the findings (PSQ - > T, β = −0.292, t = 6.366, p = 0.803), consumers' perceived trust is not affected by the PSQ. H3 determines whether perceived benefits positively correlate with consumers' perceived trust. The result (PB - > T, β = 0.230, t = 3.437, p = 0.001) shows that PB affects consumers' perceived trust.

H4 evaluates whether perceived reputation positively affects consumers' perceived trust. The results (PR - > T, β = −0.090, t = 1.976, p = 0.048) revealed that perceived reputation impacts the consumers' perceived trust that may affect their intention to use fintech. H5 determines whether perceived size positively impacts the consumers' trust. This hypothesis is validated according to the results (PS - > T, β = 0.258, t = 4.688, p = 0.000). Perceived size increases consumers' trust. H6 is whether the third-party seal positively affects consumers' trust. The results (TPS - > T, β = 0.215, t = 4.177, p = 0.000) show that the TPS has a good and positive impression on consumers' perceived trust that may let them further use fintech. H7 states that referral positively impacts the consumers' perceived trust. However, the results (R - > T, β = −0.146, t = 2.275, p = 0.023) show no relationship. Though it is significant, the referral doesn't increase consumers' trust.

H8 states that PEOU positively affects consumers' trust. The results (PEOU - > T, β = 0.203, t = 3.708, p = 0.000) are in favor; thus, this hypothesis is accepted. H9 states whether interaction with online consumers positively affects consumers' perceived trust. Results (IOC - > T (β = 0.247, t = 6.119, p = 0.000) are significant. H10 determines whether trust propensity affects the consumers' perceived trust positively. Results (TP - > T, β = 0.393, t = 8.387, p = 0.000) are also significant.

H11states that perceived trust positively affects the intentions to use fintech. As per findings (T - > IU, β = 0.707, t = 16.436., p = 0.000), perceived trust strongly impacts consumers' intentions to use fintech.

5. Discussion

This study explores the elements influencing consumers' perceived trust, ultimately affecting fintech usage intentions. This research offers a combined model based on TAM, IDT, Vested interest theory, and social network theory. Our study approach yielded several interesting findings, reported here in one category: positive predictors. The discriminant analysis findings reveal factors that positively affect consumers' intention to use, and they are found to be significant. Pakistani consumers' intention is positively affected by trust. The overarching conclusion aligns with the existing body of knowledge, affirming that various factors play a crucial role in shaping consumers' trust in financial technology. Among these factors, trust propensity emerges as the most significant determinant. The statistics of this research have an extended range of noteworthy relevance for fintech operators and scholars.

H1 states that perceived ubiquity positively affects the perceived trust that could further let the consumer use fintech. However, it has been found that consumers have less trust in perceived ubiquity. This shows that consumers perceived less ubiquity, which does not positively affect consumers' trust, which ultimately affects intention to use fintech. Our studies here are not aligned with Kim, Mirusmonov [37], and Lee [72]. Previous research has focused on the importance of perceived ubiquity and found it strongly affects the intentions to use fintech [73]. Fintech service providers should pay more attention to ubiquity for a better response from fintech consumers. Perceiving less ubiquity can lead consumers toward traditional ways of performing transactions.

H2 findings depict that consumers' perceived trust is positively affected by the PSQ. This showed that PSQ did not increase consumers' perceived trust, affecting their intention to use fintech. It showed that the site quality of the fintech service providers is not qualified, or consumers didn't pay attention to it. However, according to previous studies, perceived site quality is essential [42]. The more professional the site is, the more consumers will be affected. Thus, based on this, service providers can focus more on the company's site quality because the site is the first look for consumers. Our study is not in line with Al-Debei, Akroush [45]. The initial impressions of a website are reflected in the site's quality.

Perceived benefits have a positive relationship with consumers' perceived trust, and the hypothesis is significant. This is due to the prevalent market competition. A consumer feels that if the company gives more benefits, it might be considered worth it or not cheap. Our study aligns with Park, Amendah [74].

Our study revealed that perceived reputation impacts the consumers' trust, which may affect their intention to use fintech. The hypothesis is significant because many factors may be considered as to why perceived reputation did not affect consumers' trust in using fintech. The biggest reason might be that well-reputed service providers now fulfill consumers' demands on the internet. Our study aligns with Srivastava, Chandra [65], who proposed that the company's reputation can increase consumer trust even if a consumer does not have experience. By contrast, perceived site quality has been found insignificant in assuring consumers of fintech usage. Even a well-reputed firm may have issues, and people may feel the difference in their trustworthiness.

On the other hand, consumers' perceived trust increases, and they intend to use fintech when it comes to the perceived size of the company. Our results show that it positively impacts the consumers' trust, and the hypothesis is significant. A more prominent company, such as more branches, a national or international brand, or widely used, will increase consumers' trust in fintech. However, some authors found that a company's perceived size does not significantly improve consumer trust [33,50]. However, Pakistani consumers are more interested in the perceived size of the company. Our study also does not align with that of Maqableh [75].

The third-party seal provides a sense of security and authenticity to consumers, which increases consumers' trust. Because there are so many online service providers, there are also chances of more fraud, but when a consumer finds a third-party seal, it will affect their trust. Our study is in line with Özpolat and Jank [76].

The results of H7 show no relationship between referral and consumers' trust. The hypothesis is found to be insignificant. The referral doesn't increase consumers' trust. Consumers think there are no more solid referrals they can rely on. Our study does not align with the Kim and Prabhakar [56]. On the other hand, our study is also in line with Kim [77], who showed that referral's impact on the consumers' perceived trust depends on the culture they have been brought up with, such as individualist or collectivist cultures. However, this study shows referral does not significantly impact consumers' perceived trust.

H8 states that PEOU affects perceived trust positively; as expected, the findings are supported. Here, it depicts that consumers do not find it challenging to use fintech. Services providers should work properly to ease the use of fintech. If consumers find it challenging, they will stop using the services. Our study aligns with Nangin, Barus [78].

Interaction with online consumers positively affects consumers' perceived trust. The results are significant. It is because a customer will reveal his experience to other consumers. A proper response from the consumers' side has a more substantial impact on other consumers. A seller or service provider will always make the efficiency of his things or services, but the reality can be counter-checked by asking other consumers questions. Our study aligns with Koufaris and Hampton-Sosa [50], Li and Kang [79], and Shen, Wu [80], who showed that interaction with online consumers affects the consumers' perceived trust.

H10determines that trust propensity positively affects consumers' perceived trust. A higher rate of trust propensity in an individual will result in high trust in him. Thus, our results are also significant. Our study aligns with Chan, Troshani [81]. H11 states that perceived trust positively affects the intentions to use fintech. According to findings, perceived trust has a strong positive impact on consumers' intentions to use fintech. Our study is in line with Ali, Raza [82], Ryu and Ko [83], and Tandon, Mittal [84].

A service company should thus concentrate on better strategies for fostering trust. Individuals will be reluctant to adopt fintech if there is a lack of trust. Additionally, the fact that fintech platforms are self-service offers significant strategic advantages for financial organizations. In order to increase service quality, retention of consumers, credibility, and web-based services earnings, financial institutions in Pakistan and other countries may leverage fintech as an exciting integrated service potential.

6. Conclusion

The elements of perceived trust considerations have been studied across several fields. This study aims to add to the body of knowledge on the use of consumer-focused financial networks in Pakistan, notably to help practitioners conceptualize, lower risk obstacles, and be ready for fintech disruption by focusing on the trust antecedents.

The overarching finding corresponds with established research, confirming that several elements are pivotal in influencing consumers' trust in financial technology. Notably, trust propensity stands out as the foremost determining factor in this regard. Some factors do not positively affect consumers' trust; thus, consumers lack trust in using fintech. Among the factors influencing consumers' trust, trust propensity is the dominant factor, as are laterally perceived size, interaction with online customers, perceived benefit, third-party seal, perceived ease of use, and perceived reputation. Whereas referral and perceived site quality are found to be significant, the hypothesis didn't support them. Perceived ubiquity didn't support and was found insignificant to influence fintech usage intention in Pakistan significantly.

Conclusively, the implications of these findings reverberate throughout the fintech industry. Operational capabilities, technical proficiency, and structure operational efficiency must be counted when providing services. Fintech adoption will be hampered by consumer dissatisfaction and mistrust due to inadequate or unsuccessful financial services operations, user difficulties, bad reputation, a lack of perceived ubiquity, or a lack of referrals. Whereas perceived benefit, reputation, company size, third party seal, PEOU, interaction with online customers, and trust propensity are considered much more important according to statistics of this study that increase consumers' perceived trust, service sponsors must sustain and manage these services in a properly manner for gaining consumers' trust.

6.1. Theoretical implication

In light of theory development, this research aims to create a theory by integrating variables into different schools of the nomological structural model, such as TAM, IDT, Social network theory, and Vested interest theory, deploying them to a new environment. This method is expected to result in a steady evolution of theory. Consequently, the suggested approach significantly yields the growing fintech exposition and research. This study's finding has wide-ranging ramifications for subsequent fintech research. The empirical results imply that trust and its antecedents have a more significant effect on consumers' judgment than the gain component, meaning that trust takes precedence over benefits such as PEOU for online banking clients when considering fintech. The empirical findings demonstrate that combining theories has solid explanatory power, which might provide a foundation for combining additional technological acceptance models. Information technology (IT) acceptance research has produced several competing models, each with its own set of acceptance determinants, such as innovation diffusion theory (IDT), social cognitive theory (SCT), theory of reasoned action (TRA), and expectation confirmation model (ECM). This finding is expected to inspire further research combining these opposing theories to create a further unified one with different approaches to the usage of fintech.

6.2. Practical implications

The study's results shed light on many crucial facets of customers' aspirations about fintech that were ignored in earlier studies. This research recommends that fintech organizations use trust-building mechanisms like guarantee declarations, enhanced familiarity via marketing, and long-term customer care to attract clients.

The study's results shed light on some crucial facets of customers' aspirations about fintech that were ignored in earlier studies. On the other side, offering clients a secure, reliable system is far more crucial. According to the findings, they should focus on PUB, PSQ, and R to prevent infiltration, theft, and identity fraud. Some practical suggestions are given below.

Fintech companies should prioritize personalized communication in fintech. They build trust through competence, integrity, reliability, and transparency in communication, privacy policies, customer support, and delivering on promises. Reduce information asymmetry by providing clear and accessible information about services, security, privacy, and user rights. Prioritize personalized communication, exceptional customer support, and prompt issue resolution to foster perceived trust in fintech.

Trust should be built through compliance and licensing in fintech. They should emphasize benefits, convenience, and user-friendly interfaces to enhance perceived trust in fintech services. Compliance with regulations and obtaining licenses enhance perceived trust in fintech services. Fintech companies should empower users with data control, customizable privacy settings, and transparent opt-in/opt-out mechanisms to foster trust in fintech services.

Fintech service providers should leverage social influence with user reviews, testimonials, and endorsements to increase fintech-related perceived trust. Manage reputation through ethical practices, customer satisfaction, and promptly addressing negative publicity or complaints. Display trust seals and certifications for external validation of security measures and compliance in fintech platforms.

6.3. Study limitations

This research has some limitations, which are described below. To begin with, this study relied on a web survey platform in which only responders with web access participated.

Second, to some extent, this research was a convenience sampling approach, not a fully target-oriented approach focusing only on those living in Pakistan. Thirdly, individuals who consider the rewards enticing (e.g., rewards for the lotto, candy, handsets, and hard cash packets) are more inclined to join, which restricts us from having more respondents.

Fourth, specific correlations between the independent variables are possible. These connections, however, are outside the focus of the research. Furthermore, SEM analysis revealed that additional pathways were not required. Fifth, the research is restricted in scope and only looks at trust variables mainly as they are viewed. It looks at how perceived trust variables positively affect Pakistani customers' willingness to utilize fintech. Sixth, this study didn't include the actual use of fintech.

Lastly, there are so many fintech types that this study is limited in this context, and it didn't select any specific fintech application. This study was one broader sense as Pakistan is in its beginning stage of using fintech.

6.4. Futuristic approach

Though there are limitations as well, this study also lists a set of possibilities for future investigation. Firstly, future research may include a wide range of ways of collecting the data, such as contacting professional companies. Secondly, the study may be recreated using samples from various nations to test the model's generalizability across diverse cultures. Thirdly, interactions between independent variables can be investigated in future studies. Fourth, future studies should elaborate on and examine other trust-related factors in understanding fintech usage intentions. Fifth, future academics should perform more analysis to analyze the actual use of financial technology in their study framework. Lastly, there are so many fintech types that future research may be conducted on other types of fintech.

Availability of data

The dataset analyzed during the current study is available from the corresponding author upon reasonable request.

Funding statement

Not applicable.

Ethics statement

Approval was obtained from the ethics committee of the School of Economics and Management, Yanshan University. The procedures used in this study adhere to the tenets of the Declaration of Helsinki. The professor of the university first approved the research questionnaire, and then we sent it to the respondents.

Informed consent

Informed consent was obtained from all participants. It was an online questionnaire-based study, and all the responders were fully aware of the academic research purpose of the study, as the purpose was clearly mentioned in the introductory part of the questionnaire. This study needed responders’ responses regarding fintech usage. The demographic information was not set as compulsory questions because of privacy concerns.

Strengths of the study

This study boasts several key strengths. Firstly, its methodological rigor, integrating diverse theoretical frameworks (TAM, IDT, Social Network Theory, Vested Interest Theory), ensures a comprehensive understanding of factors shaping consumers' trust in fintech. The literature review goes beyond summarization, critically engaging with prior research to identify gaps and establish a robust rationale.

The incorporation of discriminant analysis adds a quantitative edge, enhancing statistical robustness. Focusing on positive predictors and individual trust factors provides novel insights into fintech adoption. Lastly, the study's practical implications offer actionable guidance for industry practitioners, cementing its relevance.

In summary, the study's methodological approach, literature review, analytical techniques, and practical insights collectively position it as a valuable contribution to fintech research.

CRediT authorship contribution statement

Haifeng Zhao: Supervision. Nosherwan Khaliq: Writing – original draft, Conceptualization. Chunling Li: Validation, Data curation. Faheem Ur Rehman: Writing – review & editing, Validation. József Popp: Data curation.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following are the Supplementary data to this article: