In quest of perceived transaction cost's impact on fintech users' intention: the moderating role of situational factors

The study investigates how perceived transaction costs shape Pakistani consumers’ intention to use fintech, addressing limited adoption despite rising digitalization and mobile money trends. In Pakistan, barriers such as low literacy, limited financial inclusion, and affordability issues hinder fintech uptake. Prior local research has focused largely on technology acceptance constructs (e.g., TAM, UTAUT), leaving cost-related deterrents underexplored. This research integrates Transaction Cost Economics (TCE), Innovation Diffusion Theory (IDT), and Belk’s Theory of situational factors to analyze antecedents of perceived transaction cost (PTC) and their impact on intention to use fintech (IU), while assessing how situational contexts—pandemic and impending policy (PIP) and environmental and physical surroundings (EPS)—moderate the PTC–IU relationship. The study aims to clarify: (1) how PTC and its antecedents influence IU; (2) how PIP and EPS moderate the PTC–IU link; and (3) what actions can improve fintech adoption to strengthen users’ intention to use.

The paper synthesizes three frameworks. Innovation Diffusion Theory (IDT) explains how consumers adopt innovations through peer effects, perceived attributes, and diffusion processes, noting that adoption depends on user attitudes and perceived complexity/compatibility. Transaction Cost Economics (TCE) posits that bounded rationality, opportunism, and transaction characteristics (frequency, uncertainty, asset specificity) influence exchange choices and costs; applied to e-business, higher uncertainty and asset specificity can raise PTC and deter usage. Belk’s Theory introduces situational factors (e.g., physical and social surroundings, time, task, antecedent states) that systematically shape consumer behavior. The authors operationalize PTC antecedents—perceived asset specificity (PAS), perceived complexity (PCX), perceived uncertainty (PUC), dependability (DPND), and convenience (CONV)—and propose: H1 PAS→PTC (+); H2 PCX→PTC (+); H3 PUC→PTC (+); H4 DPND→PTC (−); H5 CONV→PTC (−); H6 PTC→IU (−). They further hypothesize moderation by situational factors: H7 PIP moderates PTC→IU; H8 EPS moderates PTC→IU.

Design: Cross-sectional online survey of Pakistani adults (≥20 years) with internet access and exposure to financial transactions; convenience sampling targeted literate individuals familiar with fintech. Recruitment via WhatsApp, Facebook, email, and messages. N = 276 valid responses. Data collection window reported as September 1–30, 2022. Instruments: Two-part questionnaire—demographics (gender, age, education, occupation, income, fintech experience) and constructs measured on five-point Likert scales (1 = strongly disagree to 5 = strongly agree). Measurement sources: PCX (3 items; Lee 2021); PAS (1 item; Pan et al. 2022); PUC (3 items; Lee 2021); DPND (3 items; Swan et al. 1988); CONV (3 items; Zhang & Kim 2020); PTC (3 items; Li & Fang 2022; Nirmawan & Astiwardhani 2021; Urumsah et al. 2022); IU (3 items; Alshari & Lokhande 2022); PIP (5 items; De Haas et al. 2020; Sumaedi et al. 2020; plus one new item); EPS (4 items; Humpel et al. 2002; Ashraf et al. 2014). Pretesting: Ten respondents reviewed language and length; two management science experts validated content; questionnaire simplified to reduce fatigue. Ethics: Institutional approval (Yanshan University), informed consent, anonymity; demographics not mandatory. Analysis: Partial least squares SEM (SmartPLS 3), two-step approach (Anderson & Gerbing, 1988). Measurement model: Indicator loadings ≥0.70; Cronbach’s alpha ≥0.77; composite reliability ≥0.87; AVE ≥0.593, supporting internal consistency and convergent validity. Discriminant validity met via Fornell–Larcker and HTMT (≤0.90). Collinearity: Inner VIFs 1.00–2.475 (below thresholds). Structural model: Bootstrapping with 5000 subsamples; assessment of path coefficients, t-values, p-values, and R². Model fit: R² = 0.667 for PTC, 0.404 for IU, indicating good explanatory power.

- Antecedents of PTC (all significant): - PAS → PTC: β = 0.203, t = 2.725, p = 0.007 - PCX → PTC: β = 0.181, t = 2.806, p = 0.005 - PUC → PTC: β = 0.204, t = 3.384, p = 0.001 - DPND → PTC: β = −0.223, t = 2.384, p = 0.017 - CONV → PTC: β = −0.186, t = 2.853, p = 0.005 - PTC → IU: β = −0.280, t = 4.398, p < 0.001 (higher perceived transaction cost lowers intention). - Situational moderators and direct effects: - PIP → IU: β = 0.242, t = 3.685, p < 0.001; interaction PTC×PIP → IU: β = −0.149, t = 2.610, p = 0.009 (authors interpret PIP as mitigating barriers by encouraging fintech use under restrictive conditions). - EPS → IU: β = 0.339, t = 5.161, p < 0.001; interaction PTC×EPS → IU: β = 0.139, t = 2.245, p = 0.025 (EPS counteracts the negative impact of PTC on IU; reported as fully neutralizing in narrative). - Variance explained: R² = 0.667 (PTC), 0.404 (IU). - Relative impacts on PTC (by absolute β): DPND (0.223) strongest, followed by PUC (0.204), PAS (0.203), CONV (0.186), and PCX (0.181).

The findings confirm that perceived transaction costs significantly deter Pakistani consumers’ intention to use fintech. Cost-enhancing antecedents (asset specificity, complexity, uncertainty) elevate PTC, while trust-related dependability and convenience reduce PTC, thereby supporting adoption. This addresses the research questions by identifying which cost drivers matter and demonstrating that situational contexts can change behavior: during pandemics or under adverse environmental/physical conditions, consumers may turn to fintech despite cost concerns. Practically, the results call for cost management strategies focused on lowering perceived complexity, reducing uncertainty through transparent information and security assurances, and strengthening reliability and convenience. Theoretically, integrating IDT, TCE, and Belk’s situational lens offers a comprehensive view of fintech acceptance beyond conventional TAM-based studies in Pakistan, showing that both structural (costs) and contextual (situations) forces shape intention.

This study advances understanding of fintech adoption in Pakistan by centering on perceived transaction costs and their antecedents within an integrated IDT–TCE–Belk framework. It establishes that higher PTC suppresses intention to use fintech, while dependability and convenience lower PTC. Situational factors—particularly EPS—can neutralize PTC’s negative effect on intention, and PIP also influences usage under restrictive conditions. Contributions include highlighting cost-oriented levers for practitioners (e.g., simplifying processes, enhancing reliability and transparency, improving service convenience) and demonstrating the value of combining theoretical perspectives for stronger explanatory power. Future research should validate cultural generalizability across countries, examine real usage behaviors, explore interrelationships among antecedents, and broaden the scope to additional cost-related and non-cost determinants, potentially using more targeted sampling strategies.

- Sample and mode: Online, convenience sampling targeted literate individuals with fintech exposure; limits generalizability and may overrepresent higher education/income groups. - Incentivization: Use of incentives may influence participation and responses. - Scope: Focused on cost-related constructs; does not include other important determinants of fintech adoption. - Measurement: Did not measure actual usage behavior; relies on intentions. - Model relations: Potential interrelations among independent variables were not modeled; SEM indicated no additional paths required, but future work could test cross-links. - Context specificity: Results pertain to Pakistan; replication needed to assess cross-cultural applicability.