The cashless cushion: A descriptive study of discretionary spending patterns via UPI and cash among undergraduate students

Introduction

India's payment infrastructure has undergone one of the most rapid and significant transformations in contemporary financial history. The Unified Payments Interface (UPI), developed under the auspices of the National Payments Corporation of India (NPCI), now serves as the primary channel for real-time, mobile-based financial transactions across the country. By 2024, India accounted for approximately 46% of all global digital payment transaction, with UPI alone processing nearly 84% of domestic digital transfer (Dev et al., 2024) [5] . This architectural shift carries extensive implications not merely for transactional efficiency but, more critically, for the psychological and behavioural dimensions of consumer financial decision-making. Among all demographic segments, undergraduate students represent a cohesive cohort for investigation. As digitally native individuals whose formative financial experiences are mediated predominantly by mobile applications, they inhabit an ecosystem where payment frictions are systematically minimised. Classical behavioural economics establishes that cash payments activate a 'pain of paying' mechanism, a psychophysiological aversion response that inhibits expenditure and promotes restrain (Prelec and Lowenstein, 1998) [11] . When this mechanism is attenuated through cashless systems, as UPI's seamless, one-tap architecture is designed to achieve, the psychological cost of spending diminishes, and individuals become susceptible to elevate non-essential disbursements. This study terms this dynamic the 'cashless cushion effect'. Empirical evidence increasingly supports this proposition. Dev et al., (2024) [5] found that 74.2% of UPI adopters selfreported increased expenditure post-adoption, attributing the change to payment intangibility and diminished spending salience. Kanwal and Singh (20250 documented a 9-15% rise in discretionary expenditures following digital payment adoption, with convenience and cashback incentivisation identified as a primary-mediators. Bhanushali (2025) [2] observed that UPI driven spending frequency increased among young urban users precisely because psychological detachment from physical currency eroded budgetary discipline. Among undergraduate students specifically, Balaji et al., (2025) [1] found that UPI users exhibited higher impulsive buying frequencies and weaker financial planning behaviours, particularly those with limited formal financial literacy. The present study investigates the cashless cushion effect by systematically within the undergraduate population. It's contributions are threefold: first, it applies a theoretically integrated framework unifying UPI convenience, impulse www.extensionjournal.com behavioural tendencies, peer-normative pressures, and financial awareness as a prospective moderator: second, it's employs a statistically rigorous stratified sample derived via Conchran's finite-population correction formula (N = 155); and third, it generates category-specific discretionary spending data across eight consumption domains, enabling granular insight into where the cashless cushion effect manifests most accurately among the following.

Problem Statement

The rapid expansion of digital payment systems has significantly transformed consumer transaction behaviour. While UPI has become one of the most widely adopted payment systems in India due to its convenience, speed, and accessibility, the psychological and behavioural consequences of this transition remain understudied, particularly among undergraduate students, who represent one of the most active and digitally embedded segments of digital payment users. Cash payments involve a physical exchange of money that typically increases expenditure awareness. In contrast, digital payments are intangible and frictionless, which may reduce spending consciousness and encourage more frequent discretionary purchases. Among undergraduate students, discretionary spending commonly occurs in areas such as food, entertainment, online shopping, subscriptions, and social activities. However, limited empirical evidence exists on whether UPI usage creates a 'cashless cushion effect' whereby digital payment systems psychologically reduce the perceived cost of spending relative to cash transactions. Understanding how payment modes influence discretionary spending patterns is important for promoting financial awareness, responsible consumption, and a deeper behavioural understanding of digital payment systems among the student population.

Aim, Objectives, and Research Question 3.1 Aim of the Study

To examine how payment modes, specifically UPI and cash, influence discretionary spending patterns among undergraduate students.

Research Objective

To analyse the influence of UPI payment usage on discretionary spending patterns among undergraduate students.

Research Questions

How does the use of UPI payments compared to cash affect discretionary spending patterns among undergraduate students?

Rationale of the Study

Digital payment system has rapidly eased everyday financial transactions. Undergraduate students increasingly rely on mobile payment applications for daily purchases owing to their convenience and flexibility of use. However, the transition from cash to digital transactions may alter spending behaviour by reducing the psychological awareness associated with physical money exchange. This study seeks to examine whether digital payment systems influence discretionary spending behaviour among undergraduate students. By analysing spending patterns across different expenditure categories and payment modes, it contributes to a better understanding of how digital payment adoption affects financial behaviour among youth. The findings may further provide insights for financial literacy programmes and behavioural economic studies focusing on digital payment systems. From a broader perspective, the study is secured in three theoretical traditions: Prelec and Loewenstein's (1998) [11] coupling hypothesis, which links payment tangibility to spending inhibition; the impulse-buying paradigm, which identifies reduced transactional friction as a precipitating condition for unplanned purchases (Rook, 1987) [13] ; and Rogers' (2003) [12] Diffusion of Innovations theory, which accounts for the peer-normative dimension of UPI adoption. Together, these frameworks provide a theoretically coherent lens for the quantitative patterns this study seeks to uncover.

Literature Review 5.1 Digital payment systems and consumer behaviour

The central connection between payment modality and consumer behaviour has been the focus of sustained scholarly attention across behavioural economics, consumer psychology and marketing science. The basic premise, established by Feinberg (1986) [6] and further developed by Soman (2003) [14] , is that payment salience, or the vividness with which expenditure is psychologically encoded, acts as an inhibitory mechanism on spending. Cash payments are the most essential because of their physical tangibility and immediate depletion from one's possession. It has decreased by electronic payment systems. Prelec and Loewenstein (1998) [11] formalise this relationship as the 'coupling hypothesis': the tighter the coupling in time and experience between consumption and payment, the more difficult the payment will be experienced and the more conservative the expenditure that results. India's post-demonetisation financial ecosystem provided a natural accelerator for studying this dynamic at the population scale. The government's 2016 withdrawal of high-denomination currency notes catalysed UPI adoption across diverse demographic segments. Bhanushali (2025) [2] demonstrated that post-demonetisation UPI adoption among urban youth was associated with measurable increases in expenditure frequency, particularly for self-indulgent categories, as the psychological coupling between consumption and cost was systematically weakened by digital payment architectures.

UPI and the Indian Youth Segment

Young adults represent the critical mass and foremost in India's cashless transition. A cross-sectional survey of postgraduate students by Moirangthem et al. (2026) [10] showed a significant preference for UPI across both microtransactions and larger discretionary outlays, with adoption statistically related to modified budget discipline across 18 expenditure categories. The architectural features of UPI-QR code presence, instant confirmation notifications and frictionless interface design appear to function as cognitive reinforcers of payment habituation and reduced deliberative processing at point of purchase (Dev et al., 2024) [5] . Balaji et al. (2025) [1] extended these findings to university www.extensionjournal.com students, finding that UPI users enjoyed higher impulsive buying frequencies and weaker financial planning behaviours. This is particularly obvious where students have low levels of formal financial literacy, suggesting a compounding vulnerability: minimal transactional friction in addition to minimal regulatory knowledge, with predictable expenditure consequences.

Impulse spending and the role of payment friction

Impulse buying (Rook, 1987) [13] is explained as a spontaneous, emotion-driven decision that does not require thoughtful cognitive evaluation. Digital payment environments reduce the time lag between the arousal of desire and the completion of a transaction, thus decreasing the threshold for such behaviour. The reward architecture embedded within UPI-affiliated applications, comprising cashback incentives, gamified loyalty programmes and algorithmically personalised promotional content, activates dopaminergic reinforcement cycles (Thaler, 1999) [15] . This establishes a self-reinforcing loop in which each discounted transaction primes neurological reward pathways for subsequent impulsive spending.

Peer Influence and social normalisation

Rogers' (2003) [12] Diffusion of Innovations theory provides a sociological framework for understanding the diffusion of payment behaviours through student peer networks. The adoption of digital payment methods among undergraduates is not only individually motivated but also is a significantly social negotiated. Social proof and conformity pressure job through peer payment norms that are among the strongest determinants of UPI uptake (Bhanushali, 2025) [2] . When the majority of a social reference group employs UPI for collective consumption activities such as dining out, ridesharing or shared subscriptions, without the use of UPI results in social costs in the form of transactional friction and implicit social deviance.

Financial awareness as a moderating variable

The main defence against UPI-facilitated overspending is financial understanding and awareness. Even in settings with high UPI adoption, Kanwal and Singh (2025) [8] showed that people with greater financial literacy scores maintained more strict budgetary adherence. In order to restore psychological salience diminished by cashless systems, Dev et al. (2024) [5] suggested incorporating expenditure-tracking dashboards and spending visualisation tools within UPI applications. Most importantly, though, basic understanding of digital technologies does not replace in-depth financial knowledge-a point that directly affects the design of literacy programs.

Research Gap

In spite of the increasing number of studies regarding the UPI adoption and aggregate transactions, there still seems to be a lack of analysis. In most cases, either the antecedents of adoption or the effects on the aggregate spending changes are explored; the mechanisms are not considered. Moreover, undergraduates who are susceptible to the cushion effect do not often become the focus of such researches. However, the current research aims at bridging the gap through exploring theoretically grounded antecedents of discretionary spending on eight different consumption categories.

Conceptual Framework and Hypotheses 6.1 Conceptual Model

This study adopts a simple and unified conceptual framework in which four independent constructs converge to predict the dependent variable, discretionary spending pattern (UPI vs cash):

• UPI Convenience -the perceived ease and speed associated with digital payments. • Impulse Spending Tendency -the individual's predisposition toward unplanned, affect-driven purchases.

• Peer Influence -the influence of friends and the social environment on payment and spending behaviour. • Financial Awareness -the individual's level of budgeting consciousness and spending self-regulation.

The notion of financial literacy is argued to have an inhibitory effect on discretionary expenses that counteracts the effects of other three variables that increase expenses. The asymmetry is explained by empirical findings that suggest that while financial literacy impacts convenience and impulsive shopping behavior, it does not completely eliminate it (Kanwal and Singh, 2025) [8] .

Hypotheses

Research Methodology 7.1 Research Design

A descriptive quantitative research design was employed. This design is appropriate for studies that aim to characterise the nature and magnitude of relationships between variables within a defined population without experimental manipulation (Creswell, 2014) [4] . The quantitative approach permits systematic measurement, statistical analysis, and generalisable inference from structured Likert-scale survey data.

Population

The target population comprised undergraduate students enrolled across all four years of study at a single degreegranting institution in India, yielding a clearly defined sampling frame of N = 744 students: First year (164), Second year (189), Third year (186), and Final year (205).

Sampling Technique and Sample Size

Proportionate stratified random sampling was employed to ensure accurate representation of each academic year. Sample size was determined using Cochran's (1977) [3] formula for estimating proportions from large populations:

Data Collection Instrument

The primary data was collected using a well-structured selfadministered questionnaire consisting of eight thematic questions A-H. The survey tool was distributed online using Google Forms through institutional WhatsApp platforms and other academic avenues. Participants were encouraged to voluntarily take part while ensuring confidentiality.

• Section A -Demographic information (gender, age, year of study, monthly allowance, UPI usage). • Section B -Payment Behaviour (5 items): attitudes toward UPI vs cash in everyday use. • Section C -UPI Convenience (5 items): perceived ease, speed, and friction lessness of UPI. • Section D -Impulse Spending (5 items): tendency toward unplanned, affect-driven purchases.

• Section E -Peer Influence (5 items): social normative pressures on payment behaviour. • Section F -Financial Awareness (5 items): budgeting knowledge and spending self-regulation. • Section G -Spending Category Frequencies (8 items):

frequency of spending across categories. • Section H -Discretionary Spending Behaviour (5 items): dependent variable construct.

Sections B-H used a five-point Likert scale (1 = Strongly Disagree to 5 = Strongly Agree, for attitudinal items; 1 = Never to 5 = Very Often for frequency items).

Statistical Analysis Plan

All analyses were conducted using R (version 4.x). The analytical plan comprised three sequential stages:

1. Descriptive statistics: frequency distributions, means, and standard deviations were computed for all items and construct composites. 2. Reliability analysis: Cronbach's alpha (α) was calculated for each section composite and the overall scale, with the conventionally accepted threshold of α ≥ 0.70 applied (Hair et al., 2019) [7] .

Correlation analysis:

Pearson product-moment correlation coefficients were computed between each independent construct composite and the Discretionary Spending composite (DV), with two-tailed significance testing and 95% confidence intervals via Fisher's ztransformation.

Results

Sample Demographics

A total of 155 undergraduate student responses were retained for analysis, consistent with the calculated sample requirement. Table 2 presents the full demographic profile of the sample. The sample was predominantly male (61.9%), with the modal age group being 20-21 years (53.5%). The distribution across academic years closely matched the stratified allocation targets. Nearly all respondents (96.8%) confirmed active UPI application usage, and 92.3% identified UPI as their primary spending modality, confirming the sample's high relevance to the study's focus. Among the Discretionary Spending items, 'Digital payments influence my overall spending behaviour' (M = 3.81, SD = 0.93) and 'UPI payments increase my frequency of small purchases' (M = 3.79, SD = 0.94) were the highest-rated items, indicating that respondents broadly acknowledged the frequency-amplifying and behaviour-shaping effects of digital payments. Within Section G, street food/snacks (M = 3.63) was the most frequent discretionary category, followed by transportation/ride services (M = 3.48) and online shopping (M = 3.41). Entertainment and subscriptions registered the lowest means (M = 2.79 and M = 2.97, respectively).

Reliability Analysis

Correlation Analysis and Hypothesis Testing

Pearson bivariate correlation coefficients between each independent construct composite and the Discretionary Spending dependent variable are presented in Table 6. All tests are two-tailed with 95% confidence intervals computed via Fisher's z-transformation.

H1: UPI Convenience → Discretionary Spending

A statistically significant, moderate-to-strong positive correlation was observed between UPI convenience and discretionary spending (r =. 558, t(153) = 8.32, p <. 001, 95% CI [.439, .658]). Hypothesis 1 is supported. The coefficient of determination (r² ≈. 31) indicates that UPI convenience accounts for approximately 31% of variance in discretionary expenditure, establishing it as a substantively meaningful predictor consistent with payment friction models (Prelec and Loewenstein, 1998; Dev et al., 2024) [11,5] .

H2: Impulse Spending → Discretionary Spending

The strongest association in the study was found between impulse spending tendency and discretionary expenditure (r =. 584, t(153) = 8.90, p <. 001, 95% CI [.470, .679]). Hypothesis 2 is supported. With r² ≈. 34, impulse tendency is the single largest individual predictor, explaining approximately 34% of the dependent variable variance. This result strongly supports the impulse-buying literature and replicates findings from Balaji et al. (2025) [1] among Indian university populations.

H3: Peer Influence → Discretionary Spending

Peer influence demonstrated a significant positive association with discretionary spending (r =. 318, t(153) = 4.15, p <. 001, 95% CI [.169, .453]). Hypothesis 3 is supported. The comparatively modest effect size (r² ≈. 10) suggests peer influence functions as a contextual amplifier of the primary drivers, consistent with Rogers' (2003) [12] diffusion model and Bhanushali's (2025) [2] findings on social normalisation of digital payments.

H4: Financial Awareness → Discretionary Spending

Contrary to Hypothesis 4, financial awareness did not yield a statistically significant association with discretionary spending (r =. 119, t(153) = 1.48, p =. 140, 95% CI [-.039, .272]). Hypothesis 4 is not supported. The 95% confidence interval spans zero, precluding directional inference. Possible explanations include the near-threshold reliability of Section F (α = 0.69), potential range restriction given the relatively high sample mean (M = 3.46), and the possibility that awareness exerts a moderating rather than direct effect, a mechanism that bivariate correlation analysis is structurally incapable of detecting.

Discussion 9.1 The Cashless Cushion Effect: Evidence and Mechanisms

These empirical results provide systematic quantitative evidence supporting the validity of the cashless cushion effect amongst the undergraduates. First, the highly significant relationship between the convenience of the UPI system (r =. 558) and the frequency of impulse spending (r =. 584) provides a clear indication that decreases in transaction friction, which is nothing but the underlying architecture of the UPI system, lead to increases in expenditure that are statistically significant. Together, these two variables explain a good proportion of the variance in the dependent variable, implying that the cashless cushion effect can be operationalized using not only the dispositional but also the structural mechanisms. Structurally, this finding is important for our theoretical understanding of the cashless cushion effect in terms of the fact that this effect can be extended beyond simply reflecting the users' impulsivity but rather can operate via digital platforms. The architectural attributes of the UPI system, such as seamless transactions, QR codes, immediate confirmations, and even reward systems, can be seen as factors that contribute towards facilitating spending regardless of individual heterogeneity in dispositional terms.

Impulse spending as a main predictor

The evidence of the high impact of impulse buying proclivity as the best predictor (r=.584, r²≈.34) shows that UPI-based social media provide a more favourable condition than others for the development of existing impulsive inclinations. UPI becomes a context where for students having inherent tendencies toward impulse behaviours the main constraints on their behaviours, such as the possession of money, the need to physically perform transactions, and time delays, are eliminated. Implications for policy are consistent with that perspective since diminishing the cashless cushion effect requires structural modifications of the interface as well as changes in individual dispositions through financial mindfulness training and automatic default settings for saving accounts.

Peer Influence: Contextual Amplifier

The evidence of peer influence is statistically significant, even though it is moderate (r =. 318). Thus, one can conclude that the social diffusion process is contributing to the phenomenon of UPI-induced discretionary spending, although its effect is not comparable to the influence of structural convenience and individual impulsivity. The fact that the college setting involves relatively dense clusters of residential spaces, group consumption patterns, and closeknit networks of peers makes it rather close to the optimal conditions of social normalization processes suggested by Rogers (2003) [12] and Bhanushali (2025) [2] . Approximately 10% of the variation explained by the impact of peer influence implies that it is an amplifying element of the key forces behind discretionary spending behaviours.

Exploring the null result for Financial Awareness

In connection with financial awareness, there are good grounds to be attentive to the obtained result (r =. 119, p =. 140). The possible reasons behind the lack of results include, but are not limited to, the following: first, the low level of reliability of the construct (α = 0.69) may be assumed to contribute to measurement errors, thus lowering the real value. The further use of more sophisticated scales (Lusardi and Mitchell, 2011) [9] or measures suggested by OECD for financial literacy might produce different outcomes. Second, awareness related to budgeting through the use of digital devices (as defined in section F) differs conceptually from the actual level of financial literacy that Kanwal and Singh (2025) [8] see as preventive: a person might track his/her spending pattern in an application yet fail to regulate the spending behavior. Third, due to relatively large values of the variable (M = 3.46), the number of participants was not sufficient enough. Finally, theoretically, the construct does not have a direct influence but rather operates as a structural moderator that lessens the relationship between convenience and impulse spending.

Kinds of common spending

Descriptive findings for specific categories reveal the exact points at which the CFCE is likely to arise. Discretionary expenses relating to street food (M=3.63), travel (M=3.48) and e-shopping (M=3.41) can be noted as some of the areas where high frequency of discretionary spending becomes apparent, and where spending on the UPI system is typical according to the description of this model involving microtransactions. The described features make the individual transaction too small to reach the level of financial monitoring while allowing the consumer to spend enough money during the time period to result in overall significant expenditure. These findings correspond to Dev et al.'s (2024) [5] description of such spending being frequent and repetitive.

Implications for practice and policy

At the interface level, i.e., between UPI interface designers and the NPCI, it has been observed that convenience-based design, without the presence of any constraints pertaining to spending visibility features, may inadvertently result in greater levels of discretionary overspending by younger consumers. Inclusion of real-time dashboard display of spending totals, spending categorization, and self-imposed spending limits may reinstate the psychologically relevant feedback that is lacking in current UPI designs (Dev et al., 2024) [5] . From an institutional perspective, these results indicate that education modules within undergraduate studies should incorporate greater emphasis on cashless payment psychology as opposed to general budgeting techniques. Since no evidence is found for higher financial literacy with current measures of operationalization, efforts should be made to increase awareness about the mechanisms underlying the cashless cushion effect during such learning experiences. For policymakers, there exists evidence for greater protection of consumers against incentives designed into UPI applications (such as cash back and game-based rewards) that boost spending behaviors, and such incentives are already a part of the current environment and existing literature.

Conclusion

Quantitative support of the cashless cushion effect is provided here among Indian undergraduates. Descriptive statistics, Cronbach's α (α = 0.87), and Pearson correlation tests were performed on N = 155 respondents, who represent a proportionally stratified random sample (based on Cochran's finite-population correction) derived from a population size of 744 across four strata (34, 39, 39, and 43). The results show that convenience of UPI use and impulse tendency are significant determinants of discretionary spending, with peer pressure contributing an extra dimension. In this regard, awareness turned out to be non-significant. Three contributions are made here. Firstly, theoretically, a basis of the validity of the cashless cushion concept as an economic-behavioural construct can be established, particularly in view of its unique architecture and sociological environment in UPI-dominated Indian universities. Secondly, methodologically, it is demonstrated www.extensionjournal.com that proportionate stratified sampling paired with a multisection Likert-scale questionnaire with high internal reliability (α = 0.87) is effective enough for finance behaviour research at undergraduate level. Thirdly, practically, usable information is generated for interface designers, financial engineers, and policymakers concerned about consumer wellbeing in India's cashless transaction era. Future research can build further on these results. Specifically, a structural equation model can be incorporated for testing direct and moderating effects, longitudinal studies conducted to observe how expenditures change through payment-mode changes, and experiments run in manipulating interface designs to establish causations. Also, data could be examined in crossinstitutional, socio-economic stratum, regional, and genderrelated dimensions to explore cashless cushion boundaries.

References

1. Balaji R, Karthik P, Bhuvan P. Digital payment apps and impulse buying: Analysing spending behaviours among university students. EPRA International Journal of Multidisciplinary Research. 2025;11(4). https://eprajournals.com
2. Bhanushali AS. Impact of UPI on spending habits: A comparative study of traditional and modern payment methods in Panvel. Zenodo. 2025. https://zenodo.org
3. Cochran WG. Sampling techniques. 3rd ed. Wiley; 1977.
4. Creswell JW. Research design: Qualitative, quantitative, and mixed methods approaches. 4th ed. SAGE Publications; 2014.
5. Dev H, Gupta R, Dharmavaram S, Kumar D. From cash to cashless: UPI's impact on spending behavior among Indian users and prototyping financially responsible interfaces. arXiv. 2024. https://arxiv.org
6. Feinberg RA. Credit cards as spending facilitating stimuli: A conditioning interpretation. Journal of Consumer Research. 1986;13(3):348-356. https://doi.org/10.1086/209074
7. Hair JF, Black WC, Babin BJ, Anderson RE. Multivariate data analysis. 8th ed. Cengage Learning; 2019.
8. Kanwal J, Singh J. Impact of digital payment systems on consumer spending habits. European Economic Letters. 2025;15(2s):54-69. https://eelet.org.uk
9. Lusardi A, Mitchell OS. Financial literacy around the world: An overview. Journal of Pension Economics and Finance. 2011;10(4):497-508. https://doi.org/10.1017/S1474747211000448
10. Moirangthem R, et al. Exploring the spending patterns and payment preferences: A cross-sectional survey of postgraduate students in a transforming digital economy. Advances in Consumer Research. 2026;3(2):1367-1375. https://acr-journal.com
11. Prelec D, Loewenstein G. The red and the black: Mental accounting of savings and debt. Marketing Science. 1998;17(1):4-28. https://doi.org/10.1287/mksc.17.1.4
12. Rogers EM. Diffusion of innovations. 5th ed. Free Press; 2003.
13. Rook DW. The buying impulse. Journal of Consumer Research. 1987;14(2):189-199. https://doi.org/10.1086/209105
14. Soman D. The effect of payment transparency on consumption: Quasi-experiments from the field. Marketing Letters. 2003;14(3):173-183. https://doi.org/10.1023/A:1027444717586