Impact of Digital Payments and Regional Disparities in Digital

Adoption in India

Author - Khyati Kothari, SAP ID – 76022200029

Mentor – Mrs. Esha Khanna
College – Sarla Modi School of Economics, NMIMS, Mumbai, India
Date – 1st April’2024

SYNOPSIS

The research paper investigates the impact of digital payments on India's economic growth, with a particular focus on
regional disparities in digital adoption. It outlines India's transformative journey fueled by digital technologies, notably the
India Stack framework and the introduction of the Unified Payments Interface (UPI). Through multi-linear and panel
regression analysis, cluster analysis, and visualization techniques, the study examines transaction data from sources such as
the RBI Bulletin and PhonePe Pulse. Findings reveal significant relationships between GDP and specific digital payment
indicators, emphasizing the pivotal role of UPI and Aadhaar-linked payments in driving economic growth. Regional
variations in digital payment adoption are explored, shedding light on factors such as internet penetration and ATM
accessibility. The paper concludes by proposing targeted strategies for policymakers and stakeholders to address regional
disparities, enhance digital infrastructure, and foster inclusive economic growth.

1
EXECUTIVE SUMMARY

This research paper delves into the transformative impact of digital payments on India's economic growth, with a specific
focus on the emergence of the Unified Payments Interface (UPI) and regional disparities in digital payment adoption.
Through a multi-linear regression analysis spanning two distinct phases, the study uncovers significant relationships between
GDP and various digital payment indicators. In the first phase, covering the period from 2014 to 2020, the analysis reveals a
positive association between economic growth and Prepaid Payment Instruments (PPI). Conversely, the second phase,
spanning from the third quarter of 2020 to the third quarter of 2024, expands the analysis to include a broader array of digital
payment methods, showcasing the increasing prominence of retail credit transfers linked to UPI and Aadhaar. Additionally,
the study examines regional disparities in digital payment adoption using transaction data from PhonePe Pulse, highlighting
variations in transaction volumes across different states. By incorporating factors such as Per Capita Gross State Domestic
Product, internet subscribers, ATM deployment, Male/Female Urban/Rural unemployment rate into panel regression
analysis, the research elucidates their impacts on transaction volumes and underscores the role of socio-economic factors and
banking infrastructure in driving inclusive economic growth. The findings of this study offer valuable insights for
policymakers, financial institutions, and stakeholders, emphasizing the importance of tailored strategies to address regional
disparities and promote digital inclusion in underrepresented areas. Additionally, the paper identifies areas for further
research, particularly in exploring the influence of demographic factors on digital payment adoption and devising more
granular strategies to enhance financial inclusion at the grassroots level.

2
TABLE OF CONTENTS

INTRODUCTION 4
RESEARCH RATIONALE 5
LITERATURE REVIEW 6
IMPACT OF DIGITAL PAYMENTS ON ECONOMIC GROWTH OF INDIA 7
ANALYSIS OF REGIONAL DISPARITIES IN DIGITAL ADOPTION IN INDIA 11
IMPACT OF SOCIO-ECONOMIC FACTORS ON UPI TRANSACTION VOLUME STATE-WISE: 13
CLUSTER ANALYSIS OF STATES 14
CONCLUSION 15
RESEARCH LIMITATIONS AND FURTHER SCOPE 16

3
INTRODUCTION

Technological advancements are crucial to narrow the digital gap both domestically and globally, as this divide not only
hampers economic growth but also exacerbates income inequality. Ensuring equal access to digital services is underscored
by the UNDP Sustainable Development Goals. Despite significant strides, over 4 billion people, predominantly from
developing nations, lack internet access, which necessitates enhanced digital financial inclusion. Jim Yong Kim, President of
World Bank Group, in October 2014 said that “Digital Financial inclusion involves the deployment of the cost-saving digital
means to reach currently financially excluded and underserved populations with a range of formal financial services suited to
their needs that are responsibly delivered at a cost affordable to customers and sustainable for providers.” (Nandru,
Chendragiri, & Velayutham, 2021)

In the last decade, India has witnessed a monumental shift propelled by digital technologies, fundamentally altering the
landscape of economic growth and financial inclusion. At the heart of this transformation lie three interwoven pillars:
identity, payments, and data empowerment (India Stack, 2024). This research paper delves into the economic growth
implications of technological advancements in the digital payments system with the emergence of the Unified Payments
Interface. However, despite these advancements, regional disparities persist in terms of digital payment adoption and usage
across different parts of the country. The paper utilized the transaction data from PhonePe Pulse to uncover the underlying
factors driving disparities in digital payment adoption and assess their implications for financial inclusion and economic
development.

The journey for digital financial inclusion began in 2009 by addressing the challenge of access, aiming to integrate over a
billion individuals into the formal financial system. Within a span of eight years, India witnessed a remarkable rise, with 330
million new bank accounts, defying conventional timelines by achieving in years what would have taken decades. Central to
this paradigm shift was the strategic deployment of technology infrastructure, collectively known as the India Stack. This
framework encompasses identity solutions, exemplified by Aadhaar, a unique identification system that has empowered
individuals to assert their identity securely. Aadhaar's role transcends mere authentication, extending to e-KYC protocols,
enabling swift onboarding onto financial platforms (Swallow, Haksar, & Patnam, 2021). The Pradhan Mantri Jan Dhan
Yojana in 2014 further boosted bank account accessibility, equalizing urban and rural banking access from 2014 to 2017
(urban: 59% to 76%, rural: 52% to 79%). (Misra & Tankha, 2018)

Complementing identity solutions, the foundation of Unified Payments Interface (UPI) revolutionized the digital payment
system of India. Launched in 2016, UPI swiftly emerged as a cornerstone, facilitating seamless person-to-person, person-to-
merchant, and business-to-business transactions, fostering a vibrant digital payments ecosystem. The UPI transactions grew
by 450% between January 2018 and January 2019, surpassing debit cards and prepaid instruments in transaction volume.
Currently, digital payments account for over 40% of all transactions conducted in India, with the UPI dominating the
landscape. UPI is utilized by more than 300 million individuals and over 50 million merchants across the country. In terms of
global rankings, India leads the pack with the highest volume of digital transactions, claiming nearly 46% of the market
share based on data from 2022. Following India are countries like Brazil, China, Thailand, and South Korea. (RBI, 2023)

The expansion of affordable smartphones and low-cost internet access has been a major driver of digital financial inclusion
in India. Jio's entry into the mobile data market triggered price wars and increased 4G internet availability, bringing millions
of rural Indians online for the first time. This has fueled the use of digital financial services, with Paytm reporting operations
in over 300,000 villages. In terms of digital payments, the transaction value in India has increased from 2,071 crore in FY
2017-18 to 13,462 crore in FY 2022-23 growing at a 45% CAGR. The transactions have already reached 11,660 crores till

4
11th December 2023 for the FY 2023-24 (RBI, 2023). One of the major factors that have pushed India towards this shift is
demonetization, while Covid-19 added further momentum to this growing industry.

RESEARCH RATIONALE

The primary objective of this research paper is to understand the change in the impact of digital payments on economic
growth with the emergence of Unified Payments Interface (UPI). Further, the paper attempts to understand the regional
variations in the adoption of digital payment in India with the help of PhonePe transactions. A noticeable gap exists in
studies analyzing the equality of digital adoption across states, thereby hindering a nuanced understanding of the socio-
economic dynamics shaping digital financial inclusion. This research paper addresses the differential effects of UPI
implementation in diverse geographical contexts, impeding the identification of localized barriers and opportunities for
equitable digital payment uptake.

First part of the paper contains:

a. Impact of Digital Payments on Economic Growth (2014-2020):
H1: There is a significant positive relationship between the adoption of digital payments and economic growth in India from
2014 to 2020.
b. Impact of Digital Payments on Economic Growth (2020-2024):
H2: Following the introduction of UPI in 2016, there is a significant increase in the positive relationship between digital
payments and economic growth in India from 2020 to 2024 compared to the period before UPI.

Second part of the paper contains:

Analysis of Regional Disparities in Digital Adoption in India

Third part of the paper contains:

Impact of per capita income, internet subscribers, financial inclusion, and employment on UPI transaction volume state-wise:
H3: Higher income, internet, ATMs, employment in a state positively influences the volume of UPI transactions conducted
within that state.

Forth part of the research contains:

Cluster Analysis of States based on Transaction Volume, Income, Internet, and ATMs

LITERATURE REVIEW

(Kosalai, 2017) emphasizes on the significance of cashless India, viewing it as a strategic move towards
modernizing the payment system, enhancing employment opportunities, and attracting foreign investments, with the
reduction of risks and inefficiencies associated with cash transactions. Digital payment is expected to not only diminish cash-
related robberies and corruption, but it also lowers the costs of banking services, improving the banking industry’s
performance. (Sreenu, 2020) uses causality model for period between 2010-2018 to find that use of a cashless policy on
Indian economic development in the short term will be negative, whereas in the long term it will impact positively. Further,
(Rooj & Sengupta, 2020) using a multivariate Bayesian vector autoregressive finds a positive relation between RTGS and
economic growth of India with both the variables impacting each other. (Agarwal, 2023) have also showed that all variables

5
including RTGS, CCIL, and Retail electronic clearing with the exception of Cards and PPI have a positive relationship with
the GDP and have a long-run equilibrium.

Further, the government initiatives are integral to advancing India towards a digital economy by promoting financial
inclusion and bringing more individuals into the formal banking system. The push towards this direction is supported by the
expansion of India's digital payments ecosystem, exemplified by the introduction of platforms such as UPI and the Bharat
Interface for Money (BHIM) app (Sethi, 2024). UPI has significantly contributed to India's journey towards self-reliance by
reducing dependence on international payment solutions and positioning itself as a crucial instrument in achieving the vision
of Atmanirbhar Bharat. Despite long-standing efforts by the government to promote digital payments, the emergence of UPI
has become a major transformative force in accelerating this transition to digital payment adoption. Furthermore, (Sharma,
Kanoujiya, Bhimavarapu, & Rastogi, 2022) also finds that financial inclusion impacts UPI since improved access and greater
awareness encourages people to transition into new payment.

(Khuntia, 2014) has claimed PMJDY to be “a big bang action plan” which would reduce the level of financial
untouchability, fight poverty, accelerate growth and ensure that even the last person standing in the last row can be
empowered in the Indian economy. (Rastogi, Panse, Sharma, & Bhimavarapu, 2021) further found that the UPI does not only
facilitate secure transactions for PMJDY account holders, but it also plays a significant role in integrating them into the
financial ecosystem, thereby fostering economic development. Additionally, Direct Benefit Transfers, introduced to
efficiently transfer subsidies of various government schemes through Aadhaar linked bank accounts, significantly reduced
the problems of leakage and corruption. Moreover, by leveraging on Digital Public Infrastructure (DPI) initiatives, the PFMS
was able to seamlessly transfer funds to nearly 500 million beneficiaries during the COVID-19 outbreak, enhancing cash
transfer programs’ budget execution responsiveness and efficiency (Alonso, et al., 2023). In line with this, (Chhabra &
Khandelwal, 2021) also came up to a conclusion that government initiated digitalized Yojanas have significant impact on
financial inclusion in India, emphasizing on its significance for inclusive growth since affordable access of financial services
benefits all, particularly the unprivileged population.

The research paper by Observer Research foundation projecting nationwide and state-wise UPI market, concluded that
financial inclusion underscores the digitalization of payment system, emphasizing on the importance of improvement in
digital literacy and digital inclusion for the growth of digital transactions (Ghosh & D'souza, 2023). Further, the paper by
(Rastogi & Damle, 2020) proves that mobile phones, internet explosion and new bank accounts are estimated to be the main
digital payments enablers. Moreover, (Raichoudhury, 2020) through econometric analysis utilizing a panel data structure of
28 states of India over four years found that income, infrastructure, and employment opportunities are perhaps the most
important determinants of financial inclusion. The Unified Payments Interface (UPI) in India stands as a beacon for financial
inclusion, revolutionizing digital transactions and promising a brighter future ahead (Faster capital, 2024). The results of the
GMM model by (Azimi, 2022) clearly indicated that financial inclusion has a significantly positive impact on the economic
growth, implying that financial inclusion is an effective tool in fostering rapid economic growth in the world. Further, the
paper by (Ravikumar, 2019) proves that fintech companies have changed the face of payments, credit, remittances, and
insurances leading digital financial inclusion from the front using their speed, convenience, and indiscriminative approach.

IMPACT OF DIGITAL PAYMENTS ON ECONOMIC GROWTH OF INDIA

DATA AND METHODOLOGY:

6
The study investigates the causal relationship between India's economic growth and its digital payment systems through
multi-linear regression analysis. The analysis is divided into two phases: the first phase analyzes quarterly digital payment
trends spanning from 2014 to 2020, while the second phase focuses on the period from the third quarter of 2020 to the third
quarter of 2024. The reason for the second phase starting from 2020 and not 2016, that is, when UPI was introduced, is due
to the granular data availability in the RBI Bulletin. The digital payments analysis for the first phase includes values of
“Real-Time Gross Settlement (RTGS)”, “Cheque Truncation System (CTS)”, “Retail Electronic Clearing (REC)”, “Interbank
Mobile Payment Services (IMPS)”, “National Automated Clearing house (NACH)”, “Card payments”, “Prepaid Payment
Instruments (PPI)”, and “Mobile Banking”.

The data for the digital payments has been gathered from the RBI Bulletin while the data for GDP is acquired from RBI
Public Database (DBIE). “RTGS” that comprises of customer and interbank transactions is a specialized electronic funds
transfer system used by banks and financial institutions for high-value and time-sensitive transactions. While customer
transactions refer to fund transfers initiated by individuals or businesses through their bank's online or mobile banking
platforms, interbank transactions refer to fund transfers between banks or financial institutions. “CTS” is a process used by
banks for faster and more efficient clearing of cheques where instead of physically moving paper cheques from one bank to
another, only digital images of the checks are transmitted electronically. “REC” comprises of Debit Electronic Clearing
Service (ECS DR), Credit Electronic Clearing Service (ECS CR incl. NECS) and Electronic Funds Transfer/National
Electronic Funds Transfer (EFT/NEFT) where ECS DR/ECS CR is a system used for making/receiving bulk recurring
payments, such as salary, dividend, or utility bill. ECS CR also includes National Electronic Clearing Service (NECS)
introduced by National Payments Corporation of India (NPCI) which is an enhanced version of this infrastructure. Further,
EFT/NEFT are electronic payment systems used for transferring funds from one bank account to another within India.
“IMPS” is a real-time payment system that enables individuals to transfer funds instantly between bank accounts using
mobile phones, introduced by NPCI. In addition, “NACH” facilitates efficient and automated processing of bulk and
repetitive electronic transactions. The card payments comprise of credit cards and debit cards measured by their usage in
ATMs and POS. “PPI” include m-wallet and PPI cards with the former being digital wallets while latter being physical pre-
loaded cards. “MB” refers to the use of smartphones or tablets to access and manage banking services remotely.
Furthermore, “GDP” at constant prices has been taken as a proxy for measuring economic growth. This Time-Series data is
aggregated on a quarterly bases, aligning with financial year. Digital payment figures, initially recorded on a monthly basis,
have been aggregated into quarterly periods to match the financial year intervals.

The variables in the second phase include values of “Credit Transfers – RTGS”, “Credit Transfers – Retail”, “Debit Transfers
and Direct Debits”, “Cards Payments”, “Prepaid Payments Instruments”, “CTS (NPCI Managed)”, “Mobile Payments (app
based)” and “Internet Payments (Netbanking/Internet Browser Based)”. From November 2019, the format of RBI payment
system indicator changed, including more granular data in the collection. The “Credit Transfers – Retail” comprises of AePS
(Fund Transfers), APBS, IMPS, NACH Cr, NEFT, UPI where AePS is Aadhaar enabled payment system facilitating
Aadhaar-linked transactions through biometric authentication. Further, APBS is Aadhaar Payment Bridge System that takes
aid of Aadhaar for electronically channelizing the Government subsidies and benefits to the intended beneficiaries.
Moreover, Unified Payments Interface (UPI) is a part of this structure which is an instant real-time payment system to
facilitate inter-bank transactions through mobile phones. “Debit Transfers and Direct Debits” comprises of BHIM Aadhaar
Pay, ECS Dr and NACH Dr where BHIM Aadhaar Pay is a service that allows merchants to receive digital payments over
the counter through Aadhaar identification.

FIRST PHASE METHOD

7
For the initial phase, the payments data is first normalized by conversion from INR billion to INR crores to align with the
scale of GDP data. Subsequently, the missing values and outliers are identified and removed to avoid errors in the result.
Before proceeding with the regression analysis, unit root test is conducted to check for stationarity through Augmented-dicky
fuller test (ADF) in order to avoid spurious regression results. To enhance interpretability and address potential issues with
data distribution, the entire dataset is transformed using the natural logarithm. Due to spurious results in both the log-
transformed and first-differenced data, further analysis is performed by taking the second difference.

Unit Root Test – Augmented Dicky Fuller Test

first difference second difference
Variables t-statistic p-value t-statistic p-value t-statistic p-value
LRTGS -3.12 0.15 -7.39 0.01 -9.76 0.01
LCTS -3.00 0.19 -5.29 0.01 -5.65 0.01
LREC -1.02 0.92 -4.59 0.01 -12.93 0.01
LIMPS -0.94 0.93 -3.14 0.14 -2.22 0.49
LNACH -1.38 0.81 -4.45 0.01 -5.01 0.01
LCARDS -2.03 0.56 -2.52 0.37 -3.32 0.09
LPPI -1.32 0.83 -3.59 0.05 -4.78 0.01
LMB -2.16 0.51 -2.71 0.30 -3.79 0.04
LGDP -4.34 0.01 -7.05 0.01 -8.30 0.01

Moreover, multicollinearity is assessed between independent variables using Variance Inflation Factor (VIF) to get better
results. It was found that all VIF values are below 10 indicating that there is no significant multi-collinearity present in the
data.

Multicollinearity Test - VIF

RTGS CTS REC IMPS NACH CARDS PPI MB
3.960399 2.782649 5.671041 2.064792 1.952607 1.834266 1.976027 1.518598

Firstly, all variables are taken for conducting multi-linear regression. The results indicate that there is a significant
relationship between GDP and PPI, statistically significant at 5% level, suggesting PPI has a positive impact on GDP of
India. Apart from this, other digital payment variables such as RTGS, CTS, REC, IMPS, NACH, CARDS, and MB do not
exhibit statistically significant relationships with GDP, as indicated by their p-values being greater than the conventional
threshold of 0.05 but the model is significant at 5%. But due to non-stationarity of two of the variables “IMPS” and
“CARDS”, another regression is conducted removing these two variables. The formulated multiple regression model has
been presented below:

 = 0 + 1 + 2 + 3 + 4 + 5 + 6 + 

Whereby Y is the second-difference of log of real gross domestic product for India, 0 is the intercept, 1 is the second-
difference of log of RTGS for India, 2 is the second-difference of log of CTS for India, 3 is second-difference of log of
Retail electronic clearing for India, 4 is second-difference of log of NACH for India, 5 is the second-difference of log of
PPI for India and 6 is the second-difference of log of Mobile Banking for India at the time t.

8
Multiple-linear Regression

The final results are better with Adjusted R-square improving from 0.4862 to 0.5593. The estimated coefficient for PPI is
0.144572, implying that a one-unit increase in PPI is associated with an increase of approximately 0.144572 units in GDP,
holding other variables constant.

SECOND PHASE METHOD

For the second phase, same analysis is conducted but this time the data is more granular including e-commerce transactions
and digital payments like FASTags and digital bills. Here, normalization of data is not needed since both the data are
available in crores. Initially, unit root test through ADF is conducted.

9
Unit Root Test – Augmented Dicky Fuller Test
first difference second difference
Variables t-statistic p-value t-statistic p-value t-statistic p-value
LRTGSC 2.9343 0.01873 -2.5048 0.008351 -4.6628 4.23E-05
LCT 3.136 0.02705 -2.6113 0.03959 -3.9552 0.003407
LDT 6.053 0.0002142 -2.0119 0.08268 -5.9808 0.0004946
LCP 1.5924 0.2244 -7.8237 2.91E-05 -11.5983 1.56E-06
LPPI 2.1041 0.1311 -1.6349 0.004184 -2.9631 0.0004736
LCTS 1.2471 0.07281 -3.758 0.001719 -4.1202 0.001084
LMB 3.6039 0.003738 -1.3893 0.03827 -3.7535 0.0002777
LNB 2.0002 0.1728 -1.8287 0.03352 -3.0797 0.0005406
LGDP 2.5503 0.06704 -3.7187 0.009828 -4.2748 0.005127

Some of the payment indicators including card payments, PPI, CTS, Net Banking and GDP seems to not be stationary, which
is why first difference data is taken for conducting regression analysis. Before proceeding to regression analysis, multi-
collinearity test is conducted on the dataset to find that there is a significant correlation between all the independent
variables. This is why after some trial and error, only those variables are taken which removes the problem of multi-
collinearity. The final list of independent variables taken is shown in the table below:

Multicollinearity Test - VIF

RTGSC CT DT PPI NB
5.970877 7.249886 4.918672 2.757948 2.092761

The formulated multiple regression model has been presented below:

 = 0 + 1 + 2 + 3 + 4 + 5 + 

Whereby Y is the first-difference of log of real gross domestic product for India, 0 is the intercept, 1 is the first-difference
of log of RTGSC for India, 2 is the first-difference of log of Credit transfers – retail for India, 3 is first-difference of log of
Debit transfers for India, 4 is first-difference of log of PPI for India and 5 is the first-difference of log of net banking for
India at the time t.

Multiple-linear Regression

10
The results of the multi-linear regression analysis indicate that retail credit transfers have a significantly positive impact on
economic growth of India, with the relationship statistically significant at 1% level. The coefficient of 0.6272330 shows that
for each unit increase in credit transfers, GDP is expected to increase by approximately 0.6272330 units, holding other
variables constant. Apart from this, other variables do not seem to have a significant relationship with GDP since their p-
values are greater than the conventional threshold of 0.05, but the model is significant at 5%. But given that the adjusted r-
square of the new model is as high as 0.9432, it can be said that the model represents some potential issues that might be
relating to overfitting, multicollinearity, or small sample size of the data. Further, Breusch-Pagan test was conducted for
better validity of the model where since the p-value (0.2901) was greater than the commonly used significance level of 0.05,
we failed to reject the null hypothesis of homoscedasticity.

RESULTS

Given the comprehensive coverage of retail credit transfers in the analysis, it's reasonable to conclude that the increased
impact observed in the second phase is attributed to the wider adoption and utilization of digital payment methods linked to
UPI and Aadhaar across various retail transactions. Furthermore, the significant improvement in the model's explanatory
power, as indicated by the higher adjusted R-squared value, reaffirms the growing importance of digital payments as a driver
of economic growth. Overall, the inclusion of credit transfers in retail transactions from various channels underscores the
increasingly vital role of digital payments in shaping India's economic landscape.

This is when other digital payments do not show a significant relationship with economic growth. In comparison to the last
model which showed significant relation between PPI and economic growth, the new model is not consistent with the same
since the p-value is higher than the conventional threshold of 0.05. This suggests that different digital payment indicators
may have varying impacts on GDP over time or under different circumstances. This is in line with the result by the paper
(Ravikumar, Suresha, Sriram, & Rajesh, 2019) that showed that Digital payments at large and retail electronic payments do
not contribute to the economic growth in India directly in the long run. Moreover, in analyzing the impact of various digital
payment methods on economic growth, it was observed that several payment indicators did not exhibit a direct and
statistically significant relationship with GDP. However, this does not necessarily imply a lack of influence on economic
growth, rather, it suggests that their impact might be indirect or mediated through other channels. While individual payment
methods may not directly correlate with economic growth, their collective effects could manifest through mechanisms such
as increased financial inclusion, enhanced efficiency in financial transactions, reduced transaction costs, or improved overall
productivity in the economy which will impact the economic growth in the long run rather than short term as analyzed in this

11
paper. Therefore, while certain payment methods may not show a direct association with GDP, it is imperative to consider
the broader implications and potential indirect pathways through which digital payments might contribute to economic
growth.

ANALYSIS OF REGIONAL DISPARITIES IN DIGITAL ADOPTION IN INDIA

The dataset used for analysis captures quarterly transactions facilitated by PhonePe across all states in India from 2018 to
2022, encompassing various payment modes such as merchant payments, peer-to-peer payments, recharge and bill payments,
financial services, and others. The data is sourced from PhonePe Pulse, an open public API. According to NPCI data, as of
December 2023, PhonePe has 46% share in UPI transaction volumes and 48% share in UPI transaction value (Mishra &
Mukul, 2024). Given its significant market position, analyzing this data would offer valuable insights into the regional
dynamics of digital payment adoption across India. Firstly, the position of states in terms of total volume and value is
captured using plots. Subsequently, digital adoption in states has been visualized to capture the growth of transactions over
the five years studied.

Plot 1 and Plot 2 – Value and Volume of Digital Transactions

12
Plot 1 represents the total transactions by value, showing highest transaction value in Telangana, Maharashtra, Karnataka,
Andhra Pradesh, Rajasthan, and Uttar Pradesh while the lowest transactions can be seen in the UTs and Northeastern parts of
India. Further, Plot 2 shows that in volume, Maharashtra and Karnataka overtakes Telangana while the rest of the list remain
similar. Furthermore, to understand the digital payment adoption in all states, total transactions by value are summarized
with state and year using ggplot.

Plot 3 – Digital Payment Adoption Over Time by States

Plot 3 indicates that states like Delhi, Goa, Gujarat, Haryana, Himachal Pradesh, Jammu and Kashmir, and Jharkhand has
increasing trend in the adoption of digital payments since the total volume of transactions in these states are rising over time.
In comparison, other states are quite still in comparison.

IMPACT OF SOCIO-ECONOMIC FACTORS ON UPI TRANSACTION VOLUME

13
To further understand the characteristics of regional variations in digital payment adoption, other factors have been
considered. Firstly, Per Capita Gross State Domestic Product at constant price has been taken as a proxy for estimating the
income effect. The data for this has been gathered through the RBI Database. Moreover, in order to estimate the internet
penetration in all states, internet subscribers per 100 population have been taken as a proxy from TRAI yearly performance
indicators report. Data for Andhra Pradesh includes Telangana, Madhya Pradesh includes Chhattisgarh, Bihar includes
Jharkhand, Maharashtra includes Goa, Uttar Pradesh includes Uttarakhand, West Bengal includes Sikkim and North-East
includes Arunachal Pradesh, Manipur, Meghalaya, Mizoram, Nagaland & Tripura states based on Service Area providers. As
a proxy to financial inclusion, the ATM deployment for each year has been collected from the RBI database. This is in
accordance with RBI’s recognition of ATM to be an efficient and cost-effective tool for the banks to increase financial
inclusion in India (Mukherjee, Thomas, & Goswami, 2024). Finally, unemployment rate per thousand population for both
male and female from rural and urban parts of the states has been gathered from the RBI database. A panel regression
analysis including 30 states from 2018-2022 has been conducted to understand the impact of per capita GSDP, internet
subscribers, ATM deployment and unemployment on the transaction count in PhonePe dataset. The formulated multiple
regression model has been presented below:

log(Transactionit) = αi + β1log(GSDPit) + β2Internet_Penetrationit + β3ATM_Deploymentit + β4MU_Empit + β5FU_Empit +

β6MR_Empit + β7FR_Empit + εit

Whereby Y is the log of transaction count, β1log(GSDPit) is the per capita GSDP, β2Internet_Penetrationit is the internet
subscribers per 100 population, β3ATM_Deploymentit is ATM deployment, β4MU_Empit is urban male unemployment rate,
β5FU_Empit is the urban female unemployment rate, β6MR_Empit is the rural male unemployment rate and β7FR_Empit is
the rural female unemployment rate.

This panel regression analysis investigates the factors influencing the log of transaction counts, utilizing a fixed effects
model to control for unobserved heterogeneity across states over a period. The coefficient of the log of GSDP is significantly
positive (estimate = 3.026, p-value = 0.041), indicating that as the economic size or per capita income of a state increases,
the transaction count also tends to increase. Moreover, there seems to be a highly significant positive relation (estimate =
0.077, p-value < 0.001) between internet penetration and transaction count, proving it to be a crucial factor in increasing

14
transaction counts. The positive coefficient for ATM deployment (estimate = 0.00068, p-value < 0.001) signifies that an
increase in financial infrastructure, as evidenced by more ATMs, is associated with higher transaction counts. Among the
unemployment variables, male rural unemployment is significantly negatively associated with transaction counts (estimate =
-0.013, p-value = 0.002), suggesting that higher unemployment in rural areas among males negatively impacts transaction
volume. The other employment variables, including male urban, female urban, and female rural unemployment, do not show
a significant impact on the transaction counts. The overall fit of the model is strong, with an R-squared value of 0.771,
indicating that approximately 77% of the variation in the log of transaction counts is explained by the model's independent
variables. The adjusted R-squared value of 0.697 adjusts for the number of predictors in the model, still indicating a good fit.
Further, the Breusch-Pagan test for heteroscedasticity yields a p-value of 0.515, suggesting that there is no significant
evidence of heteroscedasticity in the model, which means that the variance of the error terms is constant across the entities.

CLUSTER ANALYSIS OF STATES

In order to comprehend the characteristics of digital payment system across all the states of India, cluster analysis is
conducted by grouping them based on various attributes. This methodological approach allows for a nuanced understanding
of how distinct factors ranging from internet penetration to banking infrastructure influence digital payment adoption. By
identifying and analyzing these clusters, stakeholders can tailor their strategies, from developing targeted financial products
to designing specific regulatory policies, to boost digital payment adoption effectively. This insight aids in optimizing
resource allocation, fostering innovation in payment solutions, and ultimately enhancing the accessibility and efficiency of
digital financial services across diverse regions. In this paper, k-means cluster analysis has been conducted since the data
size is above 100. In addition, the optimal number of clusters was found to be 5. The machine learning model resulted into
the following five clusters:

Cluster Centers
Cluste Transaction_cou Per_Capita_GSDP Internet_Penetrati ATM_Depoloyme MU_Em FU_Em MR_Em FR_Em
r nt on nt p p p p

1 5351229494 153721 77.67 20114 49.6 69.6 27 13.6

2 32516094 128262.8 66.255 852.4 56.6 126.6 34 48.1

3 2675689878 75740.75 57.13 14913.5 61 87 32.7 21.7

4 1379768411 132767 86.76714 12950 43.85 59.8 48.2 16.5

5 407700647 102785.67 60.24333 5884.5 53.83 103 27.5 37.5

State-Wise Clusters
Cluster 1 Cluster 2 Cluster 3 Cluster 4 Cluster 5
Karnataka Arunachal Pradesh Andhra Pradesh Bihar Assam
Maharashtra Goa Madhya Pradesh Delhi Chhattisgarh
Telangana Himachal Pradesh Rajasthan Gujarat Jharkhand
Jammu & Kashmir Uttar Pradesh Haryana Kerala
Manipur Odisha Punjab
Meghalaya Tamil Nadu Uttarakhand
Mizoram West Bengal
Nagaland
Sikkim

15
 Tripura

The k-means clustering output shows the centers of the five clusters created from the dataset. Each cluster is characterized by
its average values across various features comprising of Transaction count, Per Capita GSDP, Internet Penetration, ATM
Deployment, and unemployment rate for the year 2022. Cluster 1 stands out with exceptionally high transaction counts,
reflecting a robust economic activity, accompanied by a high per capita Gross State Domestic Product (GSDP) and
significant internet penetration. This cluster also exhibits a substantial deployment of ATMs, indicating a mature financial
infrastructure. While urban unemployment rates are moderate, rural unemployment rates are lower with lowest rural female
unemployment rate as compared to other clusters. In contrast, Cluster 2 portrays regions with lower transaction counts and
moderate economic indicators. Per capita GSDP and internet penetration are moderate, and there is a lower deployment of
ATMs. Moreover, this cluster exhibits high levels of urban and rural unemployment, particularly female. All the
Northeastern states of India are a part of this cluster showing a significant need for investment in digital infrastructure in
these states.

Further, cluster 3 represents areas with moderate transaction counts and relatively lower economic development. Despite a
lower per capita GSDP and internet penetration, there is significant ATM deployment. Urban and rural unemployment rates
are moderate, indicating a balanced distribution of workforce across sectors. Seeing that there is a high ATM deployment in
these areas, it can be said that these states are more dependent on cash. Moreover, higher ATM deployment also shows better
financial inclusion so there is an opportunity to increase the digital payment adoption in these areas. Cluster 4 comprises
regions with moderate to high transaction counts, high per capita GSDP, and strong internet penetration. However, there is a
moderate deployment of ATMs. Urban unemployment rates are varied while Rural unemployment rates are moderate, albeit
with variations. Lastly, Cluster 5 reflects areas with low transaction counts and economic indicators. Per capita GSDP and
internet penetration are low, along with minimal ATM deployment. Unemployment rates are moderate, suggesting a mix of
economic activities with limited digital integration.

CONCLUSION

The paper delved into the short-term causal relationship between India's economic growth and its digital payment systems
through multi-linear regression analysis, spanning two distinct phases. The first phase, analyzing quarterly digital payment
trends from 2014 to 2020, revealing significant relationship only between GDP and PPI. Conversely, the second phase,
focusing on data from the third quarter of 2020 to the third quarter of 2024, expanded the analysis to include a broader array
of digital payment methods, showcasing the increasing prominence of retail credit transfers and their positive association
with economic growth. This association between retail transfers and economic growth shows the prominence of UPI and
Aadhaar linked payments in the digital payment industry.

Furthermore, the study unveiled regional disparities in digital payment adoption across India through an in-depth analysis of
transaction volumes facilitated by PhonePe, showing Telangana, Maharashtra, Karnataka, Andhra Pradesh, Rajasthan, and
Uttar Pradesh to have the highest UPI transactions. Using panel regression analysis, it was found the GDP per capita, internet
penetration, ATM deployment have significant positive impact on the number of transactions conducted in a state while rural
male unemployment has significant negative impact, showing that rural unemployment has stronger relationship with
transaction volume as compared to urban and female unemployment. Given their positive association with digital payment, it
is important to increase further investments in increasing internet penetration and banking infrastructure in all the states of
India, especially considering ‘Digital India’ initiative by the government.

16
Moreover, cluster analysis provided insights into the distinct characteristics of regions based on transaction count, Per Capita
GSDP, internet penetration, and ATM deployment, facilitating a nuanced understanding of digital payment dynamics across
diverse areas. The variations in the digital payment adoption calls for better internet penetration and banking infrastructure in
the Northeastern parts of India, Goa, Assam, Chhattisgarh, Jharkhand, Kerala, Punjab, and Uttarakhand. These findings offer
valuable implications for policymakers, financial institutions, and stakeholders, emphasizing the importance of tailored
strategies to address regional disparities, enhance digital infrastructure, and promote inclusive economic growth through
digital payment systems, particularly in underrepresented regions. Ultimately, the study contributes to the evolving discourse
on the role of digital payments in shaping India's economic landscape, highlighting opportunities for further research and
policy interventions in this domain.

RESEARCH LIMITATIONS AND FURTHER SCOPE

Due to last census being conducted in 2011, there was a huge issue in using demographic factors like gender, education, and
age in understanding the changing behavior of digital payment adoption. This can further help in creating better strategies in
addressing regional disparities. A research paper by (Nandru, Chendragiri, & Velayutham, 2021) make use of the world bank
findex data that measures the financial inclusion of all countries based on factors like Gender, age, income, education,
ownership of cards, bank accounts, usage of cards, received government subsidies through mobile phone, etc. found that
individual’s characteristics significantly determines the accessibility and usage of digital financial services. Therefore, in
order to increase the usage of digital payments, it is important to study the variations in demographic factors of state. This
can be further analyzed using districts which would need more granular level of data.

17
BIBLIOGRAPHY
RBI. (2023). Total digital payment transactions volume increases from 2,071 crore in FY 2017-18 to 13,462 crore in FY
2022-23 at a CAGR of 45 per cent: MoS Finance. Delhi: Ministry of Finance.

T, R. (2019). Digital Financial Inclusion: A Payoff of Financial Technology and Digital Finance Uprising in India.
INTERNATIONAL JOURNAL OF SCIENTIFIC & TECHNOLOGY RESEARCH, 3434-3448.

Mukherjee, A., Thomas, A., & Goswami, R. (2024). Roles of ATM in Financial Inclusion. New Delhi: Indian Council for
research on international Economic relations.

Kosalai, T. S. (2017). Impact And Importance Of Cashless Transaction In India. 3rd International Conference on Status of
Women in Cashless Economy (pp. 146-151). Tiruchengode: Shanlax International Journal Of Commerce.

Sethi, I. P. (2024). Digital Payments driving the growth of Digital Economy. Deputy Director General. NIC.

Sharma, A., Kanoujiya, J., Bhimavarapu, V. M., & Rastogi, S. (2022, September). Financial Inclusion - An Impetus to the
Digitalization of Payment Services (UPI) in India. Journal of Asian Finance, Economics and Business, 9(9), 191-203.

Khuntia, R. (2014). Pradhan Mantri Jan Dhan Yojana (PMJDY): A new drive towards financial inclusion in India. Zenith
International Journal of Business Economics & Management Research, 10-20.

Rastogi, S., Panse, C., Sharma, A., & Bhimavarapu, V. M. (2021, June 4). Unified Payment Interface (UPI): A Digital
Innovation and Its Impact on Financial Inclusion and Economic Development. Universal Journal of Accounting and Finance,
9(3), 518-530.

Alonso, C., Bhojwani, T., Hanedar, E., Prihardini, D., Uña, G., & Zhabska, K. (2023, March 31). Stacking up the Benefits:
Lessons from India’s Digital Journey. 2023(078), 50.

Chhabra, B. R., & Khandelwal, S. (2021). The Impact of Government Initiated Digitalized Yojanas on Financial Inclusion.
International Journal of Behavioural Science, 35, 7-12.

Ghosh, N., & D'souza, R. (2023). India’s UPI Market: Projections for Growth Under Various GDP Scenarios. Observer
Research Foundation. Occasional Paper.

Ravikumar, T. (2019, November). Digital Financial Inclusion: A Payoff of Financial Technology and Digital Finance
Uprising in India. International Journal of Science & Technology Research, 8(11), 3434-3448.

Rastogi, A., & Damle, M. (2020). TRENDS IN THE GROWTH PATTERN OF DIGITAL PAYMENT MODES IN INDIA
AFTER DEMONITIZATION. PalArch’s Journal of Archaeology of Egypt / Egyptology, 17(6), 4896 - 4927.

18
Raichoudhury, A. (2020, May 24). Major Determinants of Financial Inclusion: State-Level Evidences from India. Vision:
The Journal of Business Perspective.

Faster capital. (2024, March 2). Financial inclusion: Driving Financial Inclusion through UPI. Retrieved from Faster Capital:
https://fastercapital.com/content/Financial-inclusion--Driving-Financial-Inclusion-through-UPI.html#Future-Prospects-of-
UPI-in-Expanding-Financial-Inclusion

Azimi, M. N. (2022, November 17). New insights into the impact of financial inclusion on economic growth: A global
perspective. Plos One, 17(11).

India Stack. (2024). India Stack . Retrieved from https://indiastack.org

Swallow, Y. C., Haksar, V., & Patnam, V. (2021, July). A digital infrastructure known as the India Stack is revolutionizing
access to finance. Retrieved from IMF: https://www.imf.org/external/pubs/ft/fandd/2021/07/india-stack-financial-access-and-
digital-inclusion.htm

Misra, A., & Tankha, A. (2018). Inclusive Finance India Report 2018. New Delhi: ACCESS.

Nandru, P., Chendragiri, M., & Velayutham, A. (2021, March 30). Determinants of digital .

Ravikumar, T., Suresha, B., Sriram, M., & Rajesh, R. (2019, October). Impact of Digital Payments on Economic Growth:
Evidence from India. International Journal of Innovative Technology and Exploring Engineering, 8(12).

Mishra, D., & Mukul, P. (2024, Februrary 5). Paytm crisis brings 30% market share cap plan back in focus. Retrieved from
The Economic Times: https://economictimes.indiatimes.com/tech/technology/paytm-crisis-upi-market-share-cap-rule-in-
focus-leaders-may-corner-a-bigger-pie/articleshow/107406233.cms?from=mdr

RBI. (2023, October 30). India’s UPI : A global front-runner in digital payment systems. Retrieved from Special Service and
Features: https://pib.gov.in/PressReleaseIframePage.aspx?PRID=1973082&ref=indiatech.com

Sreenu, N. (2020, August 3). Cashless Payment Policy and Its Effects on Economic Growth of India: An Exploratory Study.
ACM Transactions on Management Information Systems, 11(3).

Rooj, D., & Sengupta, R. (2020). The real-time impact on real economy: A multivariate bvar analysis of digital payment
systems and economic growth in India. ADBI Working Paper Series No. 1128.

Agarwal, S. (2023). INDIA’S DIGITAL PAYMENT SYSTEM AND ITS IMPACT ON ECONOMIC GROWTH: AN
EMPIRICAL STUDY. AICTE Sponsored International Online Conference Proceedings on Future Trends in Business:
Knowledge, Skills, Sustainability, Innovation and Technology. Lucknow.

19
20