4.2 Trend Analysis of Digital Finance Infrastructure and Growth of Banking Firms in Nigeria
The data presented in Figure 4.1 illustrates the total assets of commercial banks in Nigeria spanning from 2009 to the projected year 2022. Notably, certain quarters exhibit gaps in reported values, either due to data unavailability or a lack of reporting for those specific periods. In this analysis, we aim to meticulously examine the accessible data and derive meaningful insights.
The dataset encapsulates the total assets of commercial banks in Nigeria over a 13-year timeframe, segmented by year and quarter. It's crucial to acknowledge the existence of data gaps, wherein numerous quarters lack reported values—potentially stemming from unavailable data or non-reporting during those specific periods. To facilitate a meaningful analysis, our focus will initially hone in on the available data points and their discernible trends.
2009 to 2012: Witnessing a consistent upswing in total assets during this phase, the figures ascend from 17,522.86 in Q1 2009 to 21,288.14 in Q1 2012, indicating a sustained growth trajectory. 2013 to 2015: This trend persists, with total assets reaching 28,173.26 in Q1 2015. 2016 to 2019: Marked by another substantial growth period, the total assets nearly double from 31,682.82 in Q1 2016 to 42,523.85 in Q1 2019. However, a dip is discernible in Q1 2020, plummeting to 37,298.41. An abrupt surge in Q1 2021 to 57,429.38 implies a potential rebound post the 2020 downturn. Regrettably, data for the remaining quarters of 2021 and the entirety of 2022 is absent, posing challenges in evaluating the overall trend for these years.
Several variables could impact the observed trends in commercial bank total assets in Nigeria. The broader economic landscape, encompassing GDP growth, inflation rates, and government policies, holds sway over commercial banks' performance. The Central Bank of Nigeria's monetary policies, including interest rates and reserve requirements, exert influence on the lending and investment activities of commercial banks. Alterations in banking regulations and prudential standards can reshape the balance sheets of commercial banks. Global economic events, such as the 2008 financial crisis and the COVID-19 pandemic, wield substantial influence on the financial sector. While the available data delineates a general upward trajectory in commercial bank total assets in Nigeria over the past decade, it is imperative to factor in the absent data and external influences when formulating conclusions or decisions based on this information.
The data presented in Figure 4.2 illustrates the total count of Automated Teller Machines (ATMs) in Nigeria spanning a 14-year period, categorized by both year and quarter. This dataset unveils notable variations in the total number of ATMs throughout the years, prompting a critical examination of the underlying reasons for these fluctuations and their potential implications for the banking industry.
Commencing from Q1 2009, the number of ATMs in Nigeria witnessed a steady rise, escalating from 26,103,489 to 95,277,416 by Q4 2011, indicative of a burgeoning demand for ATM services during this period. Although a minor dip occurred in Q1 2012, subsequent quarters experienced a rebound, maintaining an overall positive trend that signifies sustained demand for ATM services. In 2013, the count of ATMs remained relatively constant, suggesting a potential plateau in ATM deployment. However, a substantial surge transpired, with the number peaking at 116,870,000 in Q4 2015, potentially driven by banks expanding their ATM networks to cater to an expanding customer base. A remarkable escalation unfolded in the following years, exceeding 239 million ATMs in Q4 2017, possibly linked to an increased focus on financial inclusion and technology-driven banking services.
Despite a continued increase in 2018, the pace slowed, hinting at a probable saturation point in ATM deployment. 2019 saw a decline in the number of ATMs, possibly indicative of shifts in banking strategies or a preference for alternative digital banking channels. Notably, a surge in Q1 and Q3 2020 aligns with the onset of the COVID-19 pandemic, suggesting heightened demand for cash withdrawal and reduced reliance on in-branch banking. The data takes an unexpected turn in Q1 2022, depicting a substantial and uniform surge in the number of ATMs across all quarters. Such an abrupt and significant increase raises concerns about potential anomalies or errors in the data, necessitating a thorough investigation to validate its accuracy.
Considering the broader context, economic growth may fuel increased demand for banking services, including ATMs. Government policies promoting financial inclusion and cashless transactions can influence ATM network growth. The evolution of banking technologies, such as mobile banking, may impact the necessity for physical ATMs. Events like the COVID-19 pandemic can alter banking behavior, with a surge in ATM usage for cash withdrawal during lockdowns. Given the extraordinary increase in ATMs in Q1 2022, it is imperative to scrutinize and validate this data for accuracy. Understanding the dynamics behind the observed trends is crucial for making informed decisions within the banking and financial sector.
The data depicted in Figure 4.3 outlines the 14-year trajectory of Point of Sale (POS) machines in Nigeria, segmented by year and quarter. This information is pivotal for comprehending the evolution of digital finance infrastructure and its influence on the expansion of banking enterprises within Nigeria. Notably, Figure 4.3 illustrates a noteworthy surge in the quantity of POS machines from 2009 to 2022, underlining their critical role in enabling electronic transactions.
Over this period, the number of POS machines exhibited a steady rise, with a pronounced peak in Q4 2012, indicating an escalating adoption of electronic payment methods. Subsequently, there was a substantial surge in POS machines, particularly from 2013 onward, demonstrating an increasing reliance on electronic payment systems. The figures more than doubled during this phase, suggesting a robust trend towards electronic transactions. Noteworthy growth persisted from 2016 to 2019, aligning with the global trend of digitalization in the financial sector. A remarkable upswing was observed in the first and fourth quarters of 2020, potentially linked to the COVID-19 pandemic, expediting the shift to contactless payments and diminishing cash usage. However, the data reveals an unprecedented surge in Q1 2021, with a consistent number of POS machines reported for all quarters of 2022. This raises concerns about potential data anomalies or errors, necessitating further investigation to verify data accuracy.
The proliferation of POS machines in Nigeria holds multifaceted implications for banking firms and the broader digital finance infrastructure. A higher number of POS machines translates to increased electronic payment options for customers, fostering elevated satisfaction and retention rates for banking institutions. The expansion of POS networks contributes to financial inclusion by extending access to digital payment methods across a broader demographic. Additionally, the growth of POS machines diminishes reliance on cash transactions, enhancing security and transparency in the financial system. Banking entities stand to generate revenue through transaction fees and service charges associated with POS usage.
Figure 4.4 presents a comprehensive overview of the total number of web banking account holders in Nigeria spanning a 14-year period, segmented by year and quarter. This dataset is indispensable for comprehending the evolution of digital finance infrastructure and its consequential impact on the expansion of banking institutions in Nigeria.
The data discloses a substantial upswing in the count of web banking account holders from 2009 to 2022. These accounts constitute a pivotal component of the digital finance framework, facilitating online access to banking services. Noteworthy spikes are evident in Q3 2009 and Q1 2012, indicating an early embrace of web banking services in Nigeria. The momentum persists, with a remarkable surge in Q3 2015, aligning with the global trend toward digital banking. From 2016 to 2019, consistent growth prevails, signaling a heightened demand for web banking services and a surge in digital financial transactions.
The unprecedented spike across all quarters of 2020 can be attributed to the COVID-19 pandemic, hastening the adoption of digital banking due to social distancing measures and lockdowns. Q1 2022 reveals an extraordinary and uniform increase in web banking account holders, potentially signaling data anomalies or errors, necessitating further investigation for data accuracy confirmation.
The proliferation of web banking account holders holds multifaceted implications for banking institutions and the digital finance landscape. Increased customer engagement fosters stronger bank-customer relationships, while the cost-effectiveness of digital transactions enhances bank profitability. Furthermore, the expansion of web banking services contributes to financial inclusion by broadening access to banking services. However, caution is warranted as the significant and uniform surge in Q1 2022 and throughout 2022 may introduce distortions in data analysis and interpretation. Consequently, a thorough investigation is imperative to elucidate the underlying dynamics of these trends and ensure data accuracy.
4.2 Descriptive Result
Table 4.1 Descriptive Result digital finance infrastructure and growth of banking firms
|
BTA
|
ATM
|
POS
|
WEB
|
Mean
|
33016.68
|
2.84E+09
|
1.44E+08
|
4.36E+08
|
Median
|
29928.04
|
1.21E+08
|
11058740
|
2693216.
|
Maximum
|
65459.46
|
3.77E+10
|
9.71E+08
|
3.52E+09
|
Minimum
|
17331.56
|
7762869.
|
590.6460
|
289326.0
|
Std. Dev.
|
14070.45
|
9.75E+09
|
2.83E+08
|
9.85E+08
|
Skewness
|
0.976836
|
3.327308
|
2.155988
|
2.365449
|
Kurtosis
|
3.118663
|
12.07327
|
6.197182
|
7.405985
|
Jarque-Bera
|
8.938806
|
295.4192
|
67.23526
|
97.51957
|
Probability
|
0.011454
|
0.000000
|
0.000000
|
0.000000
|
Sum
|
1848934.
|
1.59E+11
|
8.06E+09
|
2.44E+10
|
Sum Sq. Dev.
|
1.09E+10
|
5.23E+21
|
4.40E+18
|
5.34E+19
|
Observations
|
56
|
56
|
56
|
56
|
Source: E-view 9.0 output
The descriptive statistics presented in Table 4.1 shed light on the digital finance infrastructure and the growth of banking firms in Nigeria, focusing on key variables: Bank Total Assets (BTA), Total number of Automated Teller Machines (ATM), Total number of Point of Sale Machines (POS), and Total number of holders of Web banking accounts (WEB). The mean, representing the average value across observations, and the median, indicating the middle value in ascending order, provide insights. BTA's mean is approximately 33,016.68, exceeding the median (29,928.04), implying larger banks with notably higher assets. For ATM, POS, and WEB, the means surpass the medians, signaling a positive skewness caused by a few high-value observations.
Maximum values denote the dataset's peaks, while minimum values represent the lows. BTA's maximum is around 65,459.46, showcasing a bank with substantial assets, while ATM's maximum is 3.77E+10, significantly higher than mean and median. POS's maximum (9.71E+08) and WEB's maximum (3.52E+09) highlight extensive machine and account holder numbers. Standard deviations, indicating data spread, are sizable, implying significant variability across banks. Positive skewness for all variables reveals right-skewed distributions with longer tails on the right side, indicating banks with exceptionally high values. High kurtosis suggests heavy tails and more extreme values compared to a normal distribution.
The Jarque-Bera test, assessing normality, yields low p-values for all variables, reinforcing non-normal, right-skewed, and heavy-tailed distributions. These values provide overall magnitude and variability insights. This analysis uncovers substantial variations in digital finance infrastructure among Nigerian banks. [10] emphasizes embracing digital innovations for digitally-savvy customers, while [5] identifies hindering factors like inadequate infrastructure and government regulations. Positive skewness and high kurtosis hint at banks with exceptionally high digital finance infrastructure values, potentially impacting overall banking growth. Table 4.1's descriptive statistics offer valuable insights, suggesting further analysis to explore factors influencing these variations would be beneficial.
4.3 Stationarity Test Result
Table 4.2 Unit Root Test for digital finance infrastructure and growth of banking firms
|
D(BTA)
|
D(ATM)
|
D(POS)
|
D(WEB)
|
ADF Statistics
|
-7.808302
|
-5.832509
|
-6.652185
|
-6.280935
|
1%
|
-3.557472
|
-3.508508
|
-3.588509
|
-3.571310
|
5%
|
-2.916566
|
-3.184230
|
-2.929734
|
-2.922449
|
Probability
|
0.0000
|
0.0000
|
0.0000
|
0.0000
|
Source: Extracted from E-view 9.0 Output
In Table 4.2, we present the outcomes of a unit root test conducted on variables associated with digital finance infrastructure and the growth of banking firms in Nigeria. The Augmented Dickey-Fuller (ADF) test statistics, critical values at 1% and 5% significance levels, and p-values are detailed for Bank Total Assets (BTA), Total Automated Teller Machines (ATM), Total Point of Sale Machines (POS), and Total Web Banking Accounts (WEB). The primary aim of this test is to ascertain whether these variables exhibit stationarity or non-stationarity.
The ADF statistics exhibit highly negative values across all four variables, ranging approximately from -5.83 to -7.81. These negative values signify that the variables have undergone differencing (hence the "D" prefix), a common technique to transform non-stationary time series data into stationary form. Critical values serve as thresholds to assess the statistical significance of the ADF statistics. In this instance, critical values at 1% and 5% significance levels are provided. For all four variables, the ADF statistics surpass the critical values at both significance levels, indicating statistical significance and implying that the variables are stationary. P-values are supplied to gauge the statistical significance of the ADF statistics, with lower p-values indicating higher statistical significance. Remarkably, all four variables report p-values as 0.0000, suggesting that the ADF statistics are highly statistically significant, providing robust evidence against the null hypothesis positing non-stationarity.
The unit root test results affirm that variables related to digital finance infrastructure (BTA, ATM, POS, WEB) and the growth of banking firms in Nigeria have undergone differencing, rendering them stationary. Stationarity is pivotal in time series analysis as it ensures that key statistical properties, such as mean and variance, remain constant over time. Stationary data is more amenable to modeling and analysis. The findings in Table 4.2 offer compelling evidence that these variables in the context of the Nigerian banking sector are stationary, a critical foundation for robust time series analysis. However, further exploration is warranted to delve into the relationships and potential causal factors underlying these variables within the Nigerian banking landscape.
4.4 Co-integration Test
Table 4.3 Unrestricted Cointegration Rank Test (Trace)
Hypothesized
No. of CE(s)
|
Eigenvalue
|
Trace
Statistic
|
0.05
Critical Value
|
Prob.**
|
None *
|
0.858799
|
149.8008
|
47.85613
|
0.0000
|
At most 1 *
|
0.518034
|
44.09211
|
29.79707
|
0.0006
|
At most 2
|
0.076883
|
4.678500
|
15.49471
|
0.8420
|
At most 3
|
0.006618
|
0.358550
|
3.841466
|
0.5493
|
Trace test indicates 2 cointegrating eqn(s) at the 0.05 level
* denotes rejection of the hypothesis at the 0.05 level
**MacKinnon-Haug-Michelis (1999) p-values
Source: Extracted from E-view 9.0 Output
Table 4.3 outlines the Unrestricted Cointegration Rank Test (Trace) results for digital finance infrastructure and banking firms' growth in Nigeria. Cointegration, indicating a long-term relationship between variables, is explored across hypothesized numbers of cointegrating equations (CEs) from "None" to "At most 3."
Eigenvalues signify cointegration strength, with larger values indicating stronger relationships. The Trace Statistic measures this strength for each hypothesized number of CEs, comparing it to the Critical Value at the 0.05 significance level. Under "None," the Trace Statistic (149.8008) significantly exceeds the critical value (47.85613), with a Prob. of 0.0000, signifying cointegrating relationships. For "At most 1," the Statistic (44.09211) surpasses the critical value (29.79707), with a Prob. of 0.0006, indicating at least one cointegrating relationship.
However, for "At most 2" and "At most 3," the Trace Statistic drops below the critical value, yielding high probabilities and suggesting limited or no cointegrating relationships. This aligns with the notion of at least one cointegrating relationship, implying a long-term connection. This connection suggests a lasting equilibrium influenced by economic or structural factors, not implying causality. [9] notes cointegration's ability to identify common stochastic trends among financial variables. Results indicate a long-term equilibrium between digital finance infrastructure and banking firms' growth, warranting further research for a comprehensive understanding of its nature and implications.
Table 4.4 Unrestricted Cointegration Rank Test (Maximum Eigenvalue)
Hypothesized
No. of CE(s)
|
Eigenvalue
|
Max-Eigen
Statistic
|
0.05
Critical Value
|
Prob.**
|
None *
|
0.858799
|
105.7086
|
27.58434
|
0.0000
|
At most 1 *
|
0.518034
|
39.41361
|
21.13162
|
0.0001
|
At most 2
|
0.076883
|
4.319950
|
14.26460
|
0.8241
|
At most 3
|
0.006618
|
0.358550
|
3.841466
|
0.5493
|
Max-eigenvalue test indicates 2 cointegrating eqn(s) at the 0.05 level
* denotes rejection of the hypothesis at the 0.05 level
**MacKinnon-Haug-Michelis (1999) p-values
Source: Extracted from E-view 9.0 Output
Table 4.4 presents the results of the Unrestricted Cointegration Rank Test (Maximum Eigenvalue) for digital finance infrastructure and banking firm growth in Nigeria. Hypothesized cointegrating relationships range from "None" to "At most 3." A significant Maximum Eigenvalue Statistic under "None" (105.7086 vs. 27.58434) suggests cointegrating relationships. "At most 1" also shows significance (39.41361 vs. 21.13162), implying at least one relationship. However, "At most 2" and "At most 3" lack significance, hinting at limited relationships. Results align with the Trace test, indicating at least one cointegrating relationship. This implies a lasting connection between digital finance infrastructure and banking growth in Nigeria, supported by studies. In summary, advancing digital finance infrastructure can enhance the efficiency and growth of banking firms in Nigeria. Further investigation is warranted to understand this relationship.
4.5 Regression Results
Table 4.5 Regression Results of digital finance infrastructure and bank assets
Variable
|
Coefficient
|
Std. Error
|
t-Statistic
|
Prob.
|
C
|
26167.96
|
956.1897
|
27.36691
|
0.0000
|
ATM
|
-3.70E-07
|
1.84E-07
|
-2.007864
|
0.0499
|
POS
|
5.45E-05
|
6.46E-06
|
8.437033
|
0.0000
|
WEB
|
1.18E-07
|
2.22E-06
|
0.053199
|
0.9578
|
R-squared
|
0.816712
|
Mean dependent var
|
33016.68
|
Adjusted R-squared
|
0.806138
|
S.D. dependent var
|
14070.45
|
S.E. of regression
|
6195.190
|
Akaike info criterion
|
20.36968
|
Sum squared resid
|
2.00E+09
|
Schwarz criterion
|
20.51435
|
Log likelihood
|
-566.3511
|
Hannan-Quinn criter.
|
20.42577
|
F-statistic
|
77.23556
|
Durbin-Watson stat
|
1.841693
|
Prob(F-statistic)
|
0.000000
|
|
|
|
Source: Extracted from E-view 9.0 Output
Table 4.5 presents the regression results of the relationship between digital finance infrastructure variables (ATM, POS, WEB) and Bank Total Assets (BTA) in Nigeria. Let's critically discuss these findings:
Table 4.5 provides the coefficients for the intercept (C), ATM, POS, and WEB. The coefficient for the intercept (C) is 26167.96. This represents the estimated value of BTA when all other independent variables (ATM, POS, WEB) are equal to zero. The coefficients for ATM, POS, and WEB represent the estimated change in BTA for a one-unit change in each respective independent variable, holding other variables constant. ATM has a negative coefficient of -3.70E-07, indicating that an increase in the number of ATMs is associated with a decrease in BTA, although this relationship is not very strong. POS has a positive coefficient of 5.45E-05, indicating that an increase in the number of Point of Sale Machines is associated with an increase in BTA. This relationship appears to be strong. WEB has a very small positive coefficient of 1.18E-07, suggesting that the number of holders of Web banking accounts has a minimal impact on BTA.
The R-squared value is 0.816712, indicating that approximately 81.67% of the variation in BTA can be explained by the independent variables (ATM, POS, WEB) in the regression model. This suggests a relatively good fit of the model. The Adjusted R-squared value is 0.806138, which is slightly lower than the R-squared value. This value accounts for the number of independent variables and penalizes the inclusion of unnecessary variables. It is still relatively high, indicating a good fit. The standard error of regression is 6195.190. It represents the average deviation of the observed values of BTA from the values predicted by the regression model. A lower standard error indicates a better fit of the model to the data. The F-statistic is 77.23556, and the associated p-value (Prob(F-statistic)) is reported as 0.000000, which is very close to zero. A low p-value for the F-statistic suggests that the overall model is statistically significant. In this case, it indicates that at least one of the independent variables (ATM, POS, WEB) is a statistically significant predictor of BTA. The Durbin-Watson statistic is 1.841693. This statistic measures the presence of autocorrelation in the residuals (errors) of the regression model. A value between 1 and 3 is often considered acceptable, and this value falls within that range, indicating that there may not be strong autocorrelation in the residuals.
The regression results suggest that the number of Point of Sale Machines (POS) has a positive and statistically significant impact on Bank Total Assets (BTA). An increase in POS is associated with an increase in BTA. The number of Automated Teller Machines (ATM) also has an impact on BTA, but the relationship is negative and less significant. [11] found a negative but insignificant relationship between ATM transactions and return on equity in Nigerian deposit money banks. [5] also reported a negative impact of ATMs on the efficiency of Greek banks. However, [12] found that ATMs do not have any influence on the Return On Asset (ROA) of Japanese banks. These findings suggest that while there may be a negative relationship between the number of ATMs and Bank Total Asset, the significance of this relationship is not consistent across different countries and banking systems.
The number of holders of Web banking accounts (WEB) does not appear to have a meaningful impact on BTA, as indicated by the very small and statistically insignificant coefficient. [1] found that internet banking had a negative and insignificant effect on banking performance, while size and capital had a positive and significant effect. Credit risk, expense management, and economic growth had a negative and significant effect on banking performance. [7] conducted a study in Bangladesh and found that banks with online banking had higher Return on Asset (ROA) and Return on Equity (ROE) compared to banks without online banking, but the results were insignificant. Additionally, ROA and ROE were lower after the implementation of internet banking, which could be attributed to initial infrastructure development costs and a failure to attract mass-scale adoption. [1] focused on First Bank Nigeria Plc and found that internet banking, including factors such as cheap internet costs, 24-hour internet services, and ICT competence of customers, significantly contributed to the bank's performance. The regression analysis in Table 4.5 provides insights into the relationships between digital finance infrastructure variables and Bank Total Assets in Nigeria. The findings highlight the importance of Point of Sale Machines (POS) in driving the growth of bank assets, while also noting the potential influence of Automated Teller Machines (ATM). The impact of Web banking account holders (WEB) appears to be negligible in this context.
4.6 Causality Test
Table 4.6 Causality Test on the effect of Digital finance infrastructure on total bank assets
Null Hypothesis:
|
Obs
|
F-Statistic
|
Prob.
|
ATM does not Granger Cause BTA
|
54
|
0.46367
|
0.6317
|
BTA does not Granger Cause ATM
|
3.97882
|
0.0250
|
POS does not Granger Cause BTA
|
54
|
0.17204
|
0.8425
|
BTA does not Granger Cause POS
|
2.79296
|
0.0710
|
WEB does not Granger Cause BTA
|
54
|
1.91985
|
0.1575
|
BTA does not Granger Cause WEB
|
4.80165
|
0.0125
|
POS does not Granger Cause ATM
|
54
|
52.9738
|
6.E-13
|
ATM does not Granger Cause POS
|
7.05835
|
0.0020
|
WEB does not Granger Cause ATM
|
54
|
1.54502
|
0.2235
|
ATM does not Granger Cause WEB
|
0.08798
|
0.9159
|
WEB does not Granger Cause POS
|
54
|
0.83824
|
0.4386
|
POS does not Granger Cause WEB
|
16.1276
|
4.E-06
|
Source: Extracted from E-view 9.0 Output
Table 4.6 presents the results of causality tests conducted to examine the directional causality between digital finance infrastructure variables (ATM, POS, WEB) and Bank Total Assets (BTA) in Nigeria. Let's critically discuss these findings:
The first test, examining whether ATM Granger Causes BTA, results in an observed F-Statistic of 0.46367 with a probability (Prob.) of 0.6317. This suggests that there is no strong evidence to reject the null hypothesis, implying that ATM does not significantly Granger Cause BTA. The second test, examining whether BTA Granger Causes ATM, results in an observed F-Statistic of 3.97882 with a probability (Prob.) of 0.0250. In this case, there is evidence to reject the null hypothesis, suggesting that BTA Granger Causes ATM.
The first test, examining whether POS Granger Causes BTA, results in an observed F-Statistic of 0.17204 with a probability (Prob.) of 0.8425. This indicates that there is no strong evidence to reject the null hypothesis, suggesting that POS does not significantly Granger Cause BTA. The second test, examining whether BTA Granger Causes POS, results in an observed F-Statistic of 2.79296 with a probability (Prob.) of 0.0710. In this case, there is some evidence to reject the null hypothesis, suggesting that BTA might Granger Cause POS at a lower significance level (0.0710).
The first test, examining whether WEB Granger Causes BTA, results in an observed F-Statistic of 1.91985 with a probability (Prob.) of 0.1575. This suggests that there is no strong evidence to reject the null hypothesis, implying that WEB does not significantly Granger Cause BTA. The second test, examining whether BTA Granger Causes WEB, results in an observed F-Statistic of 4.80165 with a probability (Prob.) of 0.0125. In this case, there is evidence to reject the null hypothesis, suggesting that BTA Granger Causes WEB.
Granger causality tests aim to determine whether one variable can predict changes in another variable. The results suggest that while ATM and POS do not significantly Granger Cause BTA, BTA might Granger Cause ATM and POS at a lower significance level. For WEB, there is no strong evidence that it Granger Causes BTA, but BTA significantly Granger Causes WEB. The causality tests in Table 4.6 offer insights into the potential temporal relationships between digital finance infrastructure variables and Bank Total Assets in Nigeria. While ATM and POS do not seem to significantly Granger Cause BTA, there is some evidence to suggest that BTA might Granger Cause ATM and POS. For WEB, there is no strong evidence of Granger causality, but BTA significantly Granger Causes WEB. Further analysis is needed to explore the underlying mechanisms behind these temporal relationships.