Economic Journal of Emerging Markets, 17.
2025, 57-69 Economic Journal of Emerging Markets Available at https://journal.
id/jep Econ.
Emerg.
Mark.
The dynamic effect of cash and non-cash payment instruments on money velocity in Indonesia Dian Zulfa.
Sofyan Syahnur* Department of Economics Development.
Faculty of Economics and Business.
Universitas Syiah Kuala.
Banda Aceh.
Indonesia *Corresponding author: kabari_sofyan@usk.
Article Info Abstract Article history:
Received 04 April 2024 Accepted 28 April 2025 Published 28 April 2025 Purpose Ai This study explores the dynamic effect of electronic money as a non-cash payment instrument on the velocity of money in Indonesia from 2012 to 2020.
JEL Classification Code:
C32.
E51.
E52
AuthorAos email:
dianazulfa522@gmail.
DOI:
20885/ejem.
Method Ai Using quarterly time series data from 2012 to 2020, the research employs the Error Correction Model (ECM), stationarity, cointegration, and classical assumption tests to ensure the correct estimation procedure.
Findings Ai The findings reveal several essential points: .
Faster circulation of cash generally increases the velocity of M1.
Excessive money supply slows down M1 circulation.
An increase in the use of debit cards (ATM.
tends to reduce M1 velocity, while quicker credit card transactions can accelerate it.
Rapid circulation of electronic money can expedite M1, but large amounts can hinder it.
Overall, both cash and non-cash money equally influence the behavior of M1 velocity in Indonesia.
Implication Ai The government should focus more on money velocity to maintain stability, even though various payment instruments are utilized in the economy.
Originality Ai The current research focuses on the dynamic development of modern finance in Indonesia and electronic money as non-cash payment instruments that impact money velocity.
Keywords Ai Financial development, electronic money, money velocity.
ECM, cash payment.
Introduction Money velocity is a vital monetary indicator that reflects the speed of money movement (Dong & Gong, 2.
It reflects the frequency of currency units used for transactions and the efficiency of money in facilitating transactions (Oyadeyi, 2.
Money velocity is a tool to assess the impact of monetary policy on economic growth (Genemo, 2.
The money supply indicates high, low, and stable money velocity .
ereafter MV).
Wang .
highlighted that a larger number of money supply in the economy will be followed by very active and smooth economic activities, thus stimulating MV to be too high.
However, a high MV tends to result in the possibility of a high inflation rate (Okedigba et al.
, 2024.
Salas, 2.
Conversely, if the money supply is too low, it will result in low money At the same time, it impacts the economy's sluggishness, and ultimately, the possibility of deflation will be very wide open (Bozkurt, 2.
This cyclical condition underlines that the size of a country's economic activity changes quickly or slowly depending on the amount of money in P ISSN 2086-3128 | E ISSN 2502-180X Copyright A 2025 Authors.
This is an open-access article distributed under the terms of the Creative Commons Attribution-ShareAlike 4.
0 International License .
ttp://creativecommons.
org/licences/by-sa/4.
Economic Journal of Emerging Markets, 17.
2025, 57-69
8,00
7,00
6,00
5,00
4,00
3,00
2,00
1,00
GDP
VM1
Money Velocity (Spin.
and Inflation (%) Money Supply and GDP (Billion Rupia.
circulation, as indicated by the level of money velocity (Faugere, 2.
It confirms that the stability of the money velocity is a vital indicator in determining the level of economic growth and inflation rate of a country (Avdiu & Unger, 2022.
Jung, 2.
INF
Source: Indonesian Central Bank and Central Bureau of Statistics (CBS) of Indonesia, 2010-2020 Figure 1.
The Development of Money Supply (M.
Gross Domestic Product.
Inflation, and Money Velocity in Indonesia, 2012 (Q.
-2020 (Q.
Figure 1 reflects the state of money velocity.
Gross Domestic Product (GDP), money supply, and inflation rate in Indonesia, which shows a different condition than what should happen.
It is exposed by the money velocity, which tends to slow down and decrease.
However.
GDP tends to increase.
Theoretically, this condition is quite contradictory.
where GDP or economic productivity is getting higher, it should be supported by money velocity, which is also high, not getting weaker (Arkadani, 2.
In addition, the money supply in Indonesia continues to expand, but is unable to create money velocity to increase, and tends to contract.
This contradictory condition makes money velocity an important monetary indicator for predicting the inflation rate.
Figure 1 emphasizes that the state of Indonesia's money velocity during the period 2012Q1 to 2020Q4 was not very stable.
In general, this condition explains that, during the 2012-2015 period.
Indonesia's money velocity showed a stable condition.
During the 2016-2020 period, the money velocity experienced conditions that tended to be stable-low.
This is based on the instability of money velocity measured through the standard deviation over nine years movement (Benk et al.
, which is segregated into three categories: high (>2.
, stable .
, and low (<2.
It indicates that Indonesia's economic movement has recently experienced growth with a slight For this reason, the depressed money velocity in Indonesia during this time must be overcome immediately to reach a more stable condition.
Money velocity uncertainty is mainly caused by money supply and GDP.
The money supply tends to continue to increase and is difficult to decrease.
This condition can be caused, among others, by increasingly diverse means of payment that are increasingly difficult for the central bank to control (Durgun & Timur, 2015.
Luo et al.
, 2.
The primary determinant of money supply is the money multiplier, while non-cash payment instruments can create a sizable money multiplier (Abbas et al.
, 2014.
Mughal et al.
, 2021.
Ongan & Gocer, 2.
Thus, non-cash payment instruments can influence money velocity in Indonesia through the money supply.
In addition, non-cash payment instruments can also influence money velocity through GDP (Benati, 2020.
Mennuni, 2023.
Sharma & Syarifuddin, 2.
It is because the level of money velocity can also be affected by various payment instruments that change people's behavior in transactions (Jiang & Shao, 2.
Through observation of data sourced from Bank Indonesia publications, it is shown that there is also a phenomenon where during the 2012-2020 period in Indonesia, the use of cash and non-cash instruments continued to increase.
However, money velocity dominantly tends to move slower than the average speed.
The dynamic effect of cash and non-cash payment instruments on money velocity .
(Zulfa and Syahnu.
000,00 3,50 000,00 3,00 000,00 2,50 000,00 2,00 000,00 1,50
000,00
000,00
1,00
000,00
0,50
MV .
n Spin.
CY.
DAC.
CC.
EM .
n BIllion Rupia.
The analysis of money is quite critical and interesting in economics.
Compared to money, the discussion of money velocity .
ereafter MV) is no less critical (Dong & Gong, 2.
Through Irving Fisher's Quantity Theory of Money.
MV can be measured by the ratio of GDP to money supply, and it is not considered constant.
Thus.
GDP and money supply changes can directly affect MV (Sharma & Syarifuddin, 2.
Through GDP.
MV provides a picture of goods and services transaction activities between economic actors.
Several studies explore money velocity and prove that money velocity will be affected by external shocks such as interest rates, per capita income, money growth volatility, and inflation (Ardakani, 2022.
Benati et al.
, 2021.
Chen & Siklos, 2022.
Nunes et al.
, 2018.
Oyadeyi, 2.
However, the shock that occurs in money velocity is not only caused by these macroeconomic variables, but money velocity also responds to various structural changes, one of which is the escalation of payment system efficiency (Mele & Stefanski, 2.
0,00
DAC
Source: Indonesian Central Bank and Central Bureau of Statistics (CBS) of Indonesia, 2010-2020 Note: CY=Currency.
DAC= debit cards (ATM.
CC=Credit Card.
EM=Electronic Money.
MV=Money Velocity Figure 2.
Currency.
Nominal Transactions of Debit Cards (ATM.
Credit Cards, and Electronic Money in Indonesia .
illion rupiah.
and Money Velocity M1 .
2012-2020 A practical payment system facilitates the smooth flow of economic activity by reducing transaction costs and increasing convenience, thereby increasing transaction volumes (Bachas et , 2021.
Brown et al.
, 2022.
Qamruzzaman & Jianguo, 2.
As the intensity of transactions and the amount of money in circulation increase, it is only natural that money will circulate even faster.
The prevailing payment system in Indonesia uses cash and non-cash instruments in economic Cash payments use currency.
Using less currency in transactions accelerates MV (Miskhin, 2.
Based on the data collected, debit cards (ATM.
, credit cards, and electronic money are the most popular non-cash payment instruments for non-cash payment systems.
Figure 2 illustrates the development of cash and non-cash payments in Indonesia during 2012-2020.
Sharma & Syarifuddin .
revealed that Indonesia is currently facing issues in the financial sector, including financial innovation and a cashless society, and these issues are expected to affect the money velocity in Indonesia.
Figure 2 shows that currency, debit cards (ATM.
, credit cards, and electronic money have an increasing trend.
Electronic money tends to increase steadily, but since 2017, the nominal increase in electronic money transactions has been rapid.
Compared to the other three payment instruments, debit cards (ATM.
have an enormous nominal transaction Most of the money velocity during the period moved more slowly and did not have an increasing trend.
The decline in transactions with non-cash instruments seen from 2020Q1 to 2020Q2 was due to the COVID-19 pandemic.
Research related to innovation in finance and its relationship with money velocity has been conducted by several researchers, such as Akinlo .
Jung .
Nampewo & Opolot .
, and Tule & Oduh .
These studies underlined that developments in finance generate a higher MV.
However.
Hermawan et al.
and Li et al.
demonstrated that the existence of Economic Journal of Emerging Markets, 17.
2025, 57-69 digital money such as Bitcoin and Central Bank Digital Currencies (CBDC.
decreases MV.
Bitcoin tends to be speculative, and CBDCs prefer to be stored rather than actively used for consumption.
Their research examines financial innovation through proxies such as mobile money, broad money, digital money, and M2 multiplier.
Meanwhile, this study will use Indonesia's most used payment instruments: cash, electronic money, debit cards (ATM.
, and credit cards.
From the previous description, it can be analyzed that Indonesia has experienced developments in the payment system, especially non-cash, where its use shows an increasing trend.
There will be a possibility of increasing the size of economic activity through GDP through efficient payments (Shahbaz et al.
, 2017.
Sreenu, 2.
in the economy.
Money velocity is expected to increase with innovations in the payment system (Rehman et al.
, 2.
For this reason, this empirical research will investigate non-cash and cash payments and how these two payment methods affect money velocity, particularly in Indonesia.
These two payment mechanisms, namely cash and non-cash payments, must be considered.
Although non-cash payments have proliferated in Indonesia, cash payments using currency have not been abandoned (AcedaEski et al.
, 2.
Therefore, it is necessary to investigate the dynamic influence.
Furthermore, this study will analyze the money velocity in three different models.
First, how is nominal MV affected by the velocity of cash and non-cash payment instruments in nominal terms? Second, how real MV .
he velocity of money adjusted to the price level where the real money balance element is included in the mode.
is affected by cash and non-cash payment instruments, which are also in real terms.
Third, how nominal MV is affected by the number of cash and non-cash payment instruments .
n the form of the ratio of both payment system instruments to GDP).
This research is supposed to contribute to the literature related to the velocity of money.
Methods The scope of this study includes the independent variables, namely monetary base (M.
or currency, debit cards (ATM.
, credit cards, electronic money, and trade openness.
In contrast, the dependent variable is money velocity.
The data used in this research were collected from the Central Bureau of Statistics (CBS) of Indonesia and the Indonesian Central Bank in quarterly data form for ten years from 2012 to 2020.
This empirical research uses a quantitative approach, employing an error correction model (ECM).
ECM assumes that the economic variables observed that are cointegrated will experience error correction in the next period if there is an imbalance in a specific period.
This means that they can return to the equilibrium position.
This study uses ECM because variables are not expected to affect the short term directly, but their impact can occur in the long term.
Moreover, the behavior of economic actors in holding money is different, so a time lag is also needed to observe differences in individual behavior towards money.
In addition, examining the behavior of the data, it appears that ECM can be employed as one of the dynamic models in investigating the MV in Indonesia.
The critical reason is that ECM can overcome the usual problems in time series data, such as the observed non-stationary variables and spurious regression results.
Before estimating the ECM model, several prerequisite tests must be met.
This ensures the ECM method is suitable and valid for solving the issues.
Therefore, this study starts by demonstrating descriptive statistics to provide a concise overview of the data used before the specific tests are Furthermore, the validity of the research model is continued by showing classical assumption tests such as normality, autocorrelation, heteroscedasticity, and multicollinearity.
The sequences of testing stages that must be carried out are .
Stationarity test using the Augmented Dickey-Fuller test.
Cointegration test with residual-based test method.
Specification of Error Correction Term (ECT) value.
Short and long run estimation with ECM.
The determination of variables used in this study follows the quantity theory of money developed by Fisher .
, which has become an underlying tool often used in monetary analysis, especially in analyzing the money velocity and transactions of goods and services.
ycAycO = ycEycN became the equation of Irving Fisher's quantity theory of money.
M is defined as the money supply.
V is the money velocity, the money supply in the narrow type, or M1.
The reason is that M1, which consists of currency and demand deposits, has a high level of liquidity compared to M2, so its use The dynamic effect of cash and non-cash payment instruments on money velocity .
(Zulfa and Syahnu.
in making transactions will be very accessible.
The behavior of the money velocity can be verified ycEycN ycU from the Fisher equation, namely ycAycO = ycA or ycAycO = ycA.
What we need to consider is the condition where the available money is not only paper and coin, but also money in electronic form.
For this reason, this study will use cash and non-cash variables along with other variables in the form of trade openness.
This study will develop three different models to explain the velocity of money flows.
First, the money velocity will be presented by the following nominal model:
ycU ycA1 ycU ycU ycU ycU = yce .
aycU , yayaya , yaya , yaycA , ycNycC), or .
ycAycO1 = yce.
cayc, yccycayca, ycayca, yceyco, ycNycC) The equation demonstrates how different types of money, namely cash (CY), debit cards in terms of ATMs (DAC), credit cards (CC), electronic money (EM), and trade openness (TO) ycU ycA explain the velocity of money circulation M1.
Trade openness is obtained from yayaycE .
M, and GDP are the sum of oil and gas, non-oil and gas exports, imports, and gross domestic products.
The equation/model is still in nominal form.
The second model will consider the element of inflation in the model, resulting in a new equation, namely:
ycU ycA1AE = ycE ycU ycU ycU ycU ycAAE ycU yce .
aycUAE , yayayaAE , yayaAE , yaycAAE , yayaycE ycE), or ycE ycE ycE ycE ycOycA1ycE = yce.
caycycE , yccycaycaycE , ycaycaycE , yceycoycE , ycycuycE ) .
Where P is the inflation rate, by including the element of inflation, it is expected to provide a picture of the money velocity in real terms where ycA1 AE ycE is the real money balance.
This study will also create a third model that shows the ratio of cash, non-cash money, and trade to GDP.
The third model can be written as:
ycAycO1 = yce.
aycUAEyayaycE , yayayaAEyayaycE , yayaAEyayaycE , yaycAAEyayaycE , ycU ycAAEyayaycE) or ycAycO1 = yce.
aycUyayaycE , yaycUyayaycE , yaycUyayaycE , yaycAyayaycE , ycNycC) .
Based on the function equations for the three models, the economic model can be transformed into an econometric model, which provides an overview of the long-term relationship.
Model 1.
ycAycO1yc = yu0 yu1 ycaycyc yu2 yccycaycayc yu3 ycaycayc yu4 yceycoyc yu5 ycNycCyc uyc Model 2.
ycAycO1yc = yu0 yu1 ycaycycEyc yu2 yccycaycaycEyc yu3 ycaycaycEyc yu4 yceycoycEyc yu5 ycycuycEyc uyc Model 3.
ycAycO1yc = yu0 yu1 yaycUyayaycEyc yu2 yayayayayaycEyc yu3 yayayayaycEyc yu4 yaycAyayaycEyc yu5 ycNycCyc uyc where 0 is the intercept, while 1, 2, 3, 4, and 5 are regression coefficients .
s parameter.
Furthermore, to examine the short-term relationship of the variables observed in this study, the equation above can be reformulated in an ECM form as follows (Engle & Granger, 1.
Model 1.
yuuycAycOyc = yu0 yu1 yuuycaycyc yu2 yuuyccycaycayc yu3 yuuycaycayc yu4 yuuyceycoyc yu5 yuuycNycCyc yu6 ycaycycOe1 yu7 yccycaycaycOe1 yu8 ycaycaycOe1 yu9 yceycoycOe1 yu10 ycNycCycOe1 yu11 .
caycycOe1 yccycaycaycOe1 ycaycaycOe1 yceycoycOe1 ycycuycOe1 Oe ycAycOycOe1 ) uyc Model 2.
yuuycAycOycEyc = yu0 yu1 yuuycaycycEyc yu2 yuuyccycaycaycEyc yu3 yuuycaycaycEyc yu4 yuuyceycoycEyc yu5 yuuycNycCycEyc yu6 ycaycycEycOe1 yccycaycaycEycOe1 yu8 ycaycaycEycOe1 yu9 yceycoycEycOe1 yu10 ycNycCycEycOe1 yu11 .
caycycEycOe1 yccycaycaycEycOe1 ycaycaycEycOe1 yceycoycEycOe1 ycycuycEycOe1 Oe ycAycOycEycOe1 ) uyc Model 3.
yuuycAycOyc = yu0 yu1 yuuyaycUyayaycEyc yu2 yuuyayayayayaycEyc yu3 yuuyayayayaycEyc yu4 yuuyaycAyayaycEyc yu5 yuuycNycCyc yu6 yaycUyayaycEycOe1 yayayayayaycEycOe1 yu8 yayayayaycEycOe1 yu9 yaycAyayaycEycOe1 yu10 ycNycCycOe1 yu11 .
aycUyayaycEycOe1 yayayayayaycEycOe1 yayayayaycEycOe1 yaycAyayaycEycOe1 ycNycCycOe1 Oe ycAycOycEycOe1 ) uyc Where, iMV is the change in velocity of money in period t.
The variable iXt or independent variable is the change in the independent variable in period t.
Meanwhile, ycUycOe1 is the lags of the independent variable.
0 is the intercept, while 1, 2, 3, 4, 5, 6, 7, 8, and 9 are the regression coefficients of Error Correction Term (ECT).
Economic Journal of Emerging Markets, 17.
2025, 57-69 Results and Discussion Descriptive Statistics Table 1 displays the descriptive statistics of each variable used in this study.
The average values of ycAycO1 and ycAycO1ycE are not different during the study period, 2012Q1-2020Q4, with 2,661 and 2,862 spins, respectively.
When the nominal and real money velocity of all payment instruments is compared, on average, electronic money circulates very quickly, and debit cards (ATM.
and M1 have a very weak speed.
However, debit cards (ATM.
have the highest ratio to Indonesia's GDP.
Meanwhile, electronic money has the lowest ratio to GDP.
This suggests that debit cards (ATM.
are more widely used in transactions in Indonesia, hence their ratio to GDP is also significant, at 29 percent.
The low speed of rotation of real DAC and nominal DAC, which amounted to 2.
48, respectively, can be caused by the large amount of money in the form of debit cards (ATM.
available, causing debit cards (ATM.
to circulate more slowly.
Table 1.
Descriptive Statistics of All Variables Observed (Period 2012Q1-2020Q.
Variable Unit Max Min Std.
Dev Frequency .
n Tim.
Total Frequency (%) Ao12Q1.
Ao13Q3.
Ao14Q1Q3Q4.
Ao15Q1.
Ao16Q3 2 Ao13Q3Q4 ycAycO1ycE 7 Ao12Q1.
Ao13Q3.
Ao14Q4.
Ao15Q1Q2Q3, 3 Ao13Q1Q2Q3 ycaycycE 4 Ao17Q4, 20Q2Q3Q4 11.
yaycUyayaycE Spins 6 Ao12Q1Q2Q3Q4.
Ao13Q1Q3 8 Ao12Q1Q2Q3Q4, yccycaycaycE Ao13Q1Q2Q3Q4 4 Ao18Q4.
Ao19Q1Q2, yayayayayaycE percent 43.
Ao20Q4 33 All the time, except 91.
for Ao20Q2Q3Q4 Ao18Q3, 20Q2Q3Q4 11.
ycaycaycE 3 Ao14Q4.
Ao15Q2Q4 yayayayaycE percent 2.
931 4 Ao12Q1Q2Q3, 13Q1 11.
0 7 Ao12Q1Q2Q3Q4, yceycoycE Ao13Q1Q2Q4 7 Ao19Q2Q3Q4, yaycAyayaycE percent 0.
Ao20Q1Q2Q3Q4 6 Ao12Q1Q2Q4.
Ao13Q3Q4.
Ao14Q2 9 Ao12Q1Q2Q4.
Ao13Q4, 25 ycycuycE Ao14Q1Q2Q3Q4 Source: Real data and calculations based on Indonesian Central Bank and Central Bureau of Statistics (CBS) of Indonesia, 2012-2020 Note: ycAycO1=Velocity of money (M.
ycAycO1ycE = velocity of money M1 real.
cy=velocity of currency.
ycaycycE =velocity of currency real.
yaycUyayaycE =ratio currency to GDP.
dac=velocity of debit cards (ATM.
yccycaycaycE =velocity of debit cards (ATM.
transactions real.
yayayayayaycE =ratio of debit cards (ATM.
transaction to GDP.
cc=velocity of credit card transaction.
ycaycaycE =velocity of credit card transaction real.
yayayayaycE =ratio credit card transaction to GDP.
em=velocity of electronic money transactions.
yceycoycE =velocity of electronic money transactions real.
yaycAyayaycE =ratio electronic money transaction to GDP.
TO=trade openness.
ycycuycE =trade openness real.
ycAycO1 Mean The dynamic effect of cash and non-cash payment instruments on money velocity .
(Zulfa and Syahnu.
Based on the nominal model, from 2012Q1 to 2020Q4, descriptive statistics show that 19 percent of the nominal M1 money velocity data circulates very fast.
Then, 22 percent of banknotes experienced rapid turnover.
Sixteen percent of debit cards (ATM.
circulate at a high speed.
As for credit cards, 91 percent circulate at a higher speed.
Only 11 percent of the data has a high velocity for electronic money.
In general, it can be concluded that almost all variables had good values at the beginning of the study period, as seen from the frequency table over time.
Classical Assumption Test Classical assumption tests are essential in identifying a model's BLUE OLS coefficient estimates.
The classical assumption tests consist of no autocorrelation, multicollinearity, and In addition to the classical assumptions, this paper conducts a normality test to check whether the data is normally distributed.
The results of the normality test using the JarqueBera test show that the data used in this study for the three models applied are proven to be normally distributed, which is reflected in the probability value of the Jarque-Bera test, which is Then, the results of the autocorrelation test using the Breusch-Godfrey LM-test indicated no autocorrelation in the models used.
It can be seen from the Obs*R2 value that the probability of Chi-Square is greater than the 5% alpha level.
White's test is used to detect the presence of heteroscedasticity in the model, with the result that all models have no It is shown from the probability Chi-squared, which is above the 5% alpha level or above 0.
By using Variance Inflation Factors (VIF), the correlation between independent variables in the model can be detected.
The test results show no multicollinearity in models 1, 2, and 3, indicated by the centered VIF value of all the independent variables being less than 10.
Table 2.
Classical Assumption Test Result Models Variable Multicollinearity Test Normality Test JarqueBera Prob.
Autocorrelation Test ObsProb.
Squared Heteroskedasticity Test ObsProb.
Squared Model 1 ycaycycE yccycaycaycE Model 2 ycaycaycE yceycoycE ycycuycE yaycUyayaycE yayayayayaycE Model 3 yayayayaycE yaycAyayaycE Notes: Model 1: MV and payment instruments in nominal.
Model 2: MV and payment instruments in and Model 3: Ratio of all payment instruments to GDP Stationarity Test This study employs the Augmented Dickey-Fuller test to examine stationary or non-stationary data.
Stationary data will prevent the occurrence of spurious regression.
Based on Table 3, it can be concluded that all variables are stationary at the first difference level.
This implies that all variables have the same direction towards equilibrium conditions in the long term.
The unit root test at the level using the Augmented Dickey-Fuller test shows that of the 17 variables used, only three are stationary in-level, namely CCP, ycNycCycE , and yaycAyayaycE .
The probability of the three variables being below = 5 %.
A unit root test at the level found that not all variables were stationary, and there Economic Journal of Emerging Markets, 17.
2025, 57-69 were still unit root problems.
all variables must be transformed into the first difference form so that all variables can be stationary at the same level.
It is known from the calculation results that all variables are stationary at the same level, which is at first difference or I.
Table 3.
Augmented Dickey-Fuller Stationarity Test Results Level and 1st Difference Model Variable ycycya Level First Difference Conclusion 447*** I.
146***
DAC
375***
(-0.
Model 1
363***
757***
565***
578***
631***
yeEyeoyc
763***
yeIyeCyeEyc
Model 2
062***
366***
yeEyeEyc
434***
yeIyeayc
048***
222***
447***
950***
ycyeAycycyc
670***
Model 3
360***
565***
Notes: Entries in ***, **, * are significant at 1%, 5%, and 10% confidence levels, respectively.
Cointegration Test Table 4.
Cointegration Test Results Model Model 1 Model 2 Model 3 t-Stat Prob.
The cointegration test can be recognized by testing the stationarity of the residuals generated from the long-term equation model, also known as the method of residual-based test using the PhillipsPerron test.
Cointegration occurs when the residual is stationary at the level.
Table 4 shows that the residual test results of the research models used are stationary and significant at the level because The dynamic effect of cash and non-cash payment instruments on money velocity .
(Zulfa and Syahnu.
the probability value is below = 5%.
For this reason, it can be interpreted that there is a cointegration or long-term relationship in the ECM model, and the method can be continued.
Error Correction Model (ECM) Following the primary purpose of this study.
ECM is employed to estimate the short-term and long-term effects of the independent variables on the dependent variable of this study.
The test results for the short term can be represented in Table 5.
Furthermore.
Table 5 shows the results of the statistically significant ECT regression coefficient with a probability value of 0.
000, 0.
001, respectively, for model 1, model 2, and model 3.
Moreover, the coefficient of ECT has a negative value.
This indicates that the result of the ECM model used in this empirical research is valid.
The coefficient of ECT will determine the speed at which equilibrium can be re-achieved.
Table 5.
Short-term Estimation Results Model 1 Coefficient Model 2 Coefficient Model 3 (Prob.
(Prob.
icy 188*** 465*** iyeEyeoyc iycyeAycycyc .
idac yoyeIyeCyeEyc iycycycycycyc .
icc yoyeEyeEyc yoycycycycyc .
000***
yoyeIyeayc iycycycycyc .
iTO iTO yoyeiyeayc .
cy(-.
yeEyeoyc (-.
ycyeAycycyc (-.
dac(-.
yeIyeCyeEyc (-.
ycycycycycyc(-.
cc(-.
yeEyeEyc (-.
ycycycycyc(-.
em(-.
yeIyeayc (-.
ycycycycyc (-.
TO(-.
TO(-.
yeiyeayc (-.
ECT (-.
997***
ECT(-.
050***
ECT(-.
Notes: ***, **, * indicate significant at 1%, 5%, and 10% confidence level.
Coefficient (Prob.
864***
The cointegration test results show the existence of cointegration, which indicates a longterm relationship in the model, so the long-term estimate for model 1, model 2, and model 3 is shown in Table 6.
In the long and short run, the MV of both nominal and real currency (CY) positively and significantly influences the MV of M1.
The currency is part of the money in the narrow sense (M.
For this reason, the faster the currency circulates, the faster M1 rotates in This result is supported by Khavgpom's .
and Aggarwal et al.
They underlined that the currency could spur an increase in MV through increased consumption and productive investment.
Meanwhile, the estimation results of the ratio of the currency to GDP have a negative and significant relationship with the MV in both the long-term and short-term.
When the ratio of the currency to GDP increases, it means that more currency is in circulation compared to the economic activity that occurs.
If the amount of money in the economy rises without a balanced increase in economic activity, the currency tends to settle, and it is not actively used for transactions.
This Economic Journal of Emerging Markets, 17.
2025, 57-69 condition indicates that the greater the ratio of the currency to GDP, the slower the rotation of M1 will be.
This result is consistent with Irving Fisher's quantity of money theory, which states that the money supply and velocity of money have a negative relationship.
An increase in the amount of currency can reduce MV.
Table 6.
Long-term Estimation Results Model 1 Coefficient Model 2 Coefficient Model 3 (Prob.
(Prob.
862***
229***
541***
yeEyeoyc ycyeAycycyc .
676***
yeIyeCyeEyc ycycycycycyc .
yeEyeEyc ycycycycyc .
0001***
yeIyeayc ycycycycyc .
Notes: ***, **, * indicate significant at 1%, 5%, and 10% confidence levels.
Coefficient (Prob.
417***
086***
Using debit cards (ATM.
on the velocity of M1 shows a negative and significant effect for both nominal and real models.
This implies that an increase in the use of debit cards (ATM.
can trigger the velocity of M1 to decrease.
Debit cards (ATM.
are categorized into the broad money supply (M.
Debit cards (ATM.
allow individuals to reduce their use of cash and demand Bachas et al.
explained that low-income people use credit cards to accumulate savings and reduce current consumption.
Financial hoarding without using it for productive economic purposes also results in a slowdown of the MV (Fu et al.
, 2.
Credit cards in the real model show that they positively affect money velocity.
This can be interpreted that the more often credit cards are used, the greater the MV.
The results of this study support the statement of Genemo .
and Liu & Serletis .
that payments using credit cards generate much higher MV.
In the short term and long term, electronic money significantly and positively affects the MV of the currencies.
The more transactions made by using electronic money, the greater the speed at which electronic money circulates, ultimately accelerating the MV of M1.
This finding is supported by Anwar et al.
Electronic money has facilitated easy and fast transactions in the It is now often used for micro-transactions such as payment of parking fees, toll roads, public transportation, and others (Brown et al.
, 2.
Thus, although the value of electronic money transactions is relatively small, the high frequency of transactions accelerates the overall MV of the Conclusion This study explores the effect of cash and non-cash payment systems on money velocity in Indonesia from 2012Q1 to 2020Q4 by applying the Error Correction Model (ECM).
There are several main conclusions from this research, namely: .
in the short run, an increase in the money velocity of nominal and real currency and electronic money causes MV1 to increase, meaning that there is a positive and significant relationship, while the speed of circulation of real debit cards (ATM.
has a negative and significant effect, increasing the velocity of transactions with debit cards (ATM.
in real terms can reduce MV1.
in the long run, the velocity of nominal and real currency, credit cards and electronic money has a significant positive effect on MV1, while the velocity of nominal and real debit cards (ATM.
transactions has a negative and significant impact on MV1.
in the short term, the ratio of currency has an adverse effect on MV1.
in the long run, the ratio of banknotes and electronic money has a negative and The dynamic effect of cash and non-cash payment instruments on money velocity .
(Zulfa and Syahnu.
significant impact on MV1.
trade openness both in the short and long term for the three research models has no considerable effect.
Based on the results of this study, the government can consider the suggestion that banks should provide innovations and better services in the banking sector to support the publicAos use of non-cash payments and provide more EDC machines, especially for areas that still have minimal availability.
In general, non-cash payments are essential in determining MV in Indonesia.
Therefore, the government should pay more attention to MV stabilization, especially NCPI.
Acknowledgement Not Applicable References