FinTech Innovation, Stability, and Efficiency

 
F
i
n
T
e
c
h
 
I
n
n
o
v
a
t
i
o
n
,
 
S
t
a
b
i
l
i
t
y
 
a
n
d
 
E
f
f
i
c
i
e
n
c
y
:
E
v
i
d
e
n
c
e
 
f
r
o
m
 
M
a
l
a
y
s
i
a
n
 
B
a
n
k
 
I
n
d
u
s
t
r
y
 
R
u
b
i
 
A
h
m
a
d
,
 
C
h
a
n
g
q
i
a
n
 
X
i
e
,
 
F
a
u
z
i
 
Z
a
i
n
i
r
 
a
n
d
 
M
a
g
d
a
M
o
h
s
i
n
Introduction
CONTENTS
Literature Review and
Hypothesis Development
Data
 
and Methods
Results and Discussion
Conclusion
 
I
n
t
r
o
d
u
c
t
i
o
n
1
PART
Challenges & Opportunities
 
R
e
s
e
a
r
c
h
 
b
a
c
k
g
r
o
u
n
d
Increasing demand
The Increasing demand for 
online financial
services 
during the Covid-19 pandemic
 
(Murinde,
Rizopoulos, & Zachariadis, 2022).
Increasing supply
Bank Negara Malaysia issued the 
digital bank
licenses
 to five applicants on 29
th
 April 2022 
Malaysia became the second country in
ASEAN to issue digital bank licenses
H
ow financial technology influences the
stability
 and 
efficiency
 of the traditional
banking sector in Malaysia?
1) This study 
provides empirical evidence for Malaysia and similar
developing countries 
with an open hug for FinTech development
but with relatively less advanced technology infrastructure.
 
C
o
n
t
r
i
b
u
t
i
o
n
s
2) 
T
he samples in this study cover different types of banks in
Malaysia, including 
commercial banks and Islamic banks
, which
provide a comprehensive view of the Malaysian banking sector.
The data that can be used to
purely measure the banks’
technological progress
 is still
not available.
 
L
i
m
i
t
a
t
i
o
n
s
The uniqueness of the
Malaysian banking industry
limits the broad applicability
of findings.
Literature Review &
Hypothesis Development
2
PART
2.1 Definition of 
FinTech innovation
FinTech is defined as a form of financial innovation triggered by advanced technologies
such as cloud computing, big data, blockchain, and artificial intelligence, which may result
in new business models, applications, processes or products, and further has a material effect
on the provision of financial services (CGFS & FSB, 2017).
 
This definition has also been adopted by the Basel Committee on Banking Supervision
(BCBS).
 
Despite its current rapid growth 
(Goldstein, Jiang, & Karolyi, 2019), there is still no clear conclusion about its
influence (Lee et al., 2021).
 
We know innovation creates not only new markets by bringing about new products and services but also
replaces existing markets, which is the ‘creative destruction’ theory coined by Schumpeter (2010). This
theory can also apply to the study of FinTech.
 
Hence, past studies on the impact of FinTech innovation on the financial industry draw two views -- the
innovation-growth hypothesis and the innovation-fragility hypothesis (Beck, Chen, Lin, & Song, 2016).
T
he 
innovation-growth hypothesis 
supports that FinTech innovation improves the functions of the
financial system and has a positive effect on traditional finance.
 
 
Examples : FinTech innovation fosters risk sharing (Allen & Gale, 1994) and increases
 
economic growth (Beck et al., 2016; Laeven, Levine, & Michalopoulos, 2015).
 
From the perspective of the 
innovation-fragility hypothesis
, FinTech innovation disrupts the
equilibrium of the existing financial system by introducing new forms of financial risks and
challenging the existing financial regulatory system (Zhang, 
2020). Hence, the risk of the financial
system is significantly increased.
 
 
Examples 
: 
Triggers an excessive credit expansion (Brunnermeier, 2009),increases the
 
volatility of bank profitability (Beck et al., 2016), and affects the effectiveness of traditional
 
monetary policy (Odularu & Okunrinboye, 2009).
2.2 Overview of FinTech Innovation in Malaysia
We refer to the financial
inclusion index provided
by Meng (2020) to
describe the FinTech
development in Malaysia. 
The index is b
ased on the
Financial Access Survey
(FAS) data released by
IMF in 2019
It is relatively stable, but
reached a maximum value
of 0.0427 in 2009
It is between the average
levels of developing
countries and developed
countries
2.3 Hypothesis development
Hypothesis 1 (H1):
 There is a positive relationship between FinTech innovation
and bank stability.
Hypothesis 1A (H1a):
 Compared to Islamic banks, the positive impact of FinTech
innovation on banks’ stability is more significant in commercial banks.
Hypothesis 2 (H2):
 There is a positive relationship between FinTech innovation
and bank efficiency.
Data and Methods
3
PART
3.1 Sample selection
Oribis Bank Focus Database
Consists of 
36 banks 
(8 local commercial banks, 8 local Islamic banks, 15 foreign
commercial banks, and 5 foreign Islamic banks)
Sample period: 
2006 to 2020
Final sample of 
393 bank-year 
observations
,
 
unbalanced
 
panel
 
data
3.2 Variable description: DV
Table 4: T
he average DEA-Malmquist score of banks in Malaysia (2006-2020).
3.2 Variable description: FinTech innovation
Defined as total value of 
other
intangible assets
 divided by total
assets for each bank each year (%).
FinTech2
I
f the FinTech2 is 
missing data
,
equals to 
one
, otherwise, 
zero
.
Fama and French (2015); Flannery
and Rangan (2006)
Challenges:
1)
T
he data on 
R&D expenditures 
or 
patents
 are typically 
not collected or disclosed
 by financial institutions
2)
Not a unified method is concluded
- Some 
studies use FinTech indexes provided by professional research institutions and others use text-mining technology to
collect FinTech-related keywords on the Internet to construct FinTech indexes.
Fintech2_DUM
3.2 Variable description: FinTech innovation
The relationship between
FinTech1
 and 
FinTech2
a positive correlation (
0.6212
)
3.2 Definition of variables
3.3 Research method
We employ the multivariate panel data regression model to test the research hypotheses.
3.4 Descriptive statistics
FinTech1: 
mean value
(35.3051%), range from
minimum value (27.1160%) to
maximum value (183.1608%),
which suggests 
considerable
heterogeneity 
in FinTech
development in the years of
Malaysian banks.
3.4 Descriptive statistics
Table 7 reports the correlation matrix of the identified variables. Overall, the correlation matrix does not
suggest any serious multicollinearity concerns. 
Results and Discussion
4
PART
4.1.1 FinTech innovation and bank stability
1)
O
nly in the regression of
subsample (commercial banks,
column 2), supporting 
H1a
2)
 
There is obvious 
heterogeneity
 in
the impact of FinTech
development on bank stability.
4.1.2 FinTech innovation and bank efficiency
In column (1), the coefficient of
FinTech1
 is positive and
significant at the 10% level;
In column (5) it is positive and
significant at the 1% level
H2 is only significant for our full
sample and foreign bank sample
data leading us to conclude that
full sample results are driven by
foreign banks
4.2.1 Endogeneity concern: GMM
These results are consistent
with the baseline regression
results, which further
confirm 
H1a
 and 
H2
.
4.2.2 Replacing the dependent variable
Stability2
 equals to the natural log of (ROAA+CAR)/σ(ROAA), where the standard deviation of ROAA is over the sample period for each bank, and
the ROAA is the rate of return on average assets. 
Efficiency2
 equals to the DEA-Malmquist score calculated by three input variables (labour, fixed
assets, and total deposits) and three out variables (total loans, other earning assets, and non-interest income), referring to Lee et al. (2021).
4.2.4 Additional analysis: size effect
Overall, the impact of
FinTech innovation on banks’
stability and efficiency
varies by bank size.
Big-sized banks are defined as
those with total assets exceeding
the median value of 36 banks’
total assets. Small-sized banks
are those less than the median
value.
4.2.4 Additional analysis: profitability effect
Columns (1) show that banks with 
high
ROE 
have significantly 
improved their
stability 
with higher development of
FinTech.
However, banks with 
low ROE 
enjoy a
significant effect of 
improving efficiency
brought by FinTech, as shown in column
(4)
B
anks in the high ROE group are
defined as those whose ROE exceeds the
median ROE of the sample banks, and
those in the low ROE group are on the
contrary.
Conclusion
5
PART
Compared with Islamic banks, FinTech innovation
significantly improves the stability of 
commercial banks
FinTech innovation has a more significant effect on
improving 
foreign banks
’ efficiency than local banks
These baseline results are affirmed using the GMM
approach to mitigate potential endogeneity issues
Heterogeneity analysis
: 
the high-profit 
banks enjoy the
benefits of improving their stability level from
FinTech development. However, for 
the small-sized
and low-profit banks
, FinTech innovation contributes
more to improving their efficiency.
Conclusion
 
I
m
p
l
i
c
a
t
i
o
n
s
The impact of Financial innovation 
on Malaysian traditional
banks is therefore heterogeneous. For example, the
 big-sized and
high-profit banks enjoy the benefits of improving their stability
level from FinTech development while the small-sized and low-
profit banks, more to improving their efficiency.
A need for f
inancial regulatory authorities to update their financial
supervision methods in real-time, as the wide usage of FinTech
increases the business complexity of banks.
When 
looking at the growth of 
financial innovation and its impact on
bank stability and efficiency, i
t i
s essential to investigate this
relationship thoroughly from different standpoints and under various
conditions.
Main reference
Thakor, A. V. (2020). Fintech and banking: What do we know?. Journal of Financial
Intermediation, 41, 100833.
Beck, T., Chen, T., Lin, C., & Song, F. M. (2016). Financial innovation: The bright and the dark
sides. Journal of Banking & Finance, 72, 28-51.
Murinde, V., Rizopoulos, E., & Zachariadis, M. (2022). The impact of the FinTech revolution on
the future of banking Opportunities and risks. International Review of Financial Analysis,
102103.
Lee, C. C., Li, X., Yu, C. H., & Zhao, J. (2021). Does fintech innovation improve bank efficiency?
Evidence from China’s banking industry. International Review of Economics & Finance, 74,
468-483.
Hu, D., Zhao, S., & Yang, F. (2022). Will fintech development increase commercial banks risk-
taking? Evidence from China. Electronic Commerce Research, 1-31.
Ahamed, M. M., & Mallick, S. K. (2019). Is financial inclusion good for bank stability?
International evidence. Journal of Economic Behavior & Organization, 157, 403-427.
Wang, R., Liu, J., & Luo, H. (2021). Fintech development and bank risk taking in China. The
European Journal of Finance, 27(4-5), 397-418.
THANKS
Slide Note
Embed
Share

FinTech innovation influences the stability and efficiency of the traditional banking sector in Malaysia. It provides empirical evidence and covers different types of banks in Malaysia, including commercial and Islamic banks.


Uploaded on Dec 21, 2023 | 14 Views


Download Presentation

Please find below an Image/Link to download the presentation.

The content on the website is provided AS IS for your information and personal use only. It may not be sold, licensed, or shared on other websites without obtaining consent from the author. Download presentation by click this link. If you encounter any issues during the download, it is possible that the publisher has removed the file from their server.

E N D

Presentation Transcript


  1. FinTech Innovation, Stability and Efficiency: Evidence from Malaysian Bank Industry Rubi Ahmad, Changqian Xie, Fauzi Zainir and Magda Mohsin

  2. 1 Introduction Literature Review and Hypothesis Development 2 CONTENTS 3 Data and Methods 4 Results and Discussion 5 Conclusion

  3. 1 PART Introduction

  4. Research background Increasing supply Bank Negara Malaysia issued the digital bank licenses to five applicants on 29thApril 2022 Malaysia became the second country in ASEAN to issue digital bank licenses 02 Increasing demand The Increasing demand for online financial services during the Covid-19 pandemic (Murinde, Rizopoulos, & Zachariadis, 2022). 01 Challenges & Opportunities How financial technology influences the stability and efficiency of the traditional banking sector in Malaysia?

  5. Contributions 1) This study provides empirical evidence for Malaysia and similar developing countries with an open hug for FinTech development but with relatively less advanced technology infrastructure. 2) The samples in this study cover different types of banks in Malaysia, including commercial banks and Islamic banks, which provide a comprehensive view of the Malaysian banking sector.

  6. Limitations The data that can be used to purely measure technological progress is still not available. the banks The Malaysian banking industry limits the broad applicability of findings. uniqueness of the

  7. 2 PART Literature Review & Hypothesis Development

  8. 2.1 Definition of FinTech innovation FinTech is defined as a form of financial innovation triggered by advanced technologies such as cloud computing, big data, blockchain, and artificial intelligence, which may result in new business models, applications, processes or products, and further has a material effect on the provision of financial services (CGFS & FSB, 2017). This definition has also been adopted by the Basel Committee on Banking Supervision (BCBS). Despite its current rapid growth (Goldstein, Jiang, & Karolyi, 2019), there is still no clear conclusion about its influence (Lee et al., 2021). We know innovation creates not only new markets by bringing about new products and services but also replaces existing markets, which is the creative destruction theory coined by Schumpeter (2010). This theory can also apply to the study of FinTech. Hence, past studies on the impact of FinTech innovation on the financial industry draw two views -- the innovation-growth hypothesis and the innovation-fragility hypothesis (Beck, Chen, Lin, & Song, 2016).

  9. The innovation-growth hypothesis supports that FinTech innovation improves the functions of the financial system and has a positive effect on traditional finance. Examples : FinTech innovation fosters risk sharing (Allen & Gale, 1994) and increases economic growth (Beck et al., 2016; Laeven, Levine, & Michalopoulos, 2015). From the perspective of the innovation-fragility hypothesis, FinTech innovation disrupts the equilibrium of the existing financial system by introducing new forms of financial risks and challenging the existing financial regulatory system (Zhang, 2020). Hence, the risk of the financial system is significantly increased. Examples : Triggers an excessive credit expansion (Brunnermeier, 2009),increases the volatility of bank profitability (Beck et al., 2016), and affects the effectiveness of traditional monetary policy (Odularu & Okunrinboye, 2009).

  10. 2.2 Overview of FinTech Innovation in Malaysia We refer to the financial inclusion index provided by Meng describe the development in Malaysia. The index is based on the Financial Access Survey (FAS) data released by IMF in 2019 It is relatively stable, but reached a maximum value of 0.0427 in 2009 It is between the average levels of countries and developed countries (2020) to FinTech developing

  11. 2.3 Hypothesis development Hypothesis 1 (H1): There is a positive relationship between FinTech innovation and bank stability. Hypothesis 1A (H1a): Compared to Islamic banks, the positive impact of FinTech innovation on banks stability is more significant in commercial banks. Hypothesis 2 (H2): There is a positive relationship between FinTech innovation and bank efficiency.

  12. 3 PART Data and Methods

  13. 3.1 Sample selection Oribis Bank Focus Database Consists of 36 banks (8 local commercial banks, 8 local Islamic banks, 15 foreign commercial banks, and 5 foreign Islamic banks) Sample period: 2006 to 2020 Final sample of 393 bank-year observations, unbalanced panel data

  14. 3.2 Variable description: DV Bank stability Z-score=(??? + ???)/?(???) The higher Z-score implies more stability Bank efficiency ? 1??,?? ?? ? 1?? 1,?? 1 ?? ?(?? 1,?? 1) ? 1= (1) ??? ?????????? ?? ?(??,??) ?= (2) ??? ?????????? ?? 1/2 ? 1(??,??) ?(??,??) ?? ? 1(?? 1,?? 1) ?? ?(?? 1,?? 1) (3) ??? ??????????= ?? ?? According to the definition of DEA-Malmquist, if the number calculated from equation (3) is > 1, it means that the total factor productivity (TFP) of banks has increased from period t-1 to t; otherwise, it means that the banks TFP has decreased.

  15. Table 4: The average DEA-Malmquist score of banks in Malaysia (2006-2020). Period Local commercial Local Islamic Foreign commercial Foreign Islamic Full sample 2005-2006 0.9962 1.1302 1.0536 2006-2007 1.0482 1.0295 1.0402 2007-2008 1.0740 1.1485 1.1059 2008-2009 1.0141 0.8770 0.9553 2009-2010 1.0256 1.1420 1.0755 2010-2011 1.0458 0.9785 1.0170 2011-2012 1.0018 1.1770 1.1191 1.0010 1.0899 2012-2013 0.9984 0.9795 1.0001 1.0606 1.0066 2013-2014 1.0148 1.1315 1.3464 1.0399 1.1824 2014-2015 1.0372 1.0296 1.1052 1.0130 1.0605 2015-2016 1.0253 1.2159 1.0515 1.1528 1.0963 2016-2017 1.0129 1.1255 1.0008 1.0613 1.0396 2017-2018 1.0044 1.0036 1.2585 1.0223 1.1126 2018-2019 0.9958 0.9506 0.8547 0.9028 0.9140 2019-2020 1.0098 1.0108 1.1475 0.9587 1.0603 Average 1.0203 1.0693 1.0793 1.0236 1.0540

  16. 3.2 Variable description: FinTech innovation Challenges: 1) The data on R&D expenditures or patents are typically not collected or disclosed by financial institutions 2) Not a unified method is concluded - Some studies use FinTech indexes provided by professional research institutions and others use text-mining technology to collect FinTech-related keywords on the Internet to construct FinTech indexes. FinTech2 FinTech1 Defined as total value of other intangible assets divided by total assets for each bank each year (%). Defined as the total value of off- balance-sheet items divided by total assets for each bank each year (%) (Beck et al., 2016) Fintech2_DUM If the FinTech2 is missing data, equals to one, otherwise, zero. Fama and French (2015); Flannery and Rangan (2006)

  17. 3.2 Variable description: FinTech innovation The relationship between FinTech1 and FinTech2 a positive correlation (0.6212)

  18. 3.2 Definition of variables Variable Definition Equals to the natural log of (ROA+CAR)/ (ROA), the standard deviation of ROA is over the sample period for each bank. The higher Z-score implies less risk-taking and more stability. Stability Equals to the score of DEA-Malmquist. Efficiency The total value of off-balance-sheet items divided by total assets for each bank each year (%). FinTech1 The total value of other intangible assets divided by total assets for each bank each year (%). FinTech2 Dummy variable, if the FinTech2 is missing data, equals to one, otherwise, zero. FinTech2_DUM The natural logarithm of a bank's total assets (in RM Ringgit million). The value of liquid assets divided by total assets for each bank each year (%). The value of total equity divided by total assets for each bank each year (%). Equals to the Cost to income (Efficiency) ratio (%) obtained from BankFocuse. Size Liquidity CapitalStructure CostToIncome Ratio of non-interest income to total operating income for each bank each year (%). IncomeDiversification

  19. 3.3 Research method We employ the multivariate panel data regression model to test the research hypotheses. (4) ??????????,?= ??????? ?,?+ ??????????,?+ ??+ ??+ ??,? (5) ???????????,?= ??????? ?,? 1+ ??????????,? 1+ ??+ ??+ ??,? The subscripts i and t represent the individual bank and year, respectively. The coefficient ? and ? measure the effect of FinTech innovation on banks risk- taking and efficiency. The regression model includes bank fixed effect to control any potential bias due to omitted variables in the panel datasets and year fixed effect to control potential temporal effects. We winsorize the top and bottom 1% of the continuous variables to control the outlier bias.

  20. 3.4 Descriptive statistics FinTech1: mean value (35.3051%), range from minimum value (27.1160%) to maximum value (183.1608%), which suggests considerable heterogeneity in FinTech development in the years of Malaysian banks.

  21. 3.4 Descriptive statistics Table 7 reports the correlation matrix of the identified variables. Overall, the correlation matrix does not suggest any serious multicollinearity concerns. Stability Efficiency FinTech1 FinTech2 FinTech2_DUM Size Liquidity CapitalStructure CostToIncome IncomeDiversification Stability Efficiency FinTech1 FinTech2 FinTech2_ DUM Size Liquidity CapitalStr ucture CostToInc ome 1 -0.0334 -0.0579 -0.2128*** 1 -0.0536 -0.0766 1 -0.0336 1 -0.0433 0.0896* 0.0825 -0.4998*** 1 0.2001*** -0.1114** -0.1380*** 0.1481*** 0.0124 0.0741 0.1658*** -0.2264*** -0.2178*** 0.2725*** 1 -0.4845*** 1 0.0211 0.0136 0.0361 -0.0466 0.0629 -0.5430*** 0.4292*** 1 -0.2698*** -0.0003 -0.1692*** 0.3674*** -0.0778 -0.3232*** -0.0072 0.1381*** 1 IncomeDiv ersification -0.0673 0.0206 -0.2036*** -0.1143** 0.0712 -0.2259*** 0.0699 -0.0890* 0.2120*** 1

  22. 4 PART Results and Discussion

  23. 4.1.1 FinTech innovation and bank stability (1) (2) (3) (4) (5) Dependent variable: Stability Expected Sign FullSample- FE 0.0001 (0.0003) -0.0920** (0.0414) -0.0019** -0.0022*** (0.0007) 0.0747*** 0.0723*** 0.1130*** 0.0993*** 0.0694*** (0.0044) (0.0038) (0.0122) -0.0025*** -0.0022*** -0.0026*** -0.0021*** -0.0029*** (0.0005) (0.0005) (0.0006) 0.0004 0.0013** -0.0009 (0.0006) (0.0006) (0.0009) 3.9020*** 3.8401*** 2.4440*** 3.1405*** 4.0615*** (0.4625) (0.4571) (0.2370) Yes Yes Yes Yes Yes Yes 36 23 13 0.9282 0.9491 0.8659 393 275 118 Commercial- FE 0.0005* (0.0003) -0.0934** (0.0386) Foreign- FE 0.0001 (0.0004) -0.1350** (0.0578) -0.0012* (0.0006) Islamic-FE Local-FE -0.0004 (0.0005) 0.0492 (0.0297) -0.0009 (0.0008) 0.0002 (0.0004) -0.008 (0.0554) -0.0017 (0.0014) FinTech1 + Size +/- 1) Only in the regression of subsample (commercial banks, Liquidity + (0.0008) column 2), supporting H1a CapitalStructure + (0.0098) (0.0044) 2) There is obvious heterogeneity in CostToIncome + (0.0006) 0.0004 (0.0005) (0.0007) -0.0004 (0.0013) the impact of FinTech IncomeDiversification - development on bank stability. _cons (0.6911) Yes Yes 16 0.9204 193 (0.631) Yes Yes 20 0.9439 200 Individual bank Year Number of banks R-squared (within) Observations

  24. 4.1.2 FinTech innovation and bank efficiency (1) (2) (3) (4) (5) Dependent variable: Efficiency Expected Sign FullSample- FE 0.0010* (0.0005) 0.1175* (0.0644) 0.0043 (0.0030) 0.0022 (0.0017) -0.0014 (0.0014) -0.3587 (0.7007) Yes Yes 36 0.1292 357 Commercial- FE 0.0003 (0.0005) 0.0826 (0.0654) 0.0070** (0.0027) 0.0002 (0.0024) -0.0023 (0.0022) -0.0043 (0.7329) Yes Yes 23 0.1723 252 Islamic- FE 0.0009 0.0006 0.0018*** (0.0009) (0.0009) (0.0006) 0.1145 0.1447* 0.1595** (0.1475) (0.0760) (0.0678) -0.002 0.0053** 0.0053 (0.0068) (0.0024) (0.0039) 0.0042 0.0025 (0.0027) (0.0023) (0.0023) 0.0006 -0.0019 (0.0024) (0.0013) (0.0021) -0.2132 -0.723 (1.4601) (0.8814) (0.7311) Yes Yes Yes Yes 13 16 0.2125 0.1810 105 177 Local- FE Foreign- FE l.FinTech1 + In column (1), the coefficient of FinTech1 is positive and l.Size +/- significant at the 10% level; l.Liquidity + In column (5) it is positive and 0.0026 l.CostToIncome + significant at the 1% level H2 is only significant for our full -0.001 l.IncomeDiversification + sample and foreign bank sample -0.7733 _cons data leading us to conclude that Individual bank Year Number of banks R-squared (within) Observations Yes Yes 20 0.1963 180 full sample results are driven by foreign banks

  25. 4.2.1 Endogeneity concern: GMM (1) (2) (3) Foreign bank- GMM Dependent Variable: Efficiency Commercial bank-GMM Full sample-GMM Dependent Variable: Efficiency Dependent Variable: Stability 0.4126* (0.2311) L.Stability -0.3368*** (0.0533) 0.0013* (0.0007) 0.1563* (0.0802) 0.0051* (0.0030) -0.3854*** (0.0583) 0.0022*** (0.0008) 0.2406*** (0.0777) 0.0062* (0.0036) L.Efficiency These results are consistent 0.0040** (0.0016) 1.5012** (0.6912) -0.0017 (0.0014) 0.1448*** (0.0311) 0.0007 (0.0018) FinTech1 with the baseline regression Size results, which further Liquidity confirm H1a and H2. CapitalStructure 0.0018 (0.0018) 0.0023 (0.0019) CostToIncome 0.0012* -0.0016 0.0004 IncomeDiversification (0.0007) -14.8869* (8.177) Yes Yes 23 0.0335 0.3290 1.0000 229 (0.0022) -0.4427 (0.9021) Yes Yes 36 0.0001 0.6541 1.0000 321 (0.0023) -1.1642 (0.8266) Yes Yes 20 0.0017 0.2032 1.0000 160 _cons Individual bank Year Number of banks AR (1) AR (2) Sargan test Observations

  26. 4.2.2 Replacing the dependent variable Stability2 equals to the natural log of (ROAA+CAR)/ (ROAA), where the standard deviation of ROAA is over the sample period for each bank, and the ROAA is the rate of return on average assets. Efficiency2 equals to the DEA-Malmquist score calculated by three input variables (labour, fixed assets, and total deposits) and three out variables (total loans, other earning assets, and non-interest income), referring to Lee et al. (2021). (1) Full (2) (3) (4) (5) (1) Full Sample- FE 0.0004 (0.0004) 0.0938 (0.0591) 0.0017 (0.0030) 0.0043** (0.0020) (2) (3) (4) (5) Dependent variable: Stability2 Commercial- FE 0.0006** (0.0002) -0.0814* (0.0440) -0.0016** (0.0006) 0.0699*** (0.0042) Islamic- FE -0.0004 (0.0005) 0.0357 (0.0330) -0.0010 (0.0008) 0.1137*** 0.1104*** 0.0648*** (0.0082) (0.0061) - 0.0020*** 0.0030*** (0.0006) (0.0006) -0.0003 -0.0004 (0.0005) (0.0004) 2.4814*** 2.7756*** 3.9339*** (0.2823) (0.3942) Yes Yes Yes Yes 13 16 0.9206 0.9640 118 193 Foreign- FE 0.0002 (0.0004) -0.1253** (0.0560) -0.0007 (0.0005) Dependent variable: Efficiency2 Local-FE Commercial- FE Islamic- FE Local- FE Foreign- FE Sample-FE 0.0002 (0.0003) -0.0831* (0.0424) -0.0016** (0.0006) 0.0718*** (0.0048) 0.0004* (0.0002) 0.0200 (0.0324) -0.0011 (0.0010) FinTech1 0.0010* (0.0005) 0.1076 (0.0651) 0.002 (0.0040) 0.0045* (0.0026) 0 -0.0005 (0.0008) 0.114 (0.1415) -0.0081 (0.0057) 0.0057 (0.0036) 0.0002 (0.0004) 0.1886* (0.0941) 0.0030** (0.0011) 0.0050* (0.0027) - 0.0037** (0.0016) -1.172 (1.0406) Yes Yes 16 0.2177 177 l.FinTech1 (0.0004) 0.0723 (0.0555) 0.0055** (0.0026) 0.0023* (0.0013) Size l.Size Liquidity l.Liquidity CapitalStructure (0.0038) - 0.0030*** (0.0006) 0.0002 (0.0009) - l.CostToIncome -0.0031*** -0.0033*** CostToIncome (0.0006) 0.0004 (0.0005) 3.7976*** (0.4676) Yes Yes 36 0.9375 393 (0.0008) 0.0006 (0.0007) 3.7380*** (0.5124) Yes Yes 23 0.9519 275 -0.0029* -0.0033 -0.0007 -0.0025 l.IncomeDiversification IncomeDiversification (0.0016) -0.0668 (0.6058) Yes Yes 36 0.1626 357 (0.0024) 0.0771 (0.5672) Yes Yes 23 0.1838 252 (0.0028) 0.0862 (1.4218) Yes Yes 13 0.2908 105 (0.0024) -0.1936 (0.6198) Yes Yes 20 0.2228 180 _cons _cons (0.5747) Yes Yes 20 0.9511 200 Individual bank Year Number of banks R-squared (within) Observations Individual bank Year Number of banks R-squared (within) Observations

  27. 4.2.4 Additional analysis: size effect (1) (2) (3) (4) Stability Big Size 0.0003 (0.0002) 0.0249 (0.0384) -0.0002 (0.0009) 0.1104*** 0.0665*** (0.0046) -0.0031*** -0.0025*** (0.0004) -0.0001 (0.0005) 2.6305*** 3.3918*** (0.4654) Yes Yes 0.9700 240 Stability Efficiency Efficiency Small Size Big Size Small Size -0.0002 0.0001 (0.0006) (0.0005) -0.0599 0.1434** (0.0625) (0.0614) -0.0009 0.0047** (0.0007) (0.0020) 0.0024** (0.0009) 0.038 (0.1436) 0.0027 (0.0088) FinTech1 Overall, the impact of FinTech innovation on banks Size stability and efficiency Liquidity varies by bank size. CapitalStructure (0.0042) 0.0043 (0.0027) -0.0025 (0.0016) -0.7646 (0.6879) Yes Yes 0.2353 222 0.0017 (0.0034) -0.0025 (0.0038) 0.7689 (1.2432) Yes Yes 0.1235 135 Big-sized banks are defined as those with total assets exceeding the median value of 36 banks total assets. Small-sized banks are those less than the median value. CostToIncome (0.0006) 0.0005 (0.0007) IncomeDiversification _cons (0.5747) Yes Yes 0.9483 153 Individual bank Year R-squared (within) Observations

  28. 4.2.4 Additional analysis: profitability effect (1) (2) (3) (4) Columns (1) show that banks with high Stability High ROE 0.0003* (0.0002) 0.0475 (0.0305) -0.0013 (0.0008) 0.1070*** (0.0080) -0.0025*** -0.0034*** (0.0007) 0.0009* (0.0005) 2.4176*** (0.3686) Yes Yes 0.9612 227 Stability Low ROE -0.00003 (0.0005) -0.1230* (0.0651) -0.0017* (0.0009) 0.0661*** (0.0043) Efficiency Efficiency High ROE Low ROE -0.0003 (0.0006) 0.00005 (0.0884) -0.0067 (0.0094) 0.0071 (0.0045) 0.0017 (0.0024) ROE have significantly improved their 0.0020*** (0.0004) 0.1748* (0.0837) 0.0111** (0.0052) 0.0002 (0.0030) -0.0052 (0.0034) stability with higher development of FinTech1 FinTech. Size However, banks with low ROE enjoy a significant effect of improving efficiency Liquidity brought by FinTech, as shown in column (4) CapitalStructure CostToIncome Banks in the high ROE group are (0.0010) 0.0001 (0.0006) 3.9263*** (0.6349) Yes Yes 0.9446 166 defined as those whose ROE exceeds the IncomeDiversification median ROE of the sample banks, and 0.9420 (1.015) Yes Yes 0.1607 209 -0.6698 (0.8309) Yes Yes 0.1920 148 those in the low ROE group are on the _cons contrary. Individual bank Year R-squared (within) Observations

  29. 5 PART Conclusion

  30. Conclusion Compared with Islamic banks, FinTech innovation significantly improves the stability of commercial banks 1 FinTech innovation has a more significant effect on improving foreign banks efficiency than local banks 2 These baseline results are affirmed using the GMM approach to mitigate potential endogeneity issues 3 Heterogeneity analysis: the high-profit banks enjoy the benefits of improving their stability level from FinTech development. However, for the small-sized and low-profit banks, FinTech innovation contributes more to improving their efficiency. 4

  31. Implications The impact of Financial innovation on Malaysian traditional banks is therefore heterogeneous. For example, the big-sized and high-profit banks enjoy the benefits of improving their stability level from FinTech development while the small-sized and low- profit banks, more to improving their efficiency. When looking at the growth of financial innovation and its impact on bank stability and efficiency, it is essential to investigate this relationship thoroughly from different standpoints and under various conditions. A need for financial regulatory authorities to update their financial supervision methods in real-time, as the wide usage of FinTech increases the business complexity of banks.

  32. Main reference Thakor, A. V. (2020). Fintech and banking: What do we know?. Journal of Financial Intermediation, 41, 100833. Beck, T., Chen, T., Lin, C., & Song, F. M. (2016). Financial innovation: The bright and the dark sides. Journal of Banking & Finance, 72, 28-51. Murinde, V., Rizopoulos, E., & Zachariadis, M. (2022). The impact of the FinTech revolution on the future of banking Opportunities and risks. International Review of Financial Analysis, 102103. Lee, C. C., Li, X., Yu, C. H., & Zhao, J. (2021). Does fintech innovation improve bank efficiency? Evidence from China s banking industry. International Review of Economics & Finance, 74, 468-483. Hu, D., Zhao, S., & Yang, F. (2022). Will fintech development increase commercial banks risk- taking? Evidence from China. Electronic Commerce Research, 1-31. Ahamed, M. M., & Mallick, S. K. (2019). Is financial inclusion good for bank stability? International evidence. Journal of Economic Behavior & Organization, 157, 403-427. Wang, R., Liu, J., & Luo, H. (2021). Fintech development and bank risk taking in China. The European Journal of Finance, 27(4-5), 397-418.

  33. THANKS

Related


More Related Content

giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#giItT1WQy@!-/#