Académique Documents
Professionnel Documents
Culture Documents
Are foreign banks more protable than domestic banks? Home- and
host-country effects of banking market structure, governance, and supervision
Sheng-Hung Chen , Chien-Chang Liao
Department of Finance, Nanhua University, Chiayi, Taiwan
a r t i c l e
i n f o
Article history:
Available online 20 November 2010
JEL classication:
E32
L11
G21
G28
Keywords:
Bank protability
Foreign bank
Banking competition
Panzar and Rosse H
Governance
Country risk
Regulatory supervision
a b s t r a c t
Using both bank- and country-level data on banking sectors from 70 countries over the period
19922006, this paper empirically investigates the joint home- and host-country effects of banking market structure, macroeconomic condition, governance, and changes in bank supervision on foreign bank
margins. We nd that foreign banks are more protable than domestic banks when they operate in a host
country whose banking sector is less competitive and when the parent bank in the home country is
highly protable. Moreover, when foreign banks operate in a host country with lower growth rates of
GDP, higher interest and ination rates, and more stringent regulatory compliance with Basel risk
weights, their margins increase. Specically, changes in bank supervision of a parent banks ownership
restrictiveness in the home country signicantly increases foreign bank margins, while supervisory
changes in regulatory compliance with Basel risk weights in the host country enhances foreign bank
margins.
Crown Copyright 2010 Published by Elsevier B.V. All rights reserved.
1. Introduction
Over the last few decades, countries have extended their international nancial activities and opened their doors to foreign
banks, thereby increasing their levels of nancial liberalization
and integration and boosting the banking industry worldwide.
The globalization of banking has led to institutional and regulatory
improvements and benetted both local and foreign banks. At the
local level, globalization has increased the competitiveness of
banking markets by reducing administrative costs, lowering net
interest margins, and driving down bank rates of return. At the
international level, globalization has allowed foreign banksespecially those from more developed nancial systemsto expand
into emerging market economies, where they have sometimes
become dominant. Dietz et al. (2008) nd that according to 2006
survey on global industry prot revenues and prots in the banking industry amounted to $788 billion, which was higher than
prots in any other industry. The authors also indicate that
between 2000 and 2006 the prots of developed countries grew
signicantly faster than those in less developed countries. This
Corresponding author. Tel.: +886 5 2721001x56541; fax: +886 5 2427172.
E-mail addresses: shenghong@mail.nhu.edu.tw (S.-H. Chen), macrossgods@
yahoo.com.tw (C.-C. Liao).
0378-4266/$ - see front matter Crown Copyright 2010 Published by Elsevier B.V. All rights reserved.
doi:10.1016/j.jbankn.2010.11.006
820
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
One possible reason for such contradictory results may lie in the
fact that empirical analysis for the performance of foreign banks
has mainly concentrated on European Union and US banks operating abroad. Another may be that most studies have compared the
performance of domestic banks with foreign banks for a single
country rather than analyzing how cross-country differences in
banking market structure inuence a banks protability. For
example, Williams (1996, 1998a,b, 2003) studies Australia, Minh
To and Tripe (2002) studies New Zealand, Ursacki and Vertinsky
(1992) focuses on Japan and Korea, Engwall et al. (2001) studies
the Nordic countries, and Dietrich and Wanzenried (2009) studies
Switzerland.
This paper contributes to previous studies in several ways. First,
we empirically investigate the key factors in 70 countries that affect bank protability in foreign banks when compared to domestic
banks. To do so, we analyze the long-term relationship between
bank protability and banking market structure using structural
measures, such as Panzar and Rosse H (PR H) statistics and the
Lerner index, as well as static measures, such as the Herndhal Hirschman Index (HHI) and the Concentration Ratio (CR), over the
period 19922006. Second, we empirically examine whether
cross-country differences in banking market structure, macroeconomic environment, institutional governance, banking competition, and country risk between host country and home country
inuence foreign protability. Third, we use regulation and supervision variables to explore the joint home- and host-country impacts of change in bank supervision on foreign bank protability.
The rest of this paper is organized as follows. Section 2 reviews
related literature dealing with international studies of bank protability. Section 3 describes the model for analysis used in the theoretical framework of this paper. Section 4 presents the empirical
model used for estimation, explains the data collection process,
and provides summary statistics. Section 5 discusses the results
of our empirical models. Finally, Section 6 presents our concluding
remarks.
2. Related literature
Previous literature related to bank protability can be classied
into two major categories. The rst is a cross-country comparison
of bank protability, and the second is the impact of banking market structure on bank prot. We discuss our detailed ndings in the
following sections and summarize the important empirical results
in Appendix A.
2.1. International comparison of bank protability
The empirical literature on bank prot focuses strongly on European countries. One exception to this is Williams (1996), who constructs a model that attempts to explain the performance of
Japanese nancial institutions in Australia. In general the model
performs well for measures of size, but comparatively poorly for
measures of protability. Brock and Suarez (2000) report that bank
spreads in the 1990s were inuenced by liquidity and capital risk
at the bank level and by interest rate volatility, ination, and
GDP growth at the macroeconomic level. Claessens et al. (2001)
analyze 7900 banks from 80 countries for the period 19881995
and found that foreign banks enjoyed higher prots than domestic
banks when operating in developing countries. They found the
opposite effect, however, for banks operating in developed
countries.
Kosmidou et al. (2007) analyze 19 Greek bank subsidiaries
operating in 11 nations during the period 19952001. Their ndings show that the protability of the parent bank and the operating experience of the host nation had a robust and positive impact
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
821
RD R a;
RL R b
sp a b
c
h
1
Rr2I Q
2
where ch measures the banks risk neutral spread and is the ratio of
the intercept (c) and the slope (h) of the symmetric deposit and loan
arrival functions of the bank. A large c and a small h will result in a
large ch and a large spread (sp). This means that if a bank faces relatively inelastic demand and supply functions in the markets in
which it operates, it may be able to exercise monopoly power by
demanding a greater spread than it could get if banking markets
were competitive (low ch ratio).
The ratio of ch provides some measure of the producers surplus
or monopoly rent element in bank spreads or margins. The second
term is a rst-order risk-adjustment term that depends on three
factors: (i) R, the bank managements coefcient of absolute risk
aversion; (ii) Q, the size of bank transactions; and (iii) r2, the
instantaneous variance of the interest rate on deposits and loans.
Note that the second term implies that, the greater the degree of
risk aversion, the larger the size of transactions. It also implies that
the greater the variance of interest rates, the larger the bank margins. This spread equation has an important implication: Even if
banking markets are highly competitive, as long as a banks management is risk-averse and faces transaction uncertainty, positive
bank margins will exist as the price of providing deposit and loan
intermediation.
3.2. Measuring banking market structure
This paper uses four measures proxies as the market structure
in the banking industry. Two are structural: PanzarRosse H
statistics and the Lerner index, and two are static: the Herndhal
Hirschman Index and the Concentration Ratio of the top four banks
(CR4).
3.2.1. Structural measures
3.2.1.1. PanzarRosse H statistics. Panzar and Rosse (1987) create a
convenient framework with which to assess banking market structure. Their modelthe PanzarRosseuses bank-level data to measure how a change in factor input prices is reected in equilibrium
revenues earned by banks. In a state of perfect competition, marginal costs and total revenues increase proportionally to input
prices. In a state of monopoly market, however, an increase in factor input prices raises marginal costs but reduces output and total
revenues. Following Bikker and Haaf (2002), Bikker et al. (2007),
and Bikker and Spierdijk (2008), this paper uses the PR H model
to measure the degree of competition in the banking industry by
estimating, for each country, the following empirically reduced
form of bank revenues:
822
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
nj gOI ei;t
3
where t is the number of periods observed, and i is the total number
of banks. ln denotes the natural logarithmic operator, p denotes a
banks interest revenues, PF stands for annual funding rate calculated as interest expense to total funds (IE/FUN) (as the proxy for
unit price of fund), PL denotes price of personnel expenses calculated as annual personnel expenses to total assets (PE/TA) (as the
proxy for unit price of labor), PK is the price of physical capital
expenditure calculated as non-interest expenses to xed assets
(ONIE/FA) (as the proxy for unit price of capital), and OI is calculated
as the ratio of the income to total assets. In order to apply the PR
approach to our data, this paper uses the following empirical reduced form equation of bank revenues, which is in line with Eq. (3):
p MC
LERNER
p
where p is the average price of a bank (proxied as the quotient between total revenues and total assets) and MC is the marginal cost
of total assets calculated from the estimation of a translogarithmic
costs function. Where the total costs depend on the prices of three
inputs and on the banks volume of production (total assets), the
costs function estimated is:
3
X
1
ln Costit bo b1 ln Q it b2 ln Q 2it
ckt ln W k;it
2
k1
3
X
k1
uk ln Q it ln W k;it
3 X
3
X
k1 j1
where Qit represents a proxy for bank output or total assets for bank
i at time t, and Wk,it represent three input prices. W1,it, W2,it and W3,it
indicate the input prices of labor, funds, and xed capital, respectively. The input prices are dened as: (W1,it) personnel costs/total
assets; (W2,it) interest expenses/total deposits; and (W3,it) other
operating and administrative expenses/total assets.
MC TAit
"
#
3
X
Costit
b1 b2 ln Q it
u ln W k;it
Q it
k1
Finally, the Lerner index is averaged over time for each bank i for
inclusion in the regression model. The data for this measure of competition are at the bank level and are estimated country-by-country
and year-by-year.
3.2.2. Static measures
3.2.2.1. Herndhal Hirschman Index (HHI). We dene the HHI as:
HHI
3
X
s2i
i1
where Pi,t in Eq. (9) stands for bank protability for bank i in country j in year t and ei,j,t represents the error term. To represent dependent variables, we use three measures of a banks protability
alternately: net interest margin (NIM), ROA, and ROE. NIM is the
net interest margin generated by the net interest income (=interest
income interest expense) divided by current assets. This ratio suggests that the higher the net interest margin, the better the performance. ROA is dened as the net prot divided by total assets; it
represents the earning performance of a bank based on total assets.
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
823
10
Host
c4h Contoli;j;t c5 FBi;j;t DEV Home
i;j;t c6 FBi;j;t DEV i;j;t
X
Home
c7p FBi;j;t MacEconHost
j;t c7p FBi;j;t MacEconj;t
X
Home
c8q FBi;j;t RiskHost
j;t c8q FBi;j;t Riskj;t
X
Home
c9r FBi;j;t GOV Host
j;t c9r FBi;j;t GOV j;t
X
Home
c10s FBi;j;t SUPHost
ni;j;t
j;t c10s FBi;j;t SUP j;t
11
824
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
11
X
h1
X
X
Home
c8q FBi;j;t RiskHost
c8q FBi;j;t Riskj;t
j;t
X
X
5. Empirical results
12
To describe the correlation between bank ownership and performance, we investigate whether banking market structure,
developing economy, governance, and supervision for foreign
banks between host country and home country play a partial or
joint role in driving this relationship. We start by using NIM,
ROE, and ROA to estimate a baseline model that enables us to compare how both foreign ownership and banking competition affect
bank performance. Next, we test whether banking competition,
governance, and supervision from host country and home country
inuence (either individually or jointly) the association between
foreign ownership and protability.
Table 1
Summary statistics by bank ownership.
Variables
All banks
N
Domestic banks
Mean
Std. dev.
Foreign banks
Mean
Std. dev.
Mean
Std. dev.
Dependent variable
Net interest margin (NIM)
Return on assets (ROA)
Return on equity (ROE)
61,720
62,217
62,181
3.645
0.804
9.118
3.696
3.115
27.507
53,833
54,163
54,133
3.680
0.797
8.994
3.478
3.063
25.765
7887
8054
8048
3.407
0.848
9.954
4.928
3.446
37.149
Independent variables
Bank characteristics
Operation cost
Interest payments
Liquidity ratio
Opportunity cost
Credit risk
Log (bank size)
Non-interest revenues
Capital strength
Off balance sheet activity
Market shares
Other operating income
62,098
61,669
62,146
61,862
62,605
62,546
61,731
62,362
60,451
55,584
61,668
0.038
4.003
52.198
0.029
0.002
3.164
0.628
12.853
0.264
0.013
1.796
0.059
5.349
23.723
0.043
0.046
0.975
0.407
35.641
4.862
0.043
5.323
54,070
53,786
54,094
53,901
54,447
54,413
53,836
54,270
52,796
48,066
53,780
0.038
3.957
53.684
0.029
0.002
3.161
0.615
11.972
0.251
0.012
1.707
0.059
5.226
23.067
0.040
0.041
0.954
0.424
32.666
5.175
0.040
5.227
8028
7883
8052
7961
8158
8133
7895
8092
7655
7518
7888
0.041
4.312
42.217
0.032
0.005
3.187
0.715
18.757
0.349
0.025
2.405
0.062
6.119
25.584
0.058
0.070
1.108
0.245
50.927
1.408
0.060
5.900
60,069
62,111
54,914
54,914
0.435
0.356
0.306
0.492
31.287
1.467
6.477
14.996
52,547
54,080
47,534
47,534
0.420
0.285
0.291
0.333
19.438
1.3267
6.066
11.774
7522
8031
7380
7380
0.513
0.259
0.382
0.151
72.608
1.565
7.792
2.792
Macroeconomic condition
Growth rate of GDP
Ination rate
Real interest rate
62,739
62,585
55,611
2.559
7.470
6.816
2.346
81.726
7.380
54,659
54,538
48,437
2.471
6.911
6.768
2.218
79.397
6.859
8080
8047
7174
3.151
11.260
7.136
3.005
95.953
10.219
825
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
Table 2
Impact of foreign ownership on protability: results from individual country regressions.
Criteria
Signicant
Insignicant
Austria
Egypt
Italy
Panama
Singapore
Tunisia
Algeria
Indonesia
Kenya
Poland
Slovenia
United States
Australia
Bangladesh
Switzerland
Cyprus
France
Honduras
Netherlands
Peru
Venezuela
Algeria
Colombia
Denmark
Greece
Macau
Norway
Reunion
Zimbabwe
Croatia
Luxembourg
Hong Kong
Thailand
Bulgaria
Canada
Ireland
South Africa
Slovakia
Czech
Republic of Korea
Bolivia
Hungary
Mexico
Portugal
Spain
Indonesia
The table provides summarized results from examining the impact of foreign ownership on protability for each country in the sample based on regression model (9).
5.1. The effects of banking market structure: Are foreign banks more
protable than domestic banks?
Before investigating the effect of banking market structure on
bank protability, we rst use Eq. (9) to perform individual country
regression in order to examine whether foreign banks are more
protable than domestic banks. This approach enables us to
explore whether huge differences exist between countries with
respect to the performance of foreign banks. Our results are summarized in Table 2. Note that the table divides the countries in
our sample into 4 groups. The rst one (upper left quadrant) consists of countries in which the correlation of foreign ownership to
bank protability is both positive and statistically signicant. In
these countries foreign banks are, on average, more protable than
domestic banks. The second group (upper right quadrant) consists
of countries in which the correlation of foreign ownership to bank
protability is positive, but not statistically signicant. The third
group (lower left quadrant) consists of countries in which the correlation of foreign ownership to bank protability is both negative
and signicant. In other words, domestic banks in these countries
tend to outperform foreign banks. The fourth group (lower right
quadrant) consists of countries in which the correlation of foreign
ownership to bank protability is statistically negative, but
insignicant.
Table 2 indicates that all four cases occur in our group of 47
countries. Foreign banks perform better than domestic banks in
12 countries: Austria, Algeria, Egypt, Indonesia, Italy, Kenya,
Panama, Poland, Singapore, Slovenia, Tunisia, and the United
States. They perform worse in 4 countries: Croatia, Hong Kong,
Thailand, and Luxembourg. Interestingly, this nding differs
greatly from that of Claessens and Horen (2009). This may be
due to the data period covered or to the control variables used
for empirical specication. The majority of countries do not exhibit
a signicant difference between domestic and foreign banks.
Among this group, ownership has a positive sign in 14 countries
and a negative sign in 15. These results reinforce those of previous
studies: When looking at aggregate data, there is no straightforward relationship between bank ownership and performance.
Our next step was to investigate which bank and country
factors have a signicant impact on the relative performance of
foreign banks while quantifying the inuence of banking competition on bank performance. As shown in Table 3, our empirical
results from the estimation of Eq. (10) indicate that in the context
of international investigation, foreign banks correlate both positively and signicantly to higher protability relative to domestic
banks.
It is interesting to note that differences in banking market structure across countries play a signicant role in protability when
we use PR H statistics, Lerner index, HHI, and CR4 to correspond
to different independent variables such as NIM, ROA, and ROE.
From the perspective of structural measures like PR H statistics,
Table 3 shows that a signicant negative relationship exists between bank NIM and degree of competition in the banking industry. This nding indicates that bank net interest margin is more
likely to be squeezed when a bank operates in a competitive banking industry in countries with free market access to foreign banks
or new domestic banks. This result is especially robust when using
the Lerner index (another structural measure). In contrast to alternative measures such as the HHI index, we also nd the same result. This indicates that a banks interest margin correlates both
signicantly and positively to a monopolistic banking market
structure, which is consistent with the CR4 measure.
In regard to bank nancial characteristics and macroeconomic
variables, the results from both NIM and ROA as dependent variables indicate that banks with better protability correlate signicantly and negatively with operating cost. When we use ROE as a
dependent variable, however, the correlation is insignicant. This
is in line with Maudos and Fernndez de Guevara (2004), who
use NIM as a dependent variable, as well as with Pasiouras and
Kosmidou (2007), who use ROA as a dependent variable. Banks
with better protability are positively and signicantly associated
with implicit payments in the NIM model. This indicates that
charging for a banks services through lower remuneration of liabilities would lead to higher interest margins for banks, which is
consistent with previous studies by Saunders and Schumacher
(2000) and Maudos and Fernndez de Guevara (2004).
Liquidity ratios correlate signicantly and positively with NIM,
ROA and ROE, which indicates that an increase in bank liquidity ratio tends to enhance a banks protability. Furthermore, banks with
better protability are positively and signicantly correlated to
their opportunity cost. This result, which is similar to ndings by
Saunders and Schumacher (2000), suggests that the opportunity
costs of reserves suggests the average return on earning assets
foregone by holding deposits in cash. As opportunity costs increase
the cost of funds beyond the observed rate, banks gain net interest
margins by compensating for these costs.
The relationship between bank protability and off-balance
sheet is signicantly negative in all estimation results from the
models, which means the more off-balance activities a bank undertakes, the more its prots will decrease. The relationship between
size (SIZE) and bank protability is negative on the net interest
margin. The negative coefcient indicates that larger banks tend
Independent variables
Intercept
Foreign banks ownership
(1)
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
7.18***
(72.04)
0.34***
(4.25)
6.37***
(57.07)
0.37***
(3.65)
7.63***
(68.31)
0.42***
(4.77)
7.77***
(71.51)
0.45***
(5.11)
4.69***
(58.17)
0.19***
(3.38)
3.85***
(43.06)
0.18**
(2.51)
4.81***
(55.96)
0.22***
(3.77)
4.97***
(59.50)
0.26***
(4.52)
12.92***
(12.85)
1.15**
(2.26)
12.60***
(12.55)
1.57***
(3.10)
12.90***
(10.77)
2.00***
(3.81)
19.73***
(15.22)
1.11**
(2.12)
0.02***
(10.59)
0.05***
(8.70)
0.35***
(7.93)
0.06***
(11.18)
15.98***
(15.94)
1.00***
(6.07)
14.99***
(10.75)
CR4 (assets)
7.48***
(6.16)
0.25***
(6.91)
1.17***
(4.36)
0.23***
(9.11)
0.11
(0.14)
1.95***
(2.85)
0.16***
(33.25)
0.01***
(9.18)
7.89***
(19.98)
0.00***
(5.92)
6.03***
(82.41)
0.01***
(6.23)
0.13***
(32.81)
2.86***
(9.22)
0.40***
(15.93)
2.57***
(5.01)
1.09*
(1.75)
0.20***
(41.49)
0.00***
(4.34)
9.13***
(27.99)
0.00***
(8.10)
6.52***
(93.65)
0.01***
(5.96)
0.15***
(40.42)
2.88***
(9.29)
0.41***
(16.50)
3.62***
(7.13)
1.08*
(1.72)
0.20***
(41.26)
0.00***
(4.27)
9.23***
(28.33)
0.00***
(8.15)
6.51***
(93.58)
0.01***
(5.91)
0.15***
(40.24)
0.97***
(3.83)
0.20***
(11.22)
2.62***
(6.13)
1.11***
(2.79)
0.66***
(160.28)
0.00***
(8.52)
1.37***
(5.05)
0.01***
(16.85)
4.39***
(77.00)
0.02***
(10.32)
0.70***
(225.87)
0.44*
(1.83)
0.13***
(6.49)
1.12*
(1.74)
1.47***
(3.03)
0.64***
(150.96)
0.01***
(9.48)
0.77**
(2.23)
0.01***
(16.85)
3.82***
(59.63)
0.01***
(8.25)
0.66***
(193.24)
1.07***
(3.98)
0.21***
(11.76)
2.62***
(6.62)
1.06**
(2.55)
0.65***
(156.13)
0.00***
(6.56)
1.30***
(4.75)
0.01***
(16.51)
4.51***
(76.56)
0.01***
(8.36)
0.68***
(216.69)
1.09***
(4.06)
0.22***
(12.11)
3.04***
(7.67)
1.11***
(2.65)
0.65***
(156.38)
0.00***
(6.29)
1.38***
(5.03)
0.01***
(16.66)
4.48***
(76.26)
0.01***
(8.38)
0.68***
(217.05)
1.21
(0.25)
0.34*
(1.88)
33.13***
(7.03)
4.38
(1.16)
1.51***
(21.00)
0.06***
(9.07)
20.60***
(5.04)
0.02***
(4.37)
12.44***
(13.96)
0.16***
(5.75)
1.90***
(35.48)
3.44
(0.72)
0.13
(0.70)
43.69***
(10.10)
3.44
(0.89)
1.56***
(22.60)
0.05***
(6.78)
27.98***
(7.05)
0.02***
(4.54)
11.46***
(12.93)
0.11***
(4.17)
1.81***
(35.08)
3.54
(0.74)
0.17
(0.95)
48.04***
(10.25)
3.19
(0.81)
1.58***
(22.76)
0.06***
(8.21)
28.16***
(7.03)
0.02***
(4.18)
13.01***
(14.49)
0.12***
(4.36)
1.85***
(35.70)
5.27
(1.09)
0.53***
(2.82)
30.19***
(6.24)
2.79
(0.71)
1.65***
(23.55)
0.07***
(9.79)
23.73***
(5.85)
0.02***
(4.36)
14.57***
(16.10)
0.12***
(4.22)
1.90***
(36.33)
0.05***
(11.34)
0.10***
(51.74)
0.01***
(52.34)
0.07***
(17.46)
0.07***
(40.12)
0.01***
(50.28)
0.07***
(17.48)
0.07***
(39.61)
0.01***
(51.41)
0.02***
(5.64)
0.06***
(40.02)
0.01***
(27.61)
0.01**
(2.35)
0.07***
(45.97)
0.01***
(28.42)
0.00
(0.61)
0.05***
(37.83)
0.01***
(32.27)
0.00
(0.54)
0.05***
(37.35)
0.01***
(32.72)
0.39***
(6.23)
0.00
(0.08)
0.05***
(12.15)
0.06
(1.05)
0.06***
(3.04)
0.04***
(15.51)
0.19***
(3.12)
0.05**
(2.57)
0.05***
(16.34)
0.09
(1.48)
0.04**
(2.14)
0.04***
(15.64)
Observations
Number of banks
Log-likelihood
Wald v2
38,761
3847
72,003
12,789***
47,864
4731
97,297
15,362***
47,865
4731
97,248
15,464***
47,063
4671
84,575
40,559***
38,791
3851
67,186
30,170***
47,894
4735
89,749
38,000***
47,895
4735
89,750
38,000***
47,036
4671
221,124
1741***
47,865
4735
225,999
2126
47,865
4735
226,122
1879
47,866
4735
226,105
1,922
3.71***
(13.43)
0.35***
(15.19)
2.79***
(5.26)
1.01*
(1.78)
0.21***
(45.98)
0.00***
(8.06)
9.18***
(30.13)
0.00***
(8.27)
6.56***
(102.90)
0.02***
(14.14)
0.11***
(33.55)
0.39*
(1.95)
47,033
4667
89,285
17,300 ***
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
0.76***
(3.78)
HHI (assets)
826
Table 3
The impacts of banking market structure on bank protability: alternative measures of banking competition.
Table 4
The effects of parent banks protability on foreign banks performance: protability level.
Independent variables
Intercept
Foreign banks ownership
(2)
(3)
(4)
(5)
(6)
(7)
(8)
(9)
(10)
(11)
(12)
12.22***
(135.53)
0.54***
(8.58)
13.22***
(130.96)
0.66***
(9.86)
13.10***
(129.23)
0.58***
(8.78)
14.16***
(127.42)
0.47***
(7.07)
8.01***
(100.00)
0.31***
(6.91)
7.08***
(78.05)
0.36***
(6.28)
8.26***
(97.21)
0.32***
(6.96)
9.24***
(98.72)
0.23***
(5.07)
47.40***
(37.45)
1.61***
(3.55)
47.00***
(37.48)
1.98***
(4.38)
48.31***
(38.01)
2.02***
(4.36)
51.42***
(36.55)
1.79***
(3.84)
0.02***
(11.87)
0.21***
(4.02)
0.36***
(9.25)
0.01*
(1.85)
1.55***
(10.06)
HHI (assets)
11.34***
(12.61)
1.55***
(11.73)
4.24**
(2.16)
0.05***
(19.00)
0.14***
(4.23)
2.40***
(132.01)
2.63***
(128.17)
2.64***
(128.74)
2.58***
(124.88)
1.54***
(90.29)
1.43***
(74.16)
1.62***
(89.43)
1.56***
(85.66)
11.45***
(37.66)
11.28***
(37.79)
11.49***
(38.30)
11.30***
(37.34)
6.79***
(32.88)
5.41***
(128.36)
1.26***
(3.17)
0.27
(0.62)
0.27***
(74.36)
0.00***
(6.89)
8.54***
(37.33)
0.00***
(2.84)
1.69***
(27.68)
0.02***
(12.01)
0.15***
(53.09)
7.52***
(31.02)
5.95***
(124.96)
1.75***
(4.67)
0.20
(0.43)
0.27***
(68.89)
0.01***
(11.02)
8.94***
(35.48)
0.00***
(2.94)
1.43***
(20.93)
0.01***
(8.02)
0.16***
(50.84)
7.51***
(30.99)
5.97***
(125.55)
2.03***
(5.26)
0.05
(0.10)
0.27***
(68.87)
0.01***
(11.03)
8.85***
(35.17)
0.00***
(2.59)
1.47***
(21.78)
0.01***
(7.88)
0.16***
(50.89)
7.27***
(30.00)
5.75***
(118.08)
0.26
(0.66)
0.06
(0.13)
0.27***
(67.92)
0.00***
(8.38)
8.48***
(33.66)
0.00***
(3.41)
1.73***
(25.12)
0.01***
(7.99)
0.15***
(49.62)
2.47***
(12.18)
3.51***
(89.12)
1.27***
(3.76)
0.71**
(2.26)
0.59***
(169.80)
0.00***
(3.33)
0.13
(0.61)
0.00***
(11.80)
0.99***
(17.05)
0.01***
(8.41)
0.65***
(239.22)
1.92***
(9.97)
3.19***
(72.33)
0.92*
(1.83)
0.92**
(2.40)
0.54***
(144.35)
0.00***
(3.46)
0.12
(0.42)
0.00***
(13.94)
0.81***
(12.96)
0.01***
(7.74)
0.59***
(192.00)
3.00***
(13.41)
3.68***
(88.50)
1.76***
(5.63)
0.52
(1.61)
0.58***
(159.61)
0.00***
(5.44)
0.16
(0.74)
0.00***
(11.39)
1.16***
(19.08)
0.01***
(7.72)
0.64***
(226.68)
2.78***
(12.41)
3.49***
(82.19)
0.30
(0.92)
0.56*
(1.74)
0.58***
(160.92)
0.00**
(2.38)
0.22
(0.99)
0.00***
(12.24)
1.40***
(22.56)
0.01***
(7.87)
0.64***
(228.16)
18.69***
(4.32)
25.16***
(36.28)
14.11***
(3.37)
2.53
(0.76)
2.61***
(37.51)
0.00
(0.13)
6.33*
(1.71)
0.02***
(5.46)
6.39***
(6.08)
0.02
(0.72)
2.78***
(52.76)
22.58***
(5.16)
24.88***
(36.57)
28.95***
(7.55)
1.37
(0.40)
2.45***
(36.49)
0.01
(1.44)
12.20***
(3.40)
0.02***
(5.83)
5.36***
(5.21)
0.02
(0.76)
2.66***
(52.07)
22.38***
(5.11)
25.24***
(36.92)
28.05***
(7.02)
0.82
(0.24)
2.49***
(36.98)
0.00
(0.06)
11.66***
(3.23)
0.02***
(5.81)
3.83***
(3.69)
0.02
(0.74)
2.71***
(52.95)
21.48***
(4.90)
24.67***
(35.45)
21.72***
(5.10)
0.98
(0.28)
2.52***
(37.16)
0.01
(0.91)
9.86***
(2.70)
0.02***
(5.66)
2.92***
(2.74)
0.03
(0.77)
2.73***
(53.02)
0.08***
(24.66)
0.03***
(19.82)
0.01***
(51.57)
0.08***
(24.12)
0.03***
(20.22)
0.01***
(51.99)
0.09***
(26.66)
0.03***
(19.19)
0.01***
(51.23)
0.00
(0.84)
0.02***
(18.67)
0.00***
(24.85)
0.01***
(3.10)
0.03***
(22.69)
0.00***
(24.66)
0.02***
(5.19)
0.02***
(18.37)
0.00***
(28.08)
0.02***
(8.29)
0.02***
(17.32)
0.00***
(27.10)
0.39***
(6.88)
0.01
(0.64)
0.04***
(11.00)
0.12**
(2.18)
0.04*
(1.95)
0.04***
(14.82)
0.21***
(3.88)
0.03*
(1.73)
0.04***
(15.43)
0.16***
(2.85)
0.03
(1.55)
0.04***
(15.08)
Observations
Number of banks
Log-likelihood function
Wald v2
47,424
4726
83,349
32,779***
47,424
4726
83,306
32,870***
47,425
4726
83,197
33,091***
46,631
4664
72,306
40,395***
38,512
3845
56,839
28,160***
47,440
4728
78,457
38,218***
47,441
4728
78,348
38,438***
46,609
4664
213,175
3886***
47,416
4728
217,912
4238***
47,416
4728
217,989
4085***
47,417
4728
217,987
4098***
46,615
4662
74,308
35,667***
827
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
0.05***
(18.05)
CR4 (assets)
Parent banks performance
Protability level
(1)
Independent variables
828
Table 5
The effects of parent banks protability on foreign banks performance: protability ranking.
Dependent variables: bank protability
Net interest margins (NIMs)
(1)
(2)
***
Intercept
Foreign banks ownership
6.34
(57.79)
***
0.70
(7.73)
***
5.41
(43.98)
***
0.74
(6.63)
(4)
***
6.74
(54.27)
***
0.71
(7.48)
(5)
***
9.44
(72.92)
***
0.38
(4.22)
(6)
***
4.56
(52.49)
***
0.24
(4.11)
***
3.56
(36.44)
***
0.28
(3.72)
(8)
***
4.64
(48.63)
***
0.26
(4.38)
0.02***
(10.56)
0.05***
(9.10)
HHI (assets)
0.06***
(11.23)
***
***
***
0.11
(33.00)
***
***
***
***
0.38
(5.15)
**
***
**
***
0.30
(20.95)
0.32
(20.41)
0.28
(17.78)
0.44
(29.33)
0.05
(4.32)
0.09
(7.54)
0.05
(4.25)
0.19
(16.14)
0.22
(1.98)
0.01
(19.13)
0.26
(2.05)
2.55
(7.17)
3.96***
(14.53)
0.54***
(21.08)
***
4.50
(8.19)
***
1.62
(2.61)
***
0.20
(45.60)
0.00***
(4.08)
9.26***
(30.50)
0.00***
(9.36)
***
6.27
(96.70)
***
0.02
(13.51)
***
0.11
(32.41)
1.48***
(5.56)
0.42***
(15.17)
**
1.87
(2.29)
***
2.61
(3.57)
***
0.16
(33.08)
0.00***
(4.77)
7.90***
(20.13)
0.00***
(7.03)
***
5.79
(78.64)
***
0.01
(5.80)
***
0.12
(31.65)
3.11***
(10.08)
0.57***
(20.67)
***
4.25
(8.04)
**
1.57
(2.36)
***
0.20
(41.32)
0.00
(0.87)
9.25***
(28.40)
0.00***
(9.00)
***
6.27
(88.42)
***
0.01
(5.30)
***
0.14
(39.79)
2.78***
(9.15)
0.36***
(13.62)
**
1.15
(2.27)
**
1.53
(2.43)
***
0.19
(39.94)
0.00***
(4.37)
8.24***
(25.71)
0.00***
(11.41)
***
6.57
(93.89)
***
0.01
(4.80)
***
0.13
(36.99)
0.97***
(3.80)
0.22***
(11.90)
***
3.01
(6.84)
***
1.19
(2.96)
***
0.66
(159.99)
0.00***
(7.36)
1.40***
(5.16)
0.01***
(17.02)
***
4.34
(74.61)
***
0.02
(10.19)
***
0.70
(225.81)
0.49**
(2.03)
0.18***
(8.46)
***
1.87
(2.85)
***
1.65
(3.32)
***
0.64
(151.00)
0.00***
(7.33)
0.76**
(2.20)
0.01***
(17.22)
***
3.75
(57.88)
***
0.01
(8.16)
***
0.67
(193.58)
1.08***
(3.98)
0.24***
(12.40)
***
3.00
(7.37)
***
1.13
(2.68)
***
0.65
(155.77)
0.00***
(5.45)
1.34***
(4.88)
0.01***
(16.67)
***
4.46
(74.45)
***
0.01
(8.23)
***
0.68
(216.65)
0.87***
(3.25)
0.15***
(8.06)
**
0.83
(2.04)
***
1.21
(2.91)
***
0.66
(159.72)
0.00***
(8.06)
0.48*
(1.76)
0.01***
(18.66)
***
4.64
(78.13)
***
0.01
(7.80)
***
0.69
(221.73)
1.06
(0.22)
0.21
(1.11)
***
35.78
(7.30)
4.66
(1.23)
***
1.50
(20.63)
0.06***
(8.10)
21.09***
(5.14)
0.02***
(4.18)
***
12.12
(13.37)
***
0.16
(5.76)
***
1.89
(35.23)
2.09
(0.44)
0.22
(1.24)
***
54.69
(12.52)
4.85
(1.25)
***
1.41
(20.35)
0.02***
(2.84)
28.14***
(7.09)
0.01***
(2.95)
***
9.84
(11.07)
***
0.12
(4.51)
***
1.67
(32.26)
3.98
(0.82)
0.18
(0.93)
***
40.47
(8.57)
2.72
(0.69)
***
1.57
(22.40)
0.06***
(7.50)
28.16***
(7.01)
0.02***
(4.44)
***
13.40
(14.84)
***
0.12
(4.21)
***
1.83
(35.23)
3.45
(0.37)
3.42***
(6.41)
***
54.92
(7.18)
2.97
(0.55)
***
1.21
(9.35)
0.11***
(6.98)
24.74***
(3.45)
0.03***
(3.40)
***
12.98
(7.08)
0.04
(0.69)
***
1.42
(15.04)
0.06***
(11.90)
0.10***
(50.91)
***
0.01
(53.00)
0.07***
(16.93)
0.07***
(39.20)
***
0.01
(50.61)
0.09***
(21.66)
0.06***
(35.87)
***
0.01
(49.30)
0.02***
(5.84)
0.06***
(39.60)
***
0.01
(27.67)
0.01**
(2.32)
0.07***
(45.54)
***
0.01
(28.53)
0.00
(0.91)
0.05***
(37.60)
***
0.01
(32.32)
0.01***
(3.35)
0.05***
(34.41)
***
0.00
(30.71)
0.42***
(6.47)
0.00
(0.00)
***
0.05
(12.22)
0.18***
(2.85)
0.04**
(2.03)
***
0.04
(15.85)
0.22***
(3.45)
0.06***
(2.70)
***
0.04
(16.07)
0.07
(0.61)
0.10**
(2.35)
***
0.04
(9.14)
Observations
Number of banks
Log-likelihood function
Wald v2
38,761
3847
71,786
13,224
47,864
4731
97,131
15,694
47,865
4731
96,331
17,298
47,049
4669
84,545
40,569
38,777
3849
67,137
30,221
47,880
4733
89,719
38,009
47,881
4733
89,207
39,036
47,022
4669
221,063
1744
47,865
4735
225,813
2497
47,852
4733
226,064
1885
16,699
1990
85,559
592
Numbers in parenthesis
*
Statistical signicant
**
Statistical signicant
***
Statistical signicant
47,033
4667
89,050
17,770
are t-values.
at level of 10%.
at level of 5%.
at level of 1%.
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
***
***
23.44
(9.18)
**
2.54
(2.40)
11.64***
(3.47)
1.22
(7.07)
0.16
(41.27)
***
(12)
***
12.60
(10.95)
***
1.70
(3.23)
15.27***
(15.28)
***
CR4 (assets)
(11)
***
9.25
(9.08)
***
2.97
(5.77)
***
1.54
(7.43)
***
(10)
***
12.18
(11.35)
**
1.33
(2.57)
0.35***
(7.91)
***
(9)
***
6.50
(64.39)
*
0.10
(1.72)
Table 6
The impacts of home-country characteristics on foreign bank protability.
Independent variables
Macroeconomic environment
Country risk
GDP
Legal risk
Economic risk
Control variables
0.90*
(1.92)
2.73***
(45.86)
0.35*
(1.85)
Yes
0.56***
(7.76)
2.86***
(142.27)
0.01
(0.67)
Yes
0.64***
(8.90)
2.74***
(132.25)
0.00
(0.95)
Yes
0.56***
(8.27)
2.85***
(141.63)
0.00
(1.45)
Yes
4.75**
(2.07)
2.54***
(23.04)
2.50**
(2.29)
Yes
6.91**
(2.52)
2.54***
(23.03)
4.05**
(2.33)
Yes
Observations
Number of banks
Log-likelihood function
Wald v2
6496
606
13,815
4058
50,416
4741
91,378
30,948
47,068
4696
83,848
28,369
50,255
4733
91,072
30,679
1995
313
4,489
1597
1995
313
4,489
1597
Control of corruption
Regulatory quality
Government effectiveness
Bank capital
Ownership restrictiveness
Control variables
0.58***
(7.11)
2.80***
(116.67)
0.04
(0.63)
Yes
0.47***
(4.84)
2.80***
(116.71)
0.13**
(1.98)
Yes
0.48***
(4.39)
2.80***
(116.71)
0.09
(1.47)
Yes
0.91***
(10.37)
2.72***
(96.47)
0.39***
(4.96)
Yes
0.58***
(6.95)
2.72***
(96.43)
0.10*
(1.81)
Yes
0.51***
(4.21)
2.72***
(96.39)
0.15
(1.55)
Yes
Observations
Number of banks
Log-likelihood function
Wald v2
37,028
4753
66,805
21,229
37,028
4753
66,803
21,232
37,028
4753
66,804
21,231
27,943
4622
51,426
16,724
27,943
4622
51,436
16,703
27,943
4622
51,437
16,702
Governance
Bank supervision
Bank supervision
Securities activities
Certied auditor
Auditors report
Consolidated accounts
Solvency intervention
Onsite inspections
Control variables
0.67***
(8.80)
2.72***
(96.38)
0.02
(0.34)
Yes
0.52***
(4.75)
2.72***
(96.39)
0.12*
(1.78)
Yes
0.54**
(2.50)
2.72***
(96.40)
0.12
(0.58)
Yes
0.66***
(7.90)
2.77***
(97.10)
0.03
(0.40)
Yes
0.65***
(8.05)
2.72***
(96.31)
0.02
(0.29)
Yes
0.74***
(10.00)
2.71***
(96.28)
0.02***
(4.39)
Yes
Observations
Number of banks
Log-likelihood function
Wald v2
27,943
4622
51,438
16,699
27,943
4622
51,436
16,702
27,943
4622
51,438
16,700
27,795
4622
51,118
16,778
27,908
4622
51,338
16,708
27,914
4622
51,367
16,686
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
Developing country
PR H-statistics
Lerner index
HHI (assets)
CR4 (assets)
0.5436***
(8.57)
2.3968***
(132.12)
0.004**
0.61***
(8.90)
2.63***
(128.18)
0.53***
0.38***
(4.03)
2.86***
(142.07)
1.09***
1.22***
(7.98)
2.63***
(128.13)
0.03***
(continued on next page)
829
Dependent variable is net interest margin (NIM). Numbers in parenthesis are t-values.
Control variables for empirical estimation include intercept, operation cost, bank size, market share, credit risk, interest payments, liquidity, opportunity cost, capital strength, non-interest revenues, off-balance sheet, and other
operating income.
*
Statistical signicant at level of 10%.
**
Statistical signicant at level of 5%.
***
Statistical signicant at level of 1%.
47,425
4726
83,350
32,787
50,839
4778
92,513
31,203
47,424
4726
83,346
32,785
46,615
4662
74,312
35,659
Observations
Number of banks
Log-likelihood function
Wald v2
Control variables
(2.09)
Yes
HHI (assets)
Lerner index
(4.66)
Yes
PR H-statistics
(3.03)
Yes
CR4 (assets)
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
(4.06)
Yes
830
to earn lower prots, while smaller banks tend to earn higher profits. In other words, smaller banks experience economies of scale
and scope, while larger banks experience diseconomies of scale
and scope. These ndings are consistent with studies conducted
by Kosmidou et al. (2007) and Pasiouras and Kosmidou (2007).
Another nding in our study is that non-interest revenues are
signicantly negative. This means that operation income from
non-interest operating will decrease income, and other operation
activities will decrease, too. Banks with higher protability are
positively and signicantly associated with the ratio of market
share. More importantly, based on the results of the NIM and
ROA models, the relationship between bank protability and capital ratio is signicantly positive. However, in the ROE model the
relationship between bank protability and capital ratio is both
negative and signicant.
In Eq. (10), we analyze parent bank performance in home country in order to determine whether foreign banks are more protable than domestic banks. Table 4 presents the results of using
the protability level of parent bank in home country. In columns
(1)(4), the dependent variable is NIM. The coefcient on the profitability level of parent bank is 2.40, which is at a signicant level
of 1% when using PR H statistics. This implies that when a parent
bank has high levels of protability at home, its foreign iterations
will, on average, perform better than domestic banks in the host
country. Alternative measures such as the Lerner index, HHI, and
CR4 are signicant, which is similar to the ndings in Table 3.
When using ROA and ROE as dependent variables, the coefcients
of protability level of parent bank maintain an economic signicance at the 1% level.
However, protability level of parent bank in home country
does not reect the relative performance of their counterparts in
the home country. Hence, we apply another comparative indicator
as protability ranking in home country. To do so, we dene the
parent banks protability ranking as: the higher the protability
in comparison to its counterpart in its home country, the higher
the ranking for parent banks. Table 5 presents the results of our
regression using protability ranking for parent banks in home
country to explain their comparative performance. Parent banks
with higher protability rankings at home tend to experience higher protability rankings in their foreign banks than do domestic
banks in the host country. There was almost no change in our estimates when we used alternative measures of banking market
structure. Virtually all coefcients remain signicant at the 1%
and 5% level; however, the estimates are much smaller than those
in Table 4.
5.2. Home-country impacts of macroeconomic condition, governance,
and supervision on bank protability
After controlling for bank nancial characteristics and focusing
on the results from NIM as a dependent variable, we next examine
whether home-country banking market structure, macroeconomic
conditions, governance and supervision of foreign banks have any
spillover effects on NIM in the host country. To do so, we use interaction terms for a foreign banks ownership with home-country
characteristics, including: developing country, macroeconomic
conditions, country risk, governance, and bank supervision.
Table 6 presents the results of several robustness specications.
Because some country indexes are highly correlated with each
other, we include the indexes individually in the models. Each column of each panel reports coefcients from a single regression in
which the dependent variable is NIM. The control variables are
the same as in Table 3; however, to conserve space we only report
select coefcients. The empirical results are consistent with our
previous ndings. All major explanatory and control variables
maintain their sign and signicance as before. Overall, we conrm
Table 7
The Impacts of Host-Country Characteristics on Foreign Bank protability.
Independent variables
Macroeconomic environment
Country risk
GDP
Legal risk
Economic risk
Control variables
0.47***
(5.58)
2.85***
(141.67)
0.31**
(2.36)
Yes
0.94***
(13.26)
2.85***
(141.56)
0.11***
(16.97)
Yes
0.77***
(11.17)
2.82***
(134.41)
0.02***
(10.13)
Yes
0.55***
(8.00)
2.84***
(141.49)
0.01***
(26.93)
Yes
0.42***
(2.98)
2.40***
(99.28)
0.04
(0.55)
Yes
1.26***
(4.35)
2.39***
(99.01)
0.99***
(3.82)
Yes
Observations
Number of banks
Log-likelihood function
Wald v2
50,841
4778
92,521
31,194
50,712
4766
92,181
31,436
47,509
4728
85,493
29,267
50,611
4764
91,696
31,550
27,539
4565
45,233
17,087
27,539
4565
45,212
17,130
Control of corruption
Regulatory quality
Government effectiveness
Bank capital
Ownership restrictiveness
Control variables
0.86***
(11.01)
2.79***
(115.92)
0.42***
(7.24)
Yes
1.09***
(11.74)
2.79***
(116.10)
0.49***
(7.89)
Yes
1.29***
(13.00)
2.78***
(115.73)
0.57***
(9.73)
Yes
0.55***
(2.84)
2.62***
(93.80)
0.07
(0.38)
Yes
0.65***
(7.48)
2.62***
(93.76)
0.03
(0.66)
Yes
2.97***
(6.49)
2.62***
(93.79)
3.61***
(7.93)
Yes
Observations
Number of banks
Log-likelihood function
Wald v2
37,029
4753
66,781
21,280
37,029
4753
66,776
21,289
37,029
4753
66,760
21,322
27,914
4616
50,709
15,835
27,914
4616
50,709
15,836
27,914
4616
50,678
15,898
Governance
Bank supervision
Bank Supervision
Securities activities
Certied auditor
Auditors report
Consolidated accounts
Solvency intervention
Onsite inspections
Control variables
0.66***
(8.40)
2.62***
(93.72)
0.07
(1.49)
Yes
0.78***
(4.93)
2.62***
(93.76)
0.16
(1.15)
Yes
0.59***
(4.27)
2.62***
(93.80)
0.03
(0.24)
Yes
0.41***
(3.07)
2.67***
(94.36)
0.23*
(1.88)
Yes
0.59***
(7.55)
2.62***
(93.57)
0.07
(0.90)
Yes
0.70***
(7.00)
2.61***
(93.60)
0.06
(1.25)
Yes
Observations
Number of banks
Log-likelihood function
Wald v2
27,914
4616
50,708
15,838
27,914
4616
50,708
15,837
27,914
4616
50,709
15,835
27,754
4616
50,375
15,870
27,867
4616
50,590
15,802
27,879
4616
50,634
15,792
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
Developing country
PR H-statistics
Lerner index
HHI (assets)
CR4 (assets)
0.54***
(8.57)
2.40***
(132.12)
0.00**
0.61***
(8.90)
2.63***
(128.18)
0.53***
0.38***
(4.03)
2.86***
(142.07)
1.09***
1.22***
(7.98)
2.63***
(128.13)
0.03***
(continued on next page)
831
Observations
Number of banks
Log-likelihood function
Wald v2
Dependent variable is net interest margin (NIM). Numbers in parenthesis are t-values.
Control variables for empirical estimation include intercept, operation cost, bank size, market share, credit risk, interest payments, liquidity, opportunity cost, capital strength, non-interest revenues, off-balance sheet, and other
operating income.
*
Statistical signicant at level of 10%.
**
Statistical signicant at level of 5%.
***
Statistical signicant at level of 1%.
47,424
4726
83,346
32,785
47,425
4726
83,350
32,787
50,839
4778
92,513
31,203
(4.66)
Yes
46,615
4662
74,312
35,659
HHI (assets)
(3.03)
Yes
Lerner index
PR H-statistics
(2.09)
Yes
Control variables
Table 7 (continued)
CR4 (assets)
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
(4.06)
Yes
832
833
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
Table 8
Joint effects of host- and home-country characteristics on foreign bank protability.
Dependent variables
Lerner index
3.96***
1.80***
(4.33)
(23.34)
4.66***
1.72***
(4.65)
(20.77)
8.57***
2.03***
(8.60)
(24.73)
8.19***
2.01***
(7.70)
(24.34)
0.12
3.96
4.36*
(0.43)
(1.51)
(1.71)
0.30
3.20
3.76
(1.06)
(1.16)
(1.40)
0.30
2.66
3.20
(1.20)
(1.06)
(1.30)
0.29
2.62
3.29
(1.16)
(1.03)
(1.33)
Governance
FBO regulatory quality
0.00
(0.04)
0.15
(1.09)
0.19
(1.55)
0.19
(1.49)
Bank supervision
FBO bank capital requirements
FBO bank ownership restrictiveness
FBO certied auditor
FBO onsite inspections
0.38*
0.07
0.11
0.01
(1.90)
(0.93)
(0.76)
(1.28)
0.41*
0.10
0.16
0.01
(1.83)
(1.13)
(1.00)
(0.73)
0.31
0.07
0.16
0.00
(1.41)
(0.88)
(1.01)
(0.11)
0.42**
0.08
0.17
0.01
(1.96)
(0.96)
(1.11)
(0.53)
1.94***
(3.20)
0.48***
(3.34)
6.37***
(14.45)
0.84***
(13.39)
0.47
0.06***
0.04***
0.09***
0.23*
(1.58)
(4.78)
(6.35)
(11.25)
(1.93)
0.41
0.03**
0.05***
0.12***
0.09
(1.35)
(2.29)
(7.98)
(47.73)
(0.71)
0.07
0.02*
0.04***
0.12***
0.17
(0.26)
(1.70)
(7.36)
(47.95)
(1.54)
0.05
0.02*
0.04***
0.12***
0.17
(0.18)
(1.76)
(7.44)
(47.82)
(1.48)
Governance
FBO political instability
FBO regulatory quality
FBO government effectiveness
0.18
0.03
0.15
(1.40)
(0.18)
(0.75)
0.03
0.06
0.15
(0.23)
(0.30)
(0.67)
0.17
0.12
0.17
(1.31)
(0.60)
(0.80)
0.14
0.07
0.16
(1.07)
(0.38)
(0.77)
Bank supervision
FBO basel risk weights
FBO consolidated accounts
4.18***
0.09
(6.06)
(0.56)
3.72***
0.30*
(4.69)
(1.69)
3.28***
0.25
(4.25)
(1.43)
3.32***
0.25
(4.27)
(1.44)
2.10***
(3.49)
3.67***
(3.39)
47.19***
(14.75)
0.11***
(13.84)
Observations
Number of banks
Log-likelihood function
Wald v2
2197
499
3947
3871***
2380
524
4579
7160***
HHI (assets)
2379
524
4488
8280***
CR4 (assets)
2380
524
4502
8140***
834
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
Table 9
Joint effects of supervision changing in host- and home-country on foreign bank protability.
Dependent variables
Lerner index
5.84***
0.66***
(12.23)
(7.90)
8.43***
1.02***
(15.54)
(10.78)
10.70***
1.26***
(17.00)
(12.78)
10.58***
1.24***
(15.11)
(12.62)
0.10
2.75
2.38
(0.42)
(1.35)
(1.22)
0.29
4.49*
4.03
(1.05)
(1.76)
(1.64)
0.25
4.20*
3.82
(0.94)
(1.66)
(1.56)
0.24
4.21*
3.83
(0.92)
(1.66)
(1.56)
Governance
FBO regulatory quality
0.04
(0.39)
0.44***
(3.24)
0.48***
(3.55)
0.47***
(3.52)
Bank supervision
FBO D bank capital requirements
FBO D bank ownership restrictiveness
FBO D certied auditor
FBO D onsite inspections
0.62
0.49***
0.98*
0.03
(1.31)
(2.74)
(1.72)
(1.29)
0.22
0.62***
0.93
0.01
(0.36)
(2.74)
(1.27)
(0.34)
0.20
0.56**
0.88
0.01
(0.32)
(2.48)
(1.21)
(0.23)
0.21
0.56**
0.89
0.01
(0.34)
(2.46)
(1.22)
(0.23)
2.37***
(3.04)
0.62***
(2.78)
4.47***
(8.45)
0.62***
(8.20)
0.59**
0.00
0.05***
0.03***
0.27***
(2.17)
(0.14)
(7.90)
(4.35)
(6.69)
0.22
0.05***
0.00
0.00***
0.36***
(0.68)
(3.35)
(0.52)
(2.64)
(7.47)
0.08
0.05***
0.00
0.00**
0.35***
(0.26)
(3.53)
(0.22)
(2.08)
(7.42)
0.05
0.05***
0.00
0.00**
0.36***
(0.13)
(3.52)
(0.28)
(2.12)
(7.41)
Governance
FBO political instability
FBO regulatory quality
FBO government effectiveness
0.20
0.46***
0.64***
(1.51)
(3.29)
(4.03)
0.20
0.49***
0.87***
(1.29)
(2.93)
(4.57)
0.18
0.41**
0.85***
(1.19)
(2.42)
(4.49)
0.18
0.42**
0.84***
(1.19)
(2.49)
(4.44)
Bank supervision
FBO D basel risk weights
FBO D consolidated accounts
4.20***
0.40
(4.00)
(1.45)
4.56***
0.37
(3.40)
(1.09)
4.50***
0.37
(3.36)
(1.08)
4.49***
0.36
(3.34)
(1.07)
2.47***
(3.18)
5.47***
(3.23)
33.53***
(8.87)
0.08***
(8.46)
Observations
Number of banks
Log-likelihood function
Wald v2
3718
556
7794
2469***
3956
581
9251
2313***
HHI (assets)
3956
581
9224
2525***
CR4 (assets)
3956
581
9227
2512***
835
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
Table A1
Empirical results from previous studies.
Explanatory variable
Default risk
No relation with
A
Angbazo (1997)
Liquidity risk
Capital adequacy
Angbazo (1997)N
Angbazo (1997)N
Williams (2003)A
Angbazo (1997)N
Size2
Ownership
Concentration
Cyclical output
Foreign strategic
Government-owned
International participation
Foreign bank share
Ination expectations
Overhead/ta
Per capital income
Market share
X-efciency
S-efciency
Dem. and sav. deposits
GDP growth
Williams
Williams
Williams
Williams
Williams
Home GDP
Home NIM
Licence
Home GDP growth
Asset growth
Yearly growth of deposits
Williams (2003)A
Athanasoglou et al. (2008)A
Maudos and Fernndez de Guevara
(2004)N
Athanasoglou et al. (2008)A
Dietrich and Wanzenried (2009)A
Athanasoglou et al. (2008)A
Athanasoglou et al. (2008)A
Exports
GDP growth rates difference
Parents banks ROA
Parents banks SIZE
Experience in the host nation
C5
Loan loss provisions
Lerner index
Capital ow
Williams (1998b)A
Williams (2003)A
(1998b)A
(2003)A
(1998b)A
(1998b)A
(2003)A
Williams (2003)A
Williams (2003)A
Dietrich and Wanzenried (2009)A
(continued on next page)
836
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
Table A1 (continued)
Explanatory variable
No relation with
Default risk
Angbazo (1997)N
Table B1
Empirical studies using PanzarRosse model.
Authors (published year)
Period
Countries
Empirical ndings
Shaffer (1982)
Nathan and Neave (1989)
Lloyd-Williams et al. (1991)
Molyneux et al. (1994)
1979
19821984
19861988
19861989
Monopolistic competition
1982: perfect competition; 19831984: monopolistic competition
Monopoly
Monopoly: Italy; monopolistic competition: France, Germany, Spain, and UK
19701986
19851992
19861988
19881996
19891991,
19941996
19871994
19931995
19891996
19921996
19881998
19931997
23 OECD countries
Arab Middle East countries
19942000
Hempell (2002)
Coccorese (2004)
Claessens and Laeven (2004)
19931998
19971999
19942001
19982002
19832002
19962000
19932002
19802004
19932000
19922004
19952004
20002006
Switzerland
Greece
15 EU countries
France, Germany, and Italy
the opposite effect on foreign bank margins. Overall, the host country effect of banking market structure plays an important role in
explaining why foreign banks are more protable than domestic
banks when competition in a host country is weaker. This result
is extremely robust when we use alternative measures of banking
market structure.
Table C1
Variables denition and expected sign.
Description
Notation
Dependent variables
Net interest margin
Return on assets
Return on equity
NIM
ROA
ROE
Dummy variable equals to one for foreign banks and 0 for domestic banks
Calculated as equity/liability
The natural logarithm of the accounting value of the total assets
COST is total operation expenses provides information on the efciency of the management
A measure of liquidity, calculated as loans divided by customers and short-term funding
This variable is proxied by the ratio of liquid reserves/total assets
Ratio of operating expenses minus non-interest income to total assets reect extra payments to depositors through service charge remission or other types of transfers due
to competition in the market for deposits
The credit risk is calculated as loan-loss provisions to loans ratio
This variable is proxied by non-interest revenues/total revenue
This variable is calculated as off-balance sheet/total assets
It is a ratio of its assets relative to the total assets of the banking market
Other operating income/total assets
FB
EL
SIZE
COST
LI
OPC
IP
+
+
+
+
+
CR
NOI
OBS
MS
OOI
+
+
Estimates competitive behavior of banks on the basis of the comparative static properties of reduced-form revenue equations
A country-level indicator of bank competition, measured by the Lerner index, which is calculated as the average bank-level measure of the mark-up of price over marginal
costs, with higher values indicating less competition in the banking sector
The Herndhal index measures market concentration and equals the sum of the squares of each banks market share in total industry assets
The CR4 concentration measure calculated by dividing the assets of the ve largest banks with the assets of all banks operating in the market
PR H
LER
HHIA
CR4A
+
+
Macroeconomic environment
GDP growth
Growth in nation GDP
Ination rate
Current period ination rate (consumer prices)
Real interest rate
Real interest rate
Developing countries
Dummy variable equals to 1 if a bank locates in developing country
GDP
INF
IR
DC
Country risk
Legal risk
Economic risk
LR
ER
GE
Dened as the ability of the government to formulate and implement sound policies and regulations that permit and promote private sector development
Dened as the extent to which public power is exercised for private gain, including both petty and grand forms of corruption
RQ
CO
+
+
Dummy variable that equals to 1 if related parties can own capital in a bank
CAP
Dummy variable that equals to 1 if regulatory restrictiveness of ownership by one nancial rms of banks
OWN
Dummy variable that equals to1 if it is risk-weighted in line with Basle guidelines
What are the conditions under which banks can engage in securities activities?
Dummy variable that equals to 1 if auditors licensed or certied
BASEL
SEC
AUDLIC
+
+
AUDREP
CACCS
SOLACT
ONINS
+
+
+
+
Independent variables
Bank-specic
Foreign banks
Capital strength
Bank size
Operation cost
Liquidity
Opportunity cost
Interest payments
Credit risk
Non-interest revenues
Off balance sheet
Market share
Other operating inc.
Industry-specic
PR H statistics
Lerner index
HHI assets
CR4 assets
Governance
Government
effectiveness
Regulatory quality
Control of corruption
Bank supervision
Bank capital
requirement
Bank ownership
restrictiveness
Basel risk weights
Securities activities
Licensed/certied
auditor
Auditors report
Consolidated accounts
Solvency intervention
Onsite inspections
Expected
sign.
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
Variables
837
838
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
S.-H. Chen, C.-C. Liao / Journal of Banking & Finance 35 (2011) 819839
Ho, T.S.Y., Saunders, A., 1998. The determinants of bank interest margins: theory
and empirical evidence. Journal of Financial and Quantitative Analysis 16, 581
600.
Hondroyiannis, G., Lolos, S., Papetrou, E., 1999. Assessing competitive conditions in
the Greek banking system. Journal of International Financial Markets,
Institutions and Money 9, 377391.
Jaffee, A.B., 1989. Real effects of academic research. American Economic Review 79,
957970.
Kaufmann, D., Kraay, A., Mastruzzi, M., 2007. Governance matters VI: Governance
indicators for 19962006. World Bank Policy Research, Working Paper, No. 4280.
Kosmidou, K., Pasiouras, F., Doumpos, M., Zopounidis, C., 2004. Foreign versus
domestic banks performance in the UK: a multicriteria approach.
Computational Management Science 1, 329343.
Kosmidou, K., Pasiouras, F., Tsaklanganos, A., 2007. Domestic and multinational
determinants of foreign bank prots: the case of Greek banks operating abroad.
Journal of Multinational Financial Management 17, 15.
Lloyd-Williams, D.M., Molyneux, P., Thornton, J., 1991. Competition and
contestability in the Japanese commercial banking market. Institute of
European Finance, Research Papers in Banking and Finance, No. 16.
Mahajan, A., Rangan, N., Zardkoohi, A., 1996. Cost structures in multinational and
domestic banking. Journal of Banking and Finance 20, 238306.
Mamatzakis, E., Staikouras, C., Koutsomanoli-Fillipaki, N., 2005. Competition and
concentration in the banking sector of the south eastern European region.
Emerging Markets Review 6, 192209.
Matthews, K., Murinde, V., Zhao, T., 2007. Competitive conditions among the major
British banks. Journal of Banking and Finance 31, 20252042.
Maudos, J., Fernndez de Guevara, J., 2004. Factors explaining the interest margin in
the banking sectors of the European Union. Journal of Banking and Finance 28,
22592281.
Maudos, J., Solisa, L., 2009. The determinants of net interest income in the Mexican
banking system: an integrated model. Journal of Banking and Finance 33, 1920
1931.
Mian, A., 2003. Foreign, private domestic and government banks: new evidence
from emerging markets. Mimeo, University of Chicago.
Minh To, H., Tripe, D., 2002. Factors inuencing the performance of foreign owned
banks in New Zealand. Journal of International Financial Markets, Institutions
and Money 12, 341357.
Molyneux, P., Lloyd-Williams, D.M., Thornton, J., 1994. Competitive conditions in
European banking. Journal of Banking and Finance 18, 445459.
Molyneux, P., Thornton, J., Lloyd-Williams, D.M., 1996. Competition and market
contestability in Japanese commercial banking. Journal of Economics and
Business 48, 3345.
Molyneux, P., Thorton, J., 1992. Determinants of European bank protability: a note.
Journal of Banking and Finance 16, 11731178.
Murjan, W., Ruza, C., 2002. The competitive nature of the Arab Middle Eastern
banking markets. International Advances in Economics 8, 267274.
Nathan, A., Neave, E.H., 1989. Competition and contestability in Canada s nancial
system: empirical results. Canadian Journal of Economics 22, 576594.
Niimi, K. 1998. Big-bang to waga kuni ginkogyo: Jiyuka to rireshonshipppu
rendingu. Japan Research Review (in Japanese).
Nikiel, E.M., Opiela, T.P., 2002. Customer type and bank efciency in Poland:
implication for emerging market banking. Contemporary Economic Policy 20,
255271.
Panzar, J.C., Rosse, J.N., 1987. Testing for monopoly equilibrium. Journal of
Industrial Economics 35, 443456.
Park, K.H., 2009. Has bank consolidation in Korea lessened competition? Quarterly
Review of Economics and Finance 49, 651667.
839
Pasiouras, F., Kosmidou, K., 2007. Factors inuencing the protability of domestic
and commercial banks in the European Union. Journal of Banking and Finance
21, 222237.
Peek, J., Rosengren, E.S., Kasirye, F., 1999. The poor performance of foreign bank
subsidiaries: Were the problems acquired or created? Journal of Banking and
Finance 23, 579604.
Peria, M.S.M., Mody, A., 2004. How foreign participation and market concentration
impact bank spreads: evidence from Latin America. Journal of Money, Credit
and Banking 36, 511537.
Rime, B., 1999. Mesure du degr de concurrence dans le systme bancaire suisse
laide du modle de Panzar et Rosse. Swiss Journal of Economics and Statistics
135, 2140.
Sathye, M., 2001. X-efciency in Australian banking: an empirical investigation.
Journal of Banking and Finance 25, 613630.
Saunders, A., Schumacher, L., 2000. The determinants of bank interest rate margins:
an international study. Journal of International Money and Finance 19, 813
832.
Shaffer, S., 1982. A non-structural test for competition in nancial markets. In:
Proceedings of a Conference on Bank Structure and Competition, Federal
Reserve Bank of Chicago, pp. 225243.
Shaffer, S., DiSalvo, J., 1994. Conduct in banking duopoly. Journal of Banking and
Finance 18, 10631082.
Spathis, C., Kosmidoua, K., Doumposa, M., 2002. Assessing protability factors in the
Greek banking system: a multicriteria methodology. International Transactions
in Operational Research 9, 517530.
Staikouras, C., Wood, G., 2003. The determinants of bank protability in Europe. In:
Paper presented at the Proceedings of the European Applied Business Research
Conference, Venice, Italy, June 913.
Sturm, J.-E., Williams, B., 2004. Foreign bank entry, deregulation and bank
efciency: lessons from the Australian experience. Journal of Banking and
Finance 28, 17751799.
Thorne, A., 1993. Eastern Europes experience with banking reform: Is there a role
for banks in the transition? Journal of Banking and Finance 17, 9591000.
Tsutsui, Y., Kamesaka, A., 2005. Degree of competition in the Japanese securities
industry. Journal of Economics and Business 57, 360374.
Turk-Ariss, R., 2009. Competitive behavior in Middle East and North Africa banking
systems. Quarterly Review of Economics and Finance 49, 693710.
Ursacki, T., Vertinsky, I., 1992. Choice of entry timing and scale by foreign banks in
Japan and Korea. Journal of Banking and Finance 16, 405421.
Vennet, Rudi V., 1996. The effect of mergers and acquisitions on the efciency and
protability of EC credit institutions. Journal of Banking and Finance 20, 1531
1558.
Vesala, J., 1995. Testing for competition in banking: behavioural evidence from
Finland. Bank of Finland Studies, E: 1.
Williams, B., 1996. Determinants of the performance of Japanese nancial
institutions in Australia 19871992. Applied Economics 28, 11531165.
Williams, B., 1998a. Factors affecting the performance of foreign-owned banks in
Australia: a cross-sectional study. Journal of Banking and Finance 22, 197219.
Williams, B., 1998b. A pooled study of the prots and size of foreign banks in
Australia. Journal of Multinational Financial Management 8, 211231.
Williams, B., 2003. Domestic and international determinants of bank prots: foreign
banks in Australia. Journal of Banking and Finance 27, 11851210.
Yildirim, H.S., Philippatos, G.C., 2007. Restructuring, consolidation and competition
in Latin American banking markets. Journal of Banking and Finance 31, 629
639.
Yuan, Y., 2006. The state of competition of the Chinese banking industry. Journal of
Asian Economics 17, 519534.