2007 MS Guevara - Explanatory Factors of Market Power in The Banking System
2007 MS Guevara - Explanatory Factors of Market Power in The Banking System
2007 MS Guevara - Explanatory Factors of Market Power in The Banking System
3 June 2007
1463–6786 275–296
by
JUAN FERNÁNDEZ DE GUEVARA
Instituto Valenciano de Investigaciones Economicas (Ivie)
and
JOAQUÍN MAUDOS†
Universitat de València i Ivie
1 Introduction
In recent years, European banks have suffered an alteration of the competi-
tive conditions of their markets as a consequence of various factors. The
process of deregulation has created a scenario of greater competition follow-
ing the implementation of measures such as the abolition of compulsory
investment coefficients, the liberalization of interest rates, freedom of es-
tablishment etc. Furthermore, the increasing level of integration of the
European financial markets has also increased the intensity of competition
following the adoption of various measures (most recently, the Financial
Services Action Plan of the European Commission) aimed at eliminating the
barriers or obstacles (legal, fiscal, institutional etc.) that protect national
markets from outside competition.
The events described above have coexisted with a process of consolida-
tion of national markets as a consequence of a wave of mergers and acqui-
sitions, which have increased the degree of concentration of the European
banking markets. Although cross-border mergers and acquisitions can help
to increase the degree of integration and competition of financial markets,
the evidence shows a clear predominance of domestic mergers (with the
* Manuscript received 30.9.05; final version received 10.10.06.
†
The authors gratefully acknowledge two anonymous referees for their helpful comments.
Joaquín Maudos acknowledges the financial support of the Ministerio de Ciencia y
Tecnología-FEDER (projects SEJ2004-00110 and SEJ2005-02776). Juan Fernández de
Guevara acknowledges the financial support of the Banco Herrero Foundation.
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd and The University of Manchester
Published by Blackwell Publishing Ltd, 9600 Garsington Road, Oxford OX4 2DQ, UK, and 350 Main Street, Malden, MA 02148, USA.
275
276 The Manchester School
rD ∂C L0 rk B
rk (1 − β k ) − − k −
1 − α ∂ Lk 1
= (1 − β k ) N N (2)
rk rk b + B N2
1
For the Spanish banking sector, the ratio non-interest income/total income has increased from
4.94 per cent in 1986 to 12.11 per cent in 2002.
2
As Angelini and Cetorelli (2003) affirm, ‘the choice is valid under the assumption that the stock
of total assets is a good proxy for heterogeneous flow of services supplied by banks’.
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd and The University of Manchester 2007
280 The Manchester School
2 j =1
1 3 3 1 3
+ ∑∑
2 j =1 k =1
β jk ln w ji ln wki + ∑ γ j ln TAi ln wji + μ1Trend
2 j =1
3
1
+ μ2 Trend 2 + μ3 Trend ln TA i + ∑ λ j Trend ln wji + ln ui (4)
2 j =1
where TCi is the bank’s total costs including financial and operating costs. As
mentioned before, total assets (TAi) are used as a proxy variable for banking
output. The input prices (w) are defined as follows: price of labour
(w1) = personnel costs/number of employees; price of capital (w2) = operating
costs (except personnel costs)/fixed assets; price of deposits (w3) is financial
costs/deposits. In the estimation of the costs function, fixed effects are intro-
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd and The University of Manchester 2007
Explanatory Factors of Bank Market Power 281
market power will be.3 For each bank, the value of the variable is
constructed as a weighted average of the loans/GDP ratio of the prov-
inces where the bank has branches, using as weights the relative impor-
tance of each province in terms of branches.
(d) Given the fact that the coefficient of bank reserves (a) has not been
taken into account in the estimation of the Lerner index, it is explicitly
included as a determinant. It is proxied through the ratio of cash and
deposits in central banks/total deposits. As the theoretical model shows
(expression (2)), a negative sign is to be expected for this variable, as the
higher the proportion of liquid reserves (with an implicit opportunity
cost as they are remunerated at an interest below the market rate), the
lower the margin obtained.
(e) Default risk (bk) is proxied using the ratio loan loss provisions/loans as
an ex post indicator of the cost of risk.4 As suggested by the theoretical
model used, a negative sign for this variable is to be expected, given that
a higher default risk implies a lower relative margin.
Although it does not appear explicitly in the theoretical model
(expression (2)) as determinants of the Lerner index, other papers show
the importance of introducing additional explanatory variables of
market power.
(f) The evidence shows the importance for market power of specialization
in a particular type of banking activity (Fernández de Guevara et al.,
2007). The literature on integration of the financial markets shows a
lower degree of integration and competition in retail banking markets
than in wholesale markets, as a consequence, in addition to the charac-
teristics of the products themselves, of the greater importance of the
barriers or obstacles which protect markets from outside competition
(see, among others, Cabral et al., 2002; European Commission, 2002a,
2002b; Hartmann et al., 2003; Fernández de Guevara et al., 2007). For
this reason, the importance of specialization in retail products is proxied
through the income structure. Specifically, the proxy variable used is the
ratio non-interest income/total income. It is to be expected that the
lower retail activity, and thus greater relative importance of non-interest
income, will be accompanied by a lower market power.
(g) Corvosier and Gropp (2002) and Fernández de Guevara et al. (2005)
introduce the efficiency of banking firms as an explanatory variable of
market power, using the cost/income ratio. It was introduced to con-
3
For the European banking sectors, Corvosier and Gropp’s (2002) results show that the larger
the total assets of the banking system relative to GDP, the higher banks’ margins would be
expected to be.
4
A better measurement to proxy the default risk is non-performing loans/total loans ratio.
Unfortunately, this information is not available at individual bank level in the statistical
yearbooks of the AEB (Asociacion Española de Banca) and the CECA (Confederacion
Española de Cajas de Ahorros).
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd and The University of Manchester 2007
Explanatory Factors of Bank Market Power 283
5
As mentioned above, the lack of disaggregated information on interest rates obliges us to
estimate the Lerner index for the whole banking activity using total assets as a proxy
variable. Consequently, we must estimate the marginal cost of the total assets based on a
costs function with only one output. However, the estimation of X-efficiency based on a
cost function model with a vector of banking outputs is more accurate.
6
Following the studies by Berger and DeYoung (1997), DeYoung and Hasan (1998), Rogers
(1998), among others, ‘other operating income’ is introduced as a proxy variable for
off-balance sheet activities which have grown in importance in recent years.
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd and The University of Manchester 2007
284 The Manchester School
1994 0.091 0.012 0.084 0.096 0.076 0.011 0.069 0.082 0.169 0.078 0.146 0.217
1995 0.093 0.013 0.089 0.099 0.077 0.011 0.070 0.082 0.174 0.072 0.134 0.211
1996 0.090 0.008 0.086 0.093 0.074 0.008 0.069 0.076 0.181 0.079 0.119 0.228
1997 0.074 0.009 0.069 0.079 0.060 0.006 0.055 0.064 0.186 0.085 0.128 0.228
1998 0.066 0.006 0.064 0.069 0.053 0.006 0.049 0.056 0.209 0.078 0.175 0.258
1999 0.055 0.006 0.051 0.057 0.043 0.006 0.038 0.048 0.228 0.083 0.173 0.287
Journal compilation © 2007 Blackwell Publishing Ltd and The University of Manchester 2007
2000 0.060 0.007 0.056 0.062 0.046 0.006 0.043 0.049 0.225 0.070 0.187 0.242
2001 0.061 0.008 0.057 0.063 0.047 0.007 0.042 0.048 0.236 0.081 0.204 0.272
2002 0.054 0.006 0.051 0.057 0.040 0.005 0.039 0.042 0.249 0.087 0.209 0.313
Source: AEB, CECA and own elaboration.
285
286 The Manchester School
from 10.7 per cent in 1986 to 5.4 per cent in 2002 as a consequence of the
reduction of the money market rate. In parallel, marginal costs also decreased
from 8.5 per cent in 1986 to 4 per cent in 2002 as a consequence of the
reduction of both financial and operating costs. As a result of the joint
evolution of prices and marginal costs, the value of the Lerner index rose
from 0.20 in 1986 to 0.25 in 2002, market power having increased by 25 per
cent in the period considered.
The evolution of the Lerner index (see Fig. 1) shows three differentiated
subperiods: a brief subperiod of growth until 1988, a fall until 1994, and a
practically continuous increase to 2002, reaching in this last year the highest
value of the Lerner index in the period analysed. For the periods common to
those analysed in other studies, this pattern of behaviour is similar, with
growth of market power. Thus, Maudos and Pérez (2003) analyse the degree
of competition in the Spanish banking system from 1992 to 2001 by the
estimation of Lerner indexes and the Panzar and Rosse test. The results do
not permit the rejection of the hypothesis of monopolistic competition,
showing an increase in market power since 1996. Likewise, Carbó et al.
(2003a) also obtained evidence in favour of the existence of monopolistic
competition in the period 1986–99, as well as an increase in market power
since 1996.
The distinction between savings banks and commercial banks (Fig. 1)
shows that the value of the Lerner index of the savings banks is higher than
30%
25%
20%
15%
10%
5%
0%
86
87
88
89
90
91
92
93
94
95
96
97
98
99
00
01
02
19
19
19
19
19
19
19
19
19
19
19
19
19
19
20
20
20
1600
1500
1400
1300
1200
1100
1000
900
800
86
87
88
89
90
91
92
93
94
95
96
97
98
99
00
01
02
19
19
19
19
19
19
19
19
19
19
19
19
19
19
20
20
20
Saving banks Banks Banks and saving banks
10
A more detailed analysis of mergers and acquisitions in the Spanish banking system is found in
Carbó and Humphrey (2004).
11
In the estimation we introduced a dummy for the institutional group (distinction between
commercial bank and savings bank), which did not turn out to be significant, so finally it
was omitted.
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd and The University of Manchester 2007
Explanatory Factors of Bank Market Power 289
Table 2
Variable Definitions and Sample Means
estimation with the aim of capturing the influence of other possible variables
characteristic of each bank, and temporal effects.12
Column (1) of Table 3 offers the results of the estimation of equation (2).
It shows that market concentration, proxied by the HHI, is not shown to be
relevant in the explanation of the differences in market power, this result
12
As the Hausman test allows the null hypothesis of absence of correlation between individual
effects and the explanatory variables to be rejected in all cases, we use the within-group
estimator (fixed-effects model).
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd and The University of Manchester 2007
290
Table 3
Determinants of the Market Power. 1986–2002 (Dependent Variable: Lerner Index)
Journal compilation © 2007 Blackwell Publishing Ltd and The University of Manchester 2007
Non-interest income/total income -0.07 -0.08 -0.07
Cash/deposits 0.00 0.00 0.00
Cost efficiency 7.34 7.51 9.78
Source: AEB, CECA, Bank of Spain, INE and own elaboration.
Note: The data in the lower part of the table indicate the percentage variation of the Lerner index in response to a 1 per cent increase in its determinants, evaluated at average
sample values. All estimations include fixed effects (Hausman test does not reject this specification). In column (3) cost efficiency is instrumented with its lagged value.
The Manchester School
Explanatory Factors of Bank Market Power 291
13
As some studies have shown the existence of a nonlinear relationship between concentration
and market power (e.g. Jackson, 1997), we have checked the results introducing addition-
ally the square of the HHI. However, its effect is not statistically significant.
14
It can be argued that more efficient firms enjoy greater market power because of lower marginal
costs. But if the market were perfectly competitive, gains in efficiency would translate into
lower prices as the equality between marginal costs and prices must hold. The question we
posit is that if more efficient banks do not lower prices to match their marginal cost it is
because of market power.
15
DeYoung and Hasan (1998), using as reference the USA, show that newly created banks
rapidly improve their efficiency, though on average they take nine years to reach the levels
of the established banks, as a consequence, among other things, of excess capacity. In
consequence, the greater efficiency of the established banks may act as a barrier to entry
into the sector.
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd and The University of Manchester 2007
292 The Manchester School
16
This result is in line with other studies: Berger et al. (2004), Claessens and Laeven (2004, 2005),
Fernández de Guevara et al. (2005) and Carbó et al. (2006a, 2006b).
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd and The University of Manchester 2007
Explanatory Factors of Bank Market Power 293
4 Conclusions
The European banking sectors have in recent years undergone very substan-
tial changes as a result of the process of deregulation of the finance industry,
the creation of economic and monetary union, the development of informa-
tion technology and telecommunications etc. Alongside this, the European
banking industry has experienced a process of consolidation following the
wave of mergers and acquisitions leading to a reduction in the number of
competitors and an increase in market concentration.
Although the different deregulatory measures put in place create a
climate of greater competition among financial intermediaries, the increase in
concentration poses the question of the net effect of these two forces on the
degree of competition, and therefore on social welfare.
On the basis of a theoretical model of banking competition, and using as
laboratory the Spanish banking system in the period 1986–2002 (where
market concentration has increased as a result of mergers and acquisitions
and market structure variables can be proxied at regional level), the study
analyses the explanatory variables of market power through the construction
of Lerner indexes of market power.
The results show that the market power of the Spanish banking system
decreased considerably until the mid-1990s, though there has been a stage of
steady growth since then, confirming the evidence found by other studies.
Also, the data show that the savings banks enjoy greater market power than
the commercial banks.
The analysis of the explanatory factors of market power shows the
importance of specialization and of efficiency for explaining the differences in
market power among banks. Thus, the banks that specialize to a greater
extent in retail banking (with a lower proportion of non-interest income in
their total income) and that achieve greater efficiency achieve higher relative
margins, their greater efficiency acting as a barrier to entry.
17
The connection between MC (estimated from a total costs function) and X-efficiency is weaker
if the frontier costs function excludes interest expenses from the dependent variable and the
price of deposits from the input price vector, i.e. if we estimate operating cost X-efficiency.
For this reason, we analysed the sensitivity of the results using X-operating efficiency as an
explanatory variable of the Lerner index. Results (not shown) are robust.
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd and The University of Manchester 2007
294 The Manchester School
References
Angelini, P. and Cetorelli, N. (2003). ‘The Effects of Regulatory Reform on Compe-
tition in the Banking Industry’, Journal of Money, Credit and Banking, Vol. 35,
No. 5, pp. 663–684.
Berg, S. A. and Kim, M. (1994). ‘Oligopolistic Interdependence and the Structure of
Production in Banking: an Empirical Evaluation’, Journal of Money, Credit and
Banking, Vol. 26, pp. 309–322.
Berger, A. N. (1995). ‘The Profit-relationship in Banking—Test of Market-power and
Efficient-structure Hypothesis’, Journal of Money, Credit and Banking, Vol. 27,
pp. 405–431.
Berger, A. and DeYoung, R. (1997). ‘Problem Loans and Cost Efficiency in Com-
mercial Bank’, Journal of Banking and Finance, Vol. 21, pp. 849–870.
Berger, A., Demirgüç-Kunt, A., Levine, R. and Haubrich, J. (2004). ‘Bank Concen-
tration and Competition: an Evolution in the Making’, Journal of Money, Credit
and Banking, Vol. 36, pp. 435–451.
Bikker, J. A. and Haaf, K. (2002). ‘Competition, Concentration and Their Relation-
ship: an Empirical Analysis of the Banking Industry’, Journal of Banking and
Finance, Vol. 26, pp. 2191–2214.
Cabral, I., Dierick, F. and Vesala, J. (2002). ‘Banking Integration in the Euro Area’,
Occasional Paper Series 6, December, European Central Bank.
Carbó, S. and Humphrey, D. (2004). ‘Predicted and Actual Costs from Individual
Bank Mergers’, Journal of Economics and Business, Vol. 56, pp. 137–157.
Carbó, S., Gardener, T. and Williams, J. (2002). ‘Efficiency in Banking: Empirical
Evidence from the Savings Bank Sector’, The Manchester School, Vol. 70, No. 2,
pp. 204–228.
Carbó, S., Humphrey, D. and Rodríguez, F. (2003a). ‘Deregulation, Bank Competi-
tion and Regional Growth’, Regional Studies, Vol. 37, pp. 227–237.
Carbó, S., López, R. and Rodríguez, F. (2003b). ‘Medición de la competencia en los
mercados bancarios de las regiones españolas’, Revista de Economía Aplicada,
Vol. 32, pp. 5–33.
Carbó, S., Humphrey, D., Maudos, J. and Molyneux, P. (2006a). ‘Cross-country
Comparisons of Competition and Pricing Power in European Banking’, Proceed-
ings of the 42nd Annual Conference on Bank Structure and Competition, Chicago,
IL, Federal Reserve Bank of Chicago, 17–19 May.
Carbó, S., Rodriguez, F. and Udell, G. (2006b). ‘Bank Market Power and SME
Financing Constraints’, Proceedings of the 42nd Annual Conference on Bank
Structure and Competition, Chicago, IL, Federal Reserve Bank of Chicago, 17–19
May.
© 2007 The Authors
Journal compilation © 2007 Blackwell Publishing Ltd and The University of Manchester 2007
Explanatory Factors of Bank Market Power 295