- Email: [email protected]
The evolution of market power in European banking
Accepted Manuscript
The evolution of market power in European banking Paula Cruz-Garc´ıa , Juan Fernandez de Guevara , ´ Joaqu´ın Maudos PII: DOI: Reference:
S1544-6123(17)30142-3 10.1016/j.frl.2017.06.012 FRL 730
To appear in:
Finance Research Letters
Received date: Revised date: Accepted date:
10 March 2017 2 June 2017 7 June 2017
Please cite this article as: Paula Cruz-Garc´ıa , Juan Fernandez de Guevara , Joaqu´ın Maudos , ´ The evolution of market power in European banking, Finance Research Letters (2017), doi: 10.1016/j.frl.2017.06.012
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Highlights The evolution of market power in European banking in 2000-2014 is analysed.
Market power disparities are calculated using the Theil index and the Lerner index.
The market power disparity among eurozone banks has narrowed.
The reduction is attributable to the convergence in the average levels of market power of the European banking sectors.
Complementary measures at national level are needed to intensify competition.
AC
CE
PT
ED
M
AN US
CR IP T
ACCEPTED MANUSCRIPT The evolution of market power in European banking Paula Cruz-Garcíab Juan Fernández de Guevara a,b Joaquín Maudosa,b a
Ivie. C/ Guardia Civil 22, Esc. 2, 1º, 46020 Valencia, Spain. Tel: +34 96 319 00 50
[email protected] [corresponding author] b University of Valencia. Departamento de Análisis Economico, Edificio departamental Oriental, Avda.
CR IP T
de los Naranjos, s/n. 46022 Valencia, Spain.
Abstract
ED
M
AN US
This paper analyses the effect that European financial market integration has had on the evolution of the banks' market power disparities. The results show that market power disparity has narrowed among eurozone banks. The reduction is attributable to the convergence in the average levels of market power of the European banking sectors. In contrast, the disparities observed within each country have remained stable. As a result, the measures adopted to advance towards a single banking market should be complemented by measures at the national level designed to intensify competition among the banks within a given country.
AC
CE
PT
Keywords: banks, market power, financial integration, European financial markets JEL classification: F36, G15
1. Introduction
After the European Monetary Union was created in 1999, the level of financial market integration progressed at a good pace until the financial crisis broke out in 2007. As shown by the ECB, although the price-based financial integration composite indicator doubled between 1999 and 2007, in the ensuing years of crisis (against the backdrop of 2
ACCEPTED MANUSCRIPT the Lehman Brothers bankruptcy and the sovereign debt crisis), up until 2012, it fell back to levels close to those registered in 1995. The unconventional measures pursued by the ECB, coupled with the banking union plans announced in 2012, paved the way for renewed progress towards integration in subsequent years, although the integration indicator remains below pre-crisis levels.
CR IP T
Specifically, in the case of the retail banking markets, the measures taken to date have barely increased the low level of integration observed. The corresponding indicators signal a low level of cross-border lending activities, while the share of cross-border retail deposits remains low or negligible. In fact, the ECB report (2016) on financial
AN US
integration states that “further efforts should be made to improve the integration of retail banking services in quantitative terms through a common set of rules”. In the same line, the most recent ECB report (2017) affirms that “quantity-based indicators of banking
M
sector integration continued to signal fragmentation in retail banking”. The measures undertaken with the aim of making progress towards a single banking
ED
market should have reduced both financial intermediaries’ market power and the differences between countries. Therefore, the pertinent question is whether or not, given
PT
the low level of integration in the retail banking sector (the most significant segment in
CE
terms of business volumes), the market power differences between countries and/or entities have narrowed.
AC
Against this backdrop, the aim of this paper is to analyse the evolution of market power disparity among banks across the EMU founding countries between 2000 and 2014 using the BankScope database. To this end, we calculate the Lerner index of market power and analyse the disparity across the banks using the Theil index. The advantage of the latter index lies with its additive decomposability such that we can break down overall disparity in the market power of the European banks into a 'within country'
3
ACCEPTED MANUSCRIPT effect (differences between the banks of a given country) and a 'between countries' effect (differences among countries), integration being higher the lower the between effect. Accordingly, if a market is moving towards integration, the between component can be expected to decline, such that there are fewer differences in market power levels between the various countries, as lower barriers to competition at the European level
CR IP T
force more protected countries to foster competition. That is, the differences in market power across banks are not driven by a country effect, but by differences in the firms operating in each.
The paper is structured as follows: Section 2 describes the Lerner index of market
AN US
power and shows the results obtained. Section 3 analyses the market power disparity based on the Theil index. Finally, section 4 presents our conclusions and the resulting
M
policy implications.
2. Market power in the European banking sector The Lerner index of market power
ED
2.1.
To analyse the trend in market power differences among EU banks, we need an
PT
indicator at the bank level such as the Lerner index. This index has been widely used in
CE
many academic papers (Angelini and Cetorelli, 2003; Maudos and Fernandez de Guevara, 2004; Carbó et al., 2009; and Koetter et al., 2012, etc.) and is defined as the
AC
difference between price and marginal cost expressed as a percentage of price. In its original version, based on the Monti-Klein imperfect competition model, the Lerner index is derived solving a profit-maximisation problem:
(
)
( ( )
)
(
( ))
(
)
(1)
4
ACCEPTED MANUSCRIPT where profit is net interest income minus operating costs, L and D represent loans and deposits, respectively,
and
their prices (interest rates), r the money market interest
rate and C, operating costs. The first-order conditions of this problem yield the Lerner ) index of market power in lending (
(
) and deposit-making
) and measure a bank's power to set interest rates above their marginal cost: [
]
CR IP T
(
(2)
]
AN US
[
In the case of perfect competition, the value of the index is zero; in the case of a monopoly, its value is one. Thus, the higher the Lerner index reading, the higher the
M
market power. 2.2 Empirical approach and results
ED
The empirical approach to the Lerner index has limitations, most notably in terms of
PT
estimating the interest rate of loans and deposits. In the case of loans, the profit and loss account information available in the database used (BankScope, Bureau Van Dijk)
CE
provides information about interest income from loans aggregated with interest income from other financial products (such as fixed-income investments). In the case of
AC
deposits, their cost is aggregated with the financial costs of other liability products. Consequently, we use a single indicator of banking activity (total assets) following Maudos and Fernández de Guevara (2004), Carbó et al. (2009), Liu and Wilson (2013), ECB (2017), among others. Using this approach, we compute an average price (which includes interest and non-interest income) and marginal cost (interest and operating costs) in respect of total assets.
5
ACCEPTED MANUSCRIPT To calculate marginal cost, we estimate a translogarithmic cost function for the European banking sectors: (
)
∑
∑
∑
∑ ∑ where output is total assets (
),
CR IP T
(3) is the total costs (financial and operating costs) and
the three input prices are the costs of labour, capital and deposits, which are measured as follows:
AN US
w1: Price of labor = Staff costs / total assets
w2: Price of capital = operating costs (except staff costs) / fixed assets w3: Price of deposits = financial costs / deposits
This cost function is estimated by using a data panel consisting of all banks in the
M
analysis. Fixed effects are included to account for the influence of variables specific to
ED
each bank. We also include a trend to reflect the effect of technical change, which shifts the cost function over time. As usual, we impose symmetry restrictions and
PT
homogeneity of degree one in input prices.
CE
The sample encompasses the EMU founding countries between 2000 and 2014. We include all the banks included in the BankScope database for the aforementioned
AC
countries. However, banks for which there was no information on any of the variables needed to estimate the Lerner index were excluded, as were the banks for which the prices of the factors of production needed to estimate the Lerner index were outside the range of 2.5 standard deviations on either side of the mean, calculated for each year. After filtering, the panel of data used comprised 47,375 observations. Table 1 shows the evolution of Lerner index of market power weighted by total assets for the period 2000-14 by countries. The countries with a higher Lerner index are
6
ACCEPTED MANUSCRIPT Finland, Greece, Luxembourg and Spain, whereas countries with lower market power are Belgium, Denmark and Ireland. Market power has increased in 10 of the 12 countries between 2000 and 2014. If we differentiate between the period before and after the financial crisis, the countries that increased their market power before the crisis were Ireland (by far), Greece and Denmark; and those who decreased it were Italy and
CR IP T
France. However, the countries that have increased their market power after the crisis have been Luxembourg, Netherlands and Belgium and those who have decreased it have been Portugal and Ireland1.
As shown by the evolution of the coefficient of variation (see last row in table 1),
AN US
differences in market power between countries have declined by 24% in the period analysed, with a decreasing trend until 2013, where the differences between countries reached a minimum value, and a rebound later until 2015. However, differences exist
M
between the average levels of market power, without taking into account the dispersion within each country. To do this, it is necessary to use information at the bank level, so
ED
that it is possible to decompose the total inequality between the euro area banks into a within country component and another one between countries, the purpose of which is
PT
to compute the Theil index.
2000 0.11 0.05 0.09 0.19 0.34 0.08 0.28 0.04 0.23 0.16
2001 0.11 0.04 0.07 0.21 0.28 0.07 0.31 0.05 0.24 0.17
AC
Austria Belgium Germany Spain Finland France Greece Ireland Italy Luxembourg
CE
Table 1. Lerner index weighted by total assets 2002 0.12 0.04 0.09 0.21 0.31 0.06 0.26 -0.03 0.22 0.18
2003 0.14 0.07 0.08 0.23 0.30 0.10 0.32 0.04 0.23 0.19
2004 0.14 0.06 0.10 0.25 0.39 0.13 0.32 0.06 0.23 0.19
2005 0.17 0.07 0.13 0.20 0.39 0.12 0.39 -0.03 0.23 0.18
2006 0.16 0.10 0.16 0.21 0.40 0.11 0.37 0.06 0.18 0.19
2007 0.11 0.07 0.15 0.20 0.37 0.05 0.33 0.27 0.15 0.14
2008 0.16 0.05 0.06 0.16 0.34 0.07 0.26 0.27 0.10 0.18
2009 0.20 0.06 0.19 0.23 0.31 0.15 0.31 0.07 0.19 0.27
2010 0.20 0.21 0.08 0.26 0.44 0.18 0.29 0.11 0.16 0.32
2011 0.21 0.22 0.12 0.20 0.40 0.14 0.29 0.11 0.13 0.25
2012 0.19 0.19 0.13 0.24 0.32 0.16 0.16 0.17 0.17 0.29
2013 0.17 0.19 0.13 0.17 0.41 0.17 0.25 0.20 0.16 0.33
1
In general, the results are in line with those recently obtained by the ECB (2017) for the euro area countries in the period 2003-2015. If we focus on the most recent period after the start of the financial crisis in 2008, in all countries market power increases (in our case, it also increases in all but Ireland and Portugal).
7
2014 0.17 0.18 0.14 0.23 0.44 0.16 0.28 0.25 0.17 0.32
ACCEPTED MANUSCRIPT Netherlands Portugal Coeff. var.
0.14 0.20 0.59
0.16 0.15 0.59
0.15 0.15 0.66
0.18 0.18 0.53
0.11 0.27 0.57
0.10 0.16 0.69
0.17 0.20 0.52
0.13 0.17 0.54
0.15 0.14 0.55
0.17 0.20 0.41
0.21 0.18 0.44
0.25 0.14 0.42
0.26 0.12 0.32
0.30 0.05 0.46
Source: BankScope (Bureau Van Djik) and own elaboration
3. Bank market power disparity in Europe The Theil index
CR IP T
3.1.
The evolution of the level of integration is closely related with the degree of competition insofar as the goal of financial integration is to deliver a single financial
AN US
services market in which there are no barriers to the provision of these services from one country to another. Therefore, financial integration should ultimately translate into lower barriers to competition. In the case of the banking markets, the barriers are the result of differences between countries in regulation, supervision, product taxation,
M
portfolio restrictions, market structure, etc.
With the aim of assessing the significance of these specific country-by-country barriers
ED
in explaining the differences observed in market power among the European banks, we
PT
follow Fernández de Guevara et al. (2007) and we use the Theil index. The Theil index (T) has the advantage that the disparity can be decomposed into two effects: a within-
CE
group (Tw) and a between-groups (Tb) disparity factor. To obtain this decomposition, let's assume that we have a sample of i=1 to I banks, that the total sample can be
AC
separated into nG groups (in our case, countries), that each group represents a percentage
of the total sample, that the weighted average of variable
of the total
sample is and that the weighted average of variable x of each grouping is
. With
these assumptions, the Theil index can be separated into two terms:
8
0.28 0.05 0.45
ACCEPTED MANUSCRIPT (4) ∑
where ∑(
)
(
(5)
)
∑
(
)
is the external disparity (between countries) term.
3.2.
and
(6)
) are based on the market share of each bank in total assets.
AN US
The weightings (
CR IP T
is the internal disparity (within country) component and
Results
M
To analyse which component of the Theil index better explains the different levels of market power observed across firms, we compute a Theil index for the Lerner index.
ED
Figure 1 shows how the EMU banks' market power disparity fell by 35% between 2000 and 2014, although the drop was concentrated between 2000 and 2005, since when it
PT
has fluctuated around lower values. The reduction in disparity is mostly attributable to
CE
the between countries effect (which fell by 70% between 2000 and 2014), as the within effect was only slightly lower in 2014 than it was in 2000. Therefore, within a given
AC
country, significant differences in the various banks' market power persist. That is, in the years following the introduction of the euro differences in market power among European banks declined significantly. Moreover, the reduction was mostly driven by the decline in differences across countries.
Figure 1. Evolution of the Theil index
9
ACCEPTED MANUSCRIPT
0.35 0.30 0.25 0.20
CR IP T
0.15 0.10 0.05 0.00 Theil
AN US
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Within component
Between component
Source: BankScope (Bureau Van Djik) and own elaboration
M
Figure 2 shows that a very substantial portion of the market power disparity observed in
ED
the eurozone (around 35% - 40%) was explained by the between country differences during the initial years following introduction of the euro, although the majority (about
PT
60% - 65%) was attributable to within country disparity. The between countries effect subsequently (until 2006) lost significance, accounting for just 11% of disparity that
CE
year. In the following years, the portion of the disparity explained by the differences
AC
across countries fluctuated around 15%. Lastly, this component declined between 2012 and 2014, accounting for 13% of European bank market power inequality in 2014.
Figure 2. Decomposition of the Theil index
10
ACCEPTED MANUSCRIPT
100% 90% 80% 70% 60% 50%
CR IP T
40% 30% 20% 10%
AN US
0% 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Within component (%)
Between component (%)
M
Source: BankScope (Bureau Van Djik) and own elaboration
These findings suggest that the measures taken to advance towards a more integrated
ED
banking market have reduced market power disparity between banks, as the percentage of market power disparity attributable to the between countries effect has halved.
PT
Moreover, despite the adverse impact of the crisis on integration, bank market power
CE
differences have not increased in the eurozone. Indeed, the European banks' market power disparity has continued to decline in the period of ECB quantitative easing and in
AC
the wake of the banking union plans/initiative announced in 2012. However, the factors specific to each country (rivalry among banks, market structure, barriers to entry, elasticity of demand, contestability, etc.) are responsible for a large percentage of the market power disparity observed across the banks analysed, particularly during the years since the recent financial crisis. This result is consistent with the evidence obtained by Lucotte (2015), which suggests the existence of large dissimilarities between the eurozone countries in terms of banking structure in the wake of the recent 11
ACCEPTED MANUSCRIPT financial crisis. Lucotte also flags the importance of working on a banking union “that will mitigate cross-country differences in terms of banking structure”.
4. Conclusions Despite the measures rolled out since the EMU was created with the aim of progressing
CR IP T
towards a single banking market, the level of cross-border integration in the retail banking sector remains very low, as is evidenced by the reduced level of cross-border lending and deposit-taking activity. The limited integration of retail markets is related to a lack of cross-border bank mergers and acquisitions within the euro area. However,
AN US
despite the slow pace of progress on integration, the reduction of barriers to competition (particularly by harmonising regulations) and the increased contestability of the market have driven a reduction in market power disparity across the eurozone's banks.
M
By estimating the Lerner market power index and decomposing the Theil index into a within-country effect and a between-countries effect, this paper finds that market power
ED
disparity among eurozone banks has narrowed and that the reduction is attributable to the convergence towards average market power levels across the European banking
PT
sectors. In contrast, the differences within countries have remained stable and were
CE
responsible in 2014 (the last year analysed) for nearly 90% of disparity in the EMU banks' market power. As a result, the measures adopted at the European level to
AC
advance towards a single banking market (such as Banking Union) should be complemented by measures at the national level that are designed to foster competition among the banks within a given country. It would also help to generate more competition cross-border M&A within the euro area, since unlike the M&A between banks of the same country, they do not increase the concentration of the domestic markets.
12
ACCEPTED MANUSCRIPT
Acknowledgements
CR IP T
Authors thank an anonymous referee for helpful comments and suggestions. They gratefully acknowledge financial support of the Spanish Ministry of Science and Innovation (research project ECO2013-43959-R). Joaquin Maudos also acknowledges financial support of Generalitat Valenciana (research project PROMETEOII/2014/046). Paula Cruz-García acknowledges financial support of Spanish Ministry of Education (FPU2014/00936).
References
AN US
Angelini, P. and Cetorelli, N. (2003). The effects of regulatory reform on competition in the banking industry, Journal of Money, Credit and Banking 35, 663684. Carbó, S., Humphrey, D., Maudos, J. and Molyneux, P. (2009). "Cross-country
M
comparisons of competition and pricing power in European Banking", Journal of International Money and Finance, 28, 115-134.
ED
European Central Bank (2016). Financial integration in Europe, April. European Central Bank (2017). Financial integration in Europe, May.
PT
Fernández de Guevara, J., Maudos, J. and Pérez, F. (2007). "Integration and
CE
competition in the European financial markets", Journal of International Money and Finance 26, 26-45.
AC
Koetter, M., J. Kolari, and L. Spierdijk. (2012). “Enjoying the Quiet Life under
Deregulation? Evidence from Adjusted Lerner Indices for US Banks.” Review of Economics and Statistics 94, 462–480. Liu, H. and Wilson, J. (2013). “Competition and risk in Japanese banking”, European Journal of Finance, 19(1), 1-18.
13
ACCEPTED MANUSCRIPT Lucotte, Y. (2015). “Euro area banking fragmentation in the aftermath of the crisis: a cluster analysis.” Applied Economics Letters 22 (13), 1046-1050. Maudos, J. and Fernández de Guevara, J. (2004). “Factors explaining the interest margin in the banking sectors of the European Union”, Journal of Banking and Finance
AC
CE
PT
ED
M
AN US
CR IP T
28, 2259-2281.
14
The evolution of market power in European banking Paula Cruz-Garc´ıa , Juan Fernandez de Guevara , ´ Joaqu´ın Maudos PII: DOI: Reference:
S1544-6123(17)30142-3 10.1016/j.frl.2017.06.012 FRL 730
To appear in:
Finance Research Letters
Received date: Revised date: Accepted date:
10 March 2017 2 June 2017 7 June 2017
Please cite this article as: Paula Cruz-Garc´ıa , Juan Fernandez de Guevara , Joaqu´ın Maudos , ´ The evolution of market power in European banking, Finance Research Letters (2017), doi: 10.1016/j.frl.2017.06.012
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
ACCEPTED MANUSCRIPT Highlights The evolution of market power in European banking in 2000-2014 is analysed.
Market power disparities are calculated using the Theil index and the Lerner index.
The market power disparity among eurozone banks has narrowed.
The reduction is attributable to the convergence in the average levels of market power of the European banking sectors.
Complementary measures at national level are needed to intensify competition.
AC
CE
PT
ED
M
AN US
CR IP T
ACCEPTED MANUSCRIPT The evolution of market power in European banking Paula Cruz-Garcíab Juan Fernández de Guevara a,b Joaquín Maudosa,b a
Ivie. C/ Guardia Civil 22, Esc. 2, 1º, 46020 Valencia, Spain. Tel: +34 96 319 00 50
[email protected] [corresponding author] b University of Valencia. Departamento de Análisis Economico, Edificio departamental Oriental, Avda.
CR IP T
de los Naranjos, s/n. 46022 Valencia, Spain.
Abstract
ED
M
AN US
This paper analyses the effect that European financial market integration has had on the evolution of the banks' market power disparities. The results show that market power disparity has narrowed among eurozone banks. The reduction is attributable to the convergence in the average levels of market power of the European banking sectors. In contrast, the disparities observed within each country have remained stable. As a result, the measures adopted to advance towards a single banking market should be complemented by measures at the national level designed to intensify competition among the banks within a given country.
AC
CE
PT
Keywords: banks, market power, financial integration, European financial markets JEL classification: F36, G15
1. Introduction
After the European Monetary Union was created in 1999, the level of financial market integration progressed at a good pace until the financial crisis broke out in 2007. As shown by the ECB, although the price-based financial integration composite indicator doubled between 1999 and 2007, in the ensuing years of crisis (against the backdrop of 2
ACCEPTED MANUSCRIPT the Lehman Brothers bankruptcy and the sovereign debt crisis), up until 2012, it fell back to levels close to those registered in 1995. The unconventional measures pursued by the ECB, coupled with the banking union plans announced in 2012, paved the way for renewed progress towards integration in subsequent years, although the integration indicator remains below pre-crisis levels.
CR IP T
Specifically, in the case of the retail banking markets, the measures taken to date have barely increased the low level of integration observed. The corresponding indicators signal a low level of cross-border lending activities, while the share of cross-border retail deposits remains low or negligible. In fact, the ECB report (2016) on financial
AN US
integration states that “further efforts should be made to improve the integration of retail banking services in quantitative terms through a common set of rules”. In the same line, the most recent ECB report (2017) affirms that “quantity-based indicators of banking
M
sector integration continued to signal fragmentation in retail banking”. The measures undertaken with the aim of making progress towards a single banking
ED
market should have reduced both financial intermediaries’ market power and the differences between countries. Therefore, the pertinent question is whether or not, given
PT
the low level of integration in the retail banking sector (the most significant segment in
CE
terms of business volumes), the market power differences between countries and/or entities have narrowed.
AC
Against this backdrop, the aim of this paper is to analyse the evolution of market power disparity among banks across the EMU founding countries between 2000 and 2014 using the BankScope database. To this end, we calculate the Lerner index of market power and analyse the disparity across the banks using the Theil index. The advantage of the latter index lies with its additive decomposability such that we can break down overall disparity in the market power of the European banks into a 'within country'
3
ACCEPTED MANUSCRIPT effect (differences between the banks of a given country) and a 'between countries' effect (differences among countries), integration being higher the lower the between effect. Accordingly, if a market is moving towards integration, the between component can be expected to decline, such that there are fewer differences in market power levels between the various countries, as lower barriers to competition at the European level
CR IP T
force more protected countries to foster competition. That is, the differences in market power across banks are not driven by a country effect, but by differences in the firms operating in each.
The paper is structured as follows: Section 2 describes the Lerner index of market
AN US
power and shows the results obtained. Section 3 analyses the market power disparity based on the Theil index. Finally, section 4 presents our conclusions and the resulting
M
policy implications.
2. Market power in the European banking sector The Lerner index of market power
ED
2.1.
To analyse the trend in market power differences among EU banks, we need an
PT
indicator at the bank level such as the Lerner index. This index has been widely used in
CE
many academic papers (Angelini and Cetorelli, 2003; Maudos and Fernandez de Guevara, 2004; Carbó et al., 2009; and Koetter et al., 2012, etc.) and is defined as the
AC
difference between price and marginal cost expressed as a percentage of price. In its original version, based on the Monti-Klein imperfect competition model, the Lerner index is derived solving a profit-maximisation problem:
(
)
( ( )
)
(
( ))
(
)
(1)
4
ACCEPTED MANUSCRIPT where profit is net interest income minus operating costs, L and D represent loans and deposits, respectively,
and
their prices (interest rates), r the money market interest
rate and C, operating costs. The first-order conditions of this problem yield the Lerner ) index of market power in lending (
(
) and deposit-making
) and measure a bank's power to set interest rates above their marginal cost: [
]
CR IP T
(
(2)
]
AN US
[
In the case of perfect competition, the value of the index is zero; in the case of a monopoly, its value is one. Thus, the higher the Lerner index reading, the higher the
M
market power. 2.2 Empirical approach and results
ED
The empirical approach to the Lerner index has limitations, most notably in terms of
PT
estimating the interest rate of loans and deposits. In the case of loans, the profit and loss account information available in the database used (BankScope, Bureau Van Dijk)
CE
provides information about interest income from loans aggregated with interest income from other financial products (such as fixed-income investments). In the case of
AC
deposits, their cost is aggregated with the financial costs of other liability products. Consequently, we use a single indicator of banking activity (total assets) following Maudos and Fernández de Guevara (2004), Carbó et al. (2009), Liu and Wilson (2013), ECB (2017), among others. Using this approach, we compute an average price (which includes interest and non-interest income) and marginal cost (interest and operating costs) in respect of total assets.
5
ACCEPTED MANUSCRIPT To calculate marginal cost, we estimate a translogarithmic cost function for the European banking sectors: (
)
∑
∑
∑
∑ ∑ where output is total assets (
),
CR IP T
(3) is the total costs (financial and operating costs) and
the three input prices are the costs of labour, capital and deposits, which are measured as follows:
AN US
w1: Price of labor = Staff costs / total assets
w2: Price of capital = operating costs (except staff costs) / fixed assets w3: Price of deposits = financial costs / deposits
This cost function is estimated by using a data panel consisting of all banks in the
M
analysis. Fixed effects are included to account for the influence of variables specific to
ED
each bank. We also include a trend to reflect the effect of technical change, which shifts the cost function over time. As usual, we impose symmetry restrictions and
PT
homogeneity of degree one in input prices.
CE
The sample encompasses the EMU founding countries between 2000 and 2014. We include all the banks included in the BankScope database for the aforementioned
AC
countries. However, banks for which there was no information on any of the variables needed to estimate the Lerner index were excluded, as were the banks for which the prices of the factors of production needed to estimate the Lerner index were outside the range of 2.5 standard deviations on either side of the mean, calculated for each year. After filtering, the panel of data used comprised 47,375 observations. Table 1 shows the evolution of Lerner index of market power weighted by total assets for the period 2000-14 by countries. The countries with a higher Lerner index are
6
ACCEPTED MANUSCRIPT Finland, Greece, Luxembourg and Spain, whereas countries with lower market power are Belgium, Denmark and Ireland. Market power has increased in 10 of the 12 countries between 2000 and 2014. If we differentiate between the period before and after the financial crisis, the countries that increased their market power before the crisis were Ireland (by far), Greece and Denmark; and those who decreased it were Italy and
CR IP T
France. However, the countries that have increased their market power after the crisis have been Luxembourg, Netherlands and Belgium and those who have decreased it have been Portugal and Ireland1.
As shown by the evolution of the coefficient of variation (see last row in table 1),
AN US
differences in market power between countries have declined by 24% in the period analysed, with a decreasing trend until 2013, where the differences between countries reached a minimum value, and a rebound later until 2015. However, differences exist
M
between the average levels of market power, without taking into account the dispersion within each country. To do this, it is necessary to use information at the bank level, so
ED
that it is possible to decompose the total inequality between the euro area banks into a within country component and another one between countries, the purpose of which is
PT
to compute the Theil index.
2000 0.11 0.05 0.09 0.19 0.34 0.08 0.28 0.04 0.23 0.16
2001 0.11 0.04 0.07 0.21 0.28 0.07 0.31 0.05 0.24 0.17
AC
Austria Belgium Germany Spain Finland France Greece Ireland Italy Luxembourg
CE
Table 1. Lerner index weighted by total assets 2002 0.12 0.04 0.09 0.21 0.31 0.06 0.26 -0.03 0.22 0.18
2003 0.14 0.07 0.08 0.23 0.30 0.10 0.32 0.04 0.23 0.19
2004 0.14 0.06 0.10 0.25 0.39 0.13 0.32 0.06 0.23 0.19
2005 0.17 0.07 0.13 0.20 0.39 0.12 0.39 -0.03 0.23 0.18
2006 0.16 0.10 0.16 0.21 0.40 0.11 0.37 0.06 0.18 0.19
2007 0.11 0.07 0.15 0.20 0.37 0.05 0.33 0.27 0.15 0.14
2008 0.16 0.05 0.06 0.16 0.34 0.07 0.26 0.27 0.10 0.18
2009 0.20 0.06 0.19 0.23 0.31 0.15 0.31 0.07 0.19 0.27
2010 0.20 0.21 0.08 0.26 0.44 0.18 0.29 0.11 0.16 0.32
2011 0.21 0.22 0.12 0.20 0.40 0.14 0.29 0.11 0.13 0.25
2012 0.19 0.19 0.13 0.24 0.32 0.16 0.16 0.17 0.17 0.29
2013 0.17 0.19 0.13 0.17 0.41 0.17 0.25 0.20 0.16 0.33
1
In general, the results are in line with those recently obtained by the ECB (2017) for the euro area countries in the period 2003-2015. If we focus on the most recent period after the start of the financial crisis in 2008, in all countries market power increases (in our case, it also increases in all but Ireland and Portugal).
7
2014 0.17 0.18 0.14 0.23 0.44 0.16 0.28 0.25 0.17 0.32
ACCEPTED MANUSCRIPT Netherlands Portugal Coeff. var.
0.14 0.20 0.59
0.16 0.15 0.59
0.15 0.15 0.66
0.18 0.18 0.53
0.11 0.27 0.57
0.10 0.16 0.69
0.17 0.20 0.52
0.13 0.17 0.54
0.15 0.14 0.55
0.17 0.20 0.41
0.21 0.18 0.44
0.25 0.14 0.42
0.26 0.12 0.32
0.30 0.05 0.46
Source: BankScope (Bureau Van Djik) and own elaboration
3. Bank market power disparity in Europe The Theil index
CR IP T
3.1.
The evolution of the level of integration is closely related with the degree of competition insofar as the goal of financial integration is to deliver a single financial
AN US
services market in which there are no barriers to the provision of these services from one country to another. Therefore, financial integration should ultimately translate into lower barriers to competition. In the case of the banking markets, the barriers are the result of differences between countries in regulation, supervision, product taxation,
M
portfolio restrictions, market structure, etc.
With the aim of assessing the significance of these specific country-by-country barriers
ED
in explaining the differences observed in market power among the European banks, we
PT
follow Fernández de Guevara et al. (2007) and we use the Theil index. The Theil index (T) has the advantage that the disparity can be decomposed into two effects: a within-
CE
group (Tw) and a between-groups (Tb) disparity factor. To obtain this decomposition, let's assume that we have a sample of i=1 to I banks, that the total sample can be
AC
separated into nG groups (in our case, countries), that each group represents a percentage
of the total sample, that the weighted average of variable
of the total
sample is and that the weighted average of variable x of each grouping is
. With
these assumptions, the Theil index can be separated into two terms:
8
0.28 0.05 0.45
ACCEPTED MANUSCRIPT (4) ∑
where ∑(
)
(
(5)
)
∑
(
)
is the external disparity (between countries) term.
3.2.
and
(6)
) are based on the market share of each bank in total assets.
AN US
The weightings (
CR IP T
is the internal disparity (within country) component and
Results
M
To analyse which component of the Theil index better explains the different levels of market power observed across firms, we compute a Theil index for the Lerner index.
ED
Figure 1 shows how the EMU banks' market power disparity fell by 35% between 2000 and 2014, although the drop was concentrated between 2000 and 2005, since when it
PT
has fluctuated around lower values. The reduction in disparity is mostly attributable to
CE
the between countries effect (which fell by 70% between 2000 and 2014), as the within effect was only slightly lower in 2014 than it was in 2000. Therefore, within a given
AC
country, significant differences in the various banks' market power persist. That is, in the years following the introduction of the euro differences in market power among European banks declined significantly. Moreover, the reduction was mostly driven by the decline in differences across countries.
Figure 1. Evolution of the Theil index
9
ACCEPTED MANUSCRIPT
0.35 0.30 0.25 0.20
CR IP T
0.15 0.10 0.05 0.00 Theil
AN US
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Within component
Between component
Source: BankScope (Bureau Van Djik) and own elaboration
M
Figure 2 shows that a very substantial portion of the market power disparity observed in
ED
the eurozone (around 35% - 40%) was explained by the between country differences during the initial years following introduction of the euro, although the majority (about
PT
60% - 65%) was attributable to within country disparity. The between countries effect subsequently (until 2006) lost significance, accounting for just 11% of disparity that
CE
year. In the following years, the portion of the disparity explained by the differences
AC
across countries fluctuated around 15%. Lastly, this component declined between 2012 and 2014, accounting for 13% of European bank market power inequality in 2014.
Figure 2. Decomposition of the Theil index
10
ACCEPTED MANUSCRIPT
100% 90% 80% 70% 60% 50%
CR IP T
40% 30% 20% 10%
AN US
0% 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 Within component (%)
Between component (%)
M
Source: BankScope (Bureau Van Djik) and own elaboration
These findings suggest that the measures taken to advance towards a more integrated
ED
banking market have reduced market power disparity between banks, as the percentage of market power disparity attributable to the between countries effect has halved.
PT
Moreover, despite the adverse impact of the crisis on integration, bank market power
CE
differences have not increased in the eurozone. Indeed, the European banks' market power disparity has continued to decline in the period of ECB quantitative easing and in
AC
the wake of the banking union plans/initiative announced in 2012. However, the factors specific to each country (rivalry among banks, market structure, barriers to entry, elasticity of demand, contestability, etc.) are responsible for a large percentage of the market power disparity observed across the banks analysed, particularly during the years since the recent financial crisis. This result is consistent with the evidence obtained by Lucotte (2015), which suggests the existence of large dissimilarities between the eurozone countries in terms of banking structure in the wake of the recent 11
ACCEPTED MANUSCRIPT financial crisis. Lucotte also flags the importance of working on a banking union “that will mitigate cross-country differences in terms of banking structure”.
4. Conclusions Despite the measures rolled out since the EMU was created with the aim of progressing
CR IP T
towards a single banking market, the level of cross-border integration in the retail banking sector remains very low, as is evidenced by the reduced level of cross-border lending and deposit-taking activity. The limited integration of retail markets is related to a lack of cross-border bank mergers and acquisitions within the euro area. However,
AN US
despite the slow pace of progress on integration, the reduction of barriers to competition (particularly by harmonising regulations) and the increased contestability of the market have driven a reduction in market power disparity across the eurozone's banks.
M
By estimating the Lerner market power index and decomposing the Theil index into a within-country effect and a between-countries effect, this paper finds that market power
ED
disparity among eurozone banks has narrowed and that the reduction is attributable to the convergence towards average market power levels across the European banking
PT
sectors. In contrast, the differences within countries have remained stable and were
CE
responsible in 2014 (the last year analysed) for nearly 90% of disparity in the EMU banks' market power. As a result, the measures adopted at the European level to
AC
advance towards a single banking market (such as Banking Union) should be complemented by measures at the national level that are designed to foster competition among the banks within a given country. It would also help to generate more competition cross-border M&A within the euro area, since unlike the M&A between banks of the same country, they do not increase the concentration of the domestic markets.
12
ACCEPTED MANUSCRIPT
Acknowledgements
CR IP T
Authors thank an anonymous referee for helpful comments and suggestions. They gratefully acknowledge financial support of the Spanish Ministry of Science and Innovation (research project ECO2013-43959-R). Joaquin Maudos also acknowledges financial support of Generalitat Valenciana (research project PROMETEOII/2014/046). Paula Cruz-García acknowledges financial support of Spanish Ministry of Education (FPU2014/00936).
References
AN US
Angelini, P. and Cetorelli, N. (2003). The effects of regulatory reform on competition in the banking industry, Journal of Money, Credit and Banking 35, 663684. Carbó, S., Humphrey, D., Maudos, J. and Molyneux, P. (2009). "Cross-country
M
comparisons of competition and pricing power in European Banking", Journal of International Money and Finance, 28, 115-134.
ED
European Central Bank (2016). Financial integration in Europe, April. European Central Bank (2017). Financial integration in Europe, May.
PT
Fernández de Guevara, J., Maudos, J. and Pérez, F. (2007). "Integration and
CE
competition in the European financial markets", Journal of International Money and Finance 26, 26-45.
AC
Koetter, M., J. Kolari, and L. Spierdijk. (2012). “Enjoying the Quiet Life under
Deregulation? Evidence from Adjusted Lerner Indices for US Banks.” Review of Economics and Statistics 94, 462–480. Liu, H. and Wilson, J. (2013). “Competition and risk in Japanese banking”, European Journal of Finance, 19(1), 1-18.
13
ACCEPTED MANUSCRIPT Lucotte, Y. (2015). “Euro area banking fragmentation in the aftermath of the crisis: a cluster analysis.” Applied Economics Letters 22 (13), 1046-1050. Maudos, J. and Fernández de Guevara, J. (2004). “Factors explaining the interest margin in the banking sectors of the European Union”, Journal of Banking and Finance
AC
CE
PT
ED
M
AN US
CR IP T
28, 2259-2281.
14