Upload
harvey-henderson
View
213
Download
1
Embed Size (px)
Citation preview
Location Decisions of Foreign Banks and Competitive Advantage
Stijn Claessens and Neeltje van Horen
13th Dubrovnik Economic Conference
Dubrovnik June 27-28, 2007
2
IntroductionIntroduction
Aim of paper:Test how institutional differences between countries are dealt with, using information on location decisions of foreign banks (as banking is an institutionally-intensive activity)
Specifically, examine whether banks seek out those markets where institutional familiarity provides them with a competitive advantage over other foreign competitor banks
3
Several factors identified that affect decision to go abroad and enter a specific country :
Internalization of banks is traditionally closely tied to internalization of non-financial firms
Several studies show positive correlation between FDI in banking and trade and FDI between host and source country (Grosse and Goldberg, 1991, Brealey and Kaplanis, 1996, Williams, 1998, and Yamori, 1998)
Banks engage in FDI to increase profitability within acceptable risk profile and risk diversification goals
Preference for countries with high expected economic growth and inefficient domestic banks (Focarelli and Pozzolo, 2000)FDI higher between countries with similar legal origin, banking regulations and institutional set ups (Galindo, Micco, and Serra, 2003)
IntroductionIntroduction
4
This paper argues that a bank’s competitive advantage due to familiarity with working in a certain investment climate can be another important determinant of foreign bank entry
IntroductionIntroduction
5
IntroductionIntroduction
Motivation (theory):Internalization theory asserts that firms expand abroad to exploit knowledge advantage created within the firm (Casson, 1987).
To benefit most of this internal knowledge advantage, firms are best off to expand to an environment that is most equal to one they are already familiar with (Buckley and Casson, 1991)
For banks concept of internal knowledge relates especially to information asymmetries and principal agent issues, so theory would suggest banks enter countries with similar information intensities/ institutional environments (supported by empirical work of Galindo, Micco, and Serra, 2003)
6
IntroductionIntroduction
Motivation (theory) cont’d:However, argument implicitly assumes that location decision of individual banks is made independent of the location decision of other, competing, multinational banks. This is unlikely to be the case
So, to the extent that sources of internal competitive advantage are derived from the ability to work in a certain institutional environment, it should be the difference in institutional environment between host and source country taking into account institutional quality of competitors that matters
7
IntroductionIntroduction
Motivation (practice):Figure 4.5 South-South foreign bank entry in developing countries, by country income levelShare of banks in total foreign banks Share of assets in total foreign assets
Source: World Bank Staff estimates based on Bankscope
Note: "Southern foreign banks" are those banks headquartered in a developing country. A foreign bank is one that had at least 50 percent foreign ownership as of December 2005.
a. Number of southern foreign banks as a percentage of all foreign banks (left panel).
b. Bank assets held by southern foreign banks as a percentage of total foreign assets, averaged over 2000-4 (right panel).
0
5
10
15
20
2530
35
40
45
50
Low-incomecountries
Middle-incomecountries
All developingcountries
Percentage
0
5
10
15
20
25
30
35
40
45
50
Low-incomecountries
Middle-incomecountries
All developingcountries
Percentage
8
Results:
Using detailed data on bilateral banking ownership for 137 countries, we find that
Similarities in institutions between host and source country compared to competitors affect a bank’s entry decision. This is more important than differences in institutional development between host and source country
So, a bank is, among other things, attracted to a certain market, because of its ability to work within a certain institutional environment better than its competitors.
Our results suggest that competitive advantage related to institutional environment is an important driving factor in entry decisions made by foreign banks
IntroductionIntroduction
9
Bilateral data on banking sector FDI for period 1995-2006
Primary source is Bankscope, supplemented with information from banking regulatory agency/central bank, bank’s website, annual reports, reports on corporate governance, local stock exchange, SEC’s Form F-20, parent company’s website, country reports, etc.
Sample includes all active and inactive commercial banks, saving banks, cooperative banks, bank holding companies and middle and long term credit banks that are available in Bankscope and have been operating for at lease one year in 1995-2006.
Ownership is based on direct ownership structure in 2005Bank is foreign owned if at least 50% of shares are owned by foreigners.Percentage of shares are summed by country of residence of shareholder. Country with highest percentage of shares is appointed as source country.
Countries in sampleAll high-income and almost all developing countries. Host countries <5 active banks are excluded.For developing countries all banks available in Bankscope were included, for high-income countries only the largest banks (based on 2005 Top 3000 from Banker’s Almanac.
Information about 4,148 banks of which 35 percent is foreign owned (in 2006) in 137 countries.
Data
10
Formally we test the following hypothesis :Banks from countries with relatively weak institutions compared to their competitors will enter countries with relatively weak institutions, while conversely banks from countries with stronger institutions compared to their competitors will enter countries with relatively good institutions.
MethodologyMethodology
11
Measure of foreign bank presence (dependent variable):For each host country we construct country-pairs with all possible source countries in sample
Source countries are all developing and developed countries with presence in bank sector of at least one country
Host and source countries that are offshore centers are excluded. Luxembourg is also excluded.
Host countries with no entry in the sample period are excluded
For each host country we determine per source country the change between 1996 and 2005 in the number of foreign banks from source country j in host country i. Exits are not taken into account, i.e. dependent variable captures all new investments.
4. Methodology4. Methodology
12
Measure of competitive advantage:Difference between institutional quality in source country and that of all the bank’s competitor source countries times the difference between the institutional quality in the target host country and that of competing host countries
is weighted average (based on economic size) of institutional quality in each of the possible source countries is weighted average (based on economic size) of institutional quality in each of the possible host countriesInstitutional quality in host and source country and of competitors is simple average of six governance indicators of Kaufmann, Kraay and Mastruzzi (2005), linearly transformed so value is not below zero
StInstcomp
MethodologyMethodology
)(*)( HtitStjtijt InstcompInsthostInstcompInstsourceCompAdv
HtInstcomp
13
Measure of competitive advantage:
The variable can be positive or negative, with a positive value indicating a competitive advantage of a bank from source country j in host country i and a negative value indicating a competitive disadvantage.
MethodologyMethodology
Quality inst. host – Quality inst. host competitors
Quality inst. source – Quality inst source competitors
Result
Host inst.>Comp inst Source inst.>Comp inst. +
Host inst.<Comp inst. Source inst.>Comp inst. -
Host inst.>Comp inst Source inst.<Comp inst. -
Host inst.<Comp inst. Source inst.<Comp inst. +
14
Use difference-in-differences model. Allows to control for all country-pair fixed effects, host-country fixed effects, source-country fixed effects
Dependent variable is change between 1996 and 2005, explanatory variables change between 1995 and 2004
Benchmark model:
Model is estimated using Tobit and standard errors are corrected for heteroskedasticity. Weighted estimation (inverse of dollar GDP of source country).
MethodologyMethodology
ijjj
iiijij
rcedGDPcapsoudGDPSource
dEntryrestdInsthostdCompadvcedForpresen
54
321
15
Competitive advantage explains cross-border banking activity, absolute differences in institutional environment matter less
Support for competitive advantage hypothesis
Impact economic very relevant: Improvement in competitive advantage of 3 (maximum in our sample) leads to an increase in presence of banks from country j in country i of 0.13 (mean is 0.018).
ResultsResults
dCompadv 0.043 *** 0.042 *** 0.042 *** 0.039 *** 0.252 **[0.000] [0.000] [0.000] [0.000] [0.027]
dInsthost 0.047 *** 0.043 ***[0.000] [0.000]
dDiffInst -0.008[0.307]
No. Obs. 9,306 9,306 9,306 9,389 9,389
(5)
Competitive Advantage in Foreign Banking
(1) (2) (3) (4)
IV - Tobit1996-2005 1996-2005 1996-2005 2000-2005 2000-2005
Tobit Tobit Tobit Tobit
Note:Coefficients are marginal effects. The robust p-values appear in brackets and ***, ** and * correspond to one, five and ten percent level of significance respectively.
16
Competitive advantage is more important for FDI through M&A
Also improvements in institutional quality more important for M&A
Possible explanations:Greenfields often small investments with little local financial intermediation while M&As are larger and more integrated in domestic banking sectorGreenfields allow for use of same banking technology as at home, while M&A’s (at least initially) need to work with existing technology, processes and procedures
ResultsResults
dCompadv 0.013 *** 0.024 ***[0.000] [0.000]
dInsthost 0.010 ** 0.033 ***[0.030] [0.000]
No. Obs. 9,306 9,306
(1) (2)
Competitive advantage in foreign banking - Greenfield versus M&As
Tobit TobitGreenfield M&As
Note:Coefficients are marginal effects. The robust p-values appear in brackets and ***, ** and * correspond to one, five and ten percent level of significance respectively.
17
Competitive advantage with respect to regulations, rule of law and corruption seem to matter most explaining location decisions of foreign banks
ResultsResults
dCompadv_vca -0.015 *** -0.006 *[0.000] [0.092]
dInsthost_vca 0.032 *** 0.031 ***[0.000] [0.000]
dCompadv_stab 0.016 *** 0.004 *[0.000] [0.065]
dInsthost_stab 0.005 -0.010 **[0.199] [0.016]
dCompadv_gov 0.033 *** -0.007[0.000] [0.183]
dInsthost_gov 0.056 *** 0.010[0.000] [0.267]
dCompadv_reg 0.012 *** 0.009 ***[0.000] [0.000]
dInsthost_reg 0.019 *** -0.003[0.007] [0.552]
dCompadv_rule 0.031 *** 0.015 **[0.000] [0.012]
dInsthost_rule 0.044 *** 0.026 ***[0.000] [0.004]
dCompadv_corr 0.019 *** 0.007 **[0.000] [0.019]
dInsthost_corr 0.019 *** -0.005[0.007] [0.484]
No. Obs. 9,306 9,140 9,223 9,223 9,223 9,223 9,140
Competitive advantage in foreign banking - which institutions matter?(1) (2) (3) (4) (5) (6) (7)
18
Results indicate that banks that are willing to expand their business abroad seek out those markets in which their past experience in working in a certain business climate gives them a competitive advantage
Result has important policy implicationHigh institutional quality is not necessarily a prerequisite to be able to attract FDI in banking. Since foreign banks tend to have a beneficial impact on the domestic financial system which is an engine for growth, this is potentially good news for low-income countries
However, some caution is warranted for as those banks might be immiserizing. Foreign banks entering low-income countries might be a source of instability if they lack supervision in source country. Also they might take advantage of the weak institutional environment and exploit safety nets by taking on excessive risks
ConclusionConclusion
19
EndEnd