Business School

W O R K I N G P A P E R S E R I E S

IPAG working papers are circulated for discussion and comments only. They have not been

peer-reviewed and may not be reproduced without permission of the authors.

Working Paper

2014-366

Determinants of the Capital Adequacy

Ratio of a Foreign Bank’s Subsidiaries:

The Role of the Interbank Market and

Regulation of Multinational Banks

Mehdi Mili

Jean-Michel Sahut

Hatem Trimeche

http://www.ipag.fr/fr/accueil/la-recherche/publications-WP.html

IPAG Business School

184, Boulevard Saint-Germain

75006 Paris

France

1

Determinants of the Capital Adequacy Ratio of a Foreign Bank’s

Subsidiaries: The Role of the Interbank Market and Regulation of

Multinational Banks

Mehdi Milia , Jean-Michel Sahut

b and Hatem Trimeche

c

a MODESFI - University of Sfax & I.S.G. Sousse - Tunisia

b IPAG Business School Paris

b High School of Management of Sousse - Tunisia

Abstract

This paper examines the factors influencing the capital adequacy ratio (CAR) of

foreign banks’ subsidiaries. We use data from 340 subsidiaries of 123

multinational banks and test whether the subsidiaries’ capital ratio depends on the

parent banks’ fundamentals. We also study the role of the economic conditions

and regulatory environment in the bank’s home country while determining the

CAR of its foreign subsidiaries. Our results provide strong evidence that the CAR

of subsidiaries operating in developing and developed countries do not depend on

the same set of explanatory factors. We also find that the regulatory framework of

a parent bank’s home country affects the capitalization of its foreign subsidiaries

in the host countries. Finally, we show that specific variables of the parent bank

have a stronger effect on foreign banks that are closely related to the interbank

market.

Keywords: Capital Adequacy Ratio, Multinational Banks, Interbank Market

JEL codes: F15, F34, G21

1. Introduction International banking regulations based on recommendations of the Basel Committee have

been adopted by banks and international regulators with the aim of promoting the soundness

of the global banking system. These regulations require that banks properly assess the risks

they take, since a weak banking system may threaten the stability of the economy on a

national and international scale.

Generally perceived as a prerequisite for the proper functioning of global banking systems, the regulatory model recommended by the Basel Committee is fundamentally based upon

the "paradigm of supervision of bank capital," as proposed by Caprio and Honohan (1999).

2

In this context, the Capital Adequacy Ratio (CAR) has received particular attention from

international regulatory authorities; it is one of the principle regulatory tools used to control and scrutinize a bank’s financial health.

According to Sinkey (1989), regulators use the CAR as an important measure of the "safety

and soundness" of depository institutions because they consider capital a safety margin

capable of absorbing potential losses. Several studies have been conducted on the capital structure of banks located in several countries in order to identify the determinants of the

CAR, a majority of which have tried to explain the impact of the financial health of a bank on

the evolution of the CAR.

The concept of capital adequacy emerged in the mid ‘70s, due to the expansion of lending by

banks without a parallel increase in their capital. This has forced regulators to define several

control procedures and issue new reforms in order to avoid insolvency in the banking sector.

In the same context, i.e., protection against shocks, capital requirements appear as the most

frequently cited prudential regulation.

The importance of the topic of regulatory capital in ensuring the stability of banking systems

has motivated several researchers1 to study the determinants of regulatory capital banks.

Our investigation of the literature allows us to deduce that the works on the determinants of banks' CAR may be split into two methodological paths.

One path focuses on explaining the level of capitalization in a bank, using only financial variables. Demsetz and Strahan (1997) and Ayuso and Saurina (2004) showed that large banks have the ability to operate with low levels of capital; this finding indicates that large

banks could benefit from diversification and can therefore, operate with lower capital ratios.

Gropp and Heider (2009) and Kleff and Weber (2008) claimed that the most profitable banks

tend to have relatively higher regulatory capital. Yu (2000) demonstrated that liquidity is

positively and significantly related to capital ratio, which implies that banks do not consider liquidity a pure direct substitute for capital; thus, it cannot be used to cover whole portfolio

risk.

The second path has attempted to include macroeconomic variables. These studies suggest that the national economic environment has an effect on the solvency of the bank; therefore,

it should be considered a fundamental determinant of the CAR of a bank. For example,

Hortlund (2005) tested the impact of inflation on the capitalization of Swedish banks and demonstrated that inflation and the banks regulatory capital ratios were inversely related.

Williams (1998) studied the impact of macroeconomic variables on the CAR; he noted that macroeconomic variables such as inflation, real exchange rate, money supply, political instability, and return on investment are significant determinants of regulatory capital.

From this overview of the literature, we note that no previous studies have focused on

examining the determinants of regulatory capital of foreign subsidiaries. The purpose of this work is to examine the determinants of the CAR of foreign banks in order to find the role of the parent bank therein. We also contribute to this line of research by examining the role of the economic conditions and regulatory environment in the bank’s home country in the

determination of the CAR of its foreign subsidiaries.

1 Agrawal and Jacque (2001), Berger et al. (2008), and Kleff and Weber (2008).

3

Our research is based on literature showing that parent banks exert significant influence on the financial fundamentals of their foreign subsidiaries. Dee Haas and van Lelyveld (2006 &

2010) and Martinez Peria et al. (2002) suggested that parent banks support their foreign subsidiaries, especially during periods of crisis, which allows them to be resilient to economic shocks. Morgan et al. (2004) developed a theoretical framework explaining how

lending by multinational bank’s subsidiaries is influenced by the funding strategy of their parent banks; they described the funding mechanism between them, using an internal capital market. This evidence reported in the literature, concerning the parent banks’ effect on their foreign subsidiaries, led us to test how the parent banks affects the level of capitalization of the subsidiaries.

This paper is organized as follows: the first part presents a review of the literature and

theoretical perspectives to explain the determinants of the bank’s solvency ratio, while

emphasizing the role of the parent bank in its determination, and the second part presents the

methodology and results of our research.

2. Literature review and selection of variables

Because of the recent trend towards financial liberalization, as well as increased cross-border

activities by several banks, banking supervision needs to cover all banking transactions

worldwide. Indeed, for multinational banks, control of foreign banks’ subsidiaries comes from

regulatory authorities in both the home and host country. Multinational banks generally have

complex organizations which regulators find challenging. To address this, coordination between the authorities in the country of origin and destination is required. De Haas and

van Lelyveld (2010) suggested that regulatory control procedures for foreign subsidiaries

focus on financial indicators (solvency, liquidity risk) that depend on the stability of the

national and international banking systems.

On a theoretical level, our study is related to the work of Harr and Ronde (2006), who studied the capital regulation of multinational banks. In their studies, they analyzed cases

wherein multinational subsidiaries are subject to two national regulators who strategically set capital requirements; they concluded that capital requirements may increase when foreign

subsidiaries are subject to two regulators. Other studies have focused on the problem of

divergent interests and lack of coordination between the national regulators2.

There are at least two reasons justifying our hypothesis that the financial characteristics of the parent bank affect the capitalization of its foreign subsidiaries. First, the problem of the

subsidiary’s solvency is part of overall risk management for the parent bank. Thus, the capitalization of the subsidiary is not a strategy executed independently in the home

country. Second, the integration of financial systems across the world and the efforts

deployed to harmonize the system of banking supervision worldwide justifies the fact that

central banks should care about how foreign banks operate in host countries and about their

capitalization, in particular.

2 Holthausen and Ronde (2005), Acharya (2003), Dell’Ariccia and Marquez (2006), Calzolari and Loranth

(2003), and Dalen and Olsen (2003).

4

Since the effort to regulate multinational banks and their foreign subsidiaries is built around the regulatory capital of the subsidiary, in this paper, we seek to examine the determinant of the CAR for the subsidiaries.

Our methodological approach will be conducted in three parts. First, we test the impact of the financial fundamental variables of the subsidiary and the parent bank, and integrate the

macroeconomic variables of the subsidiary’s host country and the parent bank’s home country. Then, we test the impact of these variables by distinguishing the effect according to the economic maturity of the host country. This leads us to consider the impact of the

regulatory environment of the parent bank’s home country. Lastly, we test the role of the

interbank market.

Our objective is to examine the determinants of the CAR of the subsidiary, which constitutes the dependent variable in our study. CAR is defined as the capital of a bank expressed as a percentage of its risk-weighted commitments. The international standard recommends a minimum ratio that ensures that banks can absorb a reasonable level of losses before going into financial distress. The application of these minimum capital ratios protects depositors

and promotes the stability and efficiency of the financial system as a whole.

Two types of capital ratios have been proposed. The first relates to Tier 1 capital, which

must absorb the losses of the bank without leading to a cessation of activities. The second is

Tier 2 capital, usually subordinated debt, which is intended to absorb losses in the event of liquidation and thus, provides a modicum of protection to depositors. In this study, we use the CAR formula based on both Tier 1 and Tier 2 capitals.

According to Buchs and Mathisen (2005), these ratios measure the strength of a bank's capital, and its ability to cover the risks of its undertakings and protect the interests of its depositors. This could enhance the stability and efficiency of the banking system. Figuet (2003) stated that capital helps to limit bank failures and consequently, the costs of its rescue

and restructuring. Diamond and Rajan (2000) stated that high capital reduces liquidity creation by the bank but allows it to be strong and avoid bankruptcy.

In our empirical framework, the risk-adjusted capital (RAC)of the subsidiaries will be explained by two types of variables – on the one hand, we use specific financial variables of

subsidiaries and parent banks, and on the other, we consider macroeconomics variables of country of origin and host country.

The group of specific financial variables of subsidiaries and parent banks includes:

Size. The size of the bank may impact its diversification strategies, its risks, and its access to

capital markets. This is proved by Demsetz and Strahan (1997), who concluded that large banks enjoy better diversification and therefore, operate with lower capital ratios. In the

same vein, Ayuso et al. (2004) showed that large banks have lower capital levels. We use the logarithm of total assets (T_Asset) as a measure of size, and we expect an inverse relationship between size and CAR.

Deposit Ratio. An increase in bank deposits requires regulation to guarantee the rights of depositors and preserve the bank’s solvency. The ratio of total deposits to total assets is used to measure the impact of deposits upon capital. According to Kleff and Weber (2008),

deposits are generally considered cheaper sources of funds, with respect to credit and other similar financing instruments such as bond financing or securitization of loans. Based on the

5

study by Asarkaya and Özcan (2007), we expect deposits to have a negative effect on the

CAR.

Loan Ratio. Increased loans expose the bank to higher risk and therefore, the CAR should

increase. According to Hassan and Bashir (2003), higher ratios of loans may reduce the

bank’s liquidity and increase the number of defaulters. Mpuga (2002) found a positive

relationship between CAR and the share of loans. The loan ratio is measured by total loans to

total assets.

Profitability. Most of the studies in the literature claim that profitability has a significant effect on the bank’s CAR. Brown et al. (2010) and Gropp and Heider (2009) concluded that

the most profitable banks tend to have relatively lower levels of regulatory capital, as compared to the less profitable ones. On the basis of these arguments, we expect a negative

relationship between profitability ratios of assets [return on assets (ROA)] and the

capitalization of the subsidiary.

Net Interest Margin (NIM). The NIM is a parameter that reflects the profitability of banks. Moreover, it reflects both the volume and composition of assets and liabilities, and it influences the risk portfolio of the bank. According to Angbazo (1997), adequate NIMs are expected to generate sufficient income to increase capital with increasing exposure to risk. Berger (1995) and Demirguc-Kunt and Huizinga (1998 & 2002) found a positive relationship between the NIM and the ratio of equity.

The second group of variables includes macroeconomic variables of the parent bank’s home

country and the host country in which the foreign subsidiary operates.

GDP growth. In accordance with most studies, such as Ruckes (2004), we expect a negative

relationship between credit growth and capital ratios. We state that in times of strong

economic growth, bank risk is lower, which leads banks to reduce their regulatory capital.

Economic periods of low growth increase a bank’s financial risk and encourage banks to

maintain a high capital ratio.

Real interest rates. Demirgüç-Kunt and Detragiache (1998) showed that higher real interest rates can adversely affect the repayment capacity of borrowers. This can have a negative effect upon capital ratios, should many borrowers default on their payments. We therefore

expect a negative relationship between real interest rates and capital ratios.

Real effective exchange rate. Williams (1998) found that the real effective exchange rate is another important determinant of the CAR. He identified an inverse relationship between

the exchange rate and the CAR that implies that an increase in the real exchange rate

reduces foreign direct investment and thus, reduces the CAR.

Crisis. We take any crisis experienced by the parent bank as a silent variable in this study to test its impact on the determination of banking subsidiaries’ CAR. This dummy variable

equals 1 for the period 2007-2008 and 0 for others.

3. Data and preliminary statistics

Our database consists of 340 subsidiaries of 123 foreign banks ranked in a 2010 issue of

Banker magazine. Among the foreign subsidiaries finally selected, 180 operate in developed

countries and 130 in developing countries. Our database combines both financial variables

6

specific to banks and macroeconomic variables that reflect the economic conditions in the

parent bank’s home country and the subsidiary’s host country. Financial variables for

subsidiaries and parent banks are collected for the period 2000-2010 from Bankscope, and all

variables are expressed in U.S. dollars. Our macroeconomic data used for each country were

collected from the International Monetary Fund database. Table 1 rovides a summary of the

variables used in our empirical estimation, along with their sources.

Table 1.

Table 2 presents the summary statistics of specific variables for multinational parent banks

and their foreign subsidiaries. The table shows that subsidiaries differ from their parent banks in terms of financial structure and performance. On an average, parent banks appear slightly better capitalized than their subsidiaries. In addition, we find that the size of parent banks, expressed by the logarithm of total assets, is greater than that of their subsidiaries and their loan loss provisions appear higher than those of their subsidiaries. This is probably on account of the high volume of activity, in terms of loans granted by the parent banks.

Comparing subsidiaries in developed and developing countries, our descriptive statistics are consistent with those of Allen et al. (2012); we note that the subsidiaries in developed countries are more capitalized than those in developing countries. This can be explained by the fact that regulatory requirements are generally more stringent in developed countries. The table also shows that multinational banks tend to set up subsidiaries of larger sizes in developed countries, where the banking industry is more competitive. Subsidiaries in developed countries also seem to perform better, since they have a higher ROA and lower loan loss provision. Table 2.

4. Empirical results

Our econometric approach is similar to that used by Allen et al. (2012) and Dee Haas and van

Lelyveld (2010), which consists of explaining the CAR of foreign subsidiaries by the

fundamental variables of the subsidiary and its parent bank, and the economic variables of the

host country and home country. Formally, we consider the following model:

(1)

Where ΔCAR it is the change in capital adequacy of subsidiary i in year t, Bankit is a matrix of

foreign subsidiary controls of subsidiary i, and Countryit is a matrix of macroeconomic

variables of the country where subsidiary i is located. ParentBankit is a matrix of parent bank

characteristics of subsidiary i in year t, and ParentCountryit is a matrix of macroeconomic

variables of the country of the parent bank of subsidiary i.

Table 3 presents the estimation results of all foreign subsidiaries taken together. The first column in this table reports the results of the base model line when the CAR of the

subsidiary is regressed against their specific characteristics and the macroeconomic variables

of host countries. Consistent with previous works, the coefficient of loan loss provision is

7

positive and statistically significant at less than the 1% threshold, suggesting that bad loans

increased the subsidiary’s risk, which requires greater capitalization.

The increase in loans tends to reduce the CAR; this makes it clear that regulatory capital may fall after a period of credit expansion. Contrary to Ayuso et al. (2004), who found that

larger subsidiaries hold lower levels of capital, we found that the size of the subsidiary is

negative but has no significant effect on the CAR. Thus, the management of regulatory capital may depend much more on the size of the parent bank, as the management of the

subsidiary’s risk is centralized at the parent bank.

Among the macroeconomic variables, economic growth and real interest rates seem to significantly affect the capital ratio of subsidiaries. The coefficient of GDP growth exhibits a positive and significant sign which is in line with the findings of Schaeck and Čihák (2007), who suggested that a high level of economic development requires sophisticated procedures for banking supervision.

8

Table 3.

The coefficient of real interest rate appears significantly negative. One possible explanation is that rising real interest rates adversely affect borrowers’ ability to repay their bank loans. This can negatively impact capital ratios if many borrowers default on their payments. The same result is well documented in the literature by Demirgüç-Kunt and Detragiache (1998).

In the second and third columns in Table 3, we expand our baseline model by adding specific

variables of the parent bank and macro variables of the home country to the respective

columns. The coefficients of the size of the parent banks appear significantly negative in both

models. This finding implies that larger multinational banks are generally better able to

benefit from international diversification, therefore allowing their subsidiaries to operate with

lower capital ratios. This is in line with the findings of Gropp and Heider (2009) and Shrieves

and Dahl (1992), who suggested that large banks benefit from more flexible regulation of their CAR, as compared to small banks.

The significantly positive coefficient of loan loss provision of the parent bank indicates that the deterioration of the quality of its loan portfolio creates a higher risk of distress for its subsidiary.

The NIM of the parent bank has a positive and significant impact at 5% in the first model. This is consistent with previous studies conducted by Gropp and Heider (2009) and Kleff and Weber (2008). The parent bank’s NIM expresses a positive but insignificant impact upon the CAR of the subsidiary, indicating that the operating performance of the parent bank is most important in determining the level of capital for the subsidiary. The table shows that the

levels of deposits, ROA, return on equity, liquidity, and leverage have no effect on the

subsidiary’s CAR.

Regarding the host country’s macroeconomic variables, we find in both models a positive but insignificant correlation between the CAR and the GDP growth rate. In fact, only the real interest rate has a negative and significant effect at 10%. The other macroeconomic variables chosen in all models have an insignificant effect on the subsidiary’s CAR. In addition, the dummy variable (i.e., the crisis) has no significant effect on the determination of our dependent variable.

In the third model, a new dummy variable – Crisis, whose coefficient is positive – has been

added in order to capture the effect of the last financial crisis on the capitalization of the

foreign subsidiaries. This suggests that the increase in risk for a bank in periods of economic

instability leads parent banks to increase the capitalization of their foreign subsidiaries.

In order to understand the impact of the regulatory environment of the parent bank’s home

country on the capitalization of its foreign subsidiary, our sample has been divided into two

groups – foreign subsidiaries operating in developed countries, and foreign subsidiaries

operating in developing countries (World Bank in 2008). In addition, the Capital Regulation Index of the home country is included as an explanatory variable. This index, developed by the World Bank, surveys bank regulation (2003, 2007, &

2011) and combines two dimensions. First, it reflects the overall capital stringency in the

banking sector, demonstrating whether the capital requirement reflects certain risk aspects and

9

deducts certain market value losses from capital before minimum capital adequacy is

determined. Second, this index values the initial capital stringency, which indicates whether

funds may be used to initially capitalize a bank and whether they are officially used. The

Capital Regulation Index ranges from 0 to 10, with a high value indicating great stringency.

We expect that banking regulations imposed in the home country have an impact on the

capitalization of subsidiaries in foreign countries.

The results are reported in Table 4, which shows that the CAR of foreign subsidiaries in

developing and developed countries is not determined by the same set of factors. The capital

ratio of subsidiaries in developing countries, in particular, depends more on specific variables

of the bank and macroeconomics in the host country. However, in developed countries, the

parent bank’s fundamentals significantly determine the CAR level. These results show, in

some way, that banking regulation in developed countries is more stringent than in developing

countries, and takes into account the financial health of the parent bank and the home country.

The table shows that parent banks with a significant size and strong ROA are more likely to

reduce the CAR of their foreign subsidiaries operating in developed countries. While these

variables (i.e., parent bank’s size and ROA) appear insignificant for subsidiaries operating in

developed countries, the parent bank’s loan loss provisions seem to affect its foreign

subsidiaries operating in both developed and developing countries, but with higher impact on

those in developed countries.

The coefficients of loan loss provisions and ROA of the foreign subsidiaries are positive in

models for developing countries, but insignificant in developed countries. Two possible

explanations may be adopted for these. First, foreign subsidiaries in developing countries

increase their regulatory capital with increased risk in their portfolio, hence affecting the

quality of their loan. Second, profitable banks increase their capital through retained earnings.

The consequence of macroeconomics in the host country on the subsidiary is more

pronounced in developing countries due to negative GDP growth and real interest rates

(around a level of 1%). In developed countries, GDP appears significant only at the 10%

level.

In addition, GDP growth in the banks’ home countries has different effects on the CAR for

the two groups of subsidiaries. In fact, low GDP growth in a bank’s home country may

increase the lending activity of its foreign subsidiaries, which results in higher capitalization.

Consequently, higher capitalization is required for foreign subsidiaries operating in developed

countries. However, the macroeconomic variables of home countries have no significant

effect on the capitalization of foreign subsidiaries operating in developing countries.

The table also shows that the parent bank’s fragility negatively affects the CAR of its foreign

subsidiaries in developed countries. This can be noticed based on the coefficient of the NIM

and the ROA of the parent bank, which seem to be significant for the developed host countries

but insignificant for the developing host countries. Nevertheless, other bank characteristics do

not have any impact on either group of foreign subsidiaries. In addition, the size of the parent

bank has a positive and significant impact on the capitalization of its subsidiaries in developed

countries, which could indicate that large multinational banks tend to have higher

capitalization in developed countries, where the banking sector is highly competitive.

Table 4.

10

The coefficient of the dummy variable, Crisis, seems to be significantly positive only for

foreign subsidiaries operating in developed countries. This shows that the capitalization of

subsidiaries in these countries is strongly linked to the economic situation of the host country.

Taking into account the integration of financial markets and banking sectors, regulatory

authorities in developed countries require strong capitalization in the foreign subsidiaries

operating in their markets.

The regulatory index of the home country seems to affect the two groups of subsidiaries

differently.

Parent banks from countries that practice stricter banking regulations reduce the capitalization

of their foreign subsidiaries operating in developing countries, if they have the option to do

so. However, in host countries in the developed world, where banking regulation is usually

more stringent, parent banks have to increase the capitalization of their subsidiaries.

Lastly, the sample had been divided into two sub-samples in order to explore the effect of the

integration of the subsidiary’s capitalization with the interbank market and to adjust for

market discipline.The interbank ratio is measured by interbank lending to borrowing.

The first sub-sample consists of banks with an interbank rate below one and is considered

highly interbank borrowing related. This means that the first group of foreign subsidiaries

finances their foreign loans from the interbank market, which is a risky strategy for many

banks in periods such as the recent crisis. However, an interbank ratio above one means that

the foreign subsidiary is well-positioned in the interbank market, and so it hadn’t been forced

to take loans during the recent crisis. This reflects the assertion that other banks have the

ability to monitor their peers in the interbank market (Nier and Baumann, 2006) and they have

the incentive, because interbank deposits are not typically covered by deposit protection

schemes.

We expect that the integration of the foreign subsidiary with the interbank market has an

impact on its Regulatory Capitalization Ratio. The results of these estimations are presented

in Table 5.

Globally, the results show that CARs are more dependent on subsidiary fundamentals that are

highly related to the interbank market. The coefficient of loan loss provision of the

subsidiaries seems positively significant for those that are heavily related to the interbank

market. This result suggests that the foreign subsidiary’s own financial fragility leads it to

increase its CAR.

The size variable seems to be negatively significant only for foreign subsidiaries that are

highly related to the interbank market. This implies that the larger subsidiaries that are highly

integrated in the banking sector in the host country can diversify their risk portfolio, thus

opting to reduce their CAR. The profitability of the subsidiary has a significant impact on

both groups. This effect is stronger for those foreign banks that can borrow in the interbank

market.

Regarding the sub-sample of foreign subsidiaries that are slightly related to the interbank

market, the impact from their parent bank’s capitalization is significant. This may be

explained by the fact that the subsidiary’s funding strategy is dependent on its parent bank,

and its capitalization level depends on the parent bank’s global risk portfolio. Their loan loss

provision and NIM variables specifically, are significantly positive for the second group of

subsidiaries. This shows that the subsidiaries with a better position in the interbank market are

not overly controlled in the host country markets and their level of regulatory capital depends

on the loan quality and operational performance of their parent bank.

11

Table 5. Capital Adequacy Ratio and Interbank Dependency

We also document that macroeconomic variables of the host country play an important role in

the capitalization of the subsidiaries which are highly related to the interbank market.

Therefore, we find that the CAR depends significantly on the GDP growth in the country.

This implies that foreign subsidiaries operating in the interbank market are dependent on the

monetary and financial conditions in their host countries. However, all macroeconomic

variables are insignificant for foreign subsidiaries with low dependence on the interbank

market.

Conclusion

The study investigates the determinants of the CAR of foreign subsidiaries. We provide

strong evidence that the CAR of subsidiaries operating in developed and developing countries

do not depend on the same explanatory factors. We illustrate that the fragility of parent banks

affects the capitalization of their foreign subsidiaries. This is important, especially when the

foreign subsidiary operates in a developed country. This emphasizes the fundamental role

played by multinational banks in the stability of the global banking system.

In addition, our empirical findings show that the regulatory framework of a parent bank’s

home country affects the capitalization of its foreign subsidiaries in the host countries.

Another important finding relates to the role of the interbank market in determining the CAR

of subsidiaries. We conclude that the parent bank’s effect is stronger for foreign banks highly

related to the interbank market.

Our results are important due to the implications of an increased opening of the banking

sector in developing countries to multinational banks. We document that the monetary

authorities in developing countries, especially central banks seeking the stability of their

banking sectors, should impose more stringent regulations on foreign subsidiaries, as it could

result in enhanced financial risk in these countries.

References

Acharya, V., 2003, “Is the International Convergence of Capital Adequacy Regulation

desirable?”, Journal of Finance, 58(6), pp. 2745-2781.

Agarwal,R and K.Jacques. 2001. “The impact of FDICIA and prompt corrective action on

bank capital abd risk: Estimates using a simultaneous equations model”, Journal of

Banking and Finance, 15:847-874.

12

Allen F., Hryckiewicz A., Kowalewski O., and Tumer-Alkan G. , 2012. “Transmission of Bank

Liquidity Shocks in Loan and Deposit Markets: The Role of Interbank Borrowing and

Market Monitoring” January 12, 2012 Wharton Financial Institutions Center Working

Paper 10-28

Angbazo, L. 1997 . “Commercial bank net interest margins, default risk, interest-rate risk, and

off-balance sheet banking”, Journal of Banking and Finance, 21(1),1997,pp: 55-87.

Asarkaya Y. and özcan S., 2007. “Determinants of capital structure in financial institutions:

The case of Turkey”. Journal of BRSA and Financial Markets, Vol 1, 1, 91-109.

Ayuso, J., D. Perez, and J. Saurina, 2004, “Are capital buffers procyclical? Evidence from

Spanish panel data”. Journal of Financial Intermediation, Vol. 13, pp. 249–264

Buchs, T. and J. Mathisen, 2005, “Competition and Efficiency in Banking: Behavioral

Evidence from Ghana’, IMF Working Paper WP/05/17, African Department

Berger,A.N.Herring, R.J. “The role of capital in financial institutions”. Journal of Banking

and Finance,1995.

Berger N., De Young R., Mark J., Lee D. and Aozde O., 2008. “How do large banking

manage their capital ratios” Journal of financial research. Vol 34, 2-3, 123-149.

Brown R. and Octavia M., 2010. “Determinants of bank capital structure in developing

countries: regulatory capital requirement versus the standard determinants of capital

structure”. Journal of Emerging markets. Vol 15, 50-62.

Calzolari G. and G. Loranth, 2003, “On the Regulation of Multinational Banks”, Antitrust,

Regulation and Competition: Theory and Practice”, L. Lambertini Ed., Palgrave

Macmillan, pp.265-290.

Caprio, G. and P.Honohan, , 1999. "Beyond capital ideals : restoring banking stability,"

Policy Research Working Paper Series 2235, The World Bank.

Dalen, D.M. and T. Olsen, 2003, “Strategic regulation of a multi-national banking industry",

mimeo”, Norwegian School of Economics and Business Administration.

Dell’Ariccia, G. and R. Marquez, 2006, "Competition among Regulators and Credit Market

Integration”, Journal of Financial Economics, 79, pp. 401-430.

De Haas, R. and I. van Lelyveld, 2006, “Foreign Banks and Credit Stability in Central and

Eastern Europe. A Panel Data Analysis”, Journal of Banking and Finance 30, 1927–

1952.

De Haas, R. and I. van Lelyveld, 2010, “Internal Capital Markets and Lending by

Multinational Bank Subsidiaries”, Journal of Financial Intermediation, vol. 19(1), 1-25.

Demirgüç-Kunt, A., and E. Detragiache, 1998, “The determinants of banking crises in

developing and developed countries”. IMF Staff Papers, Vol. 45 (1), pp. 81–109

———, 2002, Does deposit insurance increase banking system stability? An empirical

investigation. Journal of Monetary Economics, Vol. 49, pp. 1373–1406

Demsetz, R. S., and P. E. Strahan, 1997, “Diversification, size, and risk at U.S. bank holding

companies”. Journal of Money, Credit, and Banking, Vol. 29, pp. 300–313

Diamond, D. W., and Rajan, R. G. 1999. “A theory of bank capital”, National Bureau of

Economic Research.

13

Figuet, J.M. “Le traitement du risque de crédit dans l’accord de Bâle II : Une évaluation” .

Revue d’Économie Financière, n°71, (2003),pp. 271-293.

Gropp R., and F. Heider. 2009. “The Determinants Of Bank Capital Structure, European

Central Bank Working Paper Series N° 1096 / September 2009

Hassan, m.k. and Bashir, Ahm, 2003. “Determinants of Islamic Banking profitability”, Paper

presented at the Economic Research Forum (ERF) 10th Annual Conference,

,Marrakesh, Morocco, 16-18 .

Harr T. and T. Rønde, 2006, “Branch or Subsidiary? Capital Regulation of Multinational

Banks”, mimeo, University of Copenhagen

Holthausen C. and T. Rønde, 2005, "Cooperation in international banking supervision: a

political economy approach”, mimeo, University of Copenhagen and CEPR Discussion

Paper No. 4990.

Hortlund, P., 2005, “Do Inflation and High Taxes Increase Bank Leverage?”, Stockholm

School of Economics ,Working Paper Series in Economics and Finance, No 612

Kleff, V. and Weber, M. 2008 . “How Do Banks Determine Capital? Empirical Evidence

from Research”, Journal of Finance and Accounting ,ISSN 2222-1697 (Paper) ISSN

2222-2847 (Online),Vol 3, No 1, 2008.

Martinez Peria, M.S., A. Powell, and I. Vladkova Hollar, 2002, “Banking on Foreigners: The

Behavior of International Bank Lending to Latin America, 1985–2000”, World Bank

Working Paper No. 2893. World Bank, Washington, DC.

Morgan D. , B. Rime and P.E. Strahan, 2004, “Bank integration and state business volatility”,

Quarterly Journal of Economic.119, pp. 1555-85.

Mpuga P., 2002. “The 1998-99 banking crisis in Uganda: What was the role of the new

capital requirements?” , Journal of Financial Regulation and Compliance,

10(3),2002,pp: 224- 242.

Nier, E., and U. Baumann, 2006, “Market discipline, disclosure and moral hazard in banking”.

Journal of Financial Intermediation, Vol. 15, pp. 332-361

Ruckes, M., 2004, “Bank competition and credit standards”. Review of Financial Studies, Vol.

17, pp. 1073-1102

Schaeck, K., M. Cihak, and S. Wolfe. “Are more competitive banking systems more stable?”,

IMF Working Paper, WP/06/143, Washington, D.C.: International Monetary Fund2006.

Shrieves, R. E. and D. Dahl, 1992, “The relationship between risk and capital in commercial

banks”, Journal of Banking and Finance 16, 439-457.

Sinkey, J.F. Jr. (1989). Commercial Bank Financial Management in the Financial Services

Industry, Third Edition, U.S.A.: MacMillan Publishing Company.

Williams, B. “A pooled study of the profits and size of foreign banks in Australia”, Journal of

Multinational Financial Management, 8, (1998) ,pp. 211–231.

Yu, H.C . “Banks’ capital structure and the liquid asset-policy implication of Taiwan”, Pacific

Economic Review, 5(1) 2000, 109-114.

14

Table .1 Variable Definitions

Variable Description Source

Specific variables CAR Capital adequacy ratio Bankscope

Size The logarithm of total assets of foreign

subsidiaries and parent banks. Bankscope

Deposits Total deposit divided by the total assets of

subsidiary or parent bank deposits. Bankscope

Loans The ratio of total loans to total assets Bankscope

Loan loss provision Ratio of loan loss provisions to net loans Bankscope

ROA Return on Asset Bankscope

NIM Net Interest Margin Bankscope

Macro-variables GDP Yearly change in GDP growth World Bank

Development Indicators

Interest rate

Real interest rate is the lending interest rate

adjusted for inflation as measured by the GDP

deflator.

World Bank

Development Indicators

Exchange rate Exchange Rate from Local currency to USD World Bank

Development Indicators

Crisis Dummy variable equal to 1 if the year 2007 or

2008 and 0 if not. Authors construction

Table 2. Summary statistics of financial fundamentals of multinational parent banks

and foreign subsidiaries

Subsidiaries Parent banks

All Subsidiaries

Developing Developed

Mean Std. Dev.

Mean Std. Dev. Mean Std. Dev. Mean Std. Dev.

CAR 12.423 3.739

13.847 1.782 11.753 4.742 15.847 1.487

Size 9.853 1.853

11.901 1.843 9.042 1.923 12.853 2.361

Deposit 0.654 13.01

0.536 7.019 0.742 17.572 0.526 2.803

Loan ratio 0.735 1.634

0.530 1.937 1.831 3.729 0.638 1.172

NIM 0.051 0.734

0.052 0.824 0.049 1.946 0.092 2.063

Lon_loss_Prov 12.963 1.830

11.903 1.642 14.610 2.053 17.121 1.452

ROA 0.263 2.371

0.372 2.492 0.118 2.578 0.413 1.659

15

Table 3. Capital Adequacy Ratio (CAR) estimation

Model (1) Model (2)

Model (3)

Subsidiary characteristics

Size -2.152

0.134 2.377

(1.445)

(1.750) (0.986)

Dep_Total_asset 0.021

0.625 -0.314

(0.103)

(0.183) (0.193)

Lon_loss_ Prov 0.826***

0.985*** 0.742*

(3.471)

(4.1983) (1.934)

Loan ratio_prov -0.624

0.376 0.456

(0.255)

(0.271) (0.145)

ROA 1.279

0.789 0.856

(0.587)

(0.845) (0.273)

Dep_Total_asset -0.578

0.457 0.954

(0.561)

(1.487) (1.961)

NIM 0.0 27

1.399 -0.273

(0.264)

(0.153) (0.843)

Host country variables

GPD growth 1.162***

2.298 1.324

(3.181)

(0.125) (0.894)

Interest rate - 0.874***

-3.978* 0.214*

(4.429)

(2.190) (2.104)

Exchange rate 0.267

0.295 0.964

( 0.821)

(0.538) (0.832)

Parent bank characteristics

Size

-3.742***

- 4.642***

(9.692) (11.034)

Dep_Total_asset

-0.388 0.286

(0.822) (0.528)

Lon_loss_ Prov

0.178* 0.247

(2.092) (0.217)

Loan ratio_prov

0.282 -0.964

(0.065) ( 0.959)

ROA

- 0.021 -0.779

(0.949) ( 0.347)

Dep_Total_asset

0.722 -1.386

(1.852 ) (1.275)

NIM

0.683* 0.247

( 2.248) (0.243)

Home Countryvariables

GDP Growth

- 1.247*** -2.842***

(3.842) ( 2.747)

Crisis

1.748***

(4.248)

Constante 12.279***

3.135*** 9.894***

(4.216)

(6.311) (2.389)

R-squared 0.357

0.368 0.247

Note. The table reports the fixed effects panel estimation results. The dependent variable is the change in capital

adequacy ratio of foreign subsidiaries (ΔCAR). Size is The logarithm of total assets of foreign subsidiaries and parent

banks. Deposits is the Total deposit to total assets of subsidiary or parent bank deposits Loans is The ratio of total loans

to total assets. Loan loss provision is the Ratio of loan loss provisions to net loans.ROA is the Return on Asset. NIM is

the Net Interest Margin.GDP is the Yearly change in GDP growth Real interest rate is the lending interest rate adjusted

for inflation as measured by the GDP deflator. Exchange rate is the Exchange Rate from Local currency to USD. Crisis is

a Dummy variable that equals 1 if the year 2007 or 2008 and 0 if not.. T-student statistics are reported between

parenthesis. ***,**, * denote significance at the 1%, 5% and 10% levels, respectively.

16

Table 4. The capital adequacy ratio with distinction between developed and developing countries

developing

developed

Model (1) Model (2) Model (3)

Model (1) Model (2) Model (3)

Subsidiary characteristics

Size -2.862** 1.145 -2.933*** 1.287 -3.375*** -3.876***

(1.983) (0.093) (3.234) (1.631) (2.450) (7.973)

Dep_Total_asset 0.242 -0.727 -0.838 0.982 0.347 1.463***

(0.644) (0.116) (0.426) (0.191) (0.215) (2.285)

Loan_loss_ Prov 1.734*** 1.380** 0.246*** 0.340 0.871 0.642

(11.286) (1.938) (7.921) (0.239) (0.826) (0.875)

Loan ratio -0.246 -0.237 -0.848 0.911 -0.842 -0.851

(0.796) (0.756) (0.74) (0.231) (0.309) (0.422)

ROA 3.784*** 1.197*** 3.984* 0.864 1.374 0.347

(8.765) (6.927) (1.834) (0.655) (0.235) (0.643)

Dep_Total_asset -1.315 -1.279 -1.474 -2.363 -12.153 6.678

(0.455) (1.034) (1.466) (1.268) (0.096) (0.245)

NIM -0.276 0.562 0.768 2.784 3.836 2.967

(0.257) (0.002) (0.1 38) (0.368) ( 0.724) (1.569)

Host country variables

GPD growth -2.679*** -1.231*** -4.741*** 0.853* 1.274 -3.582***

(4.376) (3.863) (11.573) (1.725) (0.967) (1.905)

Interest rate -3.724*** -1.325*** -2.262* 0.563 -0.356 -1.987

(5.281) (4.246) (1.764) (0.254) (0.425) (1.054)

Exchange rate 0.637 2.996 2.466 0.752 1.327 1.865

( 0.711) (1.017) (1.398) (0.141) (0.459) (1.245)

Parent bank characteristics

Size

2.246 4.268

-2. 974** -3.792*

(0.622) (1.124)

(2.374) (1.864)

Dep_Total_asset

- 0.735 0.853

0.624 0.727

(0.2833) (0.357)

(0.956) (0.289)

Lon_loss_ Prov

0.748** 0.482*

1.952 *** 1.630***

(2.574) (1.723)

(4.976) (3.249)

Loan ratio

1.131 0.246

-0.265 0.987

(0.724) (1.266)

(0.099) (0.644)

ROA

-0.207 0.817

-0.927*** -0.294***

(0.245) (0.294)

(3.294) ( 5.291)

Dep_Total_asset

0.762 5.874

1.127 9.736

(0.657) (0.867)

( 1.163) (4.752)

NIM

-0.225 0.865

2.756 1.259**

(0.566) (0.901)

( 1.365) (2.187)

Country of origine variables

GDP Growth

-0.732

-1.532 ***

( 0.365) ( 14.825)

Crisis

1.346 0.875***

(0.546) (4.638)

Regulation Index

-3.273** 2.786***

(2.134) (4.253)

Constant 4.629** 10.653** 5.869 9.687 8.946 17.959

(3.654) (6.267) (3.979) (6.828) (13.946) (6.546)

R-squared 0.476 0.343 0.392 0.481 0.274 0.397

Note. The table reports the fixed effects panel estimation results. The dependent variable is the change in capital adequacy ratio

of foreign subsidiaries (ΔCAR). Size is The logarithm of total assets of foreign subsidiaries and parent banks. Deposits is the

total deposit to total assets of subsidiary or parent bank deposits. Loans is The ratio of total loans to total assets. Loan loss

provision is the Ratio of loan loss provisions to net loans.ROA is the Return on Asset. NIM is the Net Interest Margin.GDP is

the Yearly change in GDP growth. Real interest rate is the lending interest rate adjusted for inflation as measured by the GDP

deflator. Exchange rate is the Exchange Rate from Local currency to USD Regulation Indexes the Capital Regulation Index of

home country as an additional explanatory variable. This index was developed by the World Bank surveys on bank regulation

(2003, 2007 and 2011). Crisis is a Dummy variable that equals 1 if the year 2007 or 2008 and 0 if not.. T-student statistics are

reported between parenthesis. ***,**, * denote significance at the 1%, 5% and 10% levels, respectively.

17

Table 5. Capital adequacy ratio and interbank dependency

Interbank<1

Interbank>1

Model (1) Model (2) Model (3)

Model (1) Model (2) Model (3)

Subsidiary characteristics

Size -1.836** -5.137** -4.931*** -4.247 -5.147 -0.861

(1.996) (2.033) (5.979) (0.934) (1.272) (1.346)

Dep_Total_asset 0.456 -0.289 -0.124 0.274 0.275 0.829

(0.153) (0.236) (0.9967) (0.941) (0.753) (0.975)

Lon_loss_ Prov 2.875*** 1.794** 4.385*** 0.975 -0.654 0.086

(3.790) (2.312) (6.876) (0.817) (0.138) (0.722)

Loan ratio -0.247 0.728* 0.497* 0.846 0.842 0.844

(0.982) (0.976) (0.147) (0.156) (0.145) (0.566)

ROA -0.866***

-

0.371*** -0.214*** -4.431*** -3.273*** -3.835***

(3.266) (6.583) (12.1 48) (4.325) ( 11.134) (13.644)

Dep_Total_asset -1.567 -1.204 -1.403 -2.234 -12.272 6.754

(0.275) (0.848) (0.375) (0.865) (0.737) (0.745)

NIM 0.967 0.649 0.257 0.552 1.754 -0.725

(0.756) (0.780) (0.870) (0.876) (0.276) (0.685)

Host country variables

GPD growth 0.623

-

1.387*** -0.972* 0.194 0.943 -0.558

(1.305) (2.476) (1.678) (0.537) (0.125) (0.937)

Interest rate 0.347 -0.969 -0.0578 0.496 0.976 0.134

(0.958) (0.365) (0.266) (0.032) (0.004) (1.067)

Exchange rate 0.724 2.024 2.772 0.624 1.321 1.688

( 0.767) (1.247) (1.127) (0.782) (0.024) (0.004)

Parent bank characteristics

Size

-2.281 4.634

2. 764*** 3.876***

(0.674) (0.268)

(4.876) (2.985)

Dep_Total_asset

- 0.357 0.256

0.275 0.377

(0.825) (0.724)

(0.024) (0.264)

Lon_loss_ Prov

0.981 -0.278

3.234*** 7.274**

(0.361) (0.482)

(2.731) (2.347)

Loan ratio

0.452 0.636

-0.258 0.255

(0.356) (0.737)

(0.266) (0.169)

ROA

-0.883 0.675

-0.573 0.767

(0.437) (0.356)

(0.357) ( 0.424)

Dep_Total_asset

0.7367 7.523

-1.536 -2.357

(1.567) (0.677)

( 1.357) (0.863)

NIM

-0.289 0.458

1.095*** 1.689***

(0.502) (0.493)

( 6.244) (5.567)

Country of origin variables

GDP Growth

0.478

0.257

(0.838)

(0.005)

Crisis

0.376

0.536

(0.257)

( 0.003)

Constant 6.836 9.964 11.536 9.857 14.477 11.834

(6.456) (4.674) (3.838) (6.674) (7.653) (6.737)

R-squared 0.3765 0.474 0.254 0.456 0.256 0.367

Note. The table reports the fixed effects panel estimation results. The dependent variable is the change in capital adequacy

ratio of foreign subsidiaries (ΔCAR). Size is The logarithm of total assets of foreign subsidiaries and parent banks. Deposits is

the Total deposit to total assets of subsidiary or parent bank deposits Loans is The ratio of total loans to total assets Loan loss

provision is the Ratio of loan loss provisions to net loans.ROA is the Return on Asset. NIM is the Net Interest Margin.GDP is

the Yearly change in GDP growth Real interest rate is the lending interest rate adjusted for inflation as measured by the GDP

deflator. Exchange rate is the Exchange Rate from Local currency to USD. Crisis is a Dummy variable that equals 1 if the

year 2007 or 2008 and 0 if not.. T-student statistics are reported between parenthesis. ***,**, * denote significance at the 1%,

5% and 10% levels, respectively.