Regulatory Arbitrage and Presence of Foreign Banks: Evidence from the Indian Banking Sector

Abstract

Presence of a heterogeneous banking system across countries provides opportunities for cross-border banks to indulge in activities of regulatory arbitrage. This article attempts to investigate whether regulatory arbitrage induces the presence of foreign banks in India. Using relevant country-level data on various aspects of banking regulations, we conduct a series of panel regressions to examine the effect of cross-country gap in banking regulations on foreign banks’ presence in India. We find regulatory arbitrage as significantly determining foreign banks’ presence in India, after controlling for other factors (income level of home country, bilateral economic relationship, colonial and linguistic commonality, and geographic proximity).

Keywords

Regulatory arbitrage foreign banks Indian banking

Introduction

Banking regulations play an important role in achieving financial sector policy objectives. As the rationale shaping the banking regulations differ across countries, regulatory systems around the globe become heterogeneous. Compliance to regulations is costly. The presence of heterogeneous regulatory system across countries provides opportunities to the cross-border banks to minimize their regulatory cost, by taking advantage of the regulation gap. Practices through which cross-border banks can take advantage of cross-country regulatory differences to reduce their overall regulatory cost are known as regulatory arbitrage activities (Cárdenas et al., 2003). Regulatory arbitrage activities include strategies like introduction of innovative financial products that are partially controlled by the existing regulations or relocation of affiliates in jurisdictions with less stringent regulatory framework.

Global integration of banking markets facilitates expansion of banks beyond their national boundaries. These markets widen and deepen with technological advancement and liberalization of banking sectors that previously operated under restrictive regulatory environment. Moreover, the operational freedom secured by the liberalization process also encourages entry of new players looking for growth opportunities. According to Herrero and Martinez Pería (2007), there are two channels through which cross-border banks function in the global banking market—first, by facilitating international capital flows from the home country and, second, by establishing their presence in a foreign country by setting up affiliates. Scholars have tried to theorize the factors driving foreign banks’ choice of adopting either one or both channels. Theoretical underpinnings in this regard comprise application of general foreign direct investment (FDI) theories to explain foreign investments in host countries’ banking sector (Battilossi, 2006; Goldberg, 2007; Herrero & Simón, 2003).¹ Among these, eclectic theory (formulated by Dunning, 1977), with its three dimensions, viz. ownership advantage, location advantage, and internalization advantage, has been extensively applied to explain the motivations behind banking sector FDI.²

Although the aforementioned advantages have universal applicability, literature exploring the three perspectives finds that their significance is specific to country, region, and bank. Theoretical and empirical studies highlight several determinants of foreign investment in the banking sector. These factors can be broadly categorized into two advantages, viz. client information and profitability. Information advantage is related to bank–client relationship translating into follow your client strategy. Profitability include opportunities arising from market size, geopolitical relations, and institutional–cultural–lingual commonalities between countries. In addition, the presence of arbitrage opportunity between markets by way of differences in regulations also impacts bank’s profitability in foreign markets (Battilossi, 2006; Claessens & Van Horen, 2013; Horen, 2007; Houston et al., 2012).

In the case of cross-country variations in banking regulations, literature is conclusive on the fact that banks generate considerable amount of their income by managing cross-country gaps in regulatory taxes, capital requirements, deposit insurance premiums, and other prudential requirements (Cumming, 1987; Dong et al., 2011; Houston et al., 2012; Jones, 2000; Pavel & Phillis, 1987). In this context, from the perspective of the banks, it is argued that in certain circumstances, regulatory arbitrage is important for a sustainable banking business. According to Pavel and Phillis (1987) and Jones (2000), regulatory arbitrage techniques enable banks to enhance their liquidity conditions, provide opportunities of risk diversification, and help them to survive the competition in financial service industry, thus indicating that regulatory arbitrage may be an important factor in encouraging banks to establish their presence in different countries.

In general, regulatory arbitrage evolves in response to differences in accounting standards, and it also is due to differences in economies of scale and scope between bank affiliates operating in different countries. These differences may encourage banks with headquarters in heavily regulated countries to position their affiliates in host countries with loosely regulated markets. In addition to the above, regulatory systems that permit use of innovative financial products also enable large, sophisticated banks to erode the regulatory cost (Berger & Udell, 1996; Cumming, 1987; Dong, 2011; Jones, 2000). The most prevalent types of financial innovations comprise risk repackaging techniques like securitization and credit derivatives.³ According to Cumming (1987) and Jones (2000), such risk repackaging techniques are driven by large divergences in notions and measures embodied in the regulatory standards. This implies that arbitrage activities may find its way in all areas of banking regulations (such as capital requirements, supervisory standards, entry and participation norms, information disclosures, etc.), wherever regulatory systems present opportunities to benefit from the divergence.

Several case studies have explored different aspects of regulatory arbitrage in the banking sector. Cumming (1987) and Pavel and Phillis (1987) examined the regulatory arbitrage achieved by securitization methods. Jones (2000) illustrates arbitrage techniques used by banks to erode risk-based capital requirements. Similarly, Jones and Kolatch (1999) and Carbo-Valverde et al. (2009) explain arbitrage opportunities created due to divergence in deposit insurance policies in different jurisdictions. Overall, findings of these studies are consistent with the view that cross-border banks benefit from arbitrage activities.

Among the theoretical and statistical studies, Morrison and White (2009) theoretically analyzed the costs and benefits of coordinated multinational banking regulation against the costs and benefits of diverse multinational regulations. They found that cross-border banks take advantage of regulatory gaps across countries, and that bank capital tends to flow from heavily regulated markets to less regulated markets. Motivated by the same issue, recent studies by Dong et al. (2011) and Houston et al. (2012) empirically test the relation between multinational banks’ cross-border expansion and cross-country gap in banking regulations. Dong et al. (2011) used data of 2,000 international bank mergers and acquisitions (M&A) between 1990 and 2007 to analyze the effects of banking regulations on banks’ cross-border M&A. Houston et al. (2012) investigated the relation of regulation differences between home and host countries with bank capital flows for a panel of 26 home countries and 120 host countries. Results of both these studies support the theoretical claims made by Morrison and White (2009) and show that banks have strong incentives to expand into countries with diverse regulations, in order to gain from the regulatory gap between countries.

In the 1990s, several emerging economies, including India, liberalized their banking sectors. A study by Cull and Martínez Pería (2010) finds that between 1996 and 2005, foreign banks’ presence has increased across all developing countries. The study reported that in countries of sub-Saharan Africa and Eastern Europe, more than half of the banks in the system were foreign owned. In Latin American countries, on an average, the share of foreign banks to total number of banks has risen from 27 percent in 1996 to 42 percent in 2005. In the Middle East, this share rose from 14 percent to 24 percent between 1996 and 2005. In East Asian countries, the percentage of foreign banks rose from 12 percent to 19 percent. In South Asian countries, this percentage increased from 6 percent to 8 percent. Furthermore, the study reported that the share of assets held by foreign banks across developing countries had increased from 22 percent in 1996 to 39 percent in 2005. At the same time, foreign bank claims on developing countries (including both loans extended by foreign bank branches and subsidiaries in the host countries and cross-border loans from the home countries) increased from 10 percent of gross domestic product (GDP) in 1996 to 26 percent in 2008. Similarly, according to the recently compiled data by Claessens and Van Horen (2013), foreign banks’ average loan, deposit, and profit shares for the year 2007 was close to 45 percent in emerging markets and about 50 percent in developing countries.

In the Indian context, a series of agreements with the World Trade Organization (WTO 1994, 1997) and revisions of regulatory guidelines by the Reserve Bank of India (RBI, 2005, 2011, 2013a) played a central role in increasing foreign participation in the Indian banking system. Since then, the number of foreign banks in India has increased to 45 banks with 332 offices in 2016 as compared to 24 foreign banks with 156 offices in 1994.⁴ Majority of these foreign banks operate as suppliers of full range of retail banking services. As shown in Figure 1, the number of foreign banks in India after experiencing a decline until 2008 saw a sharp increase in the following years. The number of foreign banks in India increased from 30 in 2009 to 45 in 2016. Due to this rise, the gap between the number of foreign and domestic banks has narrowed in the recent years. In comparison to foreign banks, domestic banks with 48 banks and 117,071 offices continue to hold a dominant market share of the Indian banking sector. Nevertheless, a clear convergence in terms of number of banks is visible between foreign and domestic banks in India. This observation provides a tentative hint to the rising presence of foreign banks in India in the post-liberalization period.

Figure 1.

Source: Statistical tables relating to banks in India, RBI, various issues.

In this backdrop, it is interesting to examine whether cross-country regulation differences explain renewed interest of foreign banks to operate in the Indian banking market. This article attempts to empirically examine whether cross-border regulatory differences induce the presence of foreign banks in India. In this attempt, we analyze how varying bank regulatory stringency across jurisdictions impact (a) foreign banks’ decision to enter and (b) profitably operate in the Indian banking sector. With this twofold classification of foreign banks’ presence, our empirical methodology estimates two sets of panel regression. First, using a logit framework we examine factors impacting foreign banks’ decision to enter the Indian banking sector. As economic integration between countries facilitate the ease of conducting international business and vice versa (Carpenter & Dunung, 2011; Plummer, 2009), we have selected 37 countries that are economically linked with India through trade and FDI/Foreign Institutional Investment (FII) to examine whether regulation disparity impacts on entry decision. For the second set of panel regressions, a subsample of 23 countries is retained to examine the effects of regulatory arbitrage on participation of foreign banks in India. This subsample is selected by identifying home countries of foreign banks that were operating in India during our study periods. The regressions are carried out using country-level data for the years 2000, 2001, 2005, and 2010. Selection of the time period is based on four Banking Regulation and Supervisory Survey conducted by World Bank (2001, 2003, 2007, 2011). The survey compiles detailed information on various aspects, comprising administrative, industrial, and supervisory structure in which the banks operate.

Going by the larger intuition derived from the literature that possibilities of regulatory arbitrage may arise in different areas of banking regulations (Claessens & Van Horen, 2013; Dong, 2011; Jones, 2000; Morrison &White, 2009), we compiled a wide range of data on various aspects of banking regulations for all countries included in the study. On the basis of this information, we constructed a banking regulation gap index (BRGI), which is a comprehensive measure of banking regulation gap between banking regulations in foreign banks’ home country and India. BRGI is our main explanatory variable; it captures the difference in stringency of banking regulations across nations that provides cross-border banks the opportunity to reduce the regulatory cost and induce expansion in the Indian banking sector.

Apart from regulation arbitrage, we also considered other socioeconomic and geopolitical factors that may provide alternative explanations for foreign banks’ presence in India. Following the literature (Barth et al. 2006; Battilossi, 2006; Herrero & Martinez Pería, 2007; Horen, 2007; Houston et al., 2012), we included a range of variables such as bilateral factors to capture the level of economic integration (such as trade, FDI, FII), governance gap index—GGI (which is a composite measure of differences in institutional capacity between India and home countries), and the colonial and linguistic commonality and geographic proximity to examine their effect on foreign banks presence. To account for the development level, we also included the income level of foreign banks’ home country. As India is a developing country with a colonial past and strong economic relations with the rest of the world, bilateral factors capturing economic integration as well as income level of home country are expected to have a positive relation with the entry of foreign banks in India.

After controlling for the impact of other factors, it was noted that foreign banks’ decision to enter the Indian banking sector could not be explained by the difference in the Indian and home country’s banking regulations. In this case, factors like bilateral FDI, colonial similarities, common language, and geographical proximities are found to be significantly affecting foreign banks’ entry into India. However, as far as post-entry presence of existing foreign banks is concerned, the bank regulation gap is found to be significant and positive, ceteris paribus. Consolidating the two results, one can say that when we consider the sample of countries with economic ties with India, regulatory arbitrage does not play out as a significant contributor for banks’ decision to enter the Indian banking sector. However, when we consider the set of countries that have already established foreign banks in India, regulatory arbitrage seems to be significant and an important determinant of foreign banks’ expansion in India. Banks from those countries having more stringent regulation seem to be the one who have a larger presence in India compared to those countries with relatively less stringent regulation. Therefore, we conclude that regulatory arbitrage is important for foreign banks’ expansion in India.

The remainder of the article is set as follows. The second section briefly describes foreign bank regulations in India. The third section discusses the data sources and the empirical model employed for investigation of the research question. The fourth section presents exploratory analysis on the linkages between regulatory arbitrage and foreign banks’ presence in India. Finally, the fifth section concludes the article with results and implications.

Foreign Bank Regulations in India

In India, foreign banks are defined as foreign owned and controlled companies carrying banking business in India. In terms of ownership, this definition refers to banking companies in India where more than 50 percent of capital is owned by nonresidents. With respect to control, it refers to the nonresident shareholders’ rights in management and decision-making of a banking company in India.

As part of its phased liberalization of the banking sector, the Indian government undertook WTO (GATS) agreement in 1994 to issue five branch licenses to existing and new foreign banks every year. This limit was later increased to 12 branch licenses a year with the signing of another WTO agreement in 1997. Thereafter, in 2005, the RBI released a Road Map for further liberalizing the entry of foreign banks in the Indian banking system. However, this road map was put on hold on account of the unprecedented global financial crisis of 2007. RBI (2011) laid out a detailed framework for foreign banks’ entry into India. In a more recent discussion paper, RBI has finalized a framework for wholly owned subsidiary (WOS) mode of entry (RBI, 2013a).⁵ Currently, all foreign banks in India are in the form of branches. Thus, further formalization of entry regulations of foreign banks’ WOS displays clear preference of RBI for the WOS mode. Within this policy framework, WOS of foreign banks are given near national treatment wherein they can establish presence in Tier 1–Tier 6 centers of India without prior approval from the RBI, except for certain sensitive areas where entry is subject to reporting the location. However, branches of foreign banks are permitted to establish presence only in Tier 1 and Tier 2 centers of India subject to prior approval from RBI and fulfilment of other regulatory guidelines specific to foreign banks.⁶

The entry guidelines as formulated by the RBI requires foreign banks to comply with regulations related to minimum paid-up capital requirement, capital adequacy ratio, priority sector lending, and other prudential norms. The initial minimum paid-up capital requirement for foreign bank branches is ₹25 million and ₹5,000 million for WOS of foreign banks. Further, WOSs are required to maintain minimum capital adequacy ratio of 10 percent of risk-weighted asset, while, for the branches, the ratio is 9 percent. Foreign banks like domestic banks are required to extend priority sector lending in certain percentage of their net bank credit. This percentage stands at 32 percent of net bank credit for foreign bank branches and 40 percent of adjusted net bank credit for WOS (RBI, 2011, 2013a).

In addition to the above, foreign banks that pass the scrutiny of international supervisory criteria can choose to establish their presence either as branches or WOS. However, banks that do not satisfy these criteria are restricted to enter via branch mode, and for such banks, WOS mode remains the only option to enter the Indian banking sector.

Policies formulated by the RBI also regulate the capital market activities of foreign banks in India. Under the guidelines prescribing foreign banks’ access to rupee resources in India, WOS of foreign banks are permitted to raise capital in India through the issue of non-equity capital instruments. However, branches of foreign banks are prohibited access to domestic capital markets. The deposit insurance policy of India also covers foreign banks functioning in India. Regulations on protection of bank deposits payable in India are formulated by Deposit Insurance and Credit Guarantee Corporation (DICGC, GoI). As per these regulations, foreign banks are required to be registered by DICGC and pay deposit insurance premiums.

Empirical Methodology and Data

This section discusses the empirical methodology and data sources of this study. Primarily, the study investigates whether differences in bank regulations influence foreign banks’ presence in the Indian banking sector, after controlling for other factors (e.g., income level of home country, bilateral economic relationship, socioeconomic commonality, geographic proximities, and quality of governance between India and each country included in the analysis). The empirical methodology involves estimating two sets of panel regressions. First, a binary variable regression examines the significance of cross-country difference in the banking regulations on foreign banks’ entry decision in the Indian banking sector. This regression is carried out using a sample of 37 countries with which India has some economic relationship in terms of trade or FDI. The question being addressed is that given a set of 37 countries that have trade relations with India, what induces some to set up banks in India. The second set of regressions examines the effect of regulatory arbitrage on foreign banks’ participation in the Indian banking sector. Here, our dependent variable captures the presence of 63 foreign banks headquartered in 23 countries in terms of their participation in the Indian banking sector. We use two indicators to measure the participation of these foreign banks in India, after aggregating them as per their home country—the asset share of foreign banks from jth home country in total assets of foreign banks in India and office share of foreign banks from jth home country in total offices of foreign banks in India.⁷ Data on 37 countries having bilateral economic relations with India were retrieved from Handbook of Statistics on Indian Economy made available by the RBI. Thereafter, identification of foreign banks’ home country is carried out by the author from the websites of foreign banks. Data on total number of foreign bank offices and total assets of foreign banks in India were obtained from Profile of Banks annually published by the RBI.

Further, to account for the joint effect of difference in various banking regulations between each country and India, we developed a BRGI. BRGI was constructed using the methodology of principal component analysis (PCA). Construction of this index involves estimation of gaps in legal procedures related to different aspects of banking operations practiced in India and each country. Thus, larger the gap, greater is the regulatory difference, and higher is the possibility of regulatory arbitrage for banks originating from the countries included in this study. In this framework, we have assembled information on nine aspects of banking regulation, viz. (a) compliance to legal submissions required to obtain a bank license, (b) statutory capital requirements, (c) liquidity ratio, (d) share of single ownership, (e) limits on banks’ capital market participation, (f) strength of external auditing, (g) presence of an explicit depositor insurance scheme, (h) provisioning for nonperforming assets (NPAs), and (i) information disclosure of risk management procedures required to be taken up by the banks in respective countries. Data on seven of these indicators have been retrieved from Survey on Bank Regulation, Supervision, and Monitoring made available by the World Bank (2001, 2003, 2007, 2011). Data on two indicators, namely statutory capital requirements and liquidity ratio were collected by the author from the websites of the central banks of each country.

The panel regression equations that were estimated are given as follows:

Our first premise that foreign banks’ decision to enter the Indian banking sector is determined by the variations in banking regulations in India and in their home jurisdiction, ceteris paribus, is examined in regression I.

Regression I

E N T R Y_{j, t} = β_{0} + β_{1} B R G I_{j, t} + β_{2} Z_{j, t} + ϵ_{j, t}

(1)

where,

$E N T R Y_{j, t}$ is a binary outcome variable, capturing the decision of foreign banks of jth home country to enter the Indian banking sector at time t. ${BRGI}_{j, t}$ is the BRGI, $Z_{j, t}$ is a vector of other control variables and ϵ_j,t is the error term for j = 1, …, , n panel of countries at time period t = 1, …, n.

In regression I, we estimate foreign banks’ entry choice probability by fitting a logit model to our panel data. Our dependent variable—ENTRY—is a binary choice variable, which takes value one if jth country is home to one or more than one foreign bank with their presence in India, and zero otherwise. The choice probabilities are defined as a function of ${BRGI}_{j, t}$ . and a vector of control variables ( $Z_{j, t}$ ), influencing the entry decision of foreign banks in India. We write the choice probability as:

Probability of foreign banks from jth country choosing to enter India at time t:

Pr ({ENTRY}_{j, t} = 1) = F ({BRGI}_{j, t}, Z_{j, t})

The logit model takes $F ({BRGI}_{j, t}, Z_{j, t})$ as a logistic cumulative distribution function. In this case, the probability of foreign banks originating from jth home country choosing to enter India, given the banking regulation gap and other control variables becomes:

The explanatory variables will be discussed later.

Regressions II and III examine our second premise that foreign banks’ expansion in the Indian banking sector is determined by cross-country differences in banking regulations, ceteris paribus.

Regression II

ln (1 + s . {Offices}_{j, t}) = β_{0} + β_{1} {BRGI}_{j, t} + β_{2} Z_{j, t} + ϵ_{j, t}

(2)

Regression III

ln {(1 + s . Assets)}_{j, t} = β_{0} + β_{1} {BRGI}_{j, t} + β_{2} Z_{j, t} + ϵ_{j, t}

where

$ln (1 + s . {Offices}_{j, t})$ is logarithmic of one plus share of offices of foreign banks originating from jth home country in total offices of foreign banks in India at time t. $ln {(1 + s . Assets)}_{j, t}$ is logarithmic of one plus share of assets of foreign banks’ affiliating from jth home country in total assets of foreign banks in India at time t. Thus, the dependent variables in Equations (2) and (3) are country-level variables, estimating the share of offices and assets of foreign banks in India grouped as per their country of origin at time t.

Explanatory variables are discussed as follows:

${BRGI}_{j, t}$ is our main explanatory variable. It measures differences in stringency of banking regulations that frame entry and operation of banks in jth home country and in India at time t. This index measures the difference in banking regulatory environment between each country and India. If the cross-country gap in banking regulations influence foreign banks’ decision to establish presence in India, then $β_{1}$ . should be positive, thus indicating that regulatory arbitrage is a significant factor inducing the presence of foreign banks in India. Construction of BRGI is discussed in more detail in the fourth section.

$Z_{j, t}$ . is a vector of control variables included in the regressions to control for macroeconomic factors that may affect decision of foreign banks originating from jth country to establish presence in India at time t. The reason to include the control variables is to exclude alternative explanations while testing our hypothesis of regulatory arbitrage, inducing presence of foreign banks in India. The control variables included in the study are as follows:

is another index measuring the gap in quality of governance between jth home country and India at time t. The index is included as a control variable to capture cross-country differences in governance standards or institutional facilities that may influence the decision of foreign banks to establish their presence in India. The index is construed using the methodology of PCA. GGI is based on four indicators, reflecting the gap in different aspects of institutional environment existing in each country and India. Data on these indicators have been retrieved from Kaufmann and Mastruzzi (2010), Survey of Worldwide Governance Indicators published by the World Bank. (Following section of the article presents a detailed description on construction of GGI.)

$ln {(PCI)}_{j, t} :$ The logarithmic of per capita GDP of jth home country of foreign banks with presence in India at time t. This variable controls for the level of economic development in the countries. Data on GDP per capita are extracted from World Development Indicators database published by World Bank.

$ln {(TOTAL TRADE)}_{j, t}$ , $ln {(FDI)}_{j, t}$ and $ln {(FII)}_{j, t}$ . are logarithmic of bilateral total trade, bilateral FDI, and bilateral FII flows from jth home country to India at time t. As conventional in the literature, bilateral total trade is defined as the sum of the value of imports and the value of exports of goods and services of each country with India. The variable FDI denotes the value of net inflow of FDI from each country to India, and FII denotes the value of net inflow of portfolio investment from each country in India. Overall, these three variables examine the level of economic relationship in terms of trade openness and financial market integration between India and the countries included in the study. In other words, these variables aim to control for the impact of cross-country economic linkages on foreign banks’ presence in India. Data on the variables of economic integration are collected form the following sources: (a) World Integrated Trade Solutio(WITS) database is used to retrieve data on bilateral exports and imports from source countries to India. (b) Data on bilateral FDI flows are extracted from SIA News Letter made available by Department of Policy and Planning (DIPP), Government of India (GoI), and (c) data on bilateral FII flows are collected from Coordinated Portfolio Investment Survey (CPIS) published by International Monetary Fund (IMF).

${COMMONALITY}_{j, t}$ . is a dummy variable estimated using a group of commonality variables, viz. common borders, common language, and common colonial past. The variable takes a value of 1 if a home country shares same language, border, or colonial past with India, and 0 otherwise. These variables are selected on theoretical grounds, mainly to account for geographic proximities and cultural and linguistic commonalities between India and foreign banks’ home countries. Information on these variables is collected from Central Intelligence Agency (CIA) The World Factbook (2010).

Conceptualizing Regulatory and Governance Quality Differentials

The following discussion illustrates the indicators used for constructing BI and governance quality gap index, and the rationale for including them in constructing the indices.

Banking Regulation Gap Index

As mentioned earlier, BRGI is constructed on the basis of nine indicators of banking regulations. Table 1 provides information on the nine aspects of bank regulations considered in the study. It also presents the survey questions included in estimation of the indicators along with their data sources. As shown in Table 1, of the nine indicators, seven indicators are qualitative in nature (Legal Submissions, Ownership, Capital Market Participation, External Audit Requirements, Depositor Protection Scheme, Provisioning, and Information Disclosure) and are constructed by summing up the responses of several survey questions related to the particular aspect, while two indicators (Statutory Capital and Liquidity Ratio) are quantitative, data on which are collected by the author.

Table 1.

List of Indicators, Survey Questions, and Data Source

No.	Regulatory Aspects	Survey Questions	Data Source	Indicator Type
1	Legal submissions	Which of the following are legally required to be submitted before issuance of the banking license?	World Bank (2001, 2003, 2007, 2011)	Qualitative
		a. Draft by-laws?b. Intended organization chart?c. Financial projections for first 3 years?d. Financial information on main potential shareholders?
		e. Background/experience of future directors?f. Background/experience of future managers?g. Sources of funds to be disbursed in the capitalization of new bank?h. Market differentiation intended for the new bank?
2	Statutory capital
	Minimum start-up capital	What is the minimum capital entry requirement?		Quantitative
	(USD mill)
	Capital adequacy(%)	What is the minimum capital–asset ratio requirement?	Collected from websites of central banks of countries considered in the study
3	Liquidity ratio (%)	What is the minimum liquidity requirement?	Collected from websites of central banks of countries considered in the study	Quantitative
4	Ownership	What is the percentage of maximum capital that can be owned by single owner?	World Bank (2001, 2003, 2007, 2011)	Qualitative
5	Capital market participation	1. Can banks engage in securities activities?	World Bank (2001, 2003, 2007, 2011)	Qualitative
		2. Can banks engage in insurance activities?
		3. Can banks engage in real estate activities?
		4. Can banks engage in owning Nonfinancial Firms activities?(4 aspects of market participation restrictiveness- Securities, Insurance, Real estate, and bank ownership of nonfinancial firms captured by 3 = unrestricted, 2 = permitted, 1 = restricted, 0 = prohibited)
6	External audit requirements	1. Is an external audit compulsory?2. Are auditors licensed or certified?	World Bank (2001, 2003, 2007, 2011)	Qualitative
		3. Is auditor’s report given to supervisory agency?4. Can supervisors meet external auditors to discuss report without bank approval?5. Are auditors legally required to report misconduct by managers/directors to supervisory agency?6. Can legal action against external auditors be taken by supervisor for negligence?7. Can supervisors force banks to change internal organizational structure?
7 8 9	Depositor protection scheme Provisioning Information disclosure	Is there an explicit deposit insurance protection system for commercial banks? Is there a formal definition of nonperforming loan? 1. Must banks disclose risk management procedures to public?2. Do regulations require credit ratings for commercial banks?	World Bank (2001, 2003, 2007, 2011) World Bank (2001, 2003, 2007, 2011) World Bank (2001, 2003, 2007, 2011)	Qualitative Qualitative Qualitative

Source. The author.

Banking regulation indicators are discussed as follows:

Legal submissions required for obtaining bank license: Our first indicator—legal submissions—examines the administrative arrangements required to be fulfilled by an applicant bank for procurement of a bank license. As illustrated in Table 1, this indicator is constructed by summing the responses to eight questions related to bank’s operational structure, business strategy, credit policies, hierarchy of decision-making bodies, and financial strength of the bank. The responses reported in yes/no format are quantified by assigning value 1 to yes and 0 to no. Accordingly, the indicator ranges between 0 and 8, where higher value indicates greater stringency in the procedure of issuing bank license in a country.

Statutory capital requirements: There are two indicators to measure regulations on statutory capital requirements—minimum start-up capital (expressed in US dollars, million) and capital adequacy ratio (in percentage). Regulators impose statutory capital requirements to ensure that banks are able to absorb the losses and retain stability in adverse conditions. Minimum start-up capital is the amount of paid-up capital required to be put down by an applicant bank at the time of entry in a country. It shows the preliminary cost of setting up the banking business in a particular jurisdiction. Thus, a higher initial capital requirement is indicative of higher cost of establishing banking business in a country. Across countries, the banking sector policies pursued by the regulator get reflected in the differentiated capital requirements imposed on domestic banks and foreign banks. For instance, countries that encourage foreign participation in their banking sector place foreign banks at par with the domestic banks and impose uniform initial capital requirements on both. However, countries that try to prevent dominance of foreign banks impose higher initial capital requirement on foreign banks. To incorporate this regulatory difference between home and host countries, we compare initial capital requirement imposed on domestic banks in each country with initial capital requirements on foreign banks in India. If the minimum start-up capital for banks in a country is greater than that imposed in India, then it would be less expensive for banks to establish business in India. Thus, due to this difference, foreign banks would prefer to open offices in India rather than expanding in their own jurisdiction.

The second indicator of statutory capital requirements is capital adequacy ratio. This indicator is defined as minimum capital to risk-weighted asset ratio (in percentage), as set by the supervisory authority of a country. The total risk-weighted assets take into account credit risk, market risk, and operational risk. A high capital adequacy ratio restricts banks to take excessive risk and ensures that banks hold sufficient reserves to absorb potential losses. Thus, in terms of management of banks’ risks, complying with higher capital adequacy is indicative of stringent regulatory system in a country.

Liquidity ratio (in percentage): Liquidity ratio measures statutorily enforced reserve ratio in a country. Definition of liquidity ratio varies across countries, as some may define it as ratio of short-term assets to short-term liabilities, while others may impose it as the ratio of short-term assets to total assets. Broadly, it indicates the proportion of cash or other liquid assets that banks in a country are required to hold in their balance sheet or with the central bank. The purpose of maintaining this ratio is to ensure that a bank has adequate balance to fund its operations under all circumstances. From the point of view of banks, compliance to high liquidity ratio implies fewer funds left for lending and investments, which ultimately affect their returns. In other words, it restricts the operational freedom of a bank in a country. Thus, higher percentage of liquidity ratio is indicative of stringent banking regulations in a country.

Share of single ownership: This is an indicator that measures the maximum percentage of bank’s equity that can be owned by a single owner. In other words, it shows the maximum permissible share of ownership allowed to be held by a single shareholder in a bank. The term single owner includes groups or families/individuals acting in a coordinated fashion. Usually, ownership of banks by large corporate groups and business families is discouraged by the regulators to minimize the risk of misuse of the leveraged funds and restrict the control of banks by a single owner. Thus, lower the percentage share of single ownership, more restrictive is the regulation.

Capital market participation: The indicator of restrictiveness on capital market participation examines banks’ permissible limits to participate in security activities (underwriting, dealing, and brokerage services and mutual funds), insurance activities ( selling all kinds of insurance and acting as an agent), real estate activities (investment, development, and management), and owning nonfinancial firms (asset finance company, loan company, investment/infrastructure company, and microfinance institution). The extent to which banks can engage in these activities is measured by assigning each activity a value from 0 to 3 on the basis of the following four criteria:

Unrestricted: If regulations in a country allow full range of these activities to be directly conducted by the banks, then it takes a value of 3.

Permitted: If banks can engage in full range of these activities, but some of these activities can be conducted only via affiliates (e.g. subsidiaries, or a common holding company or parent), then it takes a value of 2.

Restricted: If less than the full range of activities can be conducted by the banks or through its affiliates, then it takes a value of 1.

Prohibited: If banks are not permitted to engage in any aspect of the given activity, then it takes a value of 0.

Summation of the values assigned to each activity quantifies the overall restrictiveness on capital market participation by banks in a country. Hence, larger the overall value, less restrictive is the regulations on activities in which banks can engage.

Depositor insurance scheme: The indicator of depositor insurance to examine whether the regulator of a country has established an explicit deposit protection scheme, wherein, the rules regarding its coverage, funding, and procedure are well defined and enforced as a law. It reflects regulators’ policy toward protecting the savings of the depositors. This indicator ranges from 0 to 1, with 0 indicating absence of polices on protection to depositors. From the point of view of the banks, complying with deposit insurance policies implies submission of premiums on the basis of the total deposits collected by them. Thus, existence of deposit insurance schemes that require banks to pay higher premium and impose penalties on uninsured banks increases the stringency of banking regulations of a country.

Strength of external auditing: As presented in Table 1, the effectiveness of the external audit of the banks is developed by summing up the responses of the seven questions related to the autonomy of external auditors. It measures the supervisory power of the external auditors in functioning of banks in a country. The value of this indicator ranges from 0 to 7, where higher value of the indicator is indicative of better strength of the external audit.

Provisioning for nonperforming assets: This examines whether the regulations make banks accountable for the NPAs. Provisioning requires banks to report the quality of their assets (loans) by classifying them under certain categories (e.g., substandard, doubtful, and loss) and accordingly setting aside money for assets that do not generate income for the banks. We measure this indicator on a scale of 1 to 0, where the indicator takes a value of 1 if it is mandatory for banks to provision for NPAs, and 0 if banks are not required to provision for NPAs.

Information disclosure requirements: These comprise laws that require banks to publish their risk management procedures or annual financial statements, as well as regulations on submission of final accounts to the supervisor or credit rating agencies. Disclosure of information enforces transparency and accountability in the banking system. As presented in Table 1, this indicator is estimated by summing up the response on two questions reported in yes/no format, where yes is assigned a value of 1 and no is assigned a value of 0. Accordingly, the indicator ranges from 0 to 2, where higher value indicates that regulations of a country make it mandatory for banks to reveal relevant information about their functioning.

Overall, these nine indicators attempt to evaluate bank regulatory stringencies existing in each country included in our study. For foreign banks seeking entry and expansion in India, cost of complying with regulations varies due to divergence between banking regulations in their home country and India. This divergence in regulations (calculated as gaps between the values of the regulatory indicators of home country and India) is indicative of the arbitrage opportunity for banks originating from these countries. In measuring regulatory gaps, first, we compute the value of each indicator of regulatory stringency discussed earlier. Then, we compute gaps in regulatory stringency between a home country and India by simply taking the difference in value of each of these indicators for that country and for India. Finally, we combine these nine regulatory gaps into a single index by using the PC method. PCA methodology employed in construction of BRGI is provided in Appendix 1 of this article. We refer to the first principal component of gap between banking regulation indicators of each country with India as BRGI. The first principal components are estimated using a sample of 37 countries for all the years. Weights obtained in estimation of BRGI for 37 countries are retained for the subset of 23 countries. BRGI values for the sample of 23 countries are presented in Table 2.

Table 2.

Banking Regulation Gap Index (BRGI)

Countries	2000	2001	2005	2010
Australia	−16.4975	−10.5057	16.94103	26.57543
Bahrain	10.83041	10.28458	240.9844	240.9584
Bangladesh	−35.1875	−36.0551	−10.4223	−11.5142
Belgium	−35.4305	−37.3821	−20.7239	−20.4925
Bhutan	−41.3015	−40.7827	−20.3983	−20.7111
Canada	−35.8983	−39.3627	−20.3146	−19.9436
China	78.46064	−4.40652	13.52572	13.56015
Egypt	−17.1852	42.5243	67.03956	59.29555
France	−34.5931	−38.4871	−21.5738	−21.3904
Germany	−36.8262	−38.4871	−20.8939	−20.6721
Hong Kong	−3.94138	−4.38042	13.55557	13.5606
Indonesia	249.9644	306.8598	284.3384	317.0762
Italy	−35.3747	−37.3379	−20.6899	−23.5455
Japan	−25.9521	−25.6561	−5.62354	0.077969
Kenya	−39.8184	−39.6356	−21.38	−13.7356
Korea, South	35.10818	41.01288	65.74156	−22.2753
Kuwait	−32.627	−32.8387	254.0169	246.7692
Malaysia	−37.0979	483.3848	574.5532	73.04102
Maldives	−39.9585	−40.5607	−22.6293	−14.7074
Mauritius	−38.9251	−39.3483	−17.9709	−18.0329
Nepal	−35.7073	−29.7891	−10.6538	2.020841
Netherlands	−36.8262	−38.4871	−21.5738	−21.3903
Oman	9.651523	9.090199	105.0388	105.0398
Pakistan	−34.3351	−25.5902	3.43415	90.83613
Romania	85984.21	111402.6	146867	121341.9
Russia	−37.8254	−38.4853	−21.5736	5.648493
Saudi Arabia	−41.7424	−42.2387	−24.3048	−24.3138
Singapore	794.8221	818.0815	1035.178	1167.583
South Africa	–36.6013	−9.85913	5.288862	9.447893
Sri Lanka	−36.7642	−37.7517	−1.92267	2.134809
Sudan	−38.4896	−42.5498	−15.4306	−17.4993
Switzerland	−36.4836	−35.4813	−16.9733	−16.9823
Thailand	182.7114	137.8546	125.0888	138.979
UAE	−35.6017	−36.0987	−18.1647	−18.1741
UK	−35.2113	−34.7438	−15.7818	−16.9715
USA	−32.4091	−32.9064	−14.9725	−14.9815
Zambia	3280.875	2492.316	3178.713	21371.25

Source. The author.

In Table 2, positive values of the BRGI show that regulatory stringencies in home country are higher than that in India, indicating that India might be a favorable destination for foreign banks seeking to benefit from regulatory arbitrage. However, negative values of BRGI indicate that regulatory compulsions faced by foreign banks in India are more stringent than in home country. Thus, foreign banks may choose to expand business in their home country, in order to curtail the cost of regulatory compliance. Table 2 shows that for all the years, the value of BRGI for countries, such as Bangladesh, Belgium, Bhutan, Canada, France, Germany, Italy, Japan, Kenya, Mauritius, Netherland, the UK, and the USA is negative, showing that regulatory stringencies in these countries, is lower than that in India. For example, minimum start-up capital requirement for foreign banks in India (US$25 million) is much higher than in some of the countries such as Bangladesh (BDT 1 billion, approximately US$12.85 million), Japan (JPY 2 billion, approximatelyUS$16.11 million), the USA (US$10 million), Canada (CAD 5 million, approximately US$4.7 million), and countries in the European region (€5 million, approximately US$6.79 million). Similarly, capital adequacy ratio requirement in India (9%) is higher than that in most of the countries like Bhutan, Kenya, the USA, and the European countries (8%).

On the other hand, the value of BRGI for Bahrain, Indonesia, South Korea, Oman, Romania, etc., is positive for all the years, showing that regulatory stringencies in these countries have been higher than in India. For example, minimum start-up capital in Bahrain (BD 100 million, approximately US$265 million USD), Indonesia (Rp 300 billion, approximately US$250 million), Oman (OMR 50 million, approximately US$169.8 million), Thailand (฿5,000 million, approximatelyUS$146 million) is greater than that in India. Similarly, capital adequacy requirement in countries like Bahrain (12%), Russia (11%), Oman (12%), and Romania (12%) is higher than that in India. Moreover, in case of countries like Australia, Egypt, Hong Kong, Malaysia, etc., value of BRGI has been positive for some years and negative for others. This shows that the cross-country gap in regulatory stringencies keeps on changing as countries revise banking regulations in their jurisdiction from time to time.

Table 3.

Descriptive Statistics of BRGI, 23 countries

	BRGI-2000	BRGI-2001	BRGI-2005	BRGI-2010
Mean	2423.84	3104.72	4121.26	3915.90
Standard deviation	14129.80	18303.88	24125.39	20148.65
Median	−35.21	−32.91	−10.42	0.08

Source. The author.

We also present some descriptive statistics of BRGI for 23 countries in Table 3. Table 3 shows that there has been a rise in the mean value of BRGI, as well as in the median and standard deviation over the years. This indicates that there has been an overall increase in the regulatory gap between each home country and India. Also, increase in standard deviation during the years indicates that banking regulations between home countries and India seem to show high variations or large dissimilarities. Moreover, increase in median over the years (from negative values to positive value) shows that in comparison to banking regulations in home countries, regulations in India seem to have become less stringent over the years.

Governance Gap Index

We compute the GGI to evaluate the difference in quality of institutional environment between India and each country in our sample. This index is constructed by using four indicators: voice and accountability, political stability, government effectiveness, and control of corruption. According to the description provided by Kaufmann and Mastruzzi (2010), voice and accountability indicate the extent to which a country’s citizens can exercise their fundamental rights such as freedom of expression, participation in selecting their government, and existence of free media. The indicator ‘political stability’ captures the likelihood of the government being destabilized or overthrown by unconstitutional means like politically motivated violence and terrorism. A high political stability implies low possibility of government collapse due to unstable political environment. Government effectiveness measures the quality of public services and the degree of its independence from political pressure, the quality of policy formulation and implementation, and the credibility of the government’s commitment to such policies. Lastly, control of corruption indicates the extent to which public power is exercised for private gain, including both petty and grand forms of corruption. Kaufmann and Mastruzzi (2010) have computed these indicators by using the method of unobserved component model (UCM).⁸ Estimated values of the indicators obtained through this method vary from −2.5 to 2.5, where higher value indicates stronger governance performance.

In estimation of our GGI, we change the scale of the indicators by adding a value of 2.5 to each of the estimated value so that indicators range from 0 to 5. Then, we compute the gaps by comparing the new values of the indicators obtained for each country with those of India. Here, a large gap is indicative of greater differences in quality of governance between India and each country in our sample. To construct the index, we combine the values of these gaps by using the method of principal components. Appendix Tables 3 and 4 present results of PCA as explained in Appendix 1 of this article. We refer to the first principal component of gap between governance quality indicators of each country with India as GGI. The first principal component is estimated using weights obtained for our sample of 37 countries for all the years, and same weights are retained for the subset of 23 countries. Values of GGI for the sample of 23 countries are presented in Table 4.

Table 4.

Governance Gap Index (GGI)

Countries	2000	2001	2005	2010
Australia	3.901046	3.757017	3.599404	3.847052
Bahrain	0.616026	1.272732	0.547873	0.642357
Bangladesh	−0.68089	−0.97675	−1.83251	−1.04606
Belgium	3.517432	3.721213	3.165371	3.408599
Bhutan	1.026555	1.200227	1.16312	1.62647
Canada	4.097954	4.148266	3.597497	3.913497
China	−0.3338	−0.53854	−0.88587	−0.63343
Egypt	−0.10593	−0.32202	−0.80138	−0.75389
France	3.050925	3.155643	3.001757	3.137207
Germany	4.051353	3.902273	3.423107	3.475055
Hong Kong	2.465426	2.970523	3.173072	3.411556
Indonesia	−1.1962	−1.06424	−0.93424	−0.21555
Italy	2.312255	2.253268	1.619092	1.62192
Japan	2.750608	2.735063	2.8109	3.203179
Kenya	−1.06906	−1.06992	−1.07485	−0.72003
Korea, South	1.521963	1.861929	1.926147	1.939987
Kuwait	1.218993	1.201925	0.816173	0.98034
Malaysia	1.193605	1.367175	1.324051	1.199108
Maldives	0.812193	0.896472	0.317129	0.223235
Mauritius	1.923676	2.071505	1.875344	2.098737
Nepal	−0.65288	−0.8745	−1.76863	−1.07618
Netherlands	4.464782	4.258973	3.797143	3.940745
Oman	1.424403	1.507728	0.812534	1.11939
Pakistan	−1.28515	−1.30827	−1.60849	−1.93533
Romania	0.105012	0.809449	0.48583	0.911445
Russia	−1.01746	−0.49603	−1.04766	−0.90402
Saudi Arabia	−0.38444	−0.18646	−0.5606	−0.10167
Singapore	3.522376	3.60497	3.236797	3.520604
South Africa	1.495709	1.457387	1.413334	1.191776
Sri Lanka	−0.75846	0.1199	−0.53774	−0.20482
Sudan	−2.32385	−1.95215	−2.79777	−2.75158
Switzerland	4.255908	4.330181	3.874601	4.168099
Thailand	1.034899	1.063218	0.180021	−0.24251
UAE	1.244543	1.921889	1.49551	1.700795
UK	3.842301	3.76508	3.191358	3.140905
USA	3.602972	3.48758	2.636387	2.828965
Zambia	−0.53844	−0.54906	−0.64026	0.144427

Source. The author.

In Table 4, positive values of the GGI show that governance capabilities in home country of a foreign bank are higher than in India. Thus, low quality of institutional arrangements in India may discourage the entry of foreign banks from these countries. However, negative values of GGI indicate that quality of governance in India is higher than in the home country. Moreover, it may also be argued that banks from counties that have similar working conditions as India, that is, counties for which the value of GGI is 0 or close to 0, may also prefer to enter India because of similarities in institutional environment. Table 5 presents descriptive statistics of the GGI for all the years.

Table 5.

Descriptive Statistics of GGI, 23 Countries

	GGI-2000	GGI-2001	GGI-2005	GGI-2010
Mean	1.33	1.45	1.05	1.27
Standard deviation	1.92	1.87	1.92	1.89
Median	1.22	1.37	1.16	1.19

Source. The author.

Table 5 shows that there has been a decline in the mean value of GGI after 2001. Table 5 also reports a decline in the median of the index after 2001. The standard deviation, however, has not varied much over the years. This shows that, overall, the positive gap in governance quality between each home country and India (indicating higher governance quality in home countries than in India) has decreased. In other words, this means that in comparison to the governance quality in home countries, the quality of governance in India seems to have improved after 2001.

Regulatory Arbitrage and Foreign Banks in India: Exploratory Analysis

This section aims to formulate tentative inferences on linkages between regulatory arbitrage and presence of foreign banks in India. We attempt this by exploring the data on foreign banks’ participation in the Indian banking sector vis-à-vis the BRGI computed in the third section of this article.

Table 6 presents a list of 63 foreign banks that were present in India during the period 2000–2012, along with their average asset share in total assets of foreign banks in India. Based on the country in which these banks are headquartered, we identified 23 countries as home countries of the foreign banks. In terms of the income levels, these 23 countries are reported under three categories: high-income countries (Australia, Bahrain, Belgium, Canada, France, Germany, Japan, Netherlands, Oman, Singapore, South Korea, Switzerland, the UK, UAE, and the USA), middle-income countries (China, Indonesia, Mauritius, Russia, South Africa, Sri Lanka, and Thailand), and low-income countries (Bangladesh). As illustrated in Table 6, during the period 2000–2012, the Indian banking sector largely comprised foreign banks that originated from high-income countries such as France (seven banks), Japan (seven banks), the USA (seven banks) and the UK(five banks). In terms of the average asset share, three foreign banks held significant share of total foreign bank assets in India during the sample period, namely Citibank (approximately 21%), Standard Charter (approximately 20%), and Hong Kong and Shanghai Bank Corp. (HSBC, approximately 18%).

Table 6.

Average Asset Share of Foreign Banks in Total Foreign Bank Assets (2000–2012)

	Country of Origin	Income Group	Foreign Banks	Average Asset Share (2000–2012)
1	Australia	High	Australia and New Zealand Banking Group	0.04
		income	Commonwealth Bank of Australia	0.01
			Grindlays Bank	0.92
2	Bahrain	High income	Bank of Bahrain & Kuwait	0.34
3	Bangladesh	Low	Arab Bangladesh Bank	0.04
		income	Sonali Bank	0.02
4	Belgium	High	KBC Bank	0.07
		income	Antwerp Diamond Bank	0.19
5	Canada	High	Bank of Nova Scotia	1.92
		income	Toronto Dominion Bank	0.02
6	China	Middleincome	Chinatrust Commercial BankIndustrial and Commercial Bank of China	0.090.01
7	France	High	Banque Nationale de Paris	0.18
		income	BNP Paribas	2.21
			Calyon Bank	0.57
			Credit Agricole Bank	0.34
			Credit Agricole Indosuez	0.29
			Credit Lyonnais	0.44
			Societe Generale	0.64
8	Germany	High	Commerzbank	0.10
		income	Deutsche Bank	5.79
			Dresdner Bank	0.05
9	Indonesia	Middle income	Bank International Indonesia	0.05
10	Japan	High	Bank of Tokyo Mitsubishi	1.07
		income	Mizuho Corporate Bank(Erstwhile The Fuji Bank)	0.39
			Sakura Bank	0.10
			Sanwa Bank	0.04
			Sumitomo Bank	0.17
			The Fuji Bank	0.05
			UFJ Bank	0.10
11	Mauritius	Middle income	State Bank of Mauritius	0.27
12	Netherlands	High	ABN AMRO Bank	6.83
		income	International Netherland Bank	0.11
			Rabobank International	0.01
13	Oman	High	Bank Muscat International	0.06
		income	Oman International Bank	0.29
14	Russia	Middle income	JSC VTB Bank	0.01
15	Singapore	High	DBS Bank Ltd	2.08
		income	Development Bank of Singapore	0.06
			Overseas Chinese Bank	0.01
16	South Africa	Middle income	FirstRand Bank	0.03
17	South Korea	High	Cho Hung Bank	0.09
		income	Shinhan Bank	0.13
18	Sri Lanka	Middle income	Bank of Ceylon	0.10
19	Switzerland	High	Credit Suisse AG	0.03
		income	UBSAG	0.22
20	Thailand	Middle	Krung Thai Bank	0.04
		income	Siam Commercial Bank	0.02
			The Siam Commercial Bank	0.02
21	UK	High	Barclays Bank	2.14
		income	Hongkong and Shanghai Bank	18.32
			Royal Bank of Scotland (Erstwhile ABN AMRO Bank)	1.07
			Standard Chartered Bank	20.19
			Standard Chartered Grindlays Bank (Erstwhile Grindlays Bank)	2.53
22	United Arab	High	Abu-Dhabi Commercial Bank	0.77
	Emirates	income	Mashreq Bank	0.18
23	USA	High	American Express Bank	1.53
		income	American Express Banking Corp.	0.10
			Bank of America	3.49
			Chase Manhattan Bank	0.05
			Citi bank	21.34
			JP Morgan Chase Bank (Erstwhile Morgan Guaranty Trust)	1.55
			Morgan Guaranty Trust	0.06

Source. Statistical tables relating to banks in India (RBI) and World Bank Classification of Countries (2012).

Using the list of home countries, we categorized the home countries by region, viz. Europe, North America, Asia, Africa, and Australia. Figure 2 illustrates average asset share of foreign banks in total foreign bank assets in India during the period from 2000 to 2012, aggregated by the region of their home country. It is interesting to note that foreign banks from seven European countries, on an average, held 61 percent of total foreign bank assets in India, while foreign banks from two North American countries held 30 percent share in total foreign bank asset in India. Moreover, banks that originate from Asian countries together held 7 percent share in total foreign bank assets, and Australian banks accounted for 2 percent share in total foreign bank assets. Similarly, in terms of number of offices (Figure 3), a comparison of number of foreign bank offices between the years 2000 and 2012 show that the foreign banks that established majority of offices in India originated from countries of Europe and North America, followed by Asian countries, while foreign banks belonging to the African and Australian regions established relatively smaller number of offices in India. Overall, the aforementioned description suggests that the Indian banking sector has been an attractive destination for banks originating from high-income countries of Europe and North America.

Figure 2.

Source. Profile of Banks (RBI 2013b).

Figure 3.

Source. Profile of Banks (RBI 2013b).

To explore the relationship between regulatory arbitrage and presence of foreign banks in India, we plot the asset share of foreign banks in India aggregated by their country of origin vis-à-vis the BRGI for the year 2010. Figure 4 presents a scatter plot that categorizes the group of 23 home countries into four segments. The first segment (down left) shows countries with which India has small gaps in banking regulations and low presence of foreign banks headquartered in these countries (seven countries). Similarly, we have other segments showing countries with which India has small banking regulation gap and large presence of foreign banks (two countries), large gap in banking regulations and large presence of foreign banks (six countries), and large gap in banking regulations and low presence of foreign banks (eight countries). The figure clearly illustrates that countries where banking regulations are diverse from India are also home to foreign banks that held significant share in total foreign bank assets in India. As evident from the figure, banks from the UK and USA held the largest share in total foreign bank assets in India in 2010, and these were also the countries where banking regulations are different from India. In the same segment, there are countries like Germany, France, Japan, and Switzerland that are also home countries of foreign banks that held significant shares in total asset of foreign banks in India. Considering their income levels, these are high-income countries where bank regulatory stringencies are relatively diverse from that in India.

Figure 4.

Source. The author.

Collectively, these observations provide tentative evidence that difference in banking regulations between India and the home countries may be an important factor driving foreign banks’ presence in India. Thus, if regulation arbitrage is, indeed, a significant factor motivating cross-border expansion of banks, then going by the observations in Figure 4, banks from countries like Indonesia, Russia, China, and South Africa can be expected to increase their presence in India.

Results and Implications

We present the main results of our analysis in Table 7 for the two samples. One sample includes 37 countries that share economic relations with India by means of trade and commerce, while the second sample includes only home countries of the foreign banks present in India (23 countries). Both the samples are constructed for the period 2000, 2001, 2005, and 2010. Beginning with the sample of 37 countries, Regression I reports the coefficients of Equation (1) using binary variable ENTRY as the dependent variable. Regression II reports the coefficients of Equation (2) where the dependent variable is logarithmic of one plus office share of foreign banks from jth country in India by total offices of foreign banks in India (1 + s.Office). Finally, Regression III presents the coefficients of Equation (3), using logarithmic of one plus asset share of foreign banks from jth country in India by total assets of foreign banks in India (1 + s.Assets) as the dependent variable. A Hausman specification test chooses Fixed Effects over Random Effects for estimation of Equations (2) and (3). The adjusted R-square for the three regressions is 0.93, 0.90, and 0.91, respectively, indicating that the regressions explain most of the variations in foreign banks’ presence in India.

Table 7.

Foreign Banks Presence in India and Regulatory Arbitrage

Dependent Variable	37 CountriesRegression IENTRY	23 Home Countries
Dependent Variable	37 CountriesRegression IENTRY	Regression IIln(1 + s.Offices)			Regression IIIln(1 + s.Assets)
	(1)	(2)	(3)	(4)	(5)	(6)	(7)
Constant	5.18291	0.165059	0.6527*	3.614825	1.900903	9.7251***	4.056749
	(0.578)	(0.924)	(0.062)	(0.118)	(0.112)	(000)	(0.283)
BRGI	-0.00047	0.00363**	0.004477*	0.00092*	0.001501*	0.00206*	0.01685**
	(0.807)	(0.005)	(0.045)	(0.041)	(0.084)	(0.071)	(0.027)
GGI	0.658647	0.05045	0.0828513	−0.15595	0.031591	1.037347	−0.15926
	(0.388)	(0.77)	(0.469)	(0.343)	(0.79)	(0.282)	(0.784)
ln(PCI)	-0.60873	0.066073	0.2497608	0.164845	−0.12714	0.6788172	−0.00846
	(0.627)	(0.746)	(0.211)	(0.59)	(0.366)	(0.283)	(0.993)
ln(1 + TOTAL TRADE)	0.111012	0.031977	−0.1687	0.32885*	0.008277	0.97485*	0.180616
	(0.797)	(0.404)	(0.127)	(0.031)	(0.753)	(0.042)	(0.708)
ln(FDI)	0.54400*	0.016953	0.0298882	0.098992	0.002777	0.0962187	0.263072
	(0.055)	(0.619)	(0.589)	(0.149)	(0.906)	(0.645)	(0.25)
ln(FII)	0.125366	−0.12629	0.0288383	0.09024*	−0.04673	−0.03580	0.237816
	(0.548)	(0.577)	(0.475)	(0.079)	(0.764)	(0.805)	(0.171)
COMMONALITY	–7.46159**	Dropped	0.65439*	0.61428*	Dropped	2.600869	0.938529
	(0.005)		(0.038)	(0.041)		(0.204)	(0.369)
Time Dummies (refence year 2000)	Absent	Absent	Absent	All significant except 2010	Absent	Absent	All significant
Disturbances	NA	NA	AR(1)	NA	NA	AR(1)	NA
Observations	148	92	70	74	92	70	74
Hausman test	RANDOM	FIXED	FIXED	FIXED	FIXED	FIXED	FIXED
Adjusted R²	0.93	0.9	0.5117	0.4088	0.91	0.6707	0.5614

Source: The author.

Note. p-Values are computed by the heteroskedasticity-robust standard errors clustered for recipient countries in (1), (2), and (5) and are presented in parentheses. *, **, and *** represent statistical significance at the 10%, 5%, and 1% level, respectively.

In column (1) of Table 7, effect of BRGI (bank regulatory difference between each country and India) on ENTRY is found to be insignificant. This result shows that foreign bank’s entry choice is not influenced by the difference in banking regulations between their home jurisdiction and India. Thus, regulatory arbitrage does not seem to determine entry decision of foreign banks of countries sharing bilateral relations with India. However, among the control variables, effect of ln(FDI) on foreign banks’ decision to enter India is positive and statistically significant at 10 percent. This positive association seems to indicate that in terms of bilateral FDI, economic relationship between home countries of foreign banks and India is an important factor determining entry of foreign banks in India.

The effect of another control variable—COMMONALITY—is found to be negative and significant at 5 percent. This result indicates that foreign banks’ entry into India is not driven by geographical proximities, and linguistic and colonial similarities. Though this result is counterintuitive, it seems to be true in case of India. In the section on data description, we observed that a substantial share of foreign bank assets in India is held by banks that originate from countries of the European and North American region. Because banks from neighbouring Asian countries have an average participation in the Indian banking system, negative association of COMMONALITY with ENTRY seems to fit the case of India. The effect of GGI and ln(PCI) are found to be insignificant. This shows that the income levels and gap in governance quality do not play a significant role in determining foreign banks’ entry in the Indian banking sector. This may seem to be contradictory, when we expect institutional environment in home and host countries to play an important role in determining investments by international banks (Galindo et al., 2003; Gelos & Wei, 2002; Petersen & Rajan, 1994; Wei, 2000). However, our econometric analysis using cross-foreign bank and cross-country-level data indicates that, on an average, the differential impact of governance quality gap is insignificant in explaining the presence of banks from the sample of home countries considered in the analysis. This shows that perhaps the entry or expansion of foreign banks in India is not significantly impacted by governance gap.

Columns (2)–(4) of Table 7 show the estimated results obtained for the sample of 23 countries identified as home to existing foreign banks in India. The results of these columns show the impact of the main explanatory variable and control variables on intensity of foreign banks’ presence, namely share in offices (ln(1 + s.Offices)) and share in assets (ln(1 + s.Assets)). Here, effect of BRGI is found to be positive and significant at 5 percent in Regression I and significant at 10 percent in Regression II. More specifically, this positive association indicates that difference in bank regulatory stringencies between home countries and India significantly affects presence of existing foreign banks in India. Interestingly, none of the control variables are found to be significant in both the regressions, unlike the results obtained for the sample of 37 countries, Regressions I and II are also estimated with time dummies to control for time-specific effects and with regression disturbance term modelled as an AR(1) process to correct for autocorrelation between dependent and explanatory variables. The results of which are reported in columns (3), (4), (6), and (7) of Table 7. The inclusion of time dummies and AR(1) disturbances does not change the results obtained from the earlier regressions.⁹ Thus, the overall observation of banking regulation arbitrage, positively and significantly determining the presence of existing foreign banks in India, remains robust across different specifications.

Comparing the insignificant impact of BRGI on prospective entry of foreign banks originating from economically linked countries vis-à-vis the positive impact on the presence of the existing foreign banks, the insignificance need not imply that banks may not consider this as a factor; perhaps, some banks take into account arbitrage while making entry decision, but not all banks consider this as important. Therefore, the average impact of arbitrage is found to be insignificant, while bilateral economic relation reflected by FDI/FII and COMMANALITY turns out to be significant. Once entry is made, foreign banks make use of regulatory arbitrage to make profitable expansion. In all, the results do seem to suggest that opportunity of bank regulation arbitrage is an important factor explaining foreign banks’ presence in India.

Conclusion

In this article, we attempt to empirically examine the relationship between regulatory arbitrage and presence of foreign banks in India. We constructed a BRGI to capture the joint effect of differences between various banking regulations existing in India and in the sample of countries included in this study. This index is used as a comprehensive measure to examine the effect of regulatory arbitrage on entry and expansion of foreign banks in India. In the section on the exploratory analysis, we made an initial indication that foreign banks in India mainly originate from high-income OECD countries, where banking regulations are comparatively diverse from that existing in India. Since developed countries follow stringent regulations, banks from these countries are expected to expand their business activities in India, in order to benefit from the regulatory difference. Supporting this observation, our results seem to find evidence of regulatory arbitrage as positively and significantly impacting the expansion of foreign banks in India. This result is in line with earlier studies by Houston et al. (2012), and Herrero and Martinez Pería (2007) who found bank capital flows to be driven by stringencies of banking sector regulations across countries. However, as far as entry decision of foreign banks is concerned, regulatory arbitrage does not seem to affect their choice to enter the Indian banking sector. In this case, economic relationship in terms of bilateral FDI is found to be an important factor inducing entry of foreign banks in India. Also, geographical proximities, and linguistic and colonial similarities were found to have a negative association with foreign banks’ entry into India. This may be because a large number of foreign banks in India originate from high-income countries of Europe and North America, where social settings are different from India.

Footnotes

Acknowledgments

I would like to thank the participants of Second International Conference on South Asian Economic Development, organized by South Asian University in Delhi, February 2016, and International Conference on Economics and Finance, organized by Nepal Rastra Bank, Kathmandu, February 2016. Discussion with Dr Mandira Sarma and Professor B. K. Pradhan are gratefully acknowledged. All errors are mine.

Declaration of Conflicting Interests

The author declared no potential conflict of interest, with respect to research, authorship, and/or publication of this article.

Funding

The author is thankful for the financial assistance provided by the Indian Council for Social Science Research (ICSSR), Delhi, in carrying out this research.

Notes

Appendix A

Methodology of Principal Component Analysis for construction of Banking Regulation Gap Index and Governance Gap Index

To compute Banking Regulation Gap Index (BRGI) and Governance Gap Index (GGI), we use the statistical technique of Principal Component Analysis (PCA). Before applying PCA, we formulate different indicators to quantitatively measure different aspects of banking regulations and governance capabilities for our sample of countries. Then, we calculate regulatory and governance gaps by taking the difference between the values of the indicators of each country with India. The gaps are estimated as follows:

$x_{1}, \dots, x_{k}$ be k indicators. Cross-country gap in these indicators is estimated as ${Gap}_{i} = x_{j th country, t} - x_{India, t}$ . for $i = 1, \dots, k$ . in time t.

Now, using PCA, we define BRGI as the first principal component of gap between banking regulation indicators of each country and India. Similarly, GGI is defined as the first principal component of gap between governance quality indicators of each country and India. We write the first principal component as follows: (A1)

{PC}_{1} = α_{11} {Gap}_{1} + \dots + α_{1 k} {Gap}_{K}

where αs are the elements of the eigenvectors of covariance matrix corresponding to the first eigenvalue.

The principal components (PCs) in Equation (A1) are estimated as linear combination of a set of variables (Gaps in our case) with coefficients being the elements of the eigenvector of covariance matrix corresponding to its eigenvalue. Generally, studies use PCA as a dimension reduction technique by incorporating only those components that explain maximum variation in the data. The usual criteria employed to account for maximum variation is to include principal components of correlation–covariance matrix corresponding to the largest eigenvalues. As a rule of thumb, authors like Chatfield and Collins (1992), and suggest inclusion of principal components corresponding to eigenvalues that are greater than 1. For estimation of principal components, we adopt this criterion.

Using the abovementioned methodology, the following discussion illustrates the indicators and rationale for including them in constructing the indices.

References

Battilossi

(2006). The determinants of multinational banking during the First Globalization, 1870–1914. European Review of Economic History , 10(3), 361–338.

Berger

A. N.

, & Udell

G. F.

(1996). Universal banking and the future of small business lending. In: Saunders

, & Walter

(Eds.), Financial system design: The case for Universal Banking (pp. 559–627). Irwin Publishing.

Carbo-Valverde

, Edward J.

, & Rodriguez-Fernandez . (2009). Evidence of regulatory arbitrage in cross-border mergers of banks in the EU (NBER Working Paper 15447). NBER.

Cárdenas

, Graf

J. P.

, & O’Dogherty

(2003). Foreign banks entry in emerging market economies: A host country perspective . http://www.bis.org/publ/cgfs22cbpapers.htm

Carpenter

, & Dunung

S. P.

(2011). International business: Opportunities and challenges in a flattening world . Vol 1.0. Flat World Education, Inc.

Chatfield

, & Collins

A. J.

(1992), Introduction to multivariate analysis . Chapman and Hall.

Claessens

, & Van Horen

(2013). Impact of foreign banks . De Nederlandsche Bank NV (Working Paper No. 370). https://www.dnb.nl/en/binaries/working%20Paper%20370_tcm47-284840.pdf

Cull

, & Martínez Pería

M.S.

(2010). Foreign bank participation and c rises in developing countries (Policy Research Working Paper 4128). World Bank.

Cumming

(1987). The economies of securitization. FRBNY Quarterly Review , (3), 11–23.

10.

Dong

, Song

, & Tao

(2011). Regulatory arbitrage: Evidence from bank cross-border M&As . https://ssrn.com/abstract=1849723

11.

Dunning

J. H.

(1977). Trade, location of economic activity and the MNE: A search for an eclectic approach. In: Ohlin

, Hesselborn

P. O.

, & Wijkman

P. M.

(Eds.), The international allocation of economic activity (pp. 395–418). Macmillan.

12.

Everitt

B. S.

, & Dunn

(2001). Applied multivariate data analysis . 2nd ed. Arnold.

13.

Galindo

, Micco

, & Serra

(2003). Better the devil that you know: Evidence on entry costs faced by foreign banks (Inter-American Development Bank Working Paper 477). Inter-American Development Bank.

14.

Gelos

, & Shang-Jin

(2002), Transparency and international investor behavior (NBER Working Paper 9260). NBER.

15.

Goldberg

L. S.

(2007). Financial sector FDI and host countries: New and old lessons. FRBNY Economic Policy Review , 13(1), 1–17.

16.

Herrero

A. G.

, & Martinez Peria

M. S.

(2007). The mix of International Banks’ Foreign Claims: Determinants and implications. Journal of Banking & Finance , 31(6), 1613–1631.

17.

Herrero

A. G.

, & Simón

D. N.

(2003). Determinants of financial sector FDI to emerging economies: A home country’s perspective (Working Document 216). Bank of Spain.

18.

Horen

V. N.

(2007). Foreign banking in developing countries: Origin matters. Emerging Market Review , 8(2), 81–105.

19.

Houston

J. F.

, Lin

, & Ma

(2012). Regulatory arbitrage and international bank flows. Journal of Finance , 67(5), 1845–1895.

20.

Jones

(2000). Emerging problems with the Basel Capital Accord: Regulatory capital arbitrage and related issues. Journal of Banking & Finance , 24(1–2), 35–58.

21.

Jones

, & Kolatch

(1999). The federal safety net, banking subsidies, and implications for financial modernization. FDIC Banking Review , 12(1), 1–17.

22.

Kaufmann

, & Mastruzzi

(2010). The worldwide governance indicators: Methodology and analytical issues (Policy Research Working Paper No. WPS 5430). World Bank.

23.

Morrison

A. D.

, & White

(2009). Level playing fields in International Financial Regulation. The Journal of Finance , 64(3), 1099–1142.

24.

Organization.

25.

Pavel

, & Phillis

(1987). Why commercial banks sell loans: An empirical analysis. Federal Reserve Bank Chicago: Economic Perspectives , 11(2), 3–14.

26.

Petersen

M. A.

, & Rajan

R. G.

(1994). The benefits of lender relationships: Evidence from small business data. Journal of Finance , 49(1), 478–495.

27.

Plummer

M. G.

(2009). Asian economic integration, Trade, foreign direct investment, and finance . World Scientific Publisher.

28.

RBI. (2005). Road map for presence of foreign banks in India . RBI.

29.

RBI. (2011). Presence of Foreign Banks in India . Discussion Paper. RBI.

30.

RBI. (2013a). Scheme of setting up wholly owned subsidiaries (WSO) by foreign banks in India . RBI.

31.

RBI. (2013b). Profile of banks . RBI.

32.

The World Factbook. (2010). Washington, DC: Central Intelligence Agency, 2010. https://www.cia.gov/library/publications/resources/the-world-factbook/index.html

33.

Wei

S. J.

(2000). How taxing is corruption to international investors. Review of Economics and Statistics , 82, 1–11.

34.

World Bank. (2001). Banking regulation and supervision survey . World Bank, 2001. https://www.worldbank.org/en/research/brief/BRSS

35.

World Bank. (2003). Banking regulation and supervision survey . World Bank, 2003. https://www.worldbank.org/en/research/brief/BRSS

36.

World Bank. (2007). Banking regulation and supervision survey . World Bank, 2007. https://www.worldbank.org/en/research/brief/BRSS

37.

World Bank. (2011). Banking regulation and supervision survey . World Bank, 2011. https://www.worldbank.org/en/research/brief/BRSS

38.

WTO. (1994).The results of the Uruguay round of multilateral trade negotiations:The legal texts. Geneva: GATT Secretariat.

39.

WTO. (1997). Trade policy review: European union . Geneva: World Trade