The Case Of India

Published Date: 02 Nov 2017

ECS 4030

Foreign Banks entering developing countries:

The case of India

Shweta Chaudhary*

Msc Banking and Finance

Middlese - University Business School May 2013

mdx logo.jpg

*Views of the author are that of his own and does not reflect or influence the views of an individual

Acknowledgment

Abstract

Indian banking sector is a combination of different structure banks working in the same environment. Public sector units, private banks, foreign banks, scheduled and non-scheduled banks and even non-banking finance corporations are the key players of this economy. Here, in this paper we will be analysis the data and the hypothesis that did the entry of foreign banks as a part of globalisation policy accepted by India in the 1990s beneficial to the economy or it was a not needed step. And we could have done without the liberalisation and globalisation policies. We will judge the branch locations in different states and districts over a time period covering post and pre-privatisation period. Did these firms make access to finance easier for the client? Or it only made the rich richer.

Table of contents:

Introduction

A foreign bank is a bank that has been setup in a different country, which is its home country and is also serving customers of another country, the foreign countries. Foreign banks are active in India, just the way the local banks are. They need to adhere to all the rules and regulations that are applicable to banks based out of India and they are governed by the Reserve Bank of India that controls/governs all banks operating in India. In the recent years there had been a lot of transition of banks from Developed countries like US and Japan to Developing countries i.e. emerging economies of the world.

In this paper we will study the establishment of foreign banks from various countries making a mark in the Indian banking sector. As stated above these banks work under the rules and regulations of the Apex bank of India which is Reserve Bank of India. It was a conception that foreign banks benefit the healthy competition, giving a boost to accessibility to financial instruments and services, improves financial and economic efficiency of their clients, and introduce a huge financial stability (Clarke, cull, Martinez Pera and Sanchez, 2003, Claessens 2006. Chopra 2007 and cull and Martinez peria, 2011). Normally, lesser costs of financial intermediation and less profitability are noticed with huge foreign bank occurrence (claessens, demirguc-kunt, and Huizinga, 2001, mian 2003, berger, Clarke, cull, klapper and udell, 2005).

We would also like to add that foreign banks probably forced govts. To improvise regulation and management, imparts transparency, and catalyze national reforms (Levine 1996, Dobson, 2005, and mishkin, 2006).

The foreign banks working in India are as follows:

Foreign banks branches in India as on March 31, 2013

Sl.No

Name of bank

Country of

Incorporation

No of

Branches in

India

1

AB Bank Ltd.

Bangladesh

1

2

The Royal Bank of Scotland N.V.

Netherlands

31

3

Abu Dhabi Commercial Bank Ltd.

UAE

2

4

American E -press Banking Corporation

USA

1

5

Antwerp Diamond Bank N.V.

Belgium

1

6

Bank Internasional Indonesia

Indonesia

1

7

Bank of America

USA

5

8

Bank of Bahrain & Kuwait BSC

Bahrain

2

9

Bank of Ceylon

Sri Lanka

1

10

Bank of Nova Scotia

Canada

5

11

Barclays Bank Plc.

United

Kingdom

7

12

BNP Paribas

France

8

13

Credit Agricole Corporate &

Investment Bank

France

5

14

Chinatrust Commercial Bank

Taiwan

2

15

Citibank N.A.

USA

42

16

DBS Bank Ltd.

Singapore

12

17

Deutsche Bank

Germany

18

18

HSBC Ltd

Hong Kong

50

19

J.P. Morgan Chase Bank N.A.

USA

1

20

JSC VTB Bank

Russia

1

21

Krung Thai Bank Public Co. Ltd.

Thailand

1

22

Mashreq Bank PSC

UAE

1

23

Mizuho Corporate Bank Ltd.

Japan

2

24

Oman International Bank SAOG

Sultanate of

Oman

2

25

Shinhan Bank

South Korea

3

26

Societe Generale

France

3

27

Sonali Bank Ltd.

Bangladesh

2

28

Standard Chartered Bank

United

Kingdom

101

29

State Bank of Mauritius

Mauritius

3

30

The Bank of Tokyo- Mitsubishi UFJ Ltd.

Japan

4

31

UBS AG

Switzerland

1

32

FirstRand Bank Ltd

South Africa

1

33

United Overseas Bank Ltd

Singapore

1

34

Commonwealth Bank of Australia

Australia

1

35

Sberbank

Russia

1

36

Credit Suisse A.G

Switzerland

1

37

Australia and New Zealand Banking Group

Ltd.

Australia

1

38

Rabobank International

Netherlands

1

39

Industrial & Commercial Bank of China Ltd.

China

1

40

Woori Bank

South Korea

1

41

National Australia Bank

Australia

1

42

Westpac Banking Corporation

Australia

1

43

Sumitomo Mitsui Banking Corporation

Japan

1

 

 

 

331

List of Foreign banks having Representative Offices in India as on March 2013.

Sr.

No.

Name of the representative office

Country of

incorporation

Centre

Date of

opening

1

Raiffeisen Zentral Bank Osterreich AG

Austria

Mumbai

1.11.1992

2

Fortis Bank

Belgium

Mumbai

6.10.1987

3

K.B.C. Bank N.V.

Belgium

Mumbai

1.02.2003

4

Royal Bank of Canada

Canada

Mumbai

1.2.2008

5

Toronto Dominion Bank

Canada

Mumbai

16.11.2009

6

Credit Industriel et Commercial

France

New Delhi

1.04.1997

7

Natixis

France

Mumbai

4.01.1999

8

Bayerische Hypo – und

Vereinsbank

Germany

Mumbai

12.07.1995

9

DZ Bank AG Deutsche Zentral –

Genossenschafts Bank

Germany

Mumbai

22.02.1996

10

Landesbank Baden – Wurttemberg

Germany

Mumbai

1.11.1999

11

Commerzbank

Germany

Mumbai

23.12.2002

12

BayernLB

Germany

Mumbai

15.4.2008

13

Norddeutsche Landesbank

Girozentrale (NORD LB)

Germany

Mumbai

1.9.2008

14

KfW IPEX Bank GmbH

Germany

Mumbai

1.4.2009

15

DEPFA Bank

Ireland

Mumbai

9.3.2007

16

Intesa Sanpaolo S.p.A

Italy

Mumbai

1.11.1988

17

Uni Credito Italiano

Italy

Mumbai

1.08.1998

18

Banca Populare Di Verona E Novara

Italy

Mumbai

18.06.2001

19

BPU Banca –Banche Popolari Unite

S.c.r.l

Italy

Mumbai

16.01.2006

20

Monte Dei Paschi Di Sienna

Italy

Mumbai

07.04.2006

21

Banca Popolare di Vicenza

Italy

Mumbai

29.04.2006

22

CIMB Bank Berhad

Malaysia

Mumbai

23.11.2010

23

Everest Bank Ltd.

Nepal

New Delhi

24.03.2004

24

DNB Bank ASA

Norway

Mumbai

27.8.2008

25

Caixa Geral de Depositos

Portugal

Mumbai

Goa (EC)

8.11.1999

26

Vnesheconombank

(Bank for Foreign Economic Affairs)

Russia

New Delhi

1.3.1983

27

Promsvyazbank

Russia

New Delhi

25.04.2006

28

Gazprombank

Russia

New Delhi

12.7.2010

29

Hana Bank

South Korea

New Delhi

 

30

Korea Exchange Bank

South Korea

New Delhi

27.8.2008

31

Kookmin Bank

South Korea

Mumbai

1.06.2012

32

Industrial Bank of Korea

South Korea

New Delhi

22.11.2012

33

Banco de Sabadell SA

Spain

New Delhi

2.08.2004

34

Banco Bilbao Vizcaya Argentaria

Spain

Mumbai

2.4.2007

35

CaixaBank S.A.

Spain

New Delhi

1.2.2011

36

Hatton National Bank

Sri Lanka

Chennai

1.01.1999

37

Svenska Handlesbanken

Sweden

Mumbai

1.08.2006

38

Skandinaviska Enskilda Banken AB

Sweden

New Delhi

1.02.2008

39

Zurcher Kantonalbank

Switzerland

Mumbai

27.06.2006

40

Mega International commercial Bank

Taiwan

Mumbai

2.12.2008

41

Asya Katilim Bankasi AS

Turkey

Mumbai

1.9.2012

42

Emirates Bank International

UAE

Mumbai

16.06.2000

43

First Gulf Bank

UAE

Mumbai

26.10.2009

44

Duncan Lawrie Ltd

United

Kingdom

Kolkata

30.10.2009

45

The Bank of New York Mellon

USA

Mumbai

27.10.1983

46

Wells Fargo Bank N.A.

USA

Mumbai

(Sub-office at Chennai & New Delhi)

1.11.1996

Indian Banking history:

When in 1947, India got independence from East India company rule, at that time India had main five banks namely: Central Bank of India, Punjab National Bank, United Commercial Bank, Bank of Baroda, and Bank of India. This year and the preceding year were vicious for Indian Economy as the following year in 1948, India and Pakistan got separated into two different countries and it lead to division of the bigger banks.

Liberalisation was introduced in Indian economy 18 years ago. Indian banking system is an animated cluster of competence enhanced Public Sector units and progress thirsty private sector banks. The services, money instruments, efficiency, IT facilities and management would have been a far off our vision 10 years ago. The amount of conveniences banks are providing to their corporate clients and retail customers has been improving and it was something no one ever imagined or even in their thoughts. Indian banking industry has witnessed exponential escalation the CNB Bank Index has shown a growth of 1100% in absolute terms, a compounded rate annual growth rate of 25% in the time period of 2000-2010. And if we look at the sensex, it grew at an compounded annual growth rate of 14%. The year 2010 was a good year for the Indian banking sector as it contributed to the GDP by 16.35%.

Data:

The data used in this paper has been extracted from the databases of Reserve Bank of India. The data regarding the geographical locations of the new banks and the existing Indian banks is taken from the publications of the RBI issued by the Director General. The time frame of the data is 1988-2004. We have taken this time frame because it covers the pre-privatisation period and the post privatisation period also. This was the time when there have been extreme changes in the Indian Banking system and the economy as a whole.

By using this data, we will now map out the locations of the foreign bank openings in different regions of India. This is presented in the form of a table in table1.

Strategy for Empirical study and their hypothesis:

The OLS arrangement used here is:

yi ,d ,t = β0 +β1Foreign Bankd ,t +αi +δ t +ε i ,d ,t ... (1)

Here, we have taken firm level outcome y of firm i as the dependent variable, in year t based in district d. The value y will involve measures of firms, unsettled stock of loans from Banks and FIs. Other various variables will narrated below. In district d of year t, the indicator of existence of foreign bank is Foreign Bank. that is twisted on for all the firms in the district if an outsider bank was already present in that year. When district-level allocation of foreign banks is put to use instead of an indicator, every following were seen to be similar. Though, there have been a theory which says the sheer entry of foreign lenders and investors is enough to persuade a segmentation of the market and intensify information asymmetries1 . A full set of firm dummies, αi , absorb any fixed differences in firms’ use of loans in a way that the coefficient of interest, 1β , is approximately only using within firm changes.

By this we are sure of one thing that the change in the firm specific and location specific changes of the bank branch won’t influence the empirical analysis. Running of country-level pattern in lending trend that might be caused by other alterations in government policies or regulations will be time dummy variable δt. Now, the last thing being the standard errors, as we know the disparity of foreign entry appears at this district level, the standard errors are concentrated here.

Here, the consequence of foreign bank entry is captivated by the coefficient β1 . It is to be estimated using the changes in the patterns of borrowing of the firms located in the eight districts who got their foreign bank from 1991-2002 comparative to firms in districts who never got a foreign bank. There are certain benefits of this within-the boundary analysis as compared to the more criterion approach of making use of cross-sectional, country-level analysis. Here firstly the specified made usage of the spread out timing of entrance by foreign banks in the time frame of 1994-2001. The coefficient β1 .

Is found out by mere usage of variations in borrowing trends at the same time as the time of foreign bank entrance in each and every district. Secondly, we also saw that in house variation ensured that

1 (Dell’Arricia and Marquez, 2004; Sengupta, 2006; Gormley, 2006)

the results for Beta are traceable in the entry of the FB instead of other country-level economic reforms introduced during the early 90s. What happens after such country-level changes is that the consequences are absorbed in by the y(year) dummies, and the results of foreign bank entrance will be genuinely recognised under the assumption that the pattern in use and loan firm size in those eight districts should have been similar like those in control group in the absenteeism of the foreign bank’s entrance. The initial organizing group uses all firms found in the areas who soen’t have a foreign bank in the time gap of 1991-2002. Head offices in these 18 places who already had a foreign bank saw a decline in the regression. Table no.2.

Keeping in mind the robustness test, regressions seem to be also running with a tiny control group comprising of firms seen in the nine places who got their initial foreign bank in the time frame of ’03-’04. The nine districts who showed a healthy control and were seen as a potential map points for foreign bank entrance were Fardabad, Nagpur, Surat, Aurangabad, Lucknow, Bhopal,Thane, Patna and Rajkot. As this wasn’t different than the earlier eight districts who had an early foreign bank. But the problem being that in equation (1) there is a limitation, it doesn’t let us to check whether the entrance of foreign bank has an differentially influence on the firms. So, we come up with another equation to overcome the limitation as it takes into consideration the interaction , Foreign Bank X ROA, here ROA is said to be the demean percent average return of the total assets of t firm in the time gap of ‘91-‘93. The profits are calculated by putting use of profit minus taxes and the net of non-recurring operation. The whole set of year and profit interactions, δ ×ROA, will also be considered to let the firms cross the country to variants in trend as a property of their earlier profitability.

yi ,d ,t = β0 + β1Foreign Bankd ,t +β2 Foreign Bankd ,t ×ROAi +αi +δt + δ t ×ROAi +ε

... (2)

Now we move on to the next specification, β1 constantly seems to be defining the major effects of foreign bank entrance since ROA is demean. Simultaneously , βhere describes marginal effect of keeping elevated ROA in advance. Considering the interaction of Foreign Bank x ROA checks whether the firm’s profit issues matter a lot for credit accessibility after foreign bank entrance the places, under the supposition a firm’s previous ROA is optimistic forecaster of future possibilities. Hence, this permits us to analyse whether foreign bank entrance is engaged with a shit of credit from comparetivly less profitable organisations to more profitable organisations. Because it is indicated by

2 β > 0. Furthermore, we see a drop in credit accelerated predominately by lesser loans advanced to politically engaged, unproductive firms, it should also be shown by 2 β > 0.

Lastly, we see that it is absolutely supposed that the impact of foreign bank entrance is restricted and

Realised predicatively dominated by the organisations in the districts with a foreign bank. Both these

Arrangements suppose that organisations in India scrounge to banks located closer to their registered addresses and along with it they are put up in the same district. In a broad spectrum it is to be expected that this shows as a empirical work concerning lending association in various countries has shown up the mean distance between their office and their bank is normally pretty short. Yet, even when this supposition does not hold true completely. It could be only a bias approach towards the results alongside figuring out the impact of foreign bank entrance. Even if the organisation asks for advances from a different district then organisations out up outside those eight interacted districts might be actually "interacted" bring about the estimates to know the real effects. Even further, the capacity to borrow from banks from other districts may only lessen the neighbouring impact of foreign bank entrance hence making it tricky to discriminate an effect of foreign bank entrance at the local plane.

In order to calculate organisations accessibility to advances, a few number of dependant variables, y, shall be put to use. Primarily, to check the impact on the value of advances of the organisations reports, we will need to use three various variables namely: the stash of advances from the local development banks, the stash of long-term commercial bank advances and the stash of advances from both commercial and development banks which would mean total of long-term advances. These above mentioned ways are standardised by organisations total assets as they were in the earlier stages of the samples in ’91. Next, a collection of indicators equalizing one of organisations with advances from the given resources are considered to check the variations in the probability of a organisation having a loan. Meanwhile the two sets of fiscal measures are alike in character, their discrimination is important. The primary set will detain whether organisations in districts gaining a foreign bank know-how a comparative variation in amount of financing they gain, while the next set of predictors will check whether an organisations chances of having a loan is impacted by entrance of an foreign bank.

ESTIMATES OF OLS:

OLS estimate of the correspondence among long-term advances and the incidence of a foreign bank are stated in Table 3. Columns1-4 state the coefficients by means of and predictor for having a long-term advance as dependant variable, and columns5-6 state the coefficients whilst dependant variable in the stash of aggregated long-term advance standardised by assets.

Instead of being a help to the local organisations, foreign bank entrance is closely linked with a decrease in local firms possibilities of borrowing long-term advances that is discrete to organisations previous profits. In the regular regression in all organisations and no add on control or interactions – column 1. It is been seen that the foreign banks entrance coincides to a 7.5% percentile drop in organisations possibility of borrowing a long-term advance comparative to organisations put up in districts not having a foreign bank. Insertion of ROA intersects in column-2 reveals that the fall is on a bigger scale not related to organisations ROA, and even if there is anything which exists here is a fragile proof a organisations ROA is relatively of less importance pursuing foreign bank entrance. The decrease in long-term loans is vigorous to the addition of industry-year contact, column-3 and prohibiting the ultimately "interact" control group.

There are facts, moreover that foreign bank entrance is closely linked with an elevation in the comparative importance of organisations ROA in the value of long-term financing delegated. In the regressions where in usage of stash of long-term advances standardises by assets as the dependant variable, a privileged ROA matches to an elevation in the advance to asset ratio for organisations in the local districts with a foreign bank comparative to an organisation sans a foreign bank. The scale of the coefficients in column-6 entail that a one standard deviation elevates in organisations ROA is linked with an increase in their advances to asset ratio which on an average 1/10th standard deviations bigger when a foreign bank is existing in their district. Moreover, there is no noteworthy proof of an average decline in loan sizes. The advance-asset ratio being a louder way of credit relative to the predictors might explain so as to why we found a decline in average number of advances but not average size of advances.

In general, the findings and assumptions of Table 3 are constant with the theoretical structure of Gormley(2006) where in foreign entrance is linked with a restructuring of credit that doesn’t inevitably be beneficial to all borrowers. We also saw that more of profitable organisations saw a rise in their comparative value of loans, while other organisations saw a decline in their possibility of borrowing a long-term loan of any kind. In spite of all the findings the question remains as it is.

Moreover, as to whether this restructuring is efficient. In this prospect, the decline in advances doesn’t come into view to be determined by a turn down in advances owed to only the most unfeasible, politically connected organisations. This should give away a optimistic coefficient for the marginal effect of a organisations ROA, not a pessimistic coefficient as found in columns1-4 of Table 3. As an alternative, the decline in credit seems to be non comparative to organisations potential and even if has something, the relation is contradictory to what one may expect to be. The decline in the number of organisations holding a long-term loan is comparatively bigger. As seen in table 2,

Almost 88% of all the organisations had a long-term advance in ’93. By year ’02, though this declined to around 78% as there was a normal decrease in the long-term advances given by India’s development Banks. The decline however, was majorly in districts with newer foreign banks as shown in Fig. 2 and the instance of time coincides with the normal expansion of foreign banks from ’94-’02. Extra robustness confirms on the instance of time of this decline within every district provided in section 5.

If few organistions get improved financial services after entrance, a part of the decrease in advances may also be demand driven if locally domestic organisations in the same industries counter larger competition in their yield market. Known that these are comparatively larger firms, yet such local variations are not likely to make an impact on the total demand of their yield, and a robustness test in section 5.3 also says that the decrease in advances is not related to demand.

In order to understand where the variations in advance allotment are from, the regressions are now

Carried out separately for advances from banks and financial institutes. Yet again, the regressions pertaining to bank advances will substitute for the "direct" impact of new foreign bank advances, while the financial institutes advance regressions will entail the "indirect" impact of foreign bank entrance on the local advances from India’s Development Bank. The regressions for banks and financial institutes are stated in tables 4-5. In table 4, we can see that the decline in possibilities of having a long-term advance is determined entirely by a down turn in financial institutes advances column5-8 than advances from commercial banks columns. This entails that competition from foreign banks indirectly impacts the allotment of credit by India’s local domestic banks. Yet, an organisations ROA doesn’t seem to show any sort of effect on whether it is not as much of likely to gain an financial institute loan. Hence, there doesn’t seem to be any proof to carry the hypothesis that local lenders react optimistically by accepting new broadcasted technologies and improvising their credit allotment. Rather, national development banks react to the clash from foreign banks by methodically deducting the number of local organisations they expand long-term advances to, in spite of their potential, and this decline in advances from the national banks is not equalizing by an elevation in advances from foreign banks. In table 5 we can see that comparative rise in importance of organisations ROA for the value of loan is principally by an rise in slope of bank advances considerably than financial institute advances. This proposes a rise in advances to more profitable organisations is driven by newer advances from the foreign banks than national banks. For the reason that commercial bank advances are also added in the measure of "bank" advances, although, the rise in comparative important of ROA might also be forced by new national commercial bank advances. This may happen that national commercial bank’s enhanced efficiency following foreign bank entrance. Though if this holds true, we shall also expect to explore an alike enhancement in efficiency for national development banks, however as seen in table 4. There isn’t a proof. Furthermore, the restructure of bank advances seems to be driven by lending trends constant with the "cream skimming" nature and aiming of lesser informational-dense organisations that is normally related with foreign banks more so over than national banks. This fact is stated in section 4.2 and 6.1

Opposing to a normal reallocation of credit from fewer to more profitable organisations, the optimistic interactions in table 5 seems to be an impact of an elevation in bank advance to only top most percentile of organisations. But in regards to ROA. It is rather easier to see this in figure3, which splits down the drift in bank advances or assets of organisations within the time frame of ’91-’02 build on their ROA from ’91-’93. As we can see in figure 3, the drift in advances to the more profitable 10% of organisations was comparatively plane from ’91-’95. Commencing in ’96, nevertheless there is a huge expansion in advances for the top 10% in districts with a foreign bank, and the top 10% in other districts did not illustrate and rise in advances. The rise is advances to profitable organisations is prohibited to the top 10%, even so, there is no proof that foreign bank entrance is linked with an rise in advances amounts for organisations with an ROA more than the median but not also in the top 10% of organisation with an ROA lesser than the median.

The calculations in table 6 ensure that the optimistic interface on advance sizes is due to predominately because of the top 10% of organisations. In coluns1-3, the top 10% of organisations in regards of ROA have fallen from the regressions. The optimistic impact on the size of advances to more profitable organisations observed in table 5 is now entirely gone, following the judgement that the previous results were firstly driven by a rise in bank advances to very profit making organisations. The rise in advances also was found to be primarily due to only the biggest organisations. Dipping organisations with assets in ’91 greater than the median, as shown in columns 4-6 also abolishes rise in advances. These observations hold the theory that information asymmetries are normally tricky for foreign banks to beat leading them to only provide finance to the most profit making and biggest organisations in the economy.

Interpretation of the OLS estimate:

In general, the OLS estimates are confirming of models incorporating asymmetric information. The rise in advances to the most profit making 10% of organisations is indicative that these organisations were under-financed in the closed economy, and foreign bank entrance improvised the allotment of credit by aiming more advances to these organisations. Yet, the rise is advances seems to be restricted to small division of profit making organisations, which is constant with theories that asymmetric information persuades "cream skimming" nature by foreign banks and a section credit market (Dell’Ariccia and Marquez , 2004; Sengupta, 2006; Gormley, 2006). Besides, competition from foreign banks also looks to guide to a symmetric deduction in long-term by the national development banks is not counteract by a equivalent rise in advances from foreign banks which is steady with models that indicate the sectioned market may unfavourably affect some organisations (Gormley, 2006). While a portion of this decline in credit might be the result of an efficient deduction in advances to very not profit making or politically-connected organisations, the reach of the drop and its no relation to a organisations previous profits is indicative that few viable national organisations were also unlikely to gain a loan after entrance. It is tricky to discern precisely where the national capital has gone. One probable illustration is that financial institutes rose less capital on outside markets via the commercial papers, issuance of bonds, etc. Even though the data is only accessible earlier in ’96, the actual value of outside capital raised by development banks was steady from ’96-’98 and down turn thereafter. Another possible illustration is that bank capital was forwarded somewhere else. In ’92-‘93, twenty four percent of bank deposits were hypothecated as govt. Securities, but in ’94-’98 it rose to twenty nine percent, beyond the statutory requisite of twenty five percent. Data restrictions, nonetheless, did not permit to check whether either of these variations were pumped by development banks put up in districts with newer foreign banks. These results have suggested for financial policy in lower development countries, which in current years has rousingly

Drifted towards the allotment of bigger foreign bank entrances. While the possivle gains of foreign bank entrance have inference of foreign bank entrance are many, the proof implicates that information asymmetries may curb a lot of firms in these economies from getting to know these benefits. This result equates an active literature that tests the relative disadvantage of big banks in the production and usage of soft information(Berger, Miller, Petersen, Rajan, and Stein, 2005), and the unexpected results that bigger competition might have on the lending relationships that small and medium sized entrepreneurs hold on (boot and thakor 2000; Peterson and rajan 1995). On a whole this proof indicates that it might be necessary to accept extra policies exceeding permitting foreign banks entrance- to improve efficiency and to improvise credit accessibility in developing countries. Exceeding the existence of asymmetric information, it is likely that the local institutes and disappearance of banking competition in India preceding the foreign entry is partly accountable for the seen restructures of credit, preceding 10 ’91, India’s national development banks repeatedly engaged in govt. Directed lending schemes and were protected from competing via regulatory prohibitions to entrance. Meanwhile the financial reforms in 1991 on a great extent liberalised the national banking sector years beforehand the real entry of foreign banks into different Indian districts, it is likely the development banks are differently influenced by foreign bank entry owing to their history and immaturity at successfully viewing possible clients. Though, it is worth taking a note that alike directed lending schemes and regulatory prohibitions are very common in developing countries that permit larger foreign bank entrance as a means of rising banking competition. Hence, the understanding of India’s development banks might not be all that exceptional.

Verifying the Robustness and four estimates

Although the previous regressions are indicative of predictions that extra competition from foreign banks will boost a restructure of credit when information asymmetries are great, one may be thoughtful about a possible choice bias in the ols estimates. As foreign banks endogenously selected where to establish new branches in India, it is likely the foreign banks chose into districts that were either pre-trending uniquely in bank of financial institutes advances or were leading to trend uniquely in the future motivation other than the entry of the foreign banks.

Checking the pre-trends

To check for a prevailing trend in bank and financial institute loans, two variables are summed up to the key regressions of tables3-4. The primary variable is an predictor, fake, which turns optimistic in the 3 years preceding the entry of a foreign bank. Illustration, in Ludhiana where the primary foreign bank reached in ’01, this variable equalises to 1 in years ’98-’00 and 0 all other years. Secondly variable newly added is the interface fake x ROA. Findings of this illustration can be found in table 7.

But if the most profit making organisations have previously seen a rise in their bank advances in three years preceding to foreign bank’s onset, then we can figure out a optimistic coefficient for the interface term, fake X ROA. Further, if national organisations in the foreign bank districts have previously showing a decline in their accessibility to development bank advances prior the foreign bank prior the foreign bank entrance must find a depressing coefficient for fake in the regressions making use of an predictor for financial institutes as a dependant variable. Although, in table 7, the rise in bank advances values to most profit making organisations and the decline in financial institute advances to most profit making in the three years prior foreign bank entrance. In none of the case we can ignore the null hypothesis that the direct estimate for fake x ROA is 0. There is no such proof that foreign banks chosen into districts with pre-established differential patterns in bank of financial institute loans.

Henceforth, the rise in bank advances to most profit making organisations and the decline in financial institutes advances to appear 1 or 2 years preceding foreign bank entrance within every district. Fig 4 board A plots the direct estimates from OLS regression of bank advances onto predictors for years

To be relevant to foreign bank entrance for organisations with a ROA in the top 10%. As seen in fig 4, there is no such proof of a rise in bank advances in the years prior foreign bank entrance or in the year itself of the entrance. Nonetheless, 1 year preceding entrance, bank advances rose, and the rise became and stayed important at 5% level initiating 2 years after foreign bank entrance. Fig.4 board B plots the direct estimates from a alike regression making use of all organisations and an predictor for financial institutes loans as a dependant variable. Yet again, the direct estimates show the down turn in financial advances began exactly 1 or 2 years after primary entrance.

The four estimates

Although, we still concern that ofreign banks may have chosen districts that were

Leaving to trend unlikely in the future for basis unrelated to the exact entrance. A re-evaluation of press releases of the foreign banks put up newer branches in India during late ‘90s shows newer branch points in India were selected to decline the remoteness to established borrowers and to put up a presence in high-growth cities. Installation of newer branch in Surat in ’04 Citigroup country office for India said : we’re very happy to shift nearer to the customers. As the map points selection showed basis on prior clients it is indifferently to pose an recognition problem, the assortment into high-growth districts may also cause a pessimistic bias if the constant rise of new industries in those districts intersects with slow growth rate of organisations in old and more firm industries. To consider this possible recognition problem, the pre-’94 occurrence of foreign organisations is to be used as an measure for the location of newer foreign banks. I presume that foreign banks are further probable to enter districts with organisations form their home land to hold pre-established relations or to take benefit of their competitive benefits in acquiring information regarding the organisations in their home land. This behaviour of foreign banks to guide their clients outside the country is been noticed in a various number of countries and seems to happen in India too. In the sample data, there have been 52 foreign-owned organisations scattered across twenty six of the one sixty two districts. 5 of 8 districts gaining their 1st foreign bank in ‘90s had a foreign-owned organisation already existing in ’93.

In order to check the relation between the map points of foreign bank and foreign-owned organisation, the below mentioned 1st stage regression is put to use:

Foreign Bank = const + Foreign organisationd11993 × Post-1993 +α +δ +εi,d, t, .... (3)

The measure for foreign bank is the intersection in a district panel predictor variable for having a foreign-owned organisation established in ’93, foreign organisation, and a post ’93 year predictor.

The firm-level regression in organisation and time dummies, and SE are concentrated at the district panel. The consequences of the 1st stage are stated in table 8. And as it can be shown the occurrence of a foreign owned organisation in ’93 is a optimistic and important indicator of a foreign bank being established in the years ’94-’00. The findings show an occurrence of foreign-owned organisation in ’93 rose a districts possibility of gaining a foreign bank after ’94 to say about thirty four percent points comparative to districts who doesn’t have had a foreign-owned organisation. So as to see if the measurement is applicable, nonetheless, the location of foreign organisation, itself could be uncorrelated with the borrowing pattern of national firms. Whilst the unique location selection of foreign organisations might also be a strategy. This suppositions show reasonable that the median year of establishment of foreign firms used in the data is ’74. Almost twenty years preceding to liberalization in mid ‘90s. Henceforth, the location of foreign-owned organisations is unlikely to be straightly correlated to national lending trend in mid and late ‘90s. And the location selection is prohibited to foreign organisations setup 10 years before India’s liberalization in ’94. By using only those elder foreign organisations might in addition restrict endogeneity concerns in relation to the location selection of current setup foreign organisations.

Henceforth, we don’t have any proof that national firms in districts with a foreign organisation are trending unlikely to their usage of long-term advances in 5 years before signing of GATS in ’94. This may happen due to the occurrence of the foreign organisations innovating spill over impacts who indirectly affected national borrowing trend and their growth. Still, there is no such proof so as to say that the districts with foreign organisations were unlikely different with regards to national loan advancing trend before the entrance of foreign banks. It is observed in fig5, which boards the % of organisations with an financial institute advance and the mean bank advance to asset ratio of organisations between ’89-’02. Bank advances seem to be similarly down trending in 5 years before the foreign bank entrance in districts with and sand a foreign firm in ’93 and no. Of organisations with financial institute loans were upward-trend in both districts from ’89-93. So, to make the measurement to infringe this prohibiting restriction, one should imagine story of a foreign organisations occurrence in India may just happen to pump a direct variation in national bank advances at the time of foreign bank entrance in ever district, yet not prior it. Stated here doesn’t seems to be any other variations in govt. Policies that both rose the significance of foreign organisations and intersects with the timing of foreign bank entrance in each special district, the measure seems to be valid here.

Now as the measure seems to be satisfying the recognition suppositions we would now like to move further to the four estimates of eq (1). The intersection Foreign organisation x post-93 was used as a measure for the location of foreign banks, Foreign bank, and the interaction Foreign organisation x post-93 X ROA was used to measure Foreign bank X ROA.

I now proceed to the

IV estimates of equation (2). The interaction Foreign Firm × Post -1993 is used to instrument for the

location of foreign banks, Foreign Bank , and the interaction Foreign Firm × Post -1993×ROA is used to

instrument for Foreign Bank ×ROA . The IV estimates are reported in Table 9.

The IV estimates confirm the OLS estimates. The arrival of a foreign bank is still associated

with a drop in the average firm’s likelihood of receiving a long-term loan [Table 9, column (3)]. The

IV estimates suggest foreign bank entry is associated with a 12.4 percentage point reduction in firms’

likelihood of having either a bank or FI loan, which is larger than OLS estimate of 7.6 percentage

points. Moreover, foreign bank entry is still associated with a positive and significant increase in the

marginal importance of ROA for bank loan sizes [Table 9, column (5)], and the magnitude of the

effect is similar to the OLS estimate. While not shown, the IV estimates are also robust to including

industry-year interactions as done in some of the OLS specifications.

5.3 Additional robustness checks

Overall, both the drop in firms’ likelihood of having a long-term loan and the increase in the

Table 1

District and year wise no. Of foreign banks in India.

Number of foreign bank branches calculated using the Directory of Bank Offices . Bank numbers represent total branches as of

March 31 for each year.

District Name

State Name

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

Districts with Pre-Existing Foreign Bank Branches

Amritsar

Punjab

3

3

3

3

3

3

3

3

3

2

1

1

1

Bangalore Urban

Kanrataka

2

2

2

3

3

5

6

7

7

10

11

11

12

Coimbatore

Tamil Nadu

1

1

1

1

1

1

1

1

2

2

2

3

4

Darjiling

West Bengal

1

1

1

1

1

1

1

1

1

1

1

1

1

Delhi

Delhi

22

23

24

24

26

28

28

31

35

36

37

38

37

Ernakulam

Kerala

3

3

3

3

4

4

4

4

4

3

3

3

4

Greater Mumbai

Maharashtra

51

52

52

51

51

55

58

63

65

63

64

64

63

Haora

West Bengal

2

2

2

2

2

2

2

2

2

2

2

2

2

Hyderabad

Andhra Pradesh

1

1

1

1

1

2

2

2

2

4

6

8

8

Kamrup

Assam

1

1

1

1

1

1

1

1

1

1

1

1

1

Kanpur City

Uttar Pradesh

3

3

3

3

3

3

3

3

3

3

3

3

3

Kolkata

West Bengal

43

43

42

42

42

42

42

43

43

34

34

34

34

Kozhikode

Kerala

1

1

1

1

1

1

1

1

1

1

Chennai

Tamil Nadu

11

11

11

12

12

12

14

15

16

16

16

16

16

Simla

Himachal Pradesh

1

1

1

1

1

1

1

1

1

1

1

1

1

South Goa

Goa

1

1

1

1

1

1

1

1

1

1

Srinagar

Jammu & Kashmir

1

1

1

1

1

1

1

1

1

1

1

1

1

Vishakhapatnam

Andhra Pradesh

1

1

1

1

1

1

1

1

1

1

1

1

1

Thiruvananthapuram

Kerala

1

1

1

1

1

1

1

1

1

Ahmedabad

Gujarat

2

2

3

3

5

5

8

8

Pune

Maharashtra

1

1

4

5

5

5

6

Chandigarh

Chandigarh

1

1

1

1

2

2

Gurgaon

Haryana

1

1

1

2

Vadodara

Gujarat

1

1

2

2

Jaipur

Rajasthan

1

1

Ludhiana

Punjab

1

1

Total Foreign Bank B

ranches

149

151

151

152

156

167

174

187

198

196

198

209

212

Districts Receiving First Foreign Bank

Table 2

District wise summary statistics, Using 1993 Data

.

Districts with

Districts with No Foreign Bank in 1991

Pre-Existing

Foreign Banks

Foreign Bank by

2002

No Foreign

Bank by 2002

(1)

(2)

(3)

Firm Characteristics

Total Assets (10 mn. Rp.)

511.78

229.21

259.12

1991-1993 Average ROA (%)

2.48

3.75

2.07

Short-Term Bank Credit / Total Borrowings

0.380

0.344

0.350

Long-Term Bank & FI Loans / Total Borrowings

0.298

0.337

0.373

Short-Term Bank Credit / Assets

0.148

0.123

0.148

Long-Term Bank Loans / Assets

0.041

0.023

0.034

FI Loans / Assets

0.106

0.118

0.168

% of Firms with Long-Term Loan

80.2

87.5

88.1

% Firms with Bank Loan

42.2

43.1

44.1

% Firms with FI Loan

69.3

80.6

81.4

District Banking & Population Characteristics

Population / Km2

6591

1228

476

Total Banks / Million People

135

118

72

% Share of Private Banks

11.32

6.13

6.13

Number of Districts

14

8

154

Number of Firms

1047

156

500

Table 3

Impact of foreing bank entry on entrance on aggregate long-term advances.

. Standard errors, clustered at the district-level, are reported in parentheses. * = 10% level, ** = 5% level, *** = 1% level.

Dependent Variable =

Indicator for Long-Term Loan Long-Term Loans / 1991 Assets

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

Foreign Bank

-0.076***

-0.075***

-0.080***

-0.077**

-0.049

-0.078

-0.181

-0.046

(0.028)

(0.028)

(0.028)

(0.033)

(0.150)

(0.141)

(0.249)

(0.106)

Foreign Bank * ROA

-0.003*

-0.003

-0.002

0.011*

0.011

0.019***

(0.002)

(0.002)

(0.002)

(0.006)

(0.006)

(0.005)

Observations

7088

7088

7088

2617

7088

7088

7088

2617

R-squared

0.55

0.56

0.67

0.56

0.50

0.51

0.61

0.56

Number of Districts

162

162

162

17

162

162

162

17

ROA-Year Interactions

4-Digit Industry-Year Interactions

-

---

-

-

- -

-

"Treated" Control Group Used

-

-

Table 4

Impact of foreign Bank entrance on accessibility to banks and financial institutes advances.

Standard errors, clustered at the district-level, are reported in parentheses. * = 10% level, ** = 5% level, *** = 1% level.

Dependent Variable =

Indicator for Bank advances Indicator for Finc. Inst. advances

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

Foreign Bank

0.008

0.009

-0.014

0.038

-0.084**

-0.087**

-0.065*

-0.073**

(0.038)

(0.039)

(0.035)

(0.038)

(0.040)

(0.039)

(0.037)

(0.035)

Foreign Bank * ROA

-0.003*

-0.002

-0.003

-0.001

0.000

-0.001

(0.002)

(0.002)

(0.002)

(0.001)

(0.002)

(0.001)

Observations

7088

7088

7088

2617

7088

7088

7088

2617

R-squared

0.46

0.47

0.56

0.45

0.64

0.65

0.72

0.65

Number of Districts

162

162

162

17

162

162

162

17

ROA-Year Interactions

4-Digit Industry-Year Interactions

-

- -

-

-

- -

-

"Treated" Control Group Used

-

-

Table 5

Impact of foreign bank entrance on size of bank and financial institute advances.

Standard errors, clustered at the district-level, are reported in parentheses. * = 10% level, ** = 5% level, *** = 1% level.

Dependent Variable =

Bank advances / 1991 Assets FI advances / 1991 Assets

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

Foreign Bank

0.041

0.029

0.030

0.035

-0.089

-0.108

-0.211

-0.081

(0.069)

(0.063)

(0.070)

(0.046)

(0.093)

(0.090)

(0.202)

(0.081)

Foreign Bank * ROA

0.006**

0.007***

0.007***

0.005

0.004

0.012***

(0.002)

(0.003)

(0.002)

(0.006)

(0.007)

(0.004)

Observations

7088

7088

7088

2617

7088

7088

7088

2617

R-squared

0.42

0.43

0.51

0.53

0.50

0.50

0.61

0.54

Number of Districts

162

162

162

17

162

162

162

17

ROA-Year Interactions

4-Digit Industry-Year Interactions

-

- -

-

-

- -

-

"" Control Group Used

-

-

Table 6

Scope of Foreign Bank Entry Effect on Size of Bank Loans

Standard errors, clustered at the district-level, are reported in parentheses. * = 10% level, ** = 5% level, *** = 1% level.

Firms Dropped =

Dependent Variable = Bank Loans / 1991 Assets

ROA > 90th Percentile 1991 Assets > 50th Percentile

(1)

(2)

(3)

(4)

(5)

(6)

Foreign Bank

-0.012

0.008

0.005

-0.026

-0.081

-0.010

(0.029)

(0.024)

(0.020)

(0.061)

(0.074)

(0.049)

Foreign Bank * ROA

-0.001

-0.001

0.000

-0.002

-0.006

-0.002

(0.001)

(0.001)

(0.001)

(0.002)

(0.004)

(0.004)

Observations

6387

6387

2332

3412

3412

1233

R-squared

0.39

0.46

0.41

0.35

0.52

0.48

Number of Districts

162

162

17

162

162

17

4-Digit Industry-Year Interactions

"Treated" Control Group Used

-

-

-

-

Table 7

Pre-Trend Falsification Test

Standard errors, clustered at the district-level, are reported in parentheses. * = 10% level, ** = 5% level, *** = 1% level.

Dependent Variable =

Bank advances / 1991 Assets Indicator for Finc inst. advances

(5)

(6)

(7)

(8)

(1)

(2)

(3)

(4)

Fake

-0.028

-0.026

-0.050

-0.034

-0.006

-0.004

-0.018

0.004

(0.021)

(0.019)

(0.032)

(0.038)

(0.026)

(0.025)

(0.030)

(0.035)

Foreign Bank

0.023

0.012

-0.002

0.008

-0.088*

-0.090*

-0.077

-0.069

(0.060)

(0.054)

(0.057)

(0.033)

(0.046)

(0.046)

(0.052)

(0.051)

Fake * ROA

-0.001

0.001

-0.000

-0.001

-0.001

-0.001

(0.001)

(0.001)

(0.002)

(0.003)

(0.003)

(0.003)

Foreign Bank * ROA

0.005**

0.008**

0.006***

-0.001

-0.000

-0.002

(0.003)

(0.003)

(0.002)

(0.003)

(0.003)

(0.003)

Observations

7088

7088

7088

2617

7088

7088

7088

2617

R-squared

0.42

0.43

0.50

0.53

0.64

0.65

0.72

0.65

Number of Districts

162

162

162

17

162

162

162

17

ROA-Year Interactions

4-Digit Industry-Year Interactions

-

- -

-

-

---

-

"Treated" Control Group Used

-

-

Table 8

First Stage Regression

Yearly observations from ‘91 to ‘02 are included for firms with positive sales and assets in ‘91 but not situated in a district with a foreign bank by

1991. Standard errors, cconcentrated at the district-level, are reported in parentheses. * =

10% level, ** = 5% level, *** = 1% level.

Dependent Variable =

'Foreign Bank'

Foreign-Owned Firms in 1993 * Post-1993 0.341*** (0.131)

Observations 7088

R-squared 0.65

Table 9

Instrumental Variable Estimates of Foreign Bank Entry

Standard errors, concentrated at the district- level, are reported in parentheses. * = 10% level, ** = 5% level, *** = 1% level.

Dependent Variable =

FI advance

Bank

advance

Either

advance

FI advance

Bank

advance

Both

advancee

(1)

(2)

(3)

(4)

(5)

(6)

Foreign Bank

-0.131

-0.124

-0.124*

-0.346

0.011

-0.336

(0.089)

(0.137)

(0.068)

(0.279)

(0.113)

(0.360)

Foreign Bank * ROA

0.000

-0.001

-0.002

0.019

0.010***

0.030

(0.004)

(0.005)

(0.003)

(0.017)

(0.004)

(0.019)

Observations

7088

7088

7088

7088

7088

7088

R-squared

0.65

0.46

0.56

0.50

0.43

0.50

Number of Districts

162

162

162

162

162

162

Indicator for value / 1991 Assets for

Table 10

Access to FI Loans for Group versus Non-Group Firms

Measurement used in the first stage are ‘Foreign-Owned Firm in 1993’ * post-1993 year dummy and ‘Foreign-Owned Firm in 1993’ * ‘ROA’ * post-1993 year dummy. Standard errors, concentrated at the district-level, are reported in parentheses. * =

10% level, ** = 5% level, *** = 1% level.

Dependent Variable = Indicator for FI advance

Group organisation Non-Group organisation

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

Foreign Bank -0.105*** -0.107***

-0.089**

-0.080*** -0.233***

-0.034

-0.021

0.038

-0.026

0.131

(0.038) (0.037)

(0.043)

(0.027) (0.084)

(0.060)

(0.060)

(0.061)

(0.063)

(0.181)

Foreign Bank * ROA 0.003 0.004 0.003 0.007 -0.006*** -0.006*** -0.007*** -0.005

(0.004)

(0.004)

(0.004)

(0.006)

(0.001)

(0.002)

(0.001)

(0.006)

Observations

4140

4140

4140

1673

4140

2948

2948

2948

944

2948

R-squared

0.58

0.59

0.68

0.58

0.58

0.69

0.69

0.81

0.72

0.69

Number of Districts

121

121

121

17

121

115

115

115

16

115

ROA-Year Interactions

4-Digit Industry-Year Interactions

-

- -

-

-

-

- -

-

-

"Treated" Control Group Used

-

-

Specification

OLS

OLS

OLS

OLS

IV

OLS

OLS

OLS

OLS

IV

Table 11

Access to Bank Loans for Group versus Non-Group Firms

Standard errors, concentrated at the district-level, are reported in parentheses.

* = 10% level, ** = 5% level, *** = 1% level.

Dependent Variable = Indicator for Bank advances

Group organisations Non-Group organisations

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

(10)

Foreign Bank

0.015

0.015

-0.026

0.058

-0.075

-0.008

-0.026

0.077

-0.019

-0.279

(0.051)

(0.056)

(0.060)

(0.060)

(0.154)

(0.054)

(0.049)

(0.110)

(0.053)

(0.268)

Foreign Bank * ROA

-0.011***

-0.009**

-0.010**

-0.012*

0.005**

0.010**

0.004*

0.008

(0.004)

(0.004)

(0.004)

(0.006)

(0.002)

(0.004)

(0.002)

(0.007)

Observations

4140

4140

4140

1673

4140

2948

2948

2948

944

2948

R-squared

0.45

0.45

0.58

0.43

0.45

0.46

0.47

0.64

0.49

0.46

Number of Districts

121

121

121

17

121

115

115

115

16

115

ROA-Year Interactions

4-Digit Industry-Year Interactions

-

- -

-

-

-

---

-

-

"Treated" Control Group Used

-

-

Specification

OLS

OLS

OLS

OLS

IV

OLS

OLS

OLS

OLS

IV

Table 12

Effect of Foreign Bank Entry on Bankruptcy Rates and Sales

This table reports coefficients from regressions using OLS and IV with firm and year fixed effects. The dependent variable is an indicator equal to 1 if the firm has declared bankruptcy with the Board for Industrial and Financial Reconstruction in columns (1)-(5) and the log of total sales in columns (6)-(10). Yearly observations from 1991 to 2002 are included for domestic, non-financial firms with positive sales and assets in 1991 but not located in a district with a foreign bank by 1991. ‘Foreign Bank’ is equal to one for firms located in a district with a foreign bank in the given year, and zero otherwise. ‘ROA’ is a firm’s 1991-

1993 average percent return on assets, demeaned. Columns (3) & (8) include 4-digit industry-year interactions. Columns (4) & (9) restrict the sample to

‘treated’ firms located in districts with a foreign bank by 2004. Columns (5) & (10) report the IV estimates. Instruments used in the first stage are ‘Foreign- Owned Firm in 1993’ * post-1993 year dummy and ‘Foreign-Owned Firm in 1993’ * ‘ROA’ * post-1993 year dummy. Standard errors, clustered at the district- level, are reported in parentheses. * = 10% level, ** = 5% level, *** = 1% level.

Dependent Variable =

Bankruptcy Log(Sales)

(1) (2) (3) (4) (5) (6) (7) (8) (9) (10)

Foreign Bank -0.0264 -0.0244 -0.0097 -0.0001 -0.0763 0.046 0.002 -0.114 -0.128 -0.444 (0.0243) (0.0252) (0.0201) (0.0285) (0.0504) (0.101) (0.110) (0.117) (0.118) (0.329)

Foreign Bank * ROA -0.0007 0.0000 -0.0011 0.0016 0.025** 0.030** 0.038*** 0.061** (0.0016) (0.0011) (0.0015) (0.0017) (0.012) (0.012) (0.012) (0.023)

Observations 7872 7872