6

Assets as Liability? :

NPAs in the Commercial Banks of India

A

Research Project Funded By

South Asia Network of Economic Research Institutes

Meenakshi Rajeev Institute for Social and Economic Change

Nagarbhavi, Bangalore-560072, India

Ph: 080-23215468

Fax:080-23217008

Email: [email protected]

January, 2008

Non-Performing Assets in the Indian Banking Sector

(with Special Reference to the Small Industries Sector)

By

Meenakshi Rajeev

Institute for Social and Economic Change

Nagarbhavi, Bangalore-560072, India

Ph: 080-23215468

Fax:080-23217008

Email: [email protected]

186

CHAPTER 1 INTRODUCTION

1.1 Introduction

Financial resource is, no doubt, one of the most important inputs for economic

development. Higher levels of financial development are significantly and robustly

correlated with faster current and future rates of economic growth, physical capital

accumulation and economic efficiency improvements (King and Levine, 1993a). The

relationship between financial development and the economic growth has been

established by various empirical studies ( see Adelman and Morris, 1968 and Goldsmith,

1969). It has been observed historically that banks formed the major part of financial

system and thus played an important role in economic development. In India also

financial system has been synonymous with banking sector. The importance of banking

system in India is noted by the fact that the aggregate deposits stood at 55 percent of

GDP and bank credit to government and commercial sector stood at 26 percent and 33

percent of GDP respectively in 2004-05.

In the earlier stages of development, banking credit was directed towards selected

activities only. For example, in the decade of 1960s, more than 80% of credit was to the

trade and industry sector whereas agriculture and small manufacturing sectors were

completely neglected. In 1969 nationalization of banks took place. At the time of

nationalisation of private sector banks in 1969, the prime concern was to use resources

available with these institutions for supporting the growth of priority sectors, viz.

agriculture, small and village industries, artisans, etc., besides expanding the outreach to

the poor. The assumption then was that credit could ensure faster growth. The planners

adopted a supply side approach, which was possibly the need of the hour. The policy

environment created at that time was primarily to pursue the agenda of social banking.

187

The regulatory framework mainly focused on regulation of interest rate, preemption of

resources, restricted investment avenue and expansion of banking network in backward

areas. With multi-agency approach, banking network in the rural areas has made its

formidable presence in providing rural financial services. The loans provided by banks

have contributed substantially for the growth of all the priority sectors. Besides, the

banking facilities were made available in unimaginable remote areas for tapping the

latent savings of the rural masses. Though the volume of loans provided by banks has

increased substantially, the health of these institutions also took a beating with increased

thrust to financing under what is called ‘directed lending’ and by implementing various

government sponsored programs using banking as a channel of credit purveyor. At the

time of carrying out general economic reform in the country a need to initiate financial

sector reform was also been felt.

Since 1991, the Indian commercial banks have undergone the reform process

aimed at putting the Indian banking sector on par with international standards1.

Performance in terms of profitability has become the benchmark for the banking industry

like any business enterprise. In particular, due to the social banking motto of the

Government, the problem of non-performing asset (NPA) was not considered seriously in

India in the post nationalization (of banks) period. However, with the recent financial

sector liberalization drive, this issue has been taken up seriously by introducing various

prudential norms relating to income recognition, asset classification, provisioning for bad

assets and assigning risks to various kinds of assets of a bank. While the Reserve Bank of

India (RBI) as well as the banks have begun to pay considerable attention to the NPA

problem, there are only a limited number of rigorous studies in the Indian context that

look at this issue in some detail. In this project we attempt to look at the determinants of

NPA (using a panel data model with a cross section of over 100 banks) by examining

some of the external and internal factors like extent of competition, total assets of a bank,

size of operations, proportion of rural branches , investment, etc., that can influence

NPA. It is of our interest to examine, between various bank groups (viz, SBI,

1 One can see in this regard Charkavati Committee Report (1985), Narasimham Committee Report (1998), Basel I and II norms.

188

Nationalized banks, Private banks and foreign banks), which is the most efficient group

in the context of recovery of loans and what are the factors that determine this efficiency.

For determining efficiency of the different banks in their loan recovery effort, the concept

of technical efficiency will be applied, using a frontier production technique (Bettesse

and Coelli (1995)).

One of the important objectives of the nationalization of commercial banks in

1969 and in 1980 was to provide credit to till then neglected sector (what was later called

the priority sector). Since them lot of effort has been gone in to chanelising credit to

priority sector. One of the major components of the priority sector is the Small Scale

Industrial sector. It has potential to generate substantial employment and also contribute

in terms of production and exports. Unlike agriculture, there is no separate sub-target for

the SSI sector, within the priority sector lending for the Indian public and private sector

banks; and the share of credit to the SSI sector has been falling over the years in the post

reform era2. This is a matter of serious concern as availability of credit has been always

recognized as a constraint to the growth of the SSI sector, be it a women or a rural

enterprise. Government has so far tried to mitigate the problem through various measures.

However, one of the major concerns of banks is the problem of bad loans arising out of

such small and medium enterprises (SME) accounts.

While the problem of non-recovery of agricultural loan is a well-discussed issue

(Bardhan, 1989, Bell and Srinivasan, 1989), not many studies in India have focused on

the non-recovery of loans from the SSI sector. Most authors usually touched upon this

issue in passing amid other problems of the SSI sector. However, the recent figures show

that amongst different sub-sectors within the priority sector, SSI’s contribution is the

highest in total NPA of the priority sector lending (Table3). It is also worth noting in this

context that SSIs share in net bank credit went from 15.89 per cent in 1991 to 11.1 per

cent in 2003, charting a steady decline. The share of SSIs in total priority sector lending

(TPSL) decreased even more dramatically in a shorter span of time: it went down from

36.12 per cent to 26.1 percent. Thus, as expected, channeling the credit away from this

sector appears to be the solution adopted by the banks (EPW Editorial, June 5, 2004).

2 See RBI Bulletin different issues.

189

Thus in this research work it is of our interest to look at various aspects of the NPA

problem arising out of this segment for the Indian banking sector.

Given our interest in the commercial banks, to put the issues in perspective, we

first look at the development and the change that have taken place in the Indian banking

sector.

1.2 A Brief Review of Indian Financial System

The financial system of any country consists of specialised and non-specialised

financial institutions, organized and unorganized financial markets, financial instruments

and services, which facilitate transfer of funds. Procedures and practices adopted in the

markets and financial interrelationships are also part of financial system (Bhole, 1999). In

India, the financial system has undergone a significant change over time in terms of size,

diversity, sophistication and innovation. Now India has a well-developed financial system

with a variety of financial institutions, markets and instruments3. The structure of the

India’s financial system is illustrated in Fig.1.1.

Fig. 1.1 Structure of Indian Financial System

3 See Bhole (1999), Sen and Vaidya (1997) for a detailed discussion of India’s financial system.

190

∗ Reserve Bank of India (RBI) is the controller and supervisor of India's Financial System. Source: RBI

An important feature of India’s financial system (like any other developing

country) is that, until recently a financial institution has largely been synonymous with

banking4 (Table 1.1). At the time of independence India had a relatively weak economic

base and financial structure. Savings and investment were relatively low and only two

third of the economy was monetised5. And also flow of funds from outside was very

meager and savings from corporate sector were low. At this juncture, savings were

coming mainly from household sector. And banks played very important role in

transforming these savings to investment in industries and other infrastructure

development. The gross domestic saving which was 10% of GDP in 1950-51 increased to

15.7% in 1971-72 and to 25.6% in 1995-96 which further increased to 27.6% in 2001-02

and it was 32% of GDP in 2004-05.

Table 1.1 : Major Balance Sheet Components of Financial Institutions (2004) (Rs. In Crores)

4 Jadhav and Ajit (1996-97) pp 311 5 Lumas.P.S. (1990) pp 390

India’s Financial System*

Financial Markets

Capital Credit Money

Financial Institutions

Other Financial Institutions

Insurance Mutual Funds Development Banks

Commercial Co-operative

Banks

Public Private RRBs

Domestic Foreign

191

Banks NBFCs State Co-operative Banks

Central Co-operative Banks

Capital 22348 (66%) 6131 1012 4342

Total Assets 1975020(85%) 130142 71806 133331

Deposits 1575143 (91%) 17946 44316 82098

Loans and Advances 864143 (81%) 91613 37346 73091

Investment 802066(92%) 14298 23289 35830

Figures in bracket represent the percentage of the total. Source: Report on Trends and Progress of Banking in India 2003-04

1.3 Development of Indian Banking sector

Modern commercial banking made its beginning in India with the setting up of the

first Presidency bank, the Bank of Bengal, in Calcutta in 1806. Two other presidency banks

were set up in Bombay and Madras in 1840 and 1843 respectively. They were private

shareholders' banks. These banks were amalgamated into the Imperial Bank of India in

1921, which was nationalised into the State Bank of India (SBI) in 1955. The Reserve Bank

of India (RBI) was established in 1st April 1935 with the passing of the Reserve Bank of

India Act 1934.

Following Sen and Vaidya (1997) the evolution of Indian Banking sector in the post-

independence era can be divided into three distinct periods. (I) 1947-68 saw the

consolidation of Reserve Bank of India (RBI) in its role as the agency in charge of the

supervision and control of banks. In this period Indian banking sector operated in fairly

liberal environment. (II) The second period 1969-1991 marked its beginning with

nationalisation of commercial banks and their dominance in the financial system (III) the

third period starting with liberalisation (1992) of banking and financial sector.

After independence, the major development in the Indian banking sector was the

nationalisation of banks. The first to be nationalised was the Reserve Bank of India (RBI),

192

the country's central bank from 1st January 1949. Then came the take over of the (then)

Imperial Bank of India and its conversion into State Bank of India (SBI) in July 1955 and

the conversion of seven major state owned banks into subsidiary banks of the SBI in 1959.

In 1969, 14 major banks were nationalised and in April 1980 six more banks were

nationalised. After nationalisation public sector banks followed two important policies. (a)

Massive expansion of branches especially in rural and semi-urban areas and (b)

Diversification of credit to till then neglected sector (priority sector lending).

In the post nationalisation period there was a rapid expansion of banks in terms of coverage

and also of deposit moblisation. The number of bank offices multiplied rapidly from 8300 in

July 1969 to 59752 in 1990, which further increased to more than 62 thousand in 1995, and

it was 71177 in 2006. This has reduced the population served per bank branch. The number

of people served per bank branch reduced from 65 thousand in 1969 to 14 thousand in 1990

which, however has increased marginally to 16 thousand in 2006 when consolidation has

been in progress. Also the total deposit increased from Rs 4646 crores in 1969 to Rs 323632

crores in 1994 and to Rs 2109049 crores in 2006. Some of the major aggregates of Indian

commercial banks are presented in Table 1.2.

Table 1.2 Major Aggregates of Commercial Banks in India (Real Values)

1969 1990 1995 2000 2005 Number of Banks 89 274 282 298 289 Total Bank Branches 8262 59752 64234 67868 70373 Population Per Branch (thousands) 64 14 15 15 16 Deposit (Rs crore) 33025 235292 324246 536371 1124098 Credit (Rs crore) 25583 142994 177319 285993 727556 Total Investment (Rs crore) 9674 87578 125097 196320 488697 Credit to Priority Sector (Rs crore) 3583 56271 58008 98116 252216 Priority Sector Credit as Percent of Total Credit 14 40.7 33.7 35.4 36.7 Credit-Deposit Ratio 77.5 60.8 54.7 53.3 62.6 Source: Trends and progress of banking in India, RBI

193

The post-nationalisation period was also marked by the developmental role of the

banks. Government used banking sector as the instrument to finance its own deficit6. The

fiscal deficit to GDP ratio for the central government increased steadily from an average of

3.56% in the period 1971-76 to 8.29% in the period 1986-91. High Cash Reserve Ratio

(CRR) and Statutory Liquidity Ratio (SLR) are used in order to financed this. The CRR,

which was 3.5% in 1962-63, increased to 15% in 1989-90 and SLR, which was 25% in

1964-65 increased to 38.5% in 1990-91. Along with high CRR and SLR, the operational

freedom of the banks was curtailed with high priority sector lending of as high as 40% of the

total lending in 1989-90. To keep the borrowing cost of the Government low, the interest

rate on bank loan was fixed at lower than market rates. This affected profitability and the

efficiency of banks. Further, owing to the dominance of the public sector banks there was

no competition. Due to the expansionary policy followed by the RBI, the number of loss

making bank branches increased, especially in rural areas, which whittled away resources of

the banking industry. Due to all these factors, towards the end of 1980s banking industry

was badly in need of reforms.

In 1991, Indian economy faced a major balance of payment crisis. The foreign

exchange resources had almost disappeared. The fiscal deficit was high and the inflation rate

reached double digits. To overcome this crisis, Indian Government introduced many

economic reforms, which included amongst others financial sector reforms. As with general

reform private sector grew considerably and growth of the private sector made demands on

financial resources, there was a need to overhaul the financial system. Financial sector

reforms were introduced in 1992.

1.4 Financial Sector Reforms in India

The financial sector reforms in India began as early as 1985 itself with the

implementation of Chakravarti committee report. But the real momentum was given to it in

1992 with the implementation of recommendations of the Committee on Financial System

(CFS) (Narasimham, 1991). The important recommendations of the CFS were; (i)

Reduction in SLR (ii) reduction in CRR, payment of interest on CRR and use of CRR as the

monetary policy instrument (iii) phase out of directed credit (iv) deregulation of interest

6 Sen and Vaidya (1997) pp 15

194

rates in a phased manner and bringing interest rate on government borrowing in line with

market-determined rates (v) attainment of Basel norms for capital adequacy within three

years (vi) tightening of prudential norms (vii) entry of private banks and easing of restriction

on foreign banks (iix) sale of bank equity to public (ix) phasing out of development

institutions (x) Increased competition in lending between Development Financial

Institutions (DFI) and a switch from consortium lending to syndicate lending. (xi) easing of

regulation on capital markets combined with entry of foreign institutions.

Almost all of the recommendations of the CFS have been implemented in a phased

manner. In 1998 another committee, the committee on Banking Sector Reforms (BSR)

(Narasimham, 1998) was constituted. The recommendations of the BSR committee have

also been implemented in a phased manner. The important recommendations of the BSR

are:

1. A minimum target of 9% Capital Risk-Adequacy Ratio (CRAR) to be achieved by

the year 2000. The ratio should be raised to 10% for the year 2002.

2. A risk weight of 5% for market risk for government-approved securities should be

attached.

3. An asset to be classified as doubtful if it is in the category of 18 months in the first

instance and eventually for 12 months and loss if it has been so identified but not

written off.

4. Income recognition, asset classification should apply to government advances.

5. The minimum shareholding by government/RBI in the equity of nationalised banks

and SBI should be brought down from 51% to 33%.

Financial sector reforms can be broadly divided into reforms in financial institutions

and reforms in financial markets. Reforms in financial institutions are mostly related to

reforms in banking sector as the banking sector forms a very important part of financial

sector.

1.5 Reforms in Financial Institutions

195

Two main objectives of the financial sector reforms are to enhance the stability

and the efficiency of financial institutions7. To achieve these objectives various reform

measures were initiated which can be broadly grouped into three categories.

1. Enabling measures

2. Strengthening measures

3. Institutional measures

Reforms in the commercial banking sector had two distinct phases. The first

phase of reforms introduced subsequent to the release of the report of the CFS (1992)

focused mainly on enabling and strengthening measures. The second phase of reforms

introduced subsequent to the recommendation of the BSR (1998) committee report.

1) The enabling measures: These were designed to create an environment where

financial intermediaries could respond optimally to market signals on the basis of

commercial considerations. Salient among these include reduction in statutory pre-

emption so as to release greater funds for commercial lending, interest rate deregulation

to enable price discovery, greater operational autonomy to banks and liberalisation of the

entry norms for financial intermediaries.

(a) Reduction in statutory pre-emption: This includes reduction in CRR and SLR.

These are mainly used to finance the fiscal deficit of the government and are also

used as tools of credit control. At one stage CRR applicable to incremental deposit

was as high as 25% and SLR was 40% thus pre-empting 65% of incremental deposits.

These ratios were reduced in a series of steps after 1992. The SLR was 25% and CRR

was as low as 5.5% in 2002 and now less than 5% of the total deposit. However,

though the SLR has reduced to a much lower level, banks (especially public sector

banks) hold government securities more than prescribed level.

(b) Interest rate liberalisation: Before 1991, interest rates, both on deposits and loans

were controlled by RBI. With effect from October 1997 interest rates on all time

7 RBI (2001-02)

196

deposits have been freed. Only rates on saving deposits remain controlled by the RBI.

Similarly the lending rates were also freed in a series of steps. The RBI now directly

controls only interest rates charged on exports and also there is a ceiling on lending

rate on small loans up to Rs 2 lakhs. The rationale for liberalising interest rate in the

banking system was to allow banks greater flexibility and encourage competition.

(c) Increased autonomy and competition: Banks have been given more autonomy

by reducing government's stake in it. It was recognised that restoration of health of

the banking system was required. Restoration of net worth was achieved through

capital infusion from budget. Competition has been infused by allowing new private

sector banks and more liberal entry of foreign banks (at the end of march 2001, there

were 8 new private sector banks, 23 old private sector banks and 42 foreign banks as

against 23 foreign banks in 1991).

2) The strengthening measures: These (also called prudential norms) were aimed at

reducing the vulnerability of financial institutions in the face of fluctuations in the

economic environment. These include various prudential norms related to capital

adequacy and risk-weighted assets, income recognition, asset classification and

provisioning for bad assets (NPAs). Following the CFS report the capital adequacy ratio

was fixed at 8%. It was increased to 9% following the BSR recommendation. Financial

institutions are asked to assign a risk weight of 100% on those government guaranteed

securities, which are issued by defaulting entities. Further, due regard should be paid to

the record of particular government in housing its guarantees while processing any

further requests for loans to PSUs on the strength of that state governments' guarantees.

3) Institutional measures: These measures are aimed at creating an appropriate

institutional framework conducive to development and functioning of financial markets.

These measures include reforms in legal framework, particularly relating to banks. Banks

are allowed to close down loss making units and merging with other banks. Flexibility is

introduced in resource mobilisation. Financial institutions are not required to seek RBI's

approval for raising resources by way of bond/debentures (by public/private placement).

In order to have a coordinated approach in the recovery of large NPA accounts, as also

197

for institutionalising an arrangement for a systematic exchange of information in respect

of large borrowers (including defaulters and NPAs) common to banks and Financial

institutions, a standing committee was constituted in august 1999 under the aegis of

Industrial Development Bank of India (IDBI).

1.6 Non Performing Assets in India

One of the important issues that is drawing attention of policy makers and

researchers is the Non-Performing Assets of Commercial banks. High level of Non-

performing Assets (NPAs) is a concern to everyone involved as credit is very essential

for economic growth and NPAs affect the smooth flow of credit. Broadly, Non

Performing Advance is defined as an advance where payment of interest or repayment of

installment of principal (in case of Term Loans) or both remains unpaid for a certain

period8.

In India though the issue of NPAs was given more importance after the

Narasimham committee report (1991) highlighted its impact on the financial health of the

commercial banks and subsequently various asset classification norms were introduced,

the concept of classifying bank assets based on its quality began during 1985-86 itself

(see Chapter 3). A critical analysis for a comprehensive and uniform credit monitoring

was introduced in 1985-86 by the RBI by way of the Health Code System in banks

which, inter alia, provided information regarding the health of individual advances, the

quality of credit portfolio and the extent of advances causing concern in relation to total

advances. It was considered that such information would be of immense use to bank

managements for control purposes. Reserve Bank of India advised all commercial banks

(excluding foreign banks, most of which had similar coding system in their organisations)

on November 7, 1985, to introduce the Health Code classification indicating the quality

(or health) of individual advances in the categories, with a health code assigned to each

borrowal account. Under the above Health Code System RBI was further classifying

problem loans of each bank in three categories i.e. i) advances classified as bad &

8 This time duration given for an asset to consider it as a NPA varies from country to country and can change over time within a particular country.

198

doubtful by the bank (ii) advances where suits were filed/decrees obtained and (iii) those

advances with major undesirable features.

The Narasimham Committee (1991) felt that the classification of assets according

to the health codes is not in accordance with the international standards. It believed that a

policy of income recognition should be objective and based on record of recovery rather

than on any subjective considerations. Also, before the capital adequacy norms are

complied with by Indian banks it is necessary to have their assets revalued on a more

realistic basis on the basis of their realizable value. Thus the Narasimham committee

(1991) believed that a proper system of income recognition and provisioning is

fundamental to the preservation of the strength and stability of the banking system. The

committee suggested that Indian banks should follow the international practice in

defining a NPA. Thus based on the recommendations of Narasimham committee report

the non-performing assets would be defined as an advance where, as on the balance sheet

date:

1. In respect of overdraft and cash credits, accounts remain out of order for a period

of more than 180 days,

2. In respect of bills purchased and discounted, the bill remains overdue9 and unpaid

for a period of more than 180 days,

3. In respect of other accounts, any account to be received remains past due for a

period of more than 180 days.

The stricter regulations on NPA definitely reduced bad loans in the banks. Banks

are now constantly being conscious of such accounts and proper measures are taken when

an account has potential to become NPA10. The Gross NPA of the total banking industry

has increased from Rs 50815 crores in 1998 to 70861 crores in 2002 which however has

declined to Rs 58299 crores in 2005 (Table 1.3). Similarly the Net NPA has increased

from Rs 23761 crores in 1998 to Rs 35554 crores in 2002 which however has declined to

Rs 21441 crores in 2005. The growth rates of both Gross and Net NPAs also have

9 An amount is considered overdue when it remains outstanding 30 days beyond the due date. 10 Revealed during our interviews with the banks.

199

declined over time, and after 2003 they have become negative. This shows that the NPA

levels of Indian commercial banks are reducing. This is also confirmed by the fact that

the NPA (both gross and net) as percent of Gross advances as well as total assets is

declining over time. While the Gross NPA as percent of gross advance and total asset has

declined from 14.3% and 6.3% in 1998 to 5.2% and 2.5% in 2005 respectively, the Net

NPA as percent of Gross advance and total asset has declined from 6.7% and 2.9% in

1998 to 1.9% and 0.9% in 2005 respectively.

Table 1.3: Non Performing Assets of Total banking Sector

Non Performing Assets of Total Banking Sector (Rs Crore) 1998 1999 2000 2001 2002 2003 2004 2005Gross NPA 50815 58722 60841 63741 70861 68717 64787 58299Change 7907 2119 2900 7120 -2144 -3930 -6488Percentage growth 15.56 3.61 4.77 11.17 -3.03 -5.72 -10.01As Percent of Gross Advance 14.39 14.71 12.79 11.42 10.42 8.86 7.19 5.27As Percent of Gross Asset 6.36 6.18 5.49 4.91 4.62 4.04 3.27 2.57Net NPAs 23761 28020 30152 32462 35554 32670 24617 21441Change 4259 2132 2310 3092 -2884 -8053 -3176Percentage growth 17.92 7.61 7.66 9.52 -8.11 -24.65 -12.9As Percent of Gross Advance 6.73 7.02 6.34 5.82 5.23 4.21 2.73 1.94As Percent of Gross Asset 2.97 2.95 2.72 2.50 2.32 1.92 1.24 0.95Gross-net 27054 30702 30689 31279 35307 36047 40170 36858Change 3648 -13 590 4028 740 4123 -3312Percentage growth 13.48 -0.04 1.92 12.88 2.1 11.44 -8.24

Source: Report on Trends and Progress of Banks in India, various issues

When we examine the sector-wise scenario we observe that NPAs arising from

the SSI sector is comparatively higher than other sectors that fall even within the priority

sector. From RBI report it is seen that in 2002, NPAs from agriculture loans was 13.8%

and that of SSI was 18.7%. In 2004 NPAs arising from agriculture sector increased to

14.4% but still remained lower to that of SSI sector which was 17.6%. Thus directed

credit to the priority sector in general and loan to SSI sector in particular remained major

concern of the banks as far as NPA issue is concern. It is therefore of interest to look

briefly at the credit to these segments.

200

1.7 Directed Credit to Priority Sector 11

After independence it was felt that in order to achieve overall development of the

country it is essential to develop the large rural sector, for which it is necessary to

channelise required financial resources. In 1954 the ‘All India Rural Credit Survey

Committee’ found that not sufficient credit has been directed towards the rural sector of

the economy. Thus the committee recommended for the development of state sponsored

commercial banking system with branches spread in the rural areas. As a result of this a

drive to nationalize commercial banks was launched. Thus one of the main objectives of

nationalization of commercial banks was to provide credit to, what was considered as,

priority sector. As lending to these sectors was not profitable for commercial banks they

were not motivated to lend to these sectors. This was evident from the fact that the

proportion of credit for industry and trade moved up, from 83 per cent to 90 per cent

between 1951 and 1968. This rise was however at the expense of crucial segments of the

economy like agriculture and that small-scale industry. Due to this reason commercial

banks were directed to lend to these sectors by fixing targets. Apart from fixing targets of

minimum credit, banks were also asked to lend to these sectors at a concessional rates.

This was done to ensure that bank advances were confined not only to large-scale

industries and big business houses, but were also directed, in due proportion, to important

sectors such as agriculture, small-scale industries and exports.

To begin with there was no target on the priority-sector lending. It was just

emphasized that commercial banks should increase their involvement in the financing of

priority sectors, viz., agriculture and small scale industries. However, based on the

recommendations of the report submitted by the Informal Study Group on Statistics

relating to advances to the Priority Sectors, the description of the priority sectors was

later formalized in 1972. Later banks were advised to raise the share of the priority

sectors in their aggregate advances to the level of 33 1/3 per cent by March 1979. Further

it was increased to 40 percent at the end of 1985 and also sub-targets were fixed. During

the initial period, only agriculture, small scale Industries, small and marginal farmers and 11 Priority sector comprises agriculture (both direct and indirect), small scale industries, small roads and water transport, small business, retail trade, professional and self-employed persons, state sponsored organizations for Scheduled Castes/Scheduled Tribes, education, housing (both direct and indirect), consumption loans, micro-credit loans to software, and food and agro-processing sector (Repot on Trend and Progress of Banking in India, 2005-06).

201

artisans and exports were included in the priority sector. Later, based on the

recommendations of Narasimham Committee Report (1991), housing, education,

consumption, profession, I.T. Sector, food processing not falling under SSI, etc. were

also included under the priority sector based.

During 1989-90 the target of priority sector lending was fixed at 40 percent for

domestic commercial banks. Within this there were sub-targets which included 18

percent to agriculture and 10 percent to weaker sections. For foreign banks the total target

was 32 percent within which the sub-target was fixed at 10 percent to small scale

industries and 12 percent to export credit. In 1991.

Later Narasimham committee pointed out many problems related to priority

sector lending, the important one being that a large part of NPA comes from priority

sector lending. Thus the committee recommended reduction of priority sector target to

10 percent and expansion of the coverage of priority sector to include more sectors.

However, the target of priority sector was not reduced but the definition of priority sector

was expanded to include more sectors. Also a provision was made such that banks that

cannot meet the priority sector targets can deposit funds in the financial institutions like

National Bank for Agriculture and Rural Development (NABARD) under Rural

Infrastructure Development Fund (RIDF) or some banks can do so in the Small Industries

Bank of India (SIDBI) for lesser interest rates, which in turn will be lent out to the

priority sectors. The distribution of gross non-food bank real credit to various priority

sector is given in Table 1.4.

Table 1.4 Distribution of Commercial Bank Credit to Priority (Rs Crore, Real Values)

Distribution of Commercial Bank Credit to Priority (Rs Crore, Real Values)

Year

Gross Non-food Bank Credit

Total Priority Sector

Percent of 2 to 1 Agriculture

percent of 4 to 1

Small scale Industries

percent of 6 to 1

Other Priority Sector

percent of 7 to 1

1 2 3 4 5 6 7 8 9 1991-92 144564 54121 37.437 21633 14.964 21625 14.959 10863 7.514

202

1992-93 153862 54612 35.494 21878 14.219 21947 14.264 10787 7.0111993-94 145950 53880 36.917 21208 14.531 22617 15.496 10055 6.8891994-95 168793 58632 34.736 21916 12.984 25256 14.963 11460 6.7891995-96 186127 61461 33.021 22667 12.178 26724 14.358 12070 6.4851996-97 196111 66214 33.764 24528 12.507 28040 14.298 13646 6.9581997-98 210462 72768 34.575 25499 12.116 31817 15.118 15452 7.3421998-99 220324 77650 35.244 26852 12.188 32848 14.909 17950 8.1471999-00 244511 85926 35.142 28928 11.831 34425 14.079 22573 9.2322000-01 268250 96517 35.980 32454 12.098 35004 13.049 29059 10.8332001-02 291729 105910 36.304 36718 12.586 34566 11.849 34626 11.8692002-03 366226 124983 34.127 43422 11.857 35671 9.740 45890 12.5312003-04 411538 149059 36.220 51153 12.430 37206 9.041 60700 14.7502004-05 540426 206203 38.156 67703 12.528 40318 7.460 98182 18.1682005-06 720588 261492 36.289 88355 12.262 46276 6.422 126861 17.605

Source: Handbook of Statistics on Indian Economy

The total priority sector credit of commercial banks was around Rs 54121 crores

during 1991-92, which increased to Rs 96517 crores during 1999-2000 and it was Rs

261492 crores during 2005-06. It is observed that the priority sector credit has registered

higher growth rate during the recent years. While it was around 6 percent for the period

1991-92 to 1999-2000, it increased to around 21 percent during the period 1999-2000 to

2005-06. This could be because the growth rate of the total credit itself has increased

from around 7 percent during the period 1991-92 to 1999-2000 to around 20 percent

during the period 1999-2000 to 2005-06. Though the growth rate of the total priority

sector has been increasing over the years, similar trend is not observed in the case of the

percent of priority sector credit in the total non-food credit. It was around 37 percent of

total non-food credit in 1991-92, which declined to around 33 percent during 1994-95.

This however has improved in the following years and reached around 38 percent during

2004-05, but again declined marginally to 36 percent during 2005-06. The increase in the

percent of total non-priority sector credit in the total non-food credit is not substantial. It

was around 35.14 percent during the period 1991-92 to 1999-2000 which increased

marginally to around 36.17 percent during the period 1999-2000 to 2005-06.

Looking at the growth rates of the sub-sectors of the priority sector; the growth

rate of credit to agriculture and other priority sector are similar to that of the total priority

sector credit, whereas the growth rate of credit to Small Scale Industries (SSI) shows a

203

varying trend. The growth rate of the credit to agriculture sector was around 3.76 percent

during 1991-92 to 1999-2000, which increased to around 20.7 percent during 1999-2000

to 2005-06. The growth rate of credit to SSI was around 6.06 percent during the period

1991-92 to 1999-2000 which has declined marginally to around 5.17 percent during the

period 1999-2000 to 2005-06. However, the growth rate of credit to other priority sector

has registered substantial growth over time. It was around 10 percent during the period

1991-92 to 1999-2000 which has increased to around 34 percent during the period 1999-

2000 to 2005-06.

Looking at the percentage share of credit to the sub-groups of the priority sector,

it is observed that the credit to agriculture sector as well credit to SSI sector has declined

over time where the credit to other priority sector credit has increased over time. The

decline in the credit to SSI is sharper than the decline in the credit to agriculture sector.

The share of credit to agriculture sector in the total non-food credit declined from around

15 percent in 1991-92 to around 12 percent during 2005-06, whereas the credit to SSI

declined from around 15 percent in 1991-92 to around 6.4 percent during 2005-06. This

decline is sharper in the last few years. On the other hand the credit to other priority

sector has increased from around 7.5 percent during 1991-92 to around 18 percent during

2005-06. The increase in the share of priority sector credit could because of the

substantial increase in the housing credit, as housing credit also forms a part priority

sector credit.

1.8 Credit to SSI

Small Scale Industrial sector is one of the important sectors in India for a number

of reasons, prominent amongst them being the employment generation capability.

Recognizing its potential in terms of the employment generation and the production, it

has been give the priority sector status. During 1991-92 the total number of SSI units was

around 68 lakhs which increased to around 119 lakhs during 2004-05. The total

investment also increased from Rs 93555 crore during 1991-92 to Rs 178699 crores

during 2004-05. Production, measured at constant price (1993-94 base) which stood at Rs

84728 crores during 1991-92 increased to Rs 251511 crores during 2004-05. Importantly

the employment level which was around 158 lakh during 1991-91 almost doubled and

204

reached 283 lakh by 2004-05. Similarly the total export increased from Rs 9664 crores

during 1991-92 increased to Rs 86013 crores during 2002-03.

The definition of Small Scale Industries in India is decided on the basis of the

investment in plant and Machinery which has changed over time. During 1966 an

industry was considered a SSI if its investment in the plant and Machinery was not more

than Rs 7.5 lakh. This limit was increased to Rs 10 lakh in 1975. During 1998 the small

scale industry was defined as an industrial unit having an investment in Plant and

Machinery not exceeding Rs.1 crore for most of the 8000 products produced in the SSI

sector and not exceeding Rs.5 crore in respect of certain selected reserved items. In 2006

with the introduction of Micro Small and Medium Enterprises Development Act, 2006

(MSMED Act), the definition of the SSI was further revised. Now, the small scale

enterprises (engaged in manufacturing) are defined as units with investment in plant and

machinery between Rs. 25 lakh to Rs.5 crore. Within the SSI sector there are a number of

sub sectors including tiny industries sector, ancillary sector, khadi and village industries

sector, women enterprises and so on12.

According to the definition of commercial banks credit to Small Scale Industries

include financing of small, micro and unorganized non-farm sector. As mentioned above,

public and private sector banks have to lend 40 percent of their total credit to priority

sector, and for foreign banks it is 32 percent. Unlike agricultural sector there is no fixed

sub-target in the case of credit to SSI for public and private banks. However, foreign

banks are expected to lend 10 percent of their total credit to SSI sector. If they fail to

reach the target, the remaining amount should be deposited at the Small Industries

Development Bank of India (SIDBI). The distribution of commercial banks credit to SSI

sector is presented in table 1.5.

Table 1.5

Commercial Bank Credit to Small Scale Industries

Total SSI Credit

Growth Rate of 1

Total Non food Credit

Growth Rate of 3

Total Priority Sector Credit

Growth Rate of 5

Percent of 1 to 3

Percent of 1 to 5

1 2 3 4 5 6 7 8 1991-92 21625 144564 54121 14.96 39.96

12There are certain types of industries/activities wherein investment on plant and machinery up to Rs. 5 crores can also be registered under SSI category.

205

1992-93 21947 1.49 153862 6.43 54612 0.91 14.26 40.19 1993-94 22617 3.05 145950 -5.14 53880 -1.34 15.50 41.98 1994-95 25256 11.67 168793 15.65 58632 8.82 14.96 43.08 1995-96 26724 5.81 186127 10.27 61461 4.83 14.36 43.48 1996-97 28040 4.92 196111 5.36 66214 7.73 14.30 42.35 1997-98 31817 13.47 210462 7.32 72768 9.90 15.12 43.72 1998-99 32848 3.24 220324 4.69 77650 6.71 14.91 42.30 1999-00 34425 4.80 244511 10.98 85926 10.66 14.08 40.06 2000-01 35004 1.68 268250 9.71 96517 12.33 13.05 36.27 2001-02 34566 -1.25 291729 8.75 105910 9.73 11.85 32.64 2002-03 35671 3.20 366226 25.54 124983 18.01 9.74 28.54 2003-04 37206 4.30 411538 12.37 149059 19.26 9.04 24.96 2004-05 40318 8.36 540426 31.32 206203 38.34 7.46 19.55 2005-06 46276 14.78 720588 33.34 261492 26.81 6.42 17.70

Source: RBI

The total commercial bank credit to SSI sector stood at Rs 21625 crores during

1991-92 which increased to Rs 26724 crores during 1995-95 which further increased to

Rs 34566 crores during 2001-02 and it was Rs 46276 crores during 2005-06. Though

there is an increase in the credit to SSI sector over the years in terms of absolute value,

the annual growth rate shows a varying trend. During 1991-92 it was 1.49 percent which

increased to 11.67 percent during 1994-95, with a sharp decline in following two years it

again increased to 13.47 percent during 1997-98. However it has declined steadily and

reached the lowest level of -1.29 percent during 2001-02. Later it has improved steadily

and it was around 14.78 percent during 2005-06. Unlike the varying trend in the growth

rate, the percentage share of SSI credit in the total non-food bank credit has declined over

time. It was around 15 percent during 1991-92 which declined to 11.85 percent during

2001-02, which further declined to around 6.42 percent during 2005-06. The percentage

share of credit to SSI in the total non-priority sector has marginally increased from 40

percent to 43.7 percent between 1991-92 and 1997-98 which, however, has declined

steadily thereafter and reached around 17.7 percent during 2005-06.

1.9 Conclusion Controlling the occurrence of systemic banking problems is undoubtedly a prime

objective for policy-makers, and understanding the mechanisms that are behind the surge

in banking crises is of utmost importance in this regard. Amongst the problems faced by

the banks of many developing nations, occurrence of non-performing assets (NPA) is a

206

prominent one. While the origin of the problem of high level of NPAs basically lies in

the quality of managing credit risk and the extent of preventive measures adopted,

various factors like real interest rates, directed credit or inflation rate can also effect the

level of NPA. Analysis of factors that cause the ratio of NPAs to total loans to fluctuate,

for selected Asian countries, (viz., Taiwan, Hongkong , Singapore and others) reveals

that a high ratio of corporate loans to individual loans results in lower percentage of

NPA (Wu et al, 2003). In the literature it has also been cited that the reasons why NPAs

are created are sometimes systemic in nature and directly attributable to events such as

real estate bubbles (Thailand and Indonesia) or a high proportion of directed lending

(Krueger et al, 1999). The problem is significant for the Chinese banks as well and in

order to deal with the mounting NPA problem in the Chinese banks, government

constituted four asset management companies (Bonin and Huang, 2001). Thus NPA is a

problem of banking sector of many developing nations which needs to be studied

carefully.

In Indian financial system, an asset is classified as non-performing asset (NPAs) if the

borrower does not pay dues in the form of principal and interest for a period of 180 days.

However, with effect from March 2004, it has been decided that a default status would be

given to a borrower if dues were not paid for 90 days. Further, if any advance or credit

facilities granted by bank to a borrower becomes non-performing, then the bank will have

to treat all the advances/credit facilities granted to that borrower as non-performing

without having any regard to the fact that there may still exist certain advances / credit

facilities having performing status.

Due to the social banking motto of the Government, the problem of NPA had not

received due attention in India in the post nationalization (of banks) period. However,

with the recent financial sector liberalization drive, this issue has been taken up seriously

by introducing various prudential norms relating to income recognition, asset

classification, provisioning for bad assets and assigning risks to various kinds of assets of

a bank. Overtime though NPA as a percentage of total advances have reduced, it still

remains a concern for the Indian banking sector. While the Reserve Bank of India (RBI)

207

as well as the banks have begun to pay considerable attention to the NPA problem, there

are only a limited number of rigorous studies in the Indian context that look at this issue

in some detail (see Ghosh, 2005, Mor and Sharma, 2003, Rajaraman et al, 1999).

Furthermore, while reform regulations attempt to streamline banking operations, norms

of priority sector lending remains intact more or less. In particular, banks need to allocate

40% of their total credit disbursement to agriculture, small-scale industries and other such

designated priority sectors.

However, it is also well known that the small firms, besides generating manufacturing

output and foreign exchange through exports, are also a major source of employment in a

labour surplus economy like India. It is also understood that the lack of access to finance

for working capital and new investment presents a significant constraints on the ability of

small firms to carry out business and to expand (Gang, 1995). Thus it is essential to

examine the problem of NPA arising out of advances made by banks to this sector.

Therefore, at the macro level, there is a need to look at the determinants of NPA in the

Indian banking sector by examining some of the bank specific as well as macro level

indicators. At the micro level on the other hand, one needs to identify the sector specific

factors responsible for non-recovery of loans.

Given this back ground, in the current project, we attempt to look at the determinants of

NPA by examining some of the external and internal factors like, the extent of

competition, total assets of a bank, size of operations, proportion of rural branches ,

investments etc., that can influence NPA. It is of our interest to examine, between

various bank groups (viz, SBI, Nationalized banks, Private banks and Foreign banks),

which is the most efficient group in recovery of loans and what are the factors that

determine this efficiency.

To have a micro perspective of the problem, as a case study, we have taken up a field

survey based exercise concerning the SSI sector to understand the actual workings of the

loan recovery process and the associated problems. In particular, we are interested in

208

examining the factors that have influence on recovery of loans in this segment of the

Indian economy.

Given this background the report is arranged as follows. Credit being our main area of

focus, we concentrate on this aspect in some detail in Chapter 2 , mainly focusing on the

credit to the SSI sector. The issue of non-performing asset in the Indian banking sector is

discussed in general with trends of NPAs in Chapter 3. In Chapter 4 we analyse data

NPA of commercial banks in a panel framework to identify the determinants of NPA.

This is done for the total advances and also for advances to the SSI sector. In Chapter 5

we concentrate on the efficiency issue. In particular we examine the profit efficiency of

the Indian banking sector and in particular check for the significance of NPA as a

determinant of efficiency. Chapter 6 and Chapter 7 are based on our primary data

collection from small firms and banks respectively. A concluding section is presented at

the end.

209

CAHPTER 2

Credit Operations of Indian Banking Sector and Credit to SSI

2.1 Introduction

Lending is the core activity of the banking sector. This is the activity through which a

bank as a firm earns profit.

To make credit available to all sections of the society, at the initial stages of development

a need was felt for a wider diffusion of banking facilities, mainly the credit facilities (see

also Chapter 1). This was due to the fact that at that time, when activities were left

entirely to the banking sector, banking operations concentrated in selected locations and

sectors; for example, credit for industry and trade were as high as 83% to begin with and

moved up further to 90% between 1951 and 1968. Thus credit to some of the core sectors

like agriculture or SSI, that were in dire need of financial assistance were not

forthcoming from the institutional lending agencies like commercial banks. Consequently

certain controlling measures were perceived as necessary at that time. By 1969, 14 major

banks were nationalized and slowly lending norms for the specific , hitherto neglected

sectors were formalized. By 1979 banks were advised to lend to the priority sector (which

210

includes agriculture and SSI sectors) to the tune of 33.33% of total credit. By 1985 this

percentage was raised to 40%. While credit to the SSI sector is treated as a part of priority

sector lending, no specific target was set for this sector for the Indian banks. For the

foreign banks however, 32% of total lending is earmarked for the priority sector of which

10% is needed to go towards the small industries. Any short fall in such lending by

foreign banks has to be deposited with the Small Industries Development Bank of India

(SIDBI).

With liberalization, new changes have been brought into the banking sector (see also

Chapter 1). Efficiency in banking operations was given utmost priority and profit became

a measuring yardstick of performance. Such new emphasis has slowly been changing the

focus of the banking sector. In particular, there was a decline in the percentage of credit

to the small-scale sector and it has been observed that the banks fail to adhere to the

lending norms prescribed for the agriculture sector. Side by side rural branches also

started to decline.

In this background the present chapter looks at the current scenario of bank lending and

in particular lending to the small-scale sector. The next section concentrates on the trends

in bank lending over the years. Section 3 discusses the role of the small-scale sector for

the Indian economy and the problems faced by the sector especially relating to the

financial assistance. Section 4 examines the kind of credit facilities enjoyed by the sector

especially from the commercial banks. A concluding section follows thereafter.

2.2 Credit Operation of Indian Commercial Banks

Lending operations of the Indian commercial banks started much before independence.

According to Beckhart (1967), from the 59 reporting banks in 1939, total advances were

observed to be Rs 151 crores; thereafter there had been some fluctuations with regards to

total lending . These fluctuations were usually due to the overall economic fluctuations at

that time. In the year 1947 total loans and advances increased to Rs 425 crores As the

economic activities gain momentum during plan-period, credit also shows an upward

211

trend with credit deposit ratio increasing to above 60% from about 40% to 50% in the

previous years (table 2.1).

Table 2.1 Loans and Advances of Schedule Banks 1939-1952

Year No. of Reporting Banks

Loans and Advances (in crores of rupees)

Credit* Deposit ratio

1939 59 151 59

1942 61 122 23.9

1947 97 425.2 44

1951(first

plan)

92 545.1 67

1952 91 474 60

*loans and advances plus bills purchased and discounted

• Source: Beckhart (1967)

From 1961 onwards (third plan) there was always been an increasing trend of credit

disbursement. Bank credit of scheduled banks increased to about 1320 crores of rupees

and doubled by the year 1966-67. Such trend is due to the growth in economic activities

in general and industrial activities in particular.

By the year 1969, 14 major commercial banks were nationalised and the year 1970-‘71

saw a total lending of 4684 crores of rupees. This figure more than doubled in 1975-‘76

to reach Rs 10877 crores and this trend continued thereafter. In 1980-‘81 , total credit

was Rs 25371 crores and the year 1985 saw an increase of credit disbursement to Rs

56067 crores. By the year 1990-‘91 total credit of commercial banks touched the figure

of Rs 116301 crores.

2.3 Bank Group-wise Total Credit of Commercial Banks

Indian commercial banks are classified into four broader categories, viz., State

Bank and its associates (SB &A) , Nationalised Banks (NB), Private Banks (PB) and

Foreign Banks (FB). The total credit of commercial banks, according to the bank group,

212

is presented in table 2.2. It is observed that even in real terms there has been a substantial

increase in credit disbursement. The total bank credit (in real terms) of the banking

sector as a whole was around Rs 169182 crores in 1990 which increased to Rs 175021

crores in 1995. While the extent of rise was not phenomenal during the first 5 years of

reform, the next ten years saw substantial growth of credit. In particular, credit increased

to Rs 277193 crores in 2000 and to Rs 589911 crore in 2005. Between 1991 and 2005 the

total real credit increased by around 3.5 times. Similar trend is observed in the case of

individual bank groups as well. The credit of State Bank of India and Associates (SB&A)

increased from Rs 57004 crores in 1990 to Rs 146014 crores in 2005 (around 2.5 times),

the credit of Nationalised Banks (NB) increased from Rs 98885 crores in 1991 to Rs

292279 crores in 2005 (around 2.9 times), the credit of Private Banks (PB) increased

from Rs 5747 crores in 1990 to Rs 112993 crores in 2005 (around 19.6 times) and the

credit of Foreign banks increased from Rs 7546 crores in 1990 to Rs 38625 crores in

2005 (around 5 times).

Table 2.2 Bank group-wise total commercial bank credit

Real Credit of commercial Banks (Real values, Rs crore) Share of Total Credit SB&A NB PB FB Total SB&A NB PB FB 1990 57004 98885 5747 7546 169182 0.34 0.58 0.03 0.041991 58547 98274 5897 8483 171200 0.34 0.57 0.03 0.051992 59081 98756 7065 10233 175135 0.34 0.56 0.04 0.061993 58941 95957 7984 10641 173524 0.34 0.55 0.05 0.061994 49044 85211 8968 10610 153832 0.32 0.55 0.06 0.071995 53981 95013 13262 12764 175021 0.31 0.54 0.08 0.071996 60945 100955 17453 17573 196926 0.31 0.51 0.09 0.091997 60625 100442 20938 19563 201567 0.30 0.50 0.10 0.101998 66102 109965 23998 19845 219910 0.30 0.50 0.11 0.091999 70672 123143 27841 19233 240889 0.29 0.51 0.12 0.082000 80653 139435 34842 22263 277193 0.29 0.50 0.13 0.082001 90882 159681 41128 25983 317674 0.29 0.50 0.13 0.082002 97212 186694 68768 28724 381397 0.25 0.49 0.18 0.082003 106895 203473 77803 29475 417646 0.26 0.49 0.19 0.072004 119198 222824 92100 32706 466828 0.26 0.48 0.20 0.072005 146014 292279 112993 38625 589911 0.25 0.50 0.19 0.07

SB & A: State Bank and Associates, NB: Nationalized Banks, PB: Private Banks, FB:

Foreign Banks

Source: RBI

213

Another noteworthy feature is that though for obvious reasons nationalized banks

and state bank and associates have shares over 80% of the total bank credit, over the

years the share of private banks are increasing ; foreign banks’ share, however, has

remained more or less the same during the liberalization period. This is partly because

the new foreign banks that entered the market are yet to get stabilized and operate in a

full-fledged manner .

The percent annual growth rates of commercial bank credit, according to bank

group are presented in table 2.3. The average annual growth rate of commercial bank

credit of the total banking sector for the period 1990 to 2005 is 9.06 percent. This,

however, as expected from the above discussion, differs at the bank group level. While

the annual average growth rate of credit of SB&A is around 6.85 per cent, it is around

7.93 percent for NB and around 11.94 percent for FB. Private Banks have registered the

highest annual average growth rate of around 22.83 percent for the period 1990 to 2005.

It is also interesting to note that commercial banks, across all bank groups, have

registered higher growth rate during the period 1995 to 2005 compared to the period

1990-1995.

Table 2.3 Growth rates (percent increment) of credit of commercial banks (in

real terms)

Growth rates of Commercial Bank Real Credit (in per cent) SB&A NB PB FB Total

1991 2.71 -0.62 2.61 12.42 1.19 1992 0.91 0.49 19.81 20.64 2.30 1993 -0.24 -2.83 13.01 3.99 -0.92 1994 -16.79 -11.20 12.32 -0.30 -11.35 1995 10.07 11.50 47.89 20.31 13.77 1996 12.90 6.25 31.60 37.67 12.52 1997 -0.53 -0.51 19.97 11.33 2.36 1998 9.04 9.48 14.61 1.44 9.10 1999 6.91 11.98 16.01 -3.08 9.54 2000 14.12 13.23 25.15 15.75 15.07 2001 12.68 14.52 18.04 16.71 14.60 2002 6.96 16.92 67.20 10.55 20.06 2003 9.96 8.99 13.14 2.62 9.50 2004 11.51 9.51 18.38 10.96 11.78 2005 22.50 31.17 22.69 18.09 26.37

214

Source: RBI

2.4 Total Credit of Commercial Banks According to Occupation

Banks lending activities encompass various sectors of the economy. Naturally

funds are directed to the so-called booming sectors. However, as mentioned above Indian

financial sector is also guided by certain norms prescribed by the Reserve Bank of India

(RBI), which ensures flow of funds to certain core sectors of the economy. Distribution

of total credit according to occupation is presented in table 2.4. Looking at the different

occupation-wise flow of funds one observes that credit to the agriculture sector has

declined in real terms between 1990 and 1995. It was around Rs 22546 crore in 1990

which has declined to Rs 20910 crore in 1995. It should be mentioned in this context that

banks are supposed to direct 18 % of their total lending to the agriculture sector ;

however, in reality many banks often fail to meet this norm. Subsequently , however,

total credit to the agriculture sector has increased and reached around Rs 49356 crores in

2004.

In the case of other sectors the real total credit has been increasing over the years.

Credit to industry increased marginally from Rs 68948 crores in 1990 to Rs 80639 crores

in 1995 which further increased to Rs 171694 crores in 2004. There is a remarkable

increase in the personal loans and also credit to financial institutions. While the personal

loans increased from Rs 9083 crores in 1990 to Rs 15827 crores in 1995, it increased by

around 5 times, from Rs 15827 crores to 91839 crores between 1995 and 2004. Similarly

credit to financial institutions increased from Rs 3030 crores in 1990 to Rs 6664 crores in

1995 which further increased to Rs 20238 crore in 2004 (around 5 times).

Table 2.4 Distribution of Outstanding Credit of Commercial Banks according to

Occupation Distribution of outstanding credit of commercial banks according to occupation

(Rs in Crores, in Real values)

Agriculture Industry Personal Loans

Financial Institutions Others@ Total

1990 22546 68948 9083 3030 37844 141451

215

1991 22129 70406 11436 3344 35900 147981 1992 22179 71467 12260 4385 39527 149818 1993 22060 78964 13530 3959 43954 162467 1994 20902 77390 13862 4166 44414 160734 1995 20910 80639 15827 6664 52758 176799 1996 22474 95374 18433 7031 55373 198684 1997 23134 102609 20623 8259 53333 207958 1998 23891 109144 23545 8291 58670 223541 1999 26652 122505 25805 10307 63998 249268 2000 28526 133624 32277 13672 79477 287576 2001 31261 142877 39848 15987 95407 325380 2002 37806 160431 48738 22216 118261 387452 2003 42901 175044 64374 28613 116169 427101 2004 49356 171694 91839 30238 108314 451442

@ Other include - Transport, personal and professional service, Trade and Miscellaneous

Computation of growth rates of real credit reveals that between 1995 and 2004

growth rates have been much higher compared to the same between 1991 and 1995. The

average annual growth rate of real total deposits between 1991 and 2004 is around 8.61

per cent (Table 2.5). While it is around 5.08 percent between 1991 and 1995, it has

increased to 11.06 percent between 1996 and 2004. Credit to agriculture sector has

registered the lowest annual average growth rate of around 5.50 percent between 1991

and 2004. It was even negative between 1991 and 1995 (around -1.47 per cent per

annum). However, it has increased substantially between 1995 and 2004 (around 10.05

per cent per annum). While on the one hand credit to agricultural sector has registered

low growth rate, on the other hand personal loans and credit to financial institutions has

registered remarkable growth rate. Between 1991 and 2004 personal loan has registered

around 19.77 percent annual average growth rate, while it was around 20.32 percent in

the case of credit to financial institutions. One important observation is that, unlike other

occupations which registered higher growth rate between 1995 and 2004, credit to

financial institutions registered higher growth rate between 1991 and 1995. it was around

22.35 percent during 1991 and 1995, which reduced to around 18.97 percent between

1995 and 2004.

Table 2.5 Occupation-wise growth of outstanding Credit Growth Rates of Outstanding Credit of Commercial Banks According to Occupation

(Rs in Crores, in Real values)

216

Agriculture Industry Personal Loans

Financial Institutions Others Total

1990 -1.44 10.47 38.76 37.09 -0.03 7.24 1991 -1.85 2.11 25.91 10.36 -5.14 4.62 1992 0.22 1.51 7.20 31.16 10.10 1.24 1993 -0.53 10.49 10.36 -9.72 11.20 8.44 1994 -5.25 -1.99 2.46 5.22 1.05 -1.07 1995 0.04 4.20 14.18 59.98 18.79 9.99 1996 7.48 18.27 16.46 5.50 4.96 12.38 1997 2.94 7.59 11.88 17.46 -3.68 4.67 1998 3.27 6.37 14.17 0.39 10.01 7.49 1999 11.56 12.24 9.60 24.33 9.08 11.51 2000 7.03 9.08 25.08 32.64 24.19 15.37 2001 9.59 6.92 23.46 16.94 20.04 13.15 2002 20.94 12.29 22.31 38.96 23.95 19.08 2003 13.48 9.11 32.08 28.80 -1.77 10.23 2004 15.05 -1.91 42.67 5.68 -6.76 5.70

Source: RBI

The difference in the growth rate of the real credit to different occupations has led

to the changing composition of the credit according to the occupation. The percent share

of real credit according to occupation is presented in Fig.2.1. On the one hand the share

of the credit to industry and agriculture sector has declined between 1991 and 2004, on

the other hand, as is clear from the above discussion as well, the share of the personal

loan and credit to financial institutions has increased during the same period. The share of

the credit to industrial and agriculture sector was around 48.74 and 15.94 percent

respectively, which has subsequently declined to around 38.03 and 10.93 percent. And,

the share of personal loan and credit to financial institutions was around 6.42 and 2.14

percent respectively in 1991 which increased to 20.34 and 6.70 percent in 2004.

Fig 2.1 Percentage share of Credit of Commercial Banks According to Occupation

217

0%

20%

40%

60%

80%

100%Pe

r ce

nt

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

Year

Credit of Commercial Bnaks According to Occupation (Per cent Share)

Industry

Agriculture

Personalloans

Financialinstitutions

Others @

@ Other include - Transport, personal and professional service, Trade and Miscellaneous

2.5 Distribution of Total Credit : Rural and Urban

Another important dimension of the commercial banks credit is the credit

disbursement according to location (rural vs urban). Further, since in the post

nationalization period, credit expansion to rural and semi-urban areas was given

considerable importance, it becomes essential to look at the trends of the bank credit to

these areas over time. Distribution of commercial bank real credit according to location is

presented in table 2.6. Similar to the credit to different occupations, in the case of the

credit to different population groups also the increase in the real credit is higher between

1995 and 2004 compared to the increase in credit between 1991 and 1995. The credit to

rural sector increased from Rs 21789 crores in 1991 to Rs 22632 crore in 1996 (around

1.03 time), which further increased to Rs 56397 crores in 2005 (around 2.5 times between

1996 and 2005). Similar trend is observed in the case of credit to other population groups.

Credit Semi-urban areas increased from Rs 24206 crores in 1991 to Rs 25961 crore in

1996 which further increased to Rs 66995 crore in 2005. Similarly, credit to urban and

218

metropolitan areas increased from Rs 31995 crore and Rs 63422 crore respectively in

1991 to Rs 32812 crore and Rs 99088 respectively in 1995 which further increased to Rs

97044 crores and Rs 370571 crores respectively in 2005.

Table 2.6 Population group-wise Distribution of Outstanding Credit

Distribution of Outstanding Credit of Commercial Banks According to Population Group (in Rs Crores, in Real values)

Year Rural Semi Urban Urban Metropolitan Total 1990 21789 24206 31995 63422 141411 1991 22160 24195 33090 68536 147981 1992 22677 23671 32486 70984 149818 1993 22906 23592 33020 82949 162467 1994 22544 22438 32778 82973 160734 1995 21100 23798 32812 99088 176799 1996 22632 25961 35184 114907 198684 1997 23785 27338 36514 120321 207958 1998 25473 28699 39301 130067 223541 1999 27435 31621 44427 145785 249268 2000 30707 35351 49814 173907 289779 2001 33157 37608 57082 200145 327991 2002 38987 43157 63293 238102 383538 2003 42452 48077 71130 269364 431024 2004 45940 54124 81232 294370 475665 2005 56397 66995 97044 370571 591008

On an average percentage increment of credit from period 1991 to 2005 is around

6.34, 7.09, 7.62 and 12.46 for rural, semi-urban, urban and metropolitan areas

respectively. It is important to note that between 1991 and 1995 the rural sector has

registered negative annual growth rate of around -0.503 percent. For the same period the

average growth rate is around 9.484 percent for the metropolitan areas. However from

1996 to 2005 growth rate of credit to all sectors is seen to be positive and increasing (Fig.

2.2).

Fig. 2.2 Population group with percent increment in credit

219

Per cent Growth of Total Credit of Commercial Banks

-10

-5

0

5

10

15

20

25

30

1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005

Year

Per c

ent G

row

th

RuralSemi UrbanUrbanMetropolitan

Though there is increased growth rate of credit across all population groups, the

difference in the growth rates led to change in the composition of credit to different

population groups. The percent share of credit to different population groups is presented

in Fig.2 3. As can be seen from the chart, the share of credit to rural areas in the total

credit has declined over years, whereas the share of credit to metropolitan areas has

increased. The share of credit to rural areas was around 15.08 percent in 1991 which

declined to around 11.93 percent in 1995 which further declined to around 9.54 percent in

2005. Similarly the credit to semi-urban and urban areas was around 17.11 percent and

22.62 percent respectively in 1991 which declined to 13.46 percent and 18.55 percent

respectively in 1995, which further declined to 11.34 and 16.42 percent respectively in

2005. On the other hand the credit to metropolitan areas has been increasing over time. It

was around 44.8 percent in 1991 which increased to 56 percent in 1995 which further

increased to 62.7 percent in 2005.

Fig 2.3 Share of credit to different population groups

220

0%

20%

40%

60%

80%

100%Pe

r ce

nt

1990

1991

1992

1993

1994

1995

1996

1997

1998

1999

2000

2001

2002

2003

2004

2005

Year

Distribution of Commercial Bank Credit According to Population (Per cent Share)

Rural

SemiUrban

Urban

Metropolitan

We have seen so far trends of flow of credit to different regions of the economy as per

rural or urban and also to major sectors like agriculture or industry. In this study

however we are particularly interested in credit to the SSI sector. We therefore examine

in some detail the flow of funds to the priority sector and within the priority sector to the

SSI.

2.6 Sector-wise Distribution of Credit

As mentioned above 40% of the total credit needs to be disbursed to the priority sector. It

has been observed that initially banks were unable to meet this prescription. However,

after liberalization a number of new avenues are incorporated within the purview of the

priority sector. Subsequently, banks have been complying with the priority sector norms.

At the sectoral level, credit to priority sector was around Rs 54121 crores during 1991-

221

‘91 which increased to Rs 85926 crores during 1999-2000, and, reached Rs 261492

crores during 2005-‘06. Between 1991-‘92 and 2005-‘06 credit to priority sector has

increased by around 4.5 times. Within the priority sector, credit to agriculture sector has

increased from Rs 21633 crores during 1991-92 to Rs 88355 crores during 2005-06

(around 4 times). Much of the increase in the credit to agriculture is observed during last

few years, especially during 2003-06.

It is noteworthy that the increase in credit to small-scale industries, within the priority sector is much less compared to agriculture sector. It was Rs 21625 crores during 1991-‘92 which increased to Rs 46276 during 2005-06 (around two fold). Credit to industrial sector and wholesale trade increased from Rs 56105 crore and Rs 7332 crore respectively during 1991-92 to Rs 235291 crores and Rs 20362 crore respectively during 2005-06 (Table 2.7).

Table 2.7 Setoral Deployment of Non-food Credit

Sectoral Deployment of Outstanding Non-food Gross Bank Credit (Rs Crore, Real Values)

Year Priority Sector of which Industry

Wholesale Trade

Other Sectors Total

Agriculture Small scale Industries

1991-92 54121 21633 21625 56105 7332 27005 144564 1992-93 54612 21878 21947 64260 7637 27353 153862 1993-94 53880 21208 22617 57865 7330 26875 145950 1994-95 58632 21916 25256 68237 8909 33015 168793 1995-96 61461 22667 26724 77992 10041 36633 186127 1996-97 66214 24528 28040 80041 9626 40229 196111 1997-98 72768 25499 31817 85948 9665 42081 210462

222

1998-99 77650 26852 32848 88426 9461 44787 220324 1999-00 85926 28928 34425 96024 10962 51599 244511 2000-01 96517 32454 35004 101782 11154 58797 268250 2001-02 105910 36718 34566 104137 12364 69318 291729 2002-03 124983 43422 35671 138898 13335 89009 366226 2003-04 149059 51153 37206 139667 14049 108763 411538 2004-05 206203 67703 40318 190435 17594 61640 540426 2005-06 261492 88355 46276 235291 20362 102811 720588

*Medium and large # Other than food procurement

Growth rates of gross commercial bank credit to various sectors are presented in

table 2.8. The average annual growth rate of gross non-food credit of commercial banks

is around 11.3 percent. While the average growth rate of credit to priority sector is around

11 percent per annum, it is around 10 percent and 4.82 percent respectively for the

agriculture and small-scale industries sector during 1991-92 to 2005-06. The average

growth rate of credit to industry and wholesale trade are around 10.23 and 6.8 percent per

annum respectively for the period 1991-92 and 2005-06. It is observed that the growth

rates are higher during the period 1996-97 to 2005-06 compared to the period 1991-92 to

1995-96. The growth rates of credit to priority sector, industry and wholesale trade are

around 1.24, 5.95 and 5.18 percent respectively for the period 1991-92 to 1995-96. This

has increased to 15.95, 12.36 and 7.68 percent for priority sector, industry and wholesale

trade respectively for the period 1995-96 to 2005-06.

Table 2.8 Percentage increment of Sectoral Deployment of Credit

Per cent Growth of Sectoral Deployment of Outstanding Non-food Gross Bank Credit (Rs Crore, Real Values)

Year Priority Sector Agriculture

Small Scale Industries Industry

Wholesale Trade

Other Sectors Total

1991-92 -7.00 -4.76 -7.18 -7.04 -8.51 -1.31 -6.08 1992-93 0.91 1.13 1.49 14.53 4.16 1.29 6.43 1993-94 -1.34 -3.06 3.05 -9.95 -4.03 -1.75 -5.14 1994-95 8.82 3.34 11.67 17.92 21.54 22.85 15.65 1995-96 4.82 3.43 5.81 14.30 12.71 10.96 10.27 1996-97 7.73 8.21 4.92 2.63 -4.13 9.82 5.36 1997-98 9.90 3.96 13.47 7.38 0.41 4.60 7.32 1998-99 6.71 5.31 3.24 2.88 -2.12 6.43 4.69 1999-00 10.66 7.73 4.80 8.59 15.87 15.21 10.98

223

2000-01 12.33 12.19 1.68 6.00 1.75 13.95 9.71 2001-02 9.73 13.14 -1.25 2.31 10.84 17.90 8.75 2002-03 18.01 18.26 3.20 33.38 7.86 28.41 25.54 2003-04 19.26 17.80 4.30 0.55 5.35 22.19 12.37 2004-05 38.34 32.35 8.36 36.35 25.23 -43.33 31.32 2005-06 26.81 30.50 14.78 23.55 15.73 66.79 33.34

The percentage share of gross non-food credit to various sectors, presented in Fig.

2.4 shows varying trend over the period. The share of credit to agriculture sector has

declined from 37 percent during 1991-92 to around 33 percent during 1995-96 which

however has increased to around 38 percent during 2004-05. On the other had credit to

industrial sector has increased from 38.8 percent during 1991-92 to 41.9 percent during

1995-96 which declined to 32.6 percent during 2005-06. The percent share of credit to

wholesale trade in the total gross credit has steadily declined from 5 percent in 1991-92

to 2.8 percent during 2005-06. Looking at the components of the priority sector credit,

the share of the credit to agriculture sector as well small-scale industries has declined

over time. However, the decline is sharp in the case of small-scale industries compared to

agriculture sector. While the share of the agriculture sector in the total priority sector

lending declined from around 40 percent in 1991-92 to around 33.7 percent during 2005-

06, the share of credit to small-scale industries in the total priority sector credit declined

from around 40 percent in 1991-92 to around 17.7 percent during 2005-06.

Fig. 2.5 Share of Gross Non-Food Credit

224

Percent Share of Gross Non-food Credit

0

5

10

15

20

25

30

35

40

45

5019

91-9

2

1992

-93

1993

-94

1994

-95

1995

-96

1996

-97

1997

-98

1998

-99

1999

-00

2000

-01

2001

-02

2002

-03

2003

-04

2004

-05

2005

-06

Year

Per

cent

Priority sector

Industry

Wholesale trade

% of agriculturein total prioritysector

% of SSI in totalpriority sector

Source: RBI

This indeed is a matter of concern as the SSI sector plays a crucial role for the Indian

economy in a number of aspects. By its less capital intensive and high labour absorption

nature, small-scale industries (SSI) sector has made significant contributions to

employment generation and also to rural industrialization, thereby helping in balanced

regional growth. When the performance of this sector is viewed in terms of output as well

as employment growth against other related sectors in the economy, one observes that the

growth performance of the SSI sector is far higher than the large-scale industries sector

and the manufacturing sector.

Given this background we first discuss briefly the importance of the SSI sector in a

labour surplus economy like India and the problems it faces for further growth and

development. In this context credit related norms and issues are highlighted.

2.7. The Role of the Small and Medium Industries Sector

225

Small and medium industries sector is divided into various sub-sectors to come up with

appropriate policy measures for each of them.

Definitional Issues

One of the difficulties in defining small firms is to identify what is ‘small’ (Gang, 1995).

One is often tempted to define small firms in comparative terms, that is, in comparison to

the large firms in the industry. Brock and Evans (1986) suggest that one can consider

those firms that lie on the left hand tail of the size distribution of firms, for example, say,

in the bottom quartile as ‘small’. However, such relativistic definition has certain

shortcomings. A firm that is considered large today may become small due to the entry of

a few larger firms in the market. Thus there is a need to define ‘smallness’ in absolute

terms at least for the purpose of policy formation and implementation. Naturally, the

definition can be on the basis of employment size, turn over, invested capital and so on.

In the Indian context ‘smallness’ is conceptualized on the basis of investment in plant and

machinery.

In India, a firm is considered to belong to the SSI sector, if its investment in plant and

machinery does not exceed Rs 10 million ($ 250,000 approximately)13. Within the SSI

sector there are a number of sub sectors including tiny industries sector, ancillary sector,

khadi and village industries sector, women enterprises and so on14. Since the

government’s policy incentives differ across these sub sectors, there is a need to define

them in precise terms. An ancillary industrial undertaking is a small enterprise as per the

above definition and is engaged in the manufacture of parts, components, sub-assemblies,

tooling or intermediates. On the other hand, a small scale service and business (industry

related) enterprise (SSSBE) face investment limit up to Rs 1 million in fixed assets,

excluding land and building. The tiny and village industries sector receive special

attention of the government given their vulnerability and traditional importance. Hence

the tiny, village & khadi15 and also women enterprises are specifically defined.

13 Source: Government of India website: http://www.smallindustryindia.com/ssiindia/definition.htm 14There are certain types of industries/activities wherein investment on plant and machinery up to Rs. 5 crores can also be registered under SSI category. 15 A kind of traditional handloom.

226

Investment limit in plant and machinery with respect to tiny enterprises is Rs 2.5 million

irrespective of location of the unit. A woman enterprise is the one that is owned or held

more than 51% share by a woman.

According to the Small Industries Development Organisation (SIDO), the term small-

scale industry is used in the context of modern industrial units using a mechanized

process or those engaged in the service sector. i.e, SSSBE. On the other hand, rural

industries refer to village based semi industrial activities including the production of

khadi, silk, coir etc. The coir sector is an agro-based industry relying on coconut fiber.

The nodal agency for coordinating activities related to the SSI sector is SIDO, while for

the Khadi and Village Industries Sector, it is the Khadi and Village Industries

Commission and for the Coir sector, it is the Coir Board. Thus each of these sub sectors

operate under a separate bureaucratic setup.

Considering all enterprises viz., small, tiny and traditional village enterprises both in the

registered and unregistered segments, there were about 6 million enterprises in 1990-’91

which increased to about 10 million in 2004-’05. Output of these enterprises at the 1993-

’94 prices was about rupees of 847 billions which increased to Rs 2515 billion in 2004-

’0516.

Contributions to the Economy

The small-scale industries’ sector plays a vital role in the growth of the country by

contributing almost 40% of the gross industrial value added in the Indian economy.

Some of the major statistics concerning the small scale industries are presented in

table 2.9. The number of total SSI units increased from 68 lakhs in 1990-91 to 83 lakhs

in 1995-96 which further increased 1191 lakhs in 2004-‘05. Along with the number of

units the fixed investment increased over time. It was Rs 93555 crores in 1991-92 which

increased to Rs 125750 crores in 1995-96 which further saw a rise to Rs 178699 crores in 16 These estimates are from the Ministry of Small Scale Industries. (provided by www.indiastat.com).

227

2004-‘05. It is important to note that the fixed investment per unit has also increased over

time. It was around Rs 13.78 lakh during 1991-‘92 which increased to around Rs 15.14

lakh during 1995-96, which however has declined marginally to Rs 15.06 lakh during

2004-‘05. Similarly the production of SSI, measured at constant price has also been

increasing over time. It was around Rs 84728 crores during 1991-92 which increased to

Rs 121175 crores during 1995-96 which further increased to Rs 251511 crores during

2004-05. The production per unit is also increasing over time. It was around 1.24 lakh in

1991-92 which increased to Rs 1.4 lakh during 1995-96 which further increased to Rs

2.12 lakh during 2004-05.

Table 2.9 Major Aggregates of the SSI sector

Production

Year

No. of units

(lakhs)

Fixed

investment (at current prices, Rs.

crore)

At current prices,

(Rs. crore.)

At constant

price (1993-94 base, Rs crore)

Employm

ent (in lakh

persons)

Export

(Rs.crore)

1990-91 68 93555 78802 84728 158 9664 1991-92 71 100351 80615 87355 166 13883 1992-93 74 109623 84413 92246 175 17784 1993-94 76 115795 98796 98796 183 25307 1994-95 80 123790 122154 108774 191 29068 1995-96 83 125750 147712 121175 198 36470 1996-97 86 130560 167805 134892 206 39248 1997-98 90 133242 187217 146263 213 44442 1998-99 93 135482 210454 157525 221 48979 1999-00 97 139982 233760 170379 229 54200 2000-01 101 146845 261297 184401 239 69797 2001-02 105 154349 282270 195613 249 71244 2002-03 109 162317 311952 210636 260 86013 2003-04 114 170219 357733 228730 271 NA 2004-05 119 178699 418263 251511 283 NA

Source : Annual Report 2005-06, Ministry of SSI, Govt. of India

We see from table 2.9 that the total number of SSI units has increased over time.

Also the total fixed investment of the SSI, their production, employment level and

exports have increased over time. However, their growth rates show a different picture.

The growth rate of the number of SSI has remained almost stable around 4 percent per

annum through the years. On the other hand the growth rates of investment in the fixed

228

assets show a varying trend. It was around 7.26 percent during 1991-‘92, which declined

to around 1.68 percent during 1998-‘98, which however has increased thereafter and was

around 4.98 percent during 2004-05. Contrary to this, the growth rate of the production of

the SSI has increased from around 3.10 percent during 1991-‘92 to around 11.32 percent

during 1995-‘96 which has declined to around 6 percent during 2001-‘03 which however

has increased to around 10 percent during 2004-05. Similar to the growth rate of the

number of units the growth rate of the employment level also has remained almost stable

between 4-5 percent per annum over the years. The growth rate of the exports by SSI has

declined steadily over time. It was around 43.66 percent during 1991-92 which declined

steadily and reached the lowest level of around 2 percent during 2001-02 which however

has improved and was around 21 percent during 2002-03.

Table 2.10 Growth rate of certain indicators Growth Rates

Year Units Fixed Investment

Production (Constant price) Employment Export

1991-92 4.07 7.26 3.10 4.83 43.66 1992-93 4.08 9.24 5.60 5.33 28.10 1993-94 4.05 5.63 7.10 4.46 42.30 1994-95 4.07 6.90 10.10 4.80 14.86 1995-96 4.07 1.58 11.40 3.41 25.46 1996-97 4.07 3.83 11.32 4.01 7.62 1997-98 4.06 2.05 8.43 3.55 13.23 1998-99 4.07 1.68 7.70 3.47 10.21 1999-00 4.06 3.32 8.16 3.88 10.66 2000-01 4.07 4.90 8.23 4.20 28.78 2001-02 4.07 5.11 6.08 4.44 2.07 2002-03 4.07 5.16 7.68 4.36 20.73 2003-04 4.07 4.87 8.59 4.31 2004-05 4.07 4.98 9.96 4.11

‘Jobless growth’ is a matter of concern for the Indian policy makers since liberalization.

While large industries sector has not been able to create much employment opportunities,

rather poor performance of the agricultural sector is also forcing the farmers to move out.

Services sector and SSI sector in this regard give the economy some hope. In fact, SSI

sector in India creates largest employment opportunities for the Indian populace, next

only to Agriculture. It has been estimated that 100,000 rupees of investment in fixed

229

assets in the small-scale sector generates employment for four persons. Unlike the

stagnation of employment observed in the large-scale sector, small industries

employment reveals steady rise (Table 2.9 and 2.10). Industry group-wise food products

industry has ranked first in generating employment, providing employment to 0.48

million persons (13.1%). The next two industry groups in terms of employment

generation are non-metallic mineral products with employment of 0.45 million persons

(12.2%) and metal products with 0.37 million persons (10.2%).

Next to employment, forex reserve is India’s another important concern and the role of

SSI sector cannot be undermined in this regard. The role of SSI Sector in forex earning

through exports is well recognized17 (Table 2.11). Direct exports from the SSI Sector

account for nearly 35% of total exports. Besides direct exports, it is estimated that small-

scale industrial units contribute around 15% to exports indirectly. Thus in all, 45%-50%

of the Indian exports is contributed by this Sector. This takes place through merchant

exporters, trading houses and export houses. These may also be in the form of export

orders from large units or the production of parts and components for use for finished

exportable goods.

While Indian small-scale segment is believed to be dominated by traditional goods,

which attracts the foreign consumers, in reality non-traditional products account for more

than 95% of the SSI exports. In particular, export growth has been fuelled mainly by the

performance of garments, leather and gems and jewelry units from this sector. A few

other product groups where the SSI sector dominates in exports are, sports goods, woolen

garments and knitwear, plastic products and processed food. Further, the SSI sector is

reorienting its export strategy towards the new trade regime being ushered in by the

WTO.

The total export of the SSI was Rs 9664 crores in 1991-92 which increased to Rs

36470 crores during 1995-96 which further increased to Rs 86012 during 2002-03. The

per unit exports was around Rs 14 thousand which increased to Rs 44 thousand during

17 Source: www.smallindustriesindia.com

230

1995-95 which further increased to Rs 78 thousand during 2002-03. Share of Ssi in our

total export is also considerable (Table 2.11).

Table 2.11 Share of SSI exports in India

Total Exports

Year (in crores* of

rupees)

Exports from SSI sector (in

crores* of rupees) Percentage Share

1971-72 1608 155 9.6 1976-77 5142 766 14.9 1981-82 7809 2071 26.5 1986-87 12567 3644 29 1991-92 44040 13883 31.5 1992-93 53688 17785 33.1 1993-94 69547 25307 36.4 1994-95 82674 29068 35.1 1995-96 106353 36470 34.2 1996-97 118817 39249 33.4 1997-98 126286 44442.18 35.19 1998-99 141603.53 48979.23 34.59 1999-00 159561 54200.47 33.97 2000-01 202509.7 69796.5 34.47 2001-02 207745.56 71243.99 34.29

*1 crore = 10 million. Source : Small Scale Industries in India, Ministry of SSI, Government of India.

When the performance of this sector is viewed against the growth in the manufacturing

and the industry sector as a whole, it instills confidence in the resilience of the small-

scale sector. Growth performance of the SSI sector is far higher than the large-scale

industries sector and the manufacturing sector, as can be viewed from Table 2.12.

231

Table 2.12 Comparative real growth of overall industrial sector and SSI sector in India (1990-91 to 1999-2000) (in percent) Year Overall Industry Manufacturing

Sector SSI Sector

1990-91 8.2 9 9.1 1991-92 0.6 -0.8 3.1 1992-93 2.3 2.2 5.6 1993-94 6 6.1 7.1 1994-95 8.4 8.5 10.1 1995-96 12.8 13.8 11.4 1996-97 5.6 6.7 11.3 1997-98 6.6 6.7 8.4 1998-99 4 4.4 7.7 1999-00 E 6.4 7 8.1 Abbr.: E : Estimated. Note : Estimated figures of growth for industry and manufacturing sector based on advance estimates released by Central Statistical Organisation. Growth rates from 1994-95 onwards are as per the IIP base 1993-94 = 100 and those for earlier years are as per IIP base : 1980-81 = 100. Estimation for the SSI sector for 1999-2000 made by SIDBI. Source : Report of the Study Group on Development of Small Scale Enterprises, Planning Commission, March 2001, Govt. of India. Year: Period of fiscal year in India is April to March, e.g. year shown as 1990-91 relates to April 1990 to March 1991.

The above indicators reveal that SSIs have made significant progress over the years and

the sector has emerged as a dynamic and vibrant sector in the Indian economy. Industrial

policy of the Government, both at the center as well as at the state level, have

continuously tried to boost this sector in order to impart more vitality and growth-

impetus.

Sickness in Small Scale Industries

The Reserve Bank of India (RBI) was instrumental in appointing a number of

Committees from time to time to look into the issue of sickness affecting this sector. The

232

latest definition of ‘Sickness’ given by the ‘Working Group on Rehabilitation of Sick

Units’ set up by the RBI (also known as Kohli Committee) is given below.

“ A small scale industrial unit is considered sick when

(a) any borrowal accounts of the unit remain substandard for more than six months

or,

(b) there is erosion in the net worth due to accumulated losses to the extent of 50 percent

of its net worth during the previous accounting year, and

(c) the unit has been in commercial production for at least two years.”

In order to measure incipient sickness, the continuous decline in gross output for three

consecutive years was identified as a suitable indicator. Subsequently, the following

criteria were adapted to identify sick/ incipient sick units in the third census: i)

continuous decline in gross output compared to the previous two financial years; ii) delay

by more than 12 months in repayment of loan taken from institutional sources; and iii)

erosion in the net worth to the extent of 50 percent of the net worth during the previous

accounting year.

Magnitude of Sickness/Incipient Sickness18

Sickness identified in the registered SSI sector in terms of delay in repayment of loan or

erosion in the net worth was of the order of 2.5 %, whereas in the unregistered SSI sector,

it was 0.78 %. Out of the units having outstanding loans with institutional sources like

banks and financial institutions, sickness was about 14.08 % in the registered SSI sector

as against 13.47 % in the case of unregistered SSI sector. Incipient sickness identified in

terms of continuous decline in gross output was of the order of 13.01 % in the registered

SSI sector and 7.76 % in the unregistered SSI sector according to 2001-’02 census.

Combining the three yardsticks used to measure sickness, viz; (a) delay in repayment of

institutional loan over one year, (b) decline in net worth by 50 %, and (c) decline in 18 Source: Third All India Census of Small Scale Industries, 2001-’02.

233

output during last three years, about 14.47 % of the units in the registered SSI sector were

identified to be either sick or incipient sick, while this percentage was only 8.25 in the

case of unregistered units.

Reasons for Sickness/ Incipient Sickness

In the census of 2001-’02, the units satisfying one or more of the above criteria were

treated as not being run satisfactorily and the reasons for the same were elicited. Table

2.13 indicates the reasons as given by the units suffering from sickness/ incipient

sickness. ‘Lack of Demand’ and ‘Shortage of Working Capital’ were the main reasons for

sickness, incipient or otherwise, in the SSI sector.

Table 2.13 Reasons for sickness Percentage of Units Reason

Registered Unregistered

Lack of Demand 71.6 84.1

Shortage of working capital 48 47.1

Non availability of raw

materials

15.1 15.2

Power shortage 21.4 14.8

Labour problems 7.4 5.1

Marketing problems 44.5 41.2

Equipment Problems 10.6 12.9

Management problems 5.5 5.1

*The total will exceed 100 %, as some units reported more than one reason. Source: Census of SSIs, 2001-‘02

As is observed, both registered and unregistered units face same problems that lead to

sickness. Working capital related and Marketing problems (equivalently, lack of demand)

are the major hurdles for the registered segment (that receives government support) as

well as for the unregistered firms. The problem of working capital clearly shows that

there exists a problem of credit for the SSI sector. This has been also revealed during our

survey.

234

This led us to look at the credit policy of the SSI sector.

2.8 Credit Policy with Reference to SSI

As mentioned above , at a time when there was no restriction on the lending

activities of the banking sector, lending was directed to only a certain selected activities.

To arrive at a desirable distribution of credit to core and socially important sectors, the

concept of priority sector was developed. Subsequently, as mentioned above, at a

meeting of the Union Finance Minister with the Chief Executive Officers of public sector

banks held in March 1980, a decision was taken that banks should aim at raising the

proportion of their advances to priority sectors to 40 per cent by March 1985. Following

the recommendations of the Working Group on the Modalities of Implementation of

Priority Sector Lending and the Twenty Point Economic Programme by Banks, all

commercial banks were advised to direct 40 per cent of aggregate bank advances to the

priority sector by 1985. In addition, there were sub-targets for lending to agriculture and

to the weaker sections within the priority sector. These norms are undergoing

modifications since then. In the decade of 1990s certain specific reforms have been

brought in. The 40 per cent priority sector lending requirement for net bank credit (NBC)

as applicable to PSBs as well as private sector banks continued, but liberalization of

interest rate has been introduced on loans above Re. 2 lakhs. Few other areas are also

incorporated within the purview of priority sector lending.

Unlike agriculture, there is no separate sub-target for the SSI sector, within the priority

sector lending for the Indian public and private sector banks. It is not mandatory for the

Indian banks to deposit the shortfall in lending to the SSI sector with the Small Industries

Development Bank of India (SIDBI) (as in the case of foreign banks) or any such

organisation. Sub targets however exist for agriculture and loans to the weaker sections

consisting of small and marginal farmers, artisans and others and they are 18% and 10%

respectively. In the case of cooperative bank, 60 per cent of credit comes under priority

235

sector lending. In order to fulfill the priority sector lending targets, banks have been

permitted to adopt soft approaches like subscription to the bonds of SFCs, NABARD,

National Housing Bank, Rural Electrification Corporation, Housing & Urban

Development Corporation, etc. instead of undertaking retail lending to the SSI Sector.

Availability of credit was always recognized as a constraint to the growth of the SSI

sector, be it a women or a rural enterprise. Government has so far tried to mitigate the

problem through various measures. Few committees have been formed to understand and

to come up with appropriate measures for this sector. In 1990 a separate financial

intermediary called Small Industries Development Bank of India (SIDBI) was

established. Since then SIDBI has been acting more as a small industries development

organization rather than simply as a bank. Given the special role that the bank has come

to play, some of its activities are considered worth noting.

Role of SIDBI

Promotional Programs: SIDBI’s measures for this sector are modern in approach and

intend to improve competitiveness in the sector. Some of its measures for example

include:

• Creation of awareness on new product / process technologies

• Skill upgradation

• Development of technology related common facilities for the cluster

• Provision of unit-specific modernisation package

• Energy conservation and introduction of environment friendly technologies

• Quality upgradation in terms of systems and products

It took a cluster development approach to take up various developmental measures.

Cluster Development Approach: The first step in its approach to achieve these goals

involves the selection of clusters, which have certain homogeneity in terms of status of

technology, products, production levels, trade practices, and capacity to absorb improved

236

technology. Individual clusters are then assigned to expert consultancy agencies that

assess the technology upgradation needs and prepare unit-specific modernisation

packages including scope for consolidation of technical capabilities of existing units.

Technology Upgradation Program: The competitiveness of the products of SSI units

both in the domestic and international markets is dependent to a large extent on their

productivity levels, price factors and quality characteristics. SIDBI's technology

upgardation and modernisation programme are aimed at improving the technical

capabilities and competitiveness of SSI units located in clusters by introducing

commercially proven technologies which will result in significant improvement in

quality, productivity, cost reduction, saving of energy and raw materials and reduction in

the level of pollution.

Funding: SIDBI provides support and co-ordinates the services of consultants, and backs

up their efforts by arranging financial assistance, through banks or State Financial

Corporations (SFCs), under its refinance assistance schemes. The Bank also provides

direct financial assistance through its Rs. 2 billion Technology Development and

Modernisation Fund (TDMF) scheme. The Bank undertakes regular follow-up and

monitoring of the programmes and the implementing agencies are suitably compensated

by way of professional fee for undertaking the assignment.

Progress: The Bank in more than 25 clusters has launched technology upgradation

program. The clusters identified for intervention range from Sea Food Processing

Industry (Coastal Kerala) to Brass and Bell Metal Industry (Hajo in Assam) and from

scientific instrument industry (Ambala, Haryana) to artisan based Blacksmithy units at

Mylliem, Meghalaya and so on. In addition to this, the Bank is to implement the National

Programme for Rural Industrialisation in 25 clusters of which 12 initiatives are already

underway (www.sidbi.in).

237

Even though SIDBI has taken up various measures to aid the SSI sector number of

SIDBI offices are limited and therefore they could reach only a small proportion of the

SSI units. Rest of the sector still depends n various other sources for credit.

Multi-level Financial Institutional Structure

A large number of institutions are engaged in the task of credit dispensation to the small

and other non-farm enterprises. Major national/state-level institutions operating in the

country in addition to SIDBI are:

Commercial Banks

State Financial Corporations (SFCs)

Regional Rural Banks

Cooperative Banks

Credit in Direct/indirect Form by Other Agencies.

In addition to the creation of such specialized financial intermediaries, to improve credit

flows to the SSI sector a few committees have also been constituted and in turn the

Reserve Bank of India (RBI) has taken various measures . Focus and recommendations

of few committees and RBI measures are discussed briefly below.

Nayak Committee

This committee was formed under the Chairmanship of Ex-Deputy Governor, of

RBI Shri R.R. Nayak, to look into the problems of credit, sickness and other relevant

aspects of the SSI sector. The committee submitted its report in September 1992. Based

on the Nayak Committee recommendations, Reserve Bank of India has directed the

commercial banks to modify the definition of sick SSI units and to reduce rate of interest

for rehabilitation. The committee has also suggested various rehabilitation packages.

238

Kapur Committee

The Kapur Committee was set up by the RBI to review the working of the credit

delivery system for SSIs with a view to making the system much more effective, simple

and efficient to administer. The committee had also examined the sickness related issues

of the SSI sector. For quick rehabilitation of sick SSI units, the Committee has

recommended the following:

Changing the definition of classifying the SSI unit as sick by reducing the non-

performing period of the SSI account from 2 ½ years to one year.

Converting State Level Inter Institutional Committees (SLIICs) into statutory

bodies under a special statute to enable them to play effective role in

rehabilitation of sick SSIs.

Setting up branches of SLIICs in districts having large concentration of SSIs.

Providing relaxation in income recognition and asset classification amounts to

encourage banks to take up rehabilitation of potentially viable sick SSIs.

Measures taken by Reserve Bank of India

RBI has issued detailed guidelines vide their circular dated 17th April, 1993 and

3rd July, 1993 to banks for rehabilitation of sick SSI units including detection at the

incipient stage and to take remedial measures, including the broad parameters for grant of

relief and concessions such as:

Interest on Working Capital 1.5 % below the prevailing fixed/PLR, wherever

applicable Funded Interest Term Loan Interest free Working Capital Term Loan 1.5% below the prevailing fixed /PLR, wherever

applicable. Term Loan Concession up to 2% ( Not more than 3% in the case of

tiny/decentralised sector units) below the document rate.Contingency/Loan Assistance

Concessional rates for working capital assistance.

2.9 Commercial Banks and Credit to SSI

239

One of the important sources of credit to small-scale industries is the ercial bank.

The total credit from commercial banks to SSI sector has increased from Rs 18939 crores

in 1991-92 to Rs 34246 crores during 1995-96, which further increased to Rs 60141

crores during 2000-01 and to 90239 crores during 2005-06. Though at the absolute level,

credit has been increasing over time the growth rate shows a varying trend. The annual

growth rate of commercial banks’ credit to SSI sector increased from around 5.58 percent

in 1991-92 to around 21.67 percent during 1994-95 which almost steadily declined

thereafter and reached the lowest level of -3.58 percent during 2003-04. This however

improved in the subsequent years. The credit to SSI sector as a percent of total net bank

credit has been steadily declining over time. It was around 16.13 percent during 1991-92

which declined to around 15 percent during 1995-96 which further declined to 12.87

percent during 2000-01 and to 6.66 percent during 2005-06 (Table 2.14).

Table 2.14 Flow of Credit from Commercial Banks to SSI (Rs crore)

Net Bank Credit

Annual growth

Credit to SSI

Annual growth

Credit to SSI as percent of Net Bank Credit

1991-92 117443 7.45 18939 5.58 16.13 1992-93 141800 20.74 20975 10.75 14.79 1993-94 152501 7.55 23978 14.32 15.72 1994-95 192424 26.18 29175 21.67 15.16 1995-96 228198 18.59 34246 17.38 15.01 1996-97 245999 7.80 38196 11.53 15.53 1997-98 297265 20.84 45771 19.83 15.40 1998-99 339477 14.20 51679 12.91 15.22 1999-00 398205 17.30 57035 10.36 14.32 2000-01 467206 17.33 60141 5.45 12.87 2001-02 535063 14.52 67107 11.58 12.54 2002-03 668576 24.95 64707 -3.58 9.68 2003-04 763855 14.25 71209 10.05 9.32 2004-05 971809 27.22 83179 16.81 8.56 2005-06 1354603 39.39 90239 8.49 6.66

At the bank group level , the State Bank and Associates group also shows similar trend.

There has been substantial decline of share of SSI credit in total priority sector lending

(Table 2.15). This decline is even more striking for the private sector banks. Share of SSI

credit in the case of Foreign banks however, remained more or less stable due to the

240

special norms apply to them regarding priority sector lending. We recall that for the other

bank groups there is no special norm fixed for the SSI sector separately.

Table 2.15

Bank GroupWise Advances to Small Scale Industries 1997 1998 1999 2000 2001 2002 2003 2004

Public Sector Banks Small Scale Industries (Rs Crore) 31542 38109 42674 45788 48445 49743 52988 58278 Per cent to Total Priority sector credit 39.86 41.73 39.81 35.83 33.06 29.06 26.09 23.72 Per cent to total non-food credit 16.63 17.46 17.33 15.63 14.21 12.53 11.09 10.43

Private Banks Small Scale Industries (Rs Crore) 4754 5848 6451 7313 8158 8613 6857 7897 Per cent to Total Priority sector credit 53.83 50.35 45.57 40.58 37.86 33.50 18.68 14.94 Per cent to total non-food credit 22.15 20.60 18.86 15.71 14.40 13.70 8.29 7.08

Foreign Banks Small Scale Industries (Rs Crore) 1836 2084 2460 2871 3716 4561 3809 5438 Per cent to Total Priority sector credit 29.91 30.03 29.75 29.60 31.40 34.00 25.65 29.75 Per cent to total non-food credit 11.29 10.30 11.01 10.36 10.83 11.56 8.70 10.35

Diversion of credit away from the SSI sector may be due to the prevalence of

sickness of the SSI units. If one examines carefully it is observed that the number of sick

SSI units increased from 221472 in 1991 to around 268815 in 1995 (Table 2.16), which

has declined thereafter to 221536 in 1998. After a sharp increase to 306221 in 1999, the

total number of sick units has declined steadily thereafter and it was 138811 in 2004.

However, the value involved in the sick SSI units increased steadily over the years. It was

Rs 2792 crores in 1991, which increased to Rs 3547 crores in 1995 and further climbed

up to Rs 4608 in 2000 and to Rs 5285 in 2004. Since on the one hand the number of sick

SSI units is declining, and on the other hand the amount involved is increasing, the

241

amount involved per units has increased steadily over time. It was around Rs 1.26 lakh

per unit in 1991, which increased to Rs 1.74 lakh in 1998 which further increased to Rs

3.80 lakh in 2004.

Though the total sick SSI units show a varying trend, the sick SSI units that are

potentially viable are steadily declining over time. It was 16140 in 1991 that has declined

to 15539 in 1995, which has further gone down to 14373 in 2000 and to 2385 in 2004.

Unlike the case of total sick SSI units, the amount involved in the potentially viable sick

SSI units has been declining over time. It was Rs 693 crores in 1991 which reduced to Rs

597.93 crores in 1995, which further declined to Rs 369 corores in 2006; One however,

observe a marginal increase to Rs 421 crores in 2004. The amount involved per unit of

the potentially viable sick SSI has steadily declined from Rs 4.29 lakh in 1991 to Rs 2

lakh in 1999, however, it has increased sharply there after and it was Rs 17.6 lakh in

2004.

Table 2.16 Sickness in Small Scale industrial Sector

Total sick units Potentially viable Per cent of

Potentially Viable in Total Sick Units

Year No. Amount (Rs.

Crores)

Amount Per unit

(Rs Lakh)

No.* Amount O/s (Rs. Crores)

Amount Per unit

(Rs Lakh)

No. Amount (Rs.

Crores)

1991 221472 2792 1.261 16140 693.12 4.294 7.29 24.83 1992 245575 3101 1.263 19210 728.9 3.794 7.82 23.51 1993 238176 3443 1.446 21649 798.79 3.690 9.09 23.20 1994 256452 3680 1.435 16580 685.93 4.137 6.47 18.64 1995 268815 3547 1.320 15539 597.93 3.848 5.78 16.86 1996 262376 3722 1.419 16424 635.82 3.871 6.26 17.08 1997 235032 3609 1.536 16220 479.31 2.955 6.90 13.28 1998 221536 3857 1.741 18686 455.96 2.440 8.43 11.82 1999 306221 4313 1.409 18692 376.96 2.017 6.10 8.74 2000 304235 4608 1.515 14373 369.45 2.570 4.72 8.02 2001 249630 4506 1.805 13076 399.17 3.053 5.24 8.86 2002 177336 4819 2.717 4493 416.41 9.268 2.53 8.64 2003 167980 5706 3.397 3626 624.71 17.229 2.16 10.95 2004 138811 5285 3.807 2385 421.18 17.660 1.72 7.97

Source: Annual Report 2005-06; Ministry of SSI, Government of India *These units include village industries as well

242

The major reason for the non performing assets arising from the SSI sector is their

sickness. While there may be some proportion of willful default, unless one handles the

sickness issue appropriately , reducing the bad loan from the sector becomes difficult.

Banks will then take recourse to diverting their resources to other sectors in the economy.

In the next Chapter therefore we discuss the NPA issue in general and that arising from

the SSI sector in particular.

2.9 Concluding Remarks

The Indian banking sector has come a long way since the last century. Even since the

nationalization of banks took place, the operations of commercial banks spread like never

before. It covered all nook and corner of the nation. Accordingly credit disbursement and

deposit mobilization have increased at a phenomenal rate. Various sectors especially

agriculture sector was given priority in the lending of the commercial banks. Such norms

have helped the sector to come out of the grasp of moneylenders to a large extent if not

completely. Such social banking norms also however, have affected the financial health

of the banks.

243

With the liberalization of the Indian economy, various financial sector liberalization

norms have also been introduced. These are aimed at improving the efficiency and

profitability of the public sector banks. A number of reform measures have been

introduced to provide autonomy to the banks (see Chapter 1). While the liberalized

measures improved efficiency of the commercial banks, rural branches, credit share to

some of the priority sectors such as SSI sector declined over time. It has often been

argued that the NPA accounts mainly arise from the rural branches and priority sector

loans. While there is some truth to these allegations, moving away from these sectors

may not be an ideal solution for the economy. Further it is also important to ask whether

these sectors alone need to be blamed for NPA? To examine this we devote our attention

to the nature and extent NPA of the Indian commercial banks in the next chapter.

244

CHAPTER 3

Non-Performing Assets of Indian Commercial Banks

3.1 Introduction

High level of Non-performing Assets (NPAs) is a matter of concern for everyone

involved as credit is essential for economic growth and NPAs affect the smooth flow of

credit. Banks raise resources not just from fresh deposits, but they also create credit by

recycling the funds received back from the borrowers. Thus when a loan becomes non-

performing, it affects recycling of credit and in turn credit creation. Apart from the credit

creation, NPAs affect the profitability as well, since higher NPAs require higher

provisioning, which means a large part of the profits needs to be kept aside as provisions

for bad loans. Therefore, the problem of NPAs is not the concern of the lenders alone,

but it a concern of policy makers as well who are involved in putting economic growth on

the fast track. In India the concept of NPA came into existence after the financial sector

reforms were introduced following the recommendations of the Report of the Committee

on the Financial System (Narasimham, 1991).

Broadly, Non Performing Advance is defined as an advance where payment of

interest or repayment of installment of principal (in case of term loans) or both remains

unpaid for a certain period19. In India, the definition of NPAs has changed over time.

According to the Narasimham (1991) committee report, those assets (advances, bills

discounted, overdrafts, cash credit etc) for which the interest remain due for a period of

four quarters (180 days) should be considered as NPAs. Subsequently this period was

reduced, and from March 1995 onwards the assets for which the interest has remained

unpaid for 90 days should be considered as NPAs.

NPA being our prime concern in this study we intend to look at the trends and nature of

NPAs for the Indian economy in some detail in this chapter. However, before coming to

19 This time duration given for an asset to consider it as a NPA varies from country to country and can change over time within a particular country.

245

the Indian scenario it is useful to examine the NPA problem for other nations across the

globe to see where India stands vis-à-vis others.

3.2 NPAs at the Global Level

In order to get a global picture it is essential to look at NPA levels of different

countries in the world. Since the concept of NPA is developed in India only in the post-

reform era , it is useful to look at the recent figures rather than taking a historical account.

A closer look at the Non-performing Loans (NPL) as they are called in many nations,

reveals that during 2003 the NPL at the global level was US$ 1300 billion. India ranks

fourth with NPL of around US$ 30 billion ( 2.3 percent of the global NPL), while Japan

has the highest NPL of US$ 330 billion (25.4 percent of the global NPL) Turkey has the

lowest NPL of US$ 8 billion (0.6 percent of global NPL, Table 3.1).

Table 3.1

Global Non-performing Loans: 2003 Countries NPLs (US$ billion) Share in Global (per cent) Japan 330 25.4 China 307 23.6 Taiwan 19.1 1.5 Thailand 18.8 1.5 Philippines 9 0.7 Indonesia 16.9 1.3 India 30 2.3 Korea 15 1.2

Germany 283

21.8 Turkey 8 0.6 Global 1300 100

Source: Global NPL Report 2004, Ernst and Young.

Though one can get an idea about the magnitude of NPA by looking at the

absolute values, it does not reveal the complete picture. This is mainly because absolute

level of NPA depends on total advances. A country with larger population or GDP may

have large advances and in turn NPA as well. Thus apart from the absolute value, it is

also important to look at what proportion of the total loan has become non-performing.

The NPL levels of various countries as per cent of their total loan are presented in table

3.2. It can be seen from the table that the NPA/NPL as a percent of total loans has been

246

declining for almost all countries over the years. The average NPL as percent of total

loans across the countries was around 11.89 percent in 2001 which has declined to

around 6.44 percent in 2005. This shows that the quality of bank asset has been

improving across countries over the years. This could be because of the stringent

regulations prescribed by the BASEL norms. Examining the countries in terms of NPA

as percentage of total loans, we observe that for around 16 countries NPA percentage is

below 10 and for around 5 countries it is between 10-20 percent; for another 5 countries

NPA percentage is rather high and above 20 percent. Australia has the lowest NPA to

total loan ratio which is just 0.34 percent whereas Philippines tops the list with 25

percent. India attains the 11th highest position with around 8.6 per cent. One interesting

observation is that, most of the countries which fall under higher ‘NPL/Total Loan’ ratio

category belong to the Asian region. Out of 10 countries which have this ratio above 10

percent 8 countries belong to Asia. The improvement in the quality of the assets across

countries is also shown by the fact that in 2001 there were around 11 countries whose

NPA/Total Loan ratio was above 10 percent, by 2005 this number reduced to 7.

Table 3.2 Bank Non-performing Loans to Total Loans

Countries 2001 2002 2003 2004 2005 Australia 0.6 0.4 0.3 0.2 0.2 Bangladesh 31.5 28 22.1 17.6 15.3 Brazil 5.6 4.8 4.8 3.8 4.4 Canada 1.5 1.6 1.2 0.7 0.5

247

Chile 1.6 1.8 1.6 1.2 0.9 China 29.8 25.6 20.1 15.6 10.5 Egypt 16.9 20.2 24.2 16.3 25 France 5 5 4.8 4.2 3.5 Germany 4.6 5 5.3 5.1 4.8 Hong Kong 6.5 5 3.9 2.3 1.5 India 11.4 10.4 8.8 7.2 5.2 Indonesia 31.9 24 19.4 14.2 15.6 Japan 8.4 7.2 5.2 2.9 1.8 Korea 3.4 2.4 2.6 1.9 1.2 Malaysia 17.8 15.8 13.9 11.8 9.9 Mexico 5.1 4.6 3.2 2.5 1.8 Pakistan 23.4 21.8 17 11.6 10.6 Philippines 27.7 26.5 26.1 24.7 20 Russia 6.2 5.6 5 3.8 3.2 Singapore 8 7.7 6.7 5 3.8 Sri Lanka 15.3 15.3 13.7 9.1 9.6 Switzerland 2.3 1.9 1.4 0.9 0.5 Thailand 11.5 16.5 13.5 11.8 11.1 Turkey 29.3 17.6 11.5 6 4.8 United Kingdom 2.6 2.6 2.5 1.9 1 United States 1.3 1.4 1.1 0.8 0.7 Source: Global Financial Stability Report, May 2006, IMF

While comparing the NPA levels of different countries with each other one should

remember that the features relating to the NPA reporting/evaluation practices are not

uniform across countries. In some countries the NPA level may be low because their

losses are written off at an early stage. In some of the developing countries of ‘Asia

Pacific Countries belonging to Economic co-operation (APEC) forum’, a loan is

classified as non-performing only after it has been in arrears for at least six months;

whereas, in India, currently an asset is considered NPA if it is due for 90 days. Also in

India due to the lengthy legal process, it takes considerably long time to recover the loan.

And, due to many safeguards/procedures, even after a NPA is written off banks continue

to hold them in their books, many a time along with the provision made for those loans.

Even the classification of NPA into Gross NPA and Net NPA is not uniform because, in

some countries the provisions made are general provisions whereas, in India NPAs are

considered GNPA for some time even after making provisions. Thus while comparing the

NPA level of India with other countries one should remember that in many respect, asset

248

classification norms in India are considerably tighter than the international best practices.

The classification standards adopted in a few countries are given in the Appendix (Table

A3.7).

In addition, countries also do differ in various other respects. In that sense a strict

comparison across countries cannot be arrived at. Nonetheless the global picture do

reflect a comprehensive view of NPAs across the world.

3.3 NPA norms and Non Performing Assets in India

Though the issue of NPA was given more importance after the Narasimham

committee report (1991) highlighted its impact on the financial health of the commercial

banks and subsequently various asset classification norms were introduced, the concept

of classifying bank assets based on its quality began during 1985-86 itself. A critical

analysis for a comprehensive and uniform credit monitoring was introduced in 1985-86

by the RBI by way of the Health Code System in banks which, inter alia, provided

information regarding the health of individual advances, the quality of credit portfolio

and the extent of advances causing concern in relation to total advances. It was

considered that such information would be of immense use to bank managements for

control purposes. Reserve Bank of India advised all commercial banks (excluding foreign

banks, most of which had similar coding system in their organisations) on November 7,

1985, to introduce the Health Code classification system indicating the quality (or health)

of individual advances in the following eight categories, with a health code assigned to

each borrowal account:

1. Satisfactory - conduct is satisfactory; all terms and conditions are complied with;

all accounts are in order; and safety of the advance is not in doubt.

2. Irregular- the safety of the advance is not suspected, though there may be

occasional irregularities which may be considered as a short term phenomenon.

3. Sick, viable - advances to units which are sick but viable - under nursing and units

in respect of which nursing/revival programmes are taken up.

249

4. Sick: nonviable/sticky - the irregularities continue to persist and there are no

immediate prospects of regularisation; the accounts could throw up some of the

usual signs of incipient sickness

5. Advances recalled - accounts where the repayment is highly doubtful and nursing

is not considered worth-while; includes where decision has been taken to recall

the advance

6. Suit filed accounts - accounts where legal action or recovery proceedings have

been initiated

7. Decreed debts - where decrees (verdict) have been obtained.

8. Bad and Doubtful debts - where the recoverability of the bank's dues has become

doubtful on account of short-fall in value of security; difficulty in enforcing and

realising the securities; or inability/unwillingness of the borrowers to repay the

bank's dues partly or wholly

Under the above Health Code System RBI was classifying problem loans of each

bank in three categories i.e. i) advances classified as Bad & Doubtful by the bank

(corresponding to Health Code No.8) (ii) advances where suits were filed/decrees

obtained (corresponding to Health Codes Nos.6 and 7) and (iii) those advances with

major undesirable features (broadly corresponding to Health Codes Nos.4 and 5).

The Narasimham Committee (1991) felt that the classification of assets according

to the health codes is not in accordance with the international standards. It believed that a

policy of income recognition should be objective and based on record of recovery rather

than on any subjective considerations. Also, before the capital adequacy norms are

complied with by Indian banks, it is necessary to have their assets revalued on a more

realistic basis on the basis of their realizable value. Thus the Narasimham committee

(1991) believed that a proper system of income recognition and provisioning is

fundamental to the preservation of the strength and stability of the banking system.

The international practice is that an asset is treated as non-performing when

interest is due for at least two quarters. In respect of such non-performing assets interest

is not recognized on accrual basis but is booked as income only when it is actually

received. The committee suggested that a similar practice should be followed by banks

250

and financial institutions in India and accordingly recommended that interest on non-

performing assets should be booked as income on accrual basis. The non-performing

assets would be defined as an advance where, as on the balance sheet date:

4. In respect of overdraft and cash credits, accounts remain out of order for a period

of more than 180 days,

5. In respect of bills purchased and discounted, the bill remains overdue20 and

unpaid for a period of more than 180 days,

6. In respect of other accounts, any account to be received remains past due for a

period of more than 180 days.

As mentioned earlier, the grace period was reduced, and from March 1995

onwards the assets for which the interest has remained unpaid for 90 days should be

considered as NPAs. Provisions need to be made for the NPAs and total NPA (gross)

minus the provisions is defined as net NPA.

Along with providing the detailed definition of the Non-performing Asset, the

Narasimham committee (1991) also suggested that for the purpose of provisioning, banks

and financial institutions should classify their assets by compressing the health codes into

the four broad groups; (i) Standard (ii) Sub-standard, (iii) Doubtful and (iv) Loss.

Broadly stated, sub-standard assets would be those which exhibit problems and would

include assets classified as non-performing for a period not exceeding two years.

Doubtful assets are those non-performing assets which remain as such for a period

exceeding two years and would also include loans in respect of which installments are

overdue for a period exceeding two years. Loss assets are accounts where loss has been

identified but amounts have not been written off.

20 An amount is considered overdue when it remains outstanding 30 days beyond the due date.

251

One of the related and important aspects of NPAs is the provisioning. According

to the international norms, commercial banks need to keep aside a portion of their income

as provision for the bad loans. The amount of provision depends on the type of the NPAs

and also the time duration. The Narasimham committee (1991) believed that in the Indian

context, given the delays in the legal system, there is bound to be a time lag between an

account becoming doubtful of recovery, its recognition as such, and the realization of the

security. This factor has to be kept in mind in making provisions, besides market value of

the security charged to the banks and institutions. The committee, therefore, recommends

that the basis of provisioning against bad and doubtful debts should be as under:

1. In respect of loss assets either the entire assets should be written off in the books

or if the asset is permitted to remain in the books for certain reasons, 100 percent

of the outstanding should be provided for.

2. In respect of doubtful debts, it should be necessary for the banks to provide 100

percent of the security shortfall, that is, the full extent to which the loans and

advances are not covered by the realizable value of the security. Over and above

this, it will be necessary for banks and institutions to make a further specific

provision to the extent of a certain percentage of even the secured portion. This

percentage could vary from 20 to 50 percent depending on the period for which an

asset remains in the doubtful category.

3. In respect of sub-standard assets, a general provision of 10 percent of the total

outstanding should be created.

3.4 Recovery Mechanism of NPA

It was felt by the Government of India that the usual recovery measures like issue

of notices for enforcement of securities and recovery of dues is a time consuming

process. Thus, in order to speed up the process of recovery of NPAs, the Government of

252

India constituted a committee under the chairmanship of late Shri Tiwari in 1981. The

committee examined the ways and means of recovering NPAs and recommended, inter

alia, the setting up of ‘Special Tribunals’ which could expedite the recovery process.

Later the Narasimham Committee (1991) endorsed this recommendation, and also,

suggested setting up of Asset Reconstruction Fund (ARF). The Government of India, if

necessary should establish this fund by special legislation, which would take over NPAs

from banks and financial institutions, at a discount, and follow up on the recovery of the

dues owed to them from the primary borrowers.

Based on the recommendations of the Tiwari committee and also of the

Narasimham Committee, various Debt Recovery Tribunals were established at various

parts of the country. An Asset Reconstruction Company was also established. At present,

various measures taken to reduce NPAs include reschedulement, restructuring at the bank

level, corporate debt restructuring, and recovery through Lok Adalats, Civil Courts, and

Debt Recovery Tribunals and Compromise Settlements. Some of these measures are

discussed briefly here.

3.4.1 Debt Recovery Tribunals

It was felt by the Tiwari committee (1981) that the civil courts are burdened with

diverse types of cases. Recovery of dues due to banks and financial institutions is not

given priority by the civil courts. The banks and financial institutions like any other

litigants have to go through a process of pursuing the cases for recovery through civil

courts for unduly long periods. By 30th Sept 1990, more than 15 lakh cases filed by the

public sector banks and about 304 cases filed by the financial institutions were pending in

various courts. Recovery of debts involved more than Rs 5622 crores in dues of public

sector banks and about Rs 391 crores in dues of the financial institutions. Thus this

committee recommended for establishment of Debt Recovery Tribunals (DRTs) which

was later endorsed by the Narasimham Committee. Subsequently, the “Recovery of

Debts Due to Banks and Financial Institutions Act, 1993” was passed which enabled the

establishment of DRTs. The DRT Act seeks to provide expeditious adjudication and

speedy recovery of dues to banks and financial institutions. Presently, there are 29 DRTs

and 5 DRATs functioning all over the country. The pecuniary jurisdiction of these

253

Tribunals is Rs.10 lakhs and above. Debts Recovery Tribunals also have jurisdiction over

appeals against any of measures taken by the secured creditor or by the authorized

officer.

In order to make the process faster and to correct the legal irregularity, the

Government of India passed an amendment to the DRT Act in 2000. The new statute

gives more power to the DRTs in terms of attaching the property of the defaulter and also

distribute the sale proceedings of the defaulter’s property among secured creditors.

3.4.2 Asset Reconstruction Company (India) Limited

After recognizing the problems of NPAs, various measures were taken to reduce

the NPA levels in the future. However, by then, commercial banks and other financial

institutions had already accumulated enough NPAs. It was necessary to free commercial

banks and other financial institutions from the problems of recovering NPAs so that they

can concentrate on the regular banking business. In this regard, in addition to DRTs,

Government of India also decided to set up an agency which can acquire NPAs from

commercial banks and recover them. This was also suggested by the Narasimham

Committee report where they mentioned about setting up of an Asset Reconstruction

Fund. ARCIL is established under the SARFAESI Act 2002 and is registered with

Reserve Banks of India and commenced business from 2003. The RBI has so far issued

Certificate of Registration to four Securitasation Companies/Reconstruction Companies

(SCs/RCs), of which three have commenced their operations. Assets Reconstruction

Company (India) Limited (ARCIL) set up in 2003 has provided a major boost to the

efforts to recover the NPAs of the banks. During 2005-06, ARCIL acquired 559 cases of

NPAs amounting to Rs 21,126 crores. The portfolio of assets acquired by ARCIL was

diversified across major industry segments, which were generally performing well in the

stock market. About 78 per cent of assets under management were fully/partially

operational. There are certain important legal reforms as well to expedite the process of

loan recovery and SARFAESI act is a significant step in this direction.

3.4.3 SARFAESI Act

254

The legal mechanism for recovery of default loans by usual procedure was too

cumbersome and time consuming. Thus it was felt that banks and financial institutions

should be given power to take over securities and sell in order to recover the dues. In this

regard the Government of India appointed a committee under the chairmanship of Shri T

R Andhyarujina, senior Supreme Court advocates and former Solicitor General of India

in the year 1999 to look into these matters. The Committee submitted four reports. One

among them is related to securitization.

Based on the recommendations of the Andhyarujina committee The Securitisation And

Reconstruction of Financial Assets And Enforcement of Security Interest (SARFAESI)

Act, 2002, was enacted on 17th December 2002. The act provides enforcement of security

interest without taking recourse to civil courts. This act was passed with the aim of

enabling banks and financial institutions to realise long-term assets, manage problem of

liquidity, reduce asset liability mismatches and improve recovery by exercising powers to

take possession of securities, sell them and reduce non-performing assets by adopting

measures for recovery or reconstruction. The ordinance also allows banks and financial

institutions to utilise the services of ARCs/SCs for speedily recovering dues from

defaulters and to reduce their non-performing assets. The ordinance contains provisions

that would make it possible for ARCs/SCs to directly take possession of the secured

assets and/or the management of the defaulting borrower companies without having to

resort to time consuming process of litigation and without allowing borrowers to take

shelter under provisions of SICA/BIFR. In addition to passing SARFAESI Act certain

other legal reforms have also been introduced to speed up the loan recovery process.

3.4.4 Other Legal Reforms

One of the important factors responsible for continuing high level of NPAs in the

Indian banking industry is the weak legal system. According to an international rating

agency called FITCHIBCA “The Indian legal system is sympathetic towards the

borrowers and works against the banks' interest. Despite most of their loans being backed

by security, banks are unable to enforce their claims on the collateral, when the loans turn

non-performing, and therefore, loan recoveries have been insignificant”.

255

However efforts have been made to rectify these problems in terms of judicial

process as well as laws. In 1999 a standing committee under the aegis of Industrial

Development Bank of India (IDBI) was constituted to have a coordinated approach in the

recovery of large NPA accounts, as also for institutionalising an arrangement for a

systematic exchange of information in respect of large borrowers (including defaulters

and NPAs) common to banks and financial institutions. And, as mentioned above, in

2002 Securitisation and Reconstruction of Financial Assets and Enforcement of Security

Interest Act (SARFAESI Act) was passed which empowers the creditors to foreclose

non-performing loans and the underlying collateral without going through a lengthy court

or tribunal process (Basu, 2005). All these efforts have improved the recovery of NPAs

by commercial banks which has in turn helped in reducing the NPA level. The total

NPAs recovered through various channel was around Rs 4039 crores during 2003-04

which has increased by many fold to Rs 20578 crores during 2004-05.

3.4.5 Recovery of NPA

Using the new institutions and legal options banks and financial institutions have

accelerated their recovery of NPAs. The NPAs recovered by scheduled commercial banks

through various channels is presented in table 3.3. Between 2003-04 and 2005-06, the

total cases referred to various institutions are 932377 which related to the total amount of

around Rs 70226 crores out of which around Rs 19075 crore was recovered. In terms of

the number of cases, highest number of cases (553042) were referred to Lok Adalats and

lowest cases (15812) were referred to DRTs. Whereas, in terms of the amount involved

DRTs deal with the highest amount of around Rs 32745 crores and Lok Adalats deal with

lowest amount of around Rs 2965 crores. In terms of the recovery, one-time

settlement/compromise schemes seems to be doing better as around 58 percent of the

amount involved was recovered. DRTs recovered around 29 percent and Lok Adalats

recovered around 16 percent of the amount involved. And, around 22 percent of the

amount involved was recovered under SARFAESI act.

Table 3.3

256

NPAs Recovered by SCBs Through Various Channels

One-time settlement/

compromise Scheme

Lok Adalats DRTs SARFAESI Act

2003-04 No of cases referred 139562 186100 7544 2661 Amount involved 1510 1063 12305 7847 Amount recovered 617 149 2117 1156 2004-05 No of cases referred 132781 185395 4744 39288 Amount involved 1332 801 14317 13224 Amount recovered 880 113 2688 2391 2005-06 No of cases referred 10262 181547 3524 38969 Amount involved 772 1101 6123 9831 Amount recovered 608 223 4710 3423

Source: RBI

Thus we observe that considerable attention has been paid to the NPA issue and various

regulatory as well as institutional mechanisms are put in place. How effective are these

changes? This calls for a closer look at the NPA trends in the recent past.

3.5 Non Performing Assets in India

While efforts are on for NPA classifications, refinement of accounting system

and measures to reduce NPA in the decade of 1990s , proper practice of these norms took

time. Systematic data on NPAs started to become available in a usable form from

1998 only. Though the total GNPA has increased significantly between 1998 and 2002, it

has started to decline after that (Table 3.4, Table A3.121). During 1998 the total Gross

NPA and Net NPA of the total banking sector was Rs. 34428 crore (around 14.4 percent

of gross advance) and Rs. 16098 crores (around 7 per cent of net advance) respectively.

During 2005, the GNPA increased even in real terms to Rs 38558 crores (around 5.2

percent of gross advance) whereas, NNPA has reduced to 14181 crores (around 2 percent

21 Nominal Figures are shown in the Appendix (see Table A3.1- A3.6)

257

of net advance). The growth rate of GNPA was about 11 percent in 1999 to which started

to fall drastically and became negative after 2002. Growth rate becoming negative

implies that there is a substantial decline in the GNPA level of commercial banks

showing some impact of the sensitization and regulatory changes. Similar trend is

observed in the case of Net NPA. The decline in the level of NNPA is sharper than the

GNPA. This is mainly because of the increasing level of provisions, as shown in the last

three rows of Table 3.4.

Table 3.4

Non Performing Assets of Total Banking Sector (Rs Crore, Real Values) 1998 1999 2000 2001 2002 2003 2004 2005

Gross NPA 34428 38275 38320 38828 41430 39012 35007 38558

Change 3848 45 508 2602 -2418 -4006 3551Percentage growth 11.18 0.12 1.33 6.70 -5.84 -10.27 10.14Percent to Gross Advance 14.39 14.71 12.79 11.42 10.42 8.86 7.19 5.27

percent to Gross Assets 6.36 6.18 5.49 4.91 4.62 4.04 3.27 2.57

Net NPA 16098 18264 18991 19774 20787 18548 13302 14181

Change 2165 727 783 1013 -2240 -5246 879Percentage growth 13.45 3.98 4.13 5.12 -10.77 -28.28 6.61Percent to Gross Advance 6.73 7.02 6.34 5.82 5.23 4.21 2.73 1.94

Percent to Gross Assets 2.97 2.95 2.72 2.50 2.32 1.92 1.24 0.95

Gross Advance

239188 260180 299697 340005 397502 440207 486686 731323

Change 20993 39516 40309 57496 42705 46479 244637Percentage growth 8.78 15.19 13.45 16.91 10.74 10.56 50.27

Gross Assets

541311 619827 697384 790257 896895 966570 1070292 1497479

Change 78516 77558 92873 106638 69675 103722 427187

258

Percentage growth 14.50 12.51 13.32 13.49 7.77 10.73 39.91

Gross-net 18329 20012 19329 19054 20643 20465 21705 24377

Change 1682 -683 -275 1589 -178 1241 2672Percentage growth 9.18 -3.41 -1.43 8.34 -0.86 6.06 12.31Source: Computed by author using RBI data.

At the bank group level, public sector banks have the highest non-performing assets with

an average GNPA (for the period 1998-2005) of around Rs 31,000 crore (about 11.54

percent to gross advances) and average NNPA of about Rs 16,000 crores (around 5.72

percent of net advance) (compare Table 3.5 and 3.6). While higher GNPA from the

public sector is expected, as their share in total lending is also much higher in the total

banking sector, GNPA to total advance ratios are also higher for them. In this context we

observe that the old private banks rank second in terms of percentage of NPAs in total

lending (around 9.96 percent of gross advance) and average NNPA of around 6.15

percent of net advance. This is followed by the new private sector banks with an average

GNPA to total advance ration of around 5.51 percent (see table 3.5 and 3.6) and average

NNPA to total net advance ratio of around 3.36 percent of net advance. Similarly, high

absolute level of total NPA is expected of public sector banks as their volume of lending

is also much higher. For the same reason foreign banks rank last both in terms of average

GNPA as well as NNPA. But when we compare the public sector banks against private

banks in terms of percentage of total lending (Table 3.6) we observe that public sector

banks are as good as or as bad as their private counterparts. In some years public sector

banks are indeed doing better that the private sector banks. But when compared with the

foreign banks they do not fare well. This may be partly because foreign banks are long

accustomed to the NPA norms in their parent country. Further, various credit related

welfare programs are carried out through public sector banks. They also have maximum

reach in the rural areas. Whether having maximum rural branches is indeed creating bad

loans is a matter that needs closer scrutiny and will be examined formally in Chapter 4.

One feature however is worth noting here. Growth rate of gross NPAs of the private

sector banks are higher than the public sector banks while growth of advances of public

259

sector banks are at par with the private banks. It thus appears that after reform public

sector banks are able to tackle the NPA problem more effectively than the private banks

(Table 3.6). However, one important observation from the table 3.6 is that, GNPA as

percent of gross advance as well as NNPA as percent of net advance has been declining

over time across all bank groups22.

Table 3.5

Per Cent Growth Rates of Gross and Net NPAs of Scheduled Commercial Banks (Real values)

1998 1999 2000 2001 2002 2003 2004 2005 Public Sector Banks Gross NPAs 30930 33705 33567 33304 33018 30708 27848 31300Growth Rate 8.97 -0.41 -0.78 -0.86 -7.00 -9.31 12.39Net NPAs 14385 15781 16494 17042 16346 14118 10191 11007Growth Rate 9.70 4.52 3.32 -4.09 -13.63 -27.82 8.01Old Private Banks Gross NPAs 1893 2466 2511 2647 2836 2583 2373 2782Growth Rate 30.30 1.79 5.45 7.13 -8.92 -8.13 17.22Net NPAs 1065 1520 1565 1688 1762 1556 1156 1229Growth Rate 42.72 2.93 7.89 4.36 -11.70 -25.67 6.33New Private Banks Gross NPAs 266 568 596 985 3982 4106 3222 3026Growth Rate 113.76 4.95 65.32 304.28 3.10 -21.52 -6.07Net NPAs 197 398.25 400.58 565.91 2141.64 2351.52 1468.10 1515.87Growth Rate 102.00 0.58 41.27 278.44 9.80 -37.57 3.25Foreign Banks Gross NPAs 1339 1536 1647 1892 1594 1615 1564 1450Growth Rate 14.76 7.21 14.88 -15.76 1.34 -3.18 -7.29Net NPAs 451 564 532 478 538 523 486 429Growth Rate 25.10 -5.82 -10.05 12.49 -2.79 -6.99 -11.87

Table 3.6

Percentage Growth Rates of Gross NPA and Gross Advance 1999 2000 2001 2002 2003 2004 2005 Public Sector Banks

22 For nominal figures see table A3.2 and A3.3 in the Appendix. For frequency distribution of NPAs under different classifications see Table A3.6 in the Appendix.

260

Growth of Gross Advance 13.98 17.05 15.82 15.39 13.10 14.83 25.66 Growth of Gross NPA 13.27 3.06 2.59 3.29 -4.22 -4.72 -8.17 Old Private Banks Growth of Gross Advance 12.69 22.12 13.03 10.61 15.93 13.04 21.30 Growth of Gross NPA 35.43 5.34 9.03 11.62 -6.20 -3.47 -4.23 New Private Banks Growth of Gross Advance 25.43 60.33 40.77 141.37 24.34 25.33 6.58 Growth of Gross NPA 122.19 8.61 70.93 321.21 6.18 -17.55 -23.26 Foreign Banks Growth of Gross Advance 0.45 20.46 22.27 10.52 6.33 17.20 24.43 Growth of Gross NPA 19.28 10.95 18.78 -12.23 4.37 1.72 -24.26 Source: Computed by author using RBI data. Going to the disaggregated level of NPAs it has been observed that amongst the bad

loans, maximum is substandard loan and proportion of loss assets is rather low (Table

3.7). At the total banking sector level, on an average (for the period 1998-2005) around

89 percent of the total loan assets fall under the standard asset category. When we

consider the rest, 3.3 percent under sub-standard assets, around 3 percent under doubtful

assets and around 1.3 under loss assets. And, the total NPA is around 10.6 percent of the

total loan asset. Thus, loss assets, even for the public sector banks are rather low.

Looking at the temporal behaviour of the share of various types of assets, on average, as

the share of standard asset is improving over time, naturally, the shares of sub-standard

assets, doubtful assets, loss assets and NPAs in the total loan asset are declining. This

essentially implies that the quality of the loan asset of Indian commercial banks is

improving. What is clear from Table 3.6 about low levels of NPAs of the foreign banks

is also seen from Table 3.7. From a disaggregated analysis at the bank-group level it

appears that the quality of the loan assets of foreign banks is better than the quality of the

loan asset of other bank groups. While the share of the standard asset in the total loan

asset of foreign banks is 94 percent, it is 88.5 and 91 percent in the case of public sector

and Indian private banks. On the other hand while the share of NPA in the total loan asset

of foreign banks is 5.8 percent, it is 11.5 and 8.7 percent in the case of public sector

banks and Indian private banks. However, proportion of loss assets of public sector banks

and foreign banks are quite comparable and they are comparatively lower for the private

banks.

261

Table 3.7 Bank group-wise Classification of Loan Assets of Scheduled Commercial Banks (Rs Crores)

1998 1999 2000 2001 2002 2003 2004 2005 Public Sector Banks Standard Assets 239318 273618 326783 387360 452862 523724 610435 824253 Per cent to total 84.0 84.1 86.0 87.6 88.9 90.6 92.2 94.6 Substandard Assets 14463 16033 16361 14745 15788 14909 16909 10838 Percent to Total 5.1 4.9 4.3 3.3 3.1 2.6 2.6 1.2 Doubtful assets 25819 29252 30535 33485 33658 32340 28756 29988 Per cent to total 9.1 9 8 7.6 6.6 5.6 4.3 3.4 Loss Assets 5371 6425 6398 5644 7061 6840 5876 5771 Per cent to total 1.9 2 1.7 1.5 1.4 1.2 0.9 0.7 Total NPAs 45653 51710 53294 54774 56507 54089 51541 46597 Per cent to total 16 15.9 14 12.4 11.1 9.4 7.8 5.4 Indian Private Banks Standard Assets 33567 38394 53317 65071 109216 131620 167076 216448 Per cent to total 91.3 89.2 91.5 91.5 90.3 90.8 94.2 96.1 Substandard Assets 1766 2657 2137 2585 4738 3703 3127 2213 Percent to Total 4.8 6.2 3.7 3.6 3.9 2.6 1.8 1 Doubtful assets 1077 1591 2355 3069 6539 8512 6391 5578 Per cent to total 2.9 3.7 4 4.3 5.4 5.9 3.6 2.5 Loss Assets 343 407 439 424 390 1118 825 900 Per cent to total 0.9 0.9 0.8 0.6 0.3 0.8 0.5 0.4 Total NPAs 3186 4655 4931 6078 11667 13333 10343 8691 Per cent to total 8.7 10.8 8.5 8.5 9.7 9.2 5.8 3.9 Foreign Banks Standard Assets 28996 28702 34817 42285 47838 50851 59619 72963 Per cent to total 93.6 92.4 93 93.1 94.5 94.6 95.2 97 Substandard Assets 1198 1238 1096 876 856 994 990 714 Percent to Total 3.9 4 2.9 1.9 1.7 1.8 1.6 0.9 Doubtful assets 250 507 798 1202 1004 944 1099 974 Per cent to total 0.8 1.6 2.1 2.6 2 1.8 1.8 1.3 Loss Assets 528 612 721 1033 920 954 924 569 Per cent to total 1.7 2 1.9 2.3 1.8 1.8 1.5 0.8 Total NPAs 1976 2357 2615 3111 2780 2892 3013 2257 Per cent to total 6.4 7.6 7 6.9 5.5 5.4 4.8 3

Source: Report on Trends and Progress of Banks in India, various issues

The above statistics shows the NPA problem at the aggregate level. In order to tackle the

problem a disaggregated analysis is necessary to examine what type of loans lead to more

NPAs. This necessitates an anlysis of sector-wise NPAs.

3. 6 Sector-wise NPA: NPA arising from the SSI sector

262

One of the important issue raised in the case of the NPAs of Indian commercial banks is

that the directed credit policy followed by the RBI under social banking motto of the

Government led to increase in the level of NPAs. To examine this we first look at the

share of NPAs from the priority sector vis-à-vis non priority sector loans. Table 3.8

reveals that the share of the NPA of non-priority sector is indeed higher than the share of

the NPA of priority sector and this trend is continuing over the years.

Table 3.8 Sector Wise Non Performing Assets of Indian Scheduled Commercial Banks

(Rs Crore, Real Values)

Year Item Agriculture Small scale OthersPriority Sector

Public Sector

Non priority Sector

Bank Group: State Bank and Associates 2001 Amount 1839 2317 1283 5439 739 6118 Per cent to total 15.0 18.8 10.4 44.2 6.0 49.72002 Amount 1849 2113 1278 5273 362 5908 Per cent to total 16.0 18.3 11.1 45.7 3.1 51.22003 Amount 1688 1740 1144 4572 299 4757 Per cent to total 17.3 17.8 11.7 46.7 3.1 48.62004 Amount 1351 1232 1272 3856 119 4216 Per cent to total 16.5 15.0 15.5 47.1 1.5 51.52005 Amount 1504 1175 1926 4641 111 5042 Per cent to total 15.4 12.0 19.7 47.4 1.1 51.5

Nationalised Banks 2001 Amount 2654 3982 2640 9276 303 10512 Per cent to total 14.8 22.2 14.7 51.7 1.7 58.62002 Amount 2724 4042 2659 9425 290 11779 Per cent to total 12.7 18.8 12.4 43.9 1.4 54.82003 Amount 2688 4029 2870 9586 897 11726 Per cent to total 12.0 17.9 12.8 42.7 4.0 52.22004 Amount 2561 3871 2922 9026 2379 9669 Per cent to total 13.8 20.8 15.7 48.6 12.8 52.02005 Amount 3294 3972 3569 10834 274 10120 Per cent to total 15.6 18.8 16.9 51.3 1.3 47.9

Private Banks 2001 Amount 194 599 308 1101 75 2676 Per cent to total 5.0 15.6 8.0 28.6 2.0 69.42002 Amount 257 868 364 1489 18 5314 Per cent to total 3.8 12.7 5.3 21.8 0.3 77.92003 Amount 305 716 364 1385 54 5135 Per cent to total 4.5 10.7 5.4 20.6 0.8 76.42004 Amount 248 681 408 1338 40 4204 Per cent to total 4.4 12.2 7.3 24.0 0.7 75.3

263

2005 Amount 307 638 497 1442 28 4331 Per cent to total 5.3 11.0 8.6 24.9 0.5 74.7

Source: Computed by author using RBI data

As can be seen from the table, the average share the NPA of non-priority sector in the

total NPA is around 50.5%, 53.4% and 74.7% for SB&A, NP and PB respectively,

whereas, the average share of NPA of priority sector in the total NPA is around 46.2%,

47.9% and 23.9% for SB&A, NB and PB respectively. One important observation is that

the share of priority sector NPA is less in the case of Private Banks compared to other

bank groups. In the case of sub-category of priority sector, the share of agriculture sector

NPA in the total NPA is only around 4.61 percent for Private Banks whereas it is around

16 percent for SB&A and 13 percent for NB.

While it has been often highlighted in the literature that NPA arising from the priority

sector is less than that of non priority sector related NPAs , a point often missed is that

priority sector constitute about 40% of total lending. Therefore, it is important to

examine NPA figures in proportion to the advances made on that particular sector.

Computation of sector-wise NPAs indeed reveals that NPA arising from SSI sector is

much higher that the other sectors. While NPAs from agriculture sector was about 12.4%

in 2002 (Table 3.9), it was as high as 21.16% for the SSI sector in the same year. This

percentage however declined to 6% for the agriculture sector and to 11% for the SSI

sector. Thus declining trend is prominent uniformly across all sectors (table 3.923).

Table 3.9

Sector wise NPA of Commercial Banks (Real Values, Rs Crore) 2002 2003 2004 2005

Public Sector Banks Agriculture NPA 4573.01 4375.66 3912.20 4796.06 % to credit to agriculture 12.40 10.93 8.57 6.64 SSI NPA 6155.43 5768.80 5102.91 5144.44 % to credit to SSI 21.16 19.30 16.20 11.48 Other Priority Sector NPA 3937.08 4013.53 4194.17 5493.14

23 Nominal figures are presented in table A3.4 in the Appendix.

264

% to credit to other priority sector 12.54 9.93 8.07 6.64 Total Priority Sector NPA 14698.18 14158.16 12881.84 15469.35 % to credit to total priority sector credit 14.69 12.48 9.75 7.62 Non-Priority Sector NPA 18339.32 18093.85 13891.00 15454.60 % to credit to non-priority sector credit 9.28 8.43 6.16 4.10

Private Banks Agriculture NPA 256.76 304.74 248.02 306.80 % to credit to agriculture 5.47 4.52 3.12 2.14 SSI NPA 868.39 716.39 680.87 637.55 % to credit to SSI 17.24 18.40 16.60 11.22 Other Priority Sector NPA 363.62 363.84 408.27 497.16 % to credit other priority sector 6.85 3.64 2.93 1.94 Total Priority Sector NPA 1488.76 1384.97 1338.41 1441.52 % to credit to total priority sector credit 9.90 6.65 5.06 3.12 Non-Priority Sector NPA 5332.73 5334.00 4244.56 4357.16 % to credit to non-priority sector credit 9.58 8.59 6.11 5.15

Source: Computed by author using RBI data.

However, looking at the growth rate of NPAs across sectors we observe a negative trend

for the SSI sector. On the other hand for agriculture and priority sector figures

concerning the year 2005 shows some increment (Table 3.1024).

Table 3.9

Per Cent Growth Rates of Sector Wise Gross Non Performing Assets (Real Value)

Year Agriculture Small scale Others Priority Sector

Public Sector

Non -priority Sector

State Bank and Associates 2002 0.52 -8.79 -0.34 -3.05 -50.98 -3.43 2003 -8.69 -17.66 -10.53 -13.30 -17.57 -19.48 2004 -19.96 -29.18 11.23 -15.67 -60.16 -11.37 2005 11.32 -4.68 51.42 20.36 -6.71 19.59

Nationalised Banks 2002 2.63 1.53 0.70 1.61 -4.34 12.04 2003 -1.34 -0.33 7.94 1.71 208.88 -0.45 2004 -4.71 -3.93 1.82 -5.84 165.37 -17.54 2005 28.60 2.61 22.13 20.02 -88.50 4.66

Private Banks 24 Nominal figures are presented in Table A3.5 in the Appendix.

265

2002 32.18 44.93 18.17 35.20 -75.74 98.57 2003 18.69 -17.50 0.06 -6.97 194.33 -3.38 2004 -18.61 -4.96 12.21 -3.36 -24.89 -18.12 2005 23.74 -6.33 21.81 7.74 -30.51 3.01

Source: Computed by author using RBI data. Thus though SSI sector currently has a higher NPA to total advance ratio there is

an improvement in recovery rates and NPA from this sector shows a declining trend

even in real terms.

3.7 Conclusion The problem of NPA has received considerable attention after financial sector

liberalization in India. Accounting norms have been modified substantially and

mechanisms are put in place for reduction of bad loans. Our survey of banks however

shows that (see chapter 7) mainly due to the awareness of the problem of bad loans at the

bank level, NPAs have indeed come down considerably. It remains true that NPA arising

from the priority sector lending is still higher than the non priority sector. Within priority

sector SSI’s performance is worse than the others. Given this observation a need has been

felt to study this problem in some detail. In the next section therefore, we look at the

determinants of NPA for the Indian banking sector in general and for the SSI sector in

particular.

Appendix to Chapter 3

Table A3.1

Non Performing Assets of Total Banking Sector: Nominal Values (Rs Crore) 1998 1999 2000 2001 2002 2003 2004 2005 Gross NPA 50815 58722 60841 63741 70861 68717 64787 58299 Change 7907 2119 2900 7120 -2144 -3930 -6488 Percentage growth 15.56 3.61 4.77 11.17 -3.03 -5.72 -10.01 Gross Advance 353039 399167 475827 558157 679875 775386 900706 1105760 Change 46128 76660 82330 121718 95511 125320 205054 Percentage growth 13.07 19.20 17.30 21.81 14.05 16.16 22.77

266

Gross Assets 798970 950934 1107234 1297296 1534023 1702529 1980782 2264188 Change 151964 156300 190062 236727 168506 278254 283405 Percentage growth 19.02 16.44 17.17 18.25 10.98 16.34 14.31 Net NPAs 23761 28020 30152 32462 35554 32670 24617 21441 Change 4259 2132 2310 3092 -2884 -8053 -3176 Percentage growth 17.92 7.61 7.66 9.52 -8.11 -24.65 -12.90 Gross-net 27054 30702 30689 31279 35307 36047 40170 36858 Change 3648 -13 590 4028 740 4123 -3312 Percentage growth 13.48 -0.04 1.92 12.88 2.10 11.44 -8.24 Source: RBI

Table A3.2

Per Cent Growth Rates of Gross and Net NPAs of Scheduled Commercial Banks (Nominal Terms) 1998 1999 2000 2001 2002 2003 2004 2005

Public Sector Banks Gross NPAs 45653 51710 53294 54672 56473 54090 51538 47325 Growth Rate 13.27 3.06 2.59 3.29 -4.22 -4.72 -8.17 Net NPAs 21232 24211 26188 27977 27958 24867 18860 16642 Growth Rate 14.03 8.17 6.83 -0.07 -11.06 -24.16 -11.76 Old Private Banks Gross NPAs 2794 3784 3986 4346 4851 4550 4392 4206 Growth Rate 35.43 5.34 9.03 11.62 -6.20 -3.47 -4.23 Net NPAs 1572 2332 2484 2771 3013 2740 2140 1859 Growth Rate 48.35 6.52 11.55 8.73 -9.06 -21.90 -13.13 New Private Banks Gross NPAs 392 871 946 1617 6811 7232 5963 4576 Growth Rate 122.19 8.61 70.93 321.21 6.18 -17.55 -23.26 Net NPAs 291 611 636 929 3663 4142 2717 2292 Growth Rate 109.97 4.09 46.07 294.29 13.08 -34.40 -15.64 Foreign Banks Gross NPAs 1976 2357 2615 3106 2726 2845 2894 2192 Growth Rate 16.16 9.87 15.81 -13.94 4.18 1.69 -32.03 Net NPAs 666 866 844 785 920 921 900 648 Growth Rate 23.09 -2.61 -7.52 14.67 0.11 -2.33 -38.89

Table A3.3 Gross and Net NPAs of Scheduled Commercial Banks (Nominal Values, Rs

Crore) 1998 1999 2000 2001 2002 2003 2004 2005

Public Sector Banks Gross NPAs 45653 51710 53294 54672 56473 54090 51538 47325

Percent to Gross Advance 16 15.9 14 12.4 11.1 9.4 7.8 5.7 Percent to Gross Asset 7 6.7 6 5.3 4.9 4.2 3.5 2.8

Net NPAs 21232 24211 26188 27977 27958 24867 18860 16642 Percent to Net Advance 8.2 8.1 7.4 6.7 5.8 4.5 3 2.1 Percent to Net Assets 3.3 3.1 2.9 2.7 2.4 1.9 1.3 1.0

Old Private Banks Gross NPAs 2794 3784 3986 4346 4851 4550 4392 4206

267

Percent to Gross Advance 10.9 13.1 11.3 10.9 11.0 8.9 7.6 6.0 Percent to Gross Asset 5.1 5.8 5.1 5.1 5.2 4.3 3.6 3.2

Net NPAs 1572 2332 2484 2771 3013 2740 2140 1859 Percent to Net Advance 6.5 9.0 7.3 7.3 7.1 5.5 3.8 2.7 Percent to Net Assets 2.9 3.6 3.2 3.3 3.2 2.6 1.8 1.4

New Private Banks Gross NPAs 392 871 946 1617 6811 7232 5963 4576

Percent to Gross Advance 3.5 6.2 4.2 5.1 8.9 7.6 5.0 3.6 Percent to Gross Asset 1.5 2.3 1.6 2.1 3.9 3.8 2.4 1.6

Net NPAs 291 611 636 929 3663 4142 2717 2292 Percent to Net Advance 2.6 4.5 2.9 3.1 4.9 4.6 2.4 1.9 Percent to Net Assets 1.1 1.6 1.1 1.2 2.1 2.2 1.1 0.8

Foreign Banks Gross NPAs 1976 2357 2615 3106 2726 2845 2894 2192

Percent to Gross Advance 6.4 7.6 7.0 6.8 5.4 5.3 4.6 2.8 Percent to Gross Asset 3.0 3.1 3.2 3.0 2.4 2.4 2.1 1.4

Net NPAs 666 866 844 785 920 921 900 648 Percent to Net Advance 2.2 2.9 2.4 1.8 1.9 1.8 1.5 0.9 Percent to Net Assets 1.0 1.1 1.0 0.8 0.9 0.8 0.7 0.4

Source: Report on Trends and Progress of Banks in India, RBI, various issues

Table A3.4 Sector Wise Non Performing Assets of Indian Scheduled Commercial Banks (Nominal Values)

Year Item Agriculture Small scale Others

Priority Sector

Public Sector

Non priority Sector Total

Bank Group: State Bank and Associates 2001 Amount 3019.44 3803.19 2105.72 8928.35 1212.75 10043.57 20190.70

Per cent to total 14.95 18.84 10.43 44.22 6.01 49.74 2002 Amount 3162.26 3614.21 2186.40 9018.74 619.41 10105.41 19743.57

Per cent to total 16.02 18.31 11.07 45.68 3.14 51.18 2003 Amount 2973.52 3064.80 2014.52 8052.84 525.82 8379.44 17228.10

Per cent to total 17.26 17.79 11.69 46.74 3.05 48.64 2004 Amount 2500.59 2280.54 2354.42 7135.55 220.09 7802.97 15158.61

Per cent to total 16.50 15.04 15.53 47.07 1.45 51.48 2005 Amount 2274.19 1776.01 2912.69 7016.89 167.75 7623.56 14808.32

Per cent to total 15.36 11.99 19.67 47.38 1.13 51.48

268

Nationalised Banks 2001 Amount 4357.21 6536.22 4334.46 15227.89 498.09 17257.44 29464.42

Per cent to total 14.79 22.18 14.71 51.68 1.69 58.57 2002 Amount 4659.29 6913.86 4547.46 16120.60 496.44 20145.74 36762.81

Per cent to total 12.67 18.81 12.37 43.85 1.35 54.80 2003 Amount 4733.83 7096.43 5054.96 16885.52 1579.14 20654.24 39580.99

Per cent to total 11.96 17.93 12.77 42.66 3.99 52.18 2004 Amount 4739.70 7163.38 5407.70 16704.78 4402.97 17894.77 34389.70

Per cent to total 13.78 20.83 15.72 48.57 12.80 52.04 2005 Amount 4979.85 6004.95 5395.68 16380.49 413.54 15300.80 31964.13

Per cent to total 15.58 18.79 16.88 51.25 1.29 47.87 Private Banks

2001 Amount 318.89 983.60 505.15 1807.64 123.37 4393.46 6326.95 Per cent to total 5.04 15.55 7.98 28.61 1.95 69.44

2002 Amount 439.16 1485.27 621.92 2546.34 31.18 9089.47 11667.29 Per cent to total 3.76 12.73 5.33 21.82 0.27 77.91

2003 Amount 536.78 1261.86 640.87 2439.51 94.51 9044.07 11834.88 Per cent to total 4.54 10.66 5.42 20.61 0.80 76.42

2004 Amount 459.01 1260.08 755.58 2476.98 74.58 7780.76 10332.36 Per cent to total 4.44 12.20 7.31 23.97 0.72 75.30

2005 Amount 464.04 964.30 751.96 2180.30 42.34 6547.87 8770.50 Per cent to total 5.29 10.99 8.57 24.86 0.48 74.66

Source: Report on Trends and Progress of Banks in India, various issues

Table A3.5 Per Cent Growth Rates of Sector Wise Gross Non Performing Assets

Year Agriculture

Small Scale

Industries Others Priority Sector

Public Sector

Non priority Sector Total

State Bank and Associates 2002 4.73 -4.97 3.83 1.01 -48.93 0.62 -2.21 2003 -5.97 -15.20 -7.86 -10.71 -15.11 -17.08 -12.74 2004 -15.90 -25.59 16.87 -11.39 -58.14 -6.88 -12.01 2005 -9.05 -22.12 23.71 -1.66 -23.78 -2.30 -2.31

Other Nationalised Banks 2002 6.93 5.78 4.91 5.86 -0.33 16.74 24.77 2003 1.60 2.64 11.16 4.74 218.09 2.52 7.67 2004 0.12 0.94 6.98 -1.07 178.82 -13.36 -13.12 2005 5.07 -16.17 -0.22 -1.94 -90.61 -14.50 -7.05

Private Banks 2002 1306.02 51.00 -93.55 -89.31 6426.62 106.89 84.41 2003 -88.03 -15.04 3.05 -4.20 -98.83 -0.50 1.44 2004 -14.49 -0.14 17.90 1.54 -21.09 -13.97 -12.70 2005 1.10 -23.47 -0.48 -11.98 -43.23 -15.85 -15.12

269

Table A3.6 : Frequency distribution of NPAs under various classifications

Bank Group Wise Frequency Distribution of NPA for the period 1997-2005 (Average)

Bank Group/Percent 0-5 5-10 10-15 15-20 20-25 >25 Gross NPA to Total Asset Ratio

SBI&A 41 31 0 0 0 0 NB 94 67 7 3 0 0 PB 176 80 11 0 0 0 FB 191 35 20 10 5 14

Net NPA to Total Asset Ratio SBI&A 71 1 0 0 0 0 NB 163 8 0 0 0 0 PB 244 23 0 0 0 0 FB 237 27 9 2 0 0

Gross NPA to Gross Advance Ratio SBI&A 12 20 27 13 0 0 NB 20 54 46 29 11 11 PB 64 88 71 29 4 11 FB 127 45 24 16 14 49

Net NPA to Net Advance Ratio SBI&A 29 34 9 0 0 0 NB 71 71 22 4 2 1 PB 127 107 22 4 6 1 FB 182 33 23 9 8 20

Gross NPA to Gross Advance Ratio

Year/ Percent 0-5 5-10 10-15 15-20 20-25 >25 1997 33 17 15 16 4 7 1998 26 19 21 15 5 7 1999 17 21 21 18 8 9 2000 18 23 28 12 3 10 2001 22 22 26 5 3 8 2002 21 24 21 9 2 8 2003 19 28 21 7 0 10 2004 26 30 12 3 1 7 2005 41 23 3 2 3 5

Net NPA to Net Advance Ratio

Year / Percent 0-5 5-10 10-15 15-20 20-25 >25 1997 44 34 10 2 2 0 1998 39 32 17 2 2 1 1999 35 32 17 3 5 2 2000 38 37 10 4 2 3 2001 40 33 7 2 1 3 2002 38 31 7 1 2 6 2003 50 25 5 1 1 3

270

2004 60 12 1 2 1 3 2005 65 9 2 0 0 1

Frequency Distribution of NPA for the period 1997-2005 0-5 5-10 10-15 15-20 20-25 >25

Gross NPA to Total Asset Ratio Frequency 502.00 213.00 38.00 13.00 5.00 14.00 Percent 63.95 27.13 4.84 1.66 0.64 1.78

Net NPA to Total Asset Ratio Frequency 715.00 59.00 9.00 2.00 0.00 0.00 Percent 91.08 7.52 1.15 0.25 0.00 0.00

Gross NPA to Gross Advance Ratio Frequency 223.00 207.00 168.00 87.00 29.00 71.00 Percent 28.41 26.37 21.40 11.08 3.69 9.04

Net NPA to Net Advance Ratio Frequency 409.00 245.00 76.00 17.00 16.00 22.00 Percent 52.10 31.21 9.68 2.17 2.04 2.80

Table A3.7

Prudential Norms for Asset Classification Adopted by India and Some Other Countries Country Categories Loans Classification System Provisioning requirements Indonesia Current Installment Credit with no arrears, other

credit in arrears less than 90 days, overdrafts less than 15 days.

0.5 per cent

Sub-standard Generally, loans with payments in arrears between three and six months.

10 per cent

Doubtful Non-performing loans that can be rescued and the value of collateral exceeds 75 per cent of the loan, or loans that cannot be rescued, but are fully collateralised.

50 per cent

Loss Doubtful loans that have not been serviced for 21 months; credit in process of bankruptcy/liquidation.

100 per cent

Loans must be written off 21 months after litigation, indicates the loans will not have to be repaid.

Korea Current Borrower's credit conditions (including collateral) are good and collectibility of interest and principal are certain.

0.5 per cent

Special mention Payments are past due for between three months and six months, but collection is

1 per cent

271

certain. Sub-standard Loans covered by collateral but borrower's

credit conditions are deteriorating and payments are more than six months past due.

20 per cent

Doubtful Unsecured portion of the loans that are more than six months past due and losses are expected.

75 per cent

Estimated loss Unrecoverable amounts due net of collateral.

100 per cent

Loans must be written off within six days of being declared unrecoverable; Write-offs in excess of W300 million require Bank of Korea approval.

Malaysia For loans less than RM 1 million Standard More than a normal risk of loss due to

adverse factors; past due for between 6 and 12 months.

0 per cent

Doubtful Collection in full is improbable and there is high risk of default; past due for between 12 and 24 months

50 per cent of net (of collateral) outstanding value

Bad Uncollectible; past due for more than 24 months.

100 per cent of net outstanding value

Loans must be written off when bankruptcy hearings have finished and/or partial or full repayment is unlikely.

A general provision of at least 1 per cent of total loans net of interest in suspense and specific provisions is also required.

Philippines Unclassified Borrower has the apparent ability to satisfy obligations in full; no loss in collection is anticipated.

0 per cent of net (of collateral) exposure.

Special mention Potentially weak due, for example, to inadequate collateral, credit information, or documentation.

0 per cent

Sub-standard Loans that involve a substantial degree of risk of future loss.

25 per cent

Doubtful Loans on which collection or liquidation in full is highly improbable, substantial losses are probable.

50 per cent

Loss Uncollectible or worthless. 100 per cent Interest is not accrued on past-due

loans, which are loans or other credit not paid at the prescribed maturity date or, in the case of instalment credit, in arrears by more than a prescribed amount depending upon the frequency of instalments.

Argentina Consumer Loans Commercial Loans Liquid G'tee

Preferred G'tee

Without G'tee

Normal Less than 31 days overdue

No doubt exists. 1 per cent 1 per cent 1 per cent

Potential risk 31- 89 days overdue Performing, but 1 per cent 3 per cent 5 per cent

272

sensitive to changes or more than 30 days overdue.

Problem 90 - 179 days overdue

Problems meeting obligations; or80-179 days overdue

1 per cent 12 per cent 25 per cent

High risk 180-365 days over- due or subject to judicial proceedings for default

Highly unlikely to meet obligations; or more than 180 days overdue.

1 per cent 25 per cent 50 per cent

Irrecoverable More than 365 days overdue

Obligations cannot be met; more than365 days overdue

1 per cent 50 per cent 100 per cent

Irrecoverable for technical decision

Bankruptcy/ liquidation/ insolvency

Bankruptcy/ liquidation/ insolvency

100 per cent 100 per cent 100 per cent

Chile Consumer Mortgage Commercial Minimum initial provision (for NPAs) (allowance period is 90 days in all the3 types of advances)

A - Current B B C D

Current 1 - 30 days 3 - 59 days 60-119 days >120 days

Current 1- 179 days > 179 days N.A N.A.

Probability of default = 0% Probability of default > 0%,< 5% Probability of default = > 5%, < 40% Probability of default = .> 40%, < 80% Probability of default = >80%, < 100%

on consumer loan and mortgage loan(NPAs) is required to be made @ 60% and 1% respectively whereas the provisioning requirement on commercial loan is subjective

Peru A - Normal Current Current Current with no doubts Minimum initial provision @ 30%, (allowance B - Potential 10-29 days 32-89 daysDemonstrated 1% and 15% is required to be made onperiod is 30, problems deficiencies consumer loan, mortgage loan and 30 and 15 commercial loan (NPAs) respectively.days respectively

C - Sub- standard

30-59 days 90-119 days

60-119 days

for the three types of

D - Doubtful 60-120 days 120 - 365 days

120 - 364 days

advances) E - Loss > 120 days > 365 days> 365 days India Sub-standard Loans that have been non-performing for up to

two years, term loans on which the principal has not been reduced for more than one year, and all rescheduled debts.

10 per cent

Doubtful Loans that have been non-performing for two to three years and term loans on which the principal has not been reduced for more than two years.

100 per cent of unsecured assets; for secured assets; 20 per cent if doubtful for less than one year; 30 per cent if doubtful for one to three years, 50 per cent if doubtful for more than three years.

Loss All other assets deemed irrecoverable, where the loss has been identified by internal or external auditors or by the Reserve Bank of India inspectors, but where the amount has not been written off.

100 per cent

Source: Global NPL Report

273

CAHPTER 4

Determinants of Non Performing Assets

4.1 Introduction

Given that the NPA has strong implications on the health of the commercial banks

and also the economy, it is essential to contain the NPA levels of the commercial banks.

This calls for identification of the factors that can cause an asset to become NPA. A study

conducted by the RBI, based on the information related to 33banks shows that the NPA

level in banks are generally related to the performance level of the industrial units to

which the banks have lent loan (Muniappan, 2002). The factors could be management

inefficiency of borrowal units, obsolescence, lack of demand, non- availability of inputs,

environmental factors etc. In a few cases default occurs due to the internal factors of

banks such as weak appraisal or follow-up of loans. While a micro level study will reveal

some of these causes , before going to such analysis it is essential to get a picture of the

determinants of NPA at the macro level. In this regard an attempt is made in the current

chapter to identify the factors influencing the NPA at the macro level.

Given this back ground this present chapter is designed as follows; in the next

section a brief review of studies related to NPAs at the international level, as well as in

India is presented. Section 3 spells out the methodology used in the study to analyse the

determinants of the NPAs. While section 4 will present various sources of data used in

the study, section 5 contains the empirical results and the discussion. Finally, concluding

remarks are presented in section six.

4.2. Review of Literature

Over the last several years the issue of Non-Performing Assets has received

considerable attention from the policy makers and researchers all over the world. High

level of NPA is affecting not only the developing nations in Asia but even some of the

advanced economies like Japan and Germany (see also Chapter 3) are faced with this

problem. The question of NPA was given more emphasis after the BASEL committee

stressed that non-performing assets, from the financial systems across the globe should be

reduced and also prescribed various prudential norms. A number of studies which look at

274

the bank-failures concerning different countries note that the major reason for bank

failure is their Non-Performing Assets, since just before the failure they had accumulated

large amount of NPLs (Demirguc-Kunt 1989, Whalen 1991, and Barr and Siems 1994). It

has been observed that the weak banks, in developed and developing countries alike, do

not report their true no-performing loans (NPL) level, due to which the issues become

complicated when they are hit by a crisis (Tanaka and Hoggarth, 2006).

The problem of Non-Performing Loans is predominant in Asian countries, more

specifically in the East Asian countries (see Chapter 3). In fact, the higher levels of NPL

were one of the important reasons for the East Asian Financial Crisis of 1997. After the

so-called economic bubble burst in Japan in 1990, it went through a decade of economic

depression, which, to some extent, is due of the failure of the Japanese government to

control its NPLs (Asami, 2000). As discussed in Chapter 3, Japan has the highest NPL

levels in the world. In 2004 its NPL was around US$ 330 billion, which accounted to

around 25 percent of the world NPL (Global NPL Report 2004). Japanese banks have lost

72 trillions yen due to bad loans since March 1992 until March 2001. The bad loan per

total loan ratio has risen from 2 percent in 1992 to 6 percent in 1999 (Hanah, 2005).

However, it has been observed that the real problem of Japan’s NPL lies in understanding

the problem of the corporate to which banks are giving loans. Instead of banks continuing

to roll over loans to loss making corporate they should divert the funds to profit making

corporate (Hubbard, 2002). The only East Asian country, which not only avoided the

recent financial crises, but also continued to exhibit strong growth, is China (Huang and

Bonin, 2001). However, like many East Asian countries, China also holds high level of

Non Performing Loans, which might prove dangerous. In 2003 China’s NPL was around

US$ 307 billion, which accounted for around 23 percent of the global NPA (Global NPL

Report 2004). Measured as a percent of the total outstanding loans, the NPL of China

ranged from 18 percent to 40 percent Sprayregen et al (2004). Apart from the problem of

high NPLs, Chinese banking industry suffers from low capital base also. The public

sector dominance of the Chinese banking industry, along with the policy of policy

lending led to the Public Sector Enterprise (Chinese banks are expected to lend around

30-40 percent of their total loans to PSE), increased NPLs of the Chinese banking

industry (Huang and Bonin, 2001).

275

Recognizing the extent of damage NPLs can cause on economic performance,

considerable efforts have gone in understanding the factors that effect NPLs in any

economy and mechanisms to tackle this problem.

Using financial data collected from the public listing companies in China ’s stock

market, Lu, Thangavelu and Hu (2005) empirically examine the relationship between

banks’ lending behaviour and non-performing loans. This study takes the research

strategy to infer banks’ lending behaviour from firms’ debt financing. The lending

behaviour of banks is examined in a borrowing ratio model that captures the relationship

between firms’ borrowing ratios and their default risk, collateral, state-ownership, firm

size, and industrial policy. Their results show that state-owned enterprises (SOEs) got

more loans than other firms, other things being equal, and SOEs with high default risks

were able to borrow more than the low-risk SOEs and non-SOEs. This suggests that

Chinese banks had a systemic lending bias in favour of SOEs, particularly those with

high default risks, during the period under investigation.

In the Indian context, Rajan and Dhal (2003) empirically examine how banks’

non-performing assets are influenced by three major sets of economic and financial

factors, i.e., terms of credit, bank size induced risk preference and macroeconomic

shocks. They estimate a panel regression model, where they take into consideration the

effect of the differential social and geo-political environment confronting banks’

operations. Their results show that, when the bank size is measured in terms of assets, the

bank size has negative impact on NPAs, while the measure of bank size in terms of

capital, gives somewhat opposite result. Further, measure of credit orientation has

significant negative influence on NPAs, implying that borrowers attach considerable

importance to relatively more credit (customer) oriented banks. Also increased economic

activity leads to lower financial distress of borrowers and this lowers NPAs for banks.

The exposure to priority sector has positive impact on the NPA level. Using data on

Indian manufacturing sector for 1993-94, Ghosh (2005) has examined the association

between corporate leverage and banks’ non-performing loan. In this study asset quality of

banks is modeled as a function of corporate leverage and a set of control variables, using

a simultaneous equation framework. The results show that the capital adequacy of banks

276

have a significant and negative effect on asset quality. The findings further indicate that

the lagged leverage is an important determinant of bad loans of banks. In terms of

magnitude, a 10-percentage point rise in corporate leverage is, on an average, associated

with 1.3 percentage point rise in sticky loans, with one period lag.

Rajaraman and Vasishtha (2002) perform a panel regression on the definitionally

uniform data for a five-year period ending 1999-2000, on non-performing loans of

commercial banks. The study notes that a number of exogenous factors contribute to the

general level of bank NPA in India, such as the legal and procedural obstacles the banks

face in the process of liquidation of loss-making enterprises. According to the results of

this study, there is no significant relation between the capital adequacy ratio and NPA;

whereas, operating profit has a significant negative relation with NPA.

Along with identifying the factors causing high NPA levels, an equally important

issue, is how to tackle the problem of NPA. Many measures have been taken in the

direction of reducing NPL. In the case of Japan, it has been argued that a financial safety

net could minimize the spillover effects of failures of banks and other financial

institutions on the financial system as a whole. Bank-safety-net policies could include

lending to banks (a subset of lender of last resort), explicit and implicit insurance of some

or all bank deposits, capital adequacy requirements, bank supervision, closure and

recapitalization rules (see Diamond and Dybvig, 1983; and Allen and Gale, 2000). There

are also arguments that this safety net could lead to moral hazard problem. However, in

the case of Japan it has been observed that the safety net with the government financial

assistance taken from DIC to Japanese banks is not the reason leading to moral hazard of

increasing risk-taking incentive or bad loan, rather it helps to reduce the bad loan,

especially in case of banks with high requirement capital ratio (Hanah, 2005). Similarly,

in China, the government established four Asset Management Companies (AMCs) in

order to deal with the NPLs of each of the four large public sector companies. Drawing

on the experiences of Resolution Trust Corporation in the United States and bank

restructuring in the Central European transition economies, Huang and Bonin (2001)

argue that the original AMC design will not be successful in resolving the existing non-

277

performing loans (NPLs) nor will it prevent the creation of new bad loans due to various

drawbacks in the policy.

4.3. The Model and Estimation Procedure (Methodology)

As discussed in Chapter 1 there are two main concepts of NPA used by the

banking sector, viz., gross and net NPA. Net NPA is the total NPA minus the provisions

made by the bank under new accounting norms. Thus we observe that it is the gross NPA

that gives an indication of total bad loans of a bank. Naturally if total advances are

higher, total NPA level will also be so. Keeping these facts in mind we attempt here to

understand the determinants the share of gross NPA in total advances. We consider a

panel data set of 94 banks over the years 1997- 2005, as only from 1997 bank-wise NPA

data are made available.

Thus, in order to analyse the relation between various bank specific and economy specific variables with the level of NPA of commercial banks, an econometric model is formulated, which is estimated using panel data analysis technique. The model is given as follows:

itiit

ittt

ttitit

IMFBPBNBTBBRBBCRARLRGDP

GDPGDPTAGAGNPA

υµαααααααα

αααα

++++++++++

+++=

11/3

21]/[

10987

654

3210

(4.1)

Where,

GNPA/GA = Ratio of Gross Non Performing Asset to Gross Advance

TA = Total Assets25

GDP1 = Gross Domestic Product of Agriculture Sector

GDP2 = Gross Domestic Product of Industrial Sector

GDP3 = Gross Domestic Product of Service Sector

LR = Lending Rate (SBI Average Advance Rate)

RBB/TBB = Ratio of Rural Bank Branches to Total Bank Branches

NB = dummy for private bank group

PB = dummy for foreign bank group

25 Major aggregates of Total Asset are: Cash and Balance with RBI, Balance with Banks and money at call and short notice, Total Advance, Total Investment and Fixed Assets.

278

FB = other nationalized bank group

IM= interest rate margin (between lending rate and deposit rate)

µi = unobserved bank specific effects which are assumed to be random (like, bank

specific entrepreneurial or managerial skills)

vi = stochastic term which are assumed to be identically and independently

distributed, IID(0, σ2).

i: index for bank, t: time index

Before proceeding further it is essential to discuss the rationale for including each

variable in the model and also the expected signs. Total asset is included in the model as

a proxy for the bank size. The relation between the bank size and NPA share is an

empirical question that needs to be explored. If some kind of economies of scale exists in

recovery mechanism then a bigger bank may have lower NPA share. One of the

important variables that may have strong influence on the NPA level of commercial

banks is the general economic condition. While GDP is usually considered to reflect

general economic condition, in the present study, instead of taking total GDP, the sector

wise GDPs of three important sectors viz., agriculture, industry and services sectors are

included as explanatory variables. The reason for including the sector wise GDP is that

the growth in various sectors is not uniform. While service sector has registered the

growth rate of around 8.5 percent between 1997 and 2005, it is 5.26 and 1.99 for industry

and agricultural sector. In some of the years the growth rate of agricultural sector was

even negative. Thus it is essential to understand the relationship of each sector with the

NPA level of commercial banks. Further, it is argued that the branch expansionary policy

adopted by the government and the resulting expansion of bank branches might have led

to an increase in NPA. In order to see whether the number of Rural Bank Branches

(RBB) has any effect on the level of NPA, the ratio of RBB to Total Bank Branches is

included in the model. One of the reasons for the loan default, as argued in the theory, is

the higher cost of bank loan (Stiglitz et al, 1981). It has been argued that (Stiglitz et al,

1981) if the cost of credit is higher it will attract only the risky borrowers, which would

lead to increase in the NPA level. Thus the Advance rate of SBI is included as a proxy for

the lending rate. Interest rate margin is used to capture the extent of competition. Lower

279

margin may indicate higher level of competition. To examine bank group- wise

differences bank group specific dummies are incorporated (taken State bank of India

group as the base).

The model is estimated using the panel data estimation method. One of the

important decisions in a panel data analysis is whether the model should be estimated as a

fixed effect model or a random effect model. Since the number of firms is more than the

number of time periods, the random effect estimation technique is considered to be

appropriate to estimate the model (as it will have more degrees of freedom) compared to

the fixed effects model. However, if the assumption that the individual invariant effects

)( itu are not correlated with the regressors i.e., 0)/( =itit XuE is not valid, the GLS

estimator of the random effect model becomes biased and inconsistent. In this regard

Hausman (1978) has developed a test which suggests whether a fixed effect model is

more appropriate than a random effect model. In the present study, the Hausman

specification test recommends random effects model (see Section 4.5).

When the model is estimated as a random effect model, it is essential to check for

the presence of random effects. If this is negated, the model reduces to the Classical

Linear Regression (CLR) model, which can be estimated using Ordinary Least Square

(OLS) technique. Using the Breusch-agan test one can check the presence of random

effects, which in our case gives affirmative result (see Section 4.5).

4.4. Data Description For estimating the model given in (4.1), data are collected for 94 banks the period 1997-2005. However as many private and foreign banks are established after 1997 and few are closed during the study period, data are not available consistently for all banks for all years. Thus data used is an unbalanced panel of 94 (27 public sector banks, 33 Indian private banks and 34 foreign banks) banks for 9 years (total observation used are 746).

Data on Gross Non Performing Assets as percent of Gross Advance for each bank are collected from Report on Trends and Progress of Banking in India, published by the Reserve Bank of India. Return on Assets is calculated from the data on net profit and total assets collected from Annual Accounts of Scheduled Commercial Banks published by the Reserve Bank of India. The ratio of Rural Bank Braches to Total Bank Branches is calculated from the data on total number of rural bank branches and the total bank branches collected from the Performance Highlights of Banks published by the Indian Banks’ Association. Data on the sector wise GDP and variables like lending rate are collected from Handbook of Statistics on the Indian Economy published by the Reserve Bank of India. The GDP measured at constant price (1993-94 base) considered for the analysis and all values are converted to real wherever applicable. A summary of the variables included in the analysis is presented in table 4.1.

Table 4.1 Summary of Data Included in the Model (mean of values, at constant 1993-94 prices)

280

Variables 1997 1999 2001 2003 2005 GNPA/TA (%) 5.69 6.54 5.97 5.76 4.16 TA (Rs Crore) 8367.68 10647.93 15006.05 19404.87 28609.90 RRB/TBB (%) 22.47 19.52 19.59 16.81 18.28 GRGDP1 (%) -2.43 0.31 6.28 9.60 3.93 GRGDP2 (%) 3.05 4.05 3.51 6.52 5.90 GRGDP3 (%) 9.87 9.86 6.49 8.87 11.30 LR (%) 14.00 12.00 11.50 10.25 10.25

GRGDP1 = Growth Rate of Gross Domestic Product of Agriculture Sector GRGDP2 = Growth Rate of Gross Domestic Product of Industrial Sector GRGDP3 = Growth Rate of Gross Domestic Product of Service Sector *: Return to asset ratio (measure of profitability)

4.5. Empirical results

As mentioned in the previous section, before estimating the model it is important

to decide whether the model should be estimated as a fixed effect model or a random

effect model. In this regard the Hausman specification test is conducted and the results

are presented in table 4.2. The Hausman specification test statistic fails to reject the null

hypothesis that individual invariant effects )( itu are not correlated with the regressors

i.e., 0)/( =itit XuE . This suggests that the model should be estimated using random

effect estimation technique.

Table 4.2 Hausman Specification Test Results

Chi2 Probability4.33 0.7413

After deciding that the model should be estimated using random effect model, it is

essential to check whether random effects indeed present in the data. If the random

effects are not present in the data then the model can be estimated using Ordinary Least

Squares (OLS) method. As mentioned above, presence of random effect can be checked

by using Breusch Pagan test. The test results are presented in the table 4.3.

Table 4.3 Breusch and Pagan Lagrangian Multiplier Test for Random Effects

Chi2 Probability1034.1 0.0000

281

The results presented in table 4.3 shows that the null hypothesis that the random

effects are not present in the data is rejected, validating the use of random effect model

instead of the OLS. Thus after deciding that the model should be estimated as a random

effect model and confirming the presence of random effects, the model is estimated using

Generalised Least Square (GLS) method. The results of the GLS estimation are presented

in table 4.4.

Table 4.426 Determinants of Gross NPA/Gross Advance

Random-effects GLS Regression Dependent Variable: GNPA/GA Coefficient Std. Err. Z P>|z| Constant 9.42181 15.23245 0.62 0.536 TA -0.00066 0.000091 -3.45 0.0000 GDP1 0.0000487 0.000043 1.13 0.257 GDP2 -0.0000171 0.0000912 -0.19 0.851 GDP3 -0.000009 0.0000273 -0.34 0.735 IM 0.422903 0.319696 1.32 0.185 LR -0.812509 0.4937 -1.65 0.11 RBB/TBB 29.65196 7.830 3.790 0.0000 NB -0.71188 5.2374 -0.14 0.892 PB 1.06269 5.0730 -0.21 0.834 FB 13.05654 5.563 2.35 0.019

Wald Chi2(10) 81.24 Number of Observation 746 Prob > Chi2 0.000 Number of Groups 91

R Square Observation Per Group Within 0.078 Minimum 3 Between 0.238 Average 8.2 Overall 0.1736 Maximum 9 Sigma_u = 10.768006 Sigma_e = 7.1756816 Rho = 0.6924851

26 We have also considered various combinations of independent variables such as log of total assets , growth rates of sector-wise GDP values and so on. Qualitatively the results do not change.

282

Since the model is estimated by using GLS method, the overall significance of the

model is tested by using the Wald test statistic, which is essentially a ‘Chi Square’ test.

The Wald test statistic presented in table 4.4 shows that the model as a whole is highly

significant.

Further, coming to the specific relation of each variable with Gross Non Performing Asset to Gross Advance Ratio, we can note that the coefficient of the Total Asset (TA) is significant with negative sign. Since the TA variable is used as a proxy for the size of the bank, the negative sign of the TA shows that bigger banks have lesser GNPA, in relation to their gross advance. Thus there may be some economies of scale involved in loan recovery process especially when loan is given to the borrowers who may be located near to each other. Such possibilities can arise in case of agriculture loan or, of SSI loans given in a cluster. It also could be due to the reason that bigger banks have better network and more information about the customers due to which they manage their asset well, which helps them to keep the GNPA level under check. Further, it is argued in the literature that the massive branch expansion, particularly into rural areas might have led to increased GNPA levels for the commercial banks. Our result seems to support this argument. The positive sign of the RRB/TBB ratio shows that the higher rural bank braches will lead to increased GNPA in the commercial banks. One of the important variables that has strong influence on the level of GNPA of commercial banks is the economic condition. If the economy is doing well agriculturist ,industrial entrepreneurs or other borrowers will be able to repay the borrowed loan. Many studies have used GDP as a proxy for the economic condition. However, as mentioned earlier, the total GDP may not capture the differential growth rates of various sectors. Thus in the present study the GDP growth of three important sectors viz., agriculture, industry and service sector is included in the analysis. Our results justify our approach of including the growth rates of sector wise GDP. Out of the three sector wise GDP related variables all are found to be insignificant.

As mentioned earlier, one of the reasons for the loan default, as argued in the

theory, is the higher cost of bank loan. If the cost of credit is higher it will attract only the

risky borrowers, which would lead to increase in the NPA level. In order to capture the

relation between cost of bank credit and GNPA the Advance rate of SBI was included as

a proxy for the lending rate in the model. However, it has turned out to be insignificant

showing that interest rate on credit does not have any relation with the GNPA of

commercial banks. One reason could be that the lending rate of late has already come

283

down and even though there is some variations in lending rates over the years, they are

not too high to really bring risky borrowers to the lending process. Interest margin is also

not found to be significant. Bank group dummies show that foreign bank dummy is

positively significant. That is, while we take care of other variables like rural branches

etc. there is significant difference between foreign bank and SBI in their abilities to

reduce bad loans27. This result also goes with the finding that larger bank size is better for

reducing bad loans28.

4.6 Sector Specific Determinants of NPA: SSI Sector

We next examine the determinants of NPA at the sectoral level, especially

concentrating on the SSI sector. The model under consideration is similar to the one with

aggregate NPA levels, but now the dependent variable is NPA from the SSI sector as a

percentage of total advances29. Thus the model under consideration is as follows.

itit

itittitit

LRTBBRRBCRARIIPTAGASSINPA

υµαααααα

++++++=

5

43210 /ln]/[ (4.2)

Where,

SSINPA/GA = Ratio of Gross SSI sector Non Performing Asset to Gross Advance

ln TA = log of Total Assets

IIP: Index of Industrial Production

CRAR = Capital to Risk Asset Ratio

RBB/TBB = Ratio of Rural Bank Branches to Total Bank Branches

LR = Lending Rate (SBI Average Advance Rate)

µi = are unobserved bank specific effects which are assumed to be random (like,

bank specific entrepreneurial or managerial skills)

27 This hypothesis will be checked more rigorously when we deal with efficiency issues in Chapter 5. 28 It has been mentioned during the workshop on draft report presentation at Kathmadu by the experts that there may be endogeinity problem in the panel data estimation presented above. Therefore Hausman Wu Durbin endogeneity test has been carried out for different independent variables and the problem is not seen to be present. In the Appendix we present one such results (Table A4.1). 29 However, note that the data considered for the analysis is for the period 2001 –2005, which does not include Foreign Banks.

284

vi = are stochastic term which are assumed to be identically and independently

distributed, IID(0, σ2).

While estimating the panel data model we observe that Hauman specification test

suggests a fixed effect model to be the appropriate one (Table 4.5). In a fixed effect

model bank specific dummies cannot be incorporated. Thus the former model has been

modified accordingly. The results based on this model reveal the following (Table 4.6).

Table 4.5

Hausman Specification Test Results Chi2 Probability18.66 0.0022

After deciding that the model should estimated using fixed model estimation techniques,

it is essential to check whether fixed effects in fact present in the data. In this regard the F

test was conducted, the results of which (see table 4.6) suggest that there are fixed effect

present in the data.

Table 4.6

F-Test for Testing Fixed Effects

F Probability4.72 0.0000

After deciding that the model should be estimated as a fixed effect model and confirming

the presence of fixed effects, the model is estimated using Fixed Effects (Within) method.

The results of the within estimation is presented in table 4.7.

Table 4.7

Determinants of SSI NPA/Gross Advance Fixed-effects (within) regression Dependent Variable: SSINPA/GA Coefficient Std. Err. Z P>|z| Constant 96.9025 25.4395 3.8100 0.0000 ln TA -17.5737 6.6471 -2.6400 0.0090 IIP 0.0011 0.0348 0.0300 0.9760

285

LR -0.1661 0.7573 -0.2200 0.8270 CRAR 0.1096 0.1357 0.8100 0.4200 RBB/TBB -20.4354 10.3834 -1.9700 0.0500

F(5,198) 4.38 Number of Observation 255 Prob > F 0.0008 Number of Groups 52

R Square Observation Per Group Within 0.0996 Minimum 3 Between 0.0011 Average 4.9 Overall 0.0016 Maximum 5 Sigma_u = 13.035762 Sigma_e = 5.0268054 Rho = 0.87054922

The only significant variables are total asset of a bank and the share of rural branches.

Large size banks have lower NPA levels, which is similar to the result we have obtained

above. Importantly, larger number of rural branches appears to have a negative impact on

SSI NPA. Thus rural branches are not appearing to create NPA for the sector. This is a

result of some importance. While we find RBB/TBB to be positively significant in case

of gross NPA , it is negatively significant in the case of SSI NPA. This can happen if the

rural branches provide loans to the SSI sector in cluster and there exists some economies

of scale in monitoring.

The CRAR variable captures the capital to risky asset ratio. This variable is found to be

insignificant in the case of SSI sector NPA. Thus an increase in so called risky assets

does not seem to affect the SSI sector. It is also important to note that if we remove the

CRAR variable and run the estimation , results do not change qualitatively. Only the total

assets and RBB/TBB remain the significant variables, with no change in their signs as

well.

4.7 Conclusion

286

In this chapter we made an attempt to study the relation between GNPA

and some bank specific variables such as bank size, return on assets, the number of rural

branches and also some economy specific variables such as the growth rates of sector-

wise GDP and finally the lending rate. The model was estimated using panel data

technique, specifically through a random effect model. Our results show that bank size

contribute negatively to GNPA. One of the important findings of our analysis is that the

rural bank branches have contributed for higher GNPA in commercial banks. Thus

commercial banks need to focus on the loan recovery process concerning the rural

branches. Such loans usually are small size ones and involve a large number of

borrowers. These borrowers often need other supporting inputs from the banking sector

in addition to credit. Such support may be in the form of providing necessary information

about technology, management techniques, costing and so on for the SSI sector. Similar

supports are necessary to the agriculture sector as well. Finally our result show that the

interest rate charged on the bank credit does not have impact on the GNPA level. This

indicates that current interest rates are not at a significantly high level so as to really bring

risky borrowers to the bank net.

287

Appendix to Chapter 4

A4.1 Hausman specification Test

An important question in panel data analysis is whether to use fixed effect model

or random effect model. The major drawback of the dummy variable model is that it

consumes more degrees of freedom. On the other hand when random component model is

estimated, it is assumed that the individual invariant effects )( itu are not correlated with

the regressors i.e., 0)/( =itit XuE . If this assumption is not valid then the GLS estimator

of the random effect model becomes biased and inconsistent. However, this problem will

not occur under fixed effect model as the within transformation wipes out the )( iu , which

makes the within estimator unbiased and consistent.

In order to choose between fixed effect model and random effect model Hausman

(1978) suggests comparing GLSβ̂ and withinβ̂ under the null hypothesis

that 0)/( =itit XuE . Hausman’s essential result is that the covariance of an efficient

estimator with its difference from an inefficient estimator is zero (Greene 2003). And also

under the null hypothesis the two estimates should not differ systematically. This is tested

by a statistic given by Hausman (1978) which is essentially a chi-squared test, given by,

)(ˆ)( 1/ GLSwithinGLSwithinm ββψββ −−= − ~ χ2 (k)

)()()(ˆ GLSVarwithinVarGLSwithinVar ββββψ −=−= is the difference between the

estimated covariance matrix of the parameter estimates in the LSDV model (within) and

that of the random effects model (GLS). It is notable that an intercept and dummy

variables SHOULD be excluded in computation.

If the test statistic is higher than the critical limits, the null hypothesis that

individual invariant effects are not correlated with regressors is rejected. This indicates

that the GLS estimators are biased which means the fixed effect model is a better choice.

A4.2 Testing Individual Effects in a Fixed Effect Model (F Test)

288

While estimating a fixed effect model one should check whether the fixed effects

are present in the data or not. If the fixed effects are not present the model can be

estimated using the Classical Linear Regression technique. In order to check for the

presence of fixed effect, one has to test the joint significance of the individual specific

dummies. In other words, one her to check the null hypothesis that µ1= µ2…....µn-1 = 0,

by performing an F –test, which can be calculated using the formula given below.

)/(

)1/()(0 KNNTRSS

NRSSRSSF

UR

URR

−−−−

= ~ KTNNF −−− )1(,1

If the calculated F statistic is higher than the critical limit, for the given degrees of

freedom, then the null hypothesis that individual specific dummies are zero is rejected. In

other words, there is presence of individual effect in the data.

A4.3 Breusch - Pagan Test for Random Effects

Similar to the Fixed Effects model estimation, it is essential to check for the

presence of Random Effects under random effect model estimation. If the random effects

are not present, the model can be estimated using the Classical Linear Regression

technique. Checking for the presence of random effects is essentially testing the null

hypothesis that the cross sectional variance components are zero, i.e., 020 == µσH . For

testing this hypothesis Breusch and Pagan (1980) have developed a Lagrange Multiplier

(LM) test, which can be given as (Greene, 2003):

22

1)1(2 ⎥

⎦

⎤⎢⎣

⎡−

′′

−=

eeeeT

TnTLM

In the above formula, e is the n X 1 vector of the group specific means of pooled

regression residuals, and e e' is the Sum of Squared Errors of the pooled OLS regression.

And, the LM is distributed as chi-squared with one degree of freedom.

Table A4.1 Endogeinity Test

289

Endogeinity test for rural bank branches / total bank branches Coefficient Std.Err. Z P>[Z] RRB_Total B.B 11.97679 14.14501 0.85 0.397 RES 25.58089 17.06733 1.50 0.134 GDP_Agr .0000417 .0000432 0.97 0.334 GDP_ Indust -.000026 .0000913 -0.28 0.776 GDP_Ser -8.11 .0000272 -0.30 0.766 IM .3950728 .3200173 1.23 0.217 IR -.8573449 .4942842 -1.73 0.083 NB .4839746 5.288751 0.09 0.927 PB -1.277317 5.299394 -0.24 0.810 FB 7.345348 6.734881 1.09 0.275 Total_Assets -.0000609 .0000194 -3.14 0.002 Constant 19.49389 16.63522 1.17 0.241 F (11) 50.94 Number of Observation 748 Prob>F 0.0000 Number of groups 91 R Square Observation Per Group Within 0.0652 Minimum 3 Between 0.0625 Average 8.2 Overall 0.616 Maximum 9 Sigma_u = 12.081546 Sigma_e = 7.343387 Rho = .73022373

290

CHAPTER 5

Non-Performing Assets and Profit Efficiency

5.1 Introduction

The Non-performing Assets of commercial banks adversely affects commercial

banks in many ways, one of them being the decline in the profit of commercial banks.

Higher NPA has two important impacts on the profit level of commercial banks. First, the

assets that become NPA will not yield any return; second, banks have to make provisions

for the NPA from the profit they earn, which further erodes their profit level. Thus it is

interesting to study the relation between NPA levels and the profit efficiency in a more

rigorous way. Further, may financial sector reforms introduced subsequent to the

Narasimham Committee report (1991), particularly the prudential regulation norms, are

expected to compel the banks to reduce the NPA level, which in turn improve the

efficiency of commercial banks. Thus it of interest to study whether the efficiency of the

commercial banks has improved over time, and importantly whether the reduction in

NPA level has helped commercial banks to improve their efficiency.

Thus in the present chapter an efficiency analysis is conducted using frontier

technique, and an attempt is made to check whether NPA has any relation with

efficiency. The frontier used here is the profit frontier, since profit frontier captures one

of the important objectives of commercial banks namely profit maximization, and also it

is expected that the NPA affects the profitability of commercial banks more than many

other variables concerning the commercial banks.

Given this background, the present chapter is designed as follows; Section 2

presents a brief review of literature related to various studies measuring efficiency of

commercial banks. In Section 3 various methods of measuring efficiency of commercial

banks using frontier method is discussed. Section 4 discusses the detailed methodology

used in the present study. While Section 5 gives various sources of data used in the

present study, in Section 6 the estimated results are presented. Finally, Section 7 presents

the concluding observations.

291

5.2 Review of Literature

There exist a vast literature that concentrates on at the efficiency of commercial

banks in India as well as other countries. However, studies looking at the efficiency of

commercial banks in relation to their Non Performing Assets are very few. The issue no

doubt assume considerable importance, because if there exist a relation between these

two, then the policy prescriptions need to be tailored accordingly.

Studies analyzing the general efficiency of commercial banks look at various

aspects of commercial banks. While some studies look at the efficiency of commercial

banks in producing outputs, whichever manner they are defined, some look at the profit

efficiency of commercial banks and others examine the cost efficiency.

Paster Perez and Quesada (1997), using a non-parametric approach together with

the Malmquist index, analyse the differences in the productivity and efficiency between

different European and US banking system. They further decompose the differences in

productivity of different banking systems into differences in levels of efficiency (catching

up effect) and level of technology (distance from the frontier). The study has found that,

under the assumption of a constant returns to scale production technology, France has the

highest efficiency score of (0.95) and the UK has the lowest efficiency score (0.56).

Vivas (1997) analyse the effects of deregulation on the profit efficiency of

Spanish savings banks over 1986-1991. Profit function is considered to be more

appropriate since it reflects the joint impact of revenue as well as cost effects of

deregulation. The thick frontier approach is used since it selects a relatively large subset

of firms to define frontier unlike the other frontier measures which base the efficiency

estimate on one or a very small subset of firms. The results suggest that the profit

efficiency of Spanish savings banks, which averaged 28 percent, fell by forty percent

between 1986 and 1991. Also there was no significant shift in the profit frontier itself (in

other words, there was no technological change)

Berger and Humphrey (2000) provide a comprehensive survey of 130 studies that

conduct frontier efficiency analysis ( applying different efficiency measurement

techniques) to financial institutions from 21 countries. They note that the impact of

deregulation on the efficiency of banks is mixed. The overall efficiency, in these studies,

is round 77% (median 82%).

292

Berger and Mester (2000) examine the possible sources of differences in

measured efficiency of financial institutions, including differences in efficiency concepts,

measurement methods, the number of banks considered, market, and regulatory

characteristics and others. They estimate the efficiency of almost 6000 US commercial

banks that were in continuous existence over the six-year period from 1990 to 1995. The

study employs three distinct economic efficiency concepts – cost, standard profit, and

alternative profit efficiencies. The paper analyse the effects of a number of measurement

methods, including use of the distribution-free approach versus the stochastic frontier

approach, specification of the Fourier-flexible functional form versus the translog form,

and inclusion of problem loans and financial capital in a number of different ways. Their

results show that the mean cost efficiency from the preferred model is 0.868. It is found

that the mean efficiencies for standard and alternative profit functions are similar to each

other, however, the alternative profit function does not fit the data nearly as well as the

standard profit function. Different functional forms used (translog & Fourier) yield

essentially the same average level and dispersion of measured efficiency, and both rank

the individual banks in almost the same order.

Jonathan and Nguyen (2005) have studied the relationship between commercial

bank performance and bank ownership in South East Asia (Indonesia, Korea, Malaysia,

the Philippines, and Thailand) between 1990 and 2003, where, performance is measured

using three concepts; alternative profit efficiency, technical change and productivity.

They use four types of indicator in their study, viz., changes in governance due to bank

privatization, acquisition by foreign banks, domestic M&A; and bank restructuring. Their

study finds a positive relation between the performance of commercial bank and

deregulation. In terms of state versus private ownership, state-owned banks are found to

under-perform vis-à-vis their private counterparts. Furthermore, the study observes that

bank privatization has raised bank performance to levels in excess of pre-privatisation

era.

Sensarma (2005) measures the cost and profit efficiency of the Indian commercial

banks during the period 1986-2003, using Stochastic Frontier Analysis. The study finds

that while cost efficiency of the banking industry increased during the period, profit

efficiency underwent a decline. Also, in terms of bank groups, domestic banks appear to

293

be more efficient than foreign banks. Further Bigger banks were found to be less efficient

than their smaller counterparts.

Das et al, (2005) empirically estimates and then analyses various efficiency scores

of Indian banks during 1997-2003 using Data Envelopment Analysis (DEA). Instead of

taking a single measure of efficiency, they use multiple measures, i.e. two measures of

technical efficiency, cost efficiency, revenue efficiency and profit efficiency. Their

results show that there is not much difference in the technical efficiency of various banks.

However, for the remaining two measures of efficiency relating to cost and profit banks

appear to be more differentiated; this is particularly true with respect to profit efficiency.

Also, there has been a noticeable improvement in the profit profile of banks over the

years, particularly after 1999-2000. Profit efficiency seems to have a positive relation

with bank size, which shows that bigger banks are more efficient. Further the results also

show that there is a positive association between good management practices and profit

efficiency. Their results, however, does not show any clear association between the

efficiency of commercial banks and their privatization.

Mohan and Ray (2004) compare the performance among public sector banks,

private banks and foreign banks using physical quantities of inputs and outputs, and

comparing the revenue maximization efficiency of banks during 1992-2000. The findings

show that PSBs performed significantly better than private sector banks but no differently

from foreign banks. The study also finds that there is a convergence in performance

between public and private sector banks in the post-reform era.

De (2004) has empirically investigated the interrelation between ownership-

liberalisation-efficiency of the Indian banking industry using a panel data set for the

years 1985-‘86 to 1995-‘96. The study estimates time-invariant and time-variant

technical efficiency levels of the banks in the Indian banking industry using a stochastic

frontier production function by incorporating the Cobb-Douglas technology with four

inputs and two alternative measures of output. The overall finding is that banking

industry is technically inefficient. The average inefficiency levels are 55 percent and 20

percent for the two output measures used in the study. Technical efficiency has increased,

in the post-liberalisation for only 14 banks out of 18 banks, and, for more than two-third

of the banks in our sample technical efficiency is constant over the period.

294

Studies discussed above only concentrate on measuring the performance of

commercial banks through profitability, technical efficiency and/or productivity.

However, in the recent time it is being recognized by many that NPA has strong

implications on the performance of commercial banks. Though the issue of relation

between NPA and the performance of commercial banks has not been explored

sufficiently, there are few attempts in this direction.

Berger and Young (1997) have made one of the early attempts to analyse the

relation between NPA (or problem Loans, as they call it). They employ Granger-causality

techniques to test four hypotheses regarding the relationships among loan quality, cost

efficiency, and bank capital. The four hypotheses they test are; bad luck, bad

management, skimping and moral hazard. The results of their analysis suggest that the

intertemporal relationships between loan quality and cost efficiency run in both

directions. Further provide support for the bad luck hypothesis – increases in

nonperforming loans tend to be followed by decreases in measured cost efficiency,

suggesting that high levels of problem loans cause banks to increase spending on

monitoring, working out, and/or selling off these loans. For the industry as a whole, the

data favor the bad management hypothesis over the skimping hypothesis, however, for a

subset of banks that were consistently efficient across time, the data favor the skimping

hypothesis. Their results also support the moral hazard hypothesis, and suggest that, on

an average, thinly capitalized banks take increased portfolio risk, which results in higher

levels of problem loans in the future.

Dongili and Zago (2005) estimate the technical efficiency of Italian banks by

taking into account problem loans and using directional distance functions. Their results

show that once problem loans are taken into account, the economic efficiency of banks

increase significantly, suggesting that a significant aspect of banking production, credit

quality, needs to be considered when evaluating banks’ performances.

Jordan (1998), examine whether the increase in the problem loans of the banks of

New England is because of the inefficiency of commercial banks or because of the

government policies. In order to determine the reason for the severity of the problem an

analysis of cost and profit efficiencies are conducted using parametric method. In general,

the data suggest that no relationship exists between cost efficiency and problem loans, but

295

they show a positive and statistically significant relationship between profit efficiency

and problem loans. Further, this study finds that higher levels of profit efficiency in the

1984–88 periods are associated with higher levels of problem loans in the 1989–92

periods. These results suggest that managers of these “profit-efficient” banks deliberately

adopted policies designed to generate higher returns, but by taking higher risk.

Matthews, Guo and Zhang (2006) analyse the relation between rational

inefficiency and non-performing loans of Chinese banking industry using boot-strapping

method. They argue that inefficiency relative to 'best practice' is usually blamed on bad

management, ‘rent seeking’ behaviour and poor motivation not just X-inefficiency in the

traditional sense.

Das and Ghosh (2005) examine the interrelationships among credit risk, capital

and productivity change in the Indian context, using the data on state-owned banks

(SOBs) for the period 1995-96 through 2000-01, where credit risk is measured by the

ratio of net non-performing loans to net advances. Their results show that higher

productivity leads to a decrease in credit risk, and also there is a positive relation between

productivity and bank capitalization. This finding supports the fact that poor performers

are more prone to risk taking than better performing banking organizations. Their results

also reveals that efficiency, capital and risk taking tend to be jointly determined,

reinforcing and compensating each other.

5.3 Measuring Efficiency Efficiency has two components: one is purely technical or physical component which refers to the ability to avoid waste by

producing as much output as input usage allows, or by using as little input as production allows. Thus the analysis of technical

efficiency can have an output augmenting orientation or input conserving orientation. Another is the allocative or price component,

which refers to the ability to combine inputs and outputs in optimal proportion in the light of prevailing prices (Lovell, 1993)

Koopmans (1951) was the first to provide a formal definition of technical efficiency. According to his definition a producer

is technically efficient if an increase in any output requires a reduction in at least one other output or an increase in at least one input,

and if a reduction in any input requires an increase in one other input or a reduction in at least one out put. Thus technically inefficient

producer could produce the same outputs with less of at least one input, or could use the same inputs to produce more of at least one

output.

The basic assumption underlying the measurement of technical efficiency is that a gap normally exists between a firm's

actual and potential levels of technical performance. This can be understood from the diagram 1, given in the next page. In the

diagram, the ‘FF’ curve represents the frontier which is the combination of outputs of the best performing firms in the sample.

According to the neo-classical theory, all firms operating on this frontier are technically efficient (Kalirajan and Shand, 1994). A firm

296

operating at point ‘B’, which is on the frontier, will produce Y*1 outputs, which is the maximum possible output for a given set of

inputs X1. Thus the firm operating at point ‘B’ is technically efficient.

.

Diagram 1

Measuring Efficiency

However in practice a firm may not be working at a point on the frontier due to various reasons such as incomplete

knowledge of the best technical practices or other factors that prevent it from operating on its technical frontier. Thus the firm will

operate on an actual or perceived production function, which is below the potential frontier. The firm operating at point ‘A’ is

technically inefficient as it is producing output Y1 which is less than the maximum possible output for the input vector X1. Now, the

technical efficiency of the firm operating at point ‘A’ is measured by the distance between its actual output and the maximum possible

output (which is given by the frontier at ‘B’). Thus the technical efficiency can be measured as the ratio Y1/Y*1.

The most commonly used tool of analysis for measuring technical efficiency is the primal production function. In the neo-

classical theory of production, the primal production function defines the maximum possible output of a firm for combinations of

inputs and technology, i.e., it is frontier production function. The production frontier of the firm, producing a single output with

multiple inputs can be defined as,

);( βii xfy = (1)

Where, yi and xi are output and inputs of the ith firm. Here the firm is operating on the frontier, producing maximum

possible output, thus there is no technical inefficiency. But in reality this may not be the case. A firm, say the ith firm may not be

producing its maximum possible output. Then the production function of ith firm can be written as,

iii TExfy ⋅= );( β (2)

X

Y

A

B

Y1

Y*1

F

F

X10

297

Where, TEi is the technical efficiency of the ith firm which represents the combined effects of various non-price and

organisational factors which constrain the firm from obtaining its maximum possible output. The value taken by TEi depends on the

extent to which the firm is affected by constrains. A measure of technical efficiency of the firm can be defined as.

);( βi

ii xf

yTE = (3)

The above model is a basic model generally used for measuring technical efficiency. Here yi achieves its maximum

possible level only if TEi = 1. In this model, the numerator is observable but the denominator is not. Various methods using different

assumptions have been suggested in the literature to estimate the denominator and there by TEi. Farrell (1957), who gave the definition

of efficiency, also suggested that the production frontier can be can be estimated from sample data using either a non-parametric

piece-wise-linear method or a parametric function, such as the Cobb-Douglas form. Since then the basic model has been extended by

many ways. Various methods of estimating frontier production function, and thus technical efficiency, can be conveniently grouped

under two major groups, namely, programming (deterministic) and statistical (stochastic) methods. The classification of

different frontier production function methodology is given in the chart 1.

Chart 1

Deterministic Approach:

Frontier Methodology

Stochastic

Non Neutral Shift Neutral Shift

Deterministic

Parametric

Programming Statistical

Cross Section Panel Data

Non Parametric

Time Variant Time Invariant

298

Farrell pioneered the work on the deterministic approach of measuring technical efficiency in 1957, following the notion of

Debreu (1951) and Koopmans (1951). He assumed constant returns to scale to estimate the frontier production function, and the model

included one out put and two inputs. This model was extended to one output and m-inputs, and a functional form, Cobb-Douglas, was

specified by Aigner and Chu (1968). The principal disadvantage of Farrell model is the assumption of constant returns to scale, which

is quite restrictive. This was extended to increasing returns to scale by Seitz (1970). The Farrell model was generalised in terms of

vector inputs and vector outputs by Carnes,et,al (1978), which is Known as Data Envelopment Analysis (DEA). DEA was further

developed by Varian (1985) who incorporated stochastic characteristics. The aim was to introduce two sided deviations to include

random noise and to calculate the efficiency measures free of such random noise. Land,et.al (1980) provided an alternative approach

to Varian by allowing deterministic frontier to capture the effects of random noise without themselves becoming stochastic. In

literature this method is termed as ‘chance–constrained efficiency analysis’.

Stochastic Frontier Approach:

The programming method does not take into account the statistical errors. This was first pointed out by Timmer (1971)

who provided a simple method to deal with these errors. The full fledged stochastic frontier production function estimation was first

published by Aigner,Lovell and Schmidt and Meeusen and Van den Broeck independently in 1977. Aigner, et.al used a truncated

normal (half normal) distribution for ‘u’, whereas Meeusen and Van den Broeck used exponential for the same. However, this

stochastic frontier production function approach could only provide average technical efficiency measures for the sample

observations. Although these aggregate measures are useful in a way, individual observation-specific technical efficiency measures

are more useful from a policy viewpoint. Jondrow et,at (1982) and Kalirajan and Flinn (1983) independently considered the stochastic

model introduced by Aigner et,al (1977) and Meeusen and Broeck (1977) to predict the combined random variable (ui+ vi). Estimation

of the stochastic frontier production function for a single cross-section requires the explicit specification of distribution of statistical

noise and inefficiency variable term. Several distributions have been considered in the literature, although the most commonly

employed are the positive ‘half normal’ and the ‘exponential’. Much of the criticism surrounding these estimates of the production

frontier have been concerned with the strength of the distributional assumptions. Such strong assumptions are not required when panel

data are available. Schmidt and Sickles (1984) have used panel data to estimate production frontiers. The panel data model can be a

‘fixed effect’ model or ‘random effect’ model; it can have time invariant effect and also time varying effect.

The Stochastic Varying Coefficient Frontier Approach:

Under both approaches discussed above, the firm obtains its frontier potential output by following the best practice

techniques, given the technology. In other words, frontier output is determined by the method of application of inputs, regardless of

levels of inputs. Empirical evidence shows that with the same levels of inputs, different levels of actual output are obtained by

following different methods of application. This implies that the different methods of applying various inputs will produce different

outputs. This means that diversity of individual decision making behaviour leads to variation in production response coefficients,

which include not only the intercept but also slope coefficients, across units and over time for the same unit. This idea was first

appreciated by Nerlove (1965) and was later popularised by Swamy (1970). Kalirajan and Obwana (1994) showed that this method

facilitates the estimation of firm specific and input specific technical efficiency values. Kalirajan and Shand (1994) discuss the

modelling of the frontier production function with cross sectional heterogeneity in slopes and intercepts.

5.4 Methodology

Efficiency analyses in the present study are of two kinds. First, we analyse the

impact of NPA on the efficiency of commercial banks, in particular on the profit

299

efficiency of commercial banks. Second, we analyse the efficiency of commercial banks

in minimising the NPA, where NPA is considered as a bad output. Further we also

analyse the factors that determine the NPA inefficiency.

In order to analyse the impact of Non Performing Assets on the profit efficiency

of commercial banks, we need to first obtain the profit efficiency measures using

stochastic frontier method. Then the impact of NPA on the profit efficiency will be

analysed by regressing the profit efficiency estimates over NPA along with a set of other

control variables. In order to analyse the NPA efficiency of commercial banks, a NPA

frontier is estimated using the available data, and the NPA efficiency is derived as the

ratio of ‘minimum NPA/Actual NPA’, where the minimum NPA is given by the

frontier30. In the present study efficiency is estimated using the stochastic frontier analysis. The basic model of the SFA was

discussed in the previous section. This basic model has been extended by number of ways. Earlier approaches were using a two stage

estimation procedure, where, in the first stage firm-level efficiencies were predicted using stochastic frontiers and in the second stage

they (firm-level efficiencies) were regressed upon some firm-specific variables to identify some of the reasons for differences in

predicted efficiencies between firms. Though this procedure was useful in estimating firm-level inefficiencies, this was considered as

inconsistent in its assumptions regarding the independence of the inefficiency effects in the two estimation stages [Kumbhakar, Ghosh

and McGukin, 1991; Reifschneider and Stevensosn, 1991]. To overcome this problem Battese and Coelli (1993) proposed a model in

which inefficiency effects are expressed as explicit function of a vector of firm specific variables and a random error. Present study

follows the approach used by Bettese and Coelli (1993). The frontier specification used to derive efficiency estimates can be given as

follows;

itit

itititititit

itititititit

ititititit

uvDbWbINbADb

DWbDINbWINbDADbWADbINADb

TbDbWbINbADbbP

−+++++

++++++

+++++=

]lnlnlnln[)2/1(

lnlnlnlnlnlnlnlnlnlnlnln

lnlnlnlnln

244

233

222

211

342423

141312

543210

(4)

Where, P is the Net Profit31

AD is the Advance (output)

IN is the Investment (output)

30 Note that, this method is similar to the method of measuring cost efficiency. However, here the functional form used is a production function, not cost function. It is assumed that commercial banks minimize their NPA level, given the input level. 31 Note that the dependent variable for the profit function is [π + |π|min + 1], where |π|min is the absolute of the minimum value of net profit (π) over all banks. Since the net profit of most banks are negative the constant [|π|min +1] is added to every firm’s net profit so that the natural logarithm is taken of a positive number.

300

W is the wage to capital price ratio (input price)

D is the deposit price to capital price ratio (input price)

T is the time trend

Vit are random variables, which are assumed to be iid. N (0, σv2) and independent of Uit

Uit are non-negative random variables which are assumed to account for technical

inefficiency in production and are assumed to be independently distributed as truncations at zero

of the

N (mit, σu2) distribution;

Where; mit = zitδ,

Where; zit is a 1 x p vector of variables which may influence the inefficiency of a firm

and δ is a p x 1 vector of parameters to be estimated.

The parameterization from Battese and Corra (1977) are used replacing σv2 and σu2 with σ2 = σv2 + σu2 and the parameters

are estimated by ML approach32.

Further, in order to analyse the impact of NPA on the efficiency levels of commercial banks, the estimates of technical

inefficiency (Ui) are regressed over NPA along with a set of control variables. The functional form used to analyse the impact of NPA

on the inefficiency of commercial banks can be given as follows33;

it

it

WFBPBNBGDPIMTBBRBBTAGAGNPAU

+++++++++=

876

543210 //δδδ

δδδδδδ (5)

Where,

U is the Profit Inefficiency

GNPA/GA is the ratio of Gross Non Performing Assets to Gross Advance

TA is Total Assets

RBB/TBB is the ratio of Rural Bank Branches to Total Bank Branches

IM is the Interest Margin (Measured as the difference between lending rate and deposit rate).

GDP – Gross Domestic Product

NB, PB, and FB are bank group specific dummies for Nationalised Banks, Private Banks

and Foreign Banks respectively34

32 The log-likelihood function is given in Battese and Coelli (1993) 33 Note that under Bettese and Coelli approach both frontier and inefficiency equation are estimated simultaneously using ML method. 34 The State Bank of India and Associates (SBI&A) dummy is not considered in order to avoid dummy variable trap. Therefore, the specified dummy variables should be interpreted in comparison to the SBI&A, which serves as the base.

301

As mentioned before, apart from analysing the impact of NPA on the profit efficiency of commercial banks, an attempt is

also made to analyse the efficiency of commercial banks in minimising the NPA level. Here NPAs are considered as output that

commercial banks actually try to minimise. However, this is difficult to estimate in a straightforward manner due to non-availability of

statistical package. In order to measure the NPA efficiency, first a production frontier is estimated considering NPA as the output,

using the standard production frontier approach, under which, the firm that produces the maximum output, (given a set of input and

technology) will be the most efficient one. This firm then can equivalently be interpreted as the one that is least efficient in reducing

NPA with a given level of inputs. The frontier used to derive the NPA efficiency can be specified as follows:

( ) ititititit

itititititit

itititit

uvMaLaKa

MLaMKaLKaTaMaLaKaaGNPA

−++++

+++++++=

]lnlnln[2/1

lnlnlnlnlnlnlnlnlnln

233

222

211

231312

43210

(6)

Where,

Y is Gross Non Performing Assets

K is Capital (fixed assets)

L is Labour (employees)

M is Material (stationery, postage etc)

T is the time trend

Vit are random variables, which are assumed to be iid. N (0, σv2) and independent of Uit

Uit are non-negative random variables which are assumed to account for technical

inefficiency in production and are assumed to be independently distributed as truncations at zero

of the

N (mit, σu2) distribution;

Where; mit = zitδ,

Where; zit is a 1 x p vector of variables which may influence the inefficiency of a firm

and δ is a p x 1 vector of parameters to be estimated.

The parameterization from Battese and Corra (1977) are used replacing σv2 and σu2 with σ2 = σv2 + σu2 and the parameters are

estimated by ML approach35.

Further, the functional form used to analyse the determinants of NPA efficiency can be given as follows;

it

it

WFBPBNBLRGDPIMTBBRBBTAU

+++++++++=

8765

43210 /δδδδ

δδδδδ (7)

35 The log-likelihood function is given in Battese and Coelli (1993)

302

Where,

U is the NPA inefficiency estimates

TA is Total Assets

RBB/TBB is the ratio of Rural Bank Branches to Total Bank Branches

IM is the Interest Margin (Measured as the difference between lending rate and deposit rate).

GDP – Gross Domestic Product

LR – Lending Rate

NB, PB, and FB are bank group specific dummies for Nationalised Banks, Private Banks

and Foreign Banks respectively36

W : random error.

Before proceeding further, it is essential to discuss the rationale for including each variable in the profit inefficiency

equation as well as NPA efficiency equation. In the profit inefficiency equation, the first variable on the right hand side of the equation

is the ratio of Gross Non Performing Assets to Gross Advance. Since higher level of NPA calls for higher level of provisioning, it

affects the profit level of commercial banks. Thus the GNPA/GA is expected to have a negative impact on the profit efficiency. Total

assets represent the size of the bank. It is interesting to know whether size of the commercial banks has any impact on the efficiency

levels of commercial banks. Though the relationship between the bank size and efficiency level is an empirical question, since bigger

banks would have economies of scale, size of bank can be expected to have a positive relation with the efficiency level of commercial

banks. It has been argued that branch expansion policy followed by RBI and the resulting expansion of commercial banks branches in

the rural areas has adversely affected the efficiency of Indian commercial banks. Thus it is important to understand the relationship

between the rural bank branches and the efficiency level. In this regard the ratio of rural bank branches to the total bank branches has

been included in the inefficiency equation. An important development in the post-liberalisation period is the increasing competition. It

has been argued by many that the competition increase the efficiency of commercial banks. Thus in order to capture the impact of

competition on the efficiency of commercial banks, the Interest Margin (IM) is included in the study. The rationale for including IM

as a measure of competition is as follow; when the competition increases, it increases the pressure on the competing banks to minimise

the margin they receive by financial intermediation. Thus when the IM is declining it indicates that the competition is increasing. One

more important variable that has strong influence on the efficiency of commercial banks is the economic condition. Higher economic

growth means higher productivity, which will lead to higher financial transaction, which in turn will lead to higher profit efficiency of

commercial banks. Thus GDP, which is included to capture the economic condition, is expected to have a positive impact on the profit

efficiency of commercial banks. The bank group specific dummies are included to capture the bank group specific characteristics that

can have influence on the profit efficiency of commercial banks.

Coming to the NPA efficiency equation, most of the variables are similar to the ones included in the profit inefficiency

equation. Here also, total asset is included as a size variable. Similar to the case of profit inefficiency equation, here too the ratio of

rural bank branches to total bank braches is included in order to capture the effect of the branch expansion policy of the RBI and

government on the NPA level of commercial banks. Increasing competition would compel commercial banks to reduce their NPA

level, thus Interest Margin (IM), which is included as a proxy for the competition is expected to have a positive relation with NPA

(reduction) efficiency37. Further, an improvement in the economic condition would enable the firms to repay the borrowed loans,

which in turn would reduce the NPA level of commercial banks. Thus GDP is expected to have a positive relation with NPA reduction

36 The State Bank of India and Associates (SBI&A) dummy is not considered in order to avoid dummy variable trap. Therefore, the specified dummy variables should be interpreted in comparison to the SBI&A which serves as the base. 37 By NPA efficiency here we mean efficiency in reducing NPA levels.

303

efficiency. An important variable that is expected to have strong implications on the NPA of commercial banks is the Lending Rate

(LR). In this regard the SBI advance rate is included in the NPA inefficiency equation as a proxy for the LR. Finally the bank group

specific dummies are included to capture the bank group specific characteristics that can have influence on the profit efficiency of

commercial banks.

5.5 Sources of Data Data are collected for the period 1997-2005. However as many private and foreign banks are established after 1997 and

few are closed during the study period, data are not available consistently for all banks for all years. Thus data used is an unbalanced

panel of 94 banks (27 public sector banks, 33 Indian private banks and 34 foreign banks) for 9 years (total observation used are 746).

Banks are grouped into four groups38 (i) State Bank of India and Associates (SB & A) (ii) Nationalised Banks (NB) (iii)

Private Banks (PB) and (iv) Foreign Banks (FB). Net profit is measured as Gross Profit less (-) provisions and contingencies.

Advances are measured as total advances and Investments are measured as total investment. Data on net profit, advances and

investment are measured in value terms and are collected from Annual Accounts of Scheduled Commercial Banks published by

Reserve Bank of India.

Price of labour (employees) is obtained by dividing the total expenditure on employees by total number of employees and

the price of capital is obtained by = (total operating cost – total expenses on labour)/total fixed assets39. Price of deposits is obtained

by dividing the total interest expenditure on deposits by total deposits.

Further, fixed capital (or capital stock, (K)) is the sum of premises, furniture and other fixed assets40. Number of employees

(or Labour (L) is measured as the total number of employees which include officers, sub-ordinates and clerks. Material (M) is

measured as the sum of expenditure on printing & stationeries and postage, telegrams & telephones etc. while data on the fixed capital

and material is collected from the balance sheet of commercial banks, presented in Annual Accounts of Scheduled Commercial Banks,

data on the number of employees are collected from Statistical Tables Related to Banks in India, both documents published by

Reserve Bank of India

5.6 Empirical Results

The focus of the present study is to analyse the relation between the Gross Non-

Performing assets and the efficiency levels of commercial banks. Thus emphasis will be

given on the estimates of the inefficiency equation (equation 5). However, measurement

of efficiency is a pre-requisite to understand its relation with the GNPA of commercial

banks, it would be interesting to have a look at the efficiency levels of commercial banks

before proceeding further. The estimated coefficients of the profit frontier are presented

in the appendix (Table A5.1). It can be seen from the table that majority of the

coefficients are significant. The value of the gamma shows that majority of residual

38This grouping is done following the standard classification of RBI 39 Similar method is used by Kumbhakar and Sarkar (2003) 40 Capital stock is converted into its present value using perpetual inventory method

304

variation is due to the inefficiency effect41. An interesting observation is that the value of

the time trend shows that there is technical progress in the Indian commercial banks. The

profit efficiency estimates obtained using the profit frontier are averaged across bank

group for each year and are presented in table 5.1.

Table 5.1 Profit Efficiency Estimates

Year SB&A NB PB FB TBS 1997 0.9022 0.9217 0.9667 0.7742 0.9578 1998 0.9106 0.9768 0.9711 0.7645 0.9677 1999 0.8706 0.9688 0.9551 0.7008 0.9581 2000 0.8649 0.9618 0.9621 0.6878 0.9548 2001 0.8670 0.9408 0.9517 0.7153 0.9342 2002 0.8582 0.9439 0.9552 0.6879 0.9418 2003 0.8537 0.9466 0.9524 0.6794 0.9502 2004 0.8957 0.9631 0.9611 0.7456 0.9651 2005 0.9152 0.9709 0.9691 0.7817 0.9749

Average 0.8820 0.9549 0.9605 0.7264 0.9561

After estimating the profit frontier, profit efficiency is estimated as the ratio

between actual profit (profit of bank ‘i’) and the maximum possible profit (profit of the

best performing bank in the sample). The average profit efficiency of the total banking

sector, for the total study period is 0.956, which means that Indian commercial banks are

around 96 percent efficient in earning profit in relation to the best performing bank. In

other words, Indian commercial banks are loosing around 4 percent of the net profit due

to technical inefficiency. At the bank group level private banks are the most efficient

bank group with an efficiency estimate of around 96 percent followed by Nationalised

banks with an efficiency estimate of around 95 percent. SBI&A rank third in terms of

profit efficiency with an efficiency estimate of around 88 percent and foreign banks are

the least efficient banks with an efficiency estimate of around 78 percent. The reason for

private banks being most efficient is, their non-interest income is higher than public

sector and foreign banks, which is mainly because they focus more on providing fee

based services than conventional banking activities.

41 A value zero for gamma indicates that the deviations from the frontier are due entirely to noise, whereas a value of one would indicate that all deviations are due to technical inefficiency. Gamma is estimated as; γ=σ2/σs

2; where σ2 is the variance of ui (inefficiency term), and σs2 is total variance (variance of vi plus

variance of ui).

305

Now we move on to the important question of this chapter, i.e., the impact of

GNPA on the profit efficiency of commercial banks. As discussed in section 5.4, the

impact of GNPA on the efficiency of commercial banks is analysed by regressing the

profit efficiency estimates over the GNPA/GA ratio along with some of the control

variables. The results of the regression are presented in table 5.2.

Table 5.2 Determinants of Profit Inefficiency

Coefficient Standard-Error t-ratio

Constant -1.1933 1.8230 -0.6546 GNPA / Gadv 0.0047* 0.0010 4.7355 Total Asset -0.1627* 0.0309 -5.2667 RBB/TBB -0.2555 0.1895 -1.3484 IM 0.0612** 0.0155 3.9455 GDP -0.4408*** 0.2561 -1.7213 NB 0.6291** 0.2320 2.7110 PB -0.1749 0.2993 -0.5844 FB 2.368481* 0.396584 5.972209

*, **, *** :- Significant at 1%, 5% and 10% respectively

Results presented in table 5.2 shows that majority of the coefficients are

significant and the signs are according to expectation42. The coefficient of GNPA/GA

ratio is significant with positive sign. This shows that GNPA/GA ratio has a positive

relation with profit inefficiency, which means that when the GNPA as a ratio of Gross

Advance increases the profit inefficiency of commercial banks will increase. This shows

that Gross Non Performing Assets adversely affect the efficiency of commercial banks.

Further the coefficient of Total Asset is significant with negative sign. This shows that

size of commercial bank has a negative relation with profit inefficiency. In other words,

bigger banks are more efficient compared to smaller banks in the sample in terms of

earning higher profit. This could be because since bigger banks can lend more loans, they

can diversify their risk, due to which handle the loss arising through loan loss. As

42 Note that since the inefficiency equation is estimated simultaneously with frontier, the goodness of fit statistic presented in table A5.1 in appendix also serves as the goodness of fit statistic of the results of inefficiency estimation presented in table 5.2.

306

mentioned earlier, one of the argument in the literature is that the expansionary policy

followed by the RBI in terms of expanding branch network has adversely affected the

efficiency of commercial banks. Our results, however, does not seem to provide any

evidence in favour of such argument. This could be due to the reason that the time period

considered for our analysis falls in the post-liberalization period, where many of the

branch licensing policies has been amended due to which many loss making rural bank

branches have been closed. One of the important developments in the post-liberalisation

period is the increasing competition, which is expected to increase the efficiency of

commercial banks. The coefficient of Interest Margin being significant, with positive

sign, supports this argument. As mentioned earlier, when the competition increases it

increases the pressure on commercial banks to cut down on the interest margin they earn.

Thus a declining IM is an indication of increasing competition. The positive sign of IM

shows that when the IM declines (i.e., competition increases) the inefficiency also

declines (i.e., efficiency increases). Further, the coefficient of the GDP has a negative

sign indicating that an improvement in the economic condition of the country will help

the commercial banks to reduce their inefficiency level.

As mentioned earlier, the NPA efficiency presented in table 5.3, using the

standard production frontier approach. Results of the NPA frontier are presented in the

appendix. As can be seen from the table A5.2, majority of the coefficients are significant.

The value of the gamma shows that majority of residual variation is due to the

inefficiency effect. After estimating the NPA frontier,

Table 5.3

NPA Efficiency Estimates

SB&A NB PB FB Total 1997 0.8978 0.8971 0.6047 0.4234 0.6536 1998 0.9109 0.9019 0.6401 0.5101 0.6864 1999 0.926 0.9128 0.7199 0.5405 0.7191 2000 0.9238 0.9149 0.7448 0.5346 0.7221 2001 0.9248 0.9227 0.7787 0.5624 0.7508 2002 0.9131 0.9309 0.8041 0.5402 0.7502 2003 0.8902 0.9297 0.8073 0.5403 0.7465 2004 0.8562 0.9226 0.7879 0.5462 0.7534

307

2005 0.8406 0.9032 0.7766 0.5281 0.7426 Average 0.8982 0.9151 0.7405 0.5251 0.725

The average efficiency estimate of the total banking sector for the total study period is around 0.72. This shows that Indian commercial banks are around 72% percent efficient in incurring NPA compared to the bank, which is most efficient in incurring NPA (or least efficient in reducing NPA). It was observed in the case of profit efficiency that, Private Bank group was the most efficient bank group followed by the NB and SB&A. This order seems to have changed in the case of NPA efficiency. Here, Foreign Banks are the most efficient bank group in reducing NPA as they have the lowest score on an average; followed by Private Banks (PB). The SB&A and Nationalised Banks are the least efficient banks in reducing NPA with high efficiency scores in incurring NPA (table 3). It is worth noting in this context that NB holds the highest amount of NPA, in absolute terms as well as percent of gross advance, compared to other bank groups.

Looking at the temporal behaviour, the NPA efficiency of the total banking sector

has declined steadily from 1997 till 2001, and remained almost stable there thereafter.

From 1997 till 2002, this trend of NPA efficiency follows the trend of the GNPA of total

banking industry, which has increased steadily from Rs 50815 crores in 1997 to Rs 70861

crores in 2002 (in real terms), and thereafter since 2002 GNPA has declined then reached

Rs 58299 crores in 2005. NPA efficiency scores have remained almost stable in this

period. At the bank group level the temporal behaviour of the NPA efficiency is not

uniform. The SB&A group has shown increase in efficicny in reducing NPA. This is

however, not seen to be true for the nationalized banks . Private banks efficiency scores

also show a fluctuating trend. The temporal behaviour of the NPA efficiency of FB is

distinct from other bank groups. It has remained more or less stable over the period.

We next move on to another interesting question, viz., what are factors that have

influence on the NPA efficiency level. In this regard, the NPA efficiency equation is

estimated, along with the frontier, where the NPA efficiency estimates are regressed over

a set of variables, which are expected to have influence on the NPA efficiency of

commercial banks (Table 5.4).

308

Table 5.4

Determinants of Efficiency in reducing NPA Coefficient Standard-Error t-ratio Constant -6.0298 4.9092 -1.2283Total Asset -1.0925* 0.0847 -12.9054RBB/TBB -4.2486* 0.8343 -5.0922IM 0.0041 0.0235 0.1745GDP 0.7617 0.7340 1.0377LR 0.3159* 0.0656 4.8158NB -0.5196* 0.0979 -5.3054PB 1.0243* 0.2934 3.4905FB 0.9384* 0.3378 2.7776

* - Significant at 1%, ** - Significant at 5%

Total Assets is having a negative relation with NPA reduction efficiency, which

means bigger banks are less efficient in minimizing their Gross Non Performing

Assets. The ratio of rural bank branches to total bank branches also has a negative

relation with NPA reduction efficiency, which means, when the rural bank branches

are increasing, the NPA efficiency declines. This also shows that banks, which have

higher rural bank branches, are less efficient in minimizing their GNPA. Competition

and rate of interest also do not seem to have any impact on the NPA efficiency.

Similarly, GDP also does not seem to have any impact on the NPA efficiency. Bank

309

group dummies show that, NB is less efficient than SB&A whereas PB and FB are

more efficient than SB&A43.

5.7 Conclusion

Non Performing Assets adversely affect the efficiency of

commercial banks. In particular higher NPA has strong implications on

the profitability of commercial banks. On the one hand the assets that

become NPA will not yield any return; on the other hand, banks have to

make provisions for the NPA from the profit they earn, which further

erodes their profit level. Thus higher NPA has strong implication on the

profit efficiency of the commercial banks. In the present chapter an

attempt is made to understand the relation between Gross Non

Performing Asset to Gross Advance ratio and the profit efficiency of

commercial banks and also to analyse the efficiency of commercial

banks in minimizing the Gross Non Performing assets. The relation

between GNPA/GA and profit efficiency is analysed by regressing the

profit efficiency estimates, obtained through frontier analysis, on the

GNPA/GA ration along with some of the control variables. And, the

efficiency of commercial banks in minimizing the GNPA level is

analysed by estimating the NPA efficiency through GNPA frontier,

where GNPA is considered as a bad output and it is assumed that banks

try to minimize it.

Our results suggest that GNPA/GA ratio has significant relation

with profit efficiency of commercial banks. The sign of the GNPA/GA

ratio show that when the GNPA/GA ratio increases the profit efficiency 43 Note that since SB&A dummy is not included in order to avoid a dummy variable trap, rest of the bank group specific dummies should be interpreted with reference to SB&A.

310

of commercial banks also increases. Our results also show that bigger

banks are more efficient in terms of earning profit however, they are

less efficient in minimizing the GNPA level compared to smaller banks.

Further, the analysis suggests that while the higher proportion of rural

bank branches does not have any impact on the profit efficiency of

commercial banks, it does affect the NPA efficiency adversely. Contrary

to the result of rural bank branches, the coefficient of IM shows that

increasing competition lead to increased profit efficiency of commercial

banks, which however, does not seem to have any impact on the NPA

efficiency. Finally, the improvement in the economic condition of the

country will help the commercial banks to improve their profit

efficiency, whereas it does not seem to have any impact on the NPA

efficiency44.

Appendix To Chapter 5

Table A5.1 Profit Frontier Estimates Coefficient Standard-Error t-ratio Constant 4.8865* 0.0877 55.7394 Adv 0.0089 0.0897 0.0989 Inv -0.2289** 0.0882 -2.5941 W 0.2151** 0.0662 3.2471

44 It must be kept in mind that these are preliminary set of results, which will be again re-checked.

311

D 0.7442* 0.0916 8.1262 Adv2 -0.0047 0.0257 -0.1822 Inv2 -0.0153 0.0332 -0.4627 W2 -0.0102 0.0202 -0.5031 D2 0.0643 0.0495 1.2983 Adv*Inv 0.0558 0.0569 0.9809 Adv*W 0.0992** 0.0374 2.6539 Adv*D -0.1210** 0.0545 -2.2193 Inv*W -0.1207** 0.0364 -3.3170 InvD 0.1656** 0.0540 3.0662 W*D -0.1178** 0.0385 -3.0615 T 0.0107* 0.0014 7.9187 sigma-squared 0.1308* 0.0099 13.2378 Gamma 0.9932* 0.0011 895.8645

Log likelihood function = 818.6481 LR test of the one-sided error = 1195.1023

*, **, *** :- Significant at 1%, 5% and 10% respectively

Table A5.2 NPA Frontier Estimates

Dependent Variable - log(GNPA) Coefficient Standard-Error t-ratio Constant 2.3048* 0.2796 8.2431ln PC 0.1093 0.1867 0.5857ln L -0.2354 0.2556 -0.9207ln M -0.6550** 0.2711 -2.4164ln PC2 0.3355* 0.0632 5.3087ln L2 0.0250 0.0589 0.4252ln M2 -0.2350*** 0.1278 -1.8393ln PC*ln L -0.1320*** 0.0764 -1.7269ln PC*ln M -0.5369* 0.1467 -3.6597ln L*ln M 0.6238* 0.1263 4.9393Time 0.0100*** 0.0051 1.9680sigma-squared 0.4588* 0.0457 10.0386gamma 0.9578* 0.0062 154.6655log likelihood function -105.4861LR test of the one-sided error 487.4373*, **,:- Significant at 1%, and 5% respectively

312

CHAPTER 6

Problem of Loan Repayment: Views of the SSI Units

6.1 Introduction Much has been written about the problems of the SSI sector in the Indian context. The

growth of small scale industry in India, to a large extent is induced by the lack of

alternative employment opportunities and promotional policies adopted by government

(Desai, 1983)45. Due to lack of entrepreneurial attitude and proper training a large

number of them have met with untimely closure. Some of the major problems can be

identified as follows.

• Financial Constraints: Though there are a number of efforts to provide finances

to the SSI units, the most needy ones do not get proper information about

different schemes and often depend on the informal credit market for finance.

Coming up with an appropriate proposal also becomes difficult for such small

entrepreneurs.

• Access to raw materials is another problem faced by these units.

• One of the major handicaps of the small-scale sector has been the absence of

improved technology, which alone can ensure quality and higher productivity.

Technology is the most essential factor to remain competitive in a global market.

Lack of information again plays a critical role in the choice of technology.

• Marketing remains the major stumbling block for the growth of SSI sector.

Ignorance of potential markets, in particular, unfamiliarity with export activities

contributes to this problem. Poor designing and finish also often makes the

product not salable in the international market.

45 Desai. V, 1983, Problems and Prospects of Small Scale Industries in India, Himalaya

Publishing House, Bombay.

313

• As far as production methods are concerned there is often faulty planning and

inadequate appraisal of projects (Desai, 1983). Most often no proper viability

studies, technical or economic, are carried out.

Policy makers recognized these shortcomings and created a number of offices to handle

the problems. However, proper implementation has never been accomplished.

Consequently sickness remained a major problem for this sector. To understand the

problems of the sector in general and concerning credit facilities in particular in a

liberalized regime, a survey has been taken up. Though the survey has been conducted in

three states of India viz., Karnataka, Kerala and West Bengal.

6.2 Sampling Technique

It is well recognized that industry sector is not forthcoming in providing information.

This problem has noted down by many authors (see Deshpande et al, 2004)46. In this

background we are forced to adopt snowball sampling technique. Industry Association

gave contacts of the firms and requested them to cooperate with the survey. The firms

that agreed for a discussion were later interviewed using structured questionnaires that

were personally canvassed. The sample firms are from Kerala, Karnataka and West

Bengal and the sample sizes from these locations are respectively 50, 100 and 50.

6.3 Characteristics of the Sampled Firms

One of the important findings of the survey is that 35% of the manufacturing firms in our

sample are not availing loan from the institutional sources. These firms reported to

manage their investments from their own (or borrowing from relatives) resources. Out of

the rest 68% only 2% have availed loan from private banks. Thus dependency on public

sector banks remains prevalent.

46 Deshpande, Lalit et al, 2004, Liberalization and Labour: Labour Flexibility in Indian Manufacturing, Institute of Human Development, New Delhi.

314

While examining the investment in plant and machinery one observes that most of the

firms included in the survey falls under the Government of India definition of SSI. Only

about 2% of the firms have investment above Rs 1 crore. However as far as commercial

banks are concerned, for credit purposes they combine small and medium firms together

table 6.1).

Table – 6.1.Total investment

Investment Percent10000 - 100000 5 100000 - 500000 17.5 500001 - 1000000 37.5 1000001 - 2500000 15 2500001 - 5000000 12.5 5000001 - 1 crore 10 10000001 - Above 2.5 Total 100

Average turnover is between 30 lakhs to 50 lakhs. Thus these are comparatively large

sized firms within the SSI category (table 6.2).

Table 6.2. Total turnovers of the firms.

Turnover in Rs. This year Last year On an

average 300000 - 1000000 8 (20.00) 9 (22.50) 8 (20.00) 1000001 - 3000000 11 (27.50) 12 (30.00) 11 (27.50) 3000001 - 5000000 8 (20.00) 8 (20.00) 10 (25.00) 5000001 - 7000000 5 (12.50) 2 (5.00) 2 (5.00) 7000001 - 1 crore 3 (7.50) 3 (7.50) 3 (7.50) 10000001 - Above 5 (12.50) 5 (12.50) 6 (15.00)

Total 40 (100) 40 (100) 40 (100)

These firms are managed by qualifies entrepreneurs as can be seen from table 6.3. Above

50% of the firms owners are technical degree holders.

315

Table – 6.3 Educational qualifications of the firm manager and Owners.

Qualification Percent 1 7th 3.39 2 SSLC 15.25 3 PUC 5.08 4 BA 1.69 5 Bcom 6.78 6 Bsc 6.78 7 MA(Eco) 1.69 8 deploma(e,p.m.t.tex,) 22.03 9 ITI 1.69

10 BE 27.12 11 Mtech 5.08 12 MBBS 1.69 13 MCA 1.69 14 Total 100

Source: primary data

Source: Field Survey

However, they have the typical characteristic of small firms with lesser number of

workers. Table 6. 4 shows that almost half of firms have less than 10 workers.

Table 6.4 Total Employment positions of the firms.

Employment Total employment

0-2 2.5 3-5 2.5 6-10 42.5 11-15 15 16-20 2.5 21-25 10 26-30 5.00 Total 100

Source: Field Survey

316

These firms are mostly catering to the large firms within the state of Karnataka; as high

as 87% of these firms have subcontracting relation with a firm in Karnataka itself. Only

13% of the firms are marketing their product independently. Out of the 87% of the firms

35% are also catering to firms outside Karnataka and 2.5% are engaged in exports as well

(Table 6.5).

Table 6.5 Marketing of products.

Market

Large firm in the same state

Large firm in the another state

Export

87.5 35 2.5 Source: Filed Survey

6.4 Loan Amount and Rate of Interest: An Inverse Relation

One of the major objectives in this study is to examine the relation between the

borrowings and the rate of interest charged by the banks. It is interesting to note that

about 66% of the small borrowers (below Rs 50,000) pays interest rate of 17% or higher,

whereas the large borrowers pay comparatively lower interest rate. The category of

borrowers that borrows 40 lakhs or more pays interest rate below 12% (Table 6.6). This

may be because banks find the small borrowers comparatively more risky. Banks fixes

the rate of interest based on a number of criteria involving risky ness of a project, amount

of collateral and past records. Small borrowers may have been in a disadvantageous

position if they have started a new business with less collateral. The correlation between

loan amount and rate of interest indeed shows a negative value –0.370, which is

statistically significant at 5% level.

317

Table 6.6 Percentage of firms cross tabulated with respect to Loan amount and rate of

interest.

Bank rate of interest ( in Rs.) Total

Total loan amount ( in Rs ) 9 - 10 11 - 12 13 – 14 15 - 16 17 - 24

33.33 66.67 100

upto - 50000 10 66.67 11.54

50 50 100

50000 - 200000 10 20 7.69

33.33 16.67 33.33 16.67 100

200001 - 500000 20 20 40 33.33 23.08

50 50 100

500001 - 1000000 33.33 20 7.69

50 50 100

1000001 - 2000000 20 40 15.38

20 20 20 40 100

2000001 - 4000000 33.33 10 20 40 19.23

25 75 100

4000001 - above 33.33 30 15.38462

11.54 38.46 19.23 19.23 11.54 100

Total 100 100 100 100 100 100

Source: Field Survey

318

6.5 Problem of Bad Loans : Views of the Karnataka Firms Only about 3% of the firms have admitted to be defaulter in our sample. The rest of the

firms even though did not admit themselves to be defaulter have given possible reasons

for default by the small firms. It is interesting to note that about 80 % of the firms

consider excessive competition as a reason for low profit margin, which in turn affects

repayment capability adversely. But interestingly only 76% felt that opening up of the

market and competition from China is really not hampering small firms in our study area

(Table 6.7). Problem of non repayment of loan arising mainly due to non repayment of

dues by the large firms and large amount of rejection at a point of time. The former has

been well recognized by RBI and now small firms can take actions against large firms in

case of such defaults. However, as far as possible a small firm will not initiate such a

process due to its high dependence on the large firm. The latter may arise due to

increased quality concerns of the large firms. Our discussion with the bank officials also

reveals that diversion of funds is another major reason for default.

Table – 6.7 Reasons are default of the firms.

Reasons for default Yes a) Diversion of funds 53 b) Misunderstanding amongst partners. 42 c) Too much competition in the market 80 d) Huge quantity of finished product rejected 75 e) Competition from China/other country due to opening the market. 24 f) Large firms do not pay in time. 75 g) Too much borrowing. 51 h) Dependence on one or two large units 64 i) Marketing problem 53 Source: primary data Problem of willful default is accepted to be present by all firms, though expectedly no

one admitted to have adhered to such a practice. Diversion of funds has been the major

reason for willful default. This has also been stated during our interactions with the bank

officials. Entrepreneurs sometimes borrow in the name of a project but divert funds later

for purposes that can give more returns than an SSI unit.

319

Secondly, when a firm is running by a manger, the problem of principal –agent and in

turn monitoring exists. This sometime may result in default of loan as collateral belongs

to the firm owner. It has been observed by 7% of the respondents that high influence such

as political influence may result in willful default. This was also found to be true during

our discussions with the bank officials. Borrowers having strong political connections

often hold the view that they cannot be punished in any manner when they default.

Another reason cited by the firm owners is pure negligence. This as well has been found

to be true during our discussions with the bank officials. Even when business is running

well some borrowers do not repay loans as they feel that bank can never take any

stringent action. A few respondents also complained of collusive agreement between

bank officials and borrowers, which lead to such mal practices (Table 6.8).

Table 6.8 Opinions for the willful defaulted of firms.

Opinion Yes a) Misuse of funds 85 b) Technically not confident 8 c) Managerial problem 14 d) Subcontracting firm’s fault 7 e) High influence 7 f) Negligence 5 g) Bankers corruption 10

Source: primary data

It has been felt by the industry circle that taking prompt action and being more vigilant

can control willful default (Table 6.9). Right now it takes on an average about 10 years to

take control of the collateral/security. In the process plant and machinery depreciates

considerably, the defaulter also get sufficient return from the diverted funds and bank

becomes the major loser. SARFEISI act which ensures that a bank need not go through a

regular process of litigations through courts , may help to some extent in this regard. In

certain cases when default may be due to sheer negligence, serious advice from the bank

officials are necessary. To do this effectively bank officials may be given more powers to

take actions.

320

Table 6.9 Opinion from respondents to avoid the bank willful default

Avoid Opinion Yes

Seizing collateral 92.50 Checking 65 Self with athentification 15 Counseling 10

6.6 Problem of the Lending System

There are a number of problems faced by the borrowers from the SSI segment (Table 6.10). All the respondents considered procedures rather complicated, needing too many documentations. Number of times one needs to visit banks initially is as high as 10 to 20 times (Table 6.11).

Table 6.10. Current problems exist with the lending system.

Current Problems

Yes

Collateral problem 65 Excess document/complicated procedure 100 Not enough working capital loan 12.50 High rate of interest 27.50

This needs to be noted in the background of education levels of the respondents. If such highly educated respondents find the process complicated one can easily imagine the plight of the uneducated ones. Above 50% of the respondents who have

321

taken loan needed to visit the respective banks a minimum of 8 to 10 times before getting the loan and about 12% visited between 11 to 30 times. After availing the loan on an average in six months they need to visit 8 to 10 times; another 22% visited the bank between 10 to 30 times. Interestingly, 5% needed to visit the bank about 50 times of more; which amounts to about 10 visits per month. This really adds to the transaction costs to the borrowers.

Table 6.11 Number of times respondents need to visit banks.

Excess requirement of collateral is another major problem. Some banks demand three or four times’ higher value of security, personal guarantee and collaterals vis-a –vis the loan amount. A small entrepreneur does not usually possess assets and needs to refrain from borrowing. Though rate of interest have come down to some extent small borrowers usually pay around 2 to 3% higher than the prime lending rate (PLR). However, what is to note is that, all borrowers are charged considerable amount by the banks for handling their accounts in addition to the rate of interest charged. These comprise of cheque leaf charges, currier charges, charges for returned chaques, charges for bank officers visits and so on. Our estimate from our survey reveals that of such additional charges amounts to an additional rate of interest of 6%. Thus, if for example, the charged rate of interest is 13%, the actual resource goes to the bank at the rate of about 18 to 19%.

The 35 % of the firms that did not take loan from bank have used their own funds or from relatives to finance their business ventures. Maximum investment of these firms is Rs 10 lakhs. As far as reasons for not approaching banks are concerned 84% of them have noted excessive collateral requirement as crucial one. While complicated procedures have been noted by all, 15% also stated that working capital loan to be provided by the bank was so insufficient that they have decided not to approach banks.

Given this scenario naturally, maximum number of respondents voiced for hassle

free lending mechanism (table 6.12). 10% have indicated that a separate cell for SSI

lending may reduce some of the problems. 30% also feels that bank officials do not give

equal treatment to all borrowers. Economically better off or politically linked borrowers

No of times Initially go In six month 8- 10 52.5 30 11-30 12.5 22.5 31-50 0 7.5 51 Above 0 5 No applied 35 35

322

get priority. This can not only increase willful default but also induce good borrowers to

move to informal sector even when faced with high interest rate.

Table 6.12. Suggestions for lending system (banks)

Helpful changes

Percentage of

respondents Easy procedure 77.50 Separate cell 10 No corruption 30

6.7 Problem of Bad Loans: Views of the Firms from West Bengal The firms from West Bengal, maily located in the capital city Kolkata, echoed the same

views as that of their Karnataka counterpart. In our sample 100% of the respondents

have taken loan from the public sector banks. According to them intense competition in

the market is one reason for genuine default. Indeed, they have found the rate of return

to be much lower and comparatively therefore interest rate very high. However, no firm

has considered competition from outside firms such as that from China is responsible for

this. Wrong planning in the form of too much borrowing and , marketing are also

problems of genuine default (Table 6.14).

Table 6.14 Reason for default of the firm

Reason for default Percentage of firms say ‘yes’

a) Diversion of funds 23.8 b) Misunderstanding amongst partners 4.8 c) Too much competition in the market 23.8 d) Huge quantity of finished product rejected

e) Competition from china/other country

323

due to opening the market f) Large firms do not pay in time g) Too much borrowing 23.8 h) Dependence on one or two large units

i) Marketing problem 14.3 j) Any other Wastage problem 4.8 High rate of interest 4.8 Recession 4.8

As far as willful default is concerned respondents from Kolkata has not found bankers’

corruption as a possible reason. Rather respondents feel that misuse of funds and

political influence of the borrowers often lead to such willful default (table 6.15).

Table 6.15 Opinion about willful default

Opinion Yes

Misuse/Diversion of fund 38.1 b) Technical incompetence 14.3 c) Managerial problem 14.3 d) Unavoidable for small firms 9.5 e) High influence 19.0 f) Negligence 0 g) Bankers corruption 0

Source: Survey

Above 90% of the respondent firms voiced that prompt seizing of collateral is the most

effective way to reduce such intentional default. As far as the problem of the current

banking system is concerned one issue that came up again and again in the case of

Kolkata firms is the rate of interest (Table 6.16). This may be due to the fact that in

Kolkata there are large number of SSI firms that operate in lower segment of the market

where competition is intense and price realization is less. Hence the rate of interest that

the bank changes turns out to be high for them. Procedural complications remains a

problem for the borrowers of all regions.

Table 6.16 Currently any problem with the lending system Current problem Yes

324

b) Excess document 57.1 c) Complicated procedure 57.1 d) Discouraging behavior of other staff

4.8

e) Not enough working capital loan

4.8

f) High rate of interest 85.7

Source: Survey

6.8 Problem of Bad Loans: Views of the Firms from Kerala Unlike the other two groups , Kerala firms consider all possible reasons as equally

important in causing genuine default of the SSI firms. More importantly problem caused

by the large firms and competition from inside as well as aboard are highlighted during

our survey (Table 6.17).

Table 6.17 Reason for default Reason for default Respondent Percentage a) Diversion of funds 33.3 b) Misunderstanding amongst partners 22.2 c) Too much competition in the market 44.4 d) Huge quantity of finished product rejected 44.4 e) Competition from china/other country due to opening the market.

33.3

f) Large firms do not pay in time 44.4 g) Too much borrowing 33.3 h) Dependence on one or two large units 44.4 i) Marketing problem 66.7

Source: Survey As far as willful default is concerned, in all regions politically influential borrowers tend

to avoid repayment is a concern of all genuine borrowers. Corruption on the part of the

bank officials has also been highlighted in Kerala as well as in Kranataka (Table 6.18).

Table 6.18 Opinion about willful default Opinion Yes a) Miss use of fund 22.2 b) Managerial problem

55.6

c) High influence

22.2

d) Negligence

66.7

325

e) Bankers corruption

44.4

Source: Survey Prompt seizing of the collateral is the most effective way of reducing such default.

However, 11% of the respondent firms also felt that counseling by bank officials may

help reducing default.

Excessive documentation and complicated procedures appear to be the common problem

felt by all borrowers across regions (Table 6.19). However, high rate of interest is also

another major problem faced by the borrowers. It must be noted in this context that

usually these borrowers do not receive any concessional rate and pay about 2 to 3 percent

higher than the prime lending rate. More importantly a beginner often needs to pay

higher rate of interest and our analysis shows that smaller the loan size is (which often

implies smaller the size of the firm is) higher is the rate of interest.

Table 6.19 Current problems in lending system

Current Problem Percent

a) Collateral problem 22.7

b) Excess document 33.3

c) Complicated

procedure

44.4

d) High rate of

interest

66.7

Source: Survey

Suggestions to the policy makers are many and found to be similar across regions (Table

6.20).

Table 6.20 Suggestions to the policy makers Kind of Help Percent Power rate should be low 33.3 Pay in time

11.1

Provide the subsidy

11.1

326

Reduce the sale tax

44.4

Provide the employ ESI

22.2

Provide the land

33.3

Provide the power supply continuously

66.7

To train the projects guidance of the Bank manager

66.7

Provide the information about bank.

66.7

Source: Source In the infrastructure front power is the major concern. Secondly the firms across regions

want the bank manger to play a more active role rather than being just a fund provider.

As mentioned above most of these firm owners possess technical knowledge but lacks

management oriented knowledge of costing , pricing etc. This is the area where some

help from the banking sector is sought. They are also aware that some capacity building

for the bank officials may be necessary for them to provide effective support. A separate

cell in the bank for the sector may be useful in this regard. Right now the Small Industries

Development Bank is there to cater to this sector in a more involved manner. However,

SIDBI office and few and far between and firms cannot avail their help as and when

required. Thus active role needs to be played by the commercial banking sector.

6.9 Concluding Remarks During our intensive discussions with the bank officials it has been revealed that the

problem of NPA is reducing over time for the SSI sector. On the other hand it is

becoming more prevent in the personal loan segment. From our secondary data analysis

we have also seen that banks’ credit towards the SSI sector is also declining. Our

interviews with the SSI entrepreneurs reveal that non-repayment is often genuine, that

is, due to failure in the business. In the SSI segment competition is much more intense

which results in stiff price competition. While some segments do face competition from

cheap Chinese products, our respondents did not feel that globalization has made the

situation worse. The small firm owners during our survey have suggested several

327

initiatives from the policy makers which may be helpful for the sector. One of the major

problem the sector faces is the quality power supply. Such infrastructure bottlenecks need

to be handled to improve productivity.

Globalization indeed has helped some of the large firms to export and in turn increased

subcontracting business for the small firms; the growing automobile sector is a case in

point here. Exporting firms or multinationals however are quite quality conscious and

not meeting their requirements and resulting large scale rejection of products often put

small firms in the verge of bankruptcy. Non-repayment of dues by the large firms on time

also is a serious concern, which has been well recognized in the literature. These are

some of the genuine reasons for business failure and resulting default. Some of these can

be avoided through proper planning. Bank as a lender can act as a partner of an SSI unit

than as a policeman. For example, many SSI units we interviewed admitted that they

have technological knowledge but lack expertise on management aspects. Thus costing

and pricing strategies are adhoc and faulty. Neither do they have sufficient resources to

engage professionals. The firm owners’ felt that training of bank officials is necessary

for them to impart knowledge and act as a partner. In this regard State Bank of India,

stressed asset and rehabilitation cell have been advising some of the defaulters on these

aspects. More such efforts should come from the banks.

The case of willful default however, needs to be taken rather seriously. Currently, banks

do not identify any defaulter as a willful defaulter. Thus there is no difference in terms of

actions taken by the bank between a genuine and a willful defaulter. This approach

should change. Making a confidential list of willful defaulters may deter these borrowers

to engage in such activities. Right now a defaulter can indeed go to another bank for a

fresh loan and this often goes unnoticed. More vigilance and prompt action is the need of

the hour rather than avoiding the small borrowers.

328

CHAPTER 6

Problem of Loan Repayment: Views of the SSI Units

6.1 Introduction Much has been written about the problems of the SSI sector in the Indian context. The

growth of small scale industry in India, to a large extent is induced by the lack of

alternative employment opportunities and promotional policies adopted by government

(Desai, 1983)47. Due to lack of entrepreneurial attitude and proper training a large

number of them have met with untimely closure. Some of the major problems can be

identified as follows.

• Financial Constraints: Though there are a number of efforts to provide finances

to the SSI units, the most needy ones do not get proper information about

different schemes and often depend on the informal credit market for finance.

Coming up with an appropriate proposal also becomes difficult for such small

entrepreneurs.

• Access to raw materials is another problem faced by these units.

• One of the major handicaps of the small-scale sector has been the absence of

improved technology, which alone can ensure quality and higher productivity.

Technology is the most essential factor to remain competitive in a global market.

Lack of information again plays a critical role in the choice of technology.

• Marketing remains the major stumbling block for the growth of SSI sector.

Ignorance of potential markets, in particular, unfamiliarity with export activities

contributes to this problem. Poor designing and finish also often makes the

product not salable in the international market.

47 Desai. V, 1983, Problems and Prospects of Small Scale Industries in India, Himalaya

Publishing House, Bombay.

329

• As far as production methods are concerned there is often faulty planning and

inadequate appraisal of projects (Desai, 1983). Most often no proper viability

studies, technical or economic, are carried out.

Policy makers recognized these shortcomings and created a number of offices to handle

the problems. However, proper implementation has never been accomplished.

Consequently sickness remained a major problem for this sector. To understand the

problems of the sector in general and concerning credit facilities in particular in a

liberalized regime, a survey has been taken up. Though the survey has been conducted in

three states of India viz., Karnataka, Kerala and West Bengal.

6.2 Sampling Technique

It is well recognized that industry sector is not forthcoming in providing information.

This problem has noted down by many authors (see Deshpande et al, 2004)48. In this

background we are forced to adopt snowball sampling technique. Industry Association

gave contacts of the firms and requested them to cooperate with the survey. The firms

that agreed for a discussion were later interviewed using structured questionnaires that

were personally canvassed. The sample firms are from Kerala, Karnataka and West

Bengal and the sample sizes from these locations are respectively 50, 100 and 50.

6.3 Characteristics of the Sampled Firms

One of the important findings of the survey is that 35% of the manufacturing firms in our

sample are not availing loan from the institutional sources. These firms reported to

manage their investments from their own (or borrowing from relatives) resources. Out of

the rest 68% only 2% have availed loan from private banks. Thus dependency on public

sector banks remains prevalent.

48 Deshpande, Lalit et al, 2004, Liberalization and Labour: Labour Flexibility in Indian Manufacturing, Institute of Human Development, New Delhi.

330

While examining the investment in plant and machinery one observes that most of the

firms included in the survey falls under the Government of India definition of SSI. Only

about 2% of the firms have investment above Rs 1 crore. However as far as commercial

banks are concerned, for credit purposes they combine small and medium firms together

table 6.1).

Table – 6.1.Total investment

Investment Percent10000 - 100000 5 100000 - 500000 17.5 500001 - 1000000 37.5 1000001 - 2500000 15 2500001 - 5000000 12.5 5000001 - 1 crore 10 10000001 - Above 2.5 Total 100

Average turnover is between 30 lakhs to 50 lakhs. Thus these are comparatively large

sized firms within the SSI category (table 6.2).

Table 6.2. Total turnovers of the firms.

Turnover in Rs. This year Last year On an

average 300000 - 1000000 8 (20.00) 9 (22.50) 8 (20.00) 1000001 - 3000000 11 (27.50) 12 (30.00) 11 (27.50) 3000001 - 5000000 8 (20.00) 8 (20.00) 10 (25.00) 5000001 - 7000000 5 (12.50) 2 (5.00) 2 (5.00) 7000001 - 1 crore 3 (7.50) 3 (7.50) 3 (7.50) 10000001 - Above 5 (12.50) 5 (12.50) 6 (15.00)

Total 40 (100) 40 (100) 40 (100)

These firms are managed by qualifies entrepreneurs as can be seen from table 6.3. Above

50% of the firms owners are technical degree holders.

331

Table – 6.3 Educational qualifications of the firm manager and Owners.

Qualification Percent 1 7th 3.39 2 SSLC 15.25 3 PUC 5.08 4 BA 1.69 5 Bcom 6.78 6 Bsc 6.78 7 MA(Eco) 1.69 8 deploma(e,p.m.t.tex,) 22.03 9 ITI 1.69

10 BE 27.12 11 Mtech 5.08 12 MBBS 1.69 13 MCA 1.69 14 Total 100

Source: primary data

Source: Field Survey

However, they have the typical characteristic of small firms with lesser number of

workers. Table 6. 4 shows that almost half of firms have less than 10 workers.

Table 6.4 Total Employment positions of the firms.

Employment Total employment

0-2 2.5 3-5 2.5 6-10 42.5 11-15 15 16-20 2.5 21-25 10 26-30 5.00 Total 100

Source: Field Survey

332

These firms are mostly catering to the large firms within the state of Karnataka; as high

as 87% of these firms have subcontracting relation with a firm in Karnataka itself. Only

13% of the firms are marketing their product independently. Out of the 87% of the firms

35% are also catering to firms outside Karnataka and 2.5% are engaged in exports as well

(Table 6.5).

Table 6.5 Marketing of products.

Market

Large firm in the same state

Large firm in the another state

Export

87.5 35 2.5 Source: Filed Survey

6.4 Loan Amount and Rate of Interest: An Inverse Relation

One of the major objectives in this study is to examine the relation between the

borrowings and the rate of interest charged by the banks. It is interesting to note that

about 66% of the small borrowers (below Rs 50,000) pays interest rate of 17% or higher,

whereas the large borrowers pay comparatively lower interest rate. The category of

borrowers that borrows 40 lakhs or more pays interest rate below 12% (Table 6.6). This

may be because banks find the small borrowers comparatively more risky. Banks fixes

the rate of interest based on a number of criteria involving risky ness of a project, amount

of collateral and past records. Small borrowers may have been in a disadvantageous

position if they have started a new business with less collateral. The correlation between

loan amount and rate of interest indeed shows a negative value –0.370, which is

statistically significant at 5% level.

333

Table 6.6 Percentage of firms cross tabulated with respect to Loan amount and rate of

interest.

Bank rate of interest ( in Rs.) Total

Total loan amount ( in Rs ) 9 - 10 11 - 12 13 – 14 15 - 16 17 - 24

33.33 66.67 100

upto - 50000 10 66.67 11.54

50 50 100

50000 - 200000 10 20 7.69

33.33 16.67 33.33 16.67 100

200001 - 500000 20 20 40 33.33 23.08

50 50 100

500001 - 1000000 33.33 20 7.69

50 50 100

1000001 - 2000000 20 40 15.38

20 20 20 40 100

2000001 - 4000000 33.33 10 20 40 19.23

25 75 100

4000001 - above 33.33 30 15.38462

11.54 38.46 19.23 19.23 11.54 100

Total 100 100 100 100 100 100

Source: Field Survey

334

6.5 Problem of Bad Loans : Views of the Karnataka Firms Only about 3% of the firms have admitted to be defaulter in our sample. The rest of the

firms even though did not admit themselves to be defaulter have given possible reasons

for default by the small firms. It is interesting to note that about 80 % of the firms

consider excessive competition as a reason for low profit margin, which in turn affects

repayment capability adversely. But interestingly only 76% felt that opening up of the

market and competition from China is really not hampering small firms in our study area

(Table 6.7). Problem of non repayment of loan arising mainly due to non repayment of

dues by the large firms and large amount of rejection at a point of time. The former has

been well recognized by RBI and now small firms can take actions against large firms in

case of such defaults. However, as far as possible a small firm will not initiate such a

process due to its high dependence on the large firm. The latter may arise due to

increased quality concerns of the large firms. Our discussion with the bank officials also

reveals that diversion of funds is another major reason for default.

Table – 6.7 Reasons are default of the firms.

Reasons for default Yes a) Diversion of funds 53 b) Misunderstanding amongst partners. 42 c) Too much competition in the market 80 d) Huge quantity of finished product rejected 75 e) Competition from China/other country due to opening the market. 24 f) Large firms do not pay in time. 75 g) Too much borrowing. 51 h) Dependence on one or two large units 64 i) Marketing problem 53 Source: primary data Problem of willful default is accepted to be present by all firms, though expectedly no

one admitted to have adhered to such a practice. Diversion of funds has been the major

reason for willful default. This has also been stated during our interactions with the bank

officials. Entrepreneurs sometimes borrow in the name of a project but divert funds later

for purposes that can give more returns than an SSI unit.

335

Secondly, when a firm is running by a manger, the problem of principal –agent and in

turn monitoring exists. This sometime may result in default of loan as collateral belongs

to the firm owner. It has been observed by 7% of the respondents that high influence such

as political influence may result in willful default. This was also found to be true during

our discussions with the bank officials. Borrowers having strong political connections

often hold the view that they cannot be punished in any manner when they default.

Another reason cited by the firm owners is pure negligence. This as well has been found

to be true during our discussions with the bank officials. Even when business is running

well some borrowers do not repay loans as they feel that bank can never take any

stringent action. A few respondents also complained of collusive agreement between

bank officials and borrowers, which lead to such mal practices (Table 6.8).

Table 6.8 Opinions for the willful defaulted of firms.

Opinion Yes a) Misuse of funds 85 b) Technically not confident 8 c) Managerial problem 14 d) Subcontracting firm’s fault 7 e) High influence 7 f) Negligence 5 g) Bankers corruption 10

Source: primary data

It has been felt by the industry circle that taking prompt action and being more vigilant

can control willful default (Table 6.9). Right now it takes on an average about 10 years to

take control of the collateral/security. In the process plant and machinery depreciates

considerably, the defaulter also get sufficient return from the diverted funds and bank

becomes the major loser. SARFEISI act which ensures that a bank need not go through a

regular process of litigations through courts , may help to some extent in this regard. In

certain cases when default may be due to sheer negligence, serious advice from the bank

officials are necessary. To do this effectively bank officials may be given more powers to

take actions.

336

Table 6.9 Opinion from respondents to avoid the bank willful default

Avoid Opinion Yes

Seizing collateral 92.50 Checking 65 Self with athentification 15 Counseling 10

6.6 Problem of the Lending System

There are a number of problems faced by the borrowers from the SSI segment (Table 6.10). All the respondents considered procedures rather complicated, needing too many documentations. Number of times one needs to visit banks initially is as high as 10 to 20 times (Table 6.11).

Table 6.10. Current problems exist with the lending system.

Current Problems

Yes

Collateral problem 65 Excess document/complicated procedure 100 Not enough working capital loan 12.50 High rate of interest 27.50

This needs to be noted in the background of education levels of the respondents. If such highly educated respondents find the process complicated one can easily imagine the plight of the uneducated ones. Above 50% of the respondents who have

337

taken loan needed to visit the respective banks a minimum of 8 to 10 times before getting the loan and about 12% visited between 11 to 30 times. After availing the loan on an average in six months they need to visit 8 to 10 times; another 22% visited the bank between 10 to 30 times. Interestingly, 5% needed to visit the bank about 50 times of more; which amounts to about 10 visits per month. This really adds to the transaction costs to the borrowers.

Table 6.11 Number of times respondents need to visit banks.

Excess requirement of collateral is another major problem. Some banks demand three or four times’ higher value of security, personal guarantee and collaterals vis-a –vis the loan amount. A small entrepreneur does not usually possess assets and needs to refrain from borrowing. Though rate of interest have come down to some extent small borrowers usually pay around 2 to 3% higher than the prime lending rate (PLR). However, what is to note is that, all borrowers are charged considerable amount by the banks for handling their accounts in addition to the rate of interest charged. These comprise of cheque leaf charges, currier charges, charges for returned chaques, charges for bank officers visits and so on. Our estimate from our survey reveals that of such additional charges amounts to an additional rate of interest of 6%. Thus, if for example, the charged rate of interest is 13%, the actual resource goes to the bank at the rate of about 18 to 19%.

The 35 % of the firms that did not take loan from bank have used their own funds or from relatives to finance their business ventures. Maximum investment of these firms is Rs 10 lakhs. As far as reasons for not approaching banks are concerned 84% of them have noted excessive collateral requirement as crucial one. While complicated procedures have been noted by all, 15% also stated that working capital loan to be provided by the bank was so insufficient that they have decided not to approach banks.

Given this scenario naturally, maximum number of respondents voiced for hassle

free lending mechanism (table 6.12). 10% have indicated that a separate cell for SSI

lending may reduce some of the problems. 30% also feels that bank officials do not give

equal treatment to all borrowers. Economically better off or politically linked borrowers

No of times Initially go In six month 8- 10 52.5 30 11-30 12.5 22.5 31-50 0 7.5 51 Above 0 5 No applied 35 35

338

get priority. This can not only increase willful default but also induce good borrowers to

move to informal sector even when faced with high interest rate.

Table 6.12. Suggestions for lending system (banks)

Helpful changes

Percentage of

respondents Easy procedure 77.50 Separate cell 10 No corruption 30

6.7 Problem of Bad Loans: Views of the Firms from West Bengal The firms from West Bengal, maily located in the capital city Kolkata, echoed the same

views as that of their Karnataka counterpart. In our sample 100% of the respondents

have taken loan from the public sector banks. According to them intense competition in

the market is one reason for genuine default. Indeed, they have found the rate of return

to be much lower and comparatively therefore interest rate very high. However, no firm

has considered competition from outside firms such as that from China is responsible for

this. Wrong planning in the form of too much borrowing and , marketing are also

problems of genuine default (Table 6.14).

Table 6.14 Reason for default of the firm

Reason for default Percentage of firms say ‘yes’

a) Diversion of funds 23.8 b) Misunderstanding amongst partners 4.8 c) Too much competition in the market 23.8 d) Huge quantity of finished product rejected

e) Competition from china/other country

339

due to opening the market f) Large firms do not pay in time g) Too much borrowing 23.8 h) Dependence on one or two large units

i) Marketing problem 14.3 j) Any other Wastage problem 4.8 High rate of interest 4.8 Recession 4.8

As far as willful default is concerned respondents from Kolkata has not found bankers’

corruption as a possible reason. Rather respondents feel that misuse of funds and

political influence of the borrowers often lead to such willful default (table 6.15).

Table 6.15 Opinion about willful default

Opinion Yes

Misuse/Diversion of fund 38.1 b) Technical incompetence 14.3 c) Managerial problem 14.3 d) Unavoidable for small firms 9.5 e) High influence 19.0 f) Negligence 0 g) Bankers corruption 0

Source: Survey

Above 90% of the respondent firms voiced that prompt seizing of collateral is the most

effective way to reduce such intentional default. As far as the problem of the current

banking system is concerned one issue that came up again and again in the case of

Kolkata firms is the rate of interest (Table 6.16). This may be due to the fact that in

Kolkata there are large number of SSI firms that operate in lower segment of the market

where competition is intense and price realization is less. Hence the rate of interest that

the bank changes turns out to be high for them. Procedural complications remains a

problem for the borrowers of all regions.

Table 6.16 Currently any problem with the lending system Current problem Yes

340

b) Excess document 57.1 c) Complicated procedure 57.1 d) Discouraging behavior of other staff

4.8

e) Not enough working capital loan

4.8

f) High rate of interest 85.7

Source: Survey

6.8 Problem of Bad Loans: Views of the Firms from Kerala Unlike the other two groups , Kerala firms consider all possible reasons as equally

important in causing genuine default of the SSI firms. More importantly problem caused

by the large firms and competition from inside as well as aboard are highlighted during

our survey (Table 6.17).

Table 6.17 Reason for default Reason for default Respondent Percentage a) Diversion of funds 33.3 b) Misunderstanding amongst partners 22.2 c) Too much competition in the market 44.4 d) Huge quantity of finished product rejected 44.4 e) Competition from china/other country due to opening the market.

33.3

f) Large firms do not pay in time 44.4 g) Too much borrowing 33.3 h) Dependence on one or two large units 44.4 i) Marketing problem 66.7

Source: Survey As far as willful default is concerned, in all regions politically influential borrowers tend

to avoid repayment is a concern of all genuine borrowers. Corruption on the part of the

bank officials has also been highlighted in Kerala as well as in Kranataka (Table 6.18).

Table 6.18 Opinion about willful default Opinion Yes a) Miss use of fund 22.2 b) Managerial problem

55.6

c) High influence

22.2

d) Negligence

66.7

341

e) Bankers corruption

44.4

Source: Survey Prompt seizing of the collateral is the most effective way of reducing such default.

However, 11% of the respondent firms also felt that counseling by bank officials may

help reducing default.

Excessive documentation and complicated procedures appear to be the common problem

felt by all borrowers across regions (Table 6.19). However, high rate of interest is also

another major problem faced by the borrowers. It must be noted in this context that

usually these borrowers do not receive any concessional rate and pay about 2 to 3 percent

higher than the prime lending rate. More importantly a beginner often needs to pay

higher rate of interest and our analysis shows that smaller the loan size is (which often

implies smaller the size of the firm is) higher is the rate of interest.

Table 6.19 Current problems in lending system

Current Problem Percent

a) Collateral problem 22.7

b) Excess document 33.3

c) Complicated

procedure

44.4

d) High rate of

interest

66.7

Source: Survey

Suggestions to the policy makers are many and found to be similar across regions (Table

6.20).

Table 6.20 Suggestions to the policy makers Kind of Help Percent Power rate should be low 33.3 Pay in time

11.1

Provide the subsidy

11.1

342

Reduce the sale tax

44.4

Provide the employ ESI

22.2

Provide the land

33.3

Provide the power supply continuously

66.7

To train the projects guidance of the Bank manager

66.7

Provide the information about bank.

66.7

Source: Source In the infrastructure front power is the major concern. Secondly the firms across regions

want the bank manger to play a more active role rather than being just a fund provider.

As mentioned above most of these firm owners possess technical knowledge but lacks

management oriented knowledge of costing , pricing etc. This is the area where some

help from the banking sector is sought. They are also aware that some capacity building

for the bank officials may be necessary for them to provide effective support. A separate

cell in the bank for the sector may be useful in this regard. Right now the Small Industries

Development Bank is there to cater to this sector in a more involved manner. However,

SIDBI office and few and far between and firms cannot avail their help as and when

required. Thus active role needs to be played by the commercial banking sector.

6.9 Concluding Remarks During our intensive discussions with the bank officials it has been revealed that the

problem of NPA is reducing over time for the SSI sector. On the other hand it is

becoming more prevent in the personal loan segment. From our secondary data analysis

we have also seen that banks’ credit towards the SSI sector is also declining. Our

interviews with the SSI entrepreneurs reveal that non-repayment is often genuine, that

is, due to failure in the business. In the SSI segment competition is much more intense

which results in stiff price competition. While some segments do face competition from

cheap Chinese products, our respondents did not feel that globalization has made the

situation worse. The small firm owners during our survey have suggested several

343

initiatives from the policy makers which may be helpful for the sector. One of the major

problem the sector faces is the quality power supply. Such infrastructure bottlenecks need

to be handled to improve productivity.

Globalization indeed has helped some of the large firms to export and in turn increased

subcontracting business for the small firms; the growing automobile sector is a case in

point here. Exporting firms or multinationals however are quite quality conscious and

not meeting their requirements and resulting large scale rejection of products often put

small firms in the verge of bankruptcy. Non-repayment of dues by the large firms on time

also is a serious concern, which has been well recognized in the literature. These are

some of the genuine reasons for business failure and resulting default. Some of these can

be avoided through proper planning. Bank as a lender can act as a partner of an SSI unit

than as a policeman. For example, many SSI units we interviewed admitted that they

have technological knowledge but lack expertise on management aspects. Thus costing

and pricing strategies are adhoc and faulty. Neither do they have sufficient resources to

engage professionals. The firm owners’ felt that training of bank officials is necessary

for them to impart knowledge and act as a partner. In this regard State Bank of India,

stressed asset and rehabilitation cell have been advising some of the defaulters on these

aspects. More such efforts should come from the banks.

The case of willful default however, needs to be taken rather seriously. Currently, banks

do not identify any defaulter as a willful defaulter. Thus there is no difference in terms of

actions taken by the bank between a genuine and a willful defaulter. This approach

should change. Making a confidential list of willful defaulters may deter these borrowers

to engage in such activities. Right now a defaulter can indeed go to another bank for a

fresh loan and this often goes unnoticed. More vigilance and prompt action is the need of

the hour rather than avoiding the small borrowers.

344

CHAPTER 7

Non-performing Asset from the Perspectives of Commercial Banks

7.1 Introduction In the previous chapter we have discussed the views of one set of stakeholders viz.,

the firm owners. The survey of the small firm entrepreneurs has no doubt brought out

important issues concerning non-repayment of loan and willful default. It has also gone to

the root causes of genuine default, which need to be addressed not only by the banks but

also by the policy makers at large. To get a balanced view of any situation looking at both

sides is necessary. In this case the other stakeholder is the ‘commercial bank’ that

advances the resources mobilized by them to the small firm entrepreneurs. Our

discussions with the bank officials reveal that bad loans from the SSI sectors are indeed

in decline. This is mainly due to the pressure on the bank officials to reduce over all NPA

levels.

7.2 Approach to Information In order to understand the views of the banks, we have had discussions with several bank

officials who are in charge of the credit section. In particular, we have covered State

Bank of India, State Bank of Travancore, Syndicate Bank, Canara Bank and others. We

have also collected information about SSI accounts both pertaining to NPA as well as

non-NPA accounts from the banks. For comparison purposes we have collected

information about personal loan accounts as well. Information on individual accounts is a

confidential matter and keeping this in mind banks has not revealed to us the identity of

the borrowers. Data collection is organized as follows. To collect data from a particular

bank, permission from the head office has been sought. After acquiring permissions,

which indeed took considerable time, head offices identified certain branches and we

345

have personally visited those branches for data collection49. Information on a sample of

200 accounts has been collected from different banks taken together. Another problem

has been encountered with respect to the manner in which data are preserved by the

banks. Different banks or even branches of a bank do not keep data in a uniform manner.

Further, Companies though provide data on their turnover, profit and other financial

variables to the banks a careful scrutiny revealed that these figures are not reliable. The

reliable figures one can get are on loan amount, rate of interest, value of

security/collateral, type of account (term loan or working capital loan), activity of the

unit.

7.3 An Over all Picture from State Bank of Travancore (SBT)

A disaggregated picture of NPA for small and medium enterprises at the bank level is not

generally available. However, to understand the problem better we spent considerable

time at the head office of SBT to get a sector-wise desegregation of bad loans. SBT has

about 750 branches all over India but its main concentration is in the state of Kerala with

500 branches. Table 7.1 shows the extent of bad loans across different sectors.

Table 7.1 Industry-wise classifications of sick units and NPA

Industry No. Of Units % Of total sick

units

Outstanding (Rs

crores)

% To total

outstanding

Engineering 105 6.46 6.12 8.05

Electrical 72 4.43 3.53 4.64

Textile 75 4.61 5.62 7.39

49 Since these files are confidential they cannot be taken out or photocopied; bank officials also burdened with many responsibilities do no like to entertain such additional work during office hours. This made the data collection a rather slow process. While this process is currently going on, the present chapter makes a preliminary analysis of 200 SSI accounts.

346

Paper and paper

products

18 1.11 1 1.31

Rubber and rubber

products

35 2.15 5.12 6.74

Chemical, dyes,

paints

30 1.85 1.75 2.3

Metal and Metal

products

83 5.1 5.27 6.94

Vegetable oils and

vanaspati

32 1.97 0.62 0.82

Food processing

and producing

60 3.69 1.67 2.2

Plastics 3 0.18 0.8 1.05

Bricks 204 12.55 1 1.32

Coir 55 3.38 1.49 1.96

Bamboo 10 0.62 0.53 0.7

Wood 69 4.24 4.01 5.28

Readymade

garments

43 2.64 1.62 2.13

Miscellaneous 669 41.15 33.26 43.77

Total 1626 100 75.99 100

Source: State Bank of Travancore, Head Office

From the above information we observe that the miscellaneous category comprising

various different manufacturing items such as bus seat cover, glass making and so on has

347

the highest share in total number of sick units which also gives NPA accounts from SME

segment. Industry-wise brick making industry appears to have large number of sick units

even though amount involved is not comparatively higher. The reason for this industry to

do poorly is the environmental norms imposed on it. Due to digging of soil the wells in

the vicinity of the brick making unit go dry. Due to this reason several restrictions have

been imposed on this segment that effected their normal functioning.

Thus the discussions on SSI segment in the previous Chapter and the record of the banks

indicate that reasons of sickness are many. NPA accounts are mainly due to prevalence of

such sick units.

7.4 Important Observations In the case of SME loan, the problem of genuine default is more common in nature

compared to willful default. A specific example will reveal the situation. In SBT main

branch office, out of a total of 130 SME accounts, 18 are NPA (i.e., 13%). Out of these

18, the concerned officials feel that maximum 5 (28%) may be willful50. The genuine

reasons of default are more or less same as the reasons for sickness of the small units.

According to the assessment of the banks a small entrepreneur does not have the

necessary skill to arrive at proper estimates of costs, prices etc. In order to acquire a loan,

they usually hire an accountant who comes up with figures in the proposal, which are

later found to be unrealistic. Further, once an account becomes somewhat irregular banks

take prompt action due to the pressure on them to reduce NPA levels. They curtail their

funding to the unit concerned, which in turn makes the situation worse for the small firm.

In the case of personal loan however, default is much higher. Personal loan is given

against salary certificates and collateral security is not necessary to avail them. Corporate

employees in the big cities often avail such loans and shift cities for new jobs. It then

becomes difficult for banks to trace them and retrieve loans. We observed several such

50 Due to the lack of concrete evidences, officially no account has been marked as willfully defaulted.

348

cases during our survey. Loans to purchase durables also tend to become NPA more

often.

As far as willful default is concerned no bank usually designates a borrower as willful

defaulter. However, bank officials from their field visits can assess the cases. From our

discussions with the bank officials it is observed that one can classify them into 3

categories: (1) People with political influence (2) Negligent borrowers who feel (for some

reason or other) that they need not repay. This usually happens when they get loan under

certain government scheme (during our survey this is observed in the case of SC/ST

borrowers, women getting loan under special schemes)), (3) Borrowers who divert funds

for other purposes. Some borrowers rightly understand that they can earn higher return by

diverting their funds to other businesses such as real estate. By the time bank confiscate

their security, which usually takes considerable time (8 to 10 years), they would earn

much higher return than the lost security.

During our survey one category of willful defaulters have been found which needs to be

dealt with strictly. These borrowers usually get loan under a well-planned government

scheme introduced to help a weaker section such as women or other backward sections.

There is a well organized intermediary network prevails which in collusion with the

potential borrowers (such as poor women) prepare documents to avail loan for certain

income generating purposes as per allowed by the policy. However, rather than investing

it in the stated purpose both parties share the loan amount and in turn mis-utilize it. A

banks has even given loan in a draft form in the name of the party that was supposed to

supply the capital good to the borrower. This intermediary net-work is so effective that it

establishes collusive agreements with the machine suppliers as well.

While only a small proportion of firms are willful defaulter, from a bank’s perspective

this category assumes importance. While reducing the sickness of the SSI sector need not

necessarily fall under the purview of the banking sector, combating the problem of willful

349

default is a concern of the banking sector. Observing this RBI has of late came up with a

number of measures to fight this problem.

7.4 Measures to Contain Wilful Default51 In order to disseminate information about wilful default, under the directive of RBI a

scheme was framed under which the banks and notified All India Financial Institutions

were required to submit to RBI the details of the willful defaulters. Wilful default broadly

covered the following:

a) Deliberate non-payment of the dues despite adequate cash flow and good net

worth;

b) Siphoning off of funds to the detriment of the defaulting unit;

c) Assets financed either not been purchased or been sold and proceeds have

misutilised;

d) Misrepresentation / falsification of records;

e) Disposal / removal of securities without bank's knowledge;

f) Fraudulent transactions by the borrower.

The above scheme came into force with effect from 1st April, 1999. Accordingly, banks

and FIs started reporting certain cases of wilful defaults, which occurred or were detected

after 31st March, 1999 on a quarterly basis. It covered all non-performing borrowers

accounts with outstanding aggregating Rs.25 lakhs and above identified as willful default

by a Committee of higher functionaries headed by the Executive Director and consisting

of two GMs/DGMs. Banks/FIs were advised that they should examine all cases of wilful

defaults of Rs 1.00 crore and above for filing of suits and also consider criminal action

51 Master Circular on Willful Defaulter, RBI/2006-07/35 DBOD No.DL.BC.19 /20.16.003/2006-07

350

wherever instances of cheating/fraud by the defaulting borrowers were detected. In case

of consortium/multiple lending, banks and FIs were advised that they report wilful

defaults to other participating/financing banks also. Cases of wilful defaults at overseas

branches were required be reported if such disclosure is permitted under the laws of the

host country.

The above scheme was in addition to the Scheme of ‘Disclosure of Information on

Defaulting Borrowers of banks and FIs’ introduced in April 1994; vide RBI Circular

DBOD.No.BC/CIS/47/20.16.002/94 dated 23 April 1994.

Guidelines issued on wilful defaulters (May 30, 2002)

Considering the concerns expressed over the persistence of wilful default in the financial

system in the 8th Report of the Parliament's Standing Committee on Finance on Financial

Institutions, the Reserve Bank of India, in consultation with the Government of India,

constituted in May 2001 a Working Group on Wilful Defaulters (WGWD) under the

Chairmanship of Shri S. S. Kohli, the then Chairman of the Indian Banks' Association,

for examining some of the recommendations of the Committee. The Group submitted its

report in November 2001. An In-House Working Group constituted by the Reserve Bank

further examined the recommendations of the WGWD. Accordingly, the banks/FIs were

advised on May 30, 2002 for implementation, with immediate effect.

Definition of wilful default

As per the definition of RBI a "wilful default" would be deemed to have occurred

if any of the following events is noted: -

(a) The unit has defaulted in meeting its payment / repayment obligations to the

lender even when it has the capacity to honour the said obligations.

(b) The unit has defaulted in meeting its payment / repayment obligations to the

lender and has not utilised the finance from the lender for the specific purposes

for which finance was availed of but has diverted the funds for other purposes.

(c) The unit has defaulted in meeting its payment / repayment obligations to the

lender and has siphoned off the funds so that the funds have not been utilised for

351

the specific purpose for which finance was availed of, nor are the funds available

with the unit in the form of other assets."

Diversion and siphoning of funds

The terms “diversion of funds” and “siphoning of funds” should construe to mean the

following:-

Diversion of funds, would be construed to include any one of the undernoted

occurrences:

(a) utilisation of short-term working capital funds for long-term purposes not in

conformity with the terms of sanction;

(b) deploying borrowed funds for purposes / activities or creation of assets other than

those for which the loan was sanctioned;

(c) transferring funds to the subsidiaries / Group companies or other corporates by

whatever modalities;

(d) routing of funds through any bank other than the lender bank or members of

consortium without prior permission of the lender;

(e) investment in other companies by way of acquiring equities / debt instruments without

approval of lenders;

(f) Shortfall in deployment of funds vis-à-vis the amounts disbursed / drawn and the

difference not being accounted for.

Siphoning of funds should be construed to occur if any funds borrowed from banks / FIs

are utilised for purposes un-related to the operations of the borrower, to the detriment of

the financial health of the entity or of the lender. The decision as to whether a particular

instance amounts to siphoning of funds would have to be a judgement of the lenders

based on objective facts and circumstances of the case.

352

The identification of the wilful default should be made keeping in view the track record

of the borrowers and should not be decided on the basis of isolated

transactions/incidents. The default to be categorised as wilful must be intentional,

deliberate and calculated.

End-use of Funds

In cases of project financing, the banks / FIs seek to ensure end-use of funds by, inter

alia, obtaining certification from the Chartered Accountants for the purpose. In case of

short-term corporate / clean loans, such an approach ought to be supplemented by 'due

diligence' on the part of lenders themselves, and to the extent possible, such loans should

be limited to only those borrowers whose integrity and reliability are above board. The

banks and FIs, therefore, should not depend entirely on the certificates issued by the

Chartered Accountants but strengthen their internal controls and the credit risk

management system to enhance the quality of their loan portfolio. Needless to say,

ensuring end-use of funds by the banks and the FIs should form a part of their loan policy

document for which appropriate measures should be put in place. The following are some

of the illustrative measures that could be taken by the lenders for monitoring and ensuring

end-use of funds:

(a) Meaningful scrutiny of quarterly progress reports / operating statements /

balance sheets of the borrowers;

(b) Regular inspection of borrowers’ assets charged to the lenders as security;

(c) Periodical scrutiny of borrowers’ books of accounts and the no-lien accounts

maintained with other banks;

(d) Periodical visits to the assisted units;

(e) System of periodical stock audit, in case of working capital finance;

353

(f) Periodical comprehensive management audit of the ‘Credit’ function of the

lenders, so as to identify the systemic-weaknesses in the credit-

administration.

(It may be kept in mind that this list of measures is only illustrative and by no means

exhaustive.)

Penal measures

In order to prevent the access to the capital markets by the wilful defaulters, a copy of the

list of wilful defaulters (non-suit filed accounts) and (suit-filed accounts) are forwarded

to SEBI by RBI and Credit Information Bureau (India) Ltd. (CIBIL) respectively.

The following measures initiated by the banks and FIs against the wilful defaulters

identified as per the definition indicated at paragraph 2.1 above:

a) No additional facilities should be granted by any bank / FI to the listed wilful

defaulters. In addition, the entrepreneurs / promoters of companies where banks /

FIs have identified siphoning / diversion of funds, misrepresentation, falsification

of accounts and fraudulent transactions should be debarred from institutional

finance from the scheduled commercial banks, Development Financial

Institutions, Government owned NBFCs, investment institutions etc. for floating

new ventures for a period of 5 years from the date the name of the wilful defaulter

is published in the list of wilful defaulters by the RBI.

b) The legal process, wherever warranted, against the borrowers / guarantors and

foreclosure of recovery of dues should be initiated expeditiously. The lenders may

initiate criminal proceedings against wilful defaulters, wherever necessary.

c) Wherever possible, the banks and FIs should adopt a proactive approach for a

change of management of the wilfully defaulting borrower unit.

d) A covenant in the loan agreements, with the companies in which the banks /

notified FIs have significant stake, should be incorporated by the banks / FIs to

354

the effect that the borrowing company should not induct a person who is a

promoter or director on the Board of a company which has been identified as a

wilful defaulter as per the definition at paragraph 2.1 above and that in case, such

a person is found to be on the Board of the borrower company, it would take

expeditious and effective steps for removal of the person from its Board.

It would be imperative on the part of the banks and FIs to put in place a transparent

mechanism for the entire process so that the penal provisions are not misused and the

scope of such discretionary powers are kept to the barest minimum. It should also be

ensured that a solitary or isolated instance is not made the basis for imposing the penal

action.

Guarantees furnished by group companies

While dealing with wilful default of a single borrowing company in a Group, the banks /

FIs should consider the track record of the individual company, with reference to its

repayment performance to its lenders. However, in cases where a letter of comfort and /

or the guarantees furnished by the companies within the Group on behalf of the wilfully

defaulting units are not honoured when invoked by the banks / FIs, such Group

companies should also be reckoned as wilful defaulters.

Role of auditors

In case any falsification of accounts on the part of the borrowers is observed by the banks

/ FIs, and if it is observed that the auditors were negligent or deficient in conducting the

audit, they should lodge a formal complaint against the auditors of the borrowers with the

Institute of Chartered Accountants of India (ICAI) to enable the ICAI to examine and fix

accountability of the auditors.

With a view to monitoring the end-use of funds, if the lenders desire a specific

certification from the borrowers’ auditors regarding diversion / siphoning of funds by the

borrower, the lender should award a separate mandate to the auditors for the purpose. To

facilitate such certification by the auditors the banks and FIs will also need to ensure that

355

appropriate covenants in the loan agreements are incorporated to enable award of such a

mandate by the lenders to the borrowers / auditors.

Role of Internal Audit / Inspection.

The aspect of diversion of funds by the borrowers should be adequately looked into

while conducting internal audit/inspection of their offices/branches and periodical

reviews on cases of wilful defaults should be submitted to the Audit Committee of the

bank.

Reporting to RBI / CIBIL

Bank/FIs should submit the list of suit-filed accounts of wilful defaulters of Rs.25 lakh

and above as at end-March, June, September and December every year only to Credit

Information Bureau (India) Ltd. (CIBIL) from the quarter ended on March 31, 2003.

Banks/FIs should, however, submit the quarterly list of wilful defaulters where suits have

not been filed only to RBI.

Grievances Redressal Mechanism

Banks/FIs should take the following measures in identifying and reporting instances of

wilful default:

(i) With a view to imparting more objectivity in identifying cases of wilful default,

decisions to classify the borrower as wilful defaulter should be entrusted to a Committee

of higher functionaries headed by the Executive Director and consisting of two

GMs/DGMs as decided by the Board of the concerned bank/FI.

(ii) The decision taken on classification of wilful defaulters should be well documented

and supported by requisite evidence. The decision should clearly spell out the reasons for

which the borrower has been declared as wilful defaulter vis-à-vis RBI guidelines.

(iii) The borrower should thereafter be suitably advised about the proposal to classify him

as wilful defaulter along with the reasons therefor. The concerned borrower should be

356

provided reasonable time (say 15 days) for making representation against such decision,

if he so desires, to a Committee headed by the Chairman and Managing Director.

(iv) A final declaration as ‘wilful defaulter’ should be made according to the view of the

Committee on the representation and the borrower should be suitably advised.

Criminal Action against Wilful Defaulters: J.P.C. Recommendations

Reserve Bank examined, the issues relating to checking wilful defaults in consultation

with the Standing Technical Advisory Committee on Financial Regulation in the context

of the following recommendations of the JPC and in particular, on the need for initiating

criminal action against concerned borrowers, viz.

a. It is essential that offences of breach of trust or cheating construed to have been

committed in the case of loans should be clearly defined under the existing

statutes governing the banks, providing for criminal action in all cases where the

borrowers divert the funds with malafide intentions.

b. It is essential that banks closely monitor the end-use of funds and obtain

certificates from the borrowers certifying that the funds have been used for the

purpose for which these were obtained.

c. Wrong certification should attract criminal action against the borrower.

Some of these stringent norms if implemented properly can reduce willful default. The

literature on micro finance shows that peer pressure can indeed impact repayment

behavior of a borrower without having to go through legal measures. Given the success

of the micro-finance institutions banks are also now trying to have such innovative

schemes.

7.5 Peer Pressure as a Mechanism to Reduce NPA

357

Banks are now having tie ups with the small scale industries associations and with the

guarantee of the association bank is providing small borrowers loans without collateral.

In case of non-repayment it is the industry association that is going to put the pressure on

the borrower. To avail this facility the borrower has to be a member of the industry

association. In the process both association as well as a bank gain. Such tie ups are going

on between SBT and Kerala SSI association, Canara Bank and Karnataka SSI association

and so on. If these models become successful they can ensure loans to relatively small

borrower with small or no collateral.

Thus we observe that collateral plays an important role in determining the default

tendency, especially the willful default tendency. However, as mentioned above banks

do not usually designate any borrower as willful defaulter. Our survey of banks lending

shows no defaulters , even though in our personal discussions officials raise their

suspicions about some defaulters being willful. Given this problem it is not possible to

empirically test this hypothesis. In this background we tried to formulate a theoretical

model that captures the relation between collateral and willful default.

7.6 Bayesian Game of Willful Default and Collateral From our survey it has been observe that personal loans and loans under various

government schemes face much higher rate of default than the SSI loan provided with

proper collateral. This provide a basis to infer that lack of collateral security do provide

inducement for willful default. To formalize the strategic behaviour of the borrowers we

therefore intend to formulate a theoretical model to examine the relation between the

value of collateral security , time to retrieve the same and the occurrence of willful

default.

The Model

358

The model comprises 3 players viz., ‘nature’ , ‘borrower’ and the ‘lender’ , the last player

in this case is the commercial bank. The important point to note here is that a bank does

not have information about whether a borrower is potentially a willful defaulter or not.

Thus it is appropriate to assume that nature chooses at random from two types of

borrowers, viz., ‘willful defaulters’ (W) and ‘non-defaulters (NWF)’. In a small scale

sector a genuine defaulter is usually the one with lack of sufficient knowledge about

production and marketing management and hence do not follow any strategic move for

default. Therefore in our analysis we do not bring in this group of defaulters separately.

Since default is empirically seen to have link with collateral value we consider two types

of moves followed by the borrowers. A borrower firm may be ready to either provide a

large or a small collateral. Depending on willingness to pay large (L) or small (S)

collateral, a bank may decide either to disburse fund to the borrower or not to disburse.

The game tree can be represented as follows:

Fig. 7.1

Nature

Willful

Large Col

Small col

L

Firm

Bank Disburse loan

Not disburse

D

ND

D

Nature

Willful

Large Col

Small col

L

Firm

Bank Disburse loan

Not disburse

D

ND

D

359

Next to derive the pay-offs we make the following assumptions:

Loan Period = τ Time to retrieve collateral in case of default= T Loan amount = X Per period return from investing in the project= π Discount factor = β We further assume that by diverting the loan amount a willful defaulter earns a higher

return π́ , per period.

Pay-off for the firm A firm willfully deciding to default faces the following gains and losses. Through willful default a firm may earn expected higher return π́ for life time. But it loses collateral ‘C’ after T periods where C can be either large or small Total expected discounted pay-off (for say, large C) for the firm could be written as

{π́ / ( 1-β ) }- Clarge βT

On the other hand if the firm decides not to default its expected discounted pay-off is

π / ( 1-β )- X ( 1 + r lending) τ β τ Pay-off for the bank In case of default, the bank loses the principal as well as interest but receives the

collateral after T periods.

-X ( 1 + r lending) τ βτ + Ci βT , i = large, small

360

However, in the no default case bank earns the principal with interest after τ period.

X ( 1 + r lending) τ β τ

On the other hand if the bank decides to give no loan it can invest the money in say,

another bank at the market deposit rate . Present value of X will then be X only if we

assume β to be the same as the deposit rate.

Suppose now that the bank can fix a collateral large enough such that pay-off to the

defaulter is negative:

{π́ / ( 1-β ) }- Clarge βT< 0………………

Equilibrium Strategies As mentioned above a bank does not have information about whether a borrower is

potentially a willful defaulter or not. Suppose the bank has prior belief that the

probability of a borrower is a Willful defaulter = q. Suppose it follows the following

strategy , ‘disburse loan only if collateral is large’.

If we assume that a borrower’s strategy is to give a large collateral if he is not a willful

defaulter and small otherwise, will this strategy profile constitute a Bayesian Nash

equilibrium ?

Posterior Probability

Given these beliefs the Posterior Probability as conceived by the bank that the borrower

is a willful defaulter when large collateral can be computed as

Prob ( WF Large) = { Prob ( L WF). Prob (WF) }/ { Prob ( L WF). Prob (WF) + Prob ( L NWF). Prob (NWF)}

361

= 0. q / 0. q + 1. (1-q)=0 Posterior Probability as conceived by the bank that the borrower is a willful defaulter when small collateral is given: Prob ( WF Small) = { Prob ( S WF). Prob (WF) }/ { Prob ( S WF). Prob (WF) + Prob ( S NWF). Prob (NWF)} = 1. q / 1. q + 1. (1-q) = q It can be easily computed that Prob ( NWF Large) = 1 Under condition (*) a willful defaulter will not come forward to take a loan.

Thus given the posterior beliefs, the stated strategies can be sustained as equilibrium

strategies. In the process bank can ensure a separating equilibrium whereby it separates

the willful defaulters from the non willful ones.

However, many genuine borrowers do not possess such high collateral and are forced to

stay away from the formal lending system. Thus, the important question that arises is can

the borrowers providing small collateral be sustained as an equilibrium? More precisely

we consider the following strategy profile:

•Firm: give small collateral •Bank : give loan whether collateral is small or large. Can this strategy profile constitute a Bayesian Nash equilibrium? Posterior Probability as conceived by the bank that the borrower is a willful defaulter when small collateral is given:

Prob ( WF Small) = { Prob ( S WF). Prob (WF) }/ { Prob ( S WF). Prob (WF) + Prob ( S NWF). Prob (NWF)} = 1. q / 1. q + 1. (1-q) = q

362

It is optimal for bank to play this strategy iff q(-X ( 1 + r lending) τ βτ + Csmall βT)+ (1-q)X ( 1 + r lending) τ β τ > X

Let X = X ( 1 + r lending) τ

Then we have condition for giving loan with small collateral, (1-2q) X > X - qCsmall βT………….(**) We next try to represent condition (**) in terms of a simple diagram. Here we measure q in the X-axis and pay-offs in the Y-axis. Line AB represents LHS of

(**) and line XD represents the RHS. The intersection of the two lines and the resulting q

value shows that if probability of willful default is low enough , that is, it lies within OF,

then small collateral may be acceptable for banks; otherwise bank will demand large

collateral. However, if through proper regulatory changes , time to retrieve collateral can

be reduced then the line representing the RHS of (**) shifts down to XE. This in turn

increases the tolerable limit for q for a bank to allow small collateral. Thus if the policy

makers feel that due to the insistence on large collateral many genuine borrowers are

unable to borrow then reducing the time T for collateral retrieval through legal reform is

essential. During our survey we have not seen a single record where a court case has been

decreed and a bank could actually took possession of the collateral.

363

Fig. 7.2

X

P ay-o ffs

q

- X

R H S

L H S

q 1

X

1

X - C S β T

T fa lls

S m aller co lla tera l can b e accep ted

A

B

D

E q F G

O

364

Thus naturally an optimal strategy for the bank to insist on large collateral. The policy

often followed in such circumstances is to compel the banks to lend without or minimal

collateral. Well intended loans under various government schemes fall under this

category. Banks then have to lower the collateral level and without improvement in legal

mechanism to retrieve collateral these lending becomes NPA for the bank. This has

indeed been observed during our survey.

In order to understand the problem of default empirically we have used the data collected

from the selected bank branches (see Section 7.2) in a probit model.

7.6 A Micro Level Analysis of Default Analysis of the determinants of NPA carried out in Chapter 4 is based on aggregative

data from each bank. Having collected micro level data on the SSI and other loan

accounts including personal loans, from selected banks we next tried to look at the

‘determinants’ issue more closely for such small loans in urban areas52. From the banks

we have collected information on both NPA and non-NPA accounts and several other

indicators. However, as mentioned above much of this information is not uniform across

banks and some figures are not reliable. In this background as a first step we estimated a

probit model to look at the factors that determine whether an account would be NPA or

not. Thus NPA is a binary variable and it is our dependent variables. Independent

variables are loan amount, rate of interest and year of establishment of the company.

These are the three variables for which we have consistent information from different

banks. The model under consideration is as follows:

NPAi = α0 +α1 (real_col) i +α2 (rate of interest) i + α3 (year of establishment) i + α4 (loan

under any scheme) i +α5 (loan for SME) i + εi

‘i’ stands for the ith account/ unit.

NPA = 0, if the account is an NPA account

= 1, otherwise.

52 Given the confidentiality of this data we have not presented descriptive statistics etc.

365

Similarly, SME and Scheme are dummy variables, showing whether the loan is under

SME account and given under any Government scheme.

The results of the probit estimates are presented in Table 7.2 Table 7.2 Probit estimates Number of obs = 130 LR chi2(5) = 19.37 Prob > chi2 = 0.0016 Log likelihood = -79.172511 Pseudo R2 = 0.1090 Npa_Non Npa Coefficient Std. Err. Z P>[Z] Real_collateral 5.02 1.89 2.66 0.008 R_o_I (rate of interest)

-.2353688 .0769064 -3.06 0.002

Scheme .248585 .4582996 0.54 0.588 Year .007379 .0165627 0.45 0.656 Sme_non SME -.058827 .3077299 -0.19 0.848 Constant -12.04599 33.37519 -0.36 0.718 F(5) 19.37 Number of Observation 130 Prob>F 0.0016 R Square 0.1090 Probit Estimates Log Liklihood = -79.172511

The probit analysis most importantly shows the role of collateral in case of default. As

the value of collateral increases , the loan has a greater probability of becoming non-npa.

The above results also indicate that in case of the SSI sector interest rate has a positive

relation with NPA. Probability of an account being NPA increases by .07 if interest rate

increases by 1 unit. Thus we get an indication that Stiglitz and Weiss (1981) hypothesis is

holding here. The other variables are turned out to be insignificant. But loan amount and

NPA has negative correlation. An increase in loan size reduces the probability of an

account being NPA in the SSI segment. Other variables are not found to be significant. It

is important to note here that we have experimented with a number of variables such as

loan amount , interaction effects etc. Loan amount and collateral values are found to be

366

correlated and therefore only one of them viz., collateral has been chosen. All othe

variables and interaction effects are not found to be significant.

Out of these default accounts there is no account designated as willful default. However,

the bank officials are sanguine about certain accounts being genuine default due to

business failure or other contingencies while they are doubtful about a few other

accounts. Accordingly, we have categorized the NPA accounts as genuine and not so

genuine and tried to identify the determinants through a probit model (Table 7.3). It must

be noted however, that none of these doubtful accounts are legally proved to be willful.

Within the category of genuine default, given the reason of default, we have categorize

them into two groups: genuine and not so genuine53.

Genuine_noti = α0 +α1 (real_col) i +α2 (rate of interest) i + α3 (loan under any scheme) i

+α4 (loan for SME) i + εi

The variable Genuine_noti = 0, if genuine default

= 1, if not appear to be very genuine

Table 7.3

Genuine_not Coefficient Standard Error

Z P>[Z]

Real collateral value -5.21 3.59 -1.45 0.147 Rate of Interest .1547729 .0777499 1.99 0.047 SME or Non -SME -2.031614 .5602071 -3.63 0.000 Scheme .4600827 .8556752 0.54 0.591 Constant -.4055754 1.108938 -0.37 0.715 F(4) 40.33 Number of Observation 78 Prob>F 0.000 Pseudo R Square 0.3775 Log Likelihood -33.255995

53 This classification is entirely of the researcher.

367

Two interesting results are derived from this part of the analysis. As rate of interest

increases chances of a loan being defaulted under willful category rises. Thus possibly

people do take more risky ventures intentionally with high rate of interest. Secondly, loan

from the SME accounts tend to be more of genuine default type while loan from the

personal account (with low or no collateral) tend to be of willful default category.

People tend to default somewhat intentionally when loan is given under a Government

scheme. However, possibly because such accounts are very few, this variable is not

significant.

7.7 Conclusion

In this chapter we tried to look at the problem of NPA concerning the SSI sector as well

as of the other small-size accounts such as personal loans, from the point of view of a

bank. Our discussions with the bank officials reveal that NPA concerning the SSI sector

is reducing rapidly while from the personal loan segment it is increasing. Banks are now

having tie ups with the industry associations through which they intends to create a peer

pressure on the borrowers and also try to reduce willful default. An analysis of the NPA

accounts shows that interest rate has a positive impact on the probability of an account

being NPA. In addition, the value of collateral plays a significant role in determining

whether an account will be NPA or not and more importantly whether default will be

intentional. The latter can be seen from the fact that personal loans (usually without

collateral) have higher probability of being willful default account.

368

CHAPTER 8

Concluding Remarks

The financial system of any country consists of specialised and non-specialised

financial institutions, organized and unorganized financial markets, financial instruments

and services, which facilitate transfer of funds. Commercial banks form a major part of

financial system in any country in general and in the developing nations in particular.

This is mainly due to the fact that the other financial markets are not usually well

developed. In India, financial system has been synonymous with banking sector. The

importance of banking system in India can be noted by the fact that the aggregate

deposits stood at 55 percent of GDP and bank credit to government and commercial

sector stood at 26 percent and 33 percent of GDP respectively in 2004-05.

Over time however, the Indian financial system has undergone significant

changes in terms of size, diversity, sophistication and innovation. The financial sector

reforms in India began as early as 1985 itself with the implementation of Chakravarti

committee report. But the real momentum was given to it in 1992 with the

implementation of recommendations of the Committee on Financial System (CFS)

(Narasimham, 1991). In the post reform period India has a comparatively well-developed

financial system than before, with a variety of financial institutions, markets and

instruments.

Due to the social banking motto of the Government, the efficiency of operation

and profit earning capabilities are not considered as important criteria for evaluation of

the performance of a bank. Consequently, the problem of non-performing asset (NPA)

was not an issue of serious concern in India in the post nationalization (of banks) period.

However, with the recent financial sector liberalization drive, this issue has been taken up

seriously by introducing various prudential norms relating to income recognition, asset

classification, provisioning for bad assets and assigning risks to various kinds of assets of

a bank. While the Reserve Bank of India (RBI) as well as the commercial banks have

begun to pay considerable attention to the NPA problem, there are only a limited number

369

of rigorous studies in the Indian context that look at this issue in some detail. The current

research project has been taken up in this background.

Given that the NPA has strong implications on the health of the commercial banks and also the economy, it is essential to take measures to reduce the NPA levels in the commercial banks. This calls for identification of the factors that can cause an asset to become NPA. In order to understand this, the study has looked at the data of about 94 banks from 1997-2005 in a panel data framework. We examine whether proportion of rural branches, size of a bank, state of the economy (measured by GDP) , rate of interest and other related variables have impact on NPA levels. It has been found that rural branches indeed contribute to creation of NPA in general. However, in the case of NPA arising from the SSI sector rural branches do not have a negative impact. Rate of interest on the other hand does not seem to have a significant impact on the non-repayment of loans. Some of these results need to be examined carefully once again, in order to arrive at appropriate interpretations.

We have also looked at the efficiency of the commercial banks in generation of

profit. We observe from our analysis that profit efficiency of the public sector banks have

improved over the period (1997-2005) while efficiency of the private and foreign banks

are more or less stagnant. Results show that while rural branches do not contribute to

inefficiency, NPA levels do contribute to profit efficiency (which is something to be

expected).

We have also presented some results from our survey of the small firms and the

banks that deal with such firms. Inadequate loan amount is considered to be a major

reason for default by the firms. Both banks and small firms are aware of the presence of

willful defaulters. Banks consider diversion of funds to more lucrative activities as the

prime cause of willful default. Our discussions with the bank officials reveal that as the

legal process takes a long time (at least 10 years) for confiscating the collateral, it is to

the advantage of the borrower to invest money in other lucrative business such as real

estate and earn higher profit. Our micro level analysis of data collected from the bank on

NPA accounts from the SSI units however shows that interest has positive contribution

towards creation of NPA accounts; more precisely, an increase in the rate of interest

raises the probability of a default.

370

For a comparison purpose we have also looked at the problem of NPA arising out of

personal & other categories of loans such as vehicle or home loan. The problem of NPA

given under personal loan seems a more serious problem than that of the SME sector.

Intentional default is also comparatively more common in the segment mainly due to the

fact that there is no security involved. However, the pressure tactics adopted by the public

sector banks appear to have resulted positive impacts.

186

References

Aigner, D.J and S.F.Chu (1968): On Estimating the Industry Production Function, American Economic Review, 58. Aigner, D.J., C.A.K.Lovell and P.Schmidt (1977): Formulation and Estimation of Stochastic Frontier Models” Journal of Econometrics, 6. Allen, Franklin and Douglas Gale (2000): Financial Contagion, Journal of Political Economy, 108, No.1, pp. 1-33. Asami Tadahiro (2000): Non-Performing Loans in East Asia, News Letter, Institute for International Monetary Affairs. No. 8. Bardhan, P (1989): The Economic Theory of Agrarian Institutions, Claredon Press, Oxford, NY.

Barr, Richard, and Thomas Siems, (1994): Predicting Bank Failure Using DEA to Quantify Management Quality, Federal Reserve Bank of Dallas, Financial Industry Studies Working Paper No. 1-94. Basu, Priya (2005): How Distressed is India’s Banking System?, in Basu, P (ed.) ,India’s Financial Sector: Recent Reforms, Future Challenges, Macmillan, New Delhi. Battese, George E and Tim Coelli (1993): A Stochastic Production Function Incorporating a Model for Technical Inefficiency Effects, Working paper in Econometrics and Applied Statistics, Department of Econometrics, University of New England. Armidale, No.69. Battese, George E and Greg S Corra (1977): Estimation of a Production Frontier Model: With Application to the Pastoral Zone of Eastern Australia, Australian Journal of Agricultural Economics, 21(3), pp 169-79. Beckhart, B. H (1967): Banking Systems, Times of India Press, New Delhi.

Bell, C and T N Srinivasan (1989): Some aspects of linked products and credit market contracts among risk neutral agents, in Bardhan, 1989.

Berger A. N. and Robet Young (1997): Problem Loans and Cost Efficiency of Commercial Banks, Journal of Banking and Finance, Vol. 21, 1997

187

Berger, A N and D B Humphrey (2000): Efficiency of Financial Institutions-International Survey and Directions for Future Research, in P T Harker and S A Zenois (edt) Performance of Financial Institution”Cambridge University Press. London. Berger, A N and L J Mester (2000): Inside Black Box: What Explains Differences in the Efficiencies of Financial Institutions?, in P T Harker and S A Zenois (edt) Performance of Financial Institution Cambridge University Press. London.

Battese, George E and Tim Coelli (1995): A Model for Technical Inefficiency Effects in A Stochastic Frontier Production Function for Panel Data, Empirical Economics, 20, pp 325-332. Bhole, L.M (1999): Financial Institution and Markets, Tata McGraw Hill, Mumbai. Bonin, J. P and Huang, Y (2001): Dealing With bad Loans of Chinese Banks, Working Paper, Department of Economics, Wesleyan University, Number 357 Bonin J. P. and Y. Huang (2001): Dealing with the Bad Loans of the Chinese Banks, William Davidson Institute, University of Michigan, Working Paper Number 357, January. Breusch, T., and A. Pagan (1980): The LM Test and Its Application to Model Specification in Econometrics, Review of Economic Studies, 47, 239-254. Brock, W. A. and D. A. Evans (1986): The Economics of Small Business, New York: Holms and Meier. Chakravarty, S (1985): Report of the Committee to Review the Working of the Monetary System, Reserve Bank of India, Mumbai. Das Abhiman and Saibal Ghosh (2005): Size, Non-Performing Loan, Capital and Productivity change: Evidence from Indian State Owned Banks, Journal of quantitative Economics, Vol. 3, No. 2, 2005 Das Abhiman, Ashok Nag and Subhash Ray (2005): Liberalisation, Ownership and Efficiency in Indian Banking: A Nonparametric Analysis, Economic and Political Weekly, May 19, Vol XL No 12. De Prithwis Kumar (2004): Technical efficiency, Ownership and Reforms: An Econometric Study of Indian Banking Industry, Indian Economic Review, Vol XXXIX, No 1, pp 261-294.

188

Debreu, G (1951): The Coefficient of Resource Utilisation, Econometrica, 19(3) July. Demirguc-Kunt, Asli (1989): Deposit-Institution Failures: A Review of the Empirical Literature, Federal Reserve Bank of Cleveland, Economic Review, Quarter 4. Desai, V (1983): Problems and Prospects of Small Scale Industries in India, Himalaya Publishing House, and Bombay. Deshpande, Lalit et al, (2004): Liberalization and Labour: Labour Flexibility in Indian Manufacturing, Institute of Human Development, New Delhi. Diamond, Douglas W. and Phillip H. Dybvig, (1983): Bank Runs, Deposit Insurance, and Liquidity, Journal of Political Economy, 91, pp. 40 - 419. Dongili Paola and Angelo Zago (2005): Bad Loans and Efficiency of Italian Banks,

Universit ` a Degli Studi Di Verona, Working Paper Series, No. 28.

Farrell, M.J (1957): The Measurement of productive Efficiency, Journal of Royal Statistical Society, Series A, and General, 120(3). Gang, I. N. (1995): Small Firms in India: A Discussion of Some Issues, in Mookherjee, D (eds) :Indian Industry: Policies and Performance, Oxford University Press, India. Ghosh Saibal (2005): Does Leverage Influence Banks Non-Performing Loans? Evidence from India, Applied Economic Letters, 913-918. Global Non Performing Loan Report (2004): Ernst and Young, Goldsmith, Raymond W (1969): Financial Structure and Development, Yale University

Press.

Government of India (1981): Shri Tiwari Committee Report.

Greene, W. H (2003): Econometric Analysis, Pearson Education, Singapore. Hanh T. R. (2005): Safety Net, Franchise Value and Non-performing Loan Problem in Japan. Hausman, J. A. (1978): Specification Tests in Econometrics, Econometrica, 46, 1251-1271 Hubbard. R. G (2002): Impediments to Growth in Japan, Address before the Japan Information Access Project Conference on Fixing Japan’s Economy, Washington D. C., April 8.

189

Jadhav, N and D.Ajit. (1996): The Role of Banks in Economic Development of India, Prajnan, Vol. XXV, No. 3-4. Jondrow, J., C. A. K. Lovell, I. S. Materov and Schmidt (1982): On the Estimation of Technical Efficiency in the Stochastic Frontier Production function Model, Journal of Econometrics, 19, 233-38. Jordan J. S. (1998): Problem Loans at New England Banks, 1989 to 1992: Evidence of

Aggressive Loan Policies, New England Economic Review, January/February, 1998

Kalirajan, K.P. and J.C.Flinn (1983): The Measurement of Farm-Specific Technical Efficiency: An Empirical Testing, Pakistan Journal of Applied Economics, 2. Kalirajan,K.P and M.B.Obwana (1994): A Measurement of Firm and Input-Specific Technical and Allocative Efficiencies, Applied Economics, 26. Kalirajan, K.P and R.T.Shand (1994): Economics in Disequilibrium-An Approach From the Frontier”, McMillan India Limited. King, R G and R Levine (1993): Finance and Growth: Schumpeter Might Be Right, Quarterly Journal of Economic, 108(3), 717-37. Koopmans, T.C (1951): An Analysis of Production as an Efficient Combination of Activities, in T.C.Koopmans (ed.) Activity Analysis of Production and Allocation, Cowles Commission for research in Economics, Monograph No.13, John Wiley and Sons, Inc, New York. Krueger Anne and A Tornell (1999): The role of bank restructuring in recovering from crises: Mexico 1995-98, NBER Working pare 7042, March 1999. Kumbhakar, Saubal C and Subrata Sarkar (2003): Deregulation, Ownership, and Productivity Growth in the Banking Industry: Evidence from India, Journal of Money, Credit and Banking, Vol 35, No 3, June. Kumbhakar, S. C., S. Ghosh and J. T. McGukin (1991): A Generalised Production Frontier Approach for Estimating Determinants of Inefficiency in U. S. Dairy Farms, Journal of Business and Economic Statistics, 9, 279-286. Land, K.C., C.A.K.Lovell and S.Thore (1989): Chance Constrained Efficiency Analysis, Paper Presented at The Conference of Productive Analysis, Canberra: Bureau of Transport and Communications and the Australian National University, 15-17 October.

190

Lovell, C A K (1993): Production Frontier and Productive Efficiency in H O Fried, C A K Lovell and S S Schmidt (ed.), The Measurement of Productive Efficiency-Techniques and Applications, Oxford University Press, London. Lumas, P.S (1990): Monetisation, Financial Liberalisation and Economic Development, Economic Development and Cultural Change, 38. Matthews K., J. Guo and N. Zhang (2006): Rational Inefficiency and non-performing loans in Chinese Banking: A non-parametric Bootstrapping Approach, Cardiff Economics Working Papers, E2007/5, November 2006. Meeusen, W, and B.V.V..Broeck (1977): Efficiency Estimation from Cobb-Douglas Production Function with Composed Error, International Economic Review, 18(2). Mohan, T. R. and S. Ray (2004): Comparing Performance of Public and Private Sector Banks: A Revenue Maximisation Approach, Economic and Political Weekly, March 20, 1271-75. Mor N and B. Sharma (2003): Rooting out non performing assets, Paper presented at the 5th annual conference on Money and Finance in the Indian Economy, IGIDR, Mumbai. Muniappan G. P (2002): The NPA Overhang – Magnitude, Solutions, Legal Reforms, Address at CII Banking Summit at Mumbai, Narasimham, M (1991): Report of the Committee on the Financial System, Reserve Bank of India, Mumbai. Narasimham, M (1998): Report of the Committee on Banking Sector Reforms, Reserve Bank of India, Mumbai. Nerlove, M (1965): Estimation and Identification of Cobb-Douglas Production Functions, Rand McNally, Chicago IL. Pastor, J.M., F.Perez and J.Quesada (1997): Efficiency Analysis in Banking Firm: An International Comparison, European Journal of Operational Research, 98, pp 395-407. Rajam R. and S. C. Dhal (2003): Non-Performing Loans and Terms of Credit of Public Sector Banks in India: An Empirical Assessment, RBI Occasional Papers, Vol. 24, No. 3, winter, pp 81-121. Rajaraman I, S Bhaoumik and N Bhatia (1999): NPA variations across Indian Commercial Banks: Some findings, Economic and Political Weekly, January 16.

191

Rajaraman I. and G. Vasishtha (2002): Non-Performing Loans of PSU Banks: Some Panel Results, Economic and Political Weekly, February 2. Reifschneider, D. and R. Stevensosn (1991): Systematic Departures from the Frontier A Framework for the Analysis of Firm Inefficiency”, International Economic Review, 32, 715-723. Reserve bank of India (2002): Working Group on Rehabilitation of Sick SSI Units, Kohli Committee Report, RBI, Mumbai. Reserve Bank of India Bulletin, Various Issues. Reserve Bank of India, Annual Report, Various Issues. Reserve Bank of India, Statistical tables Relating to Banks, Various Issues. Schmidt, P and R. C. Sickles (1984): Production Frontiers and Panel Data, Journal of Business and Economic studies, 2 (4) 367-74 Seitz, W.D (1970): The Measurement of Efficiency Relative to a Frontier Production Function, American Journal of Agricultural Economics, LII. Sen, K and R.Vaidya (1997): The process of financial liberalisation in India, Oxford University Press, New Delhi. Sensarma Rudra (2005): Cost and Profit Efficiency of Indian Banks During 1986-2003: A stochastic Frontier analysis, Economic and Political Weekly, May 19, Vol XL No 12. Sprayregen, J. H. M., P.C., Jonathan, P. Friedland, N. S. Miller and C Li (2004): Non-Performing Loans in China: a Potential Win-Win Opportunity for Foreign Investors and China’s Economy, FW Global Restructuring & Insolvency Review. Stiglitz, Joseph E. and Andrew Weiss (1981): Credit Rationing in Markets with Imperfect Information American Economic Review, Vol 71, 393-410. Swamy, P.A.V.B (1970): Efficient Inference in a Random coefficient regression Model, Econometrica, 38. Tanaka, M and G. Hoggarth (2006): Resolving bank Crisis – An Analysis of Policy Options, Working Paper, Bank of England, No. 293 Lu D., S. M. Thangavelu and Q. Hu (2005): Biased Lending and Non – Performing

Loans in China’s Banking Sector, The Journal of Development Studies, 41:6, 1071-1091,

August.

192

Timmer, C.P (1971): Using a Probabilistic Frontier Production Function to Measure Technical Efficiency, Journal of Political Economy, 79. Varian, H. R. (1985): Non-Parametric Analysis of Optimising Behaviour with Measurement Error, Journal of Econometrics, 30, 445-58. Vivas Ana Lozano (1997): Profit Efficiency of Spanish Savings Banks, European Journal of Operational Research, 98, pp 381-394. Whalen, Gary (1991): A Proportional Hazards Model of Bank Failure: An Examination of Its Usefulness as an Early Warning Tool, Federal Reserve Bank of Cleveland, Economic Review, Quarter 1. Williams, J and N. Nguyen (2005): Financial liberalisation, crisis and restructuring: A

comparative study of bank performance and bank governance in South East Asia, Journal

of Banking and Finance, 29, 2119-2154

Wu Wen Chien, C O Chang, Z Selvili (2003): Banking system, real estate market and non performing loans, International Real Estate Review, 6 (1), pp. 43-62.

193

1