INSTITUTIONAL FINANCE FOR

AGRICULTURE
A:\;ALYSIS AT \L\CRO A.:\;D \lIeRO LE\'ELS
A thesi.r submitted to the Ullil'ersit)' ol,\l)so,.e, ,\1)".rore,./o,. the award oltbe
olDodor ill E'ollolJli,:r
l" nder rhe.: of
Prof. D Rajasekhar
Institute for Social aM Economic Change, Bangalore
/
/
\
. ,/
\. "

---....
t I c-,: 1_'-
/.
'7Tiis is to certify that this thesis entitfeef ''Institutional 'Finance Jar )f.griculture:
)f.nalysis at 9;(acro and 9;(icro revels" is a 60naJiefe research war/( carn'ref out 6)'
'agan (]3lhari Sahu inefepenefentfy unefer my Bu!efance anef supervIsIon.
I also certify that this has not 6een previousfy su6mitteef Jar the award' oj any efegree
or efipfoma or associateship to any other university or instItution.
lDate: )f.pn[5, 2004
J) . ,
Signature
I here6y decCare tfiat tfie thesis entided "InstitutiolUl{ q:inance for }fgricufture:
}flUlfysis at Macro atufMicro fevefs" is a resu{t of research. wor/(camcd-out 6} me ilt
the Institute for Socia{ and 'Economic Cfiange, under tfie gUIdance of IDr, .0
(RiLjaseRfiar, (Professor, ([Jecentra{isation and(Velidopment Vnit, IS'EC, :13allgu{ore,
I furtfier decCare tfiat it fias not Geen prf'l'iousfy su6mltted eitfier in part orfu{{ to
tliis or any otfier university for any degree, ([Jue acl00wledjJements fiave 6eetz mad,'
wlienever any tliing fias Geen 60rrowed or cited from otlier sources,
(Date: }fpri{ 5, 2004
.,' - - .:'!.
(j'agan IBiiiari .saiiu
CONTENTS
GLOSSARY a
INTRODUCTION 1
Statement of the Problem 1
Conceptual Framework 7
Objectives of the Study 11
Scope of the Study 12
Chapter Scheme 14
POLICIES, PROGRESS AND PROBLEMS IN AGRICULTURAL
CREDIT: THEORETICAL ISSUES 17
Introduction 17
Phases of Banking Policies 18
Movement of Directed Credit in India 31
Subsidised Interest Rate and Its Impact on Agricultural Credit Flow 35
Borrowing Costs and Demand for Credit 41
Concluding Remarks 44
CREDIT FLOW TO AGRICULTURE: TRENDS AND CONTRIBUTING
FACTORS AT ALL-INDIA LEVEL 49
Introduction 49
Credit Flow to Agriculture 52
Distribution Pattern of Agricultural Credit of SCBs 57
Achievement of SCBs in Agricultural Finance 62
Credit Subsidy in Indian Agriculture 65
Number of Rural Bank Branches and Agricultural Credit 71
Determinants of Supply of Agricultural Credit 72
Conclusions 77
INSTITUTIONAL CREDIT TO AGRICULTURE: STATE AND
DISTRICT LEVEL ANALYSIS 80
Introduction 80
Rationale of State Selection and Database 81
State-wise Flow of Agricultural Credit 83
Distribution Pattern of Agricultural Credit 90
Determinants of Credit Flow to Agriculture at the State Level 97
Credit Flow to Agriculture in Orissa
District Level Analysis on Agricultural Credit Flow
Conclusions
105
110
117
HOW DID MACRO POLICY OPERATE AT THE MICRO LEVEL? AN
ANALYSIS OF CREDIT FLOW IN THE SELECTED BANK BRANCHES
IN KALAHANDI DISTRICT, ORISSA 120
Introduction 120
Database 122
Priority Sector Lending 132
Agricultural Lending 136
Security Based Lending 141
Inter-Regional Differences in Priority Sector Lending and Deposits 145
Conclusions 150
VARIATIONS IN THE FLOW OF CREDIT TO FARM HOUSEHOLDS:
Introduction
Nature of Credit Demand of Different Categories of Farmers
Sources of Data and Econometric Methodology
Emerging Patterns in Rural Credit Market
Factors Affecting Access to Formal Credit
Determinants of Supply of Agricultural Credit
Conclusions
CREDIT DELIVERY SYSTEM, CREDIT-GAP AND COPING
STRATEGY OF AGRICULTURAL BORROWERS
Introduction
Credit Delivery System
Borrowing Costs for Agricultural Credit
The Credit-Gap ICG)
The Coping Strategies of Farmer Borrowers
Conclusions
SUMMARY, CONCLUSIONS AND POLICY IMPLICATIONS
REFERENCES
i53
153
154
156
160
164
170
178
181
181
182
188
191
196
208
210
231
Figure 1.1: Factors that Determine the Accessibility of Credit.. 1:
Table 2.1: Distribution of Total Debt of Rural Households by Different Sources ................ 2\:
Table 2.2: Lending Targets for Agriculture Sector.. . ................. 2"
Box 2.1: Recommendations of the Gupta CommIttee ..
Figure 2.1: Rationing Behaviour of the Lender .........
Figure 2.2: Borrowing Costs and Demand for Credit
, .............................................
.. .....
................................... 14
Table 3.1: Annual Average Growth Rates of Credit by SClls to Different Scctors.. .54
Table 3.2: Bank Group-wise Share of Outstanding Credit of SCBs to Agriculture from
Net Bank Credit........................................... ................ . ........................ 55
Table 3.3: Distribution (%) of Outstanding Credit for Agriculture and Allied Activities
(Short and Long-Term) by Different Types of Lending Institutions. . ........ 56
Table 3.4: Annual Average Growth Rates (%) of Outstanding Credit to Agriculture. .. ..... 5;
Table 3.5: Distribution of Outstanding Credit to the Total Net Bank Credit by SClls (eo)
by Sectors and Credit Limits .................. 5R
Table 3.6: Distribution of Loan Accounts by seBs ('Yo) by Sectors and Credit Limits. .59
Table 3.7: Distribution of Agricultural Credit by seBs ('Yo) by the Type of Lending Rates
and Credit Limits ........................................................... ,........................... 61
Table 3.8: Credit Limit-wise Annual Average Growth Rates of Loan Accounts ... . ............
Table 3.9: Proportion of Credit Advanced to Agriculture from Net Bank Credit of SCBs
(Including RRBs) ............................................... ............ . ...... 63
Table 3.10: Distribution (%) of Direct Finance to Farmers by SCBs (Short and Long-Term
Loans) by Size-Classes of Landholding............................... ..64
Table 3.11: Distribution of Disbursement Loan Accounts on Direct Finance to Farmers by
SCBs (Short-Term and Long-Term Loans)...................... . . .. .. ....... .65
Table 3.12: Estimated Lending Costs Per Rs.l00 Outstanding Loan. . ....... .... 67
Table 3.13: Cost of Lending to Agricultural Sector rer Rs.l00 Outstanding Credit of Clls
Including RRBs .......... .......... .................... ..... . 6R
Table 3.14: Cost, Interest Income and Credit Subsidy to Agricultural Sector .. ...... 70
Table 3.15: Region-Wise Growth Rate of Rural Bank Branches.... .. ............ .
-,
. .... " ...... ". i_
Table 3.16: Description, Expected Sign and Coefficient of Variables Used in the Equation ... 75
Table 4.1: State-wise Annual Average Growth Rates (%) of Outstanding Credit by SCBs ..... 84
Table 4.2: State-wise Agricultural Sector Lending to Total Credit and Total Deposit. . k7
Table 4.3: State-wise and Size Class-wise Access to Direct Finance by the Farmers. '1:\
Table 4.4: State-wise and Size Class-wise Outstanding Credit (Direct Finance to Farme,,)
Per Loan Account..... . ....... 9-1
T"ble 4.5: State-wise and Size Class-wise Outstanding Credit Per Hectare of Operati\'n,,1
....... _........... . ... %
Table 4.6: Estimation Result of Sargan's Criterion.... 101
Table 4.7: Description, hpected Sign. and Coefficient of Variables in the Log-line,,,
l\1odcl[Dependent Variable = CGA].. ... .......... 102
Table 4.8: State Specific Intercepts of Fixed Effect Model.. I P .
Table 4.9: Share of Agricultural Credit ('O) from Total Bank Credit bv Each Grnul' of
Bank................................ ....................... Ilt.
Table 4.10: Broad Sector-wise Distribution of Target and Achievements undl'f Annual
Credit Plan of seBs (including RRB) ...... He
Table 4.11: Some Selected Indicators of Spatial Distribution of Development. 111
Table 4.12: Type of District-wise Share (%) of Agricultural Credit Given bv SeRs.. . ...... 111
Table 4.13: District-wise Agricultural Sector Lending ('O) to Total Bank Credit and
Deposits .................................................................................. . ...... 11.1
Table 4.14. Type of District-wise Outstanding Credit Per Hectare of Gross Cropped Mea. 114
Table 4.15: Description, Expected Sign, and Coefficient of Variables in the Log-Linear
Model [Dependent variable = CGA] .................................... . ............... 116
Table 4.16: District Specific Intercepts of Fixed Effect Model ............. .
Table 5.1: Size Class-wise Number and Area under Operational Holdings ..
Table 5.2: Area under Irrigation, Fertiliser Consumption and Area under HYV IJ1
Kalahandi District and Orissa .................................................... .
Table 5.3: Socia-Economic Features of Kalahandi District ............................. .
TE!b!e 5.4: Criteria to ~ f i n e Advanced and Back .. .. ard Blocks ............. .
Table 5.5: Sector-wise Flow of Bank Credit ................................................ .
.. 116
.124
" .. 12h
. ...... 127
'0,
.............. 1.11
............. 132
Table 5.6: Purpose-wise Distribution of Agricultural Credit from 1996-97 to 200001 ........... 137
Table 5.7: Sector-wise Recovery of Loan Amount to Demand ...... . . .......... l3B
Table 5.8: Interest Income on Advance ........................................ .. . ...... 1-10
Table 5.9: Distribution of Agricultural Credit by Type of Security .......... . ..IB
Table 5.10: Proportion of Priority Sector Lending to Total Credit and Total Deposits ........... 146
Table 5.11: Proportion of Lending to Agriculture from Total Credit and Total Deposits ..... 147
Graph 1: PSCBD and PSCBC in the Sample Bank Branches of jaipatna and Narla Block ..... 141'
Graph 2: AGLCBD and AGLCBC in the Sample Bank Branches of Jaipatna and Narl.
Block............................................................................. .150
Table 6.1: Distribution of Borrowers by Farm Size (2001-02) ............. . ..161
Table 6.2: Distribution of Loan by Farm Size in the Year 2001-02....... . ............. 162
Table 6.3: Landholding-wise Number of Times Crop Loans (Kharif) Obtained bv
Farmers for the Period 1999 - 2000 to 2001 2002...... .16:1
Table 6.4: Supply of Credit (Crop Loan) Per Hectare of Gross Cropped Area. Ih-l
Table 6.5: Definition, Measurement. Descriptive Statistics, and Expected Sign of
Variables Used in the Prohit Equation ............ . .............. 166
Table 6.6: Access to Formal Credit: Probit Results ...
107
Table 6.7: Definition, Measurement, Descriptive Statistics and Expected Sign of Variables
used in the OLS Equation...... J 7"
Table 6.8: Determinants of formal Credit: Selectivitv Correctl'd 01.5 Rl'sult.
Table 6.9: Determinants of Formal Credit: Tobit Results ....
17S
177
Table 7.1: Problems Faced by the Farmer Borrowers in Obtitining Loans 1B4
Table 7.2: Size Class of Landholding-v.:ise Frequency Distribution of Problems Fd({'d b\'
the Farmer Borrowers.......................... 1 ~ S
Table 7.3: Time Spent to get the Loan Sanctioned 187
Table 7.4: Shortage of Amount Sanctioned to Amount Demanded.. 1 8 ~
Table 7.5: Cost of Borrowing (Short-Term) from the Financial Institutions III till' Stud\,
Area.................................................................... .. .............. 190
Table 7.6: Cost of Borrowing of Institutional Credit by Farm Households According to
the Size of Loan in the Study Area....................... . ....... 191
Table 7.7: Extent of Dependency on Credit Per Acre of Gross Cropped Area (2001-02) ....... 192
Table 7.8: Landholding-wise Credit-Gap....................... .. ....... .. . ... 195
Table 7.9: Self-finance Capacity of Farmers Per Acre of Gross Cropped Area .. . ........ 196
Table 7.10: Distribution of Borrowers by Number of Sources of Borrowing .................. . .. 197
Table 7.11: The Predetermined Exchange Rate of Paddy in Interlocked Credit Market . 0 ~
Table 7.12. Regions & Landholding-Wise Estimated Income Loss Per Bag of Selling
Paddy due to Interlocked Credit Market in Kalahandi District, Orissa (2001-
02) ...................................................................................... . .................... 205
T?b!e 7,13: Farm Size and Collateral-wise Percentage of Lean Contracts and PErcEntag
of Loan (amount) Borrowed ............................................................. ................... . 207
ABC
ACRC
AR
ARC
ATC
Alls
BC
CBs
CDR
CFS
CG
Cl
CRAFICARD
CRR
CRS
CS
DAP
FFIs
FRBS
HYV
IC
IMR
LAMPs
LBO
LDB
LT
MSP
MVP
NABARD
NDC
NDP
OLS
OPS
PACSs
PLCP
PLR
RBI
RFIs
RFMs
RRB
SAPs
SBF
SC/Sl
SCBs
SlR
SSI
ST
TC
TJ.
TPS
GLOSSARY
Average Borrowing Cost
Agricultural Credit Review Committee
Average Revenue
Agricultural Refinance Corporation
Average Transaction Cost
Agricultural Term loans
Borrowing cost
Commercial Banks
Credit Deposit Ratio
Committee on Financial System
Credit-Gap
Credit limit
Committee to Review Arrangements for Institutional Credit for Agriculture and
Rural Development
Cash Reserve Ra tio
Constant Return to Scale
Credit Subsidy
Di-Ammonium Phosphate
Formal Financial Institutions
Formal Rural Banking System
High Yielding Varieties
Informal Credit
Inverse Mills Ratio
Large Area Multi Purpose Societies
Lead Bank Office
Land Development Bank
Long-Term
Minimum Support Price
Marginal Value Product
National Bank for Agriculture and Rural Development
No Dues Certificate
Net Domestic Product
Ordinary Least Squares
Other Priority Sector
Primary Agricultural Co-operative Societies
Potential Linked Credit Plan
Prime Lending Rate
Reserve Bank of India
Rural Financial Institutions
Rural Financial Markets
Regional Rural Bank
Service Area Plans
Small Business Firm
Scheduled Caste and Scheduled Tribe
Scheduled Commercial Banks
Statutory Liquidity Ratio
Small Scale Industries
Short-Term
Transaction Cost
Term-Loan
Total Priority Sector
(Dun'llg my post graauatioll aays, I usea to pOllacr tliat aoillg a aoetorate was not a smarr
job, ana was a aream for me. Toaay, my aream nas matena[uea 1I'itn tnc co-operatlOli and'
ne[p, a great aea{ of tnougnt ana efforts of many peopfe III many '!'ays. It is neltfier
possibfe nor suffident to acli.JlOwfeage tfiem in tfiis fomw{ "Wllller witfi a fcw woras.
'First alla foremost, I remaill aeepfy Illaebtea to my supervisor q'rof rD. '1(aFlSCliJiar unacr
wnose abfe guiJallce tfiis war/( fias beell camea out. J{is fie[pfu{ cntlcism, suggestions,
meticu{ous reaefillg ana eaiting, constant encouragement, ana avera{[ morae support maac It
possl6fe for me to compfete tfie tfiesis alia to see tfie figlit of tfie aay, I a1ll ever gratifll{ to
filln.
)'ls a researcfi scfiora,., I fearnt a wt from rDr. S. :Maafieswllrun, on a number of eCOllOlme
issues tfirougfi aisCUSSIOns, in genera(, ana on matters refuting to tfie app(zeation of
quantitative tecliniques, in particufur. I fOlzafy cfierisfi alla va{ue my association witfi film,
I wouUf Ms to e:;;press my sillcere tlian/(} to !Frof 'R., 5. rDeslipanae for liis concenz a60ut
my war/(since tfie beglnlllng. rBeing a teacfier ana as one of tfie pand me/n6crs for a[{ tfie
semilwrs presentea at tfie Institute, lie offerea va[ua6fe comlllmts alia suggestions, wlilcfi
lie[pea me a fot at aijferellt stages of my war/(, I am afso tfianliju[ to !Frofessors 'J( L
'J(risfina, rB. 'J(amaiali alla'J(aifas Sarap, for tlielr va{ua6(e comments alu! suggestio liS.
I sincerefy e:{press my gratifu&zess to tfie InstItute for Socia{ ana'Ecollomic Crwnge (ISLe)
for providing me all opportunity to pursue my aoctora{ studies. I tlian/( tfie preVIous
director Prof :M. C;OVlllaa '1(00 ana tlie presellt airector (Prof C;opaf'J( 7(aaefi.9difor lfielr
support ana C/!couragement to my researcli war/(,
:My sincere tlian/(} to !Frof Tom rBrass, 'Editor, 'lJie Jounzaf of Peasant Studies, Prof
:N.s.s. :Narayana, Inaian Statisticaf Institute, r]3anga{ore, !Frof '1CP. 'l(pfirajan, Japan ana
Prof Stein JfoUfm, rDepartment of )'lgn'cu{tura[ 'Economics, :Norway, for tfieir comments
ana suggest io liS Oil cfiapters 6 ana 7.
I wouUf lili. to e:;;press my aeep sense of gratitllae to Professors :M.'R., :Narayall<1, 'r::M.
'1(ao, J{elll{ata '1(ao, 1( '{J, 'Rflju, C;.'J( 7(aral/tfi, 'J(:N.:M. '1(a]u, 5.:N. Sal/glta aI/a (Drs. :M.J,
r]3nCllae, :M. :Mafiaaeva, '1CCj. Cjayatfiri (DeVl; (B.P. '(Jam, 'Vcerasfickfzarappa for (fielr
ellcouragement, liefpfu{ a lid usefu{ discussiolls. I am afso tliaII iifll{ tD alfierf'lw{l\' mem6ers
of 15'EC for 6eillg liefpfuC I am gratefu{ to Prof flravil1if 'Pallaganya (,UIll('t'TSlty or
:Maryfallli, VSfl) wlio visited 15'EC for a sliort time III 2002 for gi'l'lllg suggestiolls
I alSo tliall/(a{{ tlie participatlts of (}3i-allllua{ semitlars presetlted at IS'E( at nlnous st,lgcs
of my study for tfieir constructive commetlts.
fl researcli tfieme of tfie tlature requires discussioll witli scliofars, 6all/( staff a lid'
illfonnatioll from van'ous sources. I am immensefy gratefu{ to everyolle wlio lias assisted me
ill acliievillg tliis.
I am tliallkju{ to my respolldetlts for proviiftng refevatlt i'ifonnatiotl atld sliantlg tlielt
eJiPerimce wfiife conductIng tlie survey, WlilCIi, itlfact, lieped me coIlSldera6fy ltl atlafyslllg
tlie data 011 tfie 06jectives of tlie study.
I am tliankju{ to :Mr. J{.JV. (j(all{]anL1tfulIl, 'R.sgistrar, IS'EC, atuf:Mr. 0/. (j(amappa,
)1.ccounts Officer, !SCC, for t!!e!r fiefp.
I am gratefu{ to tlie pli. ([), Programme Co-ordinators (IS'EC) at differetlt POltltS of time
alld:Mr. 1(s. Narayan for tlieir support at tlie needed time. I sitlcerefy acli.JlOuIedge
J{.1( flmamatli forprovidi'tlg tlie tleeded data.
I sillcerefy acli.JlOwfedge tfie fiefp alld assistance received from :Mr. 7(nsfilla Cfiandratl atld
:Mr. Satisli 7(amatli ill tlie computer centre and tlie mem6ers of LI6rary staff of 15'EC
especia{fy :Mrs. Lifa, 7(afyallappa, 'lIm/(atesli alld (j(ajamw. I tliall/( :Mr. 'lis.
Partfiasaratliy for liis fiefp in edi'tillg of tfie tfiesis. I gratefu{fy acli.JlOwfedge tlie liefp
provided 6y :ManJufa in givill{] tlie tliesis its jina{ sliape.
I taR.! tfiis opportunity to tliatll( my jnelufs at IS'EC eSPec/a{fy (}3lianu,
flmar, ([)eepi/(a, Sita/(atlta, Satya, Pratuslina, (}3initlia, flsliisli, 'l(sliyama,
Su60dli, Puma, ([)ukjia, J{ruslii, fltlanda 'lIadiveu, {Bi/(asli, (jlta''laya, Sar6al/l atld
otliers for tlieir kjluf liefp at differetlt pOints of time. I am alSo tliatliifu{ to )llIatltli,
,/(nslitlagouda, :Madliavi, Saclil alld Jugafjor tliClr getlerous liefp.
Last 6ut /lot feast, I am tliallkju{ to my pare'lts, 6rotliers, sisters, atld my wife for tliClr
COllStatlt morae support Itl fu(fi[(ztlg my dream of compfeting tlie pli.v' tliesls.
qagan {}3inari Sanu
Chapter 1
INTRODUCTION
Statement of the Problem
A majority of the cultivators find borrowing necessary as their own
farm savings are inadequate to finance vanous agricultural
development activities. It has been argued that credit provides
command over resources and facilitates the needed liquidity to the
farmers (Lipton 1976). Credit is assumed to be helpful for changing
the composition and distribution of production in favour of deficit
producers. Improved access to formal credit is supposed to shift rural
borrowing from informal market to formal institutions, increasing the
use
... __ 1 ____ ' __ _
LCL.HHUIUg,y, leading [0 increased
production and higher income for the rural poor (Donald 1976; Sarap
1990).
With the introduction of new agricultural technology in India in
the mid-1970s, the need for improving the quality of institutional
credit (timely and adequate loan amount) to agriculture \\'as largely
felt. With the assumption that the new agricultural technology was
scale neutral, it was felt that increased access among the marginal
farmers to credit would improve the employment potential and
contribute to better income distribution in rural areas.
Taking the above into account, many committees recommcnckcl
improvement in the outreach of the formal financial institutions in
favour of rural areas, and small and marginal farmers. The
nationalisation of banks in 1969, various policies aimed ,,1 s()cia!
banking and the establishment of Regional Rural Banks III 1 "7,'1
resulted in an increase in the number of bank branches in rur"l
areas. During the mid-1980s, the mandatoI'- priorit\, sector lending
\'.-as introduced to ensure adequate credit flow to hitherto neglected
sectors such as agriculture and alike. Undeniably, these resulted in a
vast network of rural financial institutions, and a rapid growth c;f
lending to all sectors including agriculture.
However, the following problems and deficiencies 111 the rural
credit delivery system were also noticed,
:J The rural banking system in India made tremendous quantitative
achievement by neglecting the qualitatiye aspects of the credit
delivery system (Shivamaggi 2000), The empirical studies show
that institutional c;rFdit was Hccessed by the ,vel! to do among n . .H"3.!
people (Adams and Vogel 1986). Further, the Indian policy makers
were unable to arrive at a banking structure and operational
system, which was suitable for the credit needs of the
agriculturists (Shivamaggi 2000).
:J The inequalities in the banking system across the regIOns and
social classes persisted (Bell 1990). This was because of the
insistence on collateral (Sarap 1991) which could not be prO\'ided
by the poor, complicated administration proced u re s, lon
a
"
distances from the villages to the branches, the cultural gap
between bank officials and the poor, political interference,
inflexible lending policies and procedures, lack of prO\'ision for
consumption credit and a widespread belief that the poor were
non-bankable.
J Although some of the rural poor obtained credit from the Form;l!
Rural Banking System (FRBS), they found that the credit \\';j"
neither timely nor adequate for their needs (RaJasekhar and
Vyasulu 1990). The disbursal of loans without assessing the
feasibility, viability and entrepreneurial experience of borrowers led
to the problem of fungibilityl. There were also problems \\'ithin the
banking system such as lack of linkages between credit and
production, extension, marketing, etc.
::J The formal banking system began to face problems of higher
transaction costs
2
(due to lending small amounts to large numbers
of borrowers) and recurring losses 1I1 the late 1980s. The
managerial inadequacies within the banking system also caused
problems (Vyasulu and Rajasekhar 1993).
::J All the above culminated in an alarming growth in overdue
payments
3
. This, in turn, adversely affected the recycling of credit,
and led to lower profit margins (Singh 1991; Rajasekhar and
1 It implies the process of financial substitution. For a detmled discussion, see, Von
Pischke and Adams (1980), Ladman and Tinnermeier (1981), and Adams and
Vogel (1986).
2 A study made by Satish and Swaminath (1989) found that the average cost of
lending of CBs, Primary Agricultural Co-operative Societies (PACSs), RRl:ls and
Land Development Banks (LOBs) was greater than the average interest II1come
from the agricu1tural sector. Discussing the \'lability of rural credn institutions.
Gadgil (1986) emphasised the high cost of transaction, and the mounting loan
delinquencies, which were eroding the strength of the rural cft"dit institutions.
Desai and Namboodiri (1992) found that all Rural Financial InstitutIons (1-11'151.
except PACSs, prevailed under constant returns to scale ( C I ~ S ) in theIr transactio,"
cost from the period 1961 to 1981. PACSs suffered from scale d,seconomlCS II1
their transaction costs for the saIne reference period.
J The causes of overdue payment haye been briefl.\" disruss('ri in l-\hcs;o (1 YS91 Ti:-::"
issues relating to overdues, default rate, political Intervention 1:: sanction ar:c
recoveries of loans. and defective lending policies ha\'e bee;; discussr'd ]T'.:
Ra]asekhar and V,'asulu (1990). \'Yasulu and I<a]"sekhar 11<)c< I. and h:1hk:c
(lCj91).
Vyasulu 1991). Many Commercial Banks (CBs), Regional Rural
Banks (RRBs) and co-operatives were consequently incurring
losses.
This led to mounting pressure on the government from Indian
bankers (and some outside agencies) to initiate banking sector reforms
for the financial viability of the FRBS. Yielding to this pressure, the
Government of India appointed the Narasimham Committee in 1991,
which recommended measures to improve efficiency, profitability and
viability of the banking sector. The recommendations of the
Narasimham Committee relevant to the present study were to reduce
the share of credit to the priority sector from 40 per cent to 10 per
cent of aggregate bank credit, deregulation of interest rates, abolition
of branch licensing, gradual phasing out of directed credit
programmes, closing down of loss-making bank branches and so on.
The committee was of the view that easy and timely access to credit
was far more important than its cost. Gradual reduction of the Cash
t{eserve Ratio (eRR) and Statutory Liquidity Ratio (SLR) to enable the
banks to have larger quantum of loanable funds was also
recommended. Besides, the committee took the view that the growth
of agriculture and small industry reached a point where banks on the
basis of their commercial judgement were in a position to meet the
legitimate productive needs.
The Gupta Committee, set up in 1998, to examine the issues
relating to the quality of credit delivered by the banking system,
highlighted the issue of how an improvement in the quality of lending
helps the banking system to become financially viable. The committec
suggested measures aimed at simplification of proccdures for
agricultural credit and identification of constraints faced bv the
commercial banks to increase the flow of credit.
The government did not accept all the recommendations made
by the Narasimham Committee (1991), especially the one on reducing
the target for priority sector. Nevertheless, it appears that, the strategy
followed by Reserve Bank of India (RBI) was to keep the overall share
of credit to priority sector at 40 per cent, and broaden the scope of
priority sector to encourage diversion from direct priority sector
lending. Since September 1990, interest rates to the ultimate
borrowers were rationalised by the RBI. This policy was based on the
Narasimham committee recommendation for deregulation and
abolition of concessional interest rates so that all rates becamF
positive in real terms.
It can, therefore, be stated that the banking sector reforms may
have the following positive as well as negative impact on the credit
flow to agriculture.
[J Increased availability of funds to banks as a result of reduced CRR
and SLR may have a positive impact on credit flow to agriculture
provided the banks are unable to find alternative and profitable
non-agricultural avenues to disburse the credit.
:J The de-regulation of interest rates may increase the credit flow to
agriculture because it has been often argued that what matters to
the farmers is the access to credit rather than its costs. With de-
regulation of interest rates and the concomitant increase in returns
to banks, one can expect that the quality of credit delivery system
may be improved and the credit flow to agriculture stepped up.
o The introduction of credit norms relating to profitability are also
expected to increase the credit flow to agriculture, provided
bankers perceive that agricultural lending is profitable.
o The closure of loss-making branches may, however, lead to
reduction in credit flow to agriculture. This is because most of the
loss-making branches during the pre-reform period were found to
be located in rural and semi-urban areas. If these branches are
closed, the credit flow to agriculture may come down
4
.
o The widening scope of priority sector may result in the greater
availability of credit to priority sector, in general, and agricultural
sector, in particular.
The above lead to the following questions: What have been the
trends in credit flow to agriculture? Was there any difference in the
credit flow to agriculture between the pre-reform and reform periods?
What factors governed the credit flow to agriculture during the pre-
reform and reform periods? How did the policy of widening the scope
of priority sector bank lending enhance the credit flow to agriculture?
Did the rationalisation of interest rates contribute to more credit flow
to agriculture? What was the impact of profit-oriented lending norms
on the credit flow to agriculture? What was the impact of closure of
rural bank branches on the provision of credit to agriculture?
4 In the earlier role of 'social banking', banks became the instruments of the state
in the development process, \\:hich are now oriented towards profit maximisation
which guides the evaluation of pre-reform branch licensing polic\. The sa\"lngs
rate in India, in fact, is positively associated with the availability of bank
branches. The evidence shows that rural branches have increased their share of
deposits and credit between 1969 and 1996 from 3.0 and 0.2 per cent to 15.4
and 12.4 per cent. respectiveh' (Kohli 1999). For more detailed anahsls. see. I'r ..
(1988). Pandit (1991). and Desai (1993).
(,
Conceptual Framework
The present study seeks to analyse these issues at both macro (all
India and state levels) and micro (household) levels. Finance to
agriculture at the macro level depends upon the collective impact of
policies relating to agricultural credit, in general, interest rate and
norms relating to profitability, in particular. The policy on finance at
the macro level primarily concerns with the level of credit flow to
agriculture, terms and conditions under which it is made available,
allocation of credit between sub-sectors (production and investment
credit) within agriculture to achieve production efficiency and so on.
Besides, macro credit policies are concerned with the provision of
cheap, adequate and timely credit (for both production and
investment) to the needy farmers. encouraging them to adopt nt'w
agricultural technology, and pre and post-appraisal of loan
disbursement, etc.
These policies, In general, give guideline to micro level
institutional finance agenCies having direct contact with the
borrowers. The flow of credit to agriculture at this level depends on
financial feasibility of the concerned project, mechanisms of client
selection, repayment performance of the borrower, ability to provide
collateral security, cost of lending, cost of borrowing and so on.
Therefore, it appears that macro level credit policies may not address
the problems relating to agricultural credit at the micro level. Hence,
an examination of the constraints faced by the farmers to access
institutional credit at the micro level is essential. Furthermore, many
changes in agricultural credit policies have taken place during the
reform period (see, Chapter 2). which may have affected the flow of
7
credit to agriculture directly or indirectly at both macro and mIcro
levels.
Credit Accessibility Problems
It is well known fact that rural financial markets are imperfect in
nature. Cost involved in obtaining information on potential clients,
long distance from the village to the bank branches, the cultural gap
between bank official and the poor, inflexible lending policies are some
of the critical elements that can restrict the outreach of rural financial
institutions. Besides, rural farmers are faced with a number of
constraints that restrict their access to a given supply of formal credit.
The literature on credit markets explains the outreach from
supply side perspective and access from the viewpoint of the potential
borrower, wherein both indicate the same thing, i.e., who is obtaining
credit. The term "accessibility" is used purposefully in this study to
emphasise the link between bank-related and household-related
factors to explain why certain households borrow, while others do not.
In the context of credit accessibility, the problem has been looked at
from both breadth (the number of clients reached) and depth (type of
clients reached) of outreach. To study these two aspects, both supply
as well as demand-related factors have been taken into account (see,
for example, Sarap 1990, and Hashuliza and Kydd 1996).
Furthermore, in the context of "accessibility", the issue of
outreach (i.e., in terms of both breadth and depth) and sustainabilitv
of financial institutions are often debated (Yaron et I 1997). Critics
note that too much of importance to the breadth of outreach resulted
in the neglect of the depth of outreach (Morduch 2000). The possible
8
trade-offs between sustainability and the depth of outreach ha\'e been.
therefore, the subject of an ongoing discussion in the rural credit
markets (Rhyne 1998).
The foregoing discussion suggests that importance to the
breadth of outreach may endanger the sustainability (viability) of the
financial institutions. But, the reverse may not necessarily be true. In
other words, if the sustainability is the main agenda then the depth of
outreach seems to be crucial. If the credit transaction is looked from a
lender's perspective, it is the objective of banks that determine the
flow of credit, and, in turn, its access by different categories of
farmers.
Therefore, the credit transaction can be analysed from the
characteristics of the supply versus demand side. Since credit
generally involves a time gap between the point when credit is given,
and the future point when credit is repaid with interest, anything may
happen in-between these two time points. So, the credit transaction
involves interpersonal trust between the borrower and the lender.
However, if the credit transaction IS looked from a lender's
perspective, one can anticipate pre-contractual (at the point of
disbursement of loan), and post-contractual problems (at the point of
loan repayment). The former brings the issue of adverse selection (i.e.,
the fear of selecting a bad borrower), credit-worthiness and screening
of the borrower. In the case of latter, a lender faces the problem of
wilful and non-wilful default. Given these problems, what governs
access to institutional credit for some peasant households but not
others? A conceptual framework summarising the factors influencing
access to credit is provided in Figure 1. 1.
9
Figure 1.1: Factors that Determine the Accessibility of Credit
Financial Institutions Farm Household
- objective of - beinfJ part of th
the financial ~ ~ ~ ~ ~ service area
products - capacity to take
- mechanism of a loan
borrower - willingness to take
selection a loan
Credit Outreach
credit of the financial institutions are their objectives of the financial
products, which may have positive or negative effects on breadth and
depth of outreach. Objectives like reduction in cost of lending or
specification of target group can have a negative effect on the breadth
of outreach. The second category of factors that determine
accessibility of credit product is the mechanism of borrower selection.
To select a borrower they need either a guarantor or physical collateral
or both. Elsewhere, it has been argued that a guarantor can be
considered as an important form of social collateral, which IS more
important than physical collateral (Vaessen 2001). Collateral
requirements may have negative impact on depth of outreach.
From the borrower's perspective, factors that influence access to
credit can be put under the following three broad categories. (a) a
borrower must be living in the area of operation, (b) a borrower must
have the capacity to take a loan, and (c) a borrower should be willing
10
to take a loan. Capacity to take a loan refers to the household's
capacity to meet the formal selection criteria and banker's additional
expectation on repayment capacity. In order to win over the lender's
confidence, the borrower must be in a position to satisfy the lender
with some collateral both tangible and intangible. Willingness to take
a loan depends upon the transaction costs, availability of credit from
alternative sources with more attractive terms and conditions, interest
rates and so on.
The above theoretical framework on accessibility of institutional
credit (Figure 1.1) expects that many factors influencing the credit
accessibility are at work. Since access to formal credit is not common
to all categories of farmers, it is necessary to analyse the factors that
account for variations in the credit flow. Even if some farmers are
accessing formal credit, is it sufficient to meet their requirement? If it
is not, under what conditions do bankers decide the credit limit to a
particular farmer? If the disbursed loan amount is largely influenced
by supply-side factors, a larger gap between supply and demand for
credit is the likely possibility. Furthermore, the magnitude of this gap
may vary across different size classes of landholdings.
Objectives of the Study
In the light of these issues, the present study on credit flow to
agriculture derives its relevance and has the following objectives.
To study macro (all India) level trends of institutional credit to
agriculture before and after the introduction of the banking
sector reform, and analyse the contributing factors.
To study dis-aggregated trends (state and district level within
Orissa) and contributing factors to the credit flow to agriculture.
II
To analyse the factors that account for variations in the flow of
credit at the household level.
To measure credit gap among farm hOllseholds across the size
class of landholdings.
Scope of the Study
The macro level analysis on the credit flow to agriculture, has been
confined to all India, while the state level analysis to 14 major states.
These states consist of Andhra Pradesh, Bihar (undivided), Gujarat,
Haryana, Karnataka, Kerala, Madhya Pradesh (undivided)
Maharashtra, Orissa, Punjab, Rajasthan, Tamil Nadu, Uttar Pradesh
(undivided) and West Bengal. The rationale for the selection of these
states lies in the fact that they account for 93 per cent of the
populatIOn and ':11.5 per cent of the Net Domestic Product (NDP) in the
country (see, Rao et al 1999). Besides, these states have been
accounting, on an average, 84.7 per cent of the total deposits in the
country through Scheduled Commercial Banks (SCBs) including RRB
for the last twenty years. Out of the total agricultural credit provided
in the country through SCBs and RRBs, on an average, 95 per cent of
it goes to these states (RBI, various years). The hill states of north and
north-eastern parts of India, which are considered as special category
states by the Planning Commission, have been excluded from the
analysis due to significant differences in the structure of these
economies from the rest of the states. It is also worth mentioning that
in these hill states the important sources of income are government
activities and there has hardly been any manufacturing base (Rao et
al 1999).
12
The district level analysis has been done for the state of Orissa
because of its distinctive character of backwardness'- The Economic
Survey of Orissa for the year 2000-01 notes with concern that the
credit deposit rati0
6
(CDR) in the state declined from 80.59 per cent in
1991 to 39.75 per cent in 1999-00, while per-bank branch population
increased from 15,200 persons to 16,300 during the same period. The
reduced CDR might be adversely effecting the credit flow to different
sectors in the economy. It was, therefore, noted that there was a need
to look into the ground level credit flow to the different sectors in the
state, in general, and to agriculture, in particular.
The micro-level analysis has been carried in the specific context
of Kalahandi district, Orissa. Located in the western part of the state,
this district IS known 111 India for widespread poverty and
backwardness. For the micro level analysis, data from four different
bank branches and 200 farm households (falling within the service
area of these banks) have been collected. Since farmers in the study
area hardly get crop loans during Rabi session, the micro level
analysis is restricted to only Kharif crops for the year 2001-02.
The secondary data for undertaking analysis at all India, state
and districts level were collected from the Report on Currency and
Finance (RBI), Banking Statistics (RBI), Fertiliser Statistics (the
Fertiliser Association of India), Indian Agricultural Statistics (Ministry
5 An examination of the data on levels of per-capita income, industrialisation,
urbanisation and literacy rate shows that this state falls into the category of
'backward' .
6 The CDR in Orissa was well below the all India average of 57.05 per cent in 1999-
00, and the states with better CDR than Orissa included Tamil Nadu (88.0 %),
Maharashtra (83.79 %), Andhra Pradesh (63.60 %), Karnataka (61.01 %), Gujarat
(49.85 %), Madhya Pradesh (49.17 %), Rajasthan 146.69 %), West Bengal (45.17),
and Kerala 142.29 %). This information was collected from Bankmg StatistIcs.
Quarterly Handout, RBI (March 2000).
13
of Agriculture, Government of India) and Statistical Abstract (Central
Statistical Organisation, Department of Statistics and Programme
Implementation, Government of India). For the analysis on trends in
credit flow in Orissa state, data were collected from Stale Level
Bankers Committee Meetings (Agenda Notes) and Orissa Agricultural
Statistics (Directorate of Agriculture and Food Production, Orissa,
Bhubaneswar).
Since the banking sector reforms have not made any impact on
the financial pattern of Co-operatives, the study is exclusively based
on the finance provided by the Scheduled Commercial Banks to
agricultural sector. The Scheduled Commercial Banks consist of State
Bank or india and its Associates, Nationalised Banks, Foreign Banks,
Regional Rural Banks, and Other Scheduled Commercial Banks.
In the case of micro level (households), analysis emphasis was
gIven on the loan provided by formal institutions. Nevertheless, to
measure the credit-gap and to capture the dependency of the study-
group on informal sources of credit, loans provided by non-formal
agencies have been taken into account.
Chapter Scheme
The study comprises eight chapters.
The second chapter provides a review of literature on issues relevant
to agricultural credit. This chapter by focusing on policies, progress
and problems of agricultural credit in India, exammes the major
directions in rural banking and credit policies during the post-
\-1
independence period, and the policy performance especially in respect
of credit flow to agricultural sector.
Chapter three, on "Credit Flow to Agriculture: Trends and
Contributing Factors at All India Level", analyses the trends in credit
flow in terms of percentage and growth rates, and contributing factors
with the help of the ordinary least squares method.
The fourth chapter covers the state and district level analysis on
the credit flow to agriculture during pre-reform and reform periods. In
this chapter also, a trend analysis in terms of growth rate and
percentage has been made to examine the flow of credit to agriculture.
It has to be mentioned here that the proportion of net bank credit may
not be able to explain the actual situation in so far as the flow of
agricultural credit at the ground level is concerned. Therefore, the
availability of credit per hectare of gross cropped area has been
calculated at both state and districts levels to analyse the variations in
the supply of agricultural credit. Factors contributing to the variation
in the supply of credit per hectare of gross cropped area have been
analysed with the help of Fixed and Random Effect models at both
state and district levels.
The fifth chapter explains the impact of macro level agricultural
credit policy at micro level. To assess the impact of these policies on
the pattern of supply of credit to different sectors, the information \\as
collected from four bank branches (two each belonging to commercial
bank and RRB categories) located in advanced and backward blocks of
Kalahandi district in Orissa. The factors accounting for variations in
the flow of credit to farm households have been analysed in the sixth
IS
chapter. It is worth mentioning here that an analysis on the variations
in the flow of credit can be undertaken either in terms of access to
credit or amount obtained across the size classes of landholdings. In
this context, the farmers' access to institutional credit is estimated
using the binary choice (Pro bit) model. The factors determining the
amount of institutional credit lent were estimated with the help of
Ordinary Least Squares (OLS) methods. However, the factors
determining access to the credit to certain farmers is not random. In
order to correct this problem of selectivity bias, Heckman two-step
procedure has been used. In the first step, the Inverse Mills Ratio (A)
has been calculated from the Pro bit equation, and in the second step,
this has been introduced as an additional explanatory variable in the
eqllation on determinants of credit.
The credit delivery system and the extent of credit-gap for
different categories of farmers have been discussed in the seventh
chapter. Given the self-finance capacity and per acre paid-out cost,
the actual and relative credit-gaps have been calculated across the
size-classes of landholdings. The coping strategies adopted by farmers
in the study area have also been highlighted.
The last chapter summanses the study, and discusses
conclusions and policy implications.
16
Chapter 2
POLICIES, PROGRESS AND PROBLEMS IN
AGRICULTURAL CREDIT: THEORETICAL ISSUES
Introduction
There is a realisation In most of the low-income countries that
agricultural credit is an important element in their development
efforts. The popularity of credit is partly due to the notion that
loans are necessary to accelerate technological change in farming,
and that formal credit is required to free farmers from their
dependency on unorganised sector. The agricultural credit policies,
in general, aim to have positive influence on the total volume of
institutional credit, the use of agricultural inputs, investment on
machinery and irrigation, agricultural output and productivity,
rural income distribution and so on. These policies also aim to
direct a larger share of institutional credit to the 'weaker' sections
for their development.
The declared objectives of the public policy with regard to
rural credit, in general, and agricultural credit, in particular, in the
post-Independent India were, therefore, "to ensure that sufficient
and timely credit at reasonable rate of interest is made available to
as large a segment of the rural population as possible" (Rangarajan
1996). To achieve this objective, the expansion of formal sector
lending institutions, directed lending and subsidised credit policies
were introduced at different points of time. A multi-agency
approach was adopted as a national policy for institutional credit
where co-operatives, Commercial Banks and Regional Rural Banks
were pressed into service to ensure sufficient agricultural and rural
credit flow. Taking different types of farm credit needs into account,
17
the provision of a range of credit services (long and short-term, and
large and small loans) was envisaged in the policies.
This chapter discusses the major policy directions, progress
and problems of institutional finance for agriculture. For this
purpose, the period 1951-52 to 2000 has been divided into three
sub-periods, viz., 1951-52 to 1969-70, 1970-71 to 1989-91 and
1991-92 to 1999-2000. The rationale for sub-periodisation has been
the nationalisation of commercial banks in 1969 and the
introduction of banking sector reforms in 1991.
Phases of Banking Policies
The Pre-Nationalisation Period (1951-52 to 1969-70)
During 1950s, co-operatives and government were the mam
providers of institutional finance to agriculture, though they
accounted for a small share of the total credit (Dandeker and Wadia
1989; Katula and Gulati 1992). According to the Report of the All
India Rural Credit Survey Committee (1954) appointed by the RBI,
out of the total amount borrowed by the cultivators from different
credit agencies (both formal and informal), formal agenCIes
accounted for only 7.3 per cent. Of this, the commercial banks
provided only 0.9 per cent, while government and co-operatives
accounted for the rest. This meant that as much as 92.7 per cent of
the credit was provided by the informal sector consisting of
moneylenders, relatives, big farmers and traders.
The committee examined the rural credit delivery system m
the light of the question on how to ensure an efficient supply of
institutional credit to increase the productivity of the agricultural
18
sector - a dominant sector in the national economy. Taking the
importance of credit in achieving the targets on agricultural
production into account, the committee recommended new
initiatives like State partnership at different levels, and co-ordination
between credit and other economic activities, especially processing
and marketing. Increased financial support from the government to
co-operatives and RBI, and technical support to the co-operatives
credit structure were the other important recommendations.
Besides, financial institutions were designed at different levels to
facilitate the credit requirement of farmers to a greater extent. The
state level co-operative credit institutions consisted of State Co-
operative Bank, Central Land Mortgage Bank and State Co-operative
Marketing Society in each State. The institutions of District Centr2-!
Co-operative Bank, Primary Land Mortgage Bank and District
Marketing Society were formed at the district level for better
allocation of agricultural credit. Primary Agricultural Credit Societies
(PACSs) and Large Area Multi Purpose Societies (LAMPs) were the
primary level co-operatives. The committee also recommended the
amalgamation of the Imperial Bank of India into the State Bank of
India to extend banking facilities in the countryside. In 1963, the
Agricultural Refinance Corporation (ARC) was established in the RBI
with a view to supplement long-term resources to credit institutions.
These policies heralded a new phase in agricultural credit. During
this period, the share of credit to rural households provided by the
financial institutions - co-operatives, banks, and government rose
from 7.2 per cent to 29.2 per cent (Katula and Gulati 1992).
19
Table 2.1: Distribution of Total Debt of Rural Households by
Different Sources
(in percenta e
Sources 1951 1961 I 1971
I

Government, Co-operatives and Banks 7. 2 17. 3 I 29.2 I 6 J. 2
Relatives and Friends Il. 5 5. 8: 13. 8 : 9.0.
Moneylenders 80. 2 69.4 ! 54.2 i 24.
3'
Others l. I 7.5: 2.8: 5 5 I
Total credit fRs. crores at 1971-72 prices) NA 36,100 I 37,541 I 23,361 I
Notes: NA - Not aVaIlable.
Source: Katula and Gulati (1992).
s)
A study by the RBI on the role of co-operative credit in
increasing farm production in 12 districts during the period 1963-64
to 1965-66, however, found that:
;,. the share of credit provided by the co-operatives to total credit
was small to moderate in most of the districts;
;,. a cnn hlp nf thp ,...,...,D,.l'1- ........ ..4 +.,...
- - --"-- -- .r-'............ ....... ............ ........ .............. J.\. uJ.,.. ........ L\""U LV r "'"
purposes;
;,. the co-operative credit was found to be untimely, inadequate and
ineffective in increasing farm production; and
;,. the central co-operative bank and the PACSs in several states
remained stagnant in credit growth because of growing overdues,
and that PACS were not designed to mobilise the rural savmgs,
considered to be an important source of loanable funds.
In July 1966, the RBI constituted an All India Rural Credit
Review Committee to undertake a comprehensive revIew of
agricultural credit system so that the bottleneck could be identified
and remedied. The report, published in July 1969, found that a
number of banks and societies were weak in mobilising deposits,
recovering loans and management. The remedial measures
suggested were rehabilitation of weak branches, administrative and
policy measures for checking overdues, direct financing of cultivators
by central banks and of societies by apex banks in areas where they
were weak. The committee also suggested streamlining the lending
policies and procedures of co-operative institutions to make credit
20
more easily accessible to the small farmers according to scales of
finance needed for different crops.
Notwithstanding these efforts, the objective of meeting the
credit needs of farmers was hardly fulfilled (Dandeker and Wadia
1989). The share of public sector banks in the total agricultural
credit was 5.4 per cent of net bank credit in 1969 (RBI 1997-98) due
to their concentration in urban areas, and non-suitability of
repayment period and security to their requirement (Shajahan
1998).
Agriculture Sector Lending as a Major Policy Instrument
(1969-91)
This 'Nas the early phase of the green revolution in India. The advent
of new technology and the introduction of High Yielding Varieties
(HYV) in 1966 resulted in a greater demand for improved seeds,
fertilisers and increased application of irrigation water. This
contributed to a substantial increase in the demand for credit, and
the supply of adequate credit was considerably beyond the capacity
of the co-operatives. Commercial banks were, therefore, introduced
to a greater extent to provide agricultural credit under the policy of
Social Control over banks in 1967. The entry of commercial banks in
the sphere of agricultural credit was intensified with the
nationalisation of 14 major commercial banks in 1969 and 6 more
banks later.
The declared objectives of the new policy, known as "social and
development banking" were to provide (Wiggins and Rajendran
1987);
21
banking services in previously unbanked or under-banked rural
areas;
substantial credit to specific activities including agriculture and
cottage industries; and
credit to certain disadvantaged groups.
The Government of India and the Reserve Bank of India (RBI)
issued specific directives regarding "social and development
banking" from time to time. These policies included the setting of
targets for the expansion of rural branches, imposing ceiling on
interest rates, and setting guidelines for the sectoral allocation of
credit. During the early part of bank nationalisation, the National
Credit Council under the chairmanship of D.R. Gadgil had
considered the deficiencies in the system of institutional credit and
formulated the lending policies for the future. Specific lending
targets were fixed for different sectors in terms of amounts to each
major bank. In this context, two major steps were taken by the RBI
to encourage the priority sector lending such as provision of
refinance to commercial banks, and introduction of credit guarantee
scheme as a support measure of bank lending.
With the establishment of Regional Rural Banks in 1975,
another wing to the agricultural credit was added. RRBs were
designed to be low cost institutions like co-operatives and
professionalised in their management like commercial banks. Their
target group was the 'weaker sections' consisting of small and
marginal farmers, agricultural labourers and rural artisans, and
their area of operation was confined to specified districts.
After nationalisation, lending to priority sectors became an
essential component of bank lending. During the mid-1970s, the RBI
study group recommended expansion in the scope of priority sector
lending to include not only agriculture (direct and indirect), small-
scale industries, and export, but also road and water transport
operators, retail traders, small business, professional and self-
employed persons and education. This definition remained
unchanged until the end of 1970s. For better provisioning of credit,
the RBI advised the banks to raise the proportion of credit to the
priority sector to total net bank credit to 33.3 per cent by March
1979. In 1980, a working group set up under the chairmanship of
K.S. Krishnaswamy to review the composition of the priority sector,
included weaker sections and excluded export related lending from
the priority sector. A major recommendation of the group was that at
least 50 per cent of direct farm credit should go to weaker sections
including marginal farmers, small farmers and agricultural
labourers, or 40 per cent of the priority sector bank credit should be
given to agriculture by March 1983.
The Committee to Review Arrangements for Institutional Credit
for Agriculture and Rural Development (CRAFICARD), 1981,
endorsed that commercial banks could playa significant role in the
various programmes of rural development, and suggested measures
to improve the quality of lending through these banks. The National
Bank for Agriculture and Rural Development (NABARD) was,
therefore, set up in 1982 by the RBI for providing all types of
production and investment credit for agriculture and rural
development. NABARD is the apex institution accredited \\'ith all
matters concerning policy, planning and operations in the field of
agriculture and other economic activities in rural areas. Its prime
role is to provide credit for the promotion of agriculture, small-scale
industries, cottage and village industries, handicrafts and other
rural crafts and allied economic activities in rural areas with a view
to promoting integrated rural development and securing prosperitv
for the rural areas. Other functions include co-ordination of rural
financing activities of all institutional works at the field level and
maintaining liaison with the Government of India, State
governments, RBI and other national level institutions.
Another RBI working group under the chairmanship of A.
Ghosh was set up to review the target of priority sector lending, in
general, and agriculture, in particular, with reference to the nature
of advances. The committee suggested a separate target for both
direct and indirect farm lending from total bank credit. The direct
farm lending was fixed to reach 14 per cent of net bank credit by
March-1985. By taking the increasing need of credit for farming
operations into account, it was recommended to raise the proportion
of direct farm lending to net bank credit to 16 per cent by March
1987,17 per cent by 1989 and to 18 per cent by March 1990 (Table
2.2). Furthermore, this group provided a comprehensive redefinition
of the "weaker sections" and suggested achieving a target of either 10
per cent of net bank credit or 25 per cent of total priority sectors
lending by March-1985. Besides, total priority sector lending was
increased to 40 per cent of the net bank credit by 1985. The lending
targets for different kinds of agricultural advances are given below in
Table 2.2.
Table 2.2: Lending Targets or Agrtcu ture Sector. u __
of Advance Target from NetlWhen Target is to be :
I
Bank Credit (in Per cent) i Reached ___ !
Direct advance for 16
,

agriculture and allied 17 March- 1989
activities 18 March-1990
-j
Direct + Indirect advance 18
I
for agricu Iture and allied of which direct and indirect October-1993 I
activities finance target were fixed at
13.5 % and 4.5 %
respectively
I
Source: RBI 1987-88 and 1993-94
The Reform Period (since 1991)
Preoccupation to achieve quantitative targets, ignoring qualitative
aspects, high cost structure of operations in rural areas, and
mounting overdues, became threat to the viability of financial
institutions. Consequently, the policy of competitive financial svstem
was adopted in provisioning agricultural credit to improve the viability
of financial agencies. This phase of policy started from 1991,
particularly after liberalisation. The policy thrust of this phase can be
found in the report of the Committee on the Financial System (CFS)
chaired by M. Narasimham. In its very first paragraph, the report calls
for a vibrant and competitive financial system to sustain the ongoing
reform in the structural aspects of the economy. The committee
suggests the use of fiscal instrument for redistributive objective rather
than the credit system. Based on this, the committee recommends
that directed credit programmes be phased out. The proposals of the
committee were deregulation of interest rates, changing capital
adequacy norms (to compete with banks globally), and removing
branch licensing policy. Besides, the committee took the view that a
new institutional structure, I.e., market driven and based on
profitability be created, and that the role played by private Indian and
foreign banks need to be enlarged. In short, the Narasimham
25
Committee recommended that banking policy be guided more by the
market force than by regulations of public authority.
However, the government did not favour the abolition of
directed lending, delinking of the rural branches of public sector
commercial banks and merging them into rural banking subsidiaries,
or even the creation of National Rural Banks except gradual
deregulation of lending rate. In fact, the strategy followed by the RBI
was to keep the overall share of priority sector at 40 per cent of net
bank credit, and broaden the scope of priority sector lending by public
sector bank to encourage diversion from direct priority sector lending
to the new areas added to this sector. The whole idea was to allow the
banks to fulfil the target of priority sector, in general, flnrl
in particular, from the total advances by banks, without having to
lend directly much more to those areas included in this sector in the
early 1990s (Shajahan 1999).
Although deregulation of interest rate and other measures were
recommended by CFS to improve efficiency, profitability and viability
of the banking sector, the committee was silent on the quality of
lending to achieve the above objective. The Gupta Committee (1998)
looked into the quality of credit delivered by the banking system to
identify the constraints faced by the commercial banks in increasing
the flow of credit, introducing new products and services, and
simplifying procedures to enable rural borrowers to access adequate
and timely agricultural credit from the commercial banks.
The committee emphasised on pre-sanction appraisal to assess
the borrower status, capability for undertaking the proposed activity,
26
creditworthiness and the technical viability of the proposal. The
committee suggested freedom to bank branches in the selection of
beneficiaries and disposal of applications. To ensure quick disposal at
least 90 per cent of the loan applications should be decided at the
branch level. Short-term credit needs of the farmer should include all
requirements directly or indirectly related to production, post-harvest
and household expenses. Repayment capacity should be assessed on
the basis of aggregate household income from all sources including
crop production and ancillary activities. Where the land has already
been mortgaged, asking for additional collateral by way of guarantors
should be discouraged. Commercial banks should be free to fix the
rates of interest for loans of all amounts as has already been done in
the case of co-operatives and RRBs. To overcome the credit gap
problem, the rate of finance should be fixed for service area of each
bank on the basis of local conditions rather than for the district as a
whole. Major recommendations of the Gupta committee accepted by
the RBI are given in Box 2.1.
Box 2.1: Recommendations of the Gupta Committee
Simplification of procedures regarding loan application, agreements/
documents, etc;
Rationalisation of internal return of banks;
Delegation of power to branch managers;
Introduction of composite cash credit limit to agricultural farmers;
Introduction of new loan products with saving component;
Cash disbursement of loans;
Dispensation of No Dues Certificate; and
Discretion of the bank on matters relating to margin/ security requirements
for agricultural loans above Rs 10,000.
Nevertheless, with a view to making credit effective and to
promote agriculture and rural development, the RBI continued to take
new initiatives and reorient its credit policies. Among the other major
policies, Potential Linked Credit Plan (pLCP), relaxation in refinance
eligibility condition to State Co-operative Banks and RRBs,
27
introduction of cash credit and deregulation of interest rates were
prominent. The basic objective of PLCP (initiated by NABARD) is to
explore the existing potential and work out an appropriate credit plan
through which such potential could be exploited to the maximum
extent possible over a specified period of time. The credit estimates
made under this plan were expected to be revised with respect to
changing economic scenario, technological improvement and creation
of further infrastructure and marketing facilities. To achieve greater
efficiency in the ground level credit flow, a background paper goes to
the concerned branch manager/lead bank officer (LBO) for preparing
their Service Area Plans (SAPs).
State Co-operative Banks and RRBs, which lost their eligibility
to obtain refinance facility from NABARD on account of failure to
recover at least 40 per cent of the demand of previous years, in order
not to deny fresh credit to new and non-defaulting members of co-
operatives and RRBs this condition was relaxed in December 1993.
The NABARD agreed to provide refinance to the extent of their
recovery (RBI 1993-94). Furthermore, with a view to ensuring
adequate and timely flow of rural credit as to meet the composite
needs of farmers, the banks were advised to extend cash credit
facilities to farmers with irrigation facilities and also to other farmers
undertaking off-farm/ allied activities, but this applied strictly to
those farmers having a good track record in repayment.
28
Rationalisation of interest rate (rate to the ultimate borrowers
7
)
IS the other important aspect of policy changes in the reform phase.
The process of rationalisation of the interest rate for rural credit, in
general, and agricultural credit, in particular, was initiated by the RBI
in September 1990 based on the recommendations of the Narasimham
Committee on deregulation of interest rates and abolition of
concessional lending rates so that all rates became positive in real
terms. The details of this and the official reaction to this
recommendation are summarised below:
(i) The recommendation that the real rates should be positive IS
accepted.
(ii) The lending rates have heen mtionalised with a reduction in the
number of concessional slabs and enhancement of some rates,
thereby reducing the element of subsidy.
(iii) Since there is inadequate margin between the cost of funds and
lending rates (to cover the cost of transactions and risk), there is
need for evolving a two-slab structure of lending rates. For the
slab of credit limit above Rs. 2 lakhs, the interest rate should be
freely determined by the market, and one concessional interest
for the credit limit up to Rs. 2 lakhs could be 2 to 3 percentage
point below the Prime Lending Rate (PLR). This may lead to some
increasing in the interest rate for small loans. But the committee
suggests that what is important is the provision of timely and
adequate credit.
1 The state intervention in interest rate policies has been discussed with the
information collected from the RBI's Report 011 Currency and FlIlance Ivarious
issues). For a detailed discussion of offlcial reaction on deregulation of interest
rates as recommended by Narasimham Committee 1991, see, RBI 11991-92), and
Gadgil 1997.
29
(iv) The credit co-operatives, Commercial Banks and RRBs have all
sustained large financial losses in rural lending, mainly because
of high administrative costs, political interference, heavy overducs
and a non-viable interest structure. Given the nature of activity of
the co-operative banks, it was decided that the co-operative credit
structure should be totally freed from deposit/lending rate
control so that these rates could settle at realistic level.
(v) Accordingly, lending and deposit rates for all co-operative banks,
excluding Urban Co-operative Banks
8
, were totally deregulated
from October 1994 subject to a minimum lending rate of 12 per
cent. RRB lending rates were also deregulated from August 1996.
(vi) Lending rates of scheduled commercial banks were freed for
credit limits of over Rs. 2 lakhs. The prescription of minimum
lending rate was abolished and banks were advised to obtain the
approval of their respective Boards for the PLR which would be
the minimum rate charged by the banks for credit limit of over
Rs. 2 lakhs, and required to apply uniformly in all their branches.
(vii) For credit limit of over Rs. 25,000 and up to Rs. 2 lakhs for all
advances including term loans, a uniform lending rate of 13.5 per
cent per annum was prescribed. Effective from October 1997,
instead of a uniform lending rate, it was decided to charge a
maximum rate of 13.5 per cent per annum (RBI 1996-97).
B From June 1995, the lending rates of Urban Co-operative banks were also
deregulated subject to a minimum of 13 per cent per annum.
Thus, there was a gradual deregulation of interest rate to the
ultimate borrowers since September 1990. The deregulation of
interest rate and their likely impact have been discussed in the last
section of this chapter.
Movement of Directed Credit in India
In India, directed credit policies have been used for promoting
agriculture and small-scale industries following the nationalisation
of commercial banks in 1969 (Kohli 1997). Subsequently, banks
were directed to lend 40 per cent of their net bank credit to priority
sector. Under priority sector-lending, agriculture was targeted to
facilitate the adaptation of new technology to increase agricultural
productivity. It is also argued that agriculture was targeted beolllsf'
it was a risky activity and households were credit rationed by the
formal sector (Swaminathan 1991; Kochar 1997). Hence, the
commercial and co-operative banks were directed to expand their
rural branch networks and to intensify their lending to agriculture
during the post nationalisation periods.
However, India's directed rural credit programme was not a
success in all respects (Naastepad 2001). Ramachandran and
Swaminathan (2001) were also of the view that although advances in
the countryside increased substantially, such an increase was
uneven, as was the case with green revolution, across regions, crops
and classes. The households with an asset holding of less than Rs
10,000 depended on private agencies for 67 to 90 per cent of their
credit needs (Dandekar and Wadia 1989). It is also found that,
although moneylenders became less important after independence
due to government intervention in the agricultural credit market
31
through directed credit policies, they still played an important role
largely due to the poor quality of institutional credit (Rajasekhar and
Vyasulu 1990). Bell (1990) notes that the RBI had exaggerated the
erosion of the moneylender's power in the unorganised credit
market. This meant that financial institutions did not help small and
marginal farmers, landless labourers for whom they were primarily
set up.
It has been argued that the low lending rates set for
agriculture as compared to commercial and industrial rates is one of
the important factors for poor allocation, and distribution of credit to
this sector. The commercial banks have to cover the cost of their
agricultural loan from profit arising out of other operations and
cross-subsidise the agriculture sector (Katula and Gulati 1992; Rao
1994; Binswanger and Khandker 1995). A handful of literature on
directed credit policies also support the ineffectiveness in allocating
9
a larger share of formal loans to agriculture, in general, and the
rural poor, in particular, because of risks, return and costs of doing
so are unattractive to formal lenders (Lad man and Adams 1978; Fry
1979; Vogel and Larson 1980).
Furthermore, because of asymmetry of information between
lender and borrowers, equilibrium credit rationing is an inherent
feature of financial market (Stiglitz and Weiss 1981; Cho 1986;
Gibson and Tsakalotos 1994). Such rationing is inevitable in
situations where banks can distinguish between groups of borrowers
9 It has been also argued that operations of Rural Financial Markets (RFMs) in
most countries are resulting in inefficient allocation of resources, causing
income and asset ownership concentration, allowing financial resources to flow
out of low income areas and, in some cases, diverting resources out of
agriculture (Adams and Tommy 1974; Vogel 1977; Araujo and Me\'er 1978).
"
.'-
(for example, on the basis of farm/ firm Size or sector), but cannot
distinguish between individual (good or bad) borrower within a
group. Besides, banks cannot filter the borrowers with high risk
without incurring costs. It is also difficult for the banks to use the
interest rate as screening device due to adverse selection. This is due
to the fact that borrowers, willing to pay high interest rates, may be
less worried about repayment of the loan. Under these
circumstances, a profit maximising bank will practice credit
rationing and be reluctant to increase interest rates in response to
an excess demand (Stiglitz and Weiss 1981). Consequently, high
risky group borrower may be completely excluded from the market
although their prospective investments offer a high-expected return.
Since agriculture is a risky activity where farmers face s h o ~ k s to
their income, bankers are practising a tight credit rationing process,
which pre-empts farmers from accessing credit (Besley 1994;
Binswanger and Deininger 1997).
The subsidised agricultural credit programmes and its impact
on the performance of RFMs came under attack
lO
as early as the
mid-1970s (Helfand 2001). The literature that emerged in early
1980s focused on issues of imperfect information, adverse selection,
moral hazard, market failures, and interlinked transactions (Stiglitz
and Weiss 1981; Bardhan 1984; Hoff and Stiglitz 1993; Besley
1994). This literature focused on informal credit in part due to the
demise of cheap formal credit programme (Helfand 2001).
10 The main criticisms were allocative inefficiencies, concentration of income and
wealth in a particular class of people, and failing to achieve their objectives of
promoting agricultural modernisation. increasing agricultural output and
redistribution of income in favour of the rural poor. For more details, see, Shaw
1973; Donald 1976; Adams and Graham 1981.
33
In the directed credit policies, importance was given to th('
outreach by neglecting the quality and cost of lending. Because of
this, the formal banking system began to face the problems of higher
transaction costs (due to lending small amounts to a large number of
borrowers). The disbursal of loans without assessing the feasibility,
viability, and entrepreneurial experience of borrowers led the banks
to incur losses due to the widespread problem of fungibility and poor
repayment of the loan. Hence, notwithstanding the empirical support
for the positive growth effects of directed credit in both agriculture
and small-scale industry, the focus of India's recent financial
reforms is on the phasing-out of its directed credit programme
(Shajahan 1999; Naastepad 2001).
The financial reform (Committee on the Financial System
1991) was primarily concerned with the decline in productivity and
efficiency of the banking sector. The report points out that directed
credit had negative effects on the income of banks through
concessional interest rates and high administrative costs of such
loans. To improve efficiency, the Committee recommended reducing
directed credit from the prevailing 40 per cent to 10 per cent of net
bank credit over a three-year period. Because of strong opposition
from farmers and small-scale industry and political lobby, the
government has not officially accepted this recommendation.
Nevertheless, the priority sector lending was reduced to 33 per cent
of gross bank credit (Naastepad 2001). It is also argued that the
performance of the directed credit programme has been improved
significantly by reducing interest rate subsidies (on loans over Rs.
200,000) and giving greater emphasis on lending along commercial
lines to reduce non-performing loans (Sarkar and Agrawal 1997).
Therefore, it can be argued that the policy to increase the efficiency
of banking sector may have direct as well as indirect effect on the
rural economy. In other words, how far the efficiency of the banking
sector has been taken into account to increase the credit flow to
agriculture, employment, and income distribution needs to be
examined
Importantly, despite increasing consensus to extend credit
facility to agriculture, in general, and small and marginal farmer, in
particular, banks were unable to fund those activities with high social
return or those categories of credit-worthy borrowers who had been
marginalised in credit markets (Stiglitz 1994). It is also argued that
the unusually low setting of interest rates on credit-saving (much
lower than the opportunity cost of capital) limited the access of small
and marginal farmers to institutional credit since these low rates lead
the lenders to select only borrowers with excellent credit ratings
(Kamajou 1980). So, given these problems what makes certain
households to access institutional credit and others do not? This issue
will be discussed 111 detail at the household level analysis.
Nevertheless, the ineffectiveness of directed agricultural credit system
in terms of allocation have been discussed in greater detail more
deeply in the subsequent section with respect to interest rate policy,
cost of lending, and the process of credit rationing.
Subsidised Interest Rate and Its Impact on Agricultural Credit
Flow
Interest rate plays an important role in determining the supph'
of agricultural credit. In general, the interest rate on agricultural loans
is kept low to promote agricultural growth and to assist the rural poor
in developing countries. Several decades of experience on the impact
35
. \ l_
of low interest rates indicates that cheap loans did not appear to have
either increased agricultural output or reached the rural poor
l
I
(Adams and Vogel 1986). Gonzalez-Vega (1984), and Ladmarr and
Tinnermeier (1981) argue that subsidised interest rates are the major
contributing factors for fungibility on the part of borrowers and credit
diversion on the part of lenders, low lender revenue and pohtical
intervention into the credit market.
Thus, if the interest income is greater than or at least equal to
the cost of lending (break-even condition 12), this may positively
influence the financial institutions ll1 increasing the supply of
agricultural credit. However, default rate plays an important role ll1
determining the break-even level of interest rate for a financial
11 The question of why formal credit is used to a lesser extent by small and potential
borrowers is adequately explained in the literature. McKinon (1973) and Shaw
(1973) are of the view that the ceilings on interest rates do not permit banks to
incorporate the additional administrative costs that are involved in advancing
small loans and supervise them. The formal lenders basically ration credit to the
small borrowers in order to reduce their transaction costs (Gonzalez-Vaga 1984;
Anderson and Khambata 1985) which are high for servicing small borrowers.
Consequently, banks advance loans to those who offer lower risk and better
security. Sarap (1991) is of the view that a typical borrower in unorganised credit
market has no access to organised credit market because of the collateral he
offers is not acceptable in the organised credit market. Credit is invariably
rationed in terms of the ability to offer collateral (Rudra 1982; Binswanger and
Sillers 1983). Lele (1981) is of the view that credit worthiness criterion adopted by
formal credit institutions is alienating small farmers from borrowing.
12 (Income from loan portfolio)" (Cost of lending)
(1 + r) I (I - PJ) X
J
" I (I + i + a
J
) XJ ----------------- (*)
where X, ~ size of each loan
i ~ interest paid per unit of principal on borrowing and deposits
a ~ I a, X, I L X,
a
J
~ administrative cost per unit of capital
r ~ lending interest rate
p ~ I P, X, I I X, (expected default rate on loan)
PJ ~ per unit loss of principal due to default on a loan size for Xj
B, substituting the above values in the equation (') we can obtain
r" (i +a + p) I (l - p) ----------------- (break-even condition of intere,t rate)
For more details, see, Anderson and Khambata (1985) and Hulme and Mosel\
(1996).
36
institution. Higher the default rates, higher the break-even interest
rate that has to be charged. But, given the administrative interest
rate, the banks cannot increase their lending rates irrespective of the
default rate. Therefore, the banks will try to bring down the break-
even condition of lending rate to the economic criterion 13 of rate of
interest by reducing the default rate and the cost of lending. In order
to reduce the default rate, banks can adopt the process of credit
rationing
l4
. Furthermore, if there is rationing, there would exist an
excess demand for credit than the size of loan sanctioned at the profit
maximising rate. Even in the absence of interest rates restriction, as
the probability of default risk is more due to imperfect information in
the rural credit market, non-price credit rationing 15 is widely accepted
in most of th" Lors (Gonzalez-Vega 1984). Besides, low interest rate
leads to low return and poor supervision, which may cause overdue
and more risk to the financial institutions. To avoid these problems,
financial institutions may prefer to concentrate on a few selected
13 The marginal efficiency of capital must be greater than or at least equal to the rate
of return on lending (i.e., lending rate of interest). The economic benefit of
investment on loans depends on the amount repaid by the borrowers. For more
details, see, Anderson and Khambata (1985).
14 It refers to a situation where at the going rate of interest in the credit transaction,
the borrower likes to borrow more money but it is not permitted by the lender
(Basu 1997; Ray 1998). Fried and Howitt (1980) are of the view that credit
rationing exists as a part of an equilibrium risk sharing arrangement between
Formal Financial Institutions (F'Flsj and the borrower. Credit is invariably
rationed in terms of ability to offer collateral (Rudra 1982; Binswanger and Sillers
1983; Von Pischke et al 1983; and Sarap 1991). For a more detailed discussion
on the impact of interest rate ceiling and credit rationing, see, Gonzalez-Vega
(1984), Anderson and Khambata (1985), and Basu (1997). Eaton and Gersovitz
(1981) have suggested a model with endogenous default costs where default leads
to exclusion from the credit market. Given this. a contract is enforceable provided
that each payment is less than the value of future access from the credit market.
15 In the case of non-price credit rationing. the potential borrower is willing to pay
the full price (i.e., non-interest term and conditions of the loan contract with
interest rate charged), but the formal financial institutions are not willing to grant
the loan of the size demanded. For more details, see, Jafee (1971 J, and Gonzalez-
Vega (1984).
37
borrowers, those who have creditworthiness with excellent loan
collateral and those who take large loans. This suggests that a ceiling
on lending rate has two kinds of impact on the institutional credit
market. The first type of impact can be observed in the flow of credit
and the second one can be observed in access to credit by different
category borrowers.
The interest rate ceiling seems to have had similar impact in
India. The provision of cheap credit to agriculture in India gives rise to
two kinds of credit subsidyl6. First, the interest subsidy accrues to
agriculture because the rate of lending to this sector is lower
compared to the other sectors of the economy. The second type of
credit subsidy is the difference between the CO!lt of !lupplying credit to
the agricultural sector, including defaults and rate of interest received
from outstanding credit. Using the above conceptual framework,
Katula and Gulati's (1992) study reveals that concessional interest
rates automatically result in cross subsidisation. They are of the view
that of the three types of banks in India, i.e., Commercial Banks
(CBs), RRBs and Co-operatives, only CBs have the scope for cross-
subsidisation. It happens in the case of CBs because they lend to a
variety of sectors and have the possibility to lend to those activities
that earn interest of more than 16 per cent. Besides, they have also
estimated that during the 1980s, the total credit subsidy to
agriculture was 7.36 per cent of the outstanding. This implies that
7.36 per cent of the total loanable funds could not come back to the
banking system, and to that extent the quantum of funds available to
formal banks came down. This is further corroborated by Bishnoi's
(1991) argument that interest subsidy to priority sectors compared to
16 For a detailed discussion, see, Katula and Gulati (1992).
38
the industrial sector has been the major contributing factor for lower
credit flow to the agriculture sector.
Can financial institutions go for higher lending rate to get a
good margin out of their deployment of funds to the agriculture
sector? The Agricultural Credit Review Committee headed by Khusro
(1989) and the Committee on Financial System (1991) had
recommended enhancement of lending rates to cover risk and
additional costs associated with it. In this context, a study made bv
Gadgil (1992, 1994) reveals that, by and large, farmers could afford
higher interest rates if accompanied by an improvement in the quality
of lending.
In theory, however, a question was raised on whether one could
take higher lending rate as the criterion of the profitability. Stiglitz
and Weiss (1981) argue that increasing interest rates or increasing
collateral requirements could increase the riskiness of the bank's loan
portfolio, either by discouraging safer investors or by inducing
borrowers to invest in a riskier projects, and therefore, could decrease
the bank's profit. This implies that the bank may incur more loss with
the increased lending rate and it may adversely affect the viability of
the system. It may also happen that many good borrowers may move
out from the formal credit market with increasing lending rate.
Making a similar point on viability, Basu (1997) notes that it is the
default risk, rather than the ceiling on interest rate, which IS the
major contributing factor of non-profitability and non-viability.
Ceiling on Interest Rates and Access to Institutional Credit
Gonzalez-Vega (1984) has developed a conceptual framework on credit
rationing and their impact on access to institutional credit with the
39
help of non-price credit rationing by assuming a binding ceiling on
interest rate is imposed on the banks. The Iron Law of Interest Rate
Restrictiolls states that, when interest rate ceiling becomes more
restrictive, the size of the loan granted to non-rationed borro\\er
classes increases, while the size of loan granted to the rationed
borrower classes diminishes (Gonzalez-Vega 1984). The Iron Law of
Interest Rate Restriction assumes that demand for credit is inversely
related to the real rate of interest and the lender's marginal cost
increases with loan size (Figure 2.1).
Figure 2.1 shows that when rationing IS taking place, the
marginal cost of the loan becomes equal to the ceiling interest rate.
Hence, depending on the relative level of ceiling with respect to
marginal cost of lending, some or all of the borrowers may be
subjected to non-price credit rationing. Figure 2.1 also shows that, for
the rationed borrower, credit is available along the lender's marginal
cost curve whereas for non-rationed borrower along with his credit
demand curve.
Figure 2.1: Rationing Behaviour oftbe Lender
r
MC.
(Non-rationed borrower)
r*
MC2 r
Credit) D,
Source: Gonzalez-Vega (1984)
HI
(Rationed borrower)
D, (Credit)
From the above theoretical model, one can see that ceiling on
interest rate and cost of lending have serious implication on access to
institutional credit. Given the risks and transaction costs associated
with lending, most of the formal financial institutions try to optimise
the loan size to each class of borrowers by credit rationing devices.
Thus, when credit rationing takes place, two things can happen. First,
financial institutions may concentrate on a select few, those who have
credit worthiness with excellent loan collateral. Second, since access
to credit and loan sizes are highly correlated with income level and
assets, large farmers and the well-to-do benefit most from the cheap
credit (Lipton 1976; Lele 1981; Adams and Vogel 1986; Braverman
and Guasch 1986).
The above discussion focuses on the poor access to
institutional credit by some category of agricultural borrowers from
the supply side point of view. However, the analysis of availability and
access to institutional credit by the borrower may not be complete
without taking into account the demand side problems. It is often
argued that small and medium borrowers are indifferent as to whether
to seek a formal or informal loan because they feel that their total cost
of acquiring a loan from either source are similar (Adams and Nehman
1979). In this context, it is essential to focus the factors, which
influence the demand for institutional credit.
Borrowing Costs and Demand for Credit
The literature on formal credit contains several explanations for the
limited demand of formal credit by small and potential borrowers. The
literature focuses that most poor do not seek formal credit due to lack
of profitable investment opportunities and poor idea of credit use.
41
Schrieder (2000) argues that many poor do not approach for credit
from institutional credit agencies as they value the risk associated
with indebtedness to be higher than the possible return on
investment. The other studies (Adams and Nehman 1979; Aron 1981;
Timberg and Aiyar 1984; Sarap 1990) focus on borrowing costsl' and
borrower's behaviour on demand for credit. They argue that higher
borrowing costs incurred by small potential borrowers as compared to
large borrowers discourage them to approach the formal credit
institutions.
Ladman (1984) provides a theoretical framework on demand for
credit based on the relationship between average borrowing costs
(ABC) and average revenue (AR) of debt financed invest.!nent. Both AR
and ABC incurred by the borrower represent revenues and borrowing
costs divided by the size of the loan (Ll utilised for a particular
purpose. He explains that, with a given set of technology, the farmer,
who relies upon credit, derives demand schedule of loan from the
value of marginal product by using successive loan units. It implies
that,
BC = Total fixed transaction costs (Tl) + Other transaction costs
(T2) and thus, ABC = (TI + T2)/ L
Tl represents the amount of outlay the farmer must make in applying
for a loan. The payment for application fees, service fees, documents
and other paperwork are examples ofT\. The cost involved in payment
for commission and bribe to offlcer / technician/ bank staff, leader or
17 To obtain a loan a farmer has to go through the procedures that are required by
the lender's credit delivery system, which result in borrowing costs. The
borrowing costs (Be) by the prospective borrower may include nominal interest
payments made to the lender, additional loan transaction costs, and the changes
in the purchasing power of the money over the loan period. For more details, see,
Adams and Nehman (1979).
42
middleman for negotiation, travel to and from the financial institution
and the opportunity cost of time involved are examples of T2.
Since loan procedures and paperwork are independent, the
average fixed cost declines with loan size. The borrowing cost also
depends on the frequency of meeting bank staff, information of
borrower's status on efftciency of funds investment, credit worthiness
and the number of times that a farm household obtained loan from
the institutional sector. For large and experienced borrowers, the
transaction costs decline. These loan transaction costs appear to be
much less important for large size borrowers. However, loan
transaction costs appear to make up a very large part of borrowing
costs for many small and medium si7.p horrowprs (Adl'lms and Nphman
01 - - - - ,- - - ----- - --
1979). Since total borrowing cost is the summation of interest
payment (r) and transaction cost, ABC declines with respect to the size
of loan.
Let us assume that the borrower is a profit (n) maxlmlser and
thus seeks a loan if he expects n > 0 i.e. AR > ABC. Thus, the borrower
would borrow until the point where marginal cost of borrowing
(additional interest payment) is equal to marginal value product (MVP)
from additional resources purchased with borrowed funds (Figure
2.2). Given the interest rates r, the vertical distance between rand
ABC represent transaction cost for any size of loan. Thus, larger the
borrower's transaction cost, the higher will be the average borrowing
cost and vice-versa for any given interest rate.
43
Figure 2.2: Borrowing Costs and Demand for Credit
ABC
AR/ABC/D/MVP
IT
TC
r
D/MVP
Tl L* Demand for Credit
Source: Ladman (1984)
AR
From the above analysis, one can say that, gIVen the binding
interest rate, it is the transnction cost, v,thich determines
the level of demand for credit. Figure 2.2 shows that borrowing
threshold (TJ) is the point where ABC = AR. Below this amount there
will be no demand for credit because ABC> AR. If the borrower's TC
exceeds a certain level of their expectation, they may not approach for
institutional credit. It is observed that not all applicants for formal
credit receive a loan. The expected borrowing costs of a new formal
applicant may be increased by their rejection possibility. These
rejection costs may be quite important if the probability of getting a
new formal loan is relatively low. Besides, many are of the view that
differential borrowing cost strongly affects the willingness of the rural
poor to seek loan from the formal sector.
Conclnding Remarks
Realising the importance of credit In terms of changing the
composition and distribution of production In favour of deficit
producers, shifting rural borrowing from informal to formal
institutions, and increasing the use of improved variety inputs, the
44
supply-led approach has been adopted m the pro\'lSlOnmg of
agricultural credit in India. It is also realised that both revealed and
potential demand for credit exceeds the supply. Based on these
considerations, banks were directed to expand the share of net bank
credit to agriculture sector. In the directed credit policy, importance
was given to the outreach by neglecting the quality and cost of
lending. The high cost structure of operations in rural areas with low
interest rate was an important factor for the failure of directed credit
system. When the supply-led approach started to threaten the viability
of financial institutions, the policy thrust since 1990s shifted to make
agricultural credit a viable activity. In this phase of agricultural credit
policy, importance was given to achieving quantitative target without
neglecting the viability of the. financial institutions. Thus, the
assumption of cross-subsidisation of agricultural credit was relaxed
and the lending rate to the ultimate borrowers was rationalised in
terms of reduction In the number of concessional slabs and
enhancement of some rates. One could, therefore, expect that the
deregulation of interest rates and improved quality of lending might
have helped the formal agencies to get more profit in agricultural
lending, and enhanced the proportion of credit to agriculture to the
total net bank credit. In this context, it is necessary to discuss the
likely impact of interest rate policy in the provisioning of agricultural
credit.
(al Except the Differential Rate of Interest Scheme operated by the
commercial banks and RRBs, identical rates are prescribed for all
sectors of the economy including agriculture, which vary directly
with the size of loan and there are only three size slabs
18
now. The
18 First slab covers the credit limit up to Rs.25, 000. The second and third slabs
cover above Rs.25, 000 and up to Rs.2 lakhs, and above Rs.2 lakhs of loan
respectively.
rates are also identical for short and long-term loans for a given
size. However, with the freedom given to rural co-operative banks
since October 1994 and to RRBs since August 1996 to determine
their own lending rates for all the slabs, the uniformity of lending
rates to ultimate borrowers has probably vanished.
(b) Since all financial institutions induding commercial banks have
been given freedom to determine the lending rates for loans above
Rs. 2 lakhs, different institutions may have formulated different
interest rates on these loans.
(c) If the lending costs (sum of financial, transaction and risk cost) of
rural credit institutions [as estimated by the Agricultural Credit
Review Committee (ACRC) for the period 1983-84 through 1985-86 J
are compared with the lending rates, the presently prescribed rate
for CBs on loans up to Rs.2 lakhs is inadequate to cover the costs
of lending
l9
. Since the maximum prescribed lending rate for CB is
13.5 per cent per annum for the credit limit of up to Rs.2 lakhs, it
is not possible to increase, the lending rate to cover the costs. Thus,
the CBs have the following two options. First, they may neglect the
borrower who demands agricultural loan up to the limit of Rs.2
lakhs; second, they have to cross-subsidise this segment of
agricultural borrowers. It may also be possible that in order to
minimise the extent of cross-subsidy, CBs may be concentrating on
a particular category of agricultural borrowers to achieve the
lending target given to them. Commercial banks may also be
moving towards a higher profit-oriented portfolio through diversion
19 For CBs, the cost of lending was 14.48 per cent per annum, whereas for Co-
operatives and RRB it was 16.30 per cent and 15.70 per cent, respectively.
46
of funds from sectors with lower returns to those having higher
returns.
(d) In the case of RRBs and co-operative banks, as the lending rates
have been totally deregulated for all slabs of credit limit, it is up to
them to raise the lending rate suitably to meet the cost of lending.
However, the increasing lending rate of RRBs and co-operative
banks may induce their borrower to turn to CBs. Even if such a
shift in demand takes place for the loan amount of up to Rs 2
lakhs, the CBs are unlikely to extend their credit facility to such
borrowers. Even if CBs extend credit facility, they may not go
beyond the prescribed level of 18 per cent of net bank credit.
Therefore, it is important to examine t h ~ rli!"trihlJtion pattern of
agricultural credit by CBs for below and above the credit limits of
Rs 2 lakhs.
Since financial institutions are not allowed to charge less than
12 per cent of annual rates on farm and non-farm loans, it is not the
floor rate but the ceiling rate that has been deregulated. In this
context, the following questions arise. Does credit subsidy due to
concessionary interest rate adversely affect the credit flow to
agriculture? Can increasing lending rate be a solution to reduce the
rate of credit subsidy? What is the interest rate elasticity on the
supply of agricultural credit? These questions have been discussed in
the third chapter.
The introduction of the PLCP is likely to be helpful in removing
:he ground level problems on credit flow and bridge the gap between
.arget and achievement of credit to the priority sectors. The policy of
47
refinance facility by NABARD might have helped to restructure the
lending activity of CBs and RRBs to different sectors including
agriculture. Importantly, these policies may be helping CBs and RRBs
to achieve their lending target given for agriculture sector.
Nevertheless, setting of lending targets for agriculture might have
resulted in a positive impact on the supply of credit to hitherto
neglected sector by giving ample access for marginal and small
farmers to institutional credit. In the context of these policy changes
and recommendations made by both Narashimham and Gupta
committees, the extent to which the formal banking system has
improved the access to agricultural credit is to be examined. This
issue is taken up for discussion in the following chapters.
48
Chapter 3
CREDIT FLOW TO AGRICULTURE: TRENDS AND
CONTRIBUTING FACTORS AT ALL-INDIA LEVEL
Introduction
The administered allocation of credit to priority sectors at
concessional interest rates is an important policy dimension of
directed credit programmes. The rationale of this policy lies in the fact
that, without government interference through directed credit
programmes, banks do not fund to those activities with high social
return or those categories of credit-worthy people who are
marginalised in the credit market (Stiglitz 1994). The directed credit
policies in India have been used for promoting agriculture and small-
scale industry. Agriculture has been targeted because it is a risky
activity and households are credit rationed by the formal sector
(Swaminathan 1991; Kochar 1997). It is also felt that the availability
of concessional credit could help the farmer to adopt new technology,
encourage investment on machinery and irrigation and augment the
use of quality inputs to increase agricultural productivity. Thus,
commercial and co-operative banks were directed to expand their
rural branch networks and intensify their lending to agriculture.
After the nationalisation, lending to priority sectors became an
essential component of bank lending. By taking the increasing credit
needs for farming operations into account, specific lending targets
were fixed for each major bank. Ultimately, it was recommended to
raise the proportion of direct farm lending to 18 per cent of net bank
credit by March 1990. However, India's rural directed credit
programme has not been a success, specifically in terms of allocation
of resources, cost-effectiveness and access to credit for different
49
categories of farmers due to the following: First, the agricultural
lending rates set by the government were lower than commercial and
industrial rates, and hence, commercial banks could not cover the
cost of their agricultural loan from profit ansmg from other
operations. This led to cross-subsidisation of the agricultural
operations (Katula and Gulati 1992; Rao 1994; Binswanger and
Khandker 1995). Second, the preoccupation to achieve quantitative
targets, ignoring qualitative aspects and mounting overdues, became a
threat to the viability of financial institutions. Third, the disbursal of
loans without assessing the feasibility, viability and entrepreneurial
experience of borrowers (Rajasekhar and Vyasulu 1993) led the banks
to incur losses due to the widespread problem of fungibility and poor
repayment of the loan
The effectiveness of directed credit programmes in stimulating
investment, raising growth and reducing poverty was often debated
(Naastepad 2001). It is also argued that directed credit was inferior to
a market-based allocation of resources in achieving growth and
redistributive objectives (World Bank 1989; Odedokun 1996; Varon et
al. 1998). Hence, notwithstanding the empirical support for the
positive growth effects of directed credit in both agriculture and small-
scale industry, the focus of India's recent financial reforms has been
on the phasing-out of its directed credit programmes (Shajahan 1999;
Naastepad 2001). As explained by the Report of the Narasimham
Committee on the Financial System (1991), the financial reforms were
primarily concerned with the decline in productivity and efficiency of
the banking sector. The report points out that the directed credit had
negative effects on the income of banks through concessional interest
rates and high administrative costs of such loans. To improve the
50
efficiency, the Committee recommended reducing the directed credit
from the prevailing 40 per cent to 10 per cent of net bank credit over a
three-year period. Due to strong opposition from fanners and small-
scale industry and political lobby, the government did not officially
accepted this recommendation. In fact, the RBI has been following the
strategy of keeping the overall share of priority sector at 40 per cent of
net bank credit, and broadening the scope of priority sector lending to
encourage diversion form it to the new areas added to this sector. In
other words, importance was given to achieve the same quantitative
target without neglecting the viability of the financial institutions.
Thus, assumption of cross-subsidisation of agricultural credit was
relaxed and the lending rate to the ultimate borrowers was
rationalised in terms of reduction in the number of c:oncessiona.l slabs
and the enhancement of some rates. The other important
recommendations include the abolition of branch licensing, closing
down of loss making bank branches, gradual reduction of Cash
Reserve Ratio (CRR) and Statutory Liquidity Ratio (SLR).
One can, therefore, expect the following positive as well as
negative influences of banking sector reforms on the credit flow to
agriculture.
1. Deregulation of interest rate may have helped the formal agencies
to enhance the proportion of disbursement to agricultural sector in
the overall net bank credit.
2. Widening scope of priority sector may have resulted in the greater
availability of credit to priority sector, in general, and agriculture
sector, in particular.
3. Increasing the availability of funds to the banks as a result of
reduced CRR and SLR may have a positive impact on the credit
flow to agriculture.
51
4. Since most of the loss-making bank branches during the pre-
reform period were located in rural areas. the closure of bank
branches may lead to reduced supply of credit to agriculture.
5. The recommendation of the Gupta Committee that the quality of
lending is to be improved may provide ample access to institutional
credit by marginal and small farmers.
The above raises the following questions. Does the banking sector
reforms improve the share of net bank credit to agriculture sector?
Does profit-oriented lending norms persuade commercial banks to
neglect agriculture sector? Does credit subsidy reduce the supply of
agricultural credit? Can increasing lending rate be a means to reduce
the rate of credit subsidy? What is the interest rate elasticity on the
supply of agricultural credit? What is the impact of the closure of
rural bank branches on the provision of credit to agriculture? Does
the expansion of the scope of agriculture sector bank-lending result in
the greater availability of credit?
This chapter addresses these questions by analysing the data
collected from the Report on Currency and Finance and Banking
Statistics on the total outstanding credit provided by the Scheduled
Commercial Banks (SCBs) to the agriculture sector.
Credit Flow to Agriculture
Soon after independence. the share of Formal Rural Banking System
(FRBS) in the total credit was found to be insignificant. However. the
nationalisation of banks in 1969. introduction of RRBs and priority
sector lending has had the desired impact in stepping up the supply of
credit to agriculture. The share of agriculture in the total credit
provided by FRBS. which was only 7.1 per cent in 1969. increased to
15.72 per cent of the total net bank credit by 1980.
52
The total outstanding credit disbursed by all the SeBs increased
from Rs.28,391.73 crores in 1981 to Rs.460,080.68 crores by 2000.
The annual growth rate of total outstanding credit was 15.38 per cent
during this period (Table 3.1). The growth rate during the period 1991-
2000 was high as compared to that during the period 1981- 91. One
of the important contributing factors for rapid growth rate during the
reform period was the policy decision to slash eRR and SLR to
increase funds availability to the banking sector. The eRR declined
from 15 per cent of demand and time liability in 1991 to 8.5 per cent
in 2000, while the SLR declined from 38.5 per cent to 25 per cent
during the same period. Besides, the borrowing from RBI was stepped
up after 1991 to facilitate more funds to the SeBs (RBI 2001).
Was the rapid growth in the total credit the same across all the
sectors? Table 3.1 shows that the annual growth rates of credit to
professional services and personal loans, finance and others were
higher during the pre-reform period. On the other hand, the rate at
which credit was disbursed to industry, trade and transport operators
was high during the reform period as compared to that in pre-reform
period. The credit disbursed to agriculture sector increased at an
annual growth rate of 11.86 per cent during the period 1981-2000.
These were, however, distinct inter-period variations. While the
annual growth rate for the pre-reform period was 14.77 per cent, it
was only 10.90 per cent during the reform period. Thus, the rate at
which credit was disbursed to agriculture had declined at the all-India
level.
53
Table 3.1: Annual Average Growth Rates
2
() of Credit (%) by
SCBs to Different Sectors
Years ! Agricul-
Indus Trans- Professional Trade Financel Others i For
I ture
try port Services and
I f s e : : ~ r s Opera- Personal
tors Loans
1981- 91 14.77 15.68 8.68 22.09 10.84 26.30 19.36 15.33
:
,
1992 2000 10.90 16.21 11.48 20.68 15.16 24.76 16.44 16.14 i
1981- 2000 11.86 15.80 7.82 21.31 13.06 26.12 17.57 15.38 I
Source: Reserve Bank of India (Bankmg Statistics) for the Year 1981 to 2000.
What were the trends in the proportion of credit to agriculture
from SCBs? Out of the total agricultural credit, the RRBs provided
two-thirds to one-half whereas the proportion in the case of SBI and
its associate banks, nationalised banks and other Scheduled
Commercial Banks was much less. Table 3.2 shows that agriculture
accounted for about 16.44 per cent of the total credit in the early
1980s and suggests that the banks were meeting the targets fixed on
agricultural lending. However, the proportion progressively declined to
9.92 per cent by 2000. Such a decline was uniform across different
bank groups. Thus, one can conclude that the credit flow to
agriculture even in proportionate terms declined during the reform
period.
10 The growth rate has been calculated by using the semi-log model, such as
In Y, ~ \>1 + \>, t + U,
where, t is the time period, PI and P, are parameters, and U, is the disturbance
term.
After estimating the above regression model, annual average growth rate (over a
period of time) has been calculated by the following way, viz., Annual Average
Growth Rate ~ (Antilog of the estimated P, - 1)* 100.
The same procedure has been applied to calculate the growth rates wherever they
are required in this study.
54
Table 3.2: Bank Group-wise Share of Outstanding Credit of
SCBs to Agriculture from Net Bank Credit
(in per cent)
Other I Total I Triennium I SBI and Its
Ending I Associates
with
Nationalised Foreign
Banks Banks
Regional
Rural
Banks
Scheduled I
Commercial I i
Banks
1983 19.23 15.35 0.00 64.66 5.90 16.44
1986 21.56 14.89 0.00 59.24 5.00 17.05
1989 19.19 16.53 0.00 55.12 4.98 17.46
1990 16.90 15.44 0.34 54.37 9.60 15.94
1993 15.86 14.18 0.54 50.45 8.14 14.38
1996 12.75 11.97 0.33 49.53 5.63 11.94
1999 11.56 11. 20 0.37 48.35 5.03 10.81
2000 10.36 10.48 0.36 48.12 4.42
Source: Reserve Bank of India (Bankmg Stat,st,cs) for the Years 1981 to 2000.
Note: One-year figure
Figures up to 1981- 89 are for the period January to December and
thereafter, for the period April to March.
9.92
The figures presented in Tables 3_1 and 3.2 do not include
credit from co-operatives, which have traditionally been playing a key
role in so far as short-term credit to agriculture is concerned. Table
3.3 includes the co-operative credit in the total outstanding credit.
The total outstanding credit disbursed by co-operatives and the SCBs
increased fram an average of Rs_ 8,618 crores in 1981-84 to Rs.62,
608 crares by the year 2000. The co-operatives accounted for more
than one-half of the agricultural credit in early 1980s. Their share had
gradually declined to about one-third by late 1990s, while that of
SCBs and RRBs had increased. An important finding is that although
the share of commercial banks and RRBs in the total agricultural
credit disbursed in the country increased during the reform period
(Tables 3.3), the share of these institutions in the net bank credit
declined during this period (Table 3.2).
55
Table 303: Distribution (%) of Outstanding Credit for
Agriculture and Allied Activities (Short and Long-Term) by
off< t T fLe dO I tOt to
Dl eren L'YI'es 0 n
109
ns 1 u Ions
=---
Triennium Scheduled Regional Total
Ending with Co-operatives Commercial Banks Rural Banks IRs. in Crores)
(Excludinl( Rims)
1983 55.28 41.49 3.23 8,618
1986 45.9'1 49.01 5.01 13,811
1989 40.18 53.56 6.26 20,922
1990* 38.16 55.20 6.64 27,687
1993 38.51 55.22 6.27 31,573
1996 41.42 51.28 7.30 41,249
1999 39.39 51.95 8.66 54,282
2000* 37.02 53.41 9.57 62,608
Note: * One-year figure. FIgures up to 1981-89 are for the penod January to
December and thereafter, for the period April to March.
Sources: 1) Reserve Bank of India (Report on Currency and Finance, Vol. II) from
1981 to 1998.
2) Figures for the years 1999 and 2000 have been collected from RBI,
2001 (Hand Book of Statistics on Indian Economy).
The growth rate of credit to agriculture from all the institutions
(in absolute terms) was 11.74 per cent during the period 1981- 2000.
On the other hand, the growth rate in real terms
21
was only 3 per cent
for the same period (Table 3.4). The growth rates have been uneven
across the sub-periods as well as across the bank groups. The credit
in both absolute and real terms grew at a much faster rate during the
period 1981-91 as compared to the reform period of 1991-2000. A
similar pattern can be observed across the bank groups also, although
the growth rates of credit flow from RRBs have been higher in both
absolute and real terms probably due to a low base as compared to
other financial institutions. The situation of co-operatives appears to
be even grimmer.
" In order to calculate the growth rates in real term, the outstanding credit amount
in each year has been deflated by the Consumer Price Index of Agricultural
Labourers taking the year 198687 as a base.
56
Table 3.4: Annual Average Growth Rates (%) of Outstanding
Credit to Agriculture
I Co-operatives
Scheduled
I
Year Commercial Banks Regional Total
I
(Excludin; R R B ~ Rural Banks I
Absolute Real Absolute Real Absolute Real Absolute Real_
1981-1991 9. 83 2. 74 19.66 11. 93 26. I I 17.96 15. 26 7. 82
1992-2000
,
8. 47 0.61 9. 06 I. 16 15. 41 7. 06 9.31 I. 40'
1981-2000 9. 79 I. 91 12. 95 4. II 17.99 8. 75 II. 74 3. 00
Source: I) Reserve Bank of IndIa (Report on Currency and FInance, Vol. II) from
1981 to 1998.
2) Figures for the years 1999 and 2000 have been collected from RBI, 200 I
(Hand Book of Statistics on Indian Economy)
The inter-sub-period differences in the flow of agricultural credit
were also observed from the mean amount of outstanding credit
disbursed by the SCBs including RRBs by formulating the null
hypothesis that "there is no significant difference in the mean amount
of credit flow of the two samples". The t-test was carried out to test the
different between the means of two samples. The calculated t value
(6.26) is greater than table value (2.10) at 5 per cent level. Based on
this result, it is inferred that there is a significant difference in the
mean amount of outstanding agricultural credit between the two sub-
periods.
Distribution Pattern of Agricultural Credit of SCBs
The proportion of credit to agriculture in the net bank credit declined
from about 17 per cent during the pre-reform period to 10.39 per cent
during the reform period, while the proportion of credit to non-
agricultural sectors continuously increased between these two periods
(Table 3.5). The proportion of agricultural credit for the loans in the
size of less than Rs. 25,000 increased during the pre-reform period.
This has been attributed to the prescribed targets on agricultural
lending (Besley 1994), lack of alternative investment avenues for rural
57
bank branches (Rajasekhar and Vyasulu 1993) and different policies
without reference to the cost of funds (Gad gil 1994; Yaron 1994).
However, the proportion of credit to agriculture for both sizes of credit
limit has been declining during the reform period. In the case of non-
agriculture sector, the share of credit flow for the credit limit of less
than Rs.2S,000 increased up to the last part of 1980s and declined in
the 1990s. Importantly, the share of non-agricultural sector from the
total credit for the credit limit of above Rs. 25,000 has been increasing
during the last two decades. Thus, based on the discussion relating to
the volume of outstanding credit (Table 3.5), it can be concluded that
the banks have been showing bias in favour of non-agriculture sector
lending.
Table 3.5: Distribution of Outstanding Credit to the Total Net
Bank Credit by SCBs (%) by Sectors and Credit Limits
Triennium Agricultural Sector NonAgricultural Sector
Ending Rs.25,OOO Above Sector Rs.25,OOO Above Sector
with and Less Rs.25,OOO Total and Less Rs.25,OOO Total
1983 8. 67 7. 78 16. 44 10.23 73. 32 83. 56
1986 9. 58 7.47 17.05 12. 10 70. 85 82. 95
1989 10. 39 7. 07 17.46 14. 77 67. 77 82. 54
1991 * 8. 76 6.65 15. 40 13. 77 70. 83 84. 60
1994 7. 36 6. 36 13.72 12. 47 73.81 86. 28
1997 5. 61 5. 77 11. 39 8. 75 79. 86 88.61
2000 4. 09 6. 29 10.39 5. 78 83.83 89.61
Note: * Two-year figure
Source: Reserve Bank of India (Banking Statistics) for the Years 1981 to 2000.
The percentage of number of loan accounts in the agricultural
sector to the total bank accounts had been declining during both pre-
reform and reform periods (Table 3.6). Interestingly, although the
number of accounts for the credit limit of less than Rs. 25,000
declined In the case of the agriculture sector during the pre-reform
period, the proportion of amount of credit increased. This indicates
the credit rationing by banks to minimise transaction costs associated
58
with agricultural lending, by reducing the numbers of accounts for
better supervision and monitoring. In the case of non-agricultural
sector, however, the proportions of both accounts and amounts for the
credit limit of less than Rs. 25,000 increased during the pre-reform
period probably due to higher interest rates on loans to non-
agricultural activities.
Table 3.6: Distribution of Loan Accounts by SCBs (%) by
Sectors and Credit Limits
Triennium AJ:ricultural Sector Non-Al!ricultural Sector
Ending Rs.25,OOO Above Rs. Sector Rs.25,000 Above Rs. Sector
with and Less 25,000 Total and Less 25,000 Total
1983* 49. 30 1. 13 50. 43 46. 28 3. 29 49. 57
1986 48.61 1. 07 49. 68 47. 20 3. 12 50. 32
1989 45. II 1. 13 46. 24 50.47 3. 29 53. 76
1991 ** 43. 46 1. 25 44. 71 51. 50 3. 79 55. 29
1994 41. 02 1. 35 42. 37 53. 25 4. 38 57.63
1997 40. 08 1. 90 41. 98 51. 42 6. 60 58. 02
2000 34. 69 4. 02 38.71 45. 71 15.58 I 61. 29
Note: One-year fIgure, ** Two-year figure
Figures up to 1981-89 are for the period January to December and thereafter, for
the period April to March.
Source: Reserve Bank of India (Banking Statistics) for the Year 1981 to 2000.
In the case of agricultural loans with credit limit of less than
Rs.25,000, the proportions of accounts to total loan accounts and
amounts to the total net bank credit declined during the reform
period. Although the proportion of accounts for the credit limit of more
than Rs. 25,000 increased in the case of agricultural sector, the net
credit flow declined during the reform period. It can be, therefore,
argued that the banks have been adopting both price and non-price
credit rationing mechanisms to minimise lending risks by disbursing
less loan amounts to borrowers in the agricultural sector.
The tendency on the part of banks to provide more funds to the
agricultural activities earning higher interest income seems to have
become more prominent after the deregulation of lending rate to
59
ultimate borrowers. Since the interest rate charged was unable to
cover the cost of lending (i.e., financial, transaction and risk costs).
the lending rate of RRBs was deregulated for all slabs of credit limit
from August 1996. Prior to this, the lending rates of CBs were also
deregulated for the credit limit of over Rs. 2 lakhs from August 1994.
For the credit limit of up to Rs. 2 lakhs, all CBs were asked to charge
a maximum rate of 13.5 per cent per annum. In the case of RRBs, as
the lending rates were totally deregulated for all slabs of credit limit, it
was up to them to charge the rate suitably to meet the cost of lending.
The CBs can also follow the same strategy for a credit limit of above
Rs 2 lakhs. In this context, the following inferences can be drawn:
(1) RRBs may find it comfortable to extend their funds for
agricultural sector.
(2) CBs may give priority for agricultural loan for the credit limit of
above Rs. 2 lakhs to earn higher interest margin.
(3) After the interest rate deregulation, loans from CBs have become
cheaper as compared to those given by RRBs. Under these
circumstances, borrowers may prefer to obtain loan amounts up
to Rs.2 lakhs from CBs provided that service area regulations are
favourable. The Gupta Committee, however, recommended that a
bank could lend to borrowers outside their service if they chose to
do so.
(4) If the third inference is true, then CBs must be providing more
agricultural credit to the farmers for credit up to Rs 2 lakhs by
cross-subsidisation approach.
As the data on agricultural lending up to the credit limit of Rs. 2
lakhs are not available separately for RRBs and CBs, it is difficult to
examine the third proposition. In other words, conclusions on the
volume of credit disbursed to agriculture sector by RRB for this credit
limit are difficult to be drawn. However, in spite of the deregulation of
lending rates of RRBs, the proportion of amount disbursed to
60
agricultural loans of less than Rs. 2 lakhs by SCBs including RRB
declined from 81.4 per cent in 1992 to 69.79 per cent in 2000 (Table
3.7). In contrast, with complete deregulation of lending rates in the
case of credit limit of above Rs. 2 lakhs for both CB and RRBs, the
proportion of outstanding credit increased from 18.6 to 32.21 per cent
during the same period. This indicates the priority to agricultural
loans of above Rs. 2 lakhs in the lending portfolio of CBs. There was
an increase in the proportion of amount disbursed to agricultural
loans in the range of Rs. 25,000 to Rs. 2 lakhs. This implies that the
declining share of agricultural loans up to Rs 2 lakhs was mainly due
to a reduction in the share for the credit limit of less than Rs 25,000.
This lends further credence to the earlier finding that there was
widespread credit rationing practice by CBs for borro\vers availing less
than Rs. 25,000 of agricultural loans.
Table 3.7: Distribution of Agricultural Credit by SCBs (%) by
the Type of Lending Rates and Credit Limits
Triennium Distribution (% I of Agricultural Proportion of Total Credit
Ending Credit with Partially Regulated Credit with (Amount in !
with Lending Rate Deregulated
Rs. Lakhsl I
Lendinl( Rate
Rs.25,OOO Above Rs.25,OOO Above Rs. 2
and less & up to Rs.2 lakhs Total lakhs
1992 56. 25 25. 15 81. 40 18. 60 1,847,903
1995 52. 62 26. 02 78. 64 21.36 2,329,370
1998 47. 45 26. 07 73. 52 26.48 3.190,188
2000- 36. 64 31. 15 67. 79 32.21 4,326,377
Note: Two-year figure
Source: Reserve Bank of India (Banking Statistics) for the Year 1981 to 2000
The number of loan accounts for all the sectors and also
individual sectors, though rapidly increased during the pre-reform
period, registered negative growth rates during the reform period.
Interestingly, the negative growth rates in the loan accounts for both
agricultural and non-agricultural sectors was solely due to the
reduction in the number of accounts for the credit limit of Rs. 25,000
61
and less. This suggests the neglect of small borrowers by seBs. This
also implies that the access of formal credit to small borrowers was
adversely affected. The relevant question, therefore, is what factors
determine the access to formal credit.
Table 3.8: Credit Limit-wise Annual Average Growth Rates of
Loan Accounts
(in per cent
Agricultural Sector Sector For All
Period Rs. Above Sector Rs. Above Sector Sectors
25,000 Rs. Total 25,000 Rs. Total
and Less 25,000 and Less 25,000
1983-91 7. 94 I!. 94 8.03 I!. 83 12. 50 11. 87 10. 03
1991-00 - 5. 33 16. 29 - 4.03 - 5. 18 19.70 - 1. 52 - 2.54
1983-00 1. 05 11. 54 1. 57 3. 28 13. 66 4.63 3. 26
Sources: Reserve Bank of India (Bankmg Statistics) for the Year 1981 to 2000
Achievement of SCBs in Agricultural Finance
The outreach efforts of the FRBS into the rural areas, in general and
agriculture, in particular, yielded positive results in quantitative
terms. The outstanding credit to agriculture by SeBs (including
RRBs) increased from Rs. 4,863 crores in 1981 to Rs. 45,638 crores
by 2000 (Reserve Bank of India 1981 and 2000). However, the share
of agricultural credit to the net bank credit declined in the last two
decades. Table 3.9, which presents the data on agricultural lending
during the period 1981-2000, shows that the SeBs hardly achieved
the prescribed targets given to agricultural lending. Before October
1993, only direct finance to agriculture was targeted and banks were
supposed to lend 18 per cent of the net bank credit by 1990. But the
achievement rate of direct finance to agriculture was 13. 84 per cent
(Table 3.9), well below the target of 18 per cent.
62
Table 3.9: Proportion of Credit Advanced to Agriculture from
Net Bank Credit of SCBs _______ ---,
Type of Triennium EndinJ: With !
Advance 1986 1989 1990' 1993 1996 1999 2000';
Direct 12.91 14.28 15.21 13.84 12.59 10.23 9.20 8. 38
Finance Il+ O. 28JJ:- 1.79Ll- 4.16) I (- 5. 41) I (- 3. 2TI.J- 4. 30Jjl- 5.
Indirect 3. 53 2. 77 2. 25 2. 10 I. 79 I. 71 I. 62 I. 54
Finance I (- 2. 79) I (- 2. 88) (- 2.96) I
Total 16.44 17.05 17.46 15.94 14.38 II. 94 10.81 9.92 'I
Advance I (- 6. 06) 1(- 7. 19) 1[- 8. 081.
Net Bank
Credit (in 33,737 52.840 78,696 104,312 141,125213,841 332,248460,081
Rs. Crores)
Note: One-year figure.
1) Figures in parentheses represent shortfall or excess of target fixed for
agriculture from net bank credit at different points in time.
2) Figures up to 1981-89 are for the period January to Decem ber and
thereafter for the period April to March.
Source: Reserve Bank of India (Banking Statistics) for the Years 1981 to 2000.
Since banks could not achieve the given target of direct finance,
indirect finance to agriculture was brought under the purview of
finance to agriculture, and clubbed with direct advances for meeting
the 18 per cent sub-target to agriculture from October 1993. However,
in order to ensure that the focus of banks on direct agricultural
lending was not diluted, the "indirect" advance to agriculture was
stipulated to be one-fourth of total agricultural lending or 4.5 per cent
of net bank credit. In order to ensure the inputs to allied agricultural
activities of dairy, poultry, piggery and fishery, advances up.to Rs. 5
lakhs for financing the distribution of inputs for the allied activities
were reckoned as indirect agricultural advances under the priority
sector effective from May 1994 [RBI 1993-941. Besides, the finance
extended to dealers in drip irrigation, sprinkler irrigation and
agricultural machinery was classified as "indirect finance to
agriculture" and clubbed with the priority sector advances. Despite
these efforts, the proportion of indirect advance to agriculture to net
bank credit varied between 1.50 and 2 per cent for the last 8 years.
Importantly, the total agricultural advances rapidly declined to 9.92
6
'
.'
per cent of net bank credit by 2000 (Table 3.9), and the gap between
target and achievement rates for both direct and indirect advances
was widening in the 1990s.
The distribution of direct finance by size-classes of farmers
(Table 3.10) during the last two decades shows that the proportion of
the amount disbursed for marginal and small farmers increased up to
1980s. In 1990s, however, the share of the marginal farmers had
declined from 29.7 to 23.8 per cent. At the same time, there was a
marginal increase in the proportion of credit disbursed to small
farmers and substantial increase in the same to medium/large
farmers in this decade. This further supports the earlier argument
that formal financial ~ g e n c i e s preferred lending to better-off farmer3.
Table 3.10: Distribution (%) of Direct Finance to Farmers by
SCBs (Short and Long-Term Loans) by Size-Classes of
La db ld n 0 109
Triennium Marginal Small (2.5 Medium and Total
Ending With 1<2.5 Acres) to 5 acres) Large 1>5 acres) IRs. Crores)
1983 27. 3 19.4 53. 3 825. 0
1986 26. 8 25. 3 47. 9 1,900.1
1989 27. 8 25. 9 46. 3 2,958.5
1991' 29. 7 24. 7 45. 5 3, 722. 3
1994 28. 5 25. 1 46. 4 4, 278. 4
1997 25.8 25. 1 49. 1 7, 589. 6
2000 23. 8 25. 6 50. 6 11, 790. 1
Note: * Two year figure.
Figures gives for all the years are for the period july-june except for the
year 1981 and 1982 (April-March).
Source: Reserve Bank of India, 2001 (Handbook of Statistics on Indian Economy)
The changes in the distribution of loan accounts by size-classes
of farmers during the period 1981 to 2000 (Table 3_11) reveal that the
proportion of accounts held by marginal farmers was generally large in
the early 1980s as compared to the late 1990s. On the other hand,
the proportion of loan accounts held by small and medium/large
farmers increased, albeit with fluctuations. This suggests that a
64
larger number of medium and large sIze farmers (above 5 acres of
land) were obtaining credit from banks as compared with marginal
farmers.
Table 3.11: Distribution or Disbursement Loan Accounts on
Direct Finance to Farmers by seBs (Short-Term and Long-
Term Loans)
Triennium Up to 2. Above 2. 5 Above 5 Total No. of
Ending 5 Acres Acres to 5 Acres Accounts
With (in %1 Acres (in 0/<>1 (in %1 (in OOOsl
1983 51. 1 24. 0 24. 9 2,329.9
1986 47. 2 29. 8 22.9 3, 960. 2
1989 46. 8 31. 0 22. 2 4, 608. 4
1991' 47. 7 30. 4 21. 9 4, 209. 3
1994 44. 2 31. 2 24. 7 4,241.7
1997 39. 0 31. 1 29. 9 5,241.2
2000 39. 8 32. 8 27. 5 5,657.9
Note: Two year fIgure.
Figures given for all the years are for the period JulyJune except for the
year 1981 and 1982 (April-MarchI.
Source: Reserve Bank of India, 2001 (Handbook of Statistics on Indian EconomYI.
Credit Subsidy in Indian Agriculture
Subsidised credit to agriculture through concessional interest rates
and high defaults have been posing a threat to the financial viability of
credit institutions. If the interest income is greater than or at least
equal to the cost of lending (break-even condition), this may positively
influence the financial institutions to step up the supply of
agricultural credit. However, default rate plays an important role in
determining the break-even level of interest rate. Higher the default
rates and higher the break-even interest rate that has to be charged.
Since it was difficult to cover the costs of lending, the RBI initiated
several measures towards deregulation of interest rates in 1990s. In
the case of RRBs and co-operatives the lending rates were totally
deregulated for all slabs of credit limit. For Commercial Banks, only
the interest rate for the credit limit of above Rs. 2 lakhs was
deregulated.
65
In the deregulated interest rate structure, the lending rate
varies with the loan size. However, for the credit limit of less than Rs 2
lakhs in CBs, the hitherto prevalent distinction between preferred and
other investment was abolished. and the maximum interest rate fixed
was 13.5 per cent per annum. Now, the question is whether this
interest rate was sufficient to cover the costs. If no, the CBs might
have to cross-subsidise to the farmers taking loan amounts less than
Rs. 2 iakhs with interest income earned from lending to the credit
limit of more than Rs. 2 lakhs by pumping more funds into the latter
category belonging to either agriculture or non-agriculture. This
discussion raises the following questions:
1. Are the interest rates sufficient to cover the costs of lending?
2. If no, is there cross-subsidisation? If yes, is it confined to
agriculture or non-agriculture?
3. Does credit subsidy reduce the supply of agricultural credit?
The answer to the first question needs a discussion and
analysis on credit subsidy. Credit subsidy can be calculated in two
different ways. The first method consists of two components, namely,
interest subsidy and default subsidy. Interest subsidy accrues to
agriculture due to concessional rate of interest being charged to this
sector vis-a-vis the other sectors of the economy. Default subsidy22
accrues to agricultural lending in the form of bad debts. Credit
subsidy to agriculture can be obtained by adding these two subsidies.
The second method perceives credit subsidy as the difference between
cost of supplying credit to agricultural sector (including defaults) and
rate of interest received from agricultural sector only.
" It is worth mentioning here that default also oCCurs in non-agricultural sector
lending and that therefore. one has to take into account onl\' the additional
default In credit to agriculture while calculating default subsid\ .
66
Due to the practice of uniform interest rates on loan amounts
less than Rs. 2 lakhs in the case of CBs, the interest subsidy may not
exist in the case of this credit limit regardless of the purpose. So,
according to the first method, the credit subsidy arises mainly due to
the default subsidy. Importantly, due to difficulty in the availability of
data on bad and doubtful debts, it may be difficult to estimate the
credit subsidy by applying the first method. In the present study,
credit subsidy has been calculated by applying the second method on
the basis of certain assumptions mentioned at different places in the
ensuing paragraphs.
The Agricultural Credit Review Committee (ACRC) had
envisaged three broad c.Htegories of C'osts in the agricultural credit
system. These were financial costs
23
, transaction costs
2
4, and risk
costs or costs of bad debts
25
. Table 3.12 provides the costs estimated
by the committee for different financial institutions for the period
1983-84 to 1985-86.
Table 3.12: Estimated Lending Costs Per Rs.I00 Outstanding
Loan
(in Rs.1
In.tltutions Financial Transaction Risk Cost Total Cost
Cost Cost
PACS 9. 90 5.40 1. 00 16. 30
LDB 6. 51 4.04 1. 00 11. 55
RRB 7. 80 6. 90 1. 00 15. 70
CB 7. 48 6. 00 1. 00 14. 48
.
Source: Khusro CommIttee, 1989, Chapter XVIII .
13 A fmancial cost refers to actual costs of raising financial resources. It is computed
as the interest on borrowing and deposits.
14 Transaction costs are organisational costs for carrying out the day to dav
institutional operations such as staff salaries and allowances, bUilding rent,
travel, printing and stationary, postage and telegram, audit and training, etc.
15 R' k .
IS cost consIsts of actual write-offs or reserve created for bad and doubtful
debts.
67
Gadgil (1994) argues that the transaction costs In 1990s were
higher than the above due to rising staff salaries (80 per cent of the
transaction costs, according to Khusro Committee, was manpower
related) and escalation in the cost of office rent, transport, stationary,
postage, telephone charges, etc. This study, therefore, assumes that
the transaction costs in 1990s were higher by one percentage point
than the estimates provided by Khusro Committee, and that the risk
cost was constant while calculating the lending cost. Since our focus
is on the credit extended by all SCBs (including RRBs), we have
considered the average of financial and transaction costs as estimated
by the Khusro Committee of both categories of banks in calculating
the cost of lending to agricultural sector. Furthermore, it is also
assumed that the cost of lending is the same for short and long-term
agricultural credit. By taking all these assumptions into account, the
lending costs for agricultural credit
26
have been calculated for the
study period in Table 3.13. Given these costs of lending, the interest
income on lending to agriculture sector is estimated below to see the
extent of credit subsidy.
Table 3.13: Cost of Lending to Agricultural Sector Per Rs.I00
o tst d' C d't f C 1 d' u an
1 D ~ re 1 0 Bs Inc u lDJ:! RRBs
Year Average Average Average Total
Financial Cost Transaction Cost Risk Cost Cost
1981 to 1990 7. 64 6.45 1. 00 15. 09
1991 to 2000 7.64 7. 45 1. 00 16. 09
26 These costs, especially financial, were marginally lower than those calculated bv
Gadgil 11997). Using the data on structure-wise deposits in CBs as on March
1993 and applying the January 1997 interest rates on deposits Inil on current
deposits, 4.5 per cent on savings bank deposits and 8- I 3 per cent on term
deposits), he calculated the weighted average annual cost of deposits at 7.90 per
cent. If the cost of deposits as estimated by Gadgil had been considered as the
finanCIal costs, the lending costs would have been more than what the study has
calcubted.
68
The data on the amount of agricultural loan outstanding by
SCBs (including RRB) at different interest rates were used to estimate
the interest income on agricultural lending. These data relate to
accounts with credit limit of over Rs 25,000 for the period 1981 to
1998, and the credit limit of above Rs 2 lakhs for the years 1999 and
2000. Since the lending rate was, by and large, fixed for the credit
limit of less than Rs. 2 lakhs, the data were accordingly adjusted In
calculating the interest income on agricultural lending.
The annual average interest income for Rs. 100 outstanding credit to
agriculture can be calculated as:
II
Y (rml x X ml )
m=1
-- -------(1)
Where,
rmt = the mean of mth interest range for agriculture sector for the tth
period
Xmt = the outstanding credit in mth interest range for agriculture for
the tth
year
X, = total outstanding credit for agriculture for the tthyear
Here, we define credit subsidy as the difference between cost of
providing credit to agriculture and interest rate earned on outstanding
agricultural advances adjusted for defaults. Thus, the credit subsidy
(S,) for Rs 100 can be calculated as:
St = Ct- rt - - -- - -- - - -- - - - - - - (2)
(where Ct is the cost of lending to agriculture sector).
69
The resulting figure of S, gives the credit subsidy per Rs. 100 of
outstanding loan. Accordingly, total credit subsidy to agriculture (Lr)
sector can be obtained as:
Lr = (S, / 100) X, -- - - - - - - - - - - - - - - - (3)
Based on the above model, credit subsidy on agricultural finance has
been estimated in Table 3. 14.
Table 3.14: Cost, Interest Income and Credit Subsidy to
A . It IS t
g r l C U ura ec or
For the Credit Limit (CL) of Total Agricultural Credit
Year
C, More Than Rs. 25,000 (including CL for < Rs. 25,000)
r, 5, Lt (Rs. Crore) r, 5, Lt (Rs. Crore) .
I 2 3 4 5 6 7 8
1981 15. 09 13. 50 1. 59 41.69 13. 50 1. 59 77. 33
1982 15. 09 13. 24 1. 85 54.62 13.26 1. 83 103. 19
1983 15. 09 12. 45 2. 64 70. 10 13. 05 2. 04 125. 29
1984 15. 09 11.99 3. 10 123. 98 12. 75 2. 31
.00
""
J.vv. ~ v
1985 15. 09 12. 00 3.09 117. 17 12. 86 2. 23 197. 35
1986 15. 09 12. 06 3. 03 122. 67 12. 92 2. 17 219.27
1987 15. 09 12. 16 2.93 146. 78 12. 95 2. 14 259. 20
1988 15. 09 12. 20 2. 89 158. 70 12.97 2. 12 285. 92
1989 15. 09 11. 37 3. 72 229. 94 12. 64 2. 45 374. 01
1990 15. 09 12. 83 2. 26 154. 54 13. 22 1. 87 310.91
1991 16. 09 13. 51 2. 58 214.63 13. 50 2. 59 481. 05
1992 16. 09 14. 95 1. 14 103. 15 14. 15 1. 94 392.61
1993 16. 09 15.71 O. 38 39. 66 14. 55 1. 54 339. 73
1994 16. 09 15. 57 O. 52 55. 66 14. 47 1. 62 370. 54
1995 16. 09 15. 43 O. 66 78.95 14. 43 1. 66 414. 14
1996 16. 09 15. 85 O. 24 34.93 14. 69 1. 40 403. 33
1997 16. 09 15. 86 O. 23 38. 54 14. 75 1. 34 423. 90
1998 16. 09 15. 37 O. 72 136. 63 14. 51 1. 58 557. 15
1999 16. 09 14. 45 1. 64 414. 00 14. 09 2. 00 817.79
2000 16. 09 14. 22 1. 87 553. 08 13. 97 2. 12 967. 53
..
Note: As the cel1mg mterest rate IS stIpulated at 13.5 per cent per annum for the
credit limit of up to Rs. 2 1akhs for CBs, the same has been assumed as the
maximum interest rate while calculating agricultural credit subsidy.
Column 7 in Table 3.14 shows that the credit subsidy per
Rs.I00 outstanding loan to agriculture sector was more during pre-
reform period as compared to the reform period. It was also true for
the credit limit of more than Rs.25,000 (see column 4). This reduction
in credit subsidy during the reform period can be attributed to the
70
interest rate deregulation. The interest Income from agricultural
lending for the credit limit of more than Rs.25,000 (column 3) was
greater than the interest income for the credit limit of less than Rs.
25,000 since the year 1991. However, the interest income from the
former slab of credit limit also did not cover the costs of lending. This
implies that cross subsidisation within the agricultural sector is not
possible, since the cost of providing service is more than the income.
Thus, the possibility of cross-subsidy within the agricultural
lending in India is ruled out. This is evident in Table 3.14 that
lending costs in the agricultural sector as a whole were more than the
income for the last two decades. The inference is, therefore, that
agricultural lending in Innia needs_credit subsidy. The amount of
credit subsidy to agriculture by all SCBs (including RRB) increased
from Rs. 77.33 crares in 1981 to Rs. 967. 53 crares in 2000. Such a
rapid increase in the amount of credit subsidy to agriculture implies
that banks had been pressurised to cross-subsidise agricultural sector
from the income generated by lending to the other sectors. This might
have resulted in two responses from the banks: Allocation of more
loanable funds to non-agricultural sector and/ or disbursing less
credit to agriculture. Has this really happened? The discussion also
calls for analysis of the following questions: Is increasing lending rate
a solution to reduce the rate of credit subsidy? To what extent the
supply of agricultural credit is elastic to interest rate change? These
are analysed in section 3.7.
Number of Rural Bank Branches and Agricultural Credit
The other important aspect of agricultural credit system IS the
availability of banking facilities at reasonable distance. In the wake of
losses to rural bank branches, it was recommended to close down the
71
loss-making branches or to merge with the other banks
27
As a result
of this policy decision, there has been a marginal decline in the
number of rural branches from 1994. In the Southern, Western and
Central Regions of the country, the rate of decline was more
pronounced (Table 3.15). A reduction in the number of bank branches
has not taken place in semi-urban, urban and metro areas. The
decline in the number of rural bank branches together with increasing
population pressure may have resulted in the following: i) A reduction
in the credit flow to agriculture; ii) loan demand and deposit supply
may have been adversely affected; iii) the growth of private investment
for agricultural development and financial deepening of the rural
sector may have been severely affected; and, iv) the access to banking
facility hy rural population may have come dO'Nn.
Table 3.15: Region-wise Growth Rate of Rural Bank Branches
(in per cent)
Year Northern North Eastern Central Western Southern Total
Region Eastern Region Region Region Region
Region
1981-91 5. 17 10. 01 7. 79 7. 37 5. 19 3. 84 6. 05
1992-00 - 0.77 - 1. 12 - 0.64 -1. 36 -1. 32 -1. 50 -1. 13
1981-00 1. 97 4. 34 3. 44 2. 63 1. 79 1. 05 2. 28
Sources: Reserve Bank of IndIa (Bankmg StatIstIcs) from the year 1981 to 2000.
Determinants of Supply of Agricultural Credit
An analysis of the achievement of SCBs in the provisioning of
agricultural credit reveals that the banks did not even reach the set
targets. Importantly, the proportion of credit to agriculture in the total
net bank credit declined especially during the reform period. Is
agricultural lending not a profitable avenue for banks? Is the interest
income from agriculture sector is less for the banks? We, therefore,
examine the factors preventing banks from extending credit to
21 In the case of RRB, the bank branches were merged with the sponsoring bank.
agricultural sector. However, this analysis has been done only for
credit flow to agriculture from scheduled commercial banks, and not
from co-operatives for two reasons. First, the banking sector reforms
have hardly any impact on the financial pattern of the Co-operatives.
Second, with widespread practices of rescheduling of loans (even in
the event of non-repayment), the data on finances provided by co-
operative credit institutions is unreliable. The a priori model on
supply of credit to agriculture is specified with the following variables.
(a) Investment in Securities
In a regulated lending system, given the interest rate and default
risk, the bankers may be more interested in investing their funds
in government and othp.r approved increasing
proportion of investment in securities out of total deposits may
leave the banks with less loanable funds. It may adversely affect
the channelisation of credit to different sectors of the economy, in
general, and agriculture, in particular. This may also negatively
affect the Credit-Deposit Ratio. Thus, INVS is expected to have
negative association with credit flow to agriculture.
(b) Rural Coverage of Banks
DUring the post-nationalisation period, banks were directed by the
RBI to give priority to rural and non-banked areas in the bank-
branch expansion to enhance the credit availability to the needy.
However, the problems of the FRBS, viz., mounting overdues, poor
quality of lending and recalcitrant attitude among borrowers
contributed to the cumulative loses, adversely affected the viability
and efficiency of the banking system. During the reform years, the
banks were directed to close down their loss making branches or
73
merge them with other banks. Thus, closure of loss-making banks
together with the increasing population pressure may have reduced
the credit flow to agriculture, and hence, the variable of RBB is
expected to have negative association with credit flow to
agriculture.
(c) Lending Rate on Agricultural Credit
From the supply side, interest rate plays an important role in the
flow of credit. If the interest income is greater or at least equal to
the cost of lending, it may positively influence the financial
institutions to increase the supply of agricultural credit. Thus, a
higher lending rate is expected to increase the supply of credit to
equation (1).
(d) Credit Subsidy
Credit subsidy is the difference between cost of providing credit to
agriculture and interest rate earned on outstanding agricultural
advances adjusted for defaults. Increase in the rate of credit
subsidy may reduce the supply of agricultural credit. Equation (2)
gives the details on the measurement of this variable.
(e) Agricultural Credit Supplied by the Co-operatives
An increase in the total agricultural credit supplied by the co-
operative institutions may reduce the extent of credit supply by the
commercial banks and RRB. The commercial banks, including the
RRB, are expected to follow a policy of mutual substitutability with
co-operative institutions in the provisioning of agricultural credit.
74
The OLS method has been applied here to examine the impact of the
above on the credit flow to agriculture by the SCBs. However, as per
the Durbin Watson 'd'Test positive auto correlation was observed in
the estimation. The Cocharine-Orentt method was applied to correct
the same. The description expected sign and estimated results of the
variables used in the analysis have been reported in the Table 3.16.
Table 3.16: Description, Expected Sign and Coefficient of
Variables Used in the Equation
[Deoendent variable = Prooortion of agricultural credit from total bank credit]
Variable Description Expected Coefficient .
Sign
Constant 26.82 **
INVS Proportion of investment in
government and other approved
-
- o. 16 ***
securities from total bank deposits
RBB Proportion of rural bank from total + 0.52 * I
number of banks branches
INTR Interest rate per Rs. 100 + - 2. 44 *
outstanding credit
CS Credit subsidy per Rs. 100 - -1.57***
outstanding loan
ACSCOP Proportion of agricultural credit - O. 05
extended by the Co-operatives
R-sauared 0.97
Durbin - Watson stat 2. 30
Number of observations 20
Note: * 1 % level of SIgnIficance; ** 5 % level of SIgnIficance; *** 10 % level of
significance
t - values
2. 86
-1. 86
8. 83
- 4. 74
-1. 84
1. 09
As expected, most of the parameters have the expected sign
except INTR and ACSCOP. The negative and statistically significant
coefficient of INVS indicates that proportion of investment in
government and other approved securities from total bank deposits
was negatively associated with the share of agriculture sector from net
bank credit. An increase in the investment in securities by one per
cent reduces the supply of agricultural credit by 0.16 per cent.
Probably, the positive returns have been giving incentives to divert
more funds on investment securities, and leaving the bank with less
75
loanable funds. This, in turn, reduces the supply of credit to
agriculture sector. Coefficient of RBB is positive and significant at 1
per cent level, which implies that improved banking coverage in rural
areas steps up the supply of agricultural credit. The coefficient of INTR
is negative and significant, which is contradictory to the theoretical
argument that an increase in the lending rate will increase the supply
of agricultural credit. Hence, increasing the lending rate may not be
an important means to increase the supply of agricultural credit. It
has been argued that an increasing interest rate or collateral
requirement increases the risks of the bank's loan portfolio by either
discouraging safer investor or by inducing borrowers to invest in
riskier project and therefore, decreases the bank profit (Stiglitz and
Weiss 1981). With the deregulFltion of interest rates in 1990s, the
extent to which this would be beneficial to enhance the supply of
agricultural credit is questionable.
The negative and significant coefficient of credit subsidy (CS)
implies that 1 per cent increase in credit subsidy reduces the supply
of agricultural credit by 1.57 per cent. In this context, can we adopt
increasing lending rate as a means to reduce the rate of credit
subsidy? Since increasing the lending rate itself reduces the supply of
agricultural credit, this cannot be a sound approach to reduce the
credit subsidy. As agricultural lending needs credit subsidy, the net
interest income out of deployment of funds in the agriculture sector is
negative. Thus, the profit oriented norms probably encourage the
bankers to release a less proportion of net bank credit to the
agriculture sector.
76
The positive coefficient of ACSCOP rejects the assumption that
commercial banks are required to follow the policy of mutual
substitutability with co-operative institutions In the field of
agricultural credit. Contrary to this, the positive coefficient implies the
mutual complementary relationship between the two sources of
agricultural credit. In other words, the commercial banks envisage the
agricultural credit policy by strengthening the supply of credit where
co-operatives are weak. This is also corroborating with the findings
that of increasing share of commercial banks and RRB in the supply
of total agricultural credit. However, this coefficient is surprisingly not
significant.
Conclusions
This chapter analysed the trends in credit flow to agriculture by SCBs.
The analysis brought out that the proportion of credit flow to
agriculture had significantly declined during the last one decade.
Such a decline was more prominent during the reform years. Across
the bank groups also, a similar decline was observed. In the case of
co-operatives the situation was grimmer. More importantly, despite
the fact that the lending targets were fixed, direct and indirect finance
was clubbed, interest rates were deregulated and lending procedures
in the credit delivery system were simplified, the banks could not
achieve the targets set in the case of agricultural lending.
The analysis shows that the number of agricultural accounts in
the credit limit of Rs.25,OOO and the amount of credit in this range
declined during the reform period. The analysis also shows that,
within agriculture, banks provided larger quantum more funds to
activities earning more interest income. This seemed to be more
77
prominent after the deregulation of lending rate to ultimate borrowers.
The better-off farmers had larger access to formal credit as compared
to small and marginal farmers. Importantly, despite the increasing
consensus to extend credit facility to agriculture, in general, and small
and marginal farmers, in particular, banks were unable to lend to
those activities with high social return or those category of
creditworthy borrowers who had been traditionally marginalised in the
credit market. This finding is also backed by Stiglitz's study in 1994.
This raises the issue of what factors determine the supply of
institutional credit, and access to institutional credit. While the first
Issue is analysed in this chapter, the latter issue has been taken up
for a detailed analysis in chapter VI with the help of household level
data.
Credit flow to agriculture was negatively associated with
Investment in government securities, credit subsidy and proportion of
credit provided by the co-operatives. Credit supply to agriculture was
positively associated with rural bank branches. This implies the need
to reconsider the policy suggestion of closing loss-making rural bank
branches. If it is absolutely essential to close down the bank
branches on considerations of profitability, alternatives in terms of
farmers' clubs, self-help groups, etc., it should be introduced, and
strengthened.
The analysis shows that increasing lending rate reduces the
credit supply to agriculture sector by institutional agencies. This
leads to the policy implication that it may not be sound to increase the
interest rate. Besides, banks cannot use the interest rate as screening
de\ise due to adverse selection problem. This is due to the fact that
78
borrowers, willing to pay high interest rates, may be less worried
about repayment of the loan. In this context, given the information
asymmetry, Stiglitz and Weiss (1981) shows that a profit maximising
bank will practice credit rationing and be reluctant to increase interest
rates in response to an excess demand. Consequently, high-risk group
of borrowers may be completely excluded from the market although
their prospective investments offer a high-expected return. Since
agriculture is a risky activity where farmers face shocks to their
mcome, it appears that bankers in India have been practising a tight
credit rationing process to select a borrower. Given these problems
one of the basic questions is that what makes certain households to
access institutional credit while others cannot? This issue has been
taken up in great detail \ ~ ' h i ! e making the household level analysis.
The calculations on credit subsidy show that the total amount
of credit was increasing, although the credit subsidy per Rs. 100 of
outstanding loan declined during the reform period. It is also found
that credit subsidy adversely affected the supply of agricultural credit.
It is also noticed that increasing lending rate could not be an
appropriate measure to reduce the credit subsidy, since it adversely
affected the supply of agricultural credit. In this context, it can be
suggested that reduction in the cost of lending in terms of either
fmancial cost or transaction cost or risk cost or by any combinations
of these three should be the prime objective of the financial
institutions to reduce the burden of credit subsidy. This can be made
possible by better quality of credit delivery system (timely and
adequate) and prompt repayment of loans.
79
Chapter 4
INSTITUTIONAL CREDIT TO AGRICULTURE: STATE AND
DISTRICT LEVEL ANALYSIS
Introduction
In Chapter III, it is noted that the proportion of credit disbursed to
"griculture to total net bank credit significantly declined during the
period 1992-2000. Further, despite that lending targets were fixed,
direct and indirect finance was clubbed, interest rates were
deregulated and lending procedures were simplified, the institutional
1gencies could not meet the targets fixed in the case of agricultural
IL'rtding. The discussion on distribution of loan accounts and
n:lOunts by size class of landholdings in Chapter III suggests that the
bt'tter-off among the farmers were having larger access to the
mstitutional credit as compared to small and marginal farmers.
despite the widespread consensus to extend credit facility
t(l agriculture, in general, and small and marginal farmers, in
particular, the banks could not finance this sector with high social
;'('turns and cover those category of borrowers who were marginalised
:n 1 he formal rural credit market. Chapter III also makes it clear that
ll1stitutionallending agencies were giving more importance to non-
,c;ncultural lending. Elsewhere, it is argued that relatively backward
, had less access to institutional credit than the developed ones
''<':id\' 2001), In this context, one can assume that the flow of credit
(',IIerent categories of farmers and regions was far from uniform,
Based on the discussion and conclusions in Chapter Ill, one can
"'l' the following questions: Is the supply of agricultural credit
.,., Lrtll1g in all the states and districts? Given the inter-regional
80
l!lfferences m the irrigation infrastructure, use of high-yielding
larteties, fertiliser consumption, etc., is the decline in the credit 110\\
ll!1lform across the states and districts? Does the access to
Illstitutional credit vary by size classes of the landholdings across the
!"l'gions? Is there any variation in the supply of agricultural credit per
:\tctare of gross cropped area across the states and districts? What
dre the activities (both agriculture and non-agriculture) for which
Illstitutional credit has been disbursed? What has determined the
s,ctoral distribution of institutional credit?
Rationale of State Selection and Database
This chapter seeks to address the above questions by confining the
.1:1alvsis to 14 major states. These s t a t ~ s Flrp Andhra Pradesh, Bih3.r
I undivided), Gujarat, Haryana, Karnataka, Kerala, Madhya Pradesh
IUlldivided), Maharashtra, Orissa, Punjab, Rajasthan, Tamil Nadu,
littar Pradesh (undivided) and West Bengal. The rationale for the
selection of these states is as follows: the 14 states account for 93 per
cent of the population and 91.5 per cent of the Net Domestic Product
I\;DP) in the country (Rao et aZI999); second, these states accounted,
lin an average, 84.7 per cent of the total deposits in Scheduled
l. 'mmercial Banks (SCBs) including RRBs during the last twenty
\["rs; and third, 95 per cent of the total agricultural credit provided
1)\ seBs and RRBs went to these states
28
The hill states located in
: :u:1 h and north-eastern parts of India, considered as special category
SI,,,l'S by the Planning Commission, have been excluded from the
.iii.: lysis as there are significant differences between these states and
. "i,C1s in terms of principal sources of income. The important sources
: ;; . ". data have been obtained from Banking Statistics pubilshed by the Rill.
Sl
of income m the hill states are government activities, as there has
hardly been any manufacturing base (Rao et al 1999).
The district level analysis has been done for the state of Orissa
because of its backward status. The Economic Survey of Orissa for the
year 2000-01 notes with concern that the Credit Deposit Ratio
29
(CDR)
m the state declined from 80.59 per cent in 1991 to 39.75 per cent in
1999-2000, while per-bank branch population increased from 15,200
to 16,300 during the same period. The reduced CDR may have
adversely affected the credit flow to different sectors in the state
economy. In this context, there is need to look into the actual flow of
credit to different sectors in the state, in general, and to agriculture, in
particular.
The analysis in this Chapter is based on secondary data. For
the analysis of trends in 14 major states, the data were collected from
the Reports on Currency and Finance (RBI), Banking Statistics (RBI),
Fertiliser Statistics (the Fertiliser Association of India), Indian
Agricultural Statistics (Ministry of Agriculture, Government of India)
and Statistical Abstract (Central Statistical Organisation, Department
of Statistics and Programme Implementation, Government of India).
For an analysis of trends in Orissa, data were collected from State
Level Bankers' Committee Meetings (Agenda Notes) and Orissa
Agricultural Statistics (Directorate of Agriculture and Food Production,
Orissa, Bhubaneswar).
29 The CDR in Orissa was well below the all India average of 57.05 per cent in 1999
2000. In that year, the states with CDR higher than that in Orissa were: Tamil
l"adu (88.0%), Maharashtra (83.79%), Andhra Pradesh (63.60%), Karnataka
(61.01 %), Gujarat (49.85%), Madhya Pradesh (49.17%), Rajasthan (46.69%), West
Bengal (45.17%) and Kerala (42.29%). This information was collected from RBI
(2000)
82
State-wise Flow of Agricultural Credit
Table 4.1 provides the growth rates on agricultural credit for the
period 1981-2000 in 14 major states. It also provides growth rates for
credit to agriculture and total credit for two sub periods, viz., 1981-
1991 and 1992-2000. Total credit (for agriculture and others) in all
the 14 major states grew at the rate of 15.3 per cent during 1981-
2000. There have been temporal variations; the growth rate during
the reform period was higher than that during the pre-reform period.
The pattern of growth rates in the case of credit to agriculture
\\'as, however, different. The credit to agriculture in 14 major states
mcreased at an annual growth rate of 12 per cent during the entire
period. However, the growth ratl;' during the reform period '.'fas less
110.9 per cent) than that during the pre-reform period (15.2 per cent).
This clearly shows that the rate at which credit was disbursed to
agriculture slowed down during the reform period.
This pattern can be observed m all the states, except Punjab
and Maharashtra. A comparison of growth rates of credit to
dgriculture between pre- and reform periods reveals that most of the
'tates registered higher growth rates of credit flow to agriculture
'iuring the pre-reform period as compared to the later period.
83
Table 4.1: State-wise Annual Average Growth Rates (%) of
outstanding Credit by SCBs
States 1981 - 1991 1992 - 2000 2000 =:J
_____ --:-Agrlculture 'fotal . Agriculture _ Total
High Income States
-----------r--------T"'- - -- --- - - ----------
l,ularat . __ ' 14. 7 ' --"- __ J6,Q _._
1I";"an';' 14. 3 , 14.9: 9. 3 14. 7
________ -t ______ ---t ______ " ____ - - -;- ____ .. _
\lriharas_htr"--___ __ I:3.:!l.. i 13. 9 __
!'unjab _ 2..---+_-'1:JLJ __ 7 ___ " .14..c S. _ .. _
13. 0_--, .. 14._0_ 10. 2
12.4
9. 9
12. 2
15. 4
13.
16. 2
9. 2 13. I
11. IS. 5
- -- - ------
_____ ._ -:-_--:-::--::--"M:::: idd1e_!nco-",_e. Sta ___ --=----c -c-c-
,\ndhra Pradesh I 13. 5 _!---,1,-,7.:.--.9 I 12. 2 . IS. 3, 11. 4 15.9!
Karnataka 17. 16. 6 0-=::::15.]:
Kerala I, 15.2 14.8 j 12.6 i 16. l' 12. 1 __ 14 .. _1_,
Temil ____ 1=7'-.. cc3_-+--cl-':c7'-.. cc6_t-i' ____ 1--=0-'-. --=3_-,-1 17. 2: 13. 2 16. 7
West Bengal I 17.5 14.5 I 3.5 '11.5-1 10.1--13.:'3--
1
GrouI' Iotal , 15.9 16. 4_--.l __ __ !_'5:_Lj=-::::12:-2--15-:'3:::::;
Low Income States
i3,har- ==-_---,---:'-::6:-;2:--,-----"";15 9 ----43--, -107-' --11 2_-=X3 . ._J
\hdh.:a Pradesh; 19.2 20.8 I 12. ___ I:_3_
Onssa I 15.9 20. 7! 9.9 i II. 5 I 10. I . li-,--:?--j
i 16.1 lS8 1').2 15.6 12.7 l-r.:1
, ---,
Uttar Pradesh I 15. 16. 8 I 9.8 i II. 8 I 12. 8 14. 3
,-"oupTotal ! 16.3 17.6, 10. I 13. I 12.7 14. ?J
,
MilJor States 15.2 15.7 i . ..9...--E:-O-_-- IS. 3 '
Source: Reserve Bank of India (various issues of Banking Statistics from 1981 to
2UOOI
We have categorised the 14 major states into high, middle and
low-income groups on the basis of net state domestic product. A
comparison of growth rates across different categories of states shows
that the growth rates of credit to agriculture in 1980s were higher for
!niddle and low-income states as compared to the high-income
C dtegory of states (Table 4.1). In the case of states such as Bihar,
West Bengal, Kerala, Karnataka, and Maharashtra, the
t;r')\\'th rate of credit to agriculture was higher than that to all sectors
(i U ring the period 1981-91.
However, the growth rates of agricultural credit were much
b'lo\\' than those in the case of total credit for all the states during the
P,rtod 19922000. The growth rates of credit to all the sectors and to
8-1
agriculture declined during this period in the case of middle and low-
income states. But the decline in growth rates was steep in the case of
agricultural credit, especially in the case of Bihar, West Bengal and
c,
Gujarat. In the case of high income group states, although the growth
rate of credit to all sectors increased during 1991-2000, it decreased
in the case of agriculture. Thus, growth rates of agricultural credit
were uneven across the sub-periods as well as across the states.
What are the trends in the proportion of credit to agriculture
from SeBs for all the states? Table 4.2 provides the proportions of
credit to agriculture to total net bank credit for four five-year periods
beginning with 1981-85. In 1980s, the proportion of farm credit to
total net bank credit was around 18 per cent of the total credit in all
the major states. Thus, the institutional agencies were meeting the
target of 18 per cent fixed under the priority sector lending norms. In
nine out of 14 states, the priority sector norms were met during the
period 1981-85. The number of states meeting the priority sector
norms increased to 10 during the period 1986-90. In general, all the
low-income states, and most of the middle income states were meeting
the norms in 1980s.
There was perceptible change in the level of credit provided to
agriculture during 1990s. First, the proportion of farm credit to total
net bank credit in all the 14 major states declined from 14.7 per cent
during 1991-95 to 11.7 during the period 1996-2000. However, the
proportion of agricultural credit in the total bank credit progressively
declined to about 11.7 per cent during the period 1996-2000. Such a
decline was uniform across the states. Nevertheless, the priority sector
norms in relation to agricultural lending were met in nine and eight
85
states during 1991-95 and 1996-2000, respectively. Further, the
norms were met in all the low-income states and a few high and
middle-income states during 1996-2000. It has to be, however, noted
that if the proportion of net bank credit to agriculture in a particular
state is more, that does not necessarily mean that credit per hectare of
t.;ross cropped area is more. This implies that on the basis of
proportions of credit to agriculture it may be difficult to conclude that
a particular state is better-off in terms of amount of credit to
agricultural sector. Therefore, the formal credit available per hectare
of gross cropped area can be a better indicator to explain the
agricultural credit situation in a particular state. We have, therefore,
taken the supply of credit per hectare of gross cropped area at the
state level as the dependent variable in the
contributing to variations in the supply of credit to agriculture in
Section 4.5.
The trends in the proportion of credit to agriculture during the
last two decades, thus, suggest that the following factors have
influenced the overall figures for 14 states. First, the low proportion
of agricultural credit in the industrialised states of Maharashtra,
Gujarat, West Bengal and Tamil Nadu has had substantial influence
on the overall figures. In the case of West Bengal and Maharashtra,
the priority sector norms were not met even during both pre-reform
periods. Second, the low proportion of agricultural credit could be
attributed to a low growth rate of agricultural credit compared to the
total credit in all the states especially during the reform period. This
shows that bankers were giving priority to non-agricultural lending in
almost all the states.
86
Table 4.2: State-wise Agricultural Sector Lending (%) to Total
Credit and Total Deposit
States
\larat
lr\-ana
(II
ii.
\1 ailarashtra
nlab
Qun Total
:rlhra Pradesh
l:-nataka
'rala
Ta mll Nadu
,'st Bengal
; 11! 'oup Total
8: har
adhya Pradesh
00
nssa
Jasthan
U
Ra
.tar Pradesh
-uup Total
, AI I States
Average for the Average for the
Period of Period of
1981-85 1986-90
ACHC I ACHD ACHC ACHD
High Income States
15. 2 B. 0 14. 4 B. 4
31. 3 29. 1 28. 5 22.7
6. 4 5. 5 7.3 4.8
33. 4 19.5 28. 3 12.6
13. 0 9. 7 12.7 7.9
Middle Income State.
35. 7 25. 7 29.5 23. 1
23. 2 18.4 24. 3 20.9
17.5 11.5 lB.3 II. 8
16.9 14.5 17.4 16. 3
7. 1 3. 3 9. 7 4.8
19.5 13. 0 19.6 14. I
Low Income States
24.5 9. 2 25. 7 8. 8
h
, C
?
24.8
.< n
v &v. V iU. V
31. 3 23. 7 25. 3 23. 2
27. 9 19.8 25.9 lB. 3
24. 2 10. I 22. 2 9.0
25. B 12. 6 24. 7 12. 1
17.9 II. 7 18. I II. 2
Average for I
Average for
the Period of the Period of
1991-95 1996-2000
ACHC I ACHD I ACHC I ACBD
14. 1 7. 6 9 9 5. 4
25. 0 16. 0 19. 2 10. 4
5. 0 3. 0 4. 1 3. 1
22. B 10. 1 lB. 6 7. 6
9.6 5. 6 7. 1 4.6
22. 6 17. 5 19.7 14. I
20. B 14. 6 16. 9 II. 6
15. 8 7. 6 13. 5 5.9
14. 1 12. 4 10. 7 9. B
6. 5 3. 3 4. B 2. 2
15. 6 10.4 13. 0 8. 4
25. 7 9. 7 19. B 5. 4
",.... ,
13.4 i 9. 5 10. I J
20. 7 13. 4 19.0 9. 0
25. a 13.9 21. 3 10.5
21. 8 9. 2 19.5 6. 3
23. a 10. 9 19.8 7. 6
14. 7 8. 6 II. 7 6. B
ACBe - Agncultural sector credit as a proportIon to net bank credit.
ACBD = Agricultural sector credit as a proportion to total bank deposit.
Source: Reserve Bank of India (various issues of Banking Statistics from the year 1981 to
2000).
Let US now turn our attention to the discussion on trends in
.\gricultural Sector Credit as a proportion to the total bank deposit
l/leBD). This indicates the extent to which the deposits mobilised in a
Sidle were lent to agricultural sector. The key findings emerging from
dll analysis of trends in ACBD in 14 states put together, and in
IIldl\"ldual states are the following:


During the period 1981 to 2000, ACBD declined progressively in all
14 major states taken together.
There was also a decline in ACBD in the individual states except in
the case of a few states during 1986-90.
87
There were substantial inter-state differences in ACBD in one sub-
period as well as between four sub-periods. The poorer states such
as Bihar and Uttar Pradesh had lower ACBD all through the period
of 1981-2000 and even during the period 1981-85. In
agriculturally prosperous state of Haryana, the ACBD was high for
most part of the last two decades. The situation in Bihar and Uttar
Pradesh in terms of ACBD was even worse during the period 1996-
2000. In Bihar, ACBD dropped from 9.2 to 5.4 per cent during the
period 1981-2000, while in Uttar Pradesh it declined from 10.1 to
6.3 per cent during the same period. Interestingly, among 14 major
states, it was West Bengal, a middle income state, which had the
distinction of having the lowest ACBD throughout the last two
decades.
The gap between proportion of agricultural credit to total net bank
credit (ACBC) and ACBD was generally large in the poorer states as
compared to the agriculturally prosperous states. For instance, the
gap between these two was as high as 15.3 and 14.1 in Bihar and
Uttar Pradesh, respectively, as compared to only 2.2 in Haryana
during the period 1981-1985.
It can be also seen that ACBD was less than ACBC in all the 14
states during all time periods. A constant decline in the ACBD in
all the major states indicates that banks have been diverting their
funds to other than agricultural sector. Such a diversion seems to
be more prominent in the backward states like Bihar, Uttar
Pradesh, and West Bengal.
The extent of diversion of funds out of deposits mobilised was
equally high in developed states of Maharashtra, Kerala, and
88
Gujarat. This implies that agriculture in a state belonging to a
high-income category need not necessarily receive higher priority in
terms of allocation of credit. Thus, the income status of the state
(per capita income in this case) itself does not determine the supply
of agricultural credit.
We presume that supply side factors
determine the level of credit made available to agricultural sector.
Such a presumption is borne out by the fact that the high per
capita income in a particular state need not necessarily be due to
high agricultural income. The non-agricultural sectors may have
played an important role in the level of per capita income. If this is
the case, the bankers may give priority to non-agricultural sectors
in their lending portfolio purely on the grounds of profitability.
Since the proportion of agricultural credit to net b ~ . n k credit has
been declining over a period of time, it can be suggested that the
bankers may not be considering lending to agricultural sector as
profitable.
In the all India level analysis, we found that the proportion of
investment in government and other approved securities from total
bank deposits, interest rate, and credit subsidy were the important
itctors that were adversely affecting the supply of agricultural credit.
Smce the same type of data is not available at the state level, we have
used state's characteristics
30
to examine the supply of agricultural
credit.
'.'
30 I t includes the percentage of irrigated area to gross cropped area, area under
commercial crops to gross cropped area, credit deposit ratio, and density of bank
branches per 1,000 farmers.
89
Distribution Pattern of Agricultural Credit
The discussion on trends in the credit flow to agriculture with the help
of growth and level of credit to agriculture leads us to the following
conclusions. First, there was a rapid growth of credit to agriculture in
(he 1980s. Such a rapid flow was generally attributed to the supply-
Jed approach adopted by the formal institutional agencies (Adams and
Vogel 1986; Gadgil 1994; Shajahan 1999; Kohli 1999; Shivamagi
2000). Second, the fall-out of such an approach could be considered
as the contributing factor for the decline in the credit flow to
agriculture in the 1990s. The literature suggests that a preoccupation
to achieve quantitative targets in 1980s ignoring the qualitative
aspects resulted in high cost structure of operation and mounting
overdues within the banking system. This hec:ame a threat to the
VIability of the financial institutions. As noted earlier, the introduction
of policies relating to market-oriented and competitive financial
system since 1991 aimed to improve the viability of the financial
institutions. Although the target of agricultural credit was kept the
same during the 1990s, there was a decline in the credit to
agriculture. This, perhaps, could be because of credit rationing
practices. As was found in the review of literature in Chapter 2, the
perceptions on risks involved in agricultural lending, the banks might
be forced to adopt price and non-price credit rationing to increase the
efficiency and viability of financial institutions in the provisioning of
agricultural credit. In order to see whether bankers have adopted price
and non-price credit rationing in the allocation of credit to agriculture,
there is need to examine the access of credit to different categories of
farmers. For this purpose, the data on the distribution of credit by
dIfferent size-classes of landholdings have been examined in this
scctlon.
90
Access to credit has been defined as the number of farmers
obtaining formal credit to total farmers in a state at a particular point
of time. For the analysis of distribution of credit by size-classes of
landholdings, we have relied upon the data on operational holdings
provided in quinquenial agricultural census. As per the latest
published agricultural census data, state-wise number of operational
holdings are available only for the years 1980-81, 1985-86 and 1990-
91. The data on the total number of holdings in each state for these
time-points have been used to examine the access to credit (Direct
Finance to Farmers) by different categories of farmers. However, for
the year 1996-97, the number of farmers (operational holdings) has
been calculated by the extrapolation method
31
. Since state-wise
number of accounts on direct fmance is available up to 1996-97, we
have calculated the number of farmers for the latest year to examine
the position of access to credit. The state-wise number of outstanding
loan accounts on direct finance has been taken into account for the
analysis. The data on number of loan accounts have been obtained
from the Reports on Currency and Finance, and number of farmers
from Fertiliser Statistics.
The distribution of agricultural credit at the all India level shows
that the farmers with more than 2 ha of operational holding were
having larger access to the formal credit as compared to those with
less than 2 ha (small and marginal farmers) (Table 4.3). The same
:rend could also be observed at the state level. There were substantial
lIlter-state differences in the access to formal credit by the same
, cltegory of farmers at one time point as well as between different time
,
, .--------------------

The number of fanners in the current year (y,) = Yo (1 ... r) t, where Yo stands for
number of farmers in the base year, r stands for rate of growth, t represents time
penods, and r = (Yt/ Yo)'/' - 1.
91
ihllnts. The access to formal credit by the small and marginal farmers
ilCld gone down in the year 1996-97 as compared to 1990-91 in all the
,Lites except Gujarat. Among the states, the proportion of marginal
,me! small farmers having access to institutional credit was the
:li."hest in Punjab (26.7 per cent) In 1996-97 followed by Tamil Nadu
,JlHI Andhra Pradesh. However, In the states like Bihar, Madhya
Pradesh, Uttar Pradesh, Haryana, and Maharashtra only 5 to 6.7 per
c , ~ n t of the marginal and small farmers were having access to formal
credit in the same year. For farmers with more than 2 ha. of
(,ptTational holdings (medium and large), 63.2 per cent of them were
we-essing formal credit in Tamil Nadu followed by Kerala (45 %), and
Punjab (31.6 %) in 1996-97. The percentage of medium and large
f8rmers having access to institutional credit was substantially less in
the states like Madhya Pradesh, Rajasthan and Maharashtra. It can
be. therefore, concluded that even the same size of landholdings were
h,i\'ing uneven access to formal credit across the states. This finding
holds good for all sizes of farmers. In this context, it is necessary to
examine what factors contribute to some fann households to access
(annal credit, while others do not. Since this question cannot be
111S\\cred with the help of state level analysis due to non-availability of
ij,lid. this question is taken up for analysis at the household level.
The outstanding credit (direct finance
32
) per loan account varies
.'.,;I,in and across different categories of states at one time point as
Ilie direct finance covers both short-term and long-term loans. Short-term loans
!::kluding crop loans) are given for the purchase of production inputs, such as,
sPcds. fertiliser, pesticides, etc., and to meet the cost of cultivation, which
i::, 'eldes labour charges for carrying out agricultural operations, irrigation
'Cl..rgCS. etc. Term (medium/long) loans are granted for development purposes
",,>, development of irrigation potential, purchase of tractors and other
,"::cultural implements and machmery, improvement of land, development of
,\!.",tatlOns, construction of godowns and cold storages, purchase of pump
v:s oil engines, plough animals (bullocks). etc.
well as between different time points. The amount of credit per loan
account was higher in the states belonging to high-income category as
compared to those in the middle and low-income categories. On an
average, small and marginal farmers in Haryana obtained Rs.22,181
of credit per loan account in 1996-97. On the other hand, farmers
belonging to the same size of operational holding in Orissa obtained
only Rs.4,862 per loan account. Such variations can be found even
among medium and large farmers in so far as the amount of credit
obtained per loan account across the states.
Table 4.3: State-wise and Size Class-wise Access to Direct
Finance by the Farmers
_____ !inEer. cefl!L
States i Up to 2 Hectares I Above 2 Hectares .. '

uj,arat G
H
M
Pu
G
aryana
aharashtra
njab
rouo Total
5. 6
13. 5--
7. 2 11. 6
3.6 7.4
21. 5 38. 0
5.6 11. 1
High Income States
12.9
- - ---r
14. 1 i 6. 4
11. 2 5. 0 11. 6
9. 4 6. 7 5. 4
41. 2 26. 7 22. 7
12. 0 8. 8 8. 0
_ ___ Income States
Andhra 16.0 24. 7 23. 3 20.1! 16. 3
Pradesh
Karnataka 11.3 20.9 19.1 11.3 9.2
Kerala 15.0 21.0 17.6 15.5 21.3
Tamil Nadu 11. 8 19.6 25.8 20. 3 11. 7
West Bengal 7. 4 12. 9 12. 0 10. 8 I 9. 2
Grall]) Total 12.3 19.8 20. 1 16.3 12.6
Low-InC"O-me States
-
Alh ar
dhya Ma
Pra
Or
Raj
Cit
Pra
Gr
desh
lssa
asthan
ar
desh
oup Total
3. 6 5.6
4. 0 7. 7
8. 3 16. 2
3. 3 9. B
3. 3 5. 5
3. 8 6.9
6. 7 5. 0 5. 7
9. 3 5. 8
I
6. 0
IB.4 14. 4 9. 7
12.4 B. 2 4. 4
6. 3 5. 2 11. 3
B. 1 6. 2 7. I
I
14. 4
I
17.5
I
10.6
17.7 i 18. 7 11. 4
10.7 I 12. 0 9. 1
29. 7
,
42. 2 31. 6
I 14.2 I 17.0
i
11. 8
I 23_6
.-----
, 21. 9 i 24. 5
,
,
15.7 I
51. 2 :
24_ 5 :
11. 5 I
20. 5 I
11. 0 I
7. 2
,
I
I
,
11. 7 I
6. 8
I
lB. 2
I
:
10. 4
I
16.4 i
45.0 I
44.9 :
12.4 I
23.2 I
11. 7
,
,
8. 0
,
,
11. 7
,
15. 8 '
45. 0
63. 2
17. 1 '
30. 5 -,
11. 6
7. 1
20. 2
___ 5. 0_
21. 4
I
IB.9
I
,
i
11. 3 I 10.3
"ole: FIgures on the number of accounts for all the years have been gIven for the
period July-June except for the year 1980-81 (April to March).
Sources: 1) Reserve Bank of India (Report on Currency and Finance, Vol. 11, for the
year 1982-83, 1987-88, 1992-93, and 1997-98).
2) Fertiliser Association of India, New Delhi, 1999-2000 (Fertiliser
Statistics)_
93
Table 4.4: State-wise and Size Class-wise Outstanding Credit
(Direct Finance to Farmers) Per Loan Account
States 1 . .,!1p to 2 Hectares ' Above --_.
[19S0-S1 j19S5-S6 , ,!,;_1J196:97 _19_80-S.b
at Gujar
HaI}'a

l'unja
GrollI'.
na
rashtra
b
Total
-
,\nd hra
desh Pra
Kar
Ker
Tam
We
Gro
nataka
ala
il Nadu
st8_enga1
up Total
r
hy;::t
81ha
Mon
Prad
Ons
.RaJa
etta
Gro
esh
sa
sthan
- -
r Pradesh
\IP Total
3,145
2,938
3,280
I
2,940
3 121
1,913
2,226
1500
1,860
1577
1800
1598
3,051
1,359
2874
1603
1786
48901
5,7821
5,663
10,511
6,526
High Income States
7,
162
1
10,8001
7823
1
99271
8,349:
- -----
9,3891 14,iI49117,4 16,773:
22,1811 17,779: 20,8221 39,0
13, 1181
8,557LlI,I
21 7571 15,0671 28,3_
14,705 22,1
741 -
771 72,228
39' 33.330
63: 57,154
0241-:-681
Middle Income States
3,430 5,540
9,492'1 4,2181 7,073 11,332: 17,648
i
I
,
3,989 6,503 15,8791 6,665' 10,011 15,808
1
33,451
2645 5,095 8,897\ 7,425\ 10,695 16066\ 8,771
:t.505 5552 12,891
1
6,280 9,097 12,649' 19,857
1922 2,859 6595\ 8430' 24,813 23,448 44,562
3,172 5,278 1O,675! 5611\ 9,526 13,655 21,341
Low Income States
2581 3922 5550 6,688 6,951 20,315 26,471
4,071 7,937 10,960 5,416 10,099:
21,358
1
36,429
i
2.,.877 2,875 4,862 1,664 4,070' 8,490 8,535
3430 5405 8969 8244 15,801 27,881 47,733
5327 8,436 7733 10,552 16,895 32,096
3136 4938 7481 63731 10,172 193821 31,023
Source: Reserve Bank of IndIa (Report on Currency and Fmance, Vol. II, for the years
1982-83, 1987-88, 1992-93, and 1997-98).
One important finding from Tables 4.3 and 4.4 is that while the
percentage of farmers accessing formal credit during 1990s was
declining in most of the states, the loan amount per account kept on
increasing. This implies that there was concentration of formal loan in
favour of a particular group of borrowers, and that the banks
practised both price and non-price credit rationing methods in the
borrower selection and fixation of the credit limit. This is further
corroborated by the fact that farmers within the same size-class of
landholding obtained substantially different loan amounts across the
states. In this context, it is essential to examine what made some
farmers to obtain more credit from formal sector even when they fell
under the same category of holding. In other words, how did the
94
bankers fIx the credit limit for different categories of farmers? Since
the data on characteristics of farmers accessing credit are not
available at the state level, factors determining the amount of credit at
the household level have been taken up for discussion in Chapter VI.
However, an attempt has been made in this chapter to analyse
the variations in formal credit per hectare of gross cropped area. At
this stage, if we analyse the variations of credit per loan account, then
it may not represent the true position of agricultural credit In a
particular state. Because, this kind of analysis at the state level
explains the credit widening (percentage of farmers with access to
direct fInance), and does not shed much light on farm households
which are out of formal credit market. Against this backdrop, credit
per hectare of gross cropped area seems more relevant. Furthermore,
credit per hectare of gross cropped area takes into account
automatically the obtained amount of credit by the farmers. Thus,
higher the percentage of farmers accessing formal credit, higher will
be the credit amount per hectare of gross cropped area.
We have calculated the amount of credit per hectare of gross
cropped area in Table 4.5 to examine state-wise variations. This
analysis has been undertaken only for 1980-81, 1985-6 and 1990-91
as the data on area under different size-classes of operational holdings
are available only for these time points. As in the case of the access
and amount of credit obtained per loan account, there have been
substantial inter-state differences in the amount of credit per hectare
of operational holding within and across different sizes of farmers
(Table 4.5). Per hectare amount of credit obtained by farmers in low-
income category states was less than that obtained by the farmers
from middle and high-income category states. This holds good in the
95
case of both categories of farmers at all time points. For instance, the
,werage credit per hectare of operational holding in low-income
category states was Rs 659 for the farmer with holding up to 2
hectares in 1990-91, whereas, it was Rs 4,161 in the case of Punjab,
the high income state. In the case of other middle and high-income
category states like Kerala, Tamil Nadu, Andhra Pradesh, and
Haryana also per hectare credit was sufficiently higher than that of
low-income category states at one as well as different points of time
even for the same size of holdings.
Table 4.5: State-wise and Size Class-wise Outstanding Credit
Per Hectare of Operational Holding
__ __ -r ______ ________ .-____ ____ __
States Hectares. ____ '
____ I 199091 11980-811 1985-86 199.0_-9_1_'
Htdl. Incoft'le Sta+es
rat
I
'ana
1
Guja
HaT}
Mah
Purl)
Grou
arashtra I
ab
LP Total
ra Andh
Prade
Karna
Kerala
Tamil
\\est
Group
sh
taka
Nadu
Bengal
Total
ya
sh
B,har

Prade
Onss
liajas
Lltar

a
than
--- -
Pradesh
p Total
I
- -
181 6811
928'
231 842 1,478'
132 443 791;
629 4,115 4,161 '
189 768 1071
.
MIddle Income States
406 1,160
1,
785
1
i
282 933 1,411 i
784 1,995 3,797
1
377 1227 2,6071
183 383 524j
358 1,024 1,769
1
Low Income States
120 287 549)
147 382 888i
J
137 590 6771
110 389 775,
95 326 594'
114 358 659
108' 408 606
330 691
1,491-
89 247 456
,
618 924
2,08'!J
174 409 789,
,
141 i 355 5651
,
,
- -
1151 315 560'
3931 1,267

1,612
1731
"
528 1,39.;
222, 840 824,
1501 428: 7291
88: 188' 609',
52
1
128
319',
1
431 125

-------.<
461 141 257
220 504 972
81, 197
42-2'
oj
'/ote: Data on area under operational holdmgs has been collected from the FertilIser
Association of India, New Delhi, 1999-2000 (Fertiliser Statistics), and
corresponding loan amount for these categories of farmers has been collected
from the Reserve Bank India (Report on Currency and Finance, Vol-H) for the
year 1982-83, 1987-88, and 1992-93.
96
Determinants of Credit Flow to Agriculture at the State Level
It has been observed in Section 4.4 that there were substantial inter-
differences in credit per loan account as well as per hectare of
,lperational holding at one and different time points even for the same
Size-class of landholdings across statesc What explains these
differences? This section takes up this questionc Before we go into the
details of the analysis, two aspects need to be mentionedc First, since
the data on characteristics of households obtaining formal credit are
not available at the state levels, the analysis on variations in credit
availability per loan account is quite difficulL Second, due to
c1\Cailability of data on operational holdings, which is confined to only
three time points, a rigorous analysis is difficultc Therefore, an
attempt has been made in this section to the factors 'Nhich
,'on tribute to the variations in the amount of credit per hectare of
c;ross cropped area. Hence, credit per hectare of gross cropped area
,eGA) is the dependent variable in the analysis. The a priori model on
1 he determinants of supply of agricultural credit has been specified
\\Cith the following variables.
1:/) Percentage of Irrigated Area to Gross Cropped Area (AIR33)
From the supply side, flow of credit can be said to be depending
upon the lender's assessment as regards the repayment capacity of
the borrower. It is assumed that irrigation facility can increase the
level of production and in turn, the repayment of the loan.
Therefore, the percentage of irrigated area to gross cropped area
3- To calculate AIR we have taken data from Indian Agricultural Statistics
(Department of Agriculture and Cooperation, Ministry of Agriculture,
Government of India) for the years 198081 to 199293c For the remaining years,
the same data have been collected from Fertiliser Statistics (the Fertiliser
cc\ssoCIation of India, New Delhi), and Statistical Abstract {Director of Economics
iHld Statistics, Ministry of Agriculture)c
97
has been specified as an important variable that determines the
supply of credit. This variable is expected to be positively
associated with credit per hectare of gross cropped area.
(b) Percentage of Area under Commercial Crops to Gross Cropped Area
(CCP)
Higher the area under commercial crops34, higher will be the
demand for funds to meet the cost of production. Besides, the
bankers may assume that the commercial crop growers are safe
borrowers in terms of repayment. Thus, an increase in the
proportion of area under commercial crops to gross cropped area
may positively influence the supply of agricultural credit.
(e) Credit Deposit Ratio (CDR)
Deposit is one of the important sources of loanable funds. The
volume of credit is assumed to be positively related with the level of
deposits. The increase in the volume of deposits can enhance the
volume of agricultural credit provided that bankers perceive
lending to agriculture as profitable. However, the CDR can increase
due to either more flow of agricultural credit or non-agricultural
credit or both. Hence, even if CDR Increases, this may not
necessarily increase the supply of agricultural credit. Thus,
increasing credit deposit ratio mayor may not lead to more supply
of agricultural credit.
34 It includes area under oilseeds (viz., groundnut, seasamum, rapeseed and
mustard oil, linseed, castorseed, nigerseed, safflower, sun-flower and soyabeen).
cotton. Jute and mesta. tea, coffee. tobacco and sugarcane. These data have been
collected from Indian Agricultural Statistics (1980-81 to 1992-93). and for the
rest of the years from Statistical Abstract.
98
(d) Density of Bank Branches per 1,000 Farmers (DBB)
It is explained in the literature that the problem of mounting
overdues, poor quality of lending and recalcitrant attitude of the
borrowers contributed to the cumulative losses to formal financial
institutions during the pre-reform years
35
. This adversely affected
the viability and efficiency of the rural banking system. Therefore,
during the reform years and especially after the financial year
1993-94, the loss making bank branches were directed to close
down or merge them with their sponsored bank branches. The data
show that only the rural bank branches have been so far closed
down. Thus, with the increasing popUlation size, the access to
banking facility by rural population might have come down. Hence,
it is important to see the relationship between banking facility and
provisioning of agricultural credit.
Since we are trying to measure variation in credit per hectare of
gross cropped area across the states, we have taken into account the
density of bank branches per thousand farmers. As per the latest
agricultural census, the number of farmers for each state is available
only for the years 1980-81, 1985-86, and 1990-91. So, we have
calculated the number of farmers for the rest of the years by
extrapolation method (see, footnote 5) by using 1980-81 and 1990-91
figures, and calculated the density of bank branches for thousand
farmers in each state.
li For more detailed discussion on these issues, see Von Pischke, Adams and
Donald 1983; Bravennan and Guasch 1989; Khusro 1989; Rajasekhar and
Ylasulu 1990, Vyasulu and Rajasekhar 1991; Kah10n 1991.
99
Model Specification
Thus, in the model, the dependent variable "credit per hectare of gross
cropped area" (CGA) is a function of the explanatory variables of AIR,
CCP, DBB and CDR. Since different states have different
characteristics, we have used panel regression method to assess the
individual impact of explanatory variables. Another reason for using
the panel regression method has been to get greater flexibility in
modelling differences in behaviour across individual states. The
individual effect is assumed to be constant over time and specific to
the individual states. Hence, differences across the states can be
captured in differences in constant term
36
. In order to specify a
Functional form, two alternative functional forms, viz., linear and log-
linear models _have been compared for the pooled data.. Sargan's
criterion has been used to choose between the linear and log-linear
specification. The Sargan's criteria is computed as:
Where,
g ~ t a n d s for the geometric mean of the dependent variable of the linear
model.
,i\l stands for the Residual sum of square of the linear model.
\ is the Residual sum of square of the log-linear model, and
T represents the number of observations.
The result of Sargan's criterion is given below in Table 4.6.
It IS possible to allow the slopes to vary across the states. However, it reqUIres
considerable complexity in the calculation.
100
Table 4.6: Estimation Result of Sargan's Criterion
g Ou
v
T 5
916. 18
404491026.8 O. 132415 280 (3331471)280
According to Sargan's criterion, if '5' is less than one, the linear
model is preferred and if it is greater than one, then the log-linear
model is preferred. Since our estimation result shows that '5' is
greater than one, it indicates log-linear as the correct functional form.
The basic framework for using the pooled regression model can
be specified as
Yit = + /3' Kit + E it
There are k regressors in XiI excluding the constant term. The
individual effect, ai which is taken to be constant over time t and
specific to the individual cross-section unit i. As it stands, this model
is a classical regression model. If we take ai to be the same across all
units, then ordinary least squares provides consistent and efficient
estimates of a and p. There are two basic frameworks used to
generalise this model. The Fixed Effect approach and Random Effect
takes a, to be a group specific constant and group-specific disturbance
term in the regression model, respectively. With this background, we
have used Fixed and/ Random Effect model to estimate the pooled
regression parameters. The estimated equation is as follows:
log (GCA);t '" aj + 131 log (AIR);! + 132 log (CCP);! + 133 log (DBB);! + 134 log
(CDR);! + E it
Based on the least square residuals, we obtain a Lagrange
iVlultiplier (LM) Test statistic of 495.54 which far exceeds the 95 per
cent critical value for chi-square with one degree of freedom (3.84). At
101
this point, we conclude that the classical regression model with single
constant term is inappropriate for these data. Furthermore, the result
of the LM test is to reject the null hypothesis in favour of the Random
Effect Model. But it is best to reserve judgement on that because there
is another competing specification that might induce these same
results, the Fixed Effect model. Keeping the fundamental difference in
the two approaches in mind, we have calculated Hausman Test for the
Fixed and Random Effect regression. This is based on the parts of the
coefficient vectors and the asymptotic covariance matrices that
correspond to the slopes in the model, that is, ignoring the constant
term (s). The test statistics is 73.09. The critical value from the chi-
square table value with four degrees of freedom is 9.48, which is less
than the test value. The hypothesis t . ~ a t the individual effects are
uncorrelated with the other regressors in the model can be rejected.
Hence, we would conclude that of the two alternatives we have
considered the Fixed Effect Model as the better choice for the
interpretation, which is shown in the Table 4.7.
Table 4.7: DescriptioD, Expected Sign, aDd CoefficieDt of
Variables iD the Log-Linear Model [Dependent Variable = eGA]
Variables Expected Filled Effect Model Random Effect Model
Sign Coefficient t value Coefficient t - value
Constant
- - 2.38 ' 2.12
~ AIR + 2.85' 11.79 1.98 ' 10.36
I CCP
+ 0.26 1.29 0.64 ' 4.01
DBB + 1. 43' 6.98 1.22 ' 6.49
I CDR
+ - - 0.71 ' - 4.13 - 0.87 ' - 5.25
. Lagrange Multiplier Test = 495.54
Fixed vs. Random Effects (Hausman) a 73.09
, No of observations E 280
:-Iote: at I % level of Significance
Findings and Discussion
The determinants of supply of credit per hectare of gross cropped area
(eGA) are estimated by Fixed Effect Model (Table 4.7). Since it is a log-
log model, the coefficients can be interpreted as elasticity. In our fIxed
effect model, the explanatory variables such as area under irrigation
as a percentage of gross cropped area (AIR) and bank branches per
thousand farmers (DBB) have got positive sign as expected. Moreover,
both of them are signifIcant at 1 per cent level. The positive and
statistically significant coefficient of AIR indicates that 1 per cent
increase in the irrigated area as a percentage of gross cropped area
will lead to 2.85 per cent increase in the credit obtained per hectare of
gross cropped area. This suggests that, in any state, larger percentage
of gross area irrigated to gross cropped area leads to a greater flow of
agricultural credit by formal fmancial institutions. Since irrigation
facility reduces uncertainty of crops, bankers are probably giving
priority to lend more in the irrigated belt. The positive and significant
coefficient of DBB implies that 1 per cent increase in the bank branch
per thousand farmers will lead to 1.42 per cent increase in the credit
per hectare of gross cropped area. Thus, given the policy of directed
credit system, the states with higher DBB may be in a better position
in terms of providing qualitative credit to the agricultural borrowers.
During the reform years the loss making bank branches were directed
to close down or asked to merge with the sponsored banks. This
happened only in the rural areas. As a consequence, the DBB might
have come down, which, in tum, leading to reduced supply of
agricultural credit.
Though the variable CCP has an expected positive sign, it is not
significant. The coefficient for the credit deposit ratio is negative and
103
significant which means 1 per cent increase in the CDR leads to 0.71
per cent decrease in the credit per hectare of gross cropped area. The
negative sign of CDR can be attributed to perception among bankers
that loans to agriculture characterise high-risk. This suggests that
even if we increase the CDR it may enhance the supply of credit
outside the agriculture sector, which is considered by bankers to be a
sector with better returns. Hence, as long as the bankers do not
perceive lending to agriculture as profitable, increasing CDR itself is
not enough for better provisioning of agricultural credit. On the whole,
the farmers in the irrigated area or/and with high density of bank
branches are most likely to benefit from institutional lending agencies.
The intercepts of fixed effect model. for 14 states are glven In
Table 4.8. This difference in intercepts can be attributed to the unique
features of each state. Although the evidence supports that the Fixed
Effect estimates are generally held to be downward biased estimates of
the true effects, they are an improvement over cross-section data
estimates (Johnston and Oi Nardo 1997).
Table 4.8: State Specific Intercepts of Fixed Effect Model
!
States Coefficient t - values
IGularat - 0.78 - 0.66
!Har\'ana
- 2. 52
- 1. 86
~
,Maharashtra
2. 16
1. 88
,Punlab - 4. 41' - 3.40
,Andhra Pradesh O. 20 0.15
Ka rna taka 1. 23 1. 03
Kerala 3.58 3. 33
\:ramil Nadu 0.03 0.02
Illest Bengal
0.02 O. 02
'R1har
- 0.33 - 0.27
\l"dhya Pradesh 1. 01 O. 88
.Onssa
0.47 O. 39
PaJasthan
- 0.23 - 0.20
L' [tar Pradesh
- 1. 70 - 1. 32
, .
O l e , at I % level of slgmficance, at 10 % level of slgmficance
104
credit Flow to Agriculture in Orissa
The credit disbursed to agricultural sector in Orissa by SCBs
(including RRBs) was Rs. 10,541.9 lakhs in 1991 and it had increased
to Rs. 29,244.0 lakhs by 2000 at an annual growth rate of 14.11 per
cent. It is to be noted that the share of agricultural credit from total
bank credit has progressively declined from 55.5 per cent to 36.1 per
cent during the same reference period (Table 4.9). Interestingly, this
has been uniform across CBs and RRBs. The rate at which CBs
disbursed credit to agriculture was about 54 per cent in 1991 and it
increased to 59.8 per cent by the triennium ending with 1994. After
this period, the rate of credit disbursement to this sector gradually
declined and reached to 36.5 per cent by the triennium ending with
2000. In the case of RRBs, the disbursement rate declined from 64.7
per cent in 1991 to 35.4 per cent by 2000.
Table 4.9: Share of Agricultural Credit (%) from Total Bank
Cre
d' b E h G fB k It )y ac roup 0 an
Triennium Ending Commercial Regional Rural Total
,
With Banks Banks
,
1991' 54.0 64.7 55.5
,
1994 59.8 55.3 58.8
I
1997 50.2 41.8 47.6
I
2000 36.5 35.4 36.1
:-late: 1) * One year figure
2) The figures are from disbursement loan amount.
Source: UCO Bank (various issues of State Level Bankers' Committee Agenda Notes,
Orissa starting from the year 1991 to 2000).
The above raises the following issues. Why is that the proportion
of credit provided by both CBs and RRBs has been declining? Some of
the bankers (at the ground level) are of the view that since the retum
on agricultural investment is highly uncertain and is affected by several
jactors, and the probability of default rate is more, banks are reluctant
to extend their credit facility to agriculture sector. It is also argued that
the failure of the commercial banks in extending credit facility to the
\05
rural poor and especially to the farmers was attributed mainly to their
policy of collateral-based lending
37
(Shylendra 1996). The annual
reports of one of the RRBs in the state note that the bank has been
trying to lend more of its funds to non-agricultural purposes
particularly to those that come under the non-priority categories. This
was done mainly because the non-agricultural loans not only had
better chances of recovery but also fetched relatively higher interest
rates. Elsewhere, it has been argued that the increasing operating
costs and higher loans delinquency rates compelled the bank to adopt
such a strategy (Shylendra 1995). The less supply of agricultural
credit by RRB as a proportion to its total bank credit can also be
attributed to the policy of allowing the RRB to lend up to 60 per cent
of their incremental lending to non-target groups (RB! 1995).
From the disbursement figure one can suggest that there is
growing preference of lending to non-agricultural sector by both
commercial and regional rural banks. This preference for non-
agricultural loans can also be corroborated by the data on targets and
achievement with respect to different sectors within the priority
sectors lending. Sector-wise distribution of target and achievement of
all scheduled commercial banks in Orissa in priority sectors under
various annual credit plan during the period 1991-92 to 1999-2000
has been presented in Table 4.10.
Table 4.10 shows that within the priority sectors lending more
than 50 per cent of the targeted loan amounts were going to
37 Von Pischke, Adams and Donald (1983) and others argue that such a policy is the
result of cheap interest rates prescribed on farm loans. In order to sustain them-
selves financially the lenders adopt the credit rationing process by pursuing
collateral based lending, which ultimately leads to the elimination of small
borrowers.
106
agriculture sector up to the year 1994. After that, the targeted loan
amounts to this sector had progressively declined from 58.1 per cent
to 44.2 per cent by the triennium ending with 2000. This can be
attributed to the declined trend of proportionately targeted loan
amounts for crop loans and allied agricultural activities. Against the
total target of priority sector lending, the proportions of achievement
in the supply of all kinds of agricultural credit lies always well below
the target. Furthennore, this achievement rate under agriculture
sector has been declining since mid nineties. Therefore, as far as
agricultural credit flow is concerned, the perfonnance of scheduled
commercial banks can be summarised as lower achievement of lower
targets.
Table 4.10: Broad Distribution Target and
Achievements under Annual Credit Plan of SCBs RRB)
TrienniuJIl Crop A&rlcultural AIlled to ToW Small Services ToW
r
I
I
,
Ending Lou Term Lou
A&rlcalture Apiculture Scs1e
With (ATlAl Industries
Distribution of Targets %)
1991 21.7 16.5 13.3 51.5 21.9 26.5 100
(23,960.72)
1994 24.8 19.2 14.1 58.1 17.2 24.7 100
(23,747.46)
1997 19.2 18.5 11.4 49.1 19.4 31.5 100
1(43,807.10)
2000 17.4 19.3 7.5 44.2 17.5 38.3 100
I (76 906.58)
Distribution of Achievement to Total Targets %}
1991 17.9 16.3 9.8 44.0 6.6 28.7 79.3
1994 22.3 15.9 8.9 47.0 12.1 20.8 79.9
1997 22.1 14.1 10.9 47.1 13.6 38.2 98.9
2000 18.5 9.6 7.0 35.1 12.8 49.3 97.2
Note: 11 Figures m parenthesIs represent total target amount (Rs. m Lakhsl for the
above mentioned sectors in the Orissa state.
21 One year figure
Source: veo Bank (various issues of State Level Bankers' Committee Agenda Notes,
Orissa starting from the year 1991 to 2000).
It is, however, clear that banks show preference to lend to
services in comparison to other sectors within the priority lending. The
distribution of target and the rate of achievement to this sector show
107
that both have been increasing over a period of time. The proportion of
targeted loan amounts for the service sector range from 24.7 per cent
to 38.3 per cent, whereas, the achievement to total target vary from
28.7 to 49.3 per cent. This shows that the rate of achievement is
always more than the rate of target in the case of service sector.
Hence, the credit flow to service sector can be termed as higher
achievement of lower targets. On the whole, the higher rate of
achievement by scheduled commercial banks (including RRB) m
priority sectors lending (for instance, 99 per cent and 97 per cent in
triennium ending with 1997 and 2000 respectively) can be attributed
to a higher rate of achievement in service sector.
The foregoing discussion suggests that there is a decline in the
proportion of credit flow to agriculture sector from the total credit
disbursed. The demand side factors have been emphasised for such a
poor achievement in the supply of agricultural credit (Government of
Orissa 2000-01). First, the number of proposals coming from farmers
especially in the case of ATLs has been inadequate. Second, poor
recovery of bank loans. Third, the infrastructure in rural areas is not
adequately developed. Fourth, the credit absorption capacity among
the farmers in rural areas is very low. Fifth, banks do not have
adequate manpower to meet the growing credit needs of the farmers. It
is also argued that the lower achievement in the case of ATLs can be
attributed to the policy decision that no direct collateral security is
needed for loans up to Rs. 50,000. This suggests that the bankers
seem to be adopting non-price credit mechanisms in the case of ATLs.
Poor extension support, low irrigation coverage, untimely and
inadequate supply of inputs, and low level of fertiliser consumption
108
were identified as important constraints for a declining rate of
achievement in the case of production credit (NABARD 2001-02).
In the case of small-scale industries also a similar kind of
situation can be observed like agriculture. The rate of achievement is
very poor compared to the given target for this sector. Factors like lack
of adequate facilities for skill upgradation, inadequate infrastructure
facilities, lack of road communication to rural areas, no marketing
outlets for handicrafts items, low level of people's awareness, no
proper industrial survey, and absence of forward and backward
linkages were some of the important constraints attributed to such a
poor achievement in the supply of credit to the small-scale industrial
sector (NABARD 2001-02).
This suggests that bankers, by and large, exclude supply side
factors from their purview. Can the decline in the credit flow to
agriculture be attributed to profitability consideration getting reflected
in the form of price and non-price types of credit rationing in the
situation of interest rate controls (in the case of commercial banks)
and inability to have direct collateral security in the case of certain
loans? If this is the case, there is a possibility of having a situation
where demand exceeds the supply of agricultural credit. This raises
another set of issues. In the case of which type of farmers and
activities, the credit is rationed? Is credit rationing leading to a
situation where this is confined to a particular type of borrowers?
However, these questions have been addressed in the subsequent
chapters.
109
District Level Analysis on Agricultural Credit Flow
Prior to 1993, there were 13 districts in the state Orissa. In 1994, 17
more districts were created from out of 10 districts. Thus, the total
number of districts became 30. The bifurcation was done in such a
manner that one or more new districts were carved out of one of the
older districts. Thus, it is possible for one to rework the data for
undivided districts. The same has been done in this study as well.
Therefore, the information presented in this study relates to the 13
undivided districts.
There is considerable variation across these districts in the
levels of irrigation infrastructure, fertiliser consumption and area
under HYVs. All these will have significant influence on the flow of
credit to agriculture. After analysing the data on these variables for
each of the undivided districts, we have categorised the districts into
three groups, viz., advanced, intermediary and backward districts. The
dIstricts of Cuttack, Puri, Sambalpur and Ganjam come under the
advanced group, while Baleswar, Bolangir, Dhenkanal and Koraput
form the intermediary group. Finally, the districts of Kalahandi,
Keonjhar, Mayurbhanj, Phulbani and Sundargarh form the backward
group. Table 4.11 shows that such a grouping is justified.
The increased concentration of development can be seen from
the distribution of agricultural credit in Table 4.12. The advanced
districts take a share, on an average, 58.68 per cent of the total
outstanding loans disbursed to agriculture sector during the period
1981-85. This share ranges from 54.12 per cent to 60.05 per cent of
total outstanding credit for the remaining period mentioned in the
table. For intermediary districts, the share of agricultural credit, on an
average, vary within the range of 24.16 per cent to 26.38 per cent,
110
\\'hereas, these shares vary in between 14.95 per cent and 19.50 per
cent for backward districts during the last two decades. Seen in terms
of distribution of agricultural credit one can notice that the advanced
group of districts takes a lion share compared to intermediary and
backward districts. In this context, one can conclude that the flow of
credit to different regions is far from uniform.
Table 4.11: Some Selected Indicators of Spatial Distribution of
Development
Type of Districts % Share Irri,-ation
1987- 88 1994 - 95 1999 - 2000
: Ad\'anced 58.4 54.0 51.4
i;'termediarv 26.7 30.3 30.9
'-S'ackward 14.9 15.7 17.7
, All' 100 100 100
(2,062.831 (2,468.33) (2,512.02)
% Share of Area under High Yieldin Variety seeds
,\d.'.'unced 49.4 48.4 <15.5
Intermediary 31.5 33.5 32.7

19.1 18.1 21.8
, All ** 100 100 100
-
(1,781.07) (3,017.69) (3,577.51)
% Share of Fertiliser Consum.l'.tion
Ad\'anced 70.2 59.6 54.3
Intermediary 19.7 29.4 30.8
Backward 10.1 11.0 14.9
All **' 100 100 100
(151.62) (219.99) (359.94)
tiote: FIgures In parentheSIS In row (*) and (*') represent total area (000 hectares)
under irrigation and HYV in the state respectively. In row (***), figures in
parentheses indicate total fertiliser consumption (000 mt) in the state.
Source: Government of Orissa (Orissa Agricultural Statistics for the years 1987-88,
1994-95, and 1999-00)
Table 4.12: Type of District-wise Share (%) of Agricultural
C d't G' b SCB re 1 lven ,y
S
Type of Average for Average for Average for Average for
Districts the Period of the Period of the Period of the Period of
1981- 85 1986 - 90 1991 - 95 1996 - 2000
\dvanced
58.68 60.05 56.53 54.12
-
I
ntermediary
26.37 24.16 26.17 26.38
8
ackward
14.95 15.79 17.30 19.50
-\11
100 100 100 100
(18,234.35) (42,298.91) (50745.20) (79,996.96)
Flgures m parentheses represent total average of agncultural credlt {Rs. m lakhsl for each
penod.
-";ource: Reserve Bank of India (various issues of Banking Statistics starting from the years 1981 to
2000).
J J I
What are the trends in the share of agricultural credit from net
bank credit of SCBs for all the districts? Table 4.13 shows that
agriculture accounted for 19 per cent of the net bank credit for the
period 1996-2000 for the state of Orissa as a whole and meeting the
target fIxed under the priority sector lending. However, this share was,
on an average, 31.3 per cent during the period 1981-85, and it
declined to 20.7 per cent during 1991-95. This suggests that the
supply of agricultural credit as a proportion to net bank credit have
been declining in the state of Orissa. This was uniform across the
districts. Nevertheless, the proportion of agricultural credit from net
bank credit have been declining, still, the bank has achieved the fIxed
target for all the intermediary and backward districts except
Sundargarh, and also has achieved th.e target for Ganjam and
Sambalpur among the advanced districts. In the case of Sundargarh
district, the achievement rate of agricultural credit was far below the
prescribed rate during both pre and reform years.
During the period, 1981 to 2000, ACBD declined progressively
for the state Orissa as a whole from 23.7 per cent for the years 1981-
85 to 9 per cent during 1996-2000 (Table 4.13). However, the declined
ACBD during 1980s for Orissa could be mainly attributed to the
declined trend of ACBD among the intermediary districts. This ratio
declined uniformly for the entire district in the state during 1990s.
Table 4.13 shows that there were substantial inter-district differences
in ACBD both at one period of time as well as between four time
periods. Importantly, for the backward districts like Sundargarh, only
3.4 per cent of net bank deposit was going to agriculture sector as
credit. Even in the advanced districts like Cuttack, Gunjam and Puri
also, the ACBD varied between 6 to 8 per cent during 1996-2000.
I I")
Since both ACBC and ACBD have been declining for all the districts
during 1990s, this again confirms the result of all India and state level
imalysis that bankers were giving priority to non-agricultural lending.
Table 4.13: District-wise Agricultural Sector Lending (%) to
T k C d't d D .
otal Ban
re 1 an eposlts
Districts
'\lttack
c
.imjam
;1;1 p
S
G
ambalpur
roup Total
)aleswar
()Iangir B
[
h
G
lhenkanal
uraput
roup Total
alahandi
ecmjhar
!.'.'urbhanj
lhulbani

. roup Total
II Orissa
G
A
198185 198690
ACBC ACBD ACBC ACBO
29.1 24.1 20.0 20.4
35.8 18.2 39.1 28.4
29.0 20.6 19.0 18.9
28.9 21.7 32.8 30.8
29.8 21.5 23.4 22.3
36.6 27.4 30.8 28.7
53.2 97.4 44.1 75.4
34.2 20.2 27.9 21.6
41.3 49.1 35.8 41.6
42.2 41.8 34.4 36.1
46.1 67.2 41.9 65.4
19.1 22.1 21.9 22.4
31.6 29.7 25.7 24.9
60.8 63.4 43.7 38.3
9.3 5.1 12.7 7.9
24.8 19.4 23.5 19.6
31.3 23.7 25.3 23.2
199195

ACBC ACBO ACBC ACBO
16.8 11.4 [6.1 7.4
29.6 15.4 22.8 7.9
15.4 9.8 13.7 6.2
22.1 14.9 21.5 12.7
18.4 11.7 16.7 7.6
24.8 22.2 20.5 13.8
32.1 31.! 30.6 17.9
27.4 14.6 26.2 10.2
28.0 22.2 31.2 15.9
27.4 20.9 25.8 13.7
38.5 35.8 35.9 24.3
25.5 18.2 19.5 10.1
23.9 13.4 33.0 12.6
35.3 24.2 32.4 18.6
10.1 4.8 8.3 3.4
21.2 12.3 20.7 9.6
20.7 13.4 19.0 9.0
\,lle: ACBC = Agncultural sector credIt as a proportIon to net bank credIt.
ACBD = Agricultural sector credit as a proportion to total bank deposit.
Source: RBI (various issues of Banking Statistics starting from the year 1981 to
2000).
As we have mentioned at the state level analysis that, if more
(less) proportion of net bank credit is disbursed to the agriculture
sector that does not necessarily mean that the per capita credit
availability per hectare of gross cropped area is more (less) in that
state. Since this proportion does not take into account the gross
cropped area, it may be difficult to reach a conclusion on the basis of
proportion of credit to agriculture to total net bank credit that a
particular state/district is better-off in terms of the availability of
agricultural credit. The above proposition has been verified in Table
-+14.
J 13
Table 4.14. Type of District-wise Outstanding Credit Per
Hectare of Gross Cropped Area
Jin Rupees)
1
Type of Average for the Average for the
Average for the
Districts
Period 1986-90
Period 1991-95 Period 1996-00
~ Advanced 683 771 1,249
'Intermediary
342 423
747
r Backward
263
338 671
; All 458 537
925
Source: Government of Orissa (various Issues of Onssa Agncu/tura/ Statzstzcs from
1985-86 to 1999- 2000) and Reserve Bank of India (various issues of
Banking Statistics from 1986 to 1999-2000).
It can be seen from Table 4.14 that the credit availability per
hectare of gross cropped area varies across the regions and time
periods. Importantly, the regions/districts where banks achieved their
lending target to agriculture sector, the supply of credit per hectare of
gross cropped area was less in most of those regions. For instance, in
the advanced districts only 16.7 per cent of net bank credit was going
to agriculture sector and credit per hectare of gross cropped area was
Rs. 1,249 for the period 1996-2000 (Tables 4.13 and 4.14), whereas,
in the case of intermediary and backward districts, the share of
agricultural credit were 25.8 per cent and 20.7 per cent of net bank
credit, and the credit available per hectare of gross cropped area were
Rs. 747 and Rs. 671 respectively during the same reference period. It
shows that the per capita credit per hectare of gross cropped area in
intermediary and backward districts were far below than the advanced
districts. Therefore, the formal credit available per hectare of gross
cropped area could be a better indicator to examine the inter-districts
variations in the flow of agricultural credit.
Determinants of Supply of Agricultural Credit at the District
Level
To examine variations in the availability of credit per hectare of gross
lTopped area across the districts, we have used the same methodology
that we have adopted at the state level analysis. With the same
ckfinition, both dependent and independent variables were specified,
and by using the same functional format the fixed effect and random
effect models have been estimated at the district level analysis
36
.
Since the Langrange Multiplier Test statistics of 42.91, which
exceeds the 95 per cent critical value of chi-square with one degree of
freedom (3.84), we conclude that the classical regression model with
term in:::lnnT"('\nri-::ltp fn.,. th.:llC'A ,1.._
SI Eo \. \.. -' - - '- -- -rr- -1"" .... _ .... - .................... " .... UULQ .1 U.l Ll1\"'.l 1 " , Lile:
Hausman Test statistics of Fixed versus Random Effects is 7.45,
\\'hich is less than the chi-square table value (9.48) with four degrees
of freedom. Thus, the hypothesis that the individual effects are
uncorrelated with the other regressors in the model can be accepted.
Therefore, out of two alternatives we considered Random Effect Model
is the better choice for the interpretation, which is shown in Table
Although the evidence supports that the Random Effect
estimates are generally held to be upwards-biased estimates of the
true effects, they are an improvement over cross-section data
estimates (Johnston and Oi Nardo 1997) .
. SlIlce the specification of variables (both dependent and independent). functional
format and other details of methodology have been explained at the state level
analysis, the similar set of explanation has not been given to avoid repetition at
the district level analysis.
115
-
Table 4.15: Description, Expected Sign, and Coefficient of
Variables in the Log-Linear Model [Dependent variable = CGA]
" Variables
Expected Fixed Effect Model
Random Effect Model
,
Sign Coefficient t - value Coefficient t - value
i
, Constant -
<--
-
8.37' lO.41
AIR
+ 0.55' 3. 30
0.55' 4. 35
f----
+ - 0.45"
- 2. IS - 0.36" ' CCP
~
- 2. 21
DBB
+ O. 14 O. 30 O. 25 O. 82
'CDR
+, - - 0.69'
- 6. 31 - 0.66' - 6.85
CLagrange Multiplier Test 42.91
Fixed vs. Random Effects (Hausman) - 7. 45
i No of observations - 182
.. 0
Note: at I % level of slgmficance, at 5 Yo level of slgmficance
Table 4.16: District Specific Intercepts of Fixed Effect Model
Districts Coefficient
t - value
Cutlack
8.81 0.96
GanJam
8.44 0.93
Pun
9.15 O.Q2
Sambalpur 8.42 0.98
Baleswar 8.36 0.94
Bolangir 8.63 1.00
Dhenkanal 8.66 0.93
horaput 8.95 0.95
halahandi 9.00 0.92
heonjhar 8.66 0.92
Ma\'urbhani 8.35 0.90
Phulbani 8.62 0.94
Sundargarh 8.16 0.81
The positive and statistically significant coefficient of AIR
lOdicates that 1 per cent increase in the irrigated area as a percentage
of gross cropped area will lead to 4.35 per cent increase in the credit
obtained per hectare of gross cropped area (Table 4.15). This again
corroborates with our state level analysis that higher the percentage of
gross area irrigated to gross cropped area, higher will be the flow of
agricultural credit by formal financial institutions. The negative and
significant coefficient of CCP indicates that increase in the area under
commercial crops leads to decrease in the availability of credit per
hectare of gross cropped area. Probably, factors like poor extension
116
sl'lyice by other line departments, lack of timely availability of
ltTlgation facility, untimely and inadequate supply of inputs, and low
i,'\-ei of fertiliser consumption can be attributed to this negative
dssociation (NABARD 2001-02), As a result of this, the bankers may
ill' adopting non-price credit rationing mechanism for commercial crop
gmwers, Though the variable DBB has an expected positive sign, it is
not significant.
The coefficient of CDR is negative and significant at 1 per cent
level. This suggests that one per cent increase in the CDR reduces the
supply of agricultural credit by 6.85 per cent. This finding also
support the result of the state level analysis that increasing CDR it-
self is not enough for better provisioning of agricultural credit. On the
\\hole, the farmers in the higher irrigated regions are most likely
beneficiaries of agricultural credit. The fixed effect intercepts of 13
districts, which are presented in Table 4.16, one can argue that this
difference in intercepts can be attributed to the unique features of
each district.
Conclusions
This chapter analysed the trends in the supply of agricultural credit
')\- II1stitutional agencies in 14 major states, and across the districts in
(lrissa state. It is found that the growth rate of agricultural credit was
'l:gher during pre-reform period compared to the reform period m
rnost of the states. It is also observed that the growth rate of
clgncultural credit was uneven during the sub-periods as well as
",TOSS the states. There were also substantial inter-state differences in
.\c:ncuitural Credit as a proportion to total net Bank Credit (ACBC)
"!lei Agricultural Credit as a proportion to Bank Deposits (ACBD) at
117
one point in time as well as during different time periods. An analysis
of the ACBD demonstrates that since this has been declining for all
the major states, it shows that there was less mobilisation of deposits
for agricultural lending by scheduled commercial banks. This was
particularly the case in backward states like Bihar, Uttar Pradesh,
and West Bengal.
Most of the findings of the district level analysis were similar to
that of the state level analysis. Important similarities between the
district and state level findings include the declining share of
agricultural credit from net bank credit and uneven credit flow across
the regions. The differences in the availability of credit per hectare of
gross cropped area across the regions is another common finding in
both state and district level analysis. This suggests that the flow of
agricultural credit follows similar patterns at both state and district
levels.
With regard to the analysis of trends in credit flow in Orissa, it
is found that the rate of credit disbursed to agriculture sector by SCBs
was declining during 1990s. The analysis reveals that as far as the
disbursement of agricultural credit is concerned, the SCBs performed
poorly in so far as the achievement of targets is concerned. The views
of bankers at the state and district levels have been examined to
understand the reasons for the failure to achieve the targets on
agricultural sector. The qualitative information (collected from bankers
and NABARD reports) suggests that the demand side problems
including poor recovery, inadequate infrastructure support and low
credit absorption capacity of the agricultural borrowers are the main
,'onstraints In the supply of agricultural crediL This suggests that
bankers, by and large, exclude supply side factors from their purview,
The proportion of irrigated area to gross cropped area, and
density of bank branches per 1,000 farmers have been identified as
the most important explanatory variables for the variations in the
credit flow per hectare of gross cropped area. However, the negative
association between CDR and credit per hectare of gross cropped
indicates that an increase in CDR need not necessarily ensure
improved credit flow to agriculture. This will happen only when the
bankers perceive the agricultural lending to be profitable, The factors
like irrigation facility and CDR also behave in the similar way at the
district levd analysis. T h ~ implication of this finding is that for better
provisioning of agricultural credit the increasing CDR should be
backed by more irrigation facility and banking infrastructure.
The analysis on access to formal credit and the outstanding
loan amount per account reveal that there are inter-state differences
in the access to credit as well as the loan amount obtained by farm
households even for the same size class of landholding, In this
context, the following questions can be raised. First, what makes some
farmers to access formal credit, while others do not? Second, how do
bankers fix up the credit limit for different categories of farmers?
These questions have been taken up in the sixth chapter.
_-----'119
Chapter 5
HOW DID MACRO POLICY OPERATE AT THE MICRO
LEVEL? AN ANALYSIS OF CREDIT FLOW IN THE
SELECTED BANK BRANCHES IN KALAHANDI DISTRICT,
ORISSA
Introduction
The macro-level policies on agricultural finance have been primarily
concerned with the level of credit flow, terms and conditions under
which credit is made available, allocation of loanable funds among
sub-sectors (production and investment) within agriculture to achieve
production efficiency and so on. The macro level policy also concerns
with the provision of adequate and timely availability of credit to the
needy farmers at a cheaper rate. These policies, in general, give
guidelines to the ground level institutional financing agencies having
direct contact with the borrowers. Since institutional credit is a
cheaper source for borrowing, most of the theoretical and empirical
studies have put forward the argument that the demand for credit far
exceeds the supply (Kamajou 1980; Gonzalez-Vega 1984; Adams and
Vogel 1986; Kochar 1997; Swain 2002). Therefore, many countries
and financial institutions followed the supply-led approach to meet
the demand for credit. In the case of India, the supply-led approach in
terms of directed lending was introduced to priority sectors, in
general, and agriculture, in particular.
As a consequence to the supply-led approach at cheaper rate,
formal financial agencies began to face the problem of higher
transaction costs on account of lending small amounts to large
number of borrowers (McKinnon 1973; Shaw 1973; Anderson and
\\hambata 1985; Gadgi11986; Satish and Swaminath 1989; Desai and
Namboodiri 1992). The disbursal of loans without asseSSIng the
feasibility, viability of the project and entrepreneurial capacity of the
borrowers leading to the problem of high overdues and the problem of
recurring losses in the case of most of the rural formal financial
institutions were noticed in India during 1980s (Khusro 1989;
Rajasekhar and Vyasulu 1990; Vyasulu and Rajasekhar 1991; Kahlon
1991). This adversely affected the efficiency and viability of the rural
formal financial agencies.
The Narasimham Committee (1991) recommended for a gradual
reduction in the priority sector target from 40 to 10 per cent of the
total bank credit. However, the RBI did not fully accept this
recommendation on the grounds that the redefined priority sectors
certainly accounted for more than 10 per cent of the total credit.
Further, the recommendation, if accepted, would put a severe squeeze
on the sectors included in the priority lending. At the same time,
various policy measures aimed at diluting the norms of priority sector
lending were taken, so as to ensure its gradual phasing out in the
long-run.
On the whole, the strategy followed by the RBI was to keep the
same target in so far as priority sector was concerned and widened the
,'overage of the priority sector, in general, and agriculture, In
particular
39
. Since the coverage of the priority sector differed In
different years in 1990s, it is necessary to analyse the share of these
sectors in the total advance by public sector banks. Shajahan (1998)
shows that the share of priority sector lending had gone down after
I he banking sector reform. This declining share continued up to 1996-
For more details on the widening coverage of agriculture sector, see 3.4 section of
Chapter Ill. Also see, Shajahan (1998, 1999).
121
97. However, with the widening of the scope of priority sector by the
RBI on a substantial scale during 1997-98 and 1998-99, the share of
priority sector lending went up during these years and after (Shajahan
1999). Thus, it appears that banks allocated more funds to the newly
included sectors and activities in the priority sector to cover the
shortfall in the priority sector lending. Therefore, the extent to which
the banks diverted advances from areas already in the priority sector
to the newly included areas is an important aspect of priority sector
lending, which needs to be analysed. In this chapter, an attempt has
been made to discuss the credit flow to the priority sector, in general,
lI1d agriculture, in particular, with the help of data from four bank
branches in Kalahandi district. The objectives of the chapter are to:
1. find out the share of
agriculture, in particular
selected bank branches;
1-1
pnontj" sectors, in generaL, allY
out of the total advances of the
2. identify the activities (both agriculture and non-agriculture
which are coming under the priority sector) for which bankers
give preference in disbursing advances and the reason for the
same;
3. understand lending policies and practices In distributing the
loans at the branch level; and,
4. examine whether the policy of collateral based lending affects
the flow of credit to agriculture.
atabase
!lis chapter seeks to address these questions by analysing the
:ound level credit flow at bank branch level to different sectors, in
'neral, and to agriculture, in particular. For the purpose of this
:ercise, the data on supply of credit from four different bank
:,lI1ches of Kalahandi district for the period 1996-97 to 2000-01 have
m extracted (two branches from advanced block and the other two
",
from backward block based upon some important parameters). The
discussion relating to the economic background of the
sl'il'Ctcd district and blocks is provided below.
Economic Background o/the Selected District and Blocks
Tl:e Kalahandi district has drawn world-wide attention for poverty,
Illiteracy, underdevelopment, starvation deaths, recurring drought and
so on. The Government of India, Government of Orissa, Reserve Bank
of India and especially National Bank for Agriculture and Rural
De\'c!opment (NABARD) are seriously concerned about the economic
dCl'elopment of the district. Watershed development programme,
promotion of self-help groups and linking them with banks and
?;ot,-iding financial assistance to the state government for the creation
of necessary infrastructure are some of the important steps
, uI:dcrtaken by NABARD for the economic development of Kalahandi
,dIstrict. Since availability of credit is a part of development process,
f\alahandi district has been selected for this purpose.
laj Land Utilisation
Tc,pographically, Kalahandi district is located in the south-western
iJdrt of Orissa. It has two distinct sub-regions: plains and hilly tracts.
The total geographical area of the district spreads over 8,493 sq. kms.,
(r)l1lpnsing of 2 sub-divisions and 13 blocks. The cultivable land in
"Ill' district falls in three distinct categories. They are high, medium
,,':ld low lands consisting of 54 per cent, 28 per cent and 18 per cent
!I'spcctively of total cultivable land of 393,550 hectares (NABARD
1-02). Of the total cultivable area, 46.5 per cent is suitable for
'j"ddv and rest for other crops. The main crops grown in the district
;,Il' paddy, minor cereals, pulses, oilseeds, sugarcane, cotton, fruits
{"j Ol1lon.
1"'
--.,
Ill) Land Distribution
The ownership structure and management of land determine the state
of agricultural prosperity and equity. A majority of the cultivators in
the district are marginal and small farmers. Around 40 per cent of the
holdings in the district have operated area less than one hectare,
\lhile 48 per cent of the total number of holdings derive their
sustenance from holdings ranging between one to four hectares.
Roughly, 30 per cent of the total cultivable area is under the
occupation of small and marginal farmers. Importantly, around 12 per
cent of the total number of holdings are medium to large in size
accounting for as much as 39 per cent of the total operational area
(Table 5.1).
Table 5.1: Size Class-wise Number and Area under Operational
Hid'
s
0 mgs
ize of Landholdings
-
U
2
4
T
o I- 1. 0 (Marginal)
01-2.0 (Small)
.01-4.0 (Semi-medium)
01-10.0 (Medium)
0.01- Above (Large)

Kalahandi 19901991)
No. of Operated
Holdings Area
lin '00) (in '00 hect)
647 (39.9) 369 (11. 7)
425 (26.2) 588 (18.7)
355 (21.9) 951 (30.2)
175 (l0.8) 982 (31.2)
18 (Ll) 260 (8.2)
1,620 (100) 3,150 (100)
Orissa 119901991)
No. of Operated
Holdings Area
lin '00) (in '00 hect)
21,184 (53.7) 10,451 (19.7)
10,353 (26.2) 14,262 (26.9)
5,935 (15.0) 15,608 (29.5)
1,855 (4.7) 10,118 (19.1J_
152 (004) 2,519 (4.8)
39,479 (100) 52,958 (100)
\o[e: FIgures m parentheses refer to percentages from the correspondmg column
",CIrce: Government of Orissa (1996).
The inequality in the distribution of land ownership in the
dIstrict can be traced to nineteenth century land revenue settlements,
IlhICh led to the emergence of bipolar class structure. At the top of the
Iural hierarchy were the gountias, village headmen who distributed
:dlld for cultivation and collected land revenue for the State during the
period, and whose power was consolidated and
by the colonial government. Although the gountia system
'.is abolished in 1936, they were allowed to retain large extent of land
124
for personal cultivation (Das 1962). Over time, gountias came to wield
cIa-larger influence over their villages, where they occupied the best
land and gained control over wasteland resources. The ceiling
legislation was largely unsuccessful in the redistribution of land from
the nch gountias to the poor landless farmers. As a result, a small
section, which is very well endowed with wealth, exercises control
through credit, land and labour markets over a large mass of
peasantry, which has been unfavourably placed in so far as the access
to productive assets like land. The gountia system coupled with
dependence of a large proportion of population on agriculture resulted
in extreme inequality of land ownership.
Ie) Agrarian Conditions
The normal rainfall in Kalahandi district is around 1,378 mms.
However, it varied from as low as 1,045.5 mms to as high as 2,011.2
mms between 1994 and 1999. The erratic behaviour of the monsoon,
uneven distribution of rainfall during the crop season coupled with
occasional long dry spells at post-sowing or pre-productive stage
caused extensive damage to the standing crops almost in every
eli tcrnative year.
In the context of uncertain rainfall, the artificial irrigation
:xIiity assumed an important role. However, on an average, the
percentage of irrigated area to gross cropped area continued to be
oi!ound 18.5 per cent between 1995-96 and 1999-2000 (Table 5.2),
"h'Ch is less than the state average. It shows that around 81.5 per
":n of the gross cropped area entirely depended upon rainfall in the
Jii"rict. Out of the total irrigated area, around 55 per cent of the area
L "Tigated by major and minor projects, and the rest by lift irrigation
125
and ram dependent sources like tanks and dugwelJ. The irrigated
areas under major and minor projects are confined to Jaipatna,
Dharmagarh, Kalampur, Junagarh, and Bhawanipatna blocks. It is
also mentioned that the existing water harvesting structure in the
district is found inadequate for irrigating the land coming under the
ayacut area because of low storage of water in the reservoir and
absence of land levelling in the adjoining areas (State Bank of India
2000-01). Consequently, the area under HYV as well as fertiliser
consumption per hectare of gross cropped area in the district is lower
than that of the state average (Table 5.2).
Table 5.2: Area under Irrigation, Fertiliser Consumption and
Area under HYV in Kalahandi District and Orissa
Year
Katahandi Orissa
Area Under Area Under Fertlliaer Area Under Area Under Fertiliser
Irription HYVto Con.ump Irrigation to HYVto Consump-
to Oro .. Oro .. -tlon Gro .. Oross tion
Cropped Cropped
Ka/laec Cropped Area Cropped K&/laec
i
Area-(%\
Area(%J
-,%)
Area-'%)
1995-96 15.4 20.6 11 27.2 32.5 25
1996-97 16.2 25.5 15 27.5 38.3 31
1997-98 19.7 27.1 18 26.8 38.4 35
: 1998-99 19.7 27.2 22 28.0 38.8 36
1999-2000 22.0 29.3 28 29.5 42.0 42
: Entire 18.5 25.7
-
27.8 37.8
-
I Period
Source: Government ofOnssa (1995-96,1996-97, 1997-98, 1998-99, and 1999-
2000).
(d\ Characteristics of Population and Their Occupational Distribution
The total population in Kalahandi (excluding Nuapada sub-division of
old Kalahandi) district was 11.31 lakhs in 1991 with the sex ratio of
999 females per thousand males. Around 93 per cent of the total
population stay in the rural areas. The percentages of male and female
literates in the district were 31 and 8, while those for the state were
47 and 21 respectively. The proportion of scheduled castes and
scheduled tribes population was high, around 46 per cent. According
to the 1991 census, the total number of workers in Kalahandi district
was 4.26 lakh, constituting 37.67 per cent of the total population in
the district. Of the 4.26 lakhs workers, a bulk of them were engaged in
agriculture. The proportion of cultivators in the total workforce was 43
per cent, whereas it was 41.1 per cent in the case of agricultural
labourers (Table 5.3). Since 84 per cent of the total workforce
depended on agriculture, agriculture could be stated as the mainstay
of the district's economy.
Table 5 3: Socio-Economic Features of Kalahandi District .
i Geographical area 8439 Sq.Kms. -
I No. of blocks 13 -
I No. of villages 2,185
-
I Total population 11.31 lakhs As per 1991 census
. Total number of workers 4.261akhs - do -
f Number of cultivators 1.821akhs - do-
1 '7C:: 1_1.1-_
- uu- I of agricultural !.abourers ..... 'oJ
i Length of surface roads 12524 Kms By 1996-97
,_ No. of registered motor vehicles 17,429 Bv 1997-98
i No. of allopathic medical institutions 61 Bv 1996-97
I No.
livestock aid centers 129 - do -
No. of livestock inspectors 121 - do -
I No. of financial institutions' 92 Bv 2000-01
Note: * Consists of co-operatIves, commercial and regIOnal TUral banks.
Source: Government of Orissa (1996, 1997).
re) Infrastructure Facilities
Kalahandi is the most backward district as far as the availability of
infrastructure facility is concerned. The surfaced roads
4o
per 1,000
square kms of geographical area was 1,475 kms in 1997-98. The
condition of the existing roads is bad and poorly maintained. The
number of registered motor vehicles per lakh of population (as per the
1991 census) in the district was 1,541 in 1997-98. The number of
buses running between district and block headquarters were hardly
41l It includes national highways, State highways, Major district roads, other district
roads, forest roads, grama panchayat roads, classified village roads, P.S. Roads,
village roads and urban roads.
sufficient. In the case of some interior villages, roads were not
connected, and public transport system did not exist.
The medical facilities are virtually absent in the villages, and
['I'en if they exist, the medical staff hardly stay in the villages. For any
kmd of major or minor health problem one has to consult either the
sub-divisional or district headquarter hospitals or private doctors. For
instance, the number of allopathic medical institutions
41
per lakh of
pupulation was 5.4 during 1996-97 in the district. The veterinary
facilities are almost absent in the district. Out of 2,185 villages, only
129 livestock aid centers with 121 livestock inspectors were there in
the district by 1995-96. It shows that, on an average, a single
livestock inspector had to cover 18 vilIag('s. Importantly, cut
total number of villages only 52.7 percentage of villages was electrified
in the district by March 1997.
There were 92 branches of various banks! financial institutions
operating in the district including 43 branches of Kalahandi Anchalika
Gramya Bank, 35 commercial banks, 11 branches of central co-
operative bank, 2 Agriculture and Rural Development bank, and one
immch of Orissa State Financial Corporation. It is noticed that the
number of people depending upon per branch was 13,600, whereas
the average number of villages covered per bank branch was around
24 111 the district by the year 2001-02 (NABARD 2001-02). This shows
that the average population per bank branch in the district was less
thelD that of State average, which was 16,300 during 1999-2000.
I, ",,'Iudes district headquarters hospital, sub-divisional hospital, other hospitals,
(ulilmunity health centres, Primary health centres, and Mobile health centres.
The existing marketing facilities are insufficient in the district.
Because of poor transport and communication facilities, a great
majority of the marginal and small farmers find it inconvenient and
unprofitable to sell their agricultural produce in the regulated markets
located in the nearby urban or semi-urban localities. As a result, most
of them sell their produce to the middleman or grain trader in their
villages at much lower rates than those prevailing in the regulated
markets
42

It must be emphasised, however, that the economic
backwardness is not uniform in all parts of the district. When
compared with the hilly tracts pursuing rain-fed agriculture, the areas
possessing flat and fertile paddy -fields receiving water from Indravati
river (presently through a dam constructed across the river) permit a
more advanced form of agriculture. Of the 13 blocks in the district, six
fall in the advanced region, while the rest in the backward region. The
blocks of Jaipatna and Narla, on which the study is based, represent
the advanced and backward regions, respectively, of Kalahandi
district. The differences between the two blocks are evident from the
following details. While the proportion of irrigated are to gross cropped
area in Jaipatna varied between 50 to 58 per cent across the seasons,
the corresponding figures for Narla block were 13 to 31 per cent.
Similarly, the fertiliser consumption was only 6 kgs per hectare in
Narla block as compared to the figure of 15 kgs for Jaipatna block.
" For example, in most of the villages from where primary data were collected, the
farmers sold their paddy within the range of Rs. 265 to Rs. 325 per bag (75 kg) in
the village when the prevailing rate at that time was Rs. 380 in the year 200 \-02.
The study made by Rajasekbar and Vyasulu (\993) in the Kalahandi district also
mentions a similar kind of rmding. They notice that in one of the sample villages
located just 34 kms away from the district headquarter of Bhawanipatna, the
farmer sold paddy for Rs. 125 per quintal in the village, while the prevailing rate
at that time was Rs. 160.
These differences in agricultural inputs lead to variations In the
density of paddy cultivation and paddy yield rates. In Jaipatna block,
paddy cultivation was widespread (with almost 100 per cent in Kharif
season), whereas in Narla it accounted for only about 61 per cent of
the area during the same season
43
. The per hectare yield rates for
paddy were 6 to 14 quintals in the backward and advanced blocks,
respectively. There were also infrastructural differences - particularly
with regard to communications (measured in terms of the length of
durable roads, and population covered by post office and banks), - as
well as the number of industries and development of off-farm
economic opportunities (Table 5.4).
Although the two blocks in Kala..h.andi district differed in terms
of resources, agricultural technology, cropping pattern, yield rates,
communications and industrial development, the agrarian structure
was largely the same. In each block, therefore, it was the gountias and
large landholders who owned considerable tracts of agricultural land,
having as a result control over credit and labour markets as well as
local institutions. At the other end of the agrarian class structure, a
large number of small and marginal farmers belonging mostly to the
depressed castes depended on tiny landholdings without much access
to institutional credit. It was these smallholders who depended on the
informal credit market located at both village and town levels for their
production and consumption needs to a greater extent.
" The other important crops in Narla block are cotton and banana.
I
,
\
I
Priority Sector Lending
The proportion of credit to priority sector has expanded significantly
over time in the selected bank branches in both the blocks. In
Jaipatna block, although the share of priority sector had marginally
declined from 89.47 to 87.72 per cent from total advance for the years
1997 -98 to 1998-99, there was an overall increase in the share of
priority sector in the total advance by 2000-01 (Table 5.5). For Narla
block, it can be observed that there was a continuous increase in the
share of outstanding credit to the priority sector and it went up from
76.86 per cent in 1996-97 to 86.04 per cent in 2000-01.
Table 5.5: Sector-wise Flow of Bank Credit
(in per cent)
1996- 1997- 1998- 1999- 2000- Annual
Sectors 97 98 99 2000 01
Average
Sample BaDb of Jaipatna Block
Growth Rates
Agriculture 42.83 37.87 27.83 21.69 18.82 09.69
Other Priority Sector 46.65 50.37 60.00 72.78 76.07 56.40
Total Priority Sector 89.47 88.24 87.72 94.47 94.89 39.28
NonPriority Sector 10.53 11.76 12.28 5.53 5.11 09.71
I Total Advance
; IRs. in Lakhs} 229.13 262.49 399.27 572.45 740.75 36.70
Sample BaDb of Narla Block
, Agriculture
48.91 41.78 37.53 40.07 46.83 11.69
Other Priority Sector 27.94 36.22 40.76 41.73 39.20 22.79
Total Priority Sector 76.86 78.00 78.29 81.79 86.04 16.27
NonPriority Sector 23.14 22.00 21.71 18.21 13.96 00.34
I Total Advance
, (Rs, in Lakhs)
87.93 108.94 126.42 134.63 146.59 13.13
Source: Sample Bank Branches of Jlllpatna and Narla Block
The total volume of outstanding credit disbursed by the
selected bank branches in Jaipatna block was Rs. 229.13 lakhs in
1996-97 and it increased to Rs. 740.75 lakhs by 2000-01 at an
annual growth (in nominal term) rate of 36.70 per cent. In the case of
Narla block, this growth rate was only 16.27 per cent. In both the
blocks, a higher growth rate was not uniform across different sectors
(Table 5.5). Importantly, the higher growth rate of outstanding credit
to the total priority sector (TPS) in both the blocks could be attributed
to the higher growth rate of credit in the other priority sector (OPS)
consisting of small business firms (SBF), small-scale industries (SSI)
and so on. The growth rate of credit flow to agriculture was far below
than the growth rate of credit to the OPS, indicating that the bankers
were giving preference to non-agricultural activities. This situation
was more prominent in the advanced block like Jaipatna. It implies
that where there is scope to invest in non-agriculture sector, banks
are reluctant to extend their credit facility to the agricultural sector. If
we compare the growth rate of credit to agriculture and OPS with the
growth rate of credit to TPS in the selected bank branches of both the
blocks, Narla block has relatively less scope for the banks to deploy
their funds_ in the OPS or non-farm activity. This is due to lower
absorption capacity by the non-farm activity holder or less scope for
investment in SBF and or SSI. In this context, the bankers are of the
view that where there is scope, non-agricultural loans not only have
better recoveries but also fetch relatively higher interest rates'".
From the foregoing discussion one can conclude that the higher
growth rate of credit flow to the OPS has been taking place at the cost
of credit going to agriculture. This preference of lending for non-
agricultural activities can also be corroborated by data on the share of
agricultural credit from total advances. Table 5.5 shows that the per
cent share of advance to agriculture from net bank credit has declined
from 42.83 per cent in 1996-97 to 18.82 per cent in 2000-01 in the
case of Jaipatna block. Besides, there exists a higher gap between the
growth rate of credit to agriculture and total advance (Le. 27.01 per
cent) for the same reference period. The same appears to be true in
"' It has been argued that the increasing operating costs and higher loan
delinquency rates compel the banks to adopt such a strategy (Shylendra \996)_
the case of credit going to agriculture in Narla block except for the
years 1999-00 and 2000-01. Interestingly, although the volume is
less, the proportion of credit going to agriculture from total advance in
the backward block (Narla) is more than that of advanced block
(Jaipatna) in the last five years. This can be attributed mainly to the
non-availability of scope for deployment of funds in the OPS. Hence,
the higher growth rate of credit flow to the TPS and or total advance
cannot be attributed to an increase in agricultural finance.
Notwithstanding this, within the priority sector, there is a preference
for non-agriculture sector by the bankers.
Why is it that the growth rate of credit to agriculture is much
below the growth rate of credit to the TPS? To address this question, it
is essential to review the policies relating to the norms of priority
sector lending. Firat, direct and indirect categories of agricultural
advances within the sub-target of 18 per cent of agricultural lending
were clubbed in 1993-4 subject to the stipulation that 'indirect'
lending to agriculture must not exceed one-fourth of the agricultural
target or 4.5 per cent of net bank credit. It had also been decided to
include any indirect advance exceeding 4.5 per cent into the overall
target of priority sector. Second, increasing the ceiling of credit limit
to Rs. 5 lakhs for agricultural inputs supply and allied to agriculture
(like dairy, poultry, piggery and so on) and bringing them under the
purview of indirect agricultural advances. This widened the scope of
the priority sector, in general, and agriculture, in particular. Third,
the extension of financial facility to even dealers in drip irrigation,
sprinkler, and agricultural machinery was classified as 'indirect
finance to agriculture'. Nevertheless, short-term advances to
plantations of tea, coffee, rubber and spices, irrespective of the size of
the holdings were brought under the scope of direct agricultural
advances.
As a result of policies relating to addition of sub-sectors into the
priority sector lending, the scope of agriculture sector is much larger
at present than what it was in the early 1990s. Consequently, the
credit flow has to be channelised to new sub-sectors keeping the limit
at 18 per cent. This has led to shrinkage of loans to the initial sub-
sectors such as agricultural term loan and more specifically allied
agricultural activities (Table 5.6).
The broadening of priority sector besides agriculture with higher
credit limit is other important factor for a high growt..1-J. rate of credit
flow to OPS and, in tum, to TPS. For instance, initially, SSI were those
industries whose investment on plant and machinery did not exceed
Rs.35 lakhs. In the case of ancillary units, the investment limit was
Rs.45 lakhs. In May 1994, these limits were raised to Rs.60 lakhs and
Rs.75 lakhs respectively. According to the revised definition, all
advances to SSls were to be treated as priority sector lending. Besides,
for small units
45
a target of 40 per cent of total credit to SSI was fixed.
Furthermore, the existing ceiling limits on advances to OPS such as
retail trade, small business enterprises, housing, professional and
self-employed persons under the purview of priority sector advance
were also elevated. Thus, when the new areas were entered under the
umbrella of priority sector lending, the scope of deployment of funds
by bankers increased to a greater extent resulting in greater flow of
credit to OPS and also TPS. This proposition can be substantiated
4S Small units consist of cottage industries, khadi and village industries, artisans
and tiny industries with investment in plant and machinery up to Rs.5 lakhs. For
more detailed information, see, RBI (1992-93, 1994).
from the sector-wise flow of outstanding credit (Table 5.5) where non-
agricultural purposes accounted for the highest share of total
advances. In this context, it is necessary to examine the pattern of
agricultural lending.
Agricultural Lending
Agricultural loans are advanced for three broad purposes, VIZ., crop
production (consisting of agricultural cash credit and agricultural gold
loan), agricultural investment, and allied activities. The production
credit is generally in the form of short-term loans given for various
seasonal agricultural operations. The agricultural investment and
allied activities come under term loans given for the purposes like land
development, minor irrigation, farm mechanisation, for the purchase
of draft animals, and for pursuing activities like dairying, sheep and
goat rearing, etc.
Coming to the flow of agricultural credit, Table 5.6 gives the
purpose-wise share of credit going to agriculture. In the case of the
selected bank branches in Jaipatna block, credit extended to
agriculture was Rs. 98.13 lakhs in 1996-97 consisting of 42.83 per
cent of total advances. By 2000-01, although the volume of bank
credit to this sector had gone up to Rs.139.44 lakhs, its share from
total advances had come down to 18.82 per cent. Interestingly, on an
average, the volume of credit going to agriculture through the selected
banks in Narla block was less than 50 per cent of the volume of credit
going to this sector in Jaipatna block for the last five years. However,
one can observe a relatively greater share of credit from total advances
going to agriculture in the backward block (Narla) compared to
advanced block (Jaipatna). This phenomenon is being noticed since
1996-97 (Table 5.6). Notwithstanding this, crop production and
agricultural term loans (ATL) accounted for the highest share of loans
outstanding in agriculture in both the areas.
Table 5.6: Purpose-wise Distribution of Agricultural Credit
from 1996-97 to 2000-01
(in per cent)
I
Purpose 1996- 1997- 1998- 1999- 2000-
I
97 98 99 2000 01
I
Sample Banks of Jaipatna Block
I Agricultural Cash Credit-tAl
33.29 22.79 21.25 22.67 24.60
~ Agricultural Gold loan-fBI
11.16 12.21 18.97 17.49 15.12
Total Croo Loan fA+ BI
44.45 35.00 40.22 40.17 39.72
Agricultural Term Loan
49.13 55.04 55.80 54.88 50.36
Allied to Agriculture
06.42 09.96 03.98 04.95 09.92
Total Advance (Rs. in Lakhsl
98.13 99.40 110.71 124.15 139.44
I
Sample Banks of Narla Block
, Agricultural Cash Credit-fA) 20.41 19.67 18.12 21.06 29.18
I Agricultural Gold Loan- (B)
14.65 16.39 13.57 19.50 17.31
l10tal Crop Loan (A+BI
35.06 36.06 31.70 40.56 46.48
I Agricultural Term Loan 36.39 38.15 38.57 31.02 31.87
I Allied to Agriculture 28.55 25.80 29.74 28.42 21.65
, Total Advance (Rs. in Lakhs) 43.01 45.51 45.51 47.45 53.94
Source: Sample Bank Branches of JB1patna and Narla Block
Based on the above discussion on the credit going to agriculture
by banks in both the areas, one can raise an important question. Why
is that the proportion of credit flow to agriculture from net bank credit
is relatively more in backward area as compared to the advanced one?
What accounts for a greater share of credit to crop production and
investment loans? From the lenders' perspective, one can argue that
since better alternative opportunities are available in non-agricultural
activities in advance areas, the deployment of funds to agriculture
may not be attractive. Alternatively, less opportunity for investment in
non-agriculture sector may force the bankers to tum to agriculture for
investment. This argument seems to be true from a substantial low
growth rate of credit flow to OPS in the backward region compared to
the advanced one rrable 5.5). Nevertheless, low (greater) share of
credit going to agriculture from net bank credit may be due to low
[high) repayment rate of loan by the borrower in the concerned sector.
Hence, in order to explore more about the issues, it is essential to
examine sector-wise recovery to demand.
Table 5.7: Sector-wise Recovery of Loan Amount to Demand
(in per cent)
1996-97 1997-98 1998-99 1999-2000 2000-01
Sectors Jaipatna Block
: Agrl. Loan 62.92 63.20 66.00 65.17 59.87
for Agrl 23.88 33.50 32.12 20.86 23.17
-!ric_ulture 40.99 45.38 45.17 43.54 40.17
)n-agriculture 69.79 78.51 80.73 69.93 71.62
:al
57.89 63.38 63.43 59.49 60.22
Narla Block
:-::1
Agrl. Loan 58.21 60.38 59.02 70.70 60.85
for Agrl 26.28 34.54 37.48 41.05 22.40
lJ'iculture 40.75 51.44 52.85 62.28 49.40
',:1-agriculture 67.37 54.71 50.00 63.79 70.18
_'tal 49.65 52.51 41.01 6271 SS 7?
, ,
,.
S,)urce: Sample Bank Branches of Jalpatna and Narla Block
Table 5.7 shows that there is an improvement 111 terms of
rel'overy to demand, especially in non-agriculture sector in both the
regions. The percentage of total recovery increased from 57.89 per
cellt in 1996-97 to 60.22 per cent in 2000-01 in the selected bank
brclnches of Jaipatna block. This can be attributed to higher recovery
rale 111 the non-agricultural sector compared to agriculture.
Fl,nhermore, the percentage of recovery for the entire sector (both
"gr:C'ulture and non-agriculture) to total demand shows that the rate
of repayment was more in the selected bank branches of Jaipatna
b;,. k as compared to Narla Block except for the year 1999-00.
HIJ\\TVer, in the latter block, the rate of recovery of amount lent to
agrll'ulture as a proportion to demand was always higher than that in
thl' former block for the entire period. This high recovery rate rna\' be
,J:;,' of the important contributing factors for a greater share of credit
gOlt:" to agriculture from total net bank credit in the case of Narla
block. Based on the above discussion, one can raise the following
questions. What contributes to low recovery rate in the agricultural
sector in the advanced Jaipatna block? Is the low repayment more
prevalent in short-term or term loans to agriculture?
To understand the questions addressed here, it is essential to
examine the repayment rate of both short-term and term loans for
agriculture. Table 5.7 shows that there was high repayment rate In
short-term (ST) loans compared to long-term agricultural loans In
both the blocks. However, there was a growing tendency in the
repayment rate in short-term agricultural loans for the period 1996-97
to 1998-99 in the selected bank branches of both the regions. A
ckclining tendency of repayment in ST !oa.'1S could be obser .... ed from
the year 1999-00 onwards in both the areas. Interestingly, the low
repayment rate of long-term (LT) loans is a common phenomenon in
both the blocks. This indicates that the low repayment rate of
agricultural loans as a whole in Jaipatna block could be attributed to
relatively poor repayment of term loans.
Why is the repayment rate of long-term agricultural loan so
poor? What may be the contributing factors for declining tendency of
repayment for crop loans (ST loan)? To understand the questions
addressed here, one need to undertake a micro-level (households)
analysis. In the context of the above issues, the farmers are of the
opinion that under finance, diversion, and misutilisation of funds are
the major factors which adversely affect the repayment rate.
Taking into account the repayment rate of both agriculture and
non-agriculture sectors, one can ask, whether the share of credit to
139
agriculture form total advance by banks is positively related with the
interest income
46
on advance? Table 5.8 reveals that the interest
income on advance is more in the selected bank branches of the
Jaipatna block for the last four years (1997-98 to 2000-011, where the
share of credit going to agriculture has been decreasing continuously,
and this share is less as compared to the counterpart region. In this
context, one can arrive at a conclusion that given the repayment rate,
interest income on advance is inversely related with the flow of credit
going to agriculture sector.
Table 5.8: Interest Income on Advance
(in per cent)
ParticuJara Years
1996-97 1997-98 1998-99 1999-2000 200001
Selected Ballk Branchcs
--- -
: of Jaipatna Block
6.16 9.65 8.89 9.23 10.94
I Selected Bank Branches
7.37 7.74 8.84 8.91 8.38

Source: Sample Bank Branches of Jatpatna and Narla Block
It can be inferred that the credit norms relating to profitability is
one of the important contributing factors for lower credit flow to
agriculture. Thus, if the interest income is greater than or at least
equal to the cost of lending (what is termed as break-even level of
interest4
7
), this may positively influence the financial institutions to
increase the supply of credit. However, default rate plays an important
role in determining the break-even level of interest rate for a financial
institution. Higher the default rate, higher is the break-even interest
rate that has to be charged. But, given the administrative interest
"Income = (Interest Received on Advance I Total Advance by Bank) x 100.
,; It implies the minimum needed lending rate for the bank (i.e. Income from loan
portfolio Cost of lending). For more details, see, Anderson and Khambata
(1985), Hulme and Morsely (1996) and Satish and Gopalkrishna (1997).
140
rate, banks cannot increase their lending rates. Therefore, banks will
try to bring down the break-even condition of lending rate to
reasonable levels by reducing default rate and the cost of lending. In
our case, since the separate figure of interest income on only
agricultural lending is not available at the branch level, we have taken
interest income on total advance as the proxy variable to explain its
relation with the flow of credit to agriculture. It is observed that for the
bank branches where the interest income on advance is more (for
instance, the selected bank branches of Jaipatna Block), the share of
credit going to agriculture out of total deployment is less and vice-
versa.
Security Based Lending
In general, bank loans are sanctioned on the basis of four different
types of security. They are: land, gold, deposits, and hypothecation of
loan assets and personal security. The loan sanctioned with land as
security can be further sub-divided into (i) sanction of loan by
mortgage, and (iiI loans by a charge over land. Basically, the former
type of security is needed for agricultural term loan like improvement
of land, purchase of agricultural implements, development of minor
irrigation and so on. In the case of default of loans, which are
sanctioned on the basis of mortgage type of security, the land may be
sold for recovering the dues. To meet the requirement of working
capital in crop production the latter type of security is needed and the
dues have to be recovered only out of the proceeds of the land.
However, the borrower has to produce the record of ownership right
over the said land to obtain loan against both categories of land
security. In the case of loans with land as security, the amount of
loan obtained depends on the cost of the said project, value of land if
It is an investment loan, and the scale of finance if it is crop loan.
141
Loan sanctioned on the basis of either gold or deposits as security can
be treated as demand loan. The amount of loan sanctioned depends
upon the market value of the gold ornaments subject to the maximum
amount fixed per unit. In the case of default, the bank can auction the
particular ornaments to recover the dues with advance notice to the
borrower. On the other hand, any default of loan against deposit will
be adjusted at the time of repayment. In the context of loan
sanctioned against hypothecation of loan assets and personal
security, no primary collateral is needed as explained above. This type
of loan sanction helps the borrower who cannot afford any primary
collateral to access institutional credit and if the loan is sanctioned for
asset creation, the same works as collateral for the bank till the loan
is repaid. In the case of default, the hypothecated asset may be sold
for repayment. In the case of loan against personal security, the
borrower is expected to repay the loan on demand by the banks. This
type of loan basically plays a major role at the time of sanction in non-
farm activity.
With this background, it may be necessary to know the
distribution of agricultural credit by type of security. Since the return
on agricultural investment is highly uncertain and affected by several
factors, and the probability of default rate is high, the financial
institutions may be reluctant to extend credit facility to the
agricultural sector. Thus, to minimise the lending risk, the bankers
may prefer collateral based lending.
142
Table 5.9: Distribution of Agricultural Credit by Type of
Security
(in per cent)
Type of Security 199697 199798
199899 1999
20000il
2000
Sample Banks of Jaipatna Block
Land
47.64
31.40 44.10 32.38 41.07
Gold
11.16 12.21
18.97 17.49 15.12
Hypothecation of Loan Assets 41.20 56.39 36.93 50.12 43.80
and Personal Security
~ o t a l Advance IRs. in Lakhs) 98.13 99.40 110.71 124.15 139.44
,
Sample Banks of Narla Block
, Land
55.29 51.00 58.48 56.47 65.39
, Gold
14.69
16.39 13.57 19.50 17.31
Hypothecation of Loan Assets 30.02 22.61 27.95 24.03 17.31
and Personal Security
Total Advance (Rs. in Lakhs) 43.01 45.51 47.45 53.94 68.65
Source: Sample Bank Branches of Jrupatna and Narla Block
It may be noted from Table 5.9 that a major percentage of total
amounts of loan is given on the basis of security of land and gold (in
both the regions). However, the proportion of loans given on the basis
of hypothecation of loan assets and personal security have also
accounted for a significant share in the total outstanding advance to
agriculture in Jaipatna block. In the last five years (Le., from 1996-97
to 2000-01), loan disbursed against hypothecated loan assets and
personal security ranged from 37 to 56 per cent of total loan
outstanding in the sample bank branches of Jaipatna block. Given the
backwardness, the rate of disbursement of loan against hypothecated
loan assets and personal security is supposed to be more in the
backward region like Narla. Surprisingly, the rate of disbursement of
loan against these securities vary from the range of as low as 17 per
cent to as high as 30 per cent of total credit given for agricultural
purposes in the sample bank branches of Narla block. It must be
emphasised, however, that the proportion of total agricultural credit
disbursed against hypothecation of loan assets and personal security
is less in the selected bank branches of backward block compared to
the advanced block (Table 5.9). Thus, is it true that the bankers are
really extending their credit facility at a significant rate on the basis of
hypothecation of loan assets and personal security in the service area
of selected bank branches of advanced block?
It is known that a loan against hypothecation of assets and
personal security is supposed to be sanctioned without insisting on
any collateral under various credit-based poverty alleviation schemes.
Earlier, it has been argued that the bank followed both direct and
indirect collateral based lending except in the case of poverty
alleviation schemes under which it was compelled to lend without
insisting on any collateral (Shylendra 1996). For instance, in many
cases it is found that, although, no collateral security except
hypothecation of loan assets and personal security is to be considered
on paper, for loans other than schematic fmance, in practice the
banks are sanctioning such loans only to those borrowers who
borrowed some secured loans by pledging land or gold or maintaining
some term deposit in the bank. Hence, borrowers' opinion may look
more logical to understand this problem.
Following are some of the important problems of the borrowers
that have been observed in the field area of Kalahandi district who
had taken loans under allied agricultural activities. First, under this
category of loans, the bankers kept at least some amount out of
subsidy given (in some cases the whole subsidy amount) in the form of
deposit in the bank against the concerned borrower. It implies that, in
the case of default, the deposit amount is adjusted at the time of
repayment. Alternatively, in the worst situation, the bank can recover
at least part of total dues from the deposit amount of the concerned
borrower. Thus, one can deduce that under finance is a common
feature in the case of loans that fall under the category of allied
agricultural activities. Second, banks in general release the loan in
more than one instalment either in the form of cash or kind or both.
Under this circumstance, if there is no substantial utilisation of
released amount, the bankers stop releasing the balance amount of
loan for the concerned project. On the other hand, this unutilised
amount may be deposited in the bank by opening a new account
against the concerned borrower. Hence, in the light of the above-
mentioned point, one can assume that bankers always take up both
direct and or indirect form of collateral from the borrowers who fall
under the category of other than schematic finance.
Coming back to the selected bank branches of Narla block,
Table 5.9 shows the loan given on the basis of land and gold security
accounted for about at least 70 per cent of the total outstanding loans
during the period 1996-97 to 2000-01. This indicates that the flow of
agricultural credit is largely governed by the security that the stock of
capital (Le., assets like land or gold) offers to the financial institutions
advancing credit rather than by the factors influencing the demand for
credit. One can conclude, therefore, that to some extent collateral
based lending affects the flow of institutional credit to the agriculture
sector.
Inter-Regional Differences in Priority Sector Lending and
Deposits
Table 5.10 shows that during the period 1996 to 2001, priority sector
credit as a proportion of total bank deposits (PSCBD) had gone up
from 64.64 per cent in 1996 to 116.7 per cent in 2001 in Jaipatna
, ..
block. Surprisingly, PSCBD has always been less than that of 'priority
sector credit as a proportion of bank credit (PSCBC) except for the
year 2000-01. However, the reverse is true in the case of Narla block
except for the years 1999-2000 and 2000-01. This indicates that the
performance of selected bank branches of Narla block in terms of
mobilisation of deposits on priority sector lending was much better
during the period 1996-97 to 1998-99 rather than the period 1999-
2000 to 2000-01. Thus, as long as PSCBD is greater than PSCBC, one
can suggest that there is better performance by banks so far as the
mobilisation of deposits under the service area of the bank is
concerned. On the contrary, if PSCBD is less than PSCBC, it implies
that there is less credit flow to priority sector out of loanable funds
for institutional credit, if PSCBD is less than PSCBC of a particular
bank branch, then any initiative measures by the bankers for a
positive movement of the first ratio can facilitate more supply of credit
in the service areas.
Table 5.10: Proportion of Priority Sector Lending to Total
.
Credit and Total De )Oalts
, Sample
1996-97 1997-98 1998-99 1999-2000
I Banks
psc PSC PSC
psc.
psc PSC PSC
BC BD BC BD BC BD BC
I Jaipatna
89.47 64.64 88.24 66.17 87.72 85.55 94.47
Block
[Narla
Block
76.86 105.30 78.01 92.49 78.29 81.56 81.79
Note. PSCBC - Pnonty Sector Credlt as Proportlon of Bank Credlt
PSCBD Priority Sector Credit as Proportion of Bank Deposits
Source: Sample Bank Branches of Jaipatna and Narla Block
PSC
SO
86.85
76.27
2000-01
PSC PSC
BC so
94.89 116.70
86.04 72.35
During the period 1996-97 to 2000-01, loans to agriculture
sector by banks as a proportion of total bank deposits have been
invariably declined in both the regions except a marginal improvement
in the last year, i.e., 2000-01. In the case of Jaipatna block,
agricultural credit as proportion of bank deposits (AGLCBD) was lower
'4"
to AGLCBC (Agricultural Credit as Proportion of Bank Credit) for
almost all the referred years except for the year 2000-01 .
Furthermore, the ratio AGLCBD perpetually declined from 30.94 per
cent in 1996-97 to 23.15 per cent in 2000-01. It is, therefore, one can
say that the amount of credit going to agriculture sector out of total
deposits from the sample bank branches of Jaipatna block was not
encouraging. It seems banks were diverting their loanable funds to
non-agricultural sector. This finding can be corroborated with a
substantial high growth rate of credit flow to other priority sectors
(excluding agriculture) in Jaipatna block (Table 5.5).
In so far as Narla block is concerned, the ratio AGLCBD was
more than AGLCBC for the years 1996-97 to 1998 99 (Table 5.11).
However, the reverse trend could be observed for the remaining years
starting from 1999-2000 to 2000-01. The ratio AGLCBD declined from
67.01 per cent for the year 1996-97 to 39.38 per cent by 2000-0l.
This indicates that although the volume of deposits in the bank was
increasing, this had not induced any stimulation in the flow of
agricultural credit.
Table 5.11: Proportion of Lending to Agriculture from Total
C d d TID t re It an ota epOSl S
isamPle
1996-97 1997-98 1998-99 1999-00 2000-01
Banks
AGLC AGLC AGLC AGLC AGLC AGLC AGLC AGLC AGLC AGLC
BC BD BC BD BC BD BC BD BC BD
': .Jalpatna
.
Biock 42.83 30.94 37.87 28.40 27.83 27.04 21.69
\arla Block
48.91 67.01 41.78 49.53 37.53 39.10 40.07
\ote. AGLCBC- Agncultural Credlt as Proporhon of Bank Credlt
AGLCBD Agricultural Credit as Proportion of Bank Deposits
Source: Sample Bank Branches of Jaipatna and Narla Block.
19.94 18.82 23.15
37.36 46.83 39.38
The key findings emerging from an analysis of priority sector
lending and deposits (see Graph 1) at the bank branch levels of both
lhe blocks are the following:
The credit given to priority sector as a proportion of total bank
credit and deposits have been increasing in the advanced block.
The PSCBD ratio in the service area of sample bank branches in
the backward Narla block have been declining for the last five
years, and it is less than the PSCBD ratio of Jaipatna block since
1998-99. This indicates that bankers are having better opportunity
to mobilise their funds raised by deposits on priority sector in the
advanced region. The poor infrastructure and lack of scope for non-
farm activities can be attributed to the poor absorption capacity of
credit by the potential borrowers in the case of backward regions
like Narla.
A comparison of PSCBC ratio between the bank branches of both
the regions further confirm the argument that bankers are having
more scope for priority sector lending in the advanced region.
Graph 1: PSCBD and PSCBC in the Sample Bank Branches of
Jaipatna and Narla Block
PSCBO and PSCBC in the sample banI< branches of Jaipatna and Narla
.-_________ BIOCk
[ ._u _______ _
. .? .. '--,


-

*
70 .. .:::;:,.;;;;;;;;;;;;;;;;;;; ....... ..-::--------------
u 60 --,- ---- - ________ , -\
Ii;
Cl. 50 1----------------------------
40 \----------.---------------.
30 I
,
201---------------------------,
10 1
o L-______________________________________ _
1997 1998 1999
Year
2000 2001
__ PSCBDJ ..... -- PSCBCJ -..-PSCBDN PSCBCN
Note: PSCBDJ: Priority sector credit as proportion to total bank deposits in
sam pie bank branches of Jaipatna block.
PSCBCJ: Priority sector credit as proportion of bank credit in sample
bank branches of Jaipatna block.
PSCBDN: Priority sector credit as proportion to total bank deposits in
bank branches of :-;arla block.
PSCBCJ: Priority sector credit as proportion of bank credit in sample
bank branches of Narla block.
I-IS
However, an increasing or decreasing proportion of credit flow to
priority sector out of total bank credit or deposits will not necessarily
be accompanied by an increasing or decreasing share of credit to
agriculture sector. Our analysis suggests that although more and
more proportion of total bank credit is disbursed to priority sector as a
whole over a period of time, the share of agriculture sector from net
bank credit has been declining in both the regions. A comparison of
AGLCBD and AGLCBC ratios between two regions (Graph 2) put the
following observations:
The AGLCBD ratio is declining in the service area of sample bank
branches for both the blocks over a period of time. Surprisingly,
the AGLCBD ratios in the bank branches of backward block were
more than that
of thp hlnr-lr fnr thp ""o .... ..... + ..... - .. ...... -- _ _" _____ ___ . _____ __ ___ ..................... ......... .. " ...... ->t..UlLlll5
from 1996-97 to 2000-01. It implies that the banks in the
backward region were mobilising a higher proportion of their
deposits for agricultural lending compared to the banks in the
advanced region. This supports our previOUS finding of higher
growth rate of agricultural credit in the backward region compared
to the advanced one.
A higher AGLCBC ratio of backward blocks indicates that branches
were showing their interest to disburse a higher proportion of net
bank credit to agriculture sector compared to the advanced one.
Given the risk on agricultural lending, therefore, that the collateral
based lending policies were more prominent in the service areas of
banks in the backward regions (see Table 5.9).
149
Graph 2: AGLCBD and AGLCBC in the Sample Bank Branches
of Jaipatna and NarIa Block
-
c:
CII

...
0
"-
30
CII
a..
20
10
0
1997 1998 1999 2000 2001
Year
-+-AGLCBDJ -4-AGLCBCJ --.--.AGLCBDN AGLCBCN
Note: AGLCBDJ: Agricultural credit as proportion to bank credit in sample
bank branches of Jaipatna block
AGLCBC: Agricultural credit as proportion to total bank credit in
sample bank branches of Jaipatna block
AGLCBDN: Agricultural credit as proportion to bank credit in sample
bank branches of NarJa block
AGLCBN: Agricultural credit as proportion to total bank credit in
sample bank branches of NarJa block
Conclusions
This chapter concentrated on an analysis of the impact of broadening
the coverage of priority sectors on credit flow to agriculture at the
bank branch level in developed and backward blocks in the selected
district. The key findings are that the share of priority sector from
total bank advance had increased in both the regions, whereas the
share of credit flow to agriculture had declined during the same
period. Thus, the higher growth rate of credit to priority sector as a
whole was due to a higher growth rate of credit disbursed to sectors
48
" Activities covered under this sector include (i) Transport operators, (ii) Retail trade
and Small Business, (iii) Professional and Self Employed, (iv) Educational Loans,
(v) Housing Loans, (I) Consumption loans (g) Small Scale Industries, and (h)
Miscellaneous activities (NABARD 2001-02).
other than the agriculture within the priority sector. It indicates that
the bankers were giving preference to non-agricultural activities. This
was widespread phenomenon in the service area of banks in the
advanced region.
It is observed that the quantum of supply of agricultural credit
(in absolute figures) in the service area of selected bank branches in
the advanced region was almost two times of credit disbursed to this
sector in the backward regions for the last five years. However, a
relatively greater share of credit to agriculture from total advances was
observed in the backward block comparee! to the advanced one. Two
Important indicators, viz., interest income on advance and type of
security-wise supply of agricultural credit was adopted in the analysis
to examine this. A comparison between the two regions on interest
income on total advance and the proportion of credit going to agriculture
from the total amount of loan was found to be inversely related. In
other words, for the bank branches where the proportion of total bank
credit was found more for agriculture sector, their interest income was
found to be less and vice-versa. It is also observed that the flow of
ugricultural credit was largely governed by the collateral security (land
or gold) that the stock of capital offers to the financial institutions
odvancing credit rather than by the factors influencing the demand for
credit. Not surprisingly, more the proportion of total advance for
awicultural credit, more prominent was the form of collateral based
lending.
Inter-regional disparities in the flow of credit to the priority
"ctor, in general, and agriculture, in particular, as a proportion of the
t'll,\l deposit in the bank is a common phenomenon in the
institutional credit markets. Any analysis of inter-regional differences
in priority sector lending from net bank credit (PSCBC) and deposits
(PSCBD), in general, and agricultural lending to total bank credit
(AGLCBC) and deposits (AGLCBD), in particular, were undertaken for
both the regions. In a region, where PSCBD was greater than PSCBC,
it indicates a better performance of banks in the mobilisation of
deposits from the concerned service area for priority sector lending.
On the contrary, if PSCBD was less than PSCBC, it implied that there
was less credit flow to priority sector out of loanable funds generated
in the form of the bank deposits. Hence, given the positive demand for
institutional credit, if PSCBD was less than PSCBC in a particular
bank branch, then any initiative by the bankers to step up the first
ratio would result in more supply of credit in the service area.
Therefore, one can suggest that since the quantum of credit disbursed
to the priority sector is calculated as a percentage of net bank credit
disbursed, this may not reflect the actual and complete picture for the
region where PSCBD is less than the PSCBC. Further, if PSCBD is less
than PSCBC, bankers may not be achieving the sectoral targets. Hence,
specific target based on deposit should be made for the service area of
c'ach bank.
Based on the interest income on advance and the lenders'
expectation of return on agricultural lending, and the policy of
collateral based lending, it is observed that these factors affected the
C<)\\' of institutional credit to agriculture. These lending norms might
also be affecting farmers' access to institutional credit. The factors
ciL'lermining the supply of institutional credit to farm households ha\'e
')\'en discussed in great detail in Chapter VI.
Chapter 6
VARIATIONS IN THE FLOW OF CREDIT TO FARM
HOUSEHOLDS: A MICRO LEVEL ANALYSIS
Introduction
Poor access to institutional credit by peasant farmers In India (and
elsewhere) is one of the important in the formal credit
market. The empirical studies show that formal credit is accessed
more by the well to do among the rural people and have seldom
benefited the poorer farmers (Rao 1975; Lipton 1976; Adams and
Vogel 1986). It has also been shown that the distribution of formal
sector credit has been unequal with respect to region, class, and caste
(Ramachandran and Swaminathan 2001). Thus, despite ma.Jor
structural changes in credit institutions in terms of outreach and
directed credit policy system, the poor access to institutional credit
has been an outstanding problem for the great majority of farm
households. In this context, an attempt has been made to examine the
disparities in the flow of credit to different categories of farmers and
analyse the factors that contribute to these disparities.
Flow of credit can be examined from two different angles. First,
now of credit viewed from the supply side can be said to depend upon
the security offered and the lender's assessment as regards the
[cpayment capacity of the borrower. However, from the borrower's
perspective it is the borrowing capacity'9 that influences the access to
fr)rmal credit. This suggests that the actual flow of credit depends on
Capacity to take a loan refers to the household's capacity to meet the formal
selection criteria and banker's additional expectation on capacity. In
order to win over the lender's confidence, the borrower must be in a pO::)ltion to
satisfy the lender with some collateral both tangible and Intangible.
the factors governing both the supply and demand for credit. S1l1ce
access to formal credit is not common to all farmers, it is necessary to
analyse the factors that account for variations in the supply of
agricultural credit, Even if some farmers are accessing formal credit, it
may not be sufficient to meet their requirement. If the disbursed loan
amount is largely influenced by the supply-side factors, a larger gap
between the supply and demand for credit is the likely possibility.
Furthermore, the magnitude of this gap may vary across different size
classes of landholdings. The issue of credit-gap across the size of
landholdings has been taken up in Chapter-VII. In this chapter, an
attempt has been made to examine the factors that account for
variations in the flow of credit at the household level. It is to be noted
that the variations in the flow of institutional credit is possible in tenns
of both access and amount obtained. Thus, the analysis of these issues
is essential to understand the agrarian credit market in the study
area.
Nature of Credit Demand of Different Categories of Farmers
The nature of credit needed by different categories of farmers can be
explained from the micro level analysis on the flow of credit to
"griculture. Broadly, farm households can be grouped under four
. lasses. The first sub-category includes marginal farmers (owning less
than 2.5 acres of land), who belongs to the bottom range in terms of
:ncome and asset holdings. They are partly involved in seasonal
both as labourers and cultivators. Since the subsistence
tiny holdings is difficult, they are involved as labourers in non-
:.Irm activities like forestry, mining, construction, transport, and
)ttage and small-scale industries. Inadequate income of this
: farmers compels them to go for consumption credit during the slack
months. They also need credit to meet contingent consumption,
working capital, and to purchase small productive assets, which are
important in providing supplementary income. This sub-group obtains
credit basically under poverty alleviation scheme.
The second sub-category (owning between 2.51 and 5 acres of
land) includes small farmer households. Although agriculture is the
main source of their livelihood, they derive a part of their income from
wage employment in farm and non-farm activities. Even if this
category of farm households occasionally faces shortfalls to meet their
consumption expenditure (including contingent consumption), they
need credit mainly to meet the working capital needs of their
production activities. Nevertheless, they also require additional credit
to acquire agricultural assets such as livestock, small tools and
machines, pump set, bore wells and so on. Farmers under this sub-
category meet their credit needs partly by hypothecation of loan assets
and personal security, and partly by providing collateral security like
land, gold, etc.
Generally, the third and fourth subcategories of farmers Iviz.,
;nedium (owning 5.01-10 acres of land) and large size (owning more
chan 10 acres of land) farmersJ fall under the affluent section of farm
households. They go for commercial production of crops given their
('eonomic status. Along with farm activity they also engage themselves
Jrl allied and non-farm activities like animal husbandry, fishery, farm-
forestry, agro-processing, manufacturing, trading, re-Iending, etc.
C;lven the economic status, these categories of farmers have a steady
li<'mand for working capital and term loans from formal financial
lrlstitutions.
It can be inferred from the foregoing discussion that credit is
necessary (may be in different magnitudes) for almost all the
categories of farmers to meet their requirements as well as for their
future economic development. This study has adopted the
stratification discussed above to analyse the credit flow to different
categories of farm households from the formal financial institutions.
Since issues on agricultural term loan and allied to agriculture had
been discussed in chapters 4th and 5
th
, in this chapter focus has been
given to the production credit (short-term loan) given by the banks to
meet the working capital needs of different categories of farmers.
Sources of Data and Econometric Methodology
Data Base and the Rationale of Sample Selection
As has been noted, there is a tendency for formal credit to flow
towards agriculturally more developed regions and to relatively larger
farmers, leaving the backward regions and small farmers to the
informal market. Accordingly, data from the service area of four
different bank branches for two such regions have been collected to
analyse the patterns in access to formal credit. In the present studv,
the level of technologySO used has been considered as a proxy variable
to define the advanced and backward regions. Based on this indicator,
the blocks of the Kalahandi district have been categorised into two
clifferent groups, the two blocks - Jaipatna (from the advanced group)
;1nd Narla (from the backward group) - have been randomly selected.
()ne CB and one RRB have been randomly chosen from each block.
From the service area of each bank, on the basis of the number of
borrowers in the last three years, the villages have been grouped.
ConsumptlOn of fertiliser per acre of gross cropped area has been usccl as prox:,
!or the level of technology usecl.
Among these groups, in turn, one village group has been picked at
random, and a sample of fifty borrower households (that is, a farmer
who has borrowed at least once during the period 1999-2002) has
been randomly selected. In order to avoid the problem of under or
overrepresentation, each group of villages has been chosen on the
basis of having approximately the same population size. Since peasant
farmers in the research area rarely get crop loans during the Rabi
session, the analysis is restricted to the Kharif crop for the year 2001-
02. The per acre paid-out cost of production for all categories of
farmers has been calculated in relation to the existing technology, so
as to ascertain the extent of production costs met by formal and
informal credit sources.
Econometric Methodology
The use of probability models IS conceptually preferable to
conventional linear regression models when the dependent variable is
dichotomous, because parameter estimates from the former overcome
most weaknesses of linear probability models. In other words, they
provide parameter estimates, which are asymptotically consistent and
efficient. In this chapter, a Pro bit model has been employed to study
rhe determinants of access to credit. The general model is a binary
choice model involving estimation of the probability of access to credit
Iy) as a function of a vector of explanatory variable (x). It is assumed
that there is an underlying response variable y" defined by the
regression relationship
13'
}' = x +u
, "
(1)
In practice, y'j is unobservable, what we observe is a dummy variable
!f defined by
y= 1 if y',>O (access to credit)
=0 othen.vise (not access to credit)
(2)
From the above relations, we get
Prob (Yi= access to credit) = Prob (u;>-fJ'x;)
= l-F(-fJ'x,}
(3)
where F is the cumulative distribution function for u. In this case, the
observed values of y are just realisations of a binomial process with
probabilities given by equation (3) and varying from trial to trail
(depending on xil. Hence the likelihood function is
L = IT F(-/I'x, )IT[I-F(-/I'x,)]
(4)
),.0 yR"1
Taking the logarithm of L and maximising with respect to p, which
gives the ML estimator of slope coefficient.
Sample Selection Bias
Basically, determinants of credit equations are estimated using a sub-
sample of farmers who have access to credit. Although this may lead
to sample selection bias, Heckman (1979) developed a joint maximum
likelihood procedure to correct this. The procedure involves estimating
both the probability of access to credit and amount of credit obtained
simultaneously. Since this approach requires identification of the
credit equation, actual size of landholding has been used as the
appropriate variable. For example, peasant farmers who have access
to credit, however, are not a randomly selected sample of all farmers
in the population. Heckman (1979) developed a solution (two-step
procedure) to this problem, and this solution variable (Inverse Mills
Ratio (IMR)) will be added as an explanatory variable in the credit
function to tackle the problem of selectivity bias. To formalise the
above explanation, let the equation that determines the sample
selection be
[
' - 'Z
, - r ,+u;
and let the equation of primary interest (determinants of credit) be
The sample selection rule is that y is observed only when I is greater
than zero. Suppose, as well that E and u have a bivariate normal
distribution with zero mean and correlation p, then we may write the
model as
EfYi I Yi is observed] = EfYi I ri >OJ
Where ex .. y' Z; / au
= E[Yi I Ui > -r ' ZJ
= fJ'Xi + E[Ci I Uj >-r 'ZJ
= fJ'Xi.+p U
e
A.i (au)
=fJ'Xi + J ~ A . j (au)
and l(a.)
;(y'Z,Iu,)
",(y' Z,I u.)
1j II; > 0 = ElY,II; > 0] + v,
.. fJ' Xi + fJ1A., (a,) + V,
Analysing the determinants of credit using the Ordinary Least
Squares regression for farmers who had access to credit produces
inconsistent estimates of ~ . We can view this problem as omitted
variable. So, least squares regression of y on x and A. would produce
consistent estimates, but if A. is omitted, the specification error of an
omitted variable is committed. Based on this observation, Heckman
proposes a two stage procedures. In the first stage, the discrete choice
model is estimated by Probit on the entire sample. Using the
estimates, the lambda has been estimated and included in the second
stage estimates of the structural relationship on the selected sample
of non-censored observations.
Emerging Patterns in Rural Credit Market
Of the 200 sample farm households, 86 per cent of them depended on
credit (that is, from both formal and informal sources) to undertake
agricultural production. However, only 56 per cent of the cultivating
households borrowed from formal credit institutions during the
reference year (Table 6.1). But, this proportion varied across the
landholding categories from as low as 19.4 per cent in the case of
marginal farmers to as high as 81 per cent in the case of medium and
large size farmers. This underlines the degree to which the proportion
of households borrowing from formal institutions rose in conjunction
with an increase in the size of landholding. Table 6.1 also shows the
extent to which farmers also depended on the informal credit market.
il shows a positive association between SIZe of landholding and the
proportion of households depending on informal sector except the
large size farmers. Importantly, the proportion of formal borrowers
who had additionally taken out informal loans also increased with
respect to size of landholding, again with the exception of the large
farm category. Based on information contained in Table 6.1, the
following inference can be drawn:
(1) the credit rationing favours larger farm households that are
economically sound and in turn the access to formal credit; and
(2) even if farmers who are accessing credit from formal sources, the
amount obtained is inadequate for their requirement, necessitating
to depend on informal sector.
Table 6.1: Distribution of Borrowers br Farm Size 12001-02)
Farm Size in Acres)
I 0.01- 2.51 - 5.00 5.01 - 10.01 and Total
2.50 (Small) 10.00 Above
I
(Mare:inal1 (Medium) (Large) i
All Households" 67 (33. 5) 50125.0) 47 123. 51 36118 0)
20011001
All Borrowers 42 48 47 35 172
% to Total (62.7) (96. 0) (100) (97.21 (86.01
Households
Formal Borrowers 13 32 38 29 112
% to Total (19.4) (64.0) (80.9) (80.6) (56.0)
Households
Informal Borrowers 36 38 43 22 139
% to Total (52.9) (77.6) (91. 5) (61. 1) 169.51
. households
No. of Formal 8 21 34 16 79
! Borrower taken
! Informal Loan (57. 1\' (67.7\" 189. 5\" 155. 2)' (70. 5)'
Note: 1) The figures In parenthesIs In row (U) are percentages of farm households In
each category from the total sample of 200.
2) The figures in parenthesis in each cell are the percentages of farm
households from the respective group total.
3) The figures in parenthesis as indicated by (.) represent the percentages of
households borrowing from formal institutions what have also taken
informal loans with respect to the farm size category. This gives us the per
cent of formal borrowers in each farm size category depending on informal
credit market.
Source: Field Survey
Table 6.2 indicates the variation In the share of total
institutional credit among different categories of farmers. In the case
of marginal farmers, who constituted 33.5 per cent of the total sample
of farm households, obtained 4.2 per cent of the total credit
disbursed. The medium category farmers, consisting of 23.5 per cent
of the sample households, obtained 30.4 per cent of total formal
credit. It is interesting to note that 47.2 per cent of the funds made
available through formal credit had gone to the large size farmers.
This shows that a small proportion of the total farmers in the sample
vJllage accessed loans from financial institutions (only 56 %), and
among these a small group consisting of medium and large sIze
.
,
:
,
I
,
,
i
I
I
I
farmers engrossed a lion's share of the net volume of credit disbursed.
The above evidence confinns that in Kalahandi district - as elsewhere
in India - institutional credit tended to gravitate towards the better-off
fanners. These findings are also corroborated by some of the earlier
studies [see Lele 1981; Gonzalez-Vega 1984; Adams and Vogel 1986;
Braveman and Guasch 1986; Egger 1986; Sarap 1990; and Basu
1997).
Table 6.2: Distribution of Loan by Farm Size in the Year 2001-
02
(in per cent)
Farm Size (in Acres)
0.01 - 2.50 2.51 - 5. 00 ' 5.01 - 10.00 10.01 and
(Marginal,
(Small'
(Medium, Above
ILarge)
I Proportion of Formal Loans 45. 5 68. 6 55. 3 72.5
I to Total Loan Borrowed
Proportion of Formal Credit
obtained by the Group to 4. 2 18. 2 30. 4 47. 2
Total Formal Credit
Source: F,eld survey
Hence, the undeniable expansion of institutional lending in the
rural areas has failed to reach the marginal and small farmers.
Notwithstanding the policy of lending rate deregulation
51
, and the
measures suggested by Gupta Committee
52
[1998) aimed at improving
the quality of credit delivery system in rural areas, CBs have failed to
cover marginal and small farmers, in general, and to increase the
credit flow to agriculture sector, in particular. This, in turn, raises a
" Since the interest rate charged was unable to cover the cost of lending (i.e.,
financial, transaction and risk costs), lending rate of RRB was deregulated for all
sizes of credit limit from August 1996. Prior to this, lending rates of CBs were also
deregulated for credit limit of over Rs. 2 lakhs since August 1994.
S2 The Gupta Committee was set up in 1998
52
to look into the quality of credit
delivered by the banking system, to identify the constraints faced by CBs in
increasing the flow of credit, introduce new product and services, and simplify
procedures to enable rural borrowers to access adequate and timely agricultural
credit from CBs.
rather obvious question: why do some farm households have access to
institutional credit while others do not? This issue has been discussed
in the subsequent section.
Table 6.3 shows that out of 200 sample households, only 48
farmers had access to formal credit to grow kharif crops for three
times during the last three years. Of the 48 farmers, almost 68.5 per
cent belonged to medium and large categories, while the rest belonged
to marginal and small farmer categories. In the case of access to credit
for two times during the reference period also, the number of medium
and large size farmers were relatively more as compared to small and
marginal farmers. However, the frequency of one time credit obtained
was more in the latter group of farmers than the former one. As can
be seen from Table 6.3, the frequency of accessing formal credit by
small and marginal farmer was relatively less than the big size
farmers. The number of times credit obtained is positively associated
with the size of land holding.
Table 6.3: Landholding-wise Number of Times Crop Loans
(Kharif) Obtained by Farmers for the Period 1999 - 2000 to
2001- 2002
(in per cent)
Size or No. or Farmers 3 Time. 2 Time. 1 Time
Landholding; in the GrouL
Up to S.OO Acres 117 15 12 45
J31.2) (37.5) (60.81
5.01 Acres and 83 33 20 29
Above (68.8) (62.5) (39.2)
Total
200 48 32 74
jl001 (1001 (100)
Note: The figures In parentheSIS are In percentage tenns. Because of poor irrigation
facility. fanners hardly grew the Rabi crop. Poor support of agricultural
extension seIVices restricts the fanner to go for high yielding variety of cash
crop. Thus. bankers hardly provide any loan during Rabi season.
Source: Field Survey
I
Table 6.4 reveals that the loan amount per acre of gross cropped
area varied across size class of landholdings within and across the
regions. The farmers in the advanced region obtained, on an average,
Rs.l,249 per acre of gross cropped area, whereas it was only RS.915
in the backward region. Based on the information from Tables 6.3 and
6.4, it can be argued that there was variation in terms of both access
to credit as well as amount of credit obtained from institutional
sources. The contributing factors for these variations have been
discussed in the subsequent sections.
Table 6.4: Supply of Credit (Crop Loan) Per Hectare of Gross
Cropped Area
(in Rupees\
Size ClaD or Advanced Backward
Landholdin&& (in Acres) Region Region
0.01- 2. 50 475 435
2.51 - 5. 00 1 193 1,176
5.01 - 10.00 1308 865
lO.OI and Above 1347 927
L All Classes 1,249 915
Source: Field survey
Factors Affecting Access to Formal Credit
Since credit transaction necessarily involves a time gap between a
point when the advance is made and a future point when credit is
repaid with interest, anything may happen in-between these two time
points. Examined from the lender's viewpoint, the act of borrowing is
one, which carries both pre-contractual (at the point prior to credit is
given), and post-contractual problems (at the point when credit is
repaid). The former difficulties take the form of adverse selection (i.e.,
the fear of selecting a bad borrower), creditworthiness and screening
of the borrower. In the latter case, a lender faces the problem of wilful
and non-wilful default. Given these problems, what governs access to
institutional credit by some peasant households but not others?
It should be stressed that access to credit is influenced both by
the farmer's capacity to take a loan and the lender's willingness to
extend it. This exchange usually involves a signalling and screening
procedure in order to communicate information about the riskiness of
the loan (Petrick and Latruffe 2003). In this context, the farmers'
access to institutional credit is estimated by using a Probit Model
where the dependent variable Yj is a dichotomous (1, 0) variable
indicating whether the i-th farm household has access to credit or not.
So, the dummy dependent variable = 1, if the farm household had
access to a loan from formal sources, and = 0, otherwise.
It is assumed that the access to agricultural credit is
determined by a number of farm and household characteristics. Since
loans are advanced to farmers-who-borrow on the basis of land they
own, the latter functions as collateral from the lender's point of view.
Hence, the actual size of land holding can be taken as one of the
important factors governing access to institutional credit. The land
quality also improves the !ann household's prospect of obtaining loan.
Better the quality of land, more likely the prospect of a farm
household's access to a loan. Apart from land, other valuable assets -
particularly agriculturaI implements, and the proportion of non-farm
income to total income can be used as an indicator to judge the
creditworthiness of the borrower. Indeed, bankers may prefer a
borrower who has income sources apart from an agricultural one so
as to minimise what they see as the risk of non-repayment. These two
variables are expected to positively influence the access to credit. An
assured source of irrigation that added to the reliability of crop
production and yield is similarly perceived by banks in a positive light,
and thus as a contributory factor to borrower credit-worthiness. Thus,
the percentage of irrigated area to gross cropped area is expected to be
positively associated with access to formal credit. Another issue is that
the credit-disbursing officials may discriminate against lower caste
and tribal households because they have fewer outside connections
and higher caste decision-makers dominant in the credit institutions.
Accordingly, the caste variable is expected to be negatively correlated
with access to formal credit facility. Conversely, formal education may
have a positive influence on access to credit. The definition and
descriptive statistics of the variables used in the analysis are provided
in Table 6.5.
Table 6.5: Definition, Measurement, Descriptive Statistics,
and EXlected Sign of Variables Used in the Probit Equation
All Farmers
!
Medium &
Variable Description
Expeeted
I:tmau :'lZe Large Size
Sip
Farmers farmers
Mean Mean Mean
Dep. Var Access to Credit -1; 0,
0.560 0.385 0.81

otherwise
10.50) (0.49) (0.40)
ASLH Actual size of landholding (in + ve 6.25 2.47 II. 58
(6.20) (I. 56) 3B)
I LANDQ
Land quality + ve 0.505 0.28 0.82
- 1, for good land quality (0.50) (0.45) (0.39)
- 0 otherwise
IASSET Value of assets excluding + ye 1,547. 18 2,450.68 33,813.31
land (in Rupees) (47,362. 02) (3,260.52J
IEOFA Income excluding own farm + ve 35.64 47.79 lB. 50
activity as proportion to total (32.11) (33. 17) (20.95)
family income (in Rupeesl
PIAGCA Percentage of irrigated area to
+ ye
22.92 16. 17 32.44
area 136.29L J32.
139.70)
CASTE Caste status - ye 0.38 0.49 O. 22
-1 ,if the borrower belong to (0.49) (0. 50) (0.41)
SC or ST
= 0, otherwise
EDU
Educational quaIification of
I the borrower
+ ye
0.21 0.08 0.40
I
- 1, for 10'" or more standard (0. 41) (0.27) (0.49)
- 0, otherwise
, Number of Observations 200 117 B3

f rea in
11) Land particulars were collected from each and every farmer 10 terms a a (
Acres) under low, medium and high in the survey. It IS found that the high land
was completely rain fed and the probability of crop failure was more. In the case of
low land the possibility of crop failure was minimal. Under this assumption,. If the
area under low land was covering at least one third of total land owned, It was
considered as farmer having good quality land.
(2) Figures in parentheses indicate Standard Deviation.
Table 6.6: Access to Formal Credit: Pro bit Results
IDummy Dependent Variable = I, if the farm household has access to credit: = O.
Otherwisel
r Variables All Farmers Marginal and Small Medium and Large
-- -;
Size Farmers Size Farmers
I
I
I Coefficient Marginal Coefficient
1
Marginal Coefficient Marginal
. i
I Constant
Effect Effect ._ Effect.J
0.33221'
i
0.19000 -0.0026 - 0.00091 - 0.08633 - 0.00056 \
12.12) 1- 0.01) (- 0 16)
-
! ASLH
-0.03491 - 0.01149 0.1593 0.05544 - 0.06629 - 0.00043
I
1- 0.92) (1.35\ . (-I. 5S\ i
: LANDQ 0.75417' 0.24510 0.3192" I
O. III 051
1.51138"
0.00981-
13.391
(3.0S)
11.98) i
ASSET 0.00005" 0.00001 0.00002'" 1 0000006 0.00007S'" j 0.00000 I
(1.97) 11.671 \ 11.731 i
: IEOFA -0.02IS8' - 0.00724 - 0.0257' - 0.00895 - 0.00304 I - 0.00002
\
(-5.19) (- 3.93) (-1.62)
PIAGCA 0.00321' 0.00105 O. 0062'" 0.00214 0.00381 0.00002
.12.05) ( 1.67) {1.61)
i CASTE - 0.10025 - 0.03323 0.0873 0.03036 - 0.40360'" - 0.00262
,
_(- 0.77) {0.29) {-I.S2)
,
I EDU 0.32436***
n 'An"l
v .... vVV.l
" CI<Ae- "
v. u"T"'tJ 0.18947 - 0.31/99 - O.OO2Ut>
i
(l.60) 11.85) 1-0.711
I Log-likelihood - 91.35 - 51.17 - 28.16
: Restricted log- - 137.19 - 77.95 - 40.69
- likelihood
. Chi-square (7 dfl 91.68 53.56 25.05
Number of 200 117 83
Observations
Note: (a) , 1 % level of signtficance; .. 5 % le\'el of slgmficance; , .. 10 % level of
significance
(b) Figures in parentheses indicate t - values.
The determinants of access to formal credit are estimated by
using Probit model, which is given in Table 6.6. The assumption made
by the Pro bit model is that the probability of borrowing from the
formal sector is determined completely by the bank, which decides
II'hether or not a farm household will get a loan. This model also
dssumes that all cultivating households demand formal credit at the
,-xisting interest rate, and that the formal sector is the cheapest
'()urce of credit for farm households. This probability is represented
:)\. a univariate normal distribution representing the access equation.
\\I\ich is specified in the methodology.
,
,
I
,
,
The result shows that ASLH is negative but not significant. This
IS contradictory to the earlier findings that larger the size of land
owned by the household, greater is the probability of its access to the
formal sector (Kochar 1997 and Swain 2002). The negative sign in our
study might be attributed to uncertainty of crop income owing to poor
infrastructure facility in the study area. Importantly, the quality of
land (LANDQ) has a positive and significant effect on the access to
formal credit. It is quite interesting to note that the marginal effect of
land quality increases the likelihood of access to credit by about 24
per cent. This implies that it is not the size of landholding but the
quality of land, which matters for the bankers in the selection of a
borrower, and in tum, increases the probability of farmers to access
institutional credit. This variable behaves in the similar direction for
different sub-groups of farm households.
On the issue of non-land resources, the coefficient of ASSET is
positive and significant at 5 per cent level, which implies that larger
the value of asset excluding land, higher the probability of access to
formal credit. Ceteris Paribus, if the farmer is well equipped bv
agricultural implements, there is a possibility of getting more income
from farm activity, which, in tum, provides an indication of his (or
her) credit worthiness as a borrower. From this it is possible to infer
that peasant farmers who dispose of less valuable non-land assets may
have a correspondingly poor access to institutional credit. A farmer who
is having higher non-farm income as a proportion to total income is
less likely to have access to credit. This is evident from the fact that
IEOFA coefficient is negative and highly significant. This result
indicates that farmers with non-agricultural income may use it to
meet their working capital requirements, and thus have no need to
approach a bank for a crop loan. Alternatively, a bank may quite
simply not have adequate information about the sources and amount
of non-farm income to judge the creditworthiness of the borrower. This
suggests that bankers prefer tangible collateral (like land, quality of
land or other assets) to select the borrower.
A better irrigation facility also improves the farm household's
prospects of obtaining a loan from the banks. This explanation is
confirmed by the positive and highly significant coefficient of PIAGCA.
However, the coefficient of this variable is positive but not significant
at the conventional level of significance for medium and large size
farmers. As expected, scheduled caste and scheduled tribe farmers are
less likely to access formal credit whereas this coefficient is
surprisingly not significant except for the group of medium and large
size farmers. However, education increases the probability of access
to credit, i.e., higher the formal education, higher is the probability of
access to credit (this coefficient is being positive and significant at a
10 per cent level). In this context, it can be argued that educational
attainment gives better information regarding the available banking
facilities, and empowers the ability of farmers to approach a bank for
loan
53
. Nevertheless, this coefficient is negative and not significant for
medium and large size farmers.
Generally speaking, the results are quite in line with theoretical
expectations and draw a plausible picture of the pattern of farm
household's credit access in Kalahandi district. The highly significant
chi-square and Pseudo R-square (33 per cent) clearly shows that the
51 It is evident from the marginal effect that a one per cent increase in educational
attainment increases the likelihood of access to credit by some 10 per cent.
estimated model is having good fit. The model points to a high degree
of rationing by the formal sector. Evaluating the probability of access
at the mean levels of the explanatory variables, 74 per cent of the
households, belonging to all the categories are credit rationed by the
formal sector
54
.
Determinants of Supply of Agricultural Credit
Directed agricultural credit system assumes that all farm households
would have a positive demand for formal credit and it is a cheaper
source for borrowing. It is further assumed that both revealed and
potential demand for credit far exceeds the supply. On the basis of
this assumption, many countries and financial institutions rely on
supply-led approach to extend their credit facility. Irrespective of this
approach, most theoretical and empirical studies argue that demand
far outweighs the supply of credit. This leads to a situation where
banks are compelled to adopt wide scale credit rationing.
Credit rationing exists in two forms, VIZ., price and non-price
credit rationing. In the former case, the borrower is denied access to
institutional credit regardless of whether or not he or she has fulfilled
the stipulated terms and conditions. In the latter case, the farmer is
able to borrow, but not as much as he or she wants or needs. The first
type of credit rationing explains about variations in the access to
credit, i.e., what makes some farm households to access formal credit
and others not. The second type explains on variations in the amount
of credit flow from the formal sector. Alternatively, on what basis
bankers determine the credit limit to the borrower? In our previous
i; Based on similar assumptions as adopted in our model. Kochar (1997). and Swain
12002) find that 81 per cent and 71 per cent of the households in the respective
studies were subject to credit rationing by the formal sector.
model on access to credit (Probit Model), it was found that the farm
and household characteristics like quality of land, value of non-land
assets, the proportion of non-farm income, percentage of irrigated area
to gross cropped area, and educational qualification of the borrower
were the determining factors for accessing the formal credit facility.
The second model (OLS) in our household's level analysis
specifies the factors that determine the amount of credit given to the
farmer borrower by the formal sector. As was discussed, analysing
the determinants of credit in terms of farmers who have received this
is not a random selection of all the farmers in the population. Such a
phenomenon leads to sample selection bias. In order to correct this
problem of selectivity bias, Heckman two-step procedures have been
used. In the first step, the Inverse Mills Ratio (A) has been calculated
from the above Probit equation, and in the second step this has been
introduced as an additional explanatory variable in the determinants
of credit equation. If the (/') variable is significant, the resultant
equation is unbiased (Heckman 1979). The second model, which deals
with the amount of institutional credit lent, is set out in Table 6.7.
The fact of credit flow - once access has been gained - can be
looked at from two distinct viewpoints. From the supply side, the
amount lent depends on the security offered and an assessment about
the repayment capacity of the borrower. As noted, land is an indicator
of the degree of security involved in the transaction, not least because
in the eyes of a lender the higher the area under cultivation the more
likely it is that the peasant farmer will generate sufficient output to
repay the loan without difficulty. Given the self-finance capacity, the
operated area
55
should be positively related to the demand for credit,
so the production surplus
56
can approximate most closely a capacity
to repay. Because of the difficulty in calculating the production
surplus, the value of output by a farm household has been used to
indicate its creditworthiness. On the demand side, the percentage of
irrigated area to gross cropped area, expenditure on fertiliser and
pesticides, and the educational qualification of the borrower can all be
expected to have a positive association with the flow of credit. Caste
and the ratio of workers to family members, income excluding own
farm activity can also influence the demand for credit.
Accordingly, it can be argued that the actual flow of credit
depends not only upon the factors governing the supply of credit but
also on the factors influencing the demand for it. Between different
farm size-groups, disparities in the flow of credit can, therefore, be
viewed in terms of the relative strength of demand and supply factors.
In the financial situation characterised by the absence of credit
rationing, the factors influencing the demand for credit will obviously
determine the quantum of credit supply. But in the financial system,
where credit rationing (either price or non-price) exists, the factors
55 Operated area = Land owned + Leased in - Leased out. In the case of a financial
system where formal creditors insist on collateral (ownership rights of the land),
the credit extended to the farmer can assume to be negatively influenced under
the tenancy system. Given the poor irrigation facility and uncertainty of crops,
share tenancy contract operates in the study area. In this type of contract. the
gross output is divided equally between the landlord and tenant after deduction
of cost of production. Generally, the landowner only bears the cost of production
and later, he takes it out in terms of output from total production. In this type of
tenancy system the tenant only invests his labour force and not responsible for
all aspects of cultivation. In our total sample, we had 15 cases like this. To avoid
double counting we have not included the area under share tenancy in the
operated area of the tenant farmer.
56 It can be defined as the difference between total income from production and
expenses incurred by the farm.
influencing the demand for credit cannot be considered as ultimate
determinants of its supply. It is, therefore, necessary to investigate
whether the demand or supply side factors account for the flow of
credit, a question addressed in Table 6.7.
Table 6.7: Definition, Measurement, Descriptive Statistics and
Expected Sign of Variables used in the OLS Equation
All Marginal I Medium
Farmers and Small and Large
Variable
Expected
-
,
,
Description
Size Size
Sign
Farmers
Mean Mean i Mean
, Dep.
I Var. *
Amount of Credit given 11,653.17 6,497 15,116 I
to the farmers (in (10,972.91) (4,166.62) (12,673.89) I
Rupees)
I
I
: PIAGCA Percentage of irrigated + ve 31.68
I
34.6,
, area to gross cropped (38.81) (36.50) (40.27) !
area
CASTE Caste status - ve 0.28 0040 0.19 '
= I, if the borrower (0.45) (0049) (0.401
belongs to SC or ST i
,
= 0, otherwise ,
,EDU Educational + ve 0.29 0.11
0.40'
qualification of the (0.45) (0.32) (0.49)
borrower
= 1, for lOth and or
above
= 0, otherwise I
'INEOFA Income excluding own + ve 17,789.54 7,696 24.5691
farm activity (in Rupees) (31,660.67) (15,912.02) (37,343.80) .
WFM Workers to family - ve 0.43 0.49 0.39.
members (ratio)
(0.17) (0.18) (0.15)
-
01\ Operated areas (in + ve 8.02 3.88 10.81
Acres)
(6.65) (1.74) (7.26) 1
c
i
:EXPFP Expenditure on fertiliser + ve 7,188.20 3,211 9,859
I
and pesticides (in
(7,432.64) (2,268.92) (8,453.03
L
Rupees)
,VAO Value of farm output + ve 43,947.21 16,767 62.202 I
,
produced in the year
(43,309.81 ) (11,696.71) (47.132.37) i
. (in Rupees)

Inverse Mills Ratio (I,)
0.4594 0.6411 0.2339 '
of Observations
112 45 67
\ote: Figures m parenthesIs mdlcate Standard DeViatIOn
As we expected, most of the parameters have the expected sign
and are significant (Table 6.8). The coefficient of PIAGCA is positive
and highly significant, which implies that a unit (one per cent)
increases in proportion of irrigated area to gross cropped area, OIl an
average, leads to an increase of 47.55 rupees (0.1292 per cent) in the
amount of credit (for the entire sample). This variable also behaves in
similar direction to marginal and small, and medium and large group
of farmers. This is true and in conformity with earlier studies that
irrigation facility is an important factor to determine the amount of
credit disbursed by the formal lender. Non-Scheduled Caste and
Scheduled Tribe farmers enjoy more credit compared to Scheduled
Caste and Scheduled Tribe (SC/ST) farmers. This reflects in the
negative and slgmticant coefficient of the variable CASTE. The positive
and significant coefficient of human capital variable (EDU) clearly
shows that the farmers who has possessed a 10th standard level of
education and above have obtained more credit compared to their less
educated counterparts.
Not surprising is the link between expenditure on fertiliser and
pesticides, on the one hand, and the lender provision for more credit,
on the other. Fieldwork in Kalahandi district confirms that the per
acre expenditure by large and medium sized farmers on these inputs
was very high. This, in turn, not only raised agricultural productivity
but also enhanced their creditworthiness. Each one per cent increase
in EXPFP lead, on an average, to a 0.5329 per cent increase in the
amount of credit (for the total sample).
i
Table 6.8: Determinants of Formal Credit: Selectivity
Corrected OLS Result
(Dependent Variable = Amount of Credit given to Farmers (in Rupees))
c----.. ----.
Coefficients
- - ---- -----
-- 1
- --- .
Variables
All Farmers Marginal and Small . Medium and Large
Size Farmers I Size Farmers ,
! Constant 1.844.540
3.578. 14 '" )
--- - c;-;I
683.31.
,
(0.701 (1. 80)
(0. IILl
,
PIAGCA 47. 548 ' O. 22 *** 74. 15' .
(2.631
-<1.72) _---.J2. 58) I
CASTE - 3.595. 320 ' - 2,922. 92 ' - 5,869. 32 1
(- 2.50) (-3.26)
12 45J
EDU 3.166.493 '" 1.133.96 , .. 3.204. 80 ... i
JI. 89) (1.88)

INEOFA O. 002 0.0] ***
O. 0 I I
,
(0.061 (1.65) \
(0. 5!fl
WFM - 1.291. 389 5.227. 17 " - 2.099. 97 ,
(- 0.34)
J218)
(-0. 32) ;
OA - 197. 011 - 1.048.33
14. 91 I
(- 0.79) 1-2.681
In n4i
I EXPFP 0.864' I. 87 '
0:68'; I
I
(2 211 I
(3. 57) (5.84)
\ VAO 0.072 ..
0.41***
O. 06 .. u :
(1. 86) (1.86)

I Inverse Mills Ratio 2,464. 128 - 2 117 65 , ..
4.341.08"1
12.00)
'(-1.65)
(196)
R - squared O. 69 0.71

Adiusted R - s,quared 0.66 0.64 0.61
F(9, 102 df) 25.39' F(9, 35 dO = 9.73 F19, 57 df} = 12.57 I
, Rho O. 39 - 0.79 0.57 I
Standard error 6,000.81 2.682.9 7.6151
corrected for selection
671 ' Number of 112 45
observations
Note: (a) , I % level of SIgnIficance; ., 5 % level of slgmficance; ... 10 % le\'el of
significance
(b) Figures in parenthesis indicate t - values,
The observed negative and insignificant operated area (OAI and
significant coefficient value of agricultural output confirm that it was
the latter rather than the size of the operational holding, which was an
important factor that influenced the formal lenders to fix up the credit
:imit In the case of all sample farmers, every one per cent increase in
,he value of agricultural output raised the amount of credit by 0.27 J 5
175
i
per cent. The negative and significant coefficient of OA for marginal
and small farmers indicates that even if OA increased, it lead to less
supply of credit. This can be explained by the fact that the OA of most
of the small and marginal farmers included land leased in also. In the
case of the financial system, where formal creditors insisted on
collateral (ownership rights of land), credit extended to the farmer
would, therefore, be negatively influenced under the tenancy system
S7
.
The coefficient of worker to household members (WFM) was positive
and significant for marginal and small farmers. This indicates that
increase in WFM could lead to the supply of more credit. Bankers
might have perceived that the increase in the number of working
members could provide more income through wage or any other
economic activity (either farm or non-farm). Probably, this might
enhance their creditworthiness. This argument can also be supported
by the positive and significant coefficient of INEOFA for marginal and
small farmers.
The result shows that the Inverse Mills Ratio (A) was positive
(negative for marginal and small farmers) and significant and the
resultant equation was unbiased. The R-square values were 0.69, 0.71
and 0.66 for the estimated equation of all sample farmers, marginal
and small, and for medium and large size, respectively. This shows
that 69 per cent, 71 per cent, and 66 per cent of the variations in the
amount of credit obtained by all farmers, marginal and small, and
medium and large farmers, respectively have been explained by the
included variables in these models. Further, the whole significance of
" Out of 45 marginal and small farmers who had obtained loan from the formal
banking system, 42 per cent of them were cultivating through fixed tenancy
system.
176
the model was also highly significant at 1 per cent level, which
rejected the hypothesis that all the slope parameters are zero.
Table 6.9: Determinants of Formal Credit: Tobit Results
(Dependent Variable = Amount of Credit given to the Farmers (in RupeesJ]
All Farmers Marginal and Small Medium and Large
Variables Size Farmers Size Farmers
Coefficient Marginal Coefficient Marginal Coefficient Marginal
Effect Effect Effect
Constant - 1,305.28 - 816.78 -5,319.94 -2,312.98 1,048.86 904.74
PIAGCA 68.63' 42.94
25.14 ...
10.93
79.99 .. 69.00
(3.40) (I.61l (2.25\
CASTE -3,874.82" -2,424.67 -1,110.96 , ..
-483.02 - 6,936.21 ' -
1- 2.50\ (-1.66) (-2.40) 5,983.17
EDU 2,341.17 1,464.98
1,370.24 ,
595.74
1,743.42 ,.,
1,503.87
(1.25\ (1.781 (1.64)
INEOFA -0.001 - 0.007 -0.02 -001 0.01 0.01
(- 0.04) (0.641 (0.33)
WFM -5,819.59 -3,641.60
1,359.97 ,
591.28 - 4,891.05 -4,219.02
1- 1.381 ' 1 . 8 4 ~ f- O.62i
lOA
15.53 9.72 -327.73 -142.49 - 90.03 -77.66
,
(0.05) (-0. 63) (- 0.22\
IEXPFP
0.87 0.55 2.16 0.94
0.75 .,
0.65
12.99\ (4.321 (2.01)
iVAO
0.062'" 0.04 0.13 ." 0.05 0.06 *** 0.05
(1.61) /1.73) (1.83)
'Sigma 8,659.41' 4,335.96 10,052.59
,
(14.28) (8.68) 111.13)
, Log -1,224.72 - 470.52 - 727.80
I likelihood
. R - squared 0.63 0.65 0.62
I. Number of 200 117 83
observations
Note: 1) FIgures In parentheses mdlcate (-values
2) '1% level significance; "5% level significance, '''10 % level
3) Marginal Effect - Partial derivative of expected value with respect to vector of
characteristics. They are computed at the means of the explanatory
variables.
The ordinary least square regression is inappropriate when the
dependent variable is discontinuous (Pindyck and Rubinfield 1997).
Logit and Pro bit models are appropriate when the dependent variable
is discrete, usually taking two values, 0 or 1. These models are useful
if the question is whether or not farmers have access to credit, but are
mappropriate when it is important to measure the amount of credit
177
obtained. The Tobit Model handles censored dependent variables
(continuous between some lower and possibly upper bound) better
and, hence, is superior to Logit and Probit (Pindyck and Rubinfield
1997). The Tobit model has been, therefore, used and the results are
presented in Table 6.9. The sign and significance level of the
coefficients are almost the same as those discussed in the OLS model,
except in the case of coefficients of lNEOFA and OA. The education
variable is significant in OLS for all the three cases, whereas in Tobit
equation it is significant only in the case of marginal and small, and
for the group of medium and large size farmers. The Tobit results
again confirm that PIAGCA, V A ~ , EXPFP, CASTE are important
variables in determining the amount of credit obtained. Further, the
co-efficient of Sigma is highly significant at 1 per cent level, which
confirms that the results are unbiased.
Conclusions
The pattern of formal credit flow at farm levels shows that the credit
rationing goes in favour of large farm households, and the obtained
amount is inadequate to meet their production expenditure for most of
the farmer borrowers who had obtained credit. Since there is
inadequate availability of formal credit, the farmers tend to depend
more on informal sector. It is clear that fonnal lenders are not in a
position to replace infonnal lenders to provide credit to agricultural
borrowers, even with a widespread network of fonnal financial
institutions, and the suggested measures by Gupta Committee aiming
at improved quality of credit delivery system.
The results on the determinants of access to formal credit
suggest that several factors were inhibiting agricultural borrowers and
178
especially the marginal and small farmers from obtaining production
credit, despite the presence of number of formal credit institutions in
the study area. Apparently, the mere existence of formal credit
institutions does not guarantee that marginal and small farmers will
benefit from them
s8
. Thus, mere location of a bank branch in a village
without restructuring its management to make it more broad-based
with proper representation of agricultural borrowers belonging to
either lower castes or illiterate or both (as noted by Sen, 1975;
Ghatak, 1977, 1983; Sarap, 1990) is not enough to ensure effective
access of formal credit to the rural borrowers.
The farm, and household characteristics such as quality of land,
value of assets excludmg land, irrigated area, and educational
qualification of the borrower determined the access to formal credit. It
is quite interesting to note that the quality of land and value of output
are found to be significant in determining the access to formal credit
and the amount obtained respectively, rather than the size of
landholding. It seems that bankers are giving preference to the safer
borrower and leaving marginal and small farmers to the unorganised
credit market. The above determinants of access to credit indicate the
creditworthiness of the agricultural borrowers. Thus, it is the bankers'
assessment on the creditworthiness that decides whether or not a
farm household should have access to credit:. The factors like area
under irrigation, value of agricultural output, expenditure on inputs
(fertiliser and pesticides), caste and educational qualification of the
borrower are the important variables that determine the amount of
credit obtained. On the whole, one can argue that the asset-based
" This finding also corroborates with Sarap's study in 1990. where field
investigation was conducted during the year 1981-2 in six villages of Sambalpur
district of Orissa state in India.
179
lending policies coupled with the bankers' assessments on
creditworthiness determines both access to credit and amount of
credit obtained by the farmer borrowers.
The estimates also suggest a relatively high demand for credit
with fairly widespread credit rationing in Kalahandi district. This
requires further development of credit programmes to support the
credit requirement of different categories of farmers, and especially
those belonging to marginal and small farmer categories.
In the econometric specification, we assume that formal
borrowing is determined by bank decision on access to credit and has
been estimated by using univariate pro bit model. However, one can
also generalise (a) where the probability of borrowing from the formal
sector is jointly determined by the bank decision on access as well as
the farm household's demand for loans, and (b) where the households
may have zero demand for formal credit and they are free to choose
between informal and formal sector. This generalisation can be
estimated using bivariate Probit and Multinomial Legit model.
Clearly, there is a need for further research and subsequent attempts
in this direction, which will hopefully allow us to explore these
al ternatives, in detail.
Chapter 7
CREDIT DELIVERY SYSTEM, CREDIT-GAP AND COPING
STRATEGY OF AGRICULTURAL BORROWERS
Introduction
An analysis of the determinants of access to formal credit in our
previous chapter showed that several factors were inhibiting
agricultural borrowers, in general, and marginal and small farmers, in
particular, from obtaining production credit. The poor quality of land,
illiteracy and lower caste status had an adverse effect on the access of
agricultural borrowers to formal credit institutions. It is also found
that small proportions of the total number of farmers in rural areas
were accessing credit from financial institutions, and out of that a
small group grabbed a large share of the net volume of credit
disbursed. Importantly, the credit rationing goes in favour of large
farmers. It is also reported by most of the farmers from the study area
that the amount obtained as loan was inadequate to meet their
production costs. Furthermore, some of the farmer borrowers stated
that they had one or other problems while obtaining credit from
formal financial institutions. Some of these problems also contributed
to the higher transaction costs and delay in the disbursement of
credit.
In this context, one can raise the following questions. Who are
these agricultural borrowers who face problems in accessing formal
credit facility? Are these problems the same across various size of
landholdings? Which class of farmers face higher transaction costs
while obtaining formal loans? Since the obtained loan amount from
formal sources is inadequate, what is the magnitude of the credit-gap
181
among farm households? Given the self-finance capacity, what is the
coping strategy of the farmers to meet their cost of cultivation? To
answer these questions both qualitative and quantitative data were
collected from 200 randomly selected borrowers from the service area
of four different bank branches
59
.
Credit Delivery System
It is argued that the disturbing trends in the rural credit delivery
system can be attributed to the poor quality of lending (Rajasekhar
and Vyasulu 1990). The supply-led approach with poor quality of
lending failed to provide the credit requirement of agricultural
borrowers, on the one hand, and made agricultural lending
un . D.rn.fi.1t:::!h.1p, <1.". nt.h.p.r. T ...... f-h;C" .. "h'" DDT ,.. .... <4- " ....... "" " .... c
_ .. _ ___ _ _ _ _ _ _ _ ...... a. .... "" .t.&l. .0"10.. .. , W .... l',L.IJ. ......... L.lp C1 UJ.! lllC1it
High Level Committee (Gupta 1998) to identify the constraints faced
by the commercial banks in increasing the flow of credit, introducing
ne 1 products and services, simplifying procedures and methods of
,
v.:urking with a view to enabling rural borrowers to access adequate
and timely credit from the commercial banking system. Major
recommendations of the Gupta committee accepted by the RBI60
includes simplification of procedures regarding loan application,
agreements/documents, etc., cash disbursement of loan, dispensation
of no dues certificate, and selection of loan application at the branch
level, and so on. With this background, let us now examine whether
the agricultural borrowers in Kalahandi are satisfied with the existing
credit delivery system in the district.
," A detailed discussion of farm household selection has been given in the sixth
chapter.
,,11 For detailed discussions on the accepted recommendations, see, Box 2.1.
182
Table 7.1 shows that a majority of the borrowers had one or
more problems on the credit delivery system. Such problems could be
grouped broadly under two categories, viz., during and after obtaining
loans. Problems in obtaining loans included complicated procedures,
lack of previous experience, distant location of the bank, and bribe. As
many as 56.5 per cent of the agricultural borrowers stated that the
bank procedures were complex. They faced serious problems in
getting No Dues Certificate (NDC) from the local financial institution.
This seemed to be a widespread problem in the area where Co-
operatives, RRBs and Commercial Banks were working. As per the
recommendation, importance should have been given to the credit
status of the borrower rather than to no dues certificate, but still the
borrowers were required to submit NDC while taking loans. The
imposition of this requirement took more time and also posed problem
to few borrowers. In spite of rapid expansion of the bank branches,
since most of the borrowers (56 per cent) did not have previous
experience of borrowing from the formal sector, it aggravated the
situation. Some of the borrowers especially those who wanted loan for
the purpose of allied activities stated that the bank officials were
indifferent to them. A general view among the borrowers of this kind
was that "unless one bribes the bank officials" they would not sanction
the loan. Based on the information, it is found that 24 per cent of the
borrowers gave bribe to the bank officials to get the loan. On the
whole, most of them felt that "it is better to obtain a loan from
moneylender than to approach the bank
61
".
" Rajasekhar and Vyasulu (1990) also found a similar problem in their study 111 Pali
district of Rajasthan.
Table 7.1: Problems Faced by the Farmer Borrowers in
Obtaining Loans
.. __ ,
Before Obtllinin.KLoan i
113[ 56:-5'_87 2QO
No 4o.O ___ 200
Location of the Bankl 961 48. 01 1041 52.0 2001
]Bribe 152;76.01'-- 200\
__
__
____
Source: Field Survey
The agricultural borrowers also stated that, the loan amounts
were inadequate and untimely. About 69.5 per cent of the agricultural
borrower out of the total sample felt that the disbursed loan amount
was inadequate and 49 per cent were of the view that it was not
available at the needed hour.
Based on these findings one of the important questions that
could be asked was who were these agricultural borrowers who faced
problems in accessing formal credit facility? Table 7.2 explains that
these problems were uneven across different size of landholdings.
Importantly, most of these problems were related to marginal and
small farmers. Out of total sample about 80 per cent of the farmer
borrowers, who belonged to marginal and small categories found that
the formal banking system was complicated. These categories of
borrowers also narrated their poor experience with the institutional
credit system. This indicates how far our banking system is good
enough in covering these borrowers. Bribing to the bank officials
seems to be a widespread problem in the case of marginal category
farmers. A general feeling among the borrowers was that bribing is
tile only means to obtain a loan for allied agricultural activities under
the subsidy and or poverty alleviation schemes". The evidence shows
that the creditworthiness, viability of the proposal, and the amount of
cash disbursed were positively related with the amount of bribe, while
the time taken to get the loan sanctioned was negatively related62.
Table 7.2: Size Class of Landholding-wise Frequency
Distribution of Problems Faced by the Farmer Borrowers
0.01=-2:5012.51:5.00 -5:01-10_00 10_01 acres
Problems I I Acres I Acres ,and Above. Total
i-_________ -LtMarglll.o!lJ... JSmallJ _--' l.Med.iuml.....:.--..JLargeJ __ '__ ___ .
Before Obtaining Loan
16 ! 14jf-7-(6:2j 113 (fOOL
(27.7). 19 (17.QL __ 5...14 5Lil 2 (1"00)
Distant Location of the 31 (32.3)1 31 (32.3)'1 23 (24.0) II (11.5) 96 (100)
@ank i 1
_______ ..L-....:4,,,"--,2 (But._3 2""'(4"'"".2L .:n: -'-::48(100)
___ _
'lnade uate Loan 41 29.5 34 24.5): ___ (100)
Untimely Loan 58 59.2: 19 19.4 16 16.3),
Note: Figures in of f3rm in each category
who have faced problems in obtaining agricultural loans.
Source: Field Survey
Bribing takes place in many forms starting from the project
initiation to sanction_ Even in the case of no direct bribing, the
bankers manipulate while buying the loan assets_ Hence, the above
practice contradicts the recommendation made by the Gupta
committee, which has preferred distribution of loan amount in cash.
In this aspect, two important dimensions were derived from the
discussion with the bankers and beneficiaries in the field areas. The
"' Chaudhuri (1993) has gone into an indirect analysis of the effects of bribing on
the delay in the disbursement of formal credit. In a sample of 75 farmers from the
two villages, 57 per cent of the fanners emphasised that the delay in getting
formal credit could be reduced by increasing the bribe and the remaining 43 per
cent did not respond. However, he mentioned that a statistical correlation
analysis between the delay and the bribing rate could not be done because the
data on the latter could not be obtained from all the agricultural borrowers who
had given bribe.
There is empirical evidence for the delay in disbursing formal credit. The dela)' is
controlled by an official of the formal credit agency, and it can be reduced by the
farmer bribing the official (Chaudhuri and Gupta, 1996). In their theoretical
model they found that, in eqUilibrium. the official either would not take a bribe or
would choose a bribing rate such that the effective interest rate on formal credit
(incorporating the bribe) would be equal to the interest rate on informal credit.
bankers were of the view that the probability of misuse of loans by the
beneficiary was more in the case, where loans were disbursed in cash.
They also argued that in the case of misuse of loans, bank looses
partially or completely the hypothecated loan asset that work as
collateral for the bank till the loan was repaid. However, the borrowers
argued that the kind components lead to distortion since their choice
was restricted. In most of the cases, the bankers got the loan asset
without considering the quality, and technical viability of the project
with the given environment. The beneficiaries were of the opinion that
this kind of problem was prominent in the case of small business,
diary-firm, goat rearing, sheep rearing and so on. In this process, in
order to disburse the loan amount in terms of cash, both the bank
officials and beneficiaries got into bargaining. Bargaining between the
lender and borrower brought the joint product called "bribe" by
neglecting the very objective of the scheme and the concerned project.
That obtaining a loan was a complicated and lengthy process
was also evident from Table 7.3. About 60 per cent of the loans took
more than two to eight months to get sanctioned in the case of
marginal farmers
63
. It is observed that the time taken for obtaining
formal short-term loan was inversely related to the size of landholding.
This finding also collaborated with the earlier study made by
Rajasekhar and Vyasulu (1990). There is empirical evidence of delay
in disbursing formal credit. Sarap (1990) has found that small and
marginal farmers also face substantial delay in getting formal (short-
term) credit. The empirical analysis of Chaudhuri (1993), in the case
of two selected villages of West Bengal pointed out that farmers with
OJ Around 80 per cent of these loans were for aUied agricultural activities. The
disbursement ofiong term loans is more time consuming (see, George et aI1985).
less than 1.5 acres of landholdings had to wait for nearly eight weeks
to get loans from the date of application. This problem has also been
observed by Bedbak (1986) who made a survey of a village of Sonepur
subdivision of Orissa.
Table 7.3: Time to get the Loan Sanctioned
-- the --c _ (iet .the LEan_Sanctioned _
Landholdings (in Acres), 0 - 15. 15 - 30 30 - 60 , 60 and Above Total Farmers

5.01-10.00 42.6, 51. I, 4.3 .1.o..l.. __ }Oo.. (47)
10.01 and Above 63. 9; 30. 6, 2.81---- __ 100.. [361.
,All Classes 29. 51 4l. 5\ 6 . .Qt.. ___ 2LQ' __ 100 J200)
Note: Figures in parentheses indicate the number of farmers.
Source: Field Survey
The other important aspect of the formal credit delivery system
was lhat the obtained ioan amounts were inadequate. We have
obtained information on amount demanded and the amount
sanctioned (short-term loans). The shortage of formal credit was
assumed to be the differential amount between loan demanded and
sanctioned. The distribution of loans on the basis of percentage of
amount shortage
64
to amount sanctioned is presented in Table 7.4. It
is observed that the amount sanctioned was less than the amount
demanded in the case of a majority of the farmers. However, only 44.8
per cent of the large size farmers obtained the loan amount of what
they had asked for. In the case of other categories of farmers the
amount sanctioned was well short of amount required. This suggests
that irrespective of the landholdings, the sanctioned loan amounts
were generally inadequate in the study area. It is also corroborated
with the finding of Table 7.1 wherein as many as 69.5 per cent of the
"' The shortage amount is a proxy for the inadequate availability of formal credit,
and may not be treated as the credit-gap. This is due to the fact that the
calculation of credit-gap may not represent the shortage amount without
considering his cost of production and self-finance capacity. The issue of credit-
gap has been discussed in the subsequent section.
farmer borrowers stated that the loan amounts were inadequate In
nature.
Table 7.4: Shortage of Amount Sanctioned to Amount
Demanded
,- SiZe-Class ofthe--,.---------P-e;centa-geofShortage------ ---- -
,LandholdingsJin Acres) I 1 - 25% I 26-=-- 50% - Total No. of Farmers 0%
!0.()1-2.50 46. 2 23. II 15.41- 100- (13)
2.51-5.00 ; 31. 3
10-"----1
32
1
34.4: 28. 11 6. 3
'5.01-10.00 10. 5 47. 4 21. Ii
21 1
110.01 and Above 44. 8 13.8
3101
10.3'
All classes 29. 5 32. I 25.0, 13. 4
Note. Figures m parenthesIs mdlcate the number of farmers.
Source: Field Survey
100
P8L
100 (29}
100 (l12}
This shows that the formal credit delivery system in the study
area was poor in qUality. It was neither adequate nor timely for most
of the agricultural borrowers. Thus, based on this information it could
be concluded that the Gupta Committee recommendations had hardly
improved the quality of credit delivery system in Kalahandi district of
Orissa. The poor quality of credit delivery system, along with some
other factors might have contributed to higher transaction costs. This
is further discussed below.
Borrowing Costs for Agricultural Credit
The effective price from the borrowers' point of view was the real cost
including interest payment and other transaction costs. To obtain a
loan the farmer had to satisfy certain terms and conditions that were
required by the lenders' credit delivery system, which resulted in the
borrower transaction costs. The borrower transaction costs in the
study included application fees, service fees, documentation and
paper works, bribe to officer/ technician/ bank staff/ leader or
middleman for negotiation, travel expenses incurred by the borrower,
and the opportunity cost of the borrower's time spent to obtain the
loan.
In the study area, it is found that many small and new
borrowers were required to visit the financial institutions at least five
to six times to maintain formality, withdraw a portion of the loan and
for repayment. The cost of time has become one of the important
aspects of borrowing costs, particularly when transactions are
contracted during the planting, weeding or harvesting seasons when
the opportunity costs of the borrower's time is high (Sarap 1990;
Adams and Vogel 1986; Adams and Nehman 1979). To avoid the
complexity of calculating the time cost of one day, it has been
considered as equivalent to one day of wage labour and calculated at
the wage rate prevailing during the process of acquisition and
repayment of the loan. This process has been adopted to calculate the
time cost of one day for the fact that the possibility of getting wage
employment is relatively high during these periods. Since the
agricultural term loans (particularly, for allied agricultural activity)
have been obtained by a small proportion of sample households in the
study area, and in many cases it has been given under the poverty
alleviation scheme, we are not considering the transaction costs
relating to term loans in the analysis.
Table 7.5 brings out that the average transaction cost (ATC)
decreased with increase in the size of holding. It is interesting to note
that the average transaction cost as percentage to average amount of
loan also decreased with respect to increasing size of landholding and
loan size. The effective rate of interest was much more for small and
marginal farmers compared to medium and large size farms. The
effective rate of interest for 100 rupees of loan was 20.55 per cent for
the marginal farmer, whereas, it was 14.22 per cent for the large size
farmer. This shows that the effective rate of interest was much higher
than that of the bank rate of interest for marginal and small farmers.
Table 7.5: Cost of Borrowing (ShortTerm) from the Financial
Institutions in the Study Area
(Amounts in Rupees)
Size of
Average Average Interest Total ATe as G/o Effective Rate
Landholding
Amount Transaction Charges Cost of of Average of Interest for
(in Acres'
of Loan Coat (ATe) for 12 Loans Amount of Rs.IOO of
Months Loan Loan
Up to 2.5 3,679 271 485 756 7. 37 20. 55
12.51 - 5.00 7,909 269 1,046 1,314 3. 40 16. 62
5.01- 10.0 11,752 234 1,542 1,776 1. 99 15. 12
[10.01 and
20,926 241 2,735 2,976 1. 15 14. 22
I Above
. Average 11,373 252 1,493 1,745 2. 22 15. 34
Source: Fleld Survey
The cost of borrowing of institutional credit by farm households
according to the size of t.l:!e loa..'1 in t..'1e study areas also reveals that
the ATe as percentage of average borrowing cost (ABC) kept on
decreasing with the increasing credit limit (Table 7.6). The ATe was
34.1 per cent within the credit limit of up to Rs. 5,000, whereas, it
drastically went down to 6.6 per cent for the credit limit of more than
Rs. 20,000. Alternatively, the average interest cost as percentage to
average borrowing cost kept on increasing with the increasing size of
the loan. The effective rate of interest was much more for the lower
size credit limit than the higher size of the loan. From Tables 7.5 and
7.6 it can be concluded that the effective rate of interest was inversely
related with the size of holding and the credit limit. Based on the
effective rate of interest one can find that the institutional credit is
costly for marginal and small farmers than the medium and large size
farmers. Thus, the higher transaction costs not only increases the
effective rate of interest but also deters small borrowers from
approaching formal institutions for credit (see, Adams and Nehman
(979).
'I'
Table 7.6: Cost of Borrowing of Institutional Credit by Farm
Households According to the Size of Loan in the Study Area
(Amount In Rupees)
I
Average Average ATe as % Average TC Per I
Loan Size
Size of Borrow- of Average Interest
Rs. I
Loan ing Cost Borrowing Cost as 100
Cost % or
\ or ABC Loan
Up to 5,000 3,690 722 34.1 65.9 6.7
5,001- 10,000 7,800 1,253 18.4 8l.6 3.0
10,001 - 15,000 13,792 1,956 10.6 89.4 l.5
15,001- 20,000 19,279 2,854 9.4 90.6 1.4
; 20,001 & Above 37,611 5,315 6.6 93.4 0.9
Note: ATC - Average Transaction Cost
Average Borrowing Cost = ATC + Average Interest Payment
Source: Field Survey
The Credit-Gap (CG)
Effective
Interest
Rate for
Rs. 100 of
Loan
19.6
16.1
14.2
14.8
14.1
As we have noted earlier, borrowers stated that the disbursed loan
amounts were inadequate. If subsidised formal credit is inadequate in
supply, a market for informal credit is automatically created
(Chaudhuri and Gupta 1996). Hence, this section is devoted to an
analysis of the extent of credit-gap among farm households across size
class of landholdings.
Table 7.7 demonstrates the level of farm dependency on credit
so as to meet the cost of production. Methodologically, this is based
on the Cost A2 process
65
(the method given by the Reports of the
Commission for Agricultural Costs and Prices) to estimate the cost of
cultivation. It can be seen that per acre paid-out cost shows a positive
association with the SIze of holding. Farm expenditure, and the
corresponding credit this necessitates, is of course depend upon
6S Cost A2 z All actual expenses in cash and kind incurred in the production by the
owner (Cost AI) + rent paid for leased-in land. Since the objective of this study is
not to highlight the cost of production, other methods which include imputed
value of family labour, interest on value of owned capital assets (excluding land).
rental value of owned land etc. have not been taken into account in the cost of
production.
labour and technical input requirements (the extent to which labour-
power employed is purchased, the use of fertiliser and pesticides,
extent of mechanisation, and the type of crop grown).
Table 7.7: Extent of Dependency on Credit Per Acre of Gross
Cropped Area (2001-021
Size of Holding Per Acre Paid- Level of Dependency on Credit
(in Acres) Out Costs (in Per cent)
(in Rs.)
Total
I
Formal
Advanced Region
0.01- 2.50 1,823 64.95 26.04
2.51 5.00 2,722 69.45
44.84
5.01 10.00 3,543 68.36 35.92
10.01 - Above 3,514 52.80 38.33
Backward Region
0.01- 2.50 1,662 58.66 26.19
2.51 - 5.00 2,161 74.47 I
54.41
S.Ul 10.00 2,242 64.01 38.60
~ 10.01 - Above 2,622 48.91 35.34
Source. Field survey
i
Not surprisingly, large size farmers depended less on credit to
finance their production expenditure. This finding was also true for
marginal farmers when compared with small farmers. It is interesting
to note that among all the categories, only small farmers were covering
a relatively higher proportion of their cost of production through
formal credit in both the regions of Kalahandi district. Although
medium and large size farmers obtained higher absolute amounts of
credit, the proportion of formal credit in their overall production cost
was small. In this connection, it must be emphasised that while in
absolute terms the flow of credit might indeed be positively associated
with the size of holding, it was inversely related as a proportion to the
per acre cost of production in the study area. It seems, bankers were
fixing a maximum credit limit irrespective of medium and large size
farmers' production costs and the credit requirement.
A close observation on the pattern of credit flow reveals that
these farmers in both the regions of Kalahandi district, managed to
cover between 26 to 54 per cent of their total expenditure on
agricultural production by means of formal sources of credit. The
foregoing analysis, therefore, suggests that the flow of credit were
insufficient to meet the production costs of peasant farmers engaged
in cultivation. To meet the remaining cost they turned either to self-
finance (capital generated from elsewhere in their total economic
operation or activity) or to borrowing from the informal market or both
or may be remaining satisfied with poor investment. In this context, it
is essential to measure the magnitude of credit-gap.
The credit-gap reJers to the unmet credit needs oj potential
borrowers Jrom either Jormal or iriformal or Jrom both the sectors with
their respective term and conditions. The demand for credit is assumed
to be positively related with area under operation and the requirement
of working capital. In our analysis it is also assumed that the
maximum demand for credit could be equal to the paid-out cost (Ct *)
given the self-finance capacity of the farmer. Thus, one can expect
that the obtained amount of credit (C
t
) S; C
t
*. If C
t
< C
t
* holds good,
then there is a possibility of credit-gap. Alternatively, given the self-
finance, when the required amount of credit for the project is not
disbursed to the concerned borrower by the above sources, then the
question of credit-gap comes.
In this section, the credit-gap has been analysed with respect to
size of landholding within and across the regions. The credit-gap per
acre of gross cropped area has been calculated in two different ways.
~ m c e the formal credit is not enough, farmer borrowers are depending
upon informal credit market. Thus, given the self-finance
66
and formal
credit the amount of informal credit taken per acre of gross cropped
area can be treated as the credit-gap. The calculation of credit-gap on
the basis of informal credit can be defined as the actual credit-gap. In
the case of second method, credit-gap for each category of farmers has
been calculated by taking the difference between the maximum paid-
out cost per acre of gross cropped area in the concerned region and
the actual paid-out cost. The second method seems to be more logical
in the sense that, ceteris paribus (viz., crop failure, pnces of
agricultural product, climatic condition, risk factors, and so on), all
categories of farmers would have been interested to invest to the
extent of maximum paid-out cost per acre in the respective regions
provided the required amount of credit would have been available to
them. The measurement of credit-gap on the basis of second method
can be stated as the relative credit-gap. The issue of relative credit-
gap seems to be more prominent in the case of farmers those who are
remaining satisfied with poor investment due to lack of availability of
credit from either or both the credit markets. The amount of credit-
gap per acre of gross cropped area has been presented in Table 7.8.
It can be seen from Table 7.8 that the actual credit-gap was
more for the medium size farmers in both the regions. This could be
attributed to the high cost of production with less self-finance capacity
as compared to large size farmers. For the rest of categories of farmers
the actual credit-gap was inversely related to the size of landholdings.
This could be mainly attributed to a progressive self-finance capacity
given the formal credit. It can be seen from Table 7.8 that the self-
finance capacity per acre of gross cropped area shows a positive
''', The self-finance capacity of a farmer can be defined as per acre paId-out cost of
production less per acre formal and informal credit
association with size classes of holding. But can we rely on this self-
finance capacity of the farm households. The distribution of self-
finance capacity is given below in Table 7.9.
Table 7.8: Landholding-wise Credit-Gap
i I I i
I PAr' A r' of I Per Acre of: Gap
I
I
I
I
s Gross
lin Acres)
er ere 0 ere 0
; Cropped Acre
I
Gross Gross
I
Area = Gros
Cropped Cropped
I
Actual Cropp ed Cropped I
Mea Mea I Credit-Gllp __ Are __ _______ _
n
_
3' 859 ' 822
7 - --1-, 086 - -0 -
-- ------ - - ----\.-- - - - - --
___ _ 39_

0.01 - 2.50 1823 475
,
70,) __ , __ 1,18
2.51 - 5.00 2,722 1,193 I 670 __
5.01 - 10.00 3,544 1,308 ,

10.01& above 3514 1,347 509 I ,85
All classes 3,289 1,249
I _2,(j0 5 1,284 255
Backward
10.01 - 2.S0 1662 435 540 97
2.51 - 5.00
2 161 1 176 434
I
1,6
'5.01 - 10.00 2,242 865 570 1,4
10.01 & above 2,622 927 356

IAlI classes 2,331 915 446 1,3
Note: Sample SIZe - 200, Ie = Informal credIt
Source: Field Survey
Table 7.9 explains that there was less variation in the self-finance
capacity at different levels for each category of farmers in the
advanced region. This implies that the self-finance capacity was
proportional to the cost of production. In the case of backward region
it seems to be progressive with respect to size of landholdings. Based
on this argument one can rely on the self-finance capacity of the
farmer as defined above while calculating the credit-gap.
Table 7.9: Self-finance Capacity of Farmers Per Acre of Gross
Cropped Area
Size class of"'-- ':- (1% 1 - 25 26 - 50 % Above 50 % Total Farmers
la."dilolciings _ -=-__ ___ __ Advanced Region
16.1 22.6 32.3' 29.0 100 (31)
- 28.6 j-19.1 .1- 28.6 100 (21)
5.01- 10.00 . 19.2-15.4 -_. 30.S -- 34.6 100 (26)
-. ;gg
___ _________ ---, __ gegion_
: i -_ ;gg -
5.01 - 10.00 I 0.0 __ 42.9 __ --1 __ 23.8 100 (21)
,10.Oland Above O.o=t __ __ 5_0.0 __ 1 _ 42 . .CJ -=-- r 100 (14)
All classes LI4.0_ . ...J4,-0--,---..lO<L ___ 12.() __ -'_ 100_ (100)
Note: Figures in parenthesis indicate the number of farm households.
Source: Field Survey
However, a large amount of relative credit-gap has been noticed
for marginal and small farmers in both the regions. Lenders' actions to
reduce default risk and transaction costs could be the main
contributing factors for a wide relative credit-gap in the case of
marginal and small farmers. Aryeetey (1996) argues that many small
potential borrowers had never actively sought formal credit was
attributed to the perception that bank credit was not available to
small farms. This perception suggests that the absence of supply
creates its own lack of demand. This might have led to low revealed
demand for credit and hence, higher credit gaps.
The Coping Strategies of Farmer Borrowers
Given the status of self-finance and inadequate availability of
institutional credit the farmer resorts to informal credit market.
Generally, they take loan from four different sources in informal sector
with different terms and conditions. They are moneylenders, large
farmers, traders, and friends and relatives. Since most of the farmers
stated the inadequacy nature of formal credIt, information was also
collected on the coping strategy of the farmer borrowers to meet their
requirement. The distribution of borrowers (with access to formal
loans) by number of sources of borrowing shows that about 70.5 per
cent of farmers went for two or more sources in the unorganised
sector to meet their cost of production (Table 7.10). This shows that
how far our formal financial systems were good enough to cover the
credit needs of farmer borrowers.
Table 7.10: Distribution of Borrowers by Number of Sources of
Borrowing
I Number of Sources I Marginal: maU_
1
- Large
___ .___ _ \"ith Access t()Form.al_Lo_ans
_ (in per cent)
All Farmers
___ 46.2 31.3 10.5 44.8 295
I. 2 46.-2-'-53.1--65. 8- - 51. '7 56 3
3 7.7 12.5 21.1 3.-;;4c--_12.5
4 and. Abc' .. e "" I ') 1 " C O. 0 .L. 0
r-_-_ -_"-"_ -::-._T:o_tal ___ __ -r
Without Access to Formal Loans
._- --'--- -_.-- . - --- ---
_____ 0 __ . . ___ .14.3 : __ 318
____ =1___ ---i----'3"'6:'.:.-=-4 52. 9 22. 2 14. 3 36 4
.2.= 7L 4" _l-=--'21 7
3 and Ab ... ".6---,,_..:""5. 9 . 55. 6 o. _0. __ 9. L
Total : 100 ,100 100 100 I 100
!_ i (17) l2.l ____ (7) ____ '_J88) __
Notes: (I) Figures in parenthesis indicate the number of farmers.
(2) This table explains the number of sourceS of borrowing by farmers to meet
cost of production in the year 2001-02.
Source: Field Survey
Table 7.10 also reveals that among the farmers, who had not
availed any formal credit, 68 per cent of them were completely
depending upon informal sector. Without having access to formal
loans almost all small and medium size farmers and around 85 per
cent of large size farmers were taking informal loans to meet the cost
of production. Among the 55 marginal farmers who had not obtained
credit from institutional sources, surprisingly, 47 per cent of them
had not even taken any informal credit. It is important here to note
that it was not that the marginal farmers did not require credit. but
107
their access to informal credit market was also hi"hl\' rl'strlctcci'
,., .
This market operates on the basis of personalised relationship. \\'hil'!J
means that anyone who is prepared to pay the interest ratl' and meet
the collateral requirements, IS not likelY to reeel\T a loan
automatically from all lenders (see Basu 1997). M a n ~ ' risk-averse
lenders might not pmvide credit to certain borrowers, or use ercdll
limit, due to the problem of moral hazard and ad\'erse selection (see
Stiglitz and Weiss, 1981).
Therefore, given the poor access 10
institutional credit and self-finance capacity of small and marginal
farmers, the restricted access to informal credit market also contributed
to a larger amount of relative credit-gap. Thus, to understand the
coping strategy, an analysis on the terms and conditions in informal
credIt market IS needed.
Type of Contracts in Informal Credit Market
In a rural community, a farmer can mobilise the amount of working
capital by offering collateral, collateral substitutes, personal security
and personalised relationship, and through third party guarantor. An
important issue in this regard is that credit is invariably rationed
according to the ability of the borrowers to offer collateral (Sarap
1991). The types of contracts in the informal credit market under
different forms of collateral security are further discussed below.
(a) Contract with Collateral
The lender's preference for an asset, as collateral. depends on
the effective rate of return he expects from the sale value of assets. in
case of default of loan. Thus, higher the market value of a cullatl'f,d
asset, higher will be loan amount to a borrower. Similarl\. higher the
" This finding also collaborates \nth the earlier study mane by HaSLl I; (fll!!
lOR
assets position of the borrowers (which tentatively represent his credit
worthiness), higher would be the amount of loan he can arrange (sec,
Sarap 1991). In a loan transaction the requirement of collateral arises
since it shifts the potential capital loss from the lender to the
borrowers (Binswanger and Rosenzweig 1986).
(bl Contract with Collateral Substitutes
In the absence of marketable collateral, the borrower uses
collateral substitutes or inferior collateral to mobilise the loans. In
such deals, the lenders provide credit to the borrower against his
commitment of future sale of labour service or crops to him. For this
kind of contract, the borrower provides his labour services at the call
of the iender at an agreed wage rate or sell his crops to him at a pre-
determined price in a mutually agreed time to repay the loan (Bard han
and Rudra 1978). Hence, in the case of without collateral, the farmer
borrower can obtain credit from the interlocked credit market.
However, different forms of welfare loss of the borrower is
inevitable in the interlocked credit contract since it allows the lender
to extract full surplus attributable to the inter-locked market
(Bardhan 1984 and 1989; Gangopadhyay and Sengupta 1987;
Gangopadhyay 1997). This extract in inter-locked credit market can
be noticed usually in the form of under valuation of labour services
and output of the borrowers and overvaluation of inputs supplied by
the lenders. Thus, given the worse access to organised credit market
and self-finance capacity, if the demand for additional liquidity to
meet the working capital is relatively inelastic in nature, the borro\\'ers
may stand at the receiving end of the bargaining process in the inter-
locked credit market. Alternatively, the terms and conditions become
favourable to the lender. Therefore, the degree of extraction Illa\' \'an
from borrO\\'er to borrower depending upon the bargaining cap'tcit\
between borrowers and lenders. The pre-determined exchange' rail' in
the inter-locked credit contract and the extent of extraction has hcen
discussed in the subsequent sub-section.
c) With Personal Security and/or Third Partv Guarantor
In the absence of acceptable collateral by the lender or
interlinked deal, the borrowers mobilise credit through personal
security or third party guarantor. In the former case, the lender may
provide loans to those borrowers from whom he is sure of repaymeI1l
or able to recover the loan through some implicit or explicit
mcchanism
68
. Elsc:w-lLere, it. IS argued that loan contract \\'ithout
collateral increased with the raise in the status
69
of the borrowers
(Sarap 1991). In such a case, the preference of the lender was based
on caste, kinship connection, and the localities of the borrowers with
which the lender may have some ties (see Bardhan and Rudra 1978).
These connections and information about the borrowers worked as a
system of guarantee for the repayment of the loans. The latter case.
the third party acted as a guarantor, exerted pressure/ influenced the
borrower for repayment of the loans. Thus, the third p a r t ~ worked as
peer monitor
70
in borrower selection as well as for the repayment of
the loan.
" This mechanism could be social sanctIOn against the defaulter or loss of luturc
borro\\'ing from the same or other lcnders. For more dctatled discuSSiOll. see,
Platteau and Abrahem (1987); Basu (1986,1989); Sarap 11991).
, This status may be In the form of landholding, assets pOSItion. and other SlH'J()-
economic status.
See, Stiglitz 11990),
)()()
Interest Rate for Different forms of Contracts in Informal Credit
Market
The interest rate in the informal credit market depends upon the kll1ci
of contracts with or without collateral. Interestingly, the rdlt' "I'
interest also varies across the quality and marketability of the
collateraL In the case of loan contract with collateral, the kndcr
generally aseertams the sale pnce of the piece of collateral assets
before providing a loan. If a particular piece of collateral asset imohes
higher transaction costs while selling, then the interest rates on loans
taken by pledging such collateral were more. For instance, tIl(' cost
involved in selling for an asset like old bicycle or utensils is very high
than gold. Therefore, the asset, which has either low resale value or
involves high transaction costs in selling, the lender either prondes a
very small amount of loan with higher interest rate or may not accept
the asset as security. Contrary to this, if the lender expects a high
resale value of an asset (for example land or gold) in future, then he
may be willing to provide a higher size of loan. Thus, for poor quality
of collateral assets the terms and conditions are more harsh to
borrowers in terms of rate of interest charged, under valuation of
assets and duration of the loan as compared to borrowing against gold
or land (see Basu 1989; Sarap 1990, 1991). For poor quality of
collateral (say brass utensils) the interest rate vary from 60 per cent to
120 per cent, whereas for good quality of collateral like gold the
interest rate vary from 24 per cent to 48 per cent per annum.
Generally, the loan contract with collateral substitutes
(interlocked credit transaction) does not involve any interest ratc.
However, under valuation of labour service and output or the
borrowers and overvaluation of inputs supplied bv the lenders lS the
201
common practice in this kind of credit transaction. The asyml11ctn' of
information, moral hazard and adverse selection problems along \\'ith
the degree of risk involved in lending (for instance, crop failure in the
case of credit-output interlocked transaction) insists the lender to fix
up a low price for collateral substitutes. Hence, it is the rate of return
through low price for borrower's output in the interlocked credit
transaction compensates the interest income loss that would have
been received by the lender in alternative investment. Given this,
credit (either cash or kind) can be interlocked \\'ith either labour or
product markets where prices are predetermined.
The interest rate on loan transaction against personal security
kind), volume of credit, the personal relationship with the lender, and
the socio-economic status of the borrower. In this system, the
borrowers repay the loans ll1 terms of either cash or kind at the
interest rates, which was frxed at the time of taking loan. For kind
repayment, both the parties jointly determine the prices of the
products at the time of repayment, given the present market situation
of the product and bargaining capacity of both the parties. It has been
observed in the study areas that the lender charges a relati\'Ch' low
interest rate for the kind component of loans
7
) (say fertiliser or
pesticides) than the cash component. Given the credil\\'orthiness of
the borrower, the high competition among the fertiliser and pesticide
traders in selling can be attributed to the low interest rate charged for
kind component of loan. The other type of kind component loan like
paddy, the interest rate charged is 50 per cent of the principal. The
duration of loan of this type is generally six months or less. Bor[i)wing
"I For instance, farmers take fertiliser and pestlcicies as credit at some 1Il11";-CS' r ; l ~ ( '
from the shopkeeper, and they repa,\" it after haryest.
202
of paddy loan decreases with the rise in the status of the households
(see, Sarap 1991). Since money has wide scope for in\,cstmt:nt, the
lender charges a high interest rate for the loan borrowed in cash. The
farmer borrowers in the study areas stated that the interest raIl' on
kind component of loan varied from 48 per cent to 60 per cent,
whereas it varied from 48 to 120 per cent for cash component of
loans.
Thus, under the type of contracts agreed upon by the borrowers
and lenders the degree of variability in the level of interest rate can be
as high as 120 per cent or as low as zero per cent per annum (Rudra
1982). Bhaduri (1977) argues that usurious interest rates are the
result of monopoly power of the lender. Bottomley (1975) mentions
that high interest rate reflects the risk of default faced by the lender in
the rural credit market of the less developed countries. The other
studies
72
claim that the presence of interlocked contracts can explain
the existence of a wide spectrum of interest rates. On the whole, it is
the kind of contracts with or without collateral that determine the
interest rate in the informal credit market.
With this background, it may be useful to analyse the degree of
extraction, the number of loan contracts and the amount of loan
borrowed against different form of collateral and type of contracts in
informal credit market being undertaken as coping strategy to meet
the cost of production.
-2 For more detailed discussion, see, Brn.\'cman ann Srinl\"(lSan 119811, \11t1"d Ill)S?,I.
and Basu (1983).
203
I
Borrowing Against Different Collateral as Coping Strategy
In the study area, interlocked credit and product transaction \\'<lS a
common phenomenon, where farmer borrowers used to sell their
product after harvest at the pre-determined price. Since paddy IS the
major crop (it covers almost 100 per cent and 61 per cent of gross
cropped area in the advanced and back\\'ard regIOns respccti\'('I\I.
credit against future production of paddy was practised.
The agricultural borrowers said that in the case of input-output
interlocked credit transaction the exchange rate was fixed irrespccti\'C
of the size of landholdings. However, in the credit (cash)-output
interlocked transaction the exchange rate differed with different
borrowers depending upon the bargaining strength between the
borrower and the lender. The pre-determined exchange rate in the
interlocked credit market and approximate distress amount per bag of
paddy has been explained in Table 7.11. It can be seen from Table
7.11 that the distress amount per bag of paddy selling in interlocked
credit market varied from Rs.I00 to Rs.155 depending upon the type
of collateral offered to the lender.
Table 7.11: The Predetermined Exchange Rate of Paddy in
Interlocked Credit Market
Lenders Set Exchange Rate Type of
I
Collateral
I
Ap
(Cash/Kind)
Interlocked!
Given
I
Market I
Am
Rice mill Advance Rs. 230 - 280 I Credit I Collateral
proximate - I
Distress
ount in Rs.
, owner, grain bag paddy (cash)-
I
substitutes 15
I traders, big
o to 100,-
I
I
I
output
farmers
,
Shopkeepers I bag Urea' = 1 bag paddy' ! Input-
- do -
oU!put
1 bag Potash'= 1 bag padd\'
- do - . do -
._--
1 bag super' 1 bag
- do - - do -
I paddy
1 bag DAP' = 2 bag paddy' - do - - do -
)iate: The approximate distress amount per bag (If selling tw.s been c;L\cULl\(,(1 till'
difference between MSP and rcspcctlvc fertiliser pncc. The :-'1SP per bag of pddcl\'
IS Rs. One bag of fcrtillser (O:;'SlstS of SU
;\'amc of fcrultscr
Flcld Survcy
However, the details of income loss due to interlocked credit
have been shown in Table 7.12. This has been estimated by obtainl11g
the quantity sold at interlocked price as against the hypothetical
scenario of selling the same quantity at the open market and
minimum support price (MSP). The second and third column of Table
7.12 shows that both interlocked and the open market price
73
differed
across the farmers with different size of landholdings. It is seen that
the medium and large size farmers had better bargain in fIXing not just
the interlocked price but also in obtaining a favourable price in the open
market. Poor bargaining capacity due to high intensity of credit
compelled the marginal and small farmers to sell at low price in the
interlocked credit market.
Table 7.12. Regions & Landholding-wise Estimated Income
Loss Per Bag of Selling Paddy due to Interlocked Credit
Market in Kalahandi District, Orissa
2001-02\
Size of
Interlocked Open
LPJ LP, , Income Loss Income Loss I
Landholding
Price Market in Rs. in Rs. in Situation in Situation I
(in Acres)
In Ra/bac Price (A)
fBI
A B
in Ratbag _lin %) iin%l
Advanced R<lJ! ion
J
0.01- 2.50 252.07 283.48 31.41 127.93 12.46 50.75
i 2.51 - 5.00 246.27 308.08 61.81 133.73 2510 54.30 :
15.01 - 10.00 257.91 320.05 62.14 122.09 24.09 47.34 I
10.01& Above 264.35 325.18 60.83 115.65 23.01 43.75 I
Backward ReRion
; 0.01- 2.50 250.65 264.92 14.27 129.35 5.69 51.61 'I
12.51- 5.00 251.23 274.61 23.38 128.77 9.31 51.261
. 5.01 - 10.00 252.80 280.39 27.59 127.20 10.91 50.32
10.01-Above 270.0 298.91 28.91 110.00 10.71 40.7 -t
Source: FIeld survey
Note: MSP = Rs. 380 I bag (l bag = 75 Kgs)
A = Difference between Open Market Price and interlocked price per bag of padd\'
B = Difference between MSP and Interlocked price per bag of paddy
Although medium and large size farmers, who are the main
sellers of paddy, also face some loss of income due to the interlocked
;, The actual price received by the fanner in the market without mterlocked
transaction.
credit market - principally when substantial amounts of this
commodity (acquired from smallholders at pre-determined low prices)
are marketed simultaneously - it is less than that of marginal and
small farmers. Table 7.12 confirms both that medium and large
farmers receive a higher price for their paddy in the open market than
do marginal and small producers, and also that the difference between
the interlocked price and the minimum support price for this crop
declines with respect to landholding size. The magnitude of income
loss, however, depends on the extent of dependency on informal credit
sources to meet working capital and also on the quantity of paddy
sold at the pre-determined price. What this underlines is that the
supply of formal credit is itself inadequate to meet the needs of
peasant farmers in both regions of Kalahandi district. Thus, given the
self-finance, adequate availability of formal credit could have
minimised the income loss of the farmers who had taken loan from
interlocked credit market.
It is found that the informal credit with collateral substitutes
was not the only way to meet the cost of production. The farmers had
also mobilised credit through other collateral such as land, gold, brass
utensils, output, personal security, and third party guarantor. Loan
transactions against these collaterals have been shown in Table 7.13.
Table 7.13 shows that 37 per cent of the loans and 35 per cent
of total loan amount was borrowed by small farmers by keeping farm
output as collateral. In such cases of inter-linked markets, the rate of
extraction was more. Against output, the marginal size farmers \\Tre
having 19 per cent of their total loan contracts consisting of 1 P, per
cent of the total loan borrowed. Probably, low base of output
production \\as inhibiting the marginal iarmers to approach ttw
lenders against output as collateral. These percentages dccreased
progressively for the medium and large size farmers as compared to
small farmers since they could mobilise informal credit \,ith
alternative contracts at low interest rate. Use of gold as collateral rose
with the increase in the status of borrowers. Poorer farm households
utilised brass utensils as collateral in the absence of other alternati\e
assets, where under valuation of assets and rate of interests were
high. The marginal farmers were having 21 of their total loan
contracts against brass utensils. Borrowing (cash/ kind) against
without collateral, where the interest rates were high, declined with
the rise in the status of farm households. The decline was more
pronounced in the case of large farmers possible due to their own
surplus and their better access to formal credit (see Sarap 1990).
Table 7.13: Farm Size and Collateral-wise Percentage of Loan
Contracts and Percentae:e of Loan (amountl Borrowed
Type of No. of Loan Contracts (%) Amount of Loan (%
Collateral
Marginal Small Medium Large Marginal Small Medium Large
Offered
Farmers Farmers Farmers Farmers Farmers Farmers Farmers FaTmers
Land -
1.5 -
- 1.2
Gold 10.0 26.3 37.9 48.3 134 33.8 41.7 46.2
Brass utensils 21.4 1.8
.
10 1 1.3 - -
Output 18.6 36.8 31.8 31.0 17.7 34.8 29.6 27.1
Sub-total 50.0 64.9 71.2 79.3 41.2 69.9 72.6 73.3
Personal 30.0 33.3 28.8 20.7 37.9 29.0 27.4 26. I
Security lal
Third party 20.0 1.8
-
- 20.9 1.1 -
Iguarantor -Ibl
Without colla- 50.0 35. I 28.8 20.7 588 30.1 27.4 26.7
teral (a+ bl
Total 100.0 100.0 100.0 100.0 100.0 100.0 100.0
](1(l
I
170) 157) (66) (29) 63,340" 110.700' 320.800' 234,05'.-i
T
:
Note: I. F,gures in parentheses represent the total number of loan contracts b\ thf'
respectl\'e size of farms.
2. 'represents the total amount of informal credit by the respectl\'e
groups.
3. (-) means that there was no in\"oh"ement of a size group of farm
for borrowing against a particular collaterai.
Source: Field Surw\
Besides, this category of farmers was having alternali\'(' assets to
mobilise loans from informal sector at cheaper rate. The percentage oj"
loan lamount; borrowed against personal seeuntv and third part\"
guarantor for marginal and small farmers were 59 and 30 per cent
respectively.
On the whole, the composition of borrowing against different
collaterates provided some characteristics of different borrowers in the
sample. First, it shows that the quality of collateral increased with the
rise in the status of borrowers. Second, in the absence of alternative
assets the small and marginal farmers were compelled to mobilise
funds at higher rate of interest from informal credit market.
Conclusions
The analysis on credit delivery system shows that most of the
agricultural borrowers faced one or the other problem during as well
as after obtaining loans. It is also observed that these problems were
more prominent in the case of marginal and small farmers. The
analysis on time taken to get a loan sanctioned shows an inverse
relationship with the size of landholdings. Added to that, the loan
amounts were inadequate. This indicates the strong practice of non-
price credit rationing in the provision of agricultural credit and in
fixing the credit limit to agricultural borrowers. Since the formal credit
delivery system in the study area was found to be neither adequate
nor timely, it could be argued that the Gupta Committee
recommendations have not been successful in improving the qualIty of
credit delivery system in Kalahandi district of Orissa.
' ) r L ~
The analysis on borrO\\'ing costs for agricultural credit s!Jo\\'s
that the effectl\'e rate of interest \\'as related \nth the size u1
landholdings and the credit limit. Based on the effective rate of
interest one could find that the institutional credit \\'as cost!\' fur
marginal and small farmers than the medium and large size farmers,
The terms and conditions of loan obtained against different
collateral assets indicate that the terms of getting loan became more
unfavourable with the decline in the quality of collateral assets
provided by the borrowers, Given the nature of informal credit market
and the effective mechanism of controls on borrowers, the lender was
in a position to extract as much as possible through higher interest
rate charges and other terms and conditlOns of the loan,
An ob\'ious policy implication that can be drawn from this
analysis is that any move by the government to reduce imperfection in
the credit market will benefit the farmer. In order to reduce the credit
gap problem, a rethinking on the scale of finance is needed,
Nevertheless, a separate scale of finance should be fixed for each
service area of the concerned bank since the cost of production vanes
across the regions, Improved quality of credit delivery system with
adequate loan amount can reduce farmers' dependence on informal
sector to meet the cost of production, and in turn, the degree of
extraction in the form of interest and others especially from small and
marginal farmers,
Chapter 8
SUMMARY, CONCLUSIONS AND POLICY IMPLICATIONS
Introduction
A majority of the peasant cultivators 111 India find borrO\\ing
necessary, as their own farm savings are inadequate for both
financing their various agricultural activities, and in carrying out
improvements in productivity of their agricultural operations (Lipton
1976; Donald 1976; Sarap 1990). The provision of formal credit to
agriculture has been, therefore, central to the concern of planners and
development economists 111 post-independent India. A number of
policies have been made to instil social responsibility 111 the formal
financial institutions to extend credit to rural households, in general.
and farm households, in particular. These policies have succeeded in
establishing a vast network of financial institutions 111 the
countryside, and helped to spread the banking culture among in rural
areas. But, quantitative progress at the cost of quality (Shivamaggi
2000), high cost structure of operations in rural areas and mounting
overdues became a threat to the viability of the financial institutions.
Consequently, the policy of competitive financial system was adopted
from 1991 onwards in the provision of agricultural credit to impro\,<,'
the viability of financial agencies. This policy may have direct or
indirect influence on the credit flow to agriculture. The Gupta
Committee, appointed in 1998 to identify the constraints faced b\ the
commercial banks in increasing the flow of credit, recommended the
introduction of new products and services and simplification of
procedures to enable the rural borrowers to access adequate and
timely agricultural credit. These recommendations m,\\ have
)I()
the access of farmers, especially sll1<lll and marginal on('s. to f<lrm,d
credit.
The present stUdy sought to anal\'se the trends in the credll
Oo\\' to agriculture and contributing factors during the period 1 m O - ~ 1
to 1999-2000. This period has been divided into t\\'o sub-periods.
namely, pre-reform period (1980-81 to 1990-91) and reform penod
(1991-92 to 1999-2000). The anaiysls was earned out at all-India
level, in 14 major states and in the state of Orissa. The state of Orissa
has been selected because this is one of the most backward districts
in the country. Kalahandi district from Orissa was selected for
undertaking bank-branch and household level analvsis because this
district is known for widespread poverty and backwardness in India.
The analysis at all-India, state and district level was based on data
published in the official documents. On the other hand, the analvsis
at the bank-branch and household levels \\'as based on the pnman'
data. The data on credit Oow was collected from four bank branches
for five years beginning with 1996-97. These bank branches were
located in the advanced (in terms of irrigation, input use, productivltv
and infra-structure development) and backward blocks in Kalahandi
district. The data on different aspects of rural CTedit market \\'ere
collected from 200 households coming under the sen'ice area of the
four bank branches.
Summary
The results of the stud\' have been presented in Chapters III to VII.
The ensuing paragraphs prO\'ide summaries of these chapters.
)]1
Trends in the Credit Flow at AII-Illdia Leliel alld COlltributing Factors
The third chapter, which analysed the trends in the suppl\ of
agricultural credit by SCBs at the all India level, revealed that the
proportion of credit to agriculture to the net bank credit rapidly
declined during the reform period. The growth rates of agricultuml
credit were uneven across the sub-periods (pre- and reform years) as
well as across the bank groups. The supply of credit to this sector in
both absolute and real terms grew at much faster rate during the pre-
reform period as compared to the reform period. Across the bank
groups also, a similar pattern could be observed. Further, the
Scheduled Commercial Banks (SCBs) did not even reach the targets
set by the government.
As far as the factors determining the credit flow to agriculture
were concerned, the regression analysis showed that credit flO\\, to
agriculture at all-India level was adversely affected by investment ll1
securities, credit subsidy, agricultural credit provided by co-operati\'es
and interest rate on lending, However, credit supply to agriculture \\'as
positively associated with number of rural bank branches,
The better-off farmers had better access to formal credit as
compared to small and marginal farmers, Importantly, despite the
increasing consensus to extend credit facility to agriculture, in
general, and small and marginal farmers, in particular, banks were
unable to lend to those activities with high social return or those
category of creditworthy borrowers who had been traditionall\'
marginalised in the credit market.
Trends and Determinants of Credit FlolU to AgnclIilllrc ur 3lale and
District Level
An analysis of trends in the supply of agricultural crCdlt \),. SCBs al
the state and district level has been carried out in the fourth chapIn
The rate at which credit to agriculture was disbursed gr-cw at a rapid
rate during the pre-reform period as compared to the reform pniod in
most of the states. These growth rates were found to be unevcn both
during the sub-periods and across the states. An anal\sis of l\m
indicators, viz., the proportion of agricultural credit to net bank cred,t.
and total bank deposits, revealed that there were substantial inter-
state differences in these indicators within and across the su b-
periods. A decline in the proportion of agricultural credit to total bank
deposits in all the major states implies that the banks were diverting
their funds raised in the form of deposits to sectors other than
agriculture. This is found to be more prominent in the back"'ard
states of Bihar, Uttar Pradesh and West Bengal.
An analysis on access to credit (i.e., the number of farmers
obtaining formal credit to total number of farmers at a particular point
of time) showed that there were substantial inter-state differences in
the access to formal credit by the same category of farmers at different
points of time. While the percentages of farmers accessing formal
credit declined during the study period in most of the sta tcs, the loan
amount per account increased. This implies that there was
concentration of formal loans in fa\'our of a particular group of
borrowers, and that the banks practised both price and non-pnce
credit rationing methods rn the borro\\'er selection and fixdlIon of the
credit limit. The non-prrce rationing bccomcs c\'<'n s\ rtlnL',n
implication if one observes Ihe f8('\ th81 farmers \\'ithin \)](' ~ " n ] ( ' SlZt
class of landholding obtained substantialh difkrent loan amounts
across the states.
It is observed that most of the findings at the district ltTcI
converged with the state level analysis. The notable similarities in the
findings relating to the state and district levels analysis were the
declining share of agricultural credit from net bank credit. uneven
credit flow across regions and the differences in the availability of
credit per hectare of gross cropped area. This suggests that the flow oi
agricultural credit followed similar patterns at both state and district
levels.
Higher (or lower) the proportion of net bank credit to the
agriculture sector does not necessarily mean that the per capita credit
availability per hectare of gross cropped area was more (or less) in a
particular state or district. Since the proportion of net bank credit
does not take the gross cropped area into account, it might be difficult
to draw a conclusion on whether or not a particular state or district
was better off in terms of the availability of agricultural credit.
Therefore, credit (obtained from institutional agencies) per hectare of
gross cropped area was used as an indicator to measure the variations
in the supply of agricultural credit in the state and district level
analvsis. Variables such as area under irrigation, area under
commercial crops to gross cropped area, Credit Deposit Ratio (CDR)
and density of bank branches per 1,000 farmers were specified as
determinants of credit per hectare of gross cropped area for the
regression analysis at both state and district levels.
Since different states and districts have different ch"r;ll'lt',']stJ('s
111 terms of irrigation facilit\', area under commercial em!,,,. lTcdil
deposit ratio and density of bank branches, panel regression \IdS Liscci
to assess the separate effect of these variables, Based on the
Langrange Multiplier statistics and Hausman specification lest, the
Fixed Effect Model over Random Effect Model was preferred "t Stall'
level analysis, On the other hand. Random Effect Model ,,'as useo in
the district level analysis,
The proportion of irrigated area to gross cropped area and credit
deposit ratio were found to be the most signifIcant explanatory
variables for the variations in the credit flow per hectare of gross
cropped area in 14 major states as well as districts within Orissa,
While the relationship between the credit per hectare of gross cropped
area and the former was positive, that with the latter \\'as negau\'l',
Since assured irrigation reduced crop uncertainty, bankers were
probably giving priority to lend more to farmers with irngateo
agriculture, Surprisingly, even when CDR increased, it reduccd the
supply of credit. The negative sign of CDR could be attributed to
perception among bankers that loans to agriculture charactt'rised
high-risk. The density of bank branches positivel" affected the credit
flow per hectare of gross cropped area at the state level. Although, this
variable had positive association at the district level, it \\'as not
significant. The coeffIcient of area under commercial crops to gross
cropped area was positive but not signifIcant at the state len'!.
However, this variable was negative and highl\' significant ,,\ t 111'
district leve!. On the whole, the state and district InTI "',,,1\sls
suggest that for improved flo\\' of credit to agriculture, tl1<' policies
aimed at Skppll1g up of the CDR should be backed \)\ (hilS,' OJ)
improving irrigation and banking infrastructure.
Operation of Macro Policy at the Micro Level: An AIlCilllSIS of Crcdlf Flou:
in the Sample Bank Branches
After 1991. the scope of the pnonty sector was broaclcned to lI1dudc
activities, which were traditionally excluded from the pun'ie\\ of this
sector. The fifth chapter has concentrated on an analysis of the
impact of broadening the coverage of priority sectors on credit flo\\" to
agriculture at the bank branch level in developed and backward
blocks in Kalahandi district, Orissa. Based on the data for tlw last fin'
years ending with 2000-01, the following results haye been arnved: (il
The share of priority sector from total bank advance had 1l1creased 111
both the regions, whereas the share of credit flow to agriculture had
declined during the same period. (ii) The bankers were gIving
preference to non-agricultural activities especially in the advanced
region. Although the volume of disbursement of agricultural credit by
the sample bank branches in the advanced region \vas more, rdatively
greater share of credit to agriculture from total advances was noticed
for the sample bank branches of the backward region. The interest
ll1come on total advance and type of securitY-\\'ise supply of
agricultural credit were calculated for both the regIOns for
corroborating the above. It was found that the interest income on
total advances and the proportion of credit going to agnculture out of
total loan were inversely related in both the regions. Further. the 110\\
of agricultural credit \\'as largely governed by the collateral sccurlt\
(either land or gold). Not larger the proportion of total
advance for agricultural credit. more prominent \\'8S the colla tcral
based lending.
These two findings had an impact on credit 10 the pn()rlt\
sector. in general, and agriculture, in partlcular. as a proportl()11 ot
(olal bank deposits. In the backward reglLln. the pmportllln ot pnurll \
sector to lOtal deposits declined from 81.56 per cent 1I1 I q'Jl-l.ql) t(J
72.35 per cent in 2000-01. In contrast, the same in the advanced
region increased from 85.55 per cent to 116.70 per cent during the
same period. Agricultural credit as proportion to total deposits
declined from 27.04 per cent in 1998-99 to 23.15 pel' cent in 2000-01
for the advanced region. A marginal Improvement of the same over the
initial 39.10 per cent was observed Il1 the backward region during the
same period. The policy of collateral based lending and expc'cted
interest income on advances could be attributed to these situations.
Factors Detennining the Access to Institutional Credit
Empirical studies reveal that the beneficiaries of rural credit extension
policies and programmes were the rural well-to-do and not poor
farmers (Rao 1975; Lipton 1976; Adams and Vogel 1986). Uther
studies confirm this trend. These studies suggest that only a small
proportion of the total cultivators in rural areas obtain loans from
formal financial institutions. Further, a small group among those
with access to formal credit monopolIsed the lIon's share of the net
volume of credit disbursed (Lele 1981 ; Gonzalez-Vega 1984; Bra\'Cman
and Guasch 1986; Egger 1986). Sarap (1990) argues that a large gap
existed between, on the one hand, the overall supply of credit to small
and marginal farmers, and on the other, their need for thIS.
Factors determining the access to institutional credit b\' tilt'
farm household were analysed in the SIxth chapter. As dIscussed. the
credit transaction depended not just on the need of the' f ~ ' r m ' T t()r "
loan, and the related need to meet the conditions gOH'fning 1 his
borrowing, but also the lender's willingness to extend credit. In l his
context, farmers' access to institutional credit was estimated b, using
the Probit Model where the dependent variable was a dichotomous
variable indicating whether a particular farm household had access to
credit or not. The assumption made by the Probit Model \\as that the
probability of borrowing from the formal sector sources was
determined completely by the bank which decided whether a farm
household would get a loan or not. This model also assumed that all
cultivating households demanded institutional credit at the existing
interest rate, and that the formal sector was the cheapest source of
credit for farmer borrowers.
The analysis at the household level shows that the farm and
household characteristics like quality of land, value of non-land
assets, irrigated area and educational qualifications of the borrowers
improve the prospects of a farm household's access to formal loan,
and hence, borrowing, The results also show that the actual size of
landholding did not significantly influence the access, This finding
contradicts the findings of some of the earlier studies that larger the
size of land owned by the households, greater was the probabilitv of
its access to the formal sector (Kochar 1997; Swain 2002), The lower
caste farmers (SCjST), and a farmer who had higher non-farm income
as proportion to the total income were less likely to hm'e access to
credit, as is evident from the negative coefficient of these \'ariablcs,
It is also evident from the sample of farm households of
Kalahandi district that even where farmers had access to formal
credit, the amount obtained \\as inadequate for their f(:qll1rements.
This necessitated borrowing from informal sources. This e\'ic!encl'
supports the argument that the demand for agricultural credit fm
exceeded its supply, In this context, it is discussed in chapter VI that
in a financial situation characterised by the absence of C1'edit
rationing, the factors influencing the demand for credit would
obviously determine the quantum of credit supply, It is also argued
that where credit rationing (either in the form of price or non-price)
existed, the factors influencing the demand for credit could not be
considered as the ultimate determinants of its supply. It is, therefore,
to identify which of these factors - demand or supply, accounted for
the flow of institutional credit, the determinants were estimated.
Since the formal lenders gave importance to collateral security
and repayment capacity of the borrower, area under operation (OA)
and value of farm output (VAO) were considered as suppb;-side
factors. The percentage of irrigated area to gross cropped area
(PIAGCA), expenditure on farm inputs like fertiliser and pesticides
(INEOFA), educational qualification (EDU) and caste of the borrower
(CASTE) and the ratio of workers to family members (WFM) were
specified as demand-side factors. As elaborated in this chapter, access
to formal credit in the case of certain farming was not random. In
order to correct this problem of selectivity bias, the Inverse Mills Ratio
(IMR) was calculated from the above Probit equation and the same
was introduced as an additional explanatory variable in the equation
on the determinants of credit. The results show that PIAGCA. SOU,
WFM, EXPFP, INEOFA and VAO positively influenced the supph' of
agricultural credit, whereas OA and CASTE adversely affected it.
Logit and Probit Models are useful if the qucstion is Wh"1 her or
not there is access to credit. But they arc inappropriate \\'hen 11 is
important to measure the amount of credit obtained. The Tobil i\lllckl
handles censored dependent variables (continuous between some
lower and possibly upper bound) better and, hence, is supnior to
Logit and Probit. The Tobit results again confirm that PIAC'CA. VAO.
EXPFP, and CASTE were important variables in determining the
amount of credit obtained.
Credit Gap and Coping Strategies
Several factors were, thus, inhibiting agricultural borrO\\'ers, in
general, and marginal and small farmers in particular, from obtaining
production credit. It is observed that the credIt rationing went in
favour of large size farmers. Interestingly, not only a small fractIOn of
total number of farmers had access to formal credit, but some of them
also reported that they faced many problems while obtaining the
credit. Some of these problems also contributed to the higher
transaction costs and delay in the disbursement of credit. In this
context, the quality of credit delivery system, the extent of credit-gap,
and the coping strategy adopted by the farmers in the study area were
analysed in the seventh chapter.
The problems faced by the farmers before and after obtaining
loans have been used as indicators on the quality of the eXIsting credit
delivery system. Important problems were complicated procedures,
lack of previous experience, distant location of the bank and untlmcl\
and inadequate loan. Of the 200 sample farmers, 56.5 per cem statcd
that bank procedures were complex, 69.5 per ccnt felt t:1,,1 the
disbursl'd loan amount was inadequatl' and 49 pl'r cent \\"l'rL' of the
view thal loans were untimely.
More importantly, the quality of formal credit dC'liven system
was particularly poor in the case of marginal and small farmers.
Bribing the bank officIals was a widespread problem in the case of
marginal farmers. A general feeling among the borrowers was that
"bribing is the only means to obtain a loan for allied agricultural
activities under the subsidy and/ or poverty alleviation sehemes". The
evidence shows that the creditworthiness, viability of the proposal.
and the amount of cash disbursed were positively related with the
amount of bribe, while the time taken to get the loan sanctioned was
negatively related.
Given that the above would increase the transaction costs
(Adams and Nehman 1979; Sarap 1990J., the borrowing costs incurred
by farmers in obtaining short-term loans were worked out. The
evidence shows that the average transaction cost as a proportion to
average borrowing cost was inversely related with the size of the loan,
whereas, the average interest costs as a percentage of average
borrowing costs was positively related with the size of the loan.
The flow of formal credit was insufficient to meet the production
costs of peasant farmers engaged in culti\'ation. To meet the
remaining cost they turned either to self-finance or to borrowing from
the informal market or both or remained satisfied with poo;'
investment. It was in this context, that the credit-gap had b(TI1
estimated by size classes of landholdings within and across the
regIOns.
The term "credit-gap" has been defined as the un met credit
needs of potential borrowers from either formal or informal or fmm
both the sources with their respective terms and conditions. H()\\T\"J".
the actual and relative credit-gap per acre of gross cropped arca has
been worked out to assess the magnitude of shortage of funds to meet
the cost of production, Given tbe self-finance and formal credit the
amount of informal credit taken per acre of gross cropped area \\'as
termed as "actual credit-gap", wbereas, tbe differential amount
between the maximum and the actual paid-out cost per acre of gross
cropped area was defined as "relative credit-gap". Since ll1CITaSe 111
investment on crop production increased the level of production,
ceteris paribus (viz" crop failure, climatic condition, pnces of
agricultural product and so on), all categories of farmers \\'ould have
been interested to invest to the extent of maximum paid-out cost per
acre in the respective regions provided the required amount of credit
was available to them, Thus, the latter method assumed to be more
important in the sense that it explained the situation, where the
peasant farmer remained satisfied with poor investment due to lack of
availability of credit from either or both the sources.
It is noticed that both actual and relative credit-gaps were more
111 the advanced region as compared to the backward region for all
categories of farmers, Even if the obtained amount of formal credit per
acre of gross cropped area and self-finance capacity was more in the
advanced region, still the extent of credit-gap \\'as more. Gi\Tn the
self-finance, the higher credit-gap in the advanced region was
attributed to the higher cost of production. This indicates that the
available scale of finance was inadequate and well below the actual
cost of production,
As a coping strateg:.. farmers turned to informal credit market
with some terms and conditions. It was analysed that these contracts
could be in terms of either collateral or collateral substltutcs or
personal security or third party guarantor. The composition of
borrowing against different types of collateral sho\\'s that the quality of
collateral increased with the rise in the status of borrowers. The terms
and conditions of loan contract became unfavourable \\'ith the decline
in the quality of collateral assets provided by the borrowers, and in
turn, the extent of exploitation.
Conclusions
The overall conclusions emerging from the analysis of credit flow to
agriculture both at the macro and micro levels are briefly provided
below.
The analysis at the all India level shows that the proportion of
credit going to agriculture sector to net bank credit has been
declining from 1991. A similar situation was also found across the
bank groups.
The credit in both absolute and real terms grew at much faster rate
during the period 1981-91 as compared to the period 1992-2000,
in general, and across the financial institutions. in particular.
Despite the fixation of lending target, clubbing of direct and
indirect finances, deregulation of interest rates, reduction in the
number of slabs in lending rates and simplification of credit
delivery system, the banks could not achieve the targeted Clmoun t
in the case of agricultural lending at the all India level. Although
the consensus was that the credit to agriculturc should be stepped
up, the banks did not find the agricultural sector profitable for
deploying larger proportion of their loanable funds.
The state-wise pattern ll1 the agricultural lending shows that the
proportion of agricultural credit from net bank credit declined in
most of the states during 1981-90. However, this decline was
common for all the states during 1990s. The growth rate of
agricultural credit was higher during the pre-reform period as
compared to that after the post-1991 period in most of the states.
The growth rate was found to be uneven during the sub-periods as
well as across the states.
The distribution of loan accounts by size-class of landholdings at
the all India and State levels shows that better-off farmers had
improved access to formal credit as compared to small and
marginal farmers. The low percentage of farmers accessing formal
credit with more loan amount per account in most of the states
shows a concentration of formal loans in favour of a particular
group of borrowers. This also indicates that bankers adopted both
price and non-price credit rationing technique in the borrower
selection and fIxation of credit amount.
At the state level, irrigation and banking facility were two important
factors that determined the supply of credit per hectare of gross
napped area.
The analysis on purpose-\\'Jse flo\\" of credit at the bank-branch 1e\"C1
in Kalahandi district reveals that bankers were giving preference \u
lend for non-agricultural activilies. The analysis <.lIsa highlights
that the flow of agricultural credit was largely governed b\' the
security that the stock of capital offered to thc financial institllt ions
advancing credit. It IS also observed that for the bank branches
where the proportion of total bank credit was found to be more for
agriculture sector, their interest income was found to be less and
vIce-versa.
An analysis on credit delivery system shows that most of the
agricultural borrowers faced one or other problems before and after
obtaining loans. It is also observed that these problems were more
prominent in the case of marginal and small farmers. An analysis
on the time taken to get a loan sanctioned shows an inverse
relationship with the size of landholdings. Added to that, the loan
amounts were inadequate. This indicates the strong practice of
non-price credit rationing 111 the provision of agricultural credit 111
fixing the credit limit to a particular borrower. Since the formal
credit delivery system m the study area was found to be neither
adequate nor timely, it could be argued that the Gupta Committee
recommendations had hardly improved the quality of credit
delivery system in Kalahandi district of Orissa. It may be noted
that the study by Rajasekhar and Vyasulu (1993) on Kalahandi
district also noted that the quality of credit delivery system was
poor.
The transaction cost was a major component of the borrow1t1g costs
in the case of farmers taking smaller loan amounts. Further, the
effective rate of interest for Rs. 100 of loan was invcrseh' related
with the size of landholdings. An implication of this was thilt the
poor quality of the credit deliver.' system made the institutional
-
credit expensive for the small and marginal farmers.
The poor quality of land, illiteracy and lower caste status had an
adverse effect on the access of agricultural borrowers to formal
credit institutions. However, the value of non-land assets and area
under irrigation improved the probability of farmer's access to
formal sector.
Irrigated area, value of agricultural output, expenditure on farm
inputs and caste were important variables that determined the
amount of credit obtained.
The credit-gap was found to be higher in the advanced region due
to higher cost of production. The magnitude of credit-gap was
higher among small and marginal farmers.
Implications of the Study for Policy and Theory
The analysis carried out in the study has the following implications for
policy and theory.
(a) It is noticed from the all India level analysis that an increase in the
lending rate reduces the supply of agricultural credit. This result is
quite contradictory to the theory, which shows the
association between interest rate and the supply of credit.
Therefore, increasing lending rate may not be an important means
to increase the supply of agricultural credit.
It is argued that increasing interest rate or collateral reqUIrement
mcreases the risk of the bank's loan portfolio bv either
discouraging safer investor or by inducing borrowers to invest in
riskier projects and therefore, decreases the bank profit (Stiglitz
and Weiss 1981; Besley 1994). A lender may thus be better off
rationing access to credit at a lower interest rate rather than
raising the interest rate of agricultural credit.
Elsewhere, it is also argued that the borrowers were willing to pay
higher interest rates if they got adequate and timely loans
(Rajasekhar and Vyasulu 1990; Gadgil 1994, 1997). But in our
existing credit delivery system "that is a big if'. The borrowers,
willing to pay high interest rate, may be less worried about
repayment of the loan. In this context, given the asymmetric
information, Stiglitz and Weiss (1981) show that a profit
maximising bank will practice credit rationing and be reluctant to
increase interest rates in response to an excess demand.
(b) Our analysis shows the credit subsidy adversely affectmg the
supply of agricultural credit. It is also noticed that increasing
lending rate cannot be an appropriate measure to reduce the credit
subsidy since it also adversely affects the supply of agricultural
credit. Therefore, it can be suggested that reduction in the cost of
lending in terms of either financial cost or transaction cost or risk
cost or by any combinations of these three should be the prime
objective of the financial institutions to reduce the burden of credit
subsidy. This can be made possible by better qualIty of credit
delivery system (timely and adequate) and prompt repa\'ment of
loans.
(c) The density of bank branches positively influence the suppl\' of
formal credit. The regression results show that an increase in the
rural bank branches by one per cent leads to 0.52 increase in the
net bank credit at all India leveL This shows that the polic.v of
closing the loss-making bank branches is not really helpmg the
cause of agriculture. Any further reduction in the number of bank
branches may adversely effect the credit flow to agriculture sector.
Hence, if the rural bank branches are to be closed under
profitability consideration, importance should be given to
alternative mechanisms like mcreasmg the number of staff 111
neighbouring bank branches, simplification of procedures in loan
sanctioned, computerisation of bank branches, formation of
larmers' organisations, self-help groups and so on.
(d) The decline in the proportion of agricultural credit to net bank
credit has been alarming. The policy, in this context, should be a
stipulation of linking credit with deposits rather than stipulating
the floor level as a proportion of net bank credit. This can also be
supported by the findings of the bank branch level analysis.
(e) As of now the practice is to monitor the credit flow to agriculture in
terms of the proportion of credit to agriculture in total net bank
credit. Our analysis shows that this may not reflect the true
picture in so far as credit flow to agriculture is concerned. Hence,
given the positive demand for institutional credit, specific target
should be made in terms of proportion of credit to agriculture to
total bank deposits.
(I) It is found that the security value taken b\' thc bankcrs was much
more than the loan amount given to thc farmers. A polin' change
should be made to ensure that the loan sanctioned IS
commensurate with the value of security.
(g) At the household level, a relatively high demand for credit was
found with fairly large credit rationing. This implies that there is
need to develop credit programmes to support the credit needs of
different categories of farmers, especially the marginal and small.
Interestingly, the quality of land and the value of output arc
significant determinants of access to credit and amount of loan
obtained respectively rather than the size of land. This calls for
pollcy and programme support in the form of watershed, irrigation
development, etc., to improve the quality of land.
(h) In order to reduce the credit-gap problem, there should be
rethinking on the scale of finance. Nevertheless, a separate scale of
finance should be fixed for each service area of the concerned bank
since the cost of production varies across the regions.
Banking sector reforms were introduced to improve the quality
and viability of the credit delivery system, so that the credit flow to thc
hitherto neglected sectors such as agricultural would bc stepped up.
This study on institutional finance to agriculture at both macro and
micro levels found that credit to agriculture had in fact declined and
the better-off farmers have not been adversely affected by this. Credil
rationing is widely practised by bankers, and the small and marginal
farmers have been worse hit. The small and marginal farmers, unable
to obtain timely and adequate formal credit, turn to informal sourn's.
and, in the process, are being subjected to an exploitation in the inter-
linked credit markets. The access to institutional credit for small and
marginal farmers, therefore, continues to be an outstanding Issue
notwithstanding (or due to?) the recent banking sector reforms.
REFERENCES
Adams, D. W. and D. H. Graham (1981). A Critique of Traditional
Agricultural Credit Projects and Policies. JOimwl of Development
Economics, VIII (3): 347-66.
Adams, D. W. and G. 1. Nehman (1979). Borrowing Cost and the
Demand for Rural Credit. The Journal of Development Studies,
XV (2): 165-76.
Adams, D. W. and J. L. Tommy (1974). Financing Small Farmers: The
Brazilian Experience 1965-69. Agricultural Finance Review,
XXXV, 36-41.
Adams, D. W. and Robert C. Vogel (1986). Rural Financial Markets in
Low-Income Countries: Recent Controversies and Lessons.
World Development, XIV (4): 477 - 87.
Anderson, D. and Farida Khambata (1985). Financing Small-Scale
Industry and Agriculture in Developing Countries: The Merits
and Limitations of "Commercial" Policies. Economic Development
and Cultural Change, XXXIII (2): 349-71.
Araujo, Paulo F.e. and Richard L. Meyer (1978). Agricultural Credit
Policy in Brazil: Objectives and Results. Savings and
Development, II, 169-94.
Aron, 1. (1981). Modernisation of Agriculture in Developing Countries:
Resources, Potentials and Problems. Binghamton: John Wiley
and Sons.
Aryeetey, Ernest (1996). Rural Finance in Africa: Institutional
Development and Access the Poor. Annual World Bank
Conference on Development Economics, The World Bank.
Bardhan, P. (1984). Land, Labour, and Rural Poverty: Essays 111
Development Economics. Delhi: Oxford University Press
Bardhan, P. (ed.) (1989). The Economic Theory of Agrarian Institurions,
Oxford: Clarendon Press
231
Bardhan, P. and A. Rudra (1978). Interlinkage of Land, Labour and
Credit Relations: An Analysis of Village Survey data in East
India. Economic and Political Weekly, XIll (7): 367-84.
Bardhan, P. K. (1984). Land, Labour and Rural Poverty. New York:
Columbia University Press.
Basu, K. (1983). The Emergence of Isolation and Interlinkage in Rural
Markets. Oxford Economic Papers, XXXV (2): 262-80.
Basu, K. (1986). Market, Power and Social Norms. Economic alld
Political Weekly, XXI (43): 1893-6.
Basu, K. (1989). Rural Credit Markets: The Structure of Interest
Rates, Exploitation and Efficiency. In P.K. Bardhan (ed.). The
Economic Theory of Agrarian Institutions. Oxford: Clarendon
Press.
tlasu, Santanu (1 SJSJ7). Why Institutional Credit Agencies are
Reluctant to Lend to the Rural Poor: A Theoretical Analysis of
the Indian Credit Market. World Development, XXV (2): 267-80.
Bedbak, H. (1986). Institutional Financing for Priority Sectors: An
Empirical Assessment of Delay and Analysis of Attitudes of
Agencies Towards Loanees. Indian Co-operative Review, XXIV
(1): 65-76.
Bell, Clive (1990). Interactions between Institutional and Informal
Credit Agencies in Rural India. The World Bank Economic
Review, IV (3): 297-327.
BesIey, Timothy (1994). How do Market Failures Justify Interventions
in Rural Credit Markets? The World Bank Research Observer, IX
(1): 27-47.
Bhaduri, A. (1977). On the Formation of Usurious Interest Rates 111
Backward Agriculture. Cambridge Journal of Ecollomics, I (4).
Binswanger, Hans P. and M.R. Rosenz\\"ieg (1986). Behavioural and
Material Determinants of Production Relation in Agriculture.
Journal of Development Studies, XXXI! (3): 503-39.
232
Binswanger, Hans P. and Donald A. Sillers (1983). Risk A\'ersion and
Credit Constraints in Farmers' DeCIsion-Making: A
Reinterpretation. Journal of Development Studies, XXII (1): 5-21.
Binswanger, Hans P. and Kalus Deininger (1997).
Agricultural and Agrarian Policies In Developing
Joumal of Economic Literature, XXXV (4): 1958-2005.
Explaining
Countncs.
Binswanger, Hans P. and Shahidur R. Khandker (1995). The Impact of
Formal Finance on the Rural Economy of India. Journal of
Development Studies, XXXIl (2). 234-62,
Bishnoi, T, R. (1991), Policy Constraints and Banks' Profitability:
Impact of Priority Sector Credit, Intemauonal Journal of
Development Banking, IX (1): 39-46.
Bottomley, A, (1975). Interest Rate Determination in Underdeveloped
Rural Areas, American Journal of Agricultural Economics, 57 (2):
279-91.
Braveman, A. and J. L. Guasch (1986). Rural Credit Markets and
Institutions in Developing Countries: Lessons for Policy Analysis
from Practice and Modern Theory. World Development, XIV (10):
1253-67.
Braveman, A. and T.N, Srinivasan (1981). Credit and Sharecropping in
Agrarian Societies, Journal of Development Economics, IX (3):
289-312,
Braverman, A. and J.L. Guasch (1989). Rural Credit in Developing
Countries. PPR, Working Paper, No, 219. Washington DC: The
World Bank.
Chaudhuri, S. (1993), Problems of Institutional Credit and Small
Farmers: A Study of Two selected VIllages of West Bengal.
Unpublished.
Chaudhuri, Sarbajit and Manas Ranjan Gupta (1996). Dela\'ed Formal
Credit, Bribing and the Informal Credit Market in Agriculture: A
Theoretical Analysis. Journal of Development Economics, 51:
433-49.
233
Cho, Yoon Je (1986). Inefficiencies from Financial Liberalisation in the
Absence of Well-Functioning Equity Markets. Joumal of Money.
Credit and Banking, IX (I): 39-46.
Dandekar, V. M. and F. K. Wadia (1989). Development of Institutional
Finance for Agriculture in India. Joumal of Indian School of
Political Economy, I (2): 167-211.
Das, J. (1962). Final Report of Land Revenue Settlement in Kalahandi
District, Ex-State Khalsa, 1945-46. Bhubaneswar: Government
of Orissa.
Desai, Bhupat M. (1993). Rural Banking: Misdirected Policy Changes.
Economic and Political Weekly, XXVIlI (18): 2868 - 70.
Desai, Bhupat M. and N. V. Namboodiri (1 992). Performance of
Institutional Finance for Agricultural Development. Economic
and Political Weekly, XXVII (51 and 52): A-190 - A-196.
Donald, Gordon (1976). Credit for Small Farmers in Developing
Countries. Boulder, CU: Westview Press.
Eaton, J. and M. Gersovitz (1981). Debt with Potential RepUdiation:
Theoretical and Empirical Analysis. Review of Economic Studies,
XLVIII (2): 289-310.
Egger, P. (1986). Banking for the Rural Poor: Lessons from Some
Innovative Saving and Credit Schemes. Intemational Labour
Review, 125 (4): 447-62.
Fried, Joel and Peter Howitt (1980). Credit Rationing and Implicit
Contract Theory. Journal of Money Credit and Banking, XlII (2):
471-87.
Fry, M. (1988). Money, Interest and Banking in Economic Development.
Baltimore: Johns Hopkins Press.
Fry, Maxwell J. (1979). The Cost of Financial Repression in Turkev.
Savings and Development, 11[, 127-35.
Gadgil, M.V. (1986). Agricultural Credit m India: A Review of
Performance and Policies. Indian Joumal of Agnculluml
Economics, XLI (3): 282 - 309.
234
Gadgil, M.V. (1992). Future of Institutional Agricultural Crt'dit in
India: Likely Impact of Narasimham and Khusro Commit tee
Reports. Indian Journal of Agricultural Economics, 47(2): 255-05.
Gadgil, M.V. (1994). Formal Agricultural Credit SYstem in India:
Shape of Things to Come. Indian Journal of Agricultural
Economics, XLIX (3): 476-90.
Gadgil, M.V. (1997). National Bank for Agriculture and Rural
Development, Promise and Performance, 1982-83 to 1995-96.
Journal of Indian School of Political Economy, IX (2): 207-65.
Gangopadhyay, S. (1997). Some Issues in Interlinked Agrarian
Markets. In Kaushik Basu (ed.), Agrarian Questions. New Delhi:
Oxford University Press.
Gangopadhyay, S. and Kunal Sengupta (1987). Small Farmers, Money
Lenders and Trading Activity. Oxford Economic Papers, 39 (2):
333-342.
George, P.T., D. Namasivayam and G. Ramachandraiah (J 985). Rural
Credit and Farmer's Borrowing Cost: A Case Study. Prajnan,
XIV (3): 190-201.
Ghatak, S. (1977). Rural Credit and the Cost of Borrowing: Interstate
Variations in India. Journal of Development Studies, XIII (2):
102-24.
Ghatak, S. (1983). On Interregional Variations in Rural Interest Rates
in India. Journal of Development Areas, XVIII (1): 21-34.
Gibson, Heather D. and Euclid Taskalatos (1994). The Scope and
Limits of Financial Liberalisation in Developing Countries: A
Critical Survey. Journal of Development Studies, XXX (3): 578-
628.
Gonzalez-Vega, Claudio (1984). Credit Rationing Behaviour of
Agricultural Lenders: The Iron Law of Interest-Rate Restriction.
In Dale W. Adams, Douglas H. Graham and J. D. Von Pischke
(eds.), Undermining Rural Development with Cheap Credit.
Colorado, US: Westview Press.
Government of Orissa (1995-96). Orissa Agricultural StatistiCS.
Bhubaneswar: Directorate of Agriculture and Food Production.
235
Government of Orissa (1996). Statistical Abstract of Orissa.
Bhubaneswar: Directorate of Economics and Statistics.
Government of Orissa (1996-97). Orissa Agricultural Statistics.
Bhubaneswar: Directorate of Agriculture and Food Production.
Government of Orissa (1997). District Statistical Handbook 1997,
Kalahandi. Bhubaneswar: Directorate of Economics and
Statistics.
Government of Orissa (1997-98). Orissa Agricultural Statistics.
Bhubaneswar: Directorate of Agriculture and Food ProductlOn.
Government of Orissa (1998-99). Orissa Agricultural Slatistics.
Bhubaneswar: Directorate of Agriculture and Food Production.
Government of Orissa (1999-2000). Orissa Agricultural Statistics.
Bhubaneswar: Directorate of Agriculture and Food Production.
Government of Orissa (2000-2001). Economic Survey. Bhubaneswar:
Directorate of Economics and Statistics, Planning and Co-
ordination Department, Government of Orissa.
Gupta, R. V. (Chairman) (1998). Report of the High-Level Committee 011
Agricultural Credit through Commercial Banks. Bombay: Reserve
Bank of India.
Hashuliza, A.K. and J.G. Kydd (1996). Determinants of Bank Credit
Access for Smallholder Farmers in Tanzania: A Discriminate
Analysis Application. Savings and Development, XX (3): 285 -
304.
Heckman, J. J. (1979). Sample Selection Bias as a Specification Error.
Econometrica, XLVII (1): 153-61.
Helfand, M. Steven (2001). The Distribution of Subsidised Agricultural
Credit in Brazil: Do Interest Group Matter? Development and
Change, XXXII (3): 465-90.
Hoff. Karla and Joseph E. Stiglitz (1993). Imperfect Information and
Rural Credit Markets: Puzzles and Policy Perspectives. In Karla
Hoff. A. Braverman and Joseph E. Stiglitz (cds.), The Economics
of Rural Organisation: Theory, Practice and Policy. Ne\\' York:
Oxford University Press for the World Bank.
236
Hulme, David and Paul Mosley (1996). Finance Against Pover1y, Vol. 1.
London, UK: Routledge.
Jafee, Dwight M. (1971). Credit Rationing and the Commercial Loan
Market. New York: John Wiley.
Johnston, J. and J. Di Nardo (1997). Econometric Methods, (4tH
edition). New York: McGraw Hill.
Kahlon, A. S. (1991). Rural Overdues: Borrower's Angle. Economic and
Political Weekly, XXVI (5): 243 - 6.
Kamajou, Francois (1980). Subsidised Interest Rates and Restricted
Agricultural Credit in LDCs. Saving and Development, IV (2):
123-36.
Katula, R. and Ashok Gulati (1992). Institutional Credit for
Agriculture: Issues Related to Interest and Default Subsidy.
Journai of indian School of PolItIcal &onomy, IV (4): 701-29.
Khusro, A. M. (1989). A Review of the Agricultural Credit System m
India: Report of the Agricultural Credit Review Committee.
Bombay: Reserve Bank of India.
Kochar, Anjini (1997). An Empirical Investigation of Rationing
Constraints in Rural Credit Market in India. Journal of
Development Economics, LIII (2): 339-71.
Kohli, Renu (1997). Directed Credit and Financial System. Economic
and Political Weekly, XXXII (42): 2667-76.
Kohli, Renu (1999). Rural Bank Branches and Financial Reform.
Economic and Political Weekly, XXXIV (3 and 4): 169 - 74.
Ladman, Jerry R. (1984). Loan Transaction Costs, Credit Rationing,
and Market Structure: The Case of Bolivia. In Dale W. Adams,
Douglas H. Graham and J. D. Von Pischke (eds.), Undermining
Rural Development with Cheap Credit. Colorado, US: Westvie\,
Press.
Ladman, Jerry R. and D. W. Adams (1978). The Rural Poor and the
Recent Performance of Formal Rural Financial Markets in the
Dominican Republic. Canadian Journal of Agncultural
Economics, XXXVI (1): 43-50.
237
Ladman, Jerry R. and Ronald L. Tinnermeier (1981). The Political
Economy of Agricultural Credit: The Case of Bolivia. American
Journal of Agricultural Economics, LXIII (1): 66 - 72.
Lele, U. (1981). Co-operatives and the Poor: A Comparative
Perspective. World Development, IX (1): 55-72.
Lipton, M. (1976). Agricultural Finance and Rural Credit In Poor
Countries. World Development, IV (7): 543-53.
McKinnon, R. 1. (1973). Money and Capital in Economic Development.
Washington DC: The Brooking Institution.
Mitra, K. Pradeep (1983). A Theory of Interlinked Rural Transaction.
Journal of Public Economics, XX (2): 167 -9l.
Morduch, J. (2000). The Micro Finance System. World Development,
XXVllI (4): 617-29.
Naastepad, C. W. M. (2001). The Macro Economic Effect of Directed
Credit Policies: A Real-Financial CGE Evaluation for India.
Development and Change, XXXII (3): 491-520.
Narashimham, M. (Chairman) (1991). Report of the Committee on the
Financial System. New Delhi: Ministry of Finance, Government
of India.
National Bank for Agriculture and Rural Development (2001-02).
Potential Linked Credit Plan, Kalahandi. Bhubaneswar:
NABARD.
Odedokun, M. O. (1996). International Evidence on the Effects of
Directed Credit Programmes on Efficiency of Resource Allocation
in Developing Countries: The Case of Development Bank
Lendings. Journal of Development Economics, 48 (2): 449-60.
Pandit, B. L. (1991). The Growth and Structure of Savings in India: An
Econometric Analysis. New Delhi: Oxford University Press.
Petrick, M. and L. Latrufee (2003). Credit Access and Borrowing Costs
in Poland's Agricultural Credit Market: A Hedonic Pricillg
Approach (D.P. No. 46). Munich: Institute of Agricultural
Development in Central and Eastern Europe.
238
Pindyck, R. S. and D. L. Rubinfeld (1997). Econometric Models alld
Economic Forecasts. Fourth Edition. New York: McGraw-Hill.
Platteau, J. Ph and Anita Abraham (1987). An Inquiry into Quasi-
Credit Constraints: The Role of Reciprocal Credit and
Interlinked Deals in Small Scale Fishing Communities. The
Joumal oj Development Studies, XXIII (4): 461-90.
Rajasekhar, D. and Vinod Vyasulu (1990). Rural Credit Delivery
System: A Study in Pali District of Rajasthan. Economic and
Political Weekly, XXV (39): A-125-A-134.
Rajasekhar, D. and Vinod Vyasulu (1991). The Half-Life of Credit:
Stimulating Fund Flows in the Priority Sector, Economic and
Political Weekly, XXVI (39): A-135-A-143.
Rajasekhar, D. and Vinod Vyasulu (1993). Managerial Changes, Rural
Credit and Economic Development: A Study in Orissa, India.
Savings and Development, XVII (2): 183-208.
Ramachandran, V. K. and Madhura Swaminathan (2001). Does
InJormal Credit Provide Security? Rural Banking Policy in India.
Geneva: International Labour Office.
Rangarajan, C. (1996). Rural India, The Role of Credit. Reserve Bank
oj India Bulletin. Bombay: RBI.
Rao, C.H.H. (1975). Technological Change and Distribution of Gains in
Indian Agriculture, New Delhi: Macmillan.
Rao, C.H.H. (1994). Issues Relating to Irrigation and Rural Credit in
India. In G. S. Bhalla (ed.), Economic Liberalisation and Indian
Agriculture. New Delhi: Institute for Studies in Industrial
Development.
Rao, M.G., R.T. Shand and K.P. Kalirajan (1999). Convergence of
Incomes across Indian States. Economic and Political Weekly,
XXXIV (21: 167-88.
Ray, Devraj (1998). Development Economics. New Delhi: Oxford
University Press.
239
Reddy, Y.V. (2001). Future of Rural Banking. In Raj Kapila (cd.),
India's Banking and Financial Sectors in New Millell/llum. Ne\\"
Delhi: Academic Foundation.
Reserve Bank of India (1987-88). Report on Trends and Progress of
Banking in India. Bombay: RBI.
Reserve Bank of India (1991-92). Report on Trends and Progress of
Banking in India. Bombay: RBI.
Reserve Bank of India (1992-93). Report on Trends and Progress of
Banking in India. Mumbai: RBI.
Reserve Bank of India (1993-94). Report on Currency and Finance,
Volume -/. Mumbai: RBI.
Reserve Bank of India (1994). RBI Bulletin. Mumbai: RBI.
Reserve Bank of India (1995). Report on Currency and Finance, Vol. I.
Mumbai: RBI.
Reserve Bank of India (1996-97). Report on Currency and Finance. Vol.
I. Mumbai: RBI.
Reserve Bank of India (1997-98). Report on Trends and Progress of
Banking in India. Mumbai: RBI.
Reserve Bank of India (2000). Banking Statistics, Quarterly Handout.
Mumbai: RBI.
Reserve Bank of India (2001). Hand Book of Statistics on Indian
Economy. Mumbai: RBI.
Rhyne, E. (1998). The Yin and Yang of Micro-Finance: Reaching the
Poor and Sustainability. The Micro-banking Bulletin (2).
Rudra, A. (1982). Indian Agriculture: Myth and Realities. Ne\\' Delhi:
Allied Publishers.
Sarap, K. (1990). Factors Affecting Small Farmers' Access to
Institutional Credit in Rural Orissa, India. Development and
Change, XXI (2): 281-307.
240
Sarap, K. (1991). Collateral and other Forms of Guarantee III Rural
Credit Markets: Evidence from Eastern India. Indian Economic
Review, XXVI (2): 167-88.
Sarkar, Jayati and Pradeep Agrawal (1997). Banking: The Challenges
of Deregulation. In Kirit S Parikh (ed.), India Deve{opml?1l1
Report. Delhi: Oxford University Press.
Satish, P. and C.K. Gopalkrishna (1997). Viability of Rural Banking.
Economic and Political Weekly, XXXII(42): 2711-6.
Satish, S. and K. K. Swaminath (1989). Lending Cost and Margin,
Agricultural Credit Review, Bombay, (Study IV). Mumbai:
Reserve Bank of India.
Schrieder, Gertrud Buchenrieder (2000). Improving Bankability of
Small Farmers in Northern Vietnam. Savings and Development,
XXIV (4): 385-402.
Sen, A.K. (1975). Employment Technology and Development. Oxford:
Clarendon Press.
Shajahan, K.M. (1998). Priority Sector Bank Lending, Some Important
Issues. Economic and Political Weekly, XXXIII (42 and 43): 2749-
56.
Shajahan, K.M. (1999). Priority Sector Bank Lending: How Useful?
Economic and Political Weekly, XXXIV (51): 3572-4.
Shaw, E. K. (1973). Financial Deepening in Economic Development.
New Delhi: Oxford University Press.
Shivamaggi, H.V. (2000). Reforms in Rural Banking: Need for Bolder
Approach. Economic and Political Weekly, XXXV (20): 1714-8.
Shylendra, H .S. (1995). Lender Viability and Lender Behaviour: A
Case Study of a Regional Rural Bank in South India. Saving and
Development, XIX (2): 231-42.
Shylendra, H.S. (1996). Institutional Reform and Rural Poor: A Stud\'
on the Distributional Performance of Regional Rural Bank.
Indian Journal oj Agricultural Economics, 51 (3): 301-14.
2 ~ 1
Singh, Surjeet (ed.) (1991). Rural Credit: Issues for the Nineties. Ne\\'
Delhi: Oxford lBH.
State Bank of India (2000-01). Annual Credit Plan - 2000-2001.
Kalahandi District (Orissa). Bhubanes\\'ar: State Bank of India.
Local Head Office.
Stiglitz, J. (1990). Peer Monitoring and Credit Markets. The World
Bank Economic Review, IV (3): 351-66.
Stiglitz, Joseph E. (1994). The Role of the State in the Financial
Markets. In the Proceedings of the World Bank Annual
Conference on Development Economics 1993. Washington DC:
The World Bank.
Stiglitz, Joseph E. and Andrew Weiss (1981). Credit Rationing in
Markets with Imperfect Information. American Economic Review.
LXXI (3): 393-410.
Swain, R. B. (2002). Credit Rationing in Rural India. Journal of
Economic Development, XXVII (2): 1-20.
Swaminathan, Madhura (1991). Segmentation, Collateral Under
valuation, and the Rate of Interest in Agrarian Credit Markets:
Some Evidence from Two Villages in South India. Cambridge
Journal of Economics, XV (2): 161-78.
Timberg, T. A. and C. V. Aiyar (1984). Informal Credit Market in India.
Economic Development and Cultural Change, XXXIII (1): 43-59.
Vaessen, Jos (2001). Accessibility of Rural Credit in Northern
Nicaragua: The Importance of Networks of Information and
Recommendation. Savings and Development, XXV (1): 5 - 31.
Vogel, Robert C. (1977). The Effects of Subsidised Agricultural Credit 0/1
the Distribution of Income in Costa Rica. Carbondale, IL:
Department of Economics, Southern Illinois University
(un published).
Vogel, Robert C. and Donald W. Larson (1980). Limitations of
Agricultural Credit Planning: The Case of Colombia, Savings
and Development IV.
242
Von Pischke, J. D. and Dale W. Adams (1980). Fungibility and the
Design and Evaluation of Agricultural Credit Projects. American
Joumal of Agricultural Economics, LXII (4): 719 - 26.
Von Pischke, J. D., Dale W. Adams and Gordon Donald (1983). Rural
Financial Markets in Developing Countries: Their Use and Abuse.
Baltimore, Maryland: Johns Hopkins University Press.
Vyasulu, Vinod and D. Rajasekhar (1991). Management of Rural
Credit: Issues for 1990s. International Joumal of Developmen t
Banking, IX (1): 47-53.
Vyasulu, Vinod and D. Rajasekhar (1993).
Development: Managerial Reforms in
Development Policy Review, 11 (4): 393-412.
Credit for Rural
Indian Banks,
Wiggins, S. and S. Rajendran (1987). Rural Banking in Southern Tamil
Nadu: Performance and Management. (Final Research Repon).
Department of Agricultural Economics and Management,
University of Reading
World Bank (1989). World Development Report. Washington, DC: The
World Bank.
Yaron, J., M.P. Benjamin and G. Piprek (1997). Rural Finance: Issues,
Design, and Best Practices. Washington DC: The World Bank.
Yaron, Jacob (1994). What Makes Rural Finance Institutions
Successful? The World Bank Research Observer, IX (1): 49-70.
Yaron, Jacob, McDonald Benjamin and Stephanie Charitonenko
(1998). Promoting Efficient Rural Financial Intermediation. The
World Bank Research Observer, XIII (2): 147-70.
243