Institutional vs Non-institutional Credit to Agricultural Households in India:

Impact on Farmers' Welfare

Anjani Kumar

99th Annual Conference Indian Economic Association

27-29 December 2016Tirupati, Andhra Pradesh

India

Outline of the presentation Background Data Methodology Findings Conclusions

Background Credit plays a crucial role in agricultural

development enables farmers to undertake new investments and/or

adopt new technologies. access to credit can enhance the risk bearing ability of

the farmers and support them invest in a little risky ventures with higher potential returns

act as a catalyst to break the vicious circle of poverty in rural areas

Agricultural credit policy in India improve farmers’ access to institutional credit and

reduce their dependence on informal credit ratio of agricultural GDP increased from 10% in 1999-00

to 38% in 2012-13 Accounts for 85% of the purchased inputs in the

agriculture and allied sectors Major milestones of the rural credit before economic

reforms include acceptance of the Rural Credit Survey Committee’s Report (1954), nationalization of the large commercial banks (1969 and 1980), establishment of Regional Rural Banks (1975) and the National Bank for Agriculture and Rural Development in 1982

Contd… After financial reforms in India, the milestones

include Special agriculture credit plan (1994-95) Kisan credit cards (1998-99) Doubling agricultural credit (2004) Agricultural debt waiver and debt relief scheme (2008) Interest subvention scheme (2010-11) Jan Dhan Yojana (2014)

Other measures to strengthen formal credit programs in India

Lead bank scheme Direct lending for priority sector Banking sector’s linkage with the government sponsored

programs Differential rate of interest scheme Service area approach Self-help group banks linkage programs Special agricultural credit plans Rural infrastructure development fund

Contd..

Access to

Formal

Credit

Agricultural

Productivity

Source: Binswanger and Khandker, 1995; Carter 1989; Carter and Weibe 1990; Feder et al 1990; Pitt and Khandker, 1996, 1998; Khandker and Farooqui 2003; Awotide et al 2015; Narayanan 2016

Objective To understand and analyze

Characteristics of agricultural credit markets in India

Characteristics of institutional and non-institutional borrowers

Determinants of access to formal credit. The impact of institutional credit on agricultural

households’ welfare in India

Using unit level data from nationally representative sample of farm survey.

Net farm income and household consumption expenditure were taken as a proxies for measuring agricultural households’ welfare

Data Used “Situation Assessment Survey of Agricultural

Households” carried out by the National Sample Survey Organization (NSSO) in 2013

4529 villages spread across the country 35200 farming households Period; Agricultural year 2012-13

Comprehensive information on socio-economic well-being of agricultural households, consumption expenditure, income from productive assets, borrowing, lending and indebtedness, their farming practices and preferences, resource availability, receipts and expenses of household’s farm and non-farm

businesses, their awareness of technological developments and access to modern technology in the field of agriculture

Methodology Binary Logistic Regression

Determinants of farmers’ access to formal credit

Where; p represents the probability that the farmer

takes formal credit βs are the regression coefficients estimated

by the maximum likelihood method Xs represent the explanatory variables and

include several socio-economic and demographic characteristics of the farm households

Contd… Instrumental variable 2SLS

assess the impact of formal credit on farmers’ profits

Where; is net profit per ha received by a farm household from farming is a dummy variable (= 1 if the farmer takes formal credit and

0 otherwise) is a vector of observable farm and operator characteristics is an error term

Assumptions Estimation of the above equation using simple ordinary least

squares (OLS) may result in biased estimates. unobserved factors could be guiding farmers’ decision to access

the formal credit. Thus,, the variable representing farmer’s access to formal credit, is

likely to be endogenous and could be correlated with the error term, .

Conducted Hausman’s test for endogeneity and found access to formal credit to be endogenous

Proportion of farmers availing institutional credit in a village as the instrumental variable

Status of banking infrastructure in India (Global vis-à-vis India)

BangladeshBrazilChina

FranceGermany

IndiaIndonesia

JapanMalaysia

MexicoNepal

PakistanRussian Federation

Sri LankaUnited Kingdom

United States

0 10 20 30 40 50 60

No. of bank branches per 100,000 adult population

Cou

ntri

es

Source: WDI 2015 (World Bank)

Structure of credit delivery mechanism in India

Source: WDI 2015 (World Bank)

Institutional sources Government Cooperative banks Commercial banks

Non-institutional sources Money lenders Employer Shopkeepers/traders Relatives/friends

Share of institutional credit in rural borrowings in India (1951-2013)

1951 1961 1971 1981 1991 2003 20130.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

8.817.3

29.2

61.2 55.7 57.1 60.3

Year

Perc

ent

Source: AIRCS (RBI); AIDIS (NSSO)

International Food Policy Research Institute

Amount of borrowings in India: 1992, 2003 and 2013

Average amount of borrowing (Rs/ha at 1993-94 prices)

1992 2003 20130

1000

2000

3000

4000

5000

6000

980

3356

4850

International Food Policy Research Institute

Equity in disbursement of institutional credit in India, 2003 and 2013

Source: NSS, GoI

Landless, Marginal and Smallholders

SC & STs OBCs'0

0.2

0.4

0.6

0.8

1

1.2

0.720.44

0.980.840.49

1.13

Ratio of weaker sections in institutional credit and households

2003 2013

Distribution of loans by sources (%)

64%

36%

Formal SourcesInformal Sources

Formal Sources Informal Sources

Government Co-operative Society

Bank0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

0.020.040.060.0

Farmers’ access to credit from formal and informal sectors, 2012-13Distribution of HHs by borrowing

(%)Share of formal and informal

credit in borrowing households (%)

Marginal

Small Medium Large All0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

0.0

10.0

20.0

30.0

40.0

50.0

60.0

14.824.2

30.039.1

22.8

21.7

14.211.6

8.9

16.2

8.913.3

17.0

23.6

13.1

54.7

48.3

41.5

28.5

47.9

Formal sourcesInformal sources Marginal Small Medium Large All

0.0

10.0

20.0

30.0

40.0

50.0

60.0

70.0

80.0

55.0

64.367.1

74.9

63.6

45.1

35.732.9

25.1

36.4

Formal Credit Informal Credit

Distribution of borrowers households by operational holding (%)

Share of HHs Share of Non-borrower

39.9

30.5

22.8

6.8

45.5

30.8

19.7

4.1

Marginal

Small

Medium

Large

Source of Borrowing Share in Credit

Marginal Small Medium Large0.0

10.0

20.0

30.0

40.0

50.0

60.0Formal Credit Informal CreditBoth Simultaneously

Marginal Small Medium Large0.0

5.0

10.0

15.0

20.0

25.0

30.0

35.0Formal Credit Informal Credit

General characteristics of institutional and non-institutional borrower

Particular Institutional

Non-institutional Particular Institutio

nalNon-

institutionalSocio-demographic variables Structure of households by farm categories

(%)Age (Years) 51.1 46.9 Marginal 25.9 41.6Family size (No.) 5.2 5.2 Small 32.5 28.6Land size (Ha) 1.7 0.8 Medium 30 22.2Per Capita Monthly Expenditure (Rs) 1603.7 1298.3 Large 11.7 7.6

Male headed households (%) 93.8 91 Principal source of household income (%)% received formal training in agriculture 3.7 1.9 Agricultural Income 79.2 74.8

Social structure by caste (%)  Non-Agr. Income 17.4 22.4Schedule tribe 8.6 13.1 Pension 1.3 0.5Schedule caste 12.8 19.1 Remittance 1.7 2Other backward caste 46.1 47.2 Awareness and access to social safety nets (%)

General caste 32.2 20.7 Minimum Support Price Awareness 30.6 22.3

Social structure by religion (%) Having MGNREGA Job Card 93.1 90.7

Hindu 88.4 87.2 Have PDS Ration Card 37.4 51.6Muslim 6.1 9.7 Source of Technical AdviceChristian 1.8 1.1 Extension Agent 19.2 17.3Other 3.3 1.8 KVK & SAU 8.5 5Education level of the head of household (%)   Pvt Commercial Agents 8.1 8.2Illiterate 30.1 49.4 Progressive Farmer 20.7 22.3

Primary 27.9 24.8 Radio / TV / Newspaper / Internet 28.8 22.7

Middle 17.3 13.6 NGO 1.3 0.8Secondary 12.2 7.6Higher secondary & above 12.4 4.7

Determinants of access to institutional credit

Significant variables dy/dx SELog of age of the household head 0.192*** (0.0134)Middle (Yes = 1, otherwise = 0) 0.0876*** (0.0129)Higher Secondary (Yes = 1, otherwise = 0) 0.138*** (0.0099)Graduate and above (Yes = 1, otherwise = 0) 0.199*** (0.0227)Schedule Tribe (Yes = 1, otherwise = 0) 0.0341* (0.0189)OBC (Yes = 1, otherwise = 0) 0.0321** (0.0155)

Other Caste (Yes = 1, otherwise = 0) 0.0275* (0.0150)

Log of Per Capita Monthly Expenditure (Rs) 0.0668*** (0.0126)Small (Yes = 1, otherwise = 0) 0.100*** (0.0110)

Medium (Yes = 1, otherwise = 0) 0.126*** (0.0138)Large (Yes = 1, otherwise = 0) 0.123*** (0.0208)

MGNREGA Job Card (Yes = 1, otherwise = 0) -0.0778*** (0.0130)Have Ration Card (Yes = 1, otherwise = 0) 0.0556*** (0.0178)Extension Agent (Yes = 1, otherwise = 0) -0.0389** (0.0178)KVK & SAU (Yes = 1, otherwise = 0) 0.0561*** (0.0174)Pvt Commercial Agents (Yes = 1, otherwise = 0) -0.0284 (0.0244)

Progressive Farmer (Yes = 1, otherwise = 0) -0.0444*** (0.0141)Radio / TV / Newspaper / Internet (Yes = 1, otherwise = 0) 0.0105 (0.0135)

NGO (Yes = 1, otherwise = 0) 0.0687* (0.0389)Minimum Support Price Awareness (Yes =1, otherwise = 0) 0.0254* (0.0141)

Constant -6.534***Observations 16583District fixed effect YesLog pseudo-likelihood -10588.192

Average return of net farm income and household consumption expenditure

Net farm income (Rs/ha)

Consumption expenditure (Rs/month/person)

Marginal Small Medium Large All0

10,000

20,000

30,000

40,000

50,000

Marginal Small Medium Large All0

500

1,000

1,500

2,000

Formal BorrowerInformal Borrower

Impact of institutional credit on net farm incomeSignificant Variables Coefficient Standard

ErrorInstitutional Credit (Yes = 1, otherwise = 0) 0.171*** (0.0456)Log of household size 0.387*** (0.0307)Gender (Male = 1, otherwise =0) 0.112*** (0.0361)Graduate and above (Yes = 1, otherwise = 0) 0.121*** (0.0371)OBC (Yes = 1, otherwise = 0) 0.0814** (0.0369)Other Caste (Yes = 1, otherwise = 0) 0.132*** (0.0379)Others Religion (Yes = 1, otherwise = 0) 0.446*** (0.0567)Agricultural Income (Yes = 1, otherwise = 0) 0.785*** (0.1420)Log of Per Capita Monthly Expenditure (Rs) 0.378*** (0.0368)Small landholding (Yes = 1, otherwise = 0) 0.707*** (0.0253)Medium landholding (Yes = 1, otherwise = 0) 1.084*** (0.0295)Large landholding (Yes = 1, otherwise = 0) 1.623*** (0.0480)MGNREGA Job Card (Yes = 1, otherwise = 0) -0.0784*** (0.0290)Private Commercial Agents Source (Yes = 1, otherwise = 0) 0.178*** (0.0607)Radio / TV / Newspaper / Internet Source (Yes = 1, otherwise = 0) 0.0662* (0.0350)NGO Source (Yes = 1, otherwise = 0) 0.258** (0.1120)MSP Awareness (Yes = 1, otherwise = 0) 0.246*** (0.0241)Share of food crop 0.570*** (0.0320)Share of high value crops 0.897*** (0.0338)Share of oilseeds 0.383*** (0.0524)Share of other crops (Non-food) 0.348*** (0.0443)Proportion of HHs availed institutional credit by village wise 0.956*** (0.0030)Constant 5.366*** (0.3760)Observations 16583District Fixed Effect YesR-squared 0.528

Impact of institutional credit on household consumption expenditure

 Significant Variables Coefficient Standard ErrorInstitutional Credit (Yes = 1, otherwise = 0) 0.107*** (0.0239)Log of age of the household head 0.195*** (0.0152)Log of household size -0.482*** (0.0111)Gender (Male = 1, otherwise =0) -0.0463*** (0.0162)Middle School (Yes = 1, otherwise = 0) 0.0315*** (0.0120)Higher Secondary School (Yes = 1, otherwise = 0) 0.124*** (0.0131)Graduate and above (Yes = 1, otherwise = 0) 0.233*** (0.0217)Schedule Tribe (Yes = 1, otherwise = 0) -0.0857*** (0.0285)OBC (Yes = 1, otherwise = 0) 0.0584*** (0.0161)Other Caste (Yes = 1, otherwise = 0) 0.0939*** (0.0161)Muslim (Yes = 1, otherwise = 0) 0.137*** (0.0226)Christian (Yes = 1, otherwise = 0) 0.269*** (0.0396)Others Religion (Yes = 1, otherwise = 0) 0.450*** (0.0389)Non-Agricultural Income (Yes = 1, otherwise = 0) 0.143* (0.0790)Pension (Yes = 1, otherwise = 0) 0.249** (0.1110)Remittance (Yes = 1, otherwise = 0) 0.212*** (0.0805)Small landholding (Yes = 1, otherwise = 0) 0.0801*** (0.0123)Medium landholding (Yes = 1, otherwise = 0) 0.173*** (0.0148)Large landholding (Yes = 1, otherwise = 0) 0.311*** (0.0202)MGNREGA Job Card (Yes = 1, otherwise = 0) -0.0640*** (0.0132)Have Ration Card (Yes = 1, otherwise = 0) 0.137*** (0.0151)Krishi Vigyan Kendra & SAU Source (Yes = 1, otherwise = 0) 0.111*** (0.0246)Private Commercial Agents Source (Yes = 1, otherwise = 0) 0.0503* (0.0266)Radio / TV / Newspaper / Internet Source (Yes = 1, otherwise = 0) 0.0735*** (0.0092)MSP Awareness (Yes = 1, otherwise = 0) 0.0832*** (0.0146)Share of food crop -0.0505*** (0.0122)Share of high value crops 0.133*** (0.0173)Proportion of HHs availed institutional credit by village wise 0.957*** (0.0030)Constant 6.694*** (0.0968)Observations 16,583District Fixed Effect YesR-squared 0.360

Hausman test for endogeneity for net farm income and household consumption expenditure

 Variable 2SLS^

Coefficient Standard ErrorNet farm income 0.171*** (0.0456)Household consumption expenditure 0.107*** (0.0239)

Ehat$ for net farm income -0.118** (0.0464)

Ehat$ for household consumption expenditure -0.0843*** (0.0249)

Note: ^ Used Instrumental variable 2SLS method to investigate the role of institutional farm credit on farm income and farm household consumption expenditures. Our instrumental variable was “proportion of HHs availed institutional credit by village wise”; $ Hausman’s test for endogeneity for net farm income and household consumption expenditure; Robust standard errors in parentheses; *** p<0.01, ** p<0.05, * p<0.1

Conclusions Changing structure of rural credit market

in India increase in the flow and share of institutional

credit improvement in financial inclusion indicators land holding (marginal & small farmers) social group (SCs, STs and OBCs)

Concerns in the rural credit delivery system

disparity in disbursement of rural credit (states and social groups)

persistence of informal agencies (charging very high interest rates)

Conclusions Access to Institutional credit has positive

and significant effect Farmers’ profits Farmers’ monthly expenditure

Determinants for agricultural households’ access to institutional credit

age, education, caste affiliation, gender, occupation and assets ownership

Way forward

Flexible products and services

Emphasis on financial literacy

Simplification of lending procedure

Unique identification number for households

Convergence with extension and value chain

development

THANK YOU