Financing Smallholder Agriculture: An Experiment with - - PowerPoint PPT Presentation

Financing Smallholder Agriculture: An Experiment with Agent-Intermediated Microloans in India

Pushkar Maitra, Sandip Mitra, Dilip Mookherjee, Alberto Motta and Sujata Visaria

2nd ATE Symposium, UNSW, Sydney

15 Dec 2014

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 1 / 32

Motivation

How can we provide finance to poor farmers?

Institutional finance does not reach smallholder agriculturalists

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 2 / 32

Motivation

How can we provide finance to poor farmers?

Institutional finance does not reach smallholder agriculturalists

lack of collateral poor information and enforcement capacity high transaction costs

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 2 / 32

Motivation

How can we provide finance to poor farmers?

Institutional finance does not reach smallholder agriculturalists

lack of collateral poor information and enforcement capacity high transaction costs

Poor farmers forced to rely on credit from informal lenders at high cost Prevents poor farmers from diversifying into high-value cash crops Restricts agricultural growth

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 2 / 32

Motivation

How can we provide finance to poor farmers?

Institutional finance does not reach smallholder agriculturalists

lack of collateral poor information and enforcement capacity high transaction costs

Poor farmers forced to rely on credit from informal lenders at high cost Prevents poor farmers from diversifying into high-value cash crops Restricts agricultural growth Major development challenge: can financial institutions lend profitably to productive small farmers?

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 2 / 32

Motivation

“Traditional” microcredit does not solve the problem.

Microcredit has expanded credit access for the poor...

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 3 / 32

Motivation

“Traditional” microcredit does not solve the problem.

Microcredit has expanded credit access for the poor...

generated high repayment rates

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 3 / 32

Motivation

“Traditional” microcredit does not solve the problem.

Microcredit has expanded credit access for the poor...

generated high repayment rates

...but has not successfully promoted entrepreneurship, risk-taking and borrower incomes (Banerjee et al 2013, Fischer 2013, Karlan & Zinman 2011)

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 3 / 32

Motivation

“Traditional” microcredit does not solve the problem.

Microcredit has expanded credit access for the poor...

generated high repayment rates

...but has not successfully promoted entrepreneurship, risk-taking and borrower incomes (Banerjee et al 2013, Fischer 2013, Karlan & Zinman 2011) It has not financed agricultural working capital in developing countries.

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 3 / 32

Motivation

Why Not?

Traditional microcredit

motivated by need to ensure high repayment rates high frequency (weekly/bi-weekly/monthly) repayment intensive peer monitoring and MFI monitoring to limit risk-taking

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 4 / 32

Motivation

Why Not?

Traditional microcredit

motivated by need to ensure high repayment rates high frequency (weekly/bi-weekly/monthly) repayment intensive peer monitoring and MFI monitoring to limit risk-taking

But these rule out scope for financing agriculture

long crop cycles high-risk

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 4 / 32

Motivation

Our Question

Is there an inevitable trade-off between selection/repayment incentives and financing productive and risky activities?

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 5 / 32

Motivation

Our Question

Is there an inevitable trade-off between selection/repayment incentives and financing productive and risky activities? OR

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 5 / 32

Motivation

Our Question

Is there an inevitable trade-off between selection/repayment incentives and financing productive and risky activities? OR Can one design an alternative lending mechanism that ensures appropriate borrower selection and repayment incentives? can be cost-effective for lending institutions?

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 5 / 32

Motivation

Our Approach: TRAIL

We design and implement a new approach to financing smallholder agriculture

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 6 / 32

Motivation

Our Approach: TRAIL

We design and implement a new approach to financing smallholder agriculture Loans have certain features that encourage their use in agriculture

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 6 / 32

Motivation

Our Approach: TRAIL

We design and implement a new approach to financing smallholder agriculture Loans have certain features that encourage their use in agriculture

4-month durations (to match crop cycles) lumpsum repayments built-in crop insurance minimal monitoring by lender ‘low’ (18% p.a.) interest rate dynamic repayment incentives

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 6 / 32

Motivation

Our Approach: TRAIL

We design and implement a new approach to financing smallholder agriculture Loans have certain features that encourage their use in agriculture

4-month durations (to match crop cycles) lumpsum repayments built-in crop insurance minimal monitoring by lender ‘low’ (18% p.a.) interest rate dynamic repayment incentives

borrower selection: a local intermediary recommends borrowers

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 6 / 32

Motivation

Our Approach: TRAIL

We design and implement a new approach to financing smallholder agriculture Loans have certain features that encourage their use in agriculture

4-month durations (to match crop cycles) lumpsum repayments built-in crop insurance minimal monitoring by lender ‘low’ (18% p.a.) interest rate dynamic repayment incentives

borrower selection: a local intermediary recommends borrowers individual liability loans

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 6 / 32

Motivation

Our Approach: TRAIL

We design and implement a new approach to financing smallholder agriculture Loans have certain features that encourage their use in agriculture

4-month durations (to match crop cycles) lumpsum repayments built-in crop insurance minimal monitoring by lender ‘low’ (18% p.a.) interest rate dynamic repayment incentives

borrower selection: a local intermediary recommends borrowers individual liability loans We call this the Trader Agent Intermediated Lending (TRAIL) scheme

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 6 / 32

Motivation

Comparing TRAIL with GBL

We conduct a field experiment to evaluate the TRAIL scheme...

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 7 / 32

Motivation

Comparing TRAIL with GBL

We conduct a field experiment to evaluate the TRAIL scheme... ...in comparison to a group-based lending (GBL) scheme

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 7 / 32

Motivation

Comparing TRAIL with GBL

We conduct a field experiment to evaluate the TRAIL scheme... ...in comparison to a group-based lending (GBL) scheme The GBL scheme has the same features to encourage loan use in agriculture

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 7 / 32

Motivation

Comparing TRAIL with GBL

We conduct a field experiment to evaluate the TRAIL scheme... ...in comparison to a group-based lending (GBL) scheme The GBL scheme has the same features to encourage loan use in agriculture and also the “traditional” group-lending features

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 7 / 32

Motivation

Comparing TRAIL with GBL

We conduct a field experiment to evaluate the TRAIL scheme... ...in comparison to a group-based lending (GBL) scheme The GBL scheme has the same features to encourage loan use in agriculture and also the “traditional” group-lending features

self-forming 5-member groups monthly group meetings savings targets joint liability loans

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 7 / 32

Experiment

The Field Experiment

We collaborated with a Kolkata-based micro-lender in two potato-growing districts in eastern India, to introduce

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 8 / 32

Experiment

The Field Experiment

We collaborated with a Kolkata-based micro-lender in two potato-growing districts in eastern India, to introduce

TRAIL scheme in 24 randomly chosen villages GBL scheme in 24 randomly chosen villages

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 8 / 32

Experiment

The Field Experiment

We collaborated with a Kolkata-based micro-lender in two potato-growing districts in eastern India, to introduce

TRAIL scheme in 24 randomly chosen villages GBL scheme in 24 randomly chosen villages

We evaluate the impact of the loans on borrowers’ total borrowing, cultivation choices, output and value-added

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 8 / 32

Experiment

The Field Experiment

We collaborated with a Kolkata-based micro-lender in two potato-growing districts in eastern India, to introduce

TRAIL scheme in 24 randomly chosen villages GBL scheme in 24 randomly chosen villages

We evaluate the impact of the loans on borrowers’ total borrowing, cultivation choices, output and value-added We separately examine behaviour of those selected to receive loans

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 8 / 32

Experiment

The Field Experiment

We collaborated with a Kolkata-based micro-lender in two potato-growing districts in eastern India, to introduce

TRAIL scheme in 24 randomly chosen villages GBL scheme in 24 randomly chosen villages

We evaluate the impact of the loans on borrowers’ total borrowing, cultivation choices, output and value-added We separately examine behaviour of those selected to receive loans TRAIL scheme

in each village, the agent recommends a set of borrowers... ...and the lender offers the TRAIL loans to a randomly chosen subset of individuals

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 8 / 32

Experiment

The Field Experiment

We collaborated with a Kolkata-based micro-lender in two potato-growing districts in eastern India, to introduce

TRAIL scheme in 24 randomly chosen villages GBL scheme in 24 randomly chosen villages

We evaluate the impact of the loans on borrowers’ total borrowing, cultivation choices, output and value-added We separately examine behaviour of those selected to receive loans TRAIL scheme

in each village, the agent recommends a set of borrowers... ...and the lender offers the TRAIL loans to a randomly chosen subset of individuals

GBL scheme

in each village, individuals form groups and apply for loans... ...and the lender offers the GBL loans to a randomly chosen subset of groups

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 8 / 32

Experiment

Agent’s Incentives

Agent receives a commission = fraction of interest received from the borrowers he recommended (= 75% in experiment) Bonus: payable at end of two years conditional on satisfactory repayment record of recommended clients

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 9 / 32

Experiment

Possible Abuse of Power by Agents

Agents could:

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 10 / 32

Experiment

Possible Abuse of Power by Agents

Agents could: charge high interest rates (if permitted or through kickbacks) select cronies select unprofitable clients select non-poor clients select those willing to pay bribes extract borrower benefits by manipulating other contractual relationships recommend non-repayment of loans to borrowers and divide up loan funds coerce borrowers to repay

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 10 / 32

Experiment

Features to Limit Abuse of Power

MFI lends directly to client rather than through the agent

Agent only recommends clients, MFI officials lend and collect payments

Only landless and marginal landowners (that own  1.5 acres) are eligible to borrow Interest rate is pegged below the average informal market rate Agent forfeits deposit he posted upfront if any client does not repay Agent is terminated if less than 50% loans are repaid in any cycle Not every household recommended by agent receives the loan (randomization)

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 11 / 32

Theory

Theoretical Mechanism: Network effects

simple model based on network effects

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 12 / 32

Theory

Theoretical Mechanism: Network effects

simple model based on network effects

abstract from adverse selection, moral hazard or credit rationing... although adverse selection or moral hazard models generate qualitatively similar results

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 12 / 32

Theory

Theoretical Mechanism: Network effects

simple model based on network effects

abstract from adverse selection, moral hazard or credit rationing... although adverse selection or moral hazard models generate qualitatively similar results

segmented credit market within each village

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 12 / 32

Theory

Theoretical Mechanism: Network effects

simple model based on network effects

abstract from adverse selection, moral hazard or credit rationing... although adverse selection or moral hazard models generate qualitatively similar results

segmented credit market within each village each informal lender has a network of households

with whom he shares information about production/marketing which acts cooperatively to max. aggregate payoffs

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 12 / 32

Theory

Theoretical Mechanism: Network effects

simple model based on network effects

abstract from adverse selection, moral hazard or credit rationing... although adverse selection or moral hazard models generate qualitatively similar results

segmented credit market within each village each informal lender has a network of households

with whom he shares information about production/marketing which acts cooperatively to max. aggregate payoffs

borrowers either

belong to a network (“connected”): project succeeds with pc belong to no network (“floating”): project succeeds with pf

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 12 / 32

Theory

Theoretical Mechanism: Network effects

simple model based on network effects

abstract from adverse selection, moral hazard or credit rationing... although adverse selection or moral hazard models generate qualitatively similar results

segmented credit market within each village each informal lender has a network of households

with whom he shares information about production/marketing which acts cooperatively to max. aggregate payoffs

borrowers either

belong to a network (“connected”): project succeeds with pc belong to no network (“floating”): project succeeds with pf

connected borrowers succeed more often: pf (2 pf ) < pc connected borrowers pay lower informal interest rates

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 12 / 32

Theory

Enter the TRAIL Scheme

Lender appoints one (randomly chosen) informal lender as its TRAIL agent

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 13 / 32

Theory

Enter the TRAIL Scheme

Lender appoints one (randomly chosen) informal lender as its TRAIL agent Whom will the agent recommend for the TRAIL loans?

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 13 / 32

Theory

Enter the TRAIL Scheme

Lender appoints one (randomly chosen) informal lender as its TRAIL agent Whom will the agent recommend for the TRAIL loans? Own-network borrower Other-network borrower Floating borrower

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 13 / 32

Theory

Enter the TRAIL Scheme

Lender appoints one (randomly chosen) informal lender as its TRAIL agent Whom will the agent recommend for the TRAIL loans? Own-network borrower Other-network borrower Floating borrower Why?

Own-network borrower chooses loan size to maximize entire network profit A borrower belonging to a network has higher repayment rate than a floating borrower

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 13 / 32

Theory

Consider the GBL Scheme

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 14 / 32

Theory

Consider the GBL Scheme

GBL loans have lower interest rates than average informal rate ) both connected and floating borrowers will apply for loans No mechanism to exclude either type from borrowing

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 14 / 32

Theory

Consider the GBL Scheme

GBL loans have lower interest rates than average informal rate ) both connected and floating borrowers will apply for loans No mechanism to exclude either type from borrowing The average GBL borrower will be less likely to be connected than the average TRAIL borrower

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 14 / 32

Theory

Consider the GBL Scheme

GBL loans have lower interest rates than average informal rate ) both connected and floating borrowers will apply for loans No mechanism to exclude either type from borrowing The average GBL borrower will be less likely to be connected than the average TRAIL borrower )

GBL borrowers have greater heterogeneity in productivity average GBL borrower pays higher informal interest rate

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 14 / 32

Treatment and Selection Effects

Estimating Treatment and Selection Effects

yi = β0 + β1TRAIL + β2(TRAIL ⇥ Recommended but no loan) + β3(TRAIL ⇥ Offered loan) + β4(GBL ⇥ Formed group but no loan) + β5(GBL ⇥ Offered loan) + γ Xi + εi

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 15 / 32

Treatment and Selection Effects

Estimating Treatment and Selection Effects

yi = β0 + β1TRAIL + β2(TRAIL ⇥ Recommended but no loan) + β3(TRAIL ⇥ Offered loan) + β4(GBL ⇥ Formed group but no loan) + β5(GBL ⇥ Offered loan) + γ Xi + εi Selection effects (TRAIL, GBL): β2, β4 Treatment effects (ITT estimates) (TRAIL, GBL): β3 β2, β5 β4 Controls include land owned, year dummy, price information intervention dummy Standard errors clustered at the village level

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 15 / 32

Treatment and Selection Effects

Impacts on Borrowing

Unit Treatment Selection Sample Mean TRAIL GBL TRAIL GBL Size Control 1 All Loans: Loan Size Rs 7126*** 6464***

• 417
• 919

2758 7279 Cost of Borrowing Percent

• 0.03**
• 0.07***
• 0.01

0.04** 2428 0.24 Non Program Loans: Loan Size Rs

• 495

254

• 372
• 930

2601 7279 Cost of Borrowing Percent 0.01

• 0.01
• 0.01

0.04*** 2159 0.24

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 16 / 32

Treatment and Selection Effects

Impacts on Potato Production

Unit Treatment Selection Sample Mean TRAIL GBL TRAIL GBL Size Control 1 Cultivate 0.0545 0.0492 0.0949*** 0.0614 4163 0.677 Acreage Acres 0.0896*** 0.0402 0.0010

• 0.0421

2718 0.432 Leased-in acres Acres 0.0467** 0.0222

• 0.00265

0.00447 2718 0.111 Output Kg 888.0*** 278 145.4

• 417.9

2718 4760 Cost of production Rs 1774** 1308 372.8

• 1111

2718 9538 Family labour hours Hours 6.03 4.91

• 0.2

4.951 2718 57.86 Revenue Rs 3429*** 1637 942

• 2534

2718 19137 Value added Rs 1687** 271.8 555.6

• 1371

2718 9498

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 17 / 32

Treatment and Selection Effects

Impacts on Value Added of Other Crops, and on Total Farm Income

Unit Treatment Selection Sample Mean TRAIL GBL TRAIL GBL Size Control 1 Sesame Rs 180

• 158.3
• 115.7

73.41 2037 2126 Paddy Rs 271.6 573.6

• 469.9
• 759.6*

3047 2506 Vegetables Rs 1255

• 1955

1329

• 957.5

402 8325 Total Farm Income Rs 2621*** 53.24 11466*** 10066*** 4163 10328

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 18 / 32

Explanations

1: TRAIL borrowers were more productive

Potatoes Total Farm income Bootstrapped Estimates TRAIL 1.05*** 1.15*** (0.06) (0.02) GBL 0.09

• 0.10

(0.37) (0.29) IV Estimates of Cobb-Douglas Production Function TRAIL 0.72** 1.03*** (0.33) (0.35) GBL 0.37 0.38 (0.97) (1.23)

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 19 / 32

Explanations

2: TRAIL agents recommended own-network borrowers

(Dependent Variable: Household was recommended) (1) (2) Bought from agent 0.023 0.016 (0.044) (0.047) Borrow from agent 0.139*** 0.142*** (0.037) (0.035) Work for agent 0.003

• 0.005

(0.049) (0.055) Non Hindu 0.030 (0.143) Non Hindu × Agent Hindu

• 0.098

(0.132) SC 0.544*** (0.031) SC × Agent High Caste

• 0.610***

(0.036) ST

• 0.198*

(0.108) ST × Agent High Caste 0.218 (0.166) Sample Size 1031 1031

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 20 / 32

Explanations

3: TRAIL borrowers were lower-risk

TRAIL GBL TRAIL v GBL OLS Heckman OLS Heckman OLS Heckman (1) (2) (3) (4) (5) (6) Recommend 0.022 0.022 0.053* 0.052* (0.016) (0.017) (0.027) (0.029) Own-clientele 0.050 0.049* (0.033) (0.027) Own-clientele × Recommend

• 0.071**
• 0.071**

(0.026) (0.035) TRAIL

• 0.064
• 0.064**

(0.046) (0.027) High caste

• 0.058***
• 0.059***

0.134* 0.134*** 0.053 0.053* (0.016) (0.016) (0.071) (0.031) (0.044) (0.028) Landholding 0.091 0.090

• 0.103
• 0.071
• 0.047
• 0.023

(0.070) (0.078) (0.170) (0.142) (0.182) (0.129) Landholding Squared

• 0.063
• 0.062

0.065 0.050 0.053 0.042 (0.044) (0.052) (0.129) (0.093) (0.136) (0.086) Constant 0.238*** 0.240*** 0.196*** 0.151 0.271*** 0.235** Sample Size 438 1,032 417 1,038 412 911

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 21 / 32

Explanations

Loan Repayment Rates

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 22 / 32

Explanations

Loan Take-Up Rates

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 23 / 32

Explanations

Loan Continuation Rates, conditional on eligibility

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 24 / 32

Robustness/Sensitivity

Treatment Effect on Non-Farm Income

Treatment Selection Sample Mean TRAIL GBL TRAIL GBL Size Control 1 Rental Income (Rupees) 153.6 784.4

• 182.1
• 427.9

4162 1508 Income from Animal Products (Rupees) 166.8 49.18 62.66

• 279.1

4162 771 Labour income (year; Rupees) 393

• 5642
• 12729**
• 4941

4162 37465 Wage employment (last 2 weeks; Hours) 0.615

• 4.496
• 6.855*

1.749 4162 40.24 Self-employment (last 2 weeks; Hours) 6.884 4.294 0.215 5.914* 4162 121.8 Reported profits (Rupees) 2343 2918 100.9

• 1917

4162 5802 Current value business (Rupees) 4917 6692 952.1 353.8 4162 10465 Total Non-Farm Income (Rupees) 3056

• 1890
• 12748
• 7565

4162 45546

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 25 / 32

Robustness/Sensitivity

Sensitivity of Treatment Effects to Price Fluctuations

Dependent Variable: Value added (Actual/Imputed) Treatment Selection Sample Mean TRAIL GBL TRAIL GBL Size Control 1 1 Actual 1687** 271.8 555.6

• 1371

2718 9498 2 2011 prices 1654*** 55.11 318

• 872.7

2718 8258 3 2012 prices 3187*** 500 254.8

• 1907

2718 14311 4 2007 prices

• 194.7
• 328.5
• 45.25
• 2744

2718 4423 5 2008 prices

• 1913**

1653 1079

• 2886**

2718

• 4434

6 2011 market wage 1672** 217.3 463.5

• 1483

2718 8219 7 2012 market wage 1665** 182.6 460.4

• 1416

2718 8134

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 26 / 32

Robustness/Sensitivity

Extraction by TRAIL Agent through Input Transactions

Sample Size Mean Control 1 Treatment Effect Bought any input from agent 12,448 0.0875

• 0.00338

Share of agricultural input purchased from agent 10,196 0.0760

• 0.00359

Input Price (Rs/unit) Inorganic fertilizer 1,672 13.78

• 0.322

Organic fertilizer 370 16.12 29.39 Outside seeds 1,654 22.36 2.174 Pesticide 2,691 533.5

• 31.08

Powertiller 1,403 195.2

• 32.33***

Water/irrigation 1,230 72.30 148.3

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 27 / 32

Robustness/Sensitivity

Extraction by TRAIL Agent through Output Transactions

Sample Size Mean Control 1 Treatment Effect Sold any output to agent 2,990 0.209 0.00559 Share of output sold to agent 2,765 0.151 0.0152 Output Price (Rs/kg) Potato 1,386 4.507

• 0.0516

Paddy 498 9.282

• 0.0215

Sesame 881 28.42

• 1.003

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 28 / 32

Robustness/Sensitivity

Extraction by TRAIL Agent through Credit Transactions

Sample Size Mean Control 1 Treatment Effect Borrowed from agent 1690 0.17

• 0.072*

Share of loan from agent 1690 0.05

• 0.034**

Interest Rate (APR) 5278 0.14

• 0.003

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 29 / 32

Financial Implications

Financial Sustainability

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 30 / 32

Financial Implications

Financial Sustainability

Administrative costs in TRAIL scheme were substantially lower: 5% of the costs in GBL scheme per month per village

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 30 / 32

Financial Implications

Financial Sustainability

Administrative costs in TRAIL scheme were substantially lower: 5% of the costs in GBL scheme per month per village

no monthly group meetings among TRAIL borrowers ⇒ lower loan officers’ salaries and transport expenses

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 30 / 32

Financial Implications

Financial Sustainability

Administrative costs in TRAIL scheme were substantially lower: 5% of the costs in GBL scheme per month per village

no monthly group meetings among TRAIL borrowers ⇒ lower loan officers’ salaries and transport expenses

However, lender retained entire interest earned on GBL loans, but paid 75%

• f interest on TRAIL loans as agent commissions

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 30 / 32

Financial Implications

Financial Sustainability

Administrative costs in TRAIL scheme were substantially lower: 5% of the costs in GBL scheme per month per village

no monthly group meetings among TRAIL borrowers ⇒ lower loan officers’ salaries and transport expenses

However, lender retained entire interest earned on GBL loans, but paid 75%

• f interest on TRAIL loans as agent commissions

Lender can break even on TRAIL scheme if has a low cost of loanable funds (⇡ 4% per annum)

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 30 / 32

Financial Implications

Financial Sustainability

Administrative costs in TRAIL scheme were substantially lower: 5% of the costs in GBL scheme per month per village

no monthly group meetings among TRAIL borrowers ⇒ lower loan officers’ salaries and transport expenses

However, lender retained entire interest earned on GBL loans, but paid 75%

• f interest on TRAIL loans as agent commissions

Lender can break even on TRAIL scheme if has a low cost of loanable funds (⇡ 4% per annum)

as in Bangladesh’s scheme for financing microlenders

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 30 / 32

Financial Implications

Summary

Designed and experimentally evaluated microcredit designed to finance smallholder agriculture of high-value cash crops

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 31 / 32

Financial Implications

Summary

Designed and experimentally evaluated microcredit designed to finance smallholder agriculture of high-value cash crops

borrowers recommended by local intermediaries who were incentivised through commissions dependent on repayment rates individual liability

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 31 / 32

Financial Implications

Summary

Designed and experimentally evaluated microcredit designed to finance smallholder agriculture of high-value cash crops

borrowers recommended by local intermediaries who were incentivised through commissions dependent on repayment rates individual liability

Evidence suggests TRAIL agents successfully selected productive, low-risk farmers

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 31 / 32

Financial Implications

Summary

Designed and experimentally evaluated microcredit designed to finance smallholder agriculture of high-value cash crops

borrowers recommended by local intermediaries who were incentivised through commissions dependent on repayment rates individual liability

Evidence suggests TRAIL agents successfully selected productive, low-risk farmers TRAIL borrowers expanded cultivation of potatoes, and their own incomes (RoR in excess of 70% at 2011-12 prices)

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 31 / 32

Financial Implications

Summary

Designed and experimentally evaluated microcredit designed to finance smallholder agriculture of high-value cash crops

borrowers recommended by local intermediaries who were incentivised through commissions dependent on repayment rates individual liability

Evidence suggests TRAIL agents successfully selected productive, low-risk farmers TRAIL borrowers expanded cultivation of potatoes, and their own incomes (RoR in excess of 70% at 2011-12 prices) GBL scheme induced eligible farmers to expand cultivation of potatoes and incur higher cost of production

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 31 / 32

Financial Implications

Summary

Designed and experimentally evaluated microcredit designed to finance smallholder agriculture of high-value cash crops

borrowers recommended by local intermediaries who were incentivised through commissions dependent on repayment rates individual liability

Evidence suggests TRAIL agents successfully selected productive, low-risk farmers TRAIL borrowers expanded cultivation of potatoes, and their own incomes (RoR in excess of 70% at 2011-12 prices) GBL scheme induced eligible farmers to expand cultivation of potatoes and incur higher cost of production but GBL borrowers had insignificant increases in output or farm income

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 31 / 32

Financial Implications

Summary

Designed and experimentally evaluated microcredit designed to finance smallholder agriculture of high-value cash crops

borrowers recommended by local intermediaries who were incentivised through commissions dependent on repayment rates individual liability

Evidence suggests TRAIL agents successfully selected productive, low-risk farmers TRAIL borrowers expanded cultivation of potatoes, and their own incomes (RoR in excess of 70% at 2011-12 prices) GBL scheme induced eligible farmers to expand cultivation of potatoes and incur higher cost of production but GBL borrowers had insignificant increases in output or farm income TRAIL scheme had (weakly) superior repayment rates, higher take-up rates, and significantly lower administrative costs

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 31 / 32

Financial Implications

Summary

Designed and experimentally evaluated microcredit designed to finance smallholder agriculture of high-value cash crops

borrowers recommended by local intermediaries who were incentivised through commissions dependent on repayment rates individual liability

Evidence suggests TRAIL agents successfully selected productive, low-risk farmers TRAIL borrowers expanded cultivation of potatoes, and their own incomes (RoR in excess of 70% at 2011-12 prices) GBL scheme induced eligible farmers to expand cultivation of potatoes and incur higher cost of production but GBL borrowers had insignificant increases in output or farm income TRAIL scheme had (weakly) superior repayment rates, higher take-up rates, and significantly lower administrative costs No evidence of extraction of TRAIL borrower benefits by agent

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 31 / 32

Financial Implications

Summary

Designed and experimentally evaluated microcredit designed to finance smallholder agriculture of high-value cash crops

borrowers recommended by local intermediaries who were incentivised through commissions dependent on repayment rates individual liability

Evidence suggests TRAIL agents successfully selected productive, low-risk farmers TRAIL borrowers expanded cultivation of potatoes, and their own incomes (RoR in excess of 70% at 2011-12 prices) GBL scheme induced eligible farmers to expand cultivation of potatoes and incur higher cost of production but GBL borrowers had insignificant increases in output or farm income TRAIL scheme had (weakly) superior repayment rates, higher take-up rates, and significantly lower administrative costs No evidence of extraction of TRAIL borrower benefits by agent Borrower benefits came from expansion of agricultural production, and therefore were sensitive to agricultural price fluctuations

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 31 / 32

Thanks

Thank you!

MMMMV (Dec 2014) Financing Smallholder Agriculture Dec 2014 32 / 32