Contract Design and Microfinance Performance: A Global Analysis

1
Contract Design and Microfinance Performance A
Global Analysis

• Robert Cull, World Bank
• Asli Demirgüç-Kunt, World Bank
• Jonathan Morduch, NYU
• Prepared for Access to Finance Building
  Inclusive Financial Systems,
• The World Bank
• May 30-31, 2006

2
The microfinance promise

• Promise To meaningfully reduce poverty without
  ongoing subsidies
• requires translating high repayment rates into
  profits
• challenge that remains for most microbanks
• Q Profitability-outreach trade-off?
• Q Role of high interest rates?

3
The MIX Data

• Financial Information on 124 institutions in 49
  countries
• Adjustments for consistency (2003 MicroBanking
  Bulletin)
• Institutions share a commitment to financial
  sustainability and transparency
• Our data cross-section (1observation per
  institution, 1999-2002).
• Seventy percent of the observations are from
  2002.

4
The MIX data

• Confidentiality
• 1. MFIs not identified by name2.
  Confidentiality maintained3. Analysis completed
  at World Bank

5
The MIX DataFinancials are adjusted for

• Inflation
• Subsidized costs of funds
• Cash donations to cover current operations
• In-kind subsidies
• Low loan loss reserves and provision expenses

• Reclassification of some long-term liabilities as
  equity
• Write-offs of non-performing assets
• Reversal of interest income accrued on
  non-performing loans

6
Good news

• Over half of the institutions were profitable
  after accounting adjustments were made
• Others are approaching profitability and should
  be able to soon achieve financial
  self-sufficiency.

7
Good news

• Simple correlations show little evidence of
  agency problems, outreach-profit trade-offs, or
  mission drift.
• The correlations attest to the possibility of
• raising interest rates without undermining
  repayment rates,
• achieving both profit and substantial outreach to
  poorer populations
• staying true to initial social missions even when
  aggressively pursuing commercial goals.

8
Subsidy continues as an important part of the
funding mix

• Subsidy gt 20 as average share of funding (total
  liabilities plus total equity)
• Average return on assets is negative overall.

9
Comparison of Outcomes by Type of Lending

• Types
• Individual-based lenders,
• Solidarity group lenders,
• Village Banks
• Examine the determinants of
• Profitability
• Loan Repayment
• Cost Reduction
• Explore whether profitability leads to mission
  drift

10
The Sample
11
Financial performance
Lending mode Individuals Solidarity Groups Village Banks
Financial self-sufficiency 111 98 95
Operational self-sufficiency 123 112 109
Subsidy 11 28 36
Return on assets 1 -5 -8
Age (years) 11 years 9 7
12
Outreach indicators
Lending mode Indivi-duals Solidarity Groups Village Banks
Average loan size () 1220 431 149
Avg loan size/GNP of bottom 20 4.8 1.6 0.6
women 46 75 88
Size rank 2.2 2 1.6
13
Profit and subsidy Return on assets, real gross
portfolio yield, and total expense ratio by
lending type
Real gross portfolio yield
Expense ratio
ROA
Individual
Solidarity group
Village Banks
14
Basic Equation

• Where
• FSS is financial self-sufficiency (also OSS and
  ROA)
• Yield is real gross portfolio yield
• Cost is two variables labor costs, capital costs
• MFI history Age and size
• Orientation profit status, loans/assets, avg
  loan size

15
Caveats

• Insufficient exogenous variation to reliably
  estimate causal impacts
• focus is thus on associations that frame debates
• Cross-section, across institutions, so no fixed
  effects (controls for omitted variables)
• Self-selected sample

16
Predicted Trade-Off Between Financial
Self-Sufficiency and Real Gross Portfolio Yield
Selection?
60
Portfolio quality? Demand?
17
Yield and Portfolio at Risk at 30 Days,
Individual Lenders
Agency problems?
Selection?
18
Predicted Trade-off Between Loan Size and Costs
Explains move to individual lending?
19
Conclusions Individual lenders

• In this top-end data set, substantial progress
  is seen toward financial self-sufficiency.
• The fastest progress is made by individual-based
  lenders
• They have the advantage of scale economies in
  lending larger loans
• But larger loans do entail higher relative costs
  after a point
• Portfolio risk management is especially crucial
• Success at managing portfolio risk at high
  interest rates
• Evidence of weakening profitability at high
  interest rates (consistent with falling demand
  but causality is unclear)

20
Conclusions Group lenders

• Deepest outreach (in terms of loan size, women
  clients) is by group lenders.
• Heavier reliance on subsidies
• Effective management of portfolio risk
• Less able to reap scale economies for individual
  borrowers

21
Conclusions Costs

• Higher labor costs are positively associated with
  profitability for individual lenders (and VBs).
• Larger loan size is associated with lower total
  costs for all lenders,
• But the minimum point on the cost curve for
  individual lenders occurs at a much larger
  average loan size.

22
Conclusions Mission drift

• For individual lenders,
• size (and age) associated with less outreach
• financial self-sufficiency is associated with
  greater outreach.
• For group lenders and village banks,
• outreach tends to not be associated with age,
  size, or financial self-sufficiency
• (except for large group lenders).

23
Conclusions Overall

• Much progress here
• Disaggregation shows a richer sense of trade-offs
  and possibilities
• Trade-offs likely to be clearer in larger data
  sets

24
Profitability Results Yields
25
Profitability Results II Capital Costs
26
Profitability Results III Labor Costs
27
Raising Interest Rates
28
Trade-Offs Costs and Average Loan Size
29
Mission Drift