Financial crisis: lessons from microfinance

Strat. Change 20: 187–203 (2011)Published online in Wiley Online Library(wileyonlinelibrary.com) DOI: 10.1002/jsc.895 RESEARCH ARTICLE

Copyright © 2011 John Wiley & Sons, Ltd.Strategic Change: Briefi ngs in Entrepreneurial Finance

Strategic Change DOI: 10.1002/jsc.895

Financial Crisis: Lessons from Microfinance1,2

Dinos ConstantinouMicrofi nance Strategy, Lausanne, Switzerland

Arvind AshtaBurgundy School of Business, Dijon, France

IntroductionIn the wake of the fi nancial crisis and the continued expansion and relatively better performance of microfi nance, a question which can be posed is whether there are lessons to be learnt from microfi nance for the economy in general and the banking sector in particular. While it would be tempting to say that microfi nance has all the answers, our research and experience suggests that microfi nance itself has its own booms and busts and that microfi nance crises merit their own specifi c analy-sis. Th e lessons of such crises are as much for microfi nance institutions (MFIs) as for the broader fi nancial sector.

How does a fi nancial crisis aff ect microfi nance? On the one hand, we observe rising ratios of non-performing loans (NPL) in a number of diff erent markets. On the other hand, microfi nance is essentially informal and typically directed to entrepreneurial activities that are not linked to the global economy (with the exception of cash crops and tourism-related micro-businesses) — which suggests that the impact of global economic events should not be signifi cant. Indeed, the microfi nance sector at large is continuing to experience growth where mainstream providers of fi nancial services have been retrenching in the context of the inter-national crisis. How can we reconcile all this? What can we learn from the current global fi nancial crisis and what have we learnt from past microfi nance crises, such as those in Russia (1998), Indonesia (1997–1998), Ecuador (1999–2000), Bolivia (1999–2001), and Argentina (1999–2002)? During the current crisis, why have

Microfinance has become a rapidly expanding industry in many emerging economies.

Similar to financial crisis in the aftermath of a global economic boom, microfinance is also a victim to its own runaway success.

A global economic crisis increases the vulnerability of micro-entrepreneurs and, at the same time, could dry up the fund sources for MFIs. The lessons for the success of MFIs are similar to those for the broader financial sector.

Microfinance institutions require funding more rather than less during

periods of financial crisis.

1 JEL classifi cation codes: G20, G21.2 Th is paper is an evolution of Constantinou and Ashta (2010). We thank the editor Carlo Milana for constructive suggestions for this evolution.

188 Dinos Constantinou and Arvind Ashta

Copyright © 2011 John Wiley & Sons, Ltd. Strategic Change DOI: 10.1002/jsc

delinquency rates shot up in some countries — e.g. Nicaragua, Morocco, Bosnia & Herzegovina and Pakistan — far more than in others?

Background: the development of microfinanceTh e success of the microfi nance movement and high reim-bursement rates by micro-entrepreneurs and other low-income borrowers has been broadly discussed in the literature (Yunus, 2003; Armendàriz and Morduch, 2005; Ashta, 2009). High re-imbursement rates are based on a range of (often innovative) solutions, such as analytical credit methodologies tailored for informal borrowers;3 group/solidarity lending mechanisms; small but frequent repayment installments; non-traditional collateral; regular monthly savings as a prerequisite for receiving loans; bor-rower incentives; targeting of typically more reliable women borrowers; and the use of new sources of informa-tion (Bernasek and Stanfi eld, 1997; Bhatt and Shui-Yan, 2001). Regulated fi nancial institutions can then rely on the credit history developed over time by borrowers in the (usually less regulated) microfi nance sector to off er follow-up loans. In countries like India and Mexico, this usually involves a migration from group to individual lending, a development also fuelled by the diverging business needs of micro-borrowers depending on the degree of success of their businesses and other factors.

Th e success of microfi nanced enterprises is compara-ble to the success of small and medium enterprises fi nanced by traditional banks (Yunus, 2003; Labrune, 2010). While Yunus ascribes this to inborn entrepreneurial skills, it may

be that group attention or the support services provided by NGOs to low-income micro borrowers also play a role. For example, some Indian MFIs rely on their fi eld workers to disseminate not only loans but also practical training as well as to organize work sharing groups (Harper et al., 2008). Th e provision of associated services in microfi nance (usually described as ‘business development services’ in industry jargon) may also help to overcome adverse selec-tion and moral hazard. Th e role of religious institutions in ensuring repayments has also been suggested (Harper et al., 2008). Finally, it may be that MFIs are successfully selecting the micro-enterprises with the best business models and highest chances of success — in turn establish-ing a virtuous cycle, whereby ‘deserving’ borrowers add access-to-funding to their panoply of competitive advan-tages vis-à-vis less competent, rival micro-businesses.

Th e impact of microfi nance on poverty development is still subject to widespread debate (Copestake et al., 2001; Chowdhury et al., 2005; Roodman and Morduch, 2009). Whether microfi nance can play a role in poverty reduction (Daru et al., 2005) or not, policy-makers remain interested in enhancing access to fi nance for the poor — with many governments passing specifi c micro-fi nance legislation to achieve the latter goal over the last two decades (Meagher, 2002; Attuel-Mendes and Ashta, 2008; Ashta and Fall, 2009).

Microfi nance has grown rapidly over the last decade to become a global $40 billion industry by 2007 (Figure

1). Moreover, the sector has been transformed by the expansion of MFIs from pure micro-lending to other fi nancial products such as micro-savings, micro-insurance and remittances.

Th ere were estimated to be more than 10,000 MFIs worldwide in 2003, of which less than 5% (i.e. 500 institu-tions) were profi table. Th e main factor behind profi tability was scale, with this elite group of MFIs providing the bulk of fi nancing to the microfi nance sector. Table 1 shows that the number of borrowers grew at an average rate of 21% per annum, with gross loan portfolios growing even faster at an average rate of 34% per annum (i.e. loans sizes

3 Micro lending methodologies often involve visits to the business premises and home of prospective borrowers by MFI loan offi cers to ascertain both ability and willingness to pay. Advanced MFIs operating on the individual lending paradigm typically train their loan offi cers to construct rough fi nancial statements for the business and the family unit from (informal and formal) information acquired during such visits, thus allowing mainstream credit analysis tools to be deployed in the evaluation of informal businesses.

Lessons from Microfinance 189


Sources: The MIX; own estimates. N.B. Chart not drawn to scale

20

30

15

10

5

0

25

$ billion

19941984 2007Year

$ 2.5 B

$ 43 billion

Δ 15-30% p.a.

(average for

entire period)

$ 0.2 B

+20,000%

Figure 1. Rapid growth of the microfi nance sector, 1984–2007.

Table 1. High growth rates of MFIs

Annual growth rates (%) 2003 2004 2005 2006 2007 2008 Compound average

Number of borrowers 14.5 21.0 26.5 22.0 22.0 20.0 21.0Gross loan portfolio ($) 32.8 37.8 32.4 38.1 45.4 20.1 34.2Number of savers 13.2 11.2 12.1 11.7 1.5 25.9 12.4voluntary savings ($) 33.9 22.9 13.8 29.2 31.9 16.8 24.5Source: Gonzalez (2008), updated on December 31, 2009.

increased on average across the sample). Between them, these MFIs served 86 million borrowers with a gross loan portfolio of $44 billion and 96 million savers with total savings of $16 billion.

While MFIs reporting to the MIX (Microfi nance Information Exchange) are likely to be the largest and most profi table, our own analysis based on data provided to the MIX by 1,395 MFIs during the period 2003–2008 shows that an ever-increasing, leading pack of MFIs is sustainable or nearing sustainability (although a deeper study would take into consideration the impact of subsidies, infl ation, etc.). As Table 2 shows, around 70% of MIX-reporting MFIs show positive profi tability in terms of return on equity (ROE) in any year. Th us, 45% of the reporting MFIs

show profi tability of more than 10% and a third show profi tability of more than 15%. A quarter of the reporting MFIs are indicating profi tability of more than 20% and a few are showing profi tability of more than 30%.

Figure 2 graphs the profi tability of these MFIs over the period 1997–2009. Th e results suggest that the per-centage of profi table MFIs remained stable in 2008 and 2009, although it was perhaps still too early to see the full impact of the crisis on MFI performance at that time.

Gonzales (2009) indicates that the write-off ratio until 2008 was 1.7%, while portfolio-at-risk over 30 days (PAR30) was 6.7%. PAR30 represents total outstanding loans of clients falling behind repayment schedule for a period exceeding 30 days, as a percentage of total gross



Tab

le 2

. Pr

ofi ta

bilit

y of

MIX

-rep

ortin

g M

FIs,

1997

–200

9

RO

E of

MFI

s19

9719

9819

9920

0020

0120

0220

0320

0420

0520

0620

0720

0820

09A

vera

ge

Tota

l MFI

s rep

ortin

g to

MIX

2768

101

136

185

285

462

673

865

998

1053

1170

1033

Tota

l with

pos

itve

ROE

2140

6389

131

196

313

487

646

748

800

881

749

Perc

enta

ge78

%59

%62

%65

%71

%68

%68

%72

%75

%75

%76

%75

%73

%71

%To

tal w

ith R

OE

< 5%

1730

5175

9715

724

941

452

163

166

472

457

6Pe

rcen

tage

63%

44%

50%

55%

52%

54%

54%

62%

60%

63%

63%

62%

56%

57%

Tota

l with

RO

E <

10%

1516

3858

7212

120

034

142

150

954

556

243

4Pe

rcen

tage

56%

38%

38%

43%

39%

42%

43%

51%

49%

54%

52%

48%

42%

45%

Tota

l with

RO

E <

15%

1217

2639

5691

137

247

340

399

414

434

310

Perc

enta

ge44

%25

%26

%29

%30

%31

%30

%37

%39

%40

%39

%37

%30

%To

tal w

ith R

OE

< 20

% 7

1320

3047

6810

018

225

732

131

133

622

8Pe

rcen

tage

26%

19%

20%

22%

25%

24%

22%

27%

30%

32%

30%

29%

22%

25%

Tota

l with

RO

E <

30%

4 4

1317

2739

5397

156

181

173

212

122

Perc

enta

ge15

% 6

%13

%13

%15

%13

%11

%14

%18

%18

%16

%18

%12

%14

%

Tab

le 3

. M

IX-r

epor

ting

MFI

s with

hig

h lo

an q

ualit

y

1997

1998

1999

2000

2001

2002

2003

2004

2005

2006

2007

2008

2009

Ave

rage

Num

ber o

f MFI

s rep

ortin

g40

8212

317

424

036

760

475

092

910

7010

6711

8110

44N

B PA

R 3

0 <

5%30

5981

116

157

232

359

478

605

679

697

725

546

Perc

enta

ge75

%72

%66

%67

%65

%63

%59

%64

%65

%63

%65

%61

%52

%65

%N

B PA

R 3

0 <

3%22

4562

9313

017

929

437

348

352

254

152

838

0Pe

rcen

tage

55%

55%

50%

53%

54%

49%

49%

50%

52%

49%

51%

45%

36%

49.8

%N

B PA

R 3

0 <

2%18

3154

7710

114

224

232

038

743

043

742

527

6Pe

rcen

tage

45%

38%

44%

44%

42%

3940

%43

%42

%40

%41

%36

%26

%40

%N

B PA

R 3

0 <

1%10

1840

5678

101

170

218

284

328

298

276

181

Perc

enta

ge25

%22

%33

%32

%33

%28

%28

%29

%31

%31

%28

%23

%17

%28

%



Figure 2. Evolution of profi tability of MIX-reporting MFIs, 1997–2009.

portfolio. Th e independent policy and research center Consultative Group to Assist the Poor (CGAP) estimates that PAR30 doubled during the fi rst fi ve months of 2010 for a smaller sample of around 200 high-performance MFIs (Reille et al., 2010).

Our analysis of MIX data for portfolio-at-risk (PAR), provided in Table 3, indicates that PAR is less than 5%

for 65% of the MFIs in any year. On average, half the MFIs report a PAR of less than 3%, 40% report a PAR of less than 2% and a quarter of them report a PAR of less than 1%.

Figure 3 provides a graphical representation of the low PAR30 MFIs. Conversely, it can be interpreted that 35% of MFIs having a PAR of greater than 5% must be

Figure 3. Number of MFIs reporting PAR30 to MIX.



Sources: MIX 2008 Analysis & Benchmarking Reports N.B. Chart not drawn to scale

US $ million

EECA LAC

13,700

15,292

10,000

0

15,000

5,000

Asia

10,250

SSA

2,236

MENA

1,039

Total:

$ 43 billion

Abbreviations:

EECA = Eastern Europe & Central Asia; LAC = Latin America & Caribbean; SSA = Sub-Saharan Africa; MENA = Middle East & North Africa

Figure 4. Distribution of loan portfolios between regions in 2007.

fairly risky. Table 3 brings out that PAR30 has been dete-riorating for MFIs in 2008 and further in 2009. For example, the percentage of MFIs with PAR30 > 2% reduced from 41% in 2007 to 36% in 2008 and further down to 26% in 2009.

Th e most developed microfi nance markets, in terms of dollar loan portfolio, were to be found in Eastern Europe/Central Asia and Latin America (see Figure 4).

Crisis and microfinanceIn this section we fi rst provide a small note on fi nancial and economic crises more generally, followed by a brief analysis of how fi nancial crisis would impact MFIs a priori. We then look at what really happened in earlier crises before considering the ongoing eff ects of the current fi nancial crisis.

Economic crisisIf MFIs have indeed been relatively immune to interna-tional crisis, it may be of interest to understand the factors which create fi nancial crises in the fi rst place to determine why they do not have a similar impact on microfi nance. According to Bruner and Carr (2007), for example, the fi nancial crisis of 1907 originated in the complexity and interdependence of the fi nancial system. Asymmetry of information — in the sense that few people understood exactly what each fi nancial transaction entailed — was behind the spread of panic across US fi nancial markets. Linkages between institutions, moreover, can have snow-balling eff ects with international repercussions.

Th e Financial Crisis Inquiry Commission (2011) has described how the recent fi nancial crisis was triggered by the collapse of the housing bubble in 2007 and amplifi ed by increasingly globalized markets, where effi cient but



complex instruments render fi nancial transactions techni-cally easy and straightforward across the world. Th e fi nan-cial sector itself has become a much more dominant economic force than in the past.

High growth typically precedes a slump, either because of increased demand for liquidity choking-off growth or because there are too many competitors coming in and demand does not justify all the new entrants. In times of economic expansion banks increase lending by accepting less credit worthy borrowers. Th e resulting higher risk profi le of loan portfolios leads banks to retrench (by accel-erating loans to borrowers who no longer satisfy loan agreement covenants and by not extending loans on matu-rity) as they realize that they are overstretched (when the boom ends and the value of the collateral falls). Bank retrenchment, in turn, exacerbates the liquidity crisis in the real economy.

Behavioral reasons for fi nancial crises are usually fear and greed. Collective action led by a responsible and recognized leader may abate the crisis (Bruner and Carr, 2007) — as happened in 1907 with the system-wide intervention of private sector banks led by J. P. Morgan. Such a leader could also be the government acting in its capacity as a lender of last resort, if private sector partici-pants cannot identify solvent (illiquid) banks or are only willing to assume the risk of lending to them at prohibi-tively high rates (Flannery, 1996).

Model: how an economic crisis could impact microfinanceEach MFI is linked to the economy across the fi nancing value chain: upstream, microfi nance investment vehicles (MIVs) provide funds to MFIs; while downstream, MFIs lend to poor entrepreneurs. To answer our question, we must consider the impact of the fi nancial crisis on both sides of the fi nancing value chain.

Upstream, MFIs were essentially donor fi nanced in the early days of the microfi nance industry. However, over the last few years, MFIs have increasingly accessed fi nanc-

ing through commercial banks and MIVs. Th is is espe-cially true of the large profi table institutions that dominate the microfi nance sector. A fi nancial crisis could impact both the availability of donor funding as well as the com-mercial funding because MIVs become reluctant to take risks.

However, many MFIs (mainly regulated fi nancial institutions) are now fi nanced by the savings deposits of the poor. As a result, they should be fairly immune from the international fi nance nature of this economic crisis. Th is would explain why microfi nance, to some extent, continues to grow unabated.

Downstream, on the lending side, MFIs lend essen-tially to small traders, artisans and service providers, often in the informal economy. Th ese traders should also be immune from the fi nancial crisis, except to the extent that their demand is linked to the global economy (cash crops, tourist market).

Th e basic business model of large, profi table MFIs should allow them to weather a possible decrease in funding, provided that the scale of operations does not fall below breakeven levels as a result. At the same time, donors in particular have so far tried to maintain their levels of fi nancing to the microfi nance sector; and even (through special facilities channeled via international MIVs) to increase such funding to make up for the possible retreat of private sector money. Indeed, much of the slowing market may be due to demand on the part of micro-entrepreneurs.

Th e basic model of the MFI, explained earlier, is weakened when repayment installments by micro-bor-rowers are essentially being refi nanced through new, larger loans — themselves funded by ever-growing donor loans similar to a pyramid/Ponzi scheme. Providing ever-increasing loans to clients, per se, is not necessarily a bad thing. If the borrower is using the funds to fi nance a growing business with increasing working capital needs and can generate the cash fl ows required to service the loans, the result is virtuous expansion for both the micro-entrepreneur and the MFI.



Economic and financial crisis

Financial and banking crisis (collapse of

national currency; freezing of banking

accounts; dollarization)

Financial and banking crisis (default on

Russian government bonds and

devaluation of rouble)

Economic crisis; abandonment of dollar

parity; collapse of banking system;

default on government bonds

Delinquency crisis in microfinance

sector, exit of consumer lenders

Near-collapse of Banco Solidario,

retreat of most commercial banks

Collapse of most commercial banks

working with EBRD but resilience of

microfinance portfolios

Collapse of largest Argentine MFI

(Fundación Emprender)

Type of Crisis Impact on Microfinance

Ecuador

1999-2000

Russia

1998

Bolivia

1999-2001

Argentina

1999-2002

Financial and banking crisis (collapse of

national currency and commercial banks)

Over-lending and near collapse of rural banks – but continued growth and

resilience of savings-based microfinance

Indonesia

1997-8

Figure 5. Impact of crisis on microfi nance — examples from the late 1990s

However, matters are complicated when external events (such as the economic crisis) impact micro-businesses and render their working capital illiquid. Th e wisdom of refi nancing in these cases is more contentious, depending on whether problems faced by the micro-enterprise (e.g., on the demand side) are of a temporary or long-lasting nature. MFIs are not always equipped to make this judgment and are therefore reluctant (especially in the initial stages of the crisis) to restructure or refi nance any loans whatsoever. In this respect, loans to poor people for their enterprises does not diff er from loans to poor nations (Armendàriz de Aghion, 1993).

Th erefore, whether entrepreneurs are impacted or not by the crisis, it is diffi cult for an MFI to stagnate: there seems to be an inbuilt snowballing eff ect. High average interest rates of 28% (Rosenberg et al., 2009) make it even more important (than in mainstream credit markets) for micro-enterprise borrowers to maintain the margins and sale volumes required to service their loans. At the same time, MFIs are subject to similar economies of scale as mainstream fi nancial institutions (e.g., due to the impor-tance of information systems in managing a large client base), so that reduced lending volumes combined with rising NPL ratios can have an immediate impact on profi tability.

Impact of previous economic crises on microfinanceTh e current fi nancial crisis comes after a decade of rapid growth in microfi nance worldwide. However, it is not the fi rst of its kind to strike the nascent microfi nance indus-try: shocks and crises had already taken their toll on microfi nance sectors around the world at the turn of the millennium, in the follow-up to the East Asian crisis of 1997 to 1998.

Figure 5 outlines the type of crisis and its impact on each of fi ve microfi nance markets: Russia, Indonesia, Ecuador, Bolivia and Argentina. In this section, we look more closely at the cases of Russia, Indonesia and, espe-cially, Bolivia.

Russia, 1998Th e micro and small business fi nance sector that devel-oped in Russia from 1995 onwards, under the tutelage and with funding of the European Bank for Reconstruc-tion & Development (EBRD), was characterized by exceptional loan quality and high (relative to loan size) portfolio yields. Th e entire sector came near to demise as the private-sector, commercial banks providing fi nance to micro and small enterprise in the context of the EBRD Russia Small Business Fund collapsed under the weight of



defaulted government obligations (GKOs). Notwith-standing the collapse of the banks, micro/small loan port-folios transferred to a specialized bank established by the EBRD showed remarkable resilience (in terms of repay-ment) to a crisis that shook the entire Russian economy. Th is example shows that prudent micro (and small busi-ness) fi nance can resist the brunt of even acute fi nancial crisis. Th ere are other examples (including the growth of Compartamos in Mexico) that lend credence to the view that in times of fi nancial crisis, when bankers stop lending to the poor, the latter are forced to turn towards money-lenders or (where possible) to MFIs. Th is is the basis of the view that microfi nance may not only have a low beta — but the beta may even be negative in some countries.

Indonesia, 1997–1998Th e Asian fi nancial crisis which impacted Indonesia in 1997–1998 was caused by factors such as overvalued exchange rates, structural weaknesses in the fi nancial sector and excessive short-term borrowing, leading to asset price infl ation, speculation and increases in non-perform-ing loans (McGuire and Conroy, 1998). Th e impact of the crisis was especially strong as the country also wit-nessed a drought in the same period. Falling exchange rates led to increasing repayment diffi culties on foreign currency loans for banks at the same time as their asset performance declined. While the government stepped in to guarantee deposits for commercial banks, rural banks nearly collapsed as they could not access the same guar-antees. Within the MFI sector, bigger loans were given to fewer borrowers (McGuire and Conroy, 1998).

Both the microfi nance unit of Bank Rakyat Indonesia (BRI) (Patten et al., 2001; Seibel et al., 2011) and village-based MFIs — the Lembaga Perkreditan Desa (Seibel and Nurcahya, 2009) — showed remarkable resilience and even growth. In fact, savings deposits (number of accounts and amounts) increased at the same time as loan portfolios stagnated. While the corporate, small and medium-sized enterprises and personal loan portfolios of BRI showed

high default rates, microfi nance loans continued to enjoy low default rates (Patten et al., 2001).

Bolivia, 1999–2001Bolivia in 1999–2001 off ers possibly the most instructive case study of crisis in the microfi nance arena. Marconi and Mosley (2006) suggest that while the microfi nance sector played a ‘notably counter-cyclical role’ in Indonesia in the late 1990s, in Bolivia it mostly intensifi ed the broader economic crises. Before considering the Marconi–Mosley argument, we look briefl y at the historical devel-opment of the Bolivian microfi nance sector.

Bolivian microfi nance evolved rapidly from the 1980s onwards under the combined infl uence of regulatory change and economic crisis (see Figure 6). Th e analysis below focuses on the four leading microfi nance players, of which one was a regulated bank (BancoSol) and three others were regulated trusts or Fondos Financieros Priva-dos (FFPs), namely Caja Los Andes, PRODEM and FIE.

Economic diffi culties, intense competition and inad-equate lending practices led to a deep crisis in the micro-fi nance sector, in the midst of the broader contagion of the Bolivian economy from the international fi nancial and economic crisis. Figure 7 illustrates the key drivers of the Bolivian microfi nance crisis, namely: rising competition among existing microfi nance players and new competi-tion from consumer lenders entering the microfi nance sector; falling micro-enterprise sales due to decline in internal demand; and demonstrations by borrower asso-ciations against microfi nance and consumer lenders.

As a result of the crisis, loan delinquency (as measured by PAR30) increased by an average of 4% (i.e., from less than 5% to almost 9%) over 1999–2001 for the four main microfi nance players (i.e., excluding consumer lenders); while the number of clients served by the same MFIs fell from over 180,000 in 1999 to little over 145,000 in 2002, a loss of some 35,000 clients.

By way of comparison, the Bolivian fi nancial system as a whole lost 220,000 (or over 30%) borrowers in the period 1999–2002, while average loan delinquency (again



Key Drivers of Microfinance Crisis Impact on “Big-4” Bolivian MFIs

Aggressive cross-lending due to:

- Rising competition among microfinance players

- Entry of consumer lenders with inadequate credit technology (application of “credit factory” model to microenterprise)

Fall in sales of most small enterprises

due to fall in internal consumption

Demonstrations by borrower

associations against both consumer

and micro-lenders in favour of debt

forgiveness

1. Falling client numbers

2. Rising delinquency

Sources: Arriola 2003; Rhyne 2001 “Big-4”: BancoSol, FFP Caja Los Anders, FFP Prodem, FFP FIE

Figure 7. Microfi nance crisis in Bolivia, 1999–2001 — key drivers and impact.

N.B. Chart not drawn to scaleSources: Arriola 2003, Rhyne 2001

1980 1990 1996-1998 1999-2002

NGOs (micro credit only)

• Transformation

of existing MFIs

into FFPs

• New FFPs

Est. of BancoSol

(1992)

FFP Decree (1995)

Entry of

consumer

lenders

Microfinance

delinquency

crisis and exit of

consumer

lenders

Deepening economic

crisis (1999)

Recovery

and

renewed

strong

growth

Hyperinflation & banking

crisis (1984-5)

Radical

reform of

banking system

2003-

Figure 6. Stages in the evolution of microfi nance in Bolivia.



measured as PAR30) of the fi ve largest, traditional Boliv-ian banks reached 17% at the end of the same period. Th e delinquency crisis had far-reaching consequences for both the microfi nance segment and the broader Bolivian credit market. Consumer lenders entering the market in 1996–1998 had virtually abandoned the micro-enterprise segment only two years later. Th e crisis thus marked the failure of the ‘credit factory’ model deployed by consumer lenders in micro-enterprise and informal borrower segments — and the market exit or outright collapse of hitherto fast-growing consumer lenders (e.g., Acceso, CrediÁgil, Banco Económico) (see Rhyne, 2001).

At the same time, we fi nd limited support for the Marconi–Mosley thesis once we account for the diff erence between consumer lenders and MFIs. Although the number of clients served by the main MFIs declined during the crisis, the volume of outstanding loans continued to increase in each year of the crisis for the regulated microfi -nance sector from $141 million before the crisis, in Decem-ber 1998, to $206 million by the end of 2001 (Marconi and Mosley, 2006). Th is suggests that the main microfi nance lenders focused their fi nancial support on their best clients while weaning their portfolio of problem borrowers, thus providing an important support to micro and small busi-nesses in Bolivia in the midst of the economic crisis.

Th e loan portfolio of commercial banks, however, declined from $4,023 million in December 1998 to $2,769 million in December 2001. Th at the loan portfo-lio of consumer-credit houses FFP Acceso and FFP Fassil declined from $109 million in December 1998 to $20 million in December 2001 is not surprising, since the pro-cyclical nature of retail (consumer) lending is well known and was exacerbated in the Bolivian case by the application of inappropriate retail lending methodologies to provide consumer fi nance to informal borrowers.

Impact of the current economic crisis on microfinanceTh e previous crisis probably strengthened the microfi -nance sectors and limited the impact of the current fi nan-

cial crisis. However, the current crisis diff ers from previous crisis as it is not an internally generated crisis but an international one and may impact countries in diff erent ways (Llanto and Badiola, 2009). One complication of the 2009 crisis has been the inclusion of a food crisis with price increases, which may aff ect the deposit capacity of MFI clients (Littlefi eld and Kneiding, 2009).

CGAP undertook a four-country study of increasing delinquency (Chen et al., 2010) in Pakistan, Morocco, Bosnia and Herzegovina and Nicaragua, where PAR exceeded 10% or was coupled with aggressive write-off s. Th e study found that the delinquency crisis aff ected MFIs across the board in three countries (although the extent of this impact diff ered widely among MFIs) and a few large ones in Pakistan. In two of these, Nicaragua and Pakistan, borrowers collectively decided to not reimburse loans.

Th e CGAP study suggests that there are three main factors which led to increased delinquency: concentrated market competition and multiple borrowing; overstretched MFI systems and controls; and erosion of MFI lending discipline. Th ese factors also emerge as fundamental drivers of the microfi nance crisis in our analysis of both Bolivia in 1999 and several Eastern European markets in the current context. Moreover, in all four countries exam-ined by the CGAP paper, links with external funding were strong, with international funding playing a major role in Nicaragua and Bosnia and Herzegovina, and internal micro savings were not developed and remained less than 10% of the portfolio, compared to a global average of over 40%. Th e international economic crisis was considered an aggravator and not the primary cause of the crisis; our analysis of the Bolivian crisis would broadly confi rm this conclusion, with the international fi nancial and economic crisis acting as a trigger rather than as the fundamental cause of the problems faced by specialized microfi nance and other players entering the microfi nance market in the course of the 1990s.

Each of these factors has behavioral implications. High repayment by microfi nance borrowers, especially in



individual lending, often depends on the ‘carrot’ of receiv-ing larger loans from the same MFI in the future. Com-petition, however, erodes this dependence since the borrower can access parallel loans from other MFIs. At the same time, high MFI growth rates stretch the capacity of experienced loan offi cers to manage ever increasing client portfolios, leading to less personal interaction between credit staff and borrowers and lapses in follow-up and monitoring. A further element is the diminished credit discipline that comes with moves to increase staff productivity, reduce transaction costs and enhance the attractiveness of the credit product in a context of rising competition and falling interest rates.

For example, by reducing face-to-face contact, the increased use of technology diminishes the potential for inter-personal trust building. Similarly, growing use of monthly as opposed to weekly repayments increases the onus on loan offi cers to evaluate correctly the micro-entrepreneur’s cash fl ows, managerial ability and charac-ter. Th e reason is that borrowers with short-term horizons fi nd it both more diffi cult and less appealing to service the resulting, far-larger monthly installments; while less frequent installments mean that loan offi cer and MFI are unable to respond as swiftly to repayment problems. A well-known mantra in microfi nance is that time is a key factor in assuring the collection of non-performing loans, far more so than in traditional lending, due to the typi-cally lower level of fi xed assets and higher mobility of micro-borrowers.

Retail credit (i.e., personal loans to employees) con-stitutes an important part of MFI product lines in some regions (e.g., Eastern Europe). In Bosnia and Herze-govina, for example, housing and consumer loans to employees (as opposed to micro-entrepreneurs) may have represented anything between 20% and 40% of loan port-folios across the microfi nance sector. Exposure to retail lending to employees radically alters the risk profi le of MFI portfolios, increasing the likelihood of losses signifi -cantly over the economic cycle [Rhyne (2001), for example, suggests expected loan delinquency of 3% to 5%

for microenterprise credit as opposed to 15–20% in a Bolivian context]; increasing correlation with the main-stream economy; and making demands on MFI processes and systems that they are ill-equipped to meet.

A further important channel for risk to fl ow from the mainstream international economy to MFIs is that of remittances. Remittances are an important source of cash fl ow for micro-borrowers in several countries — and have generally suff ered as a result of international crisis. Th us although micro-enterprise (and informal small business fi nance) may have low correlation with global economic crises, family units with high dependence on remittance cash fl ows may well face payment diffi culties in a general-ized crisis (Littlefi eld and Kneiding, 2009).

Th ese two channels — retail credit to employees and remittances — imply a fundamental change in the credit risk equation for MFIs and diminish the insulation of the microfi nance sector from the mainstream and interna-tional economy. Th us it is the departure of MFIs from classical micro-enterprise lending models that increases correlation rather than faults in earlier analytical thinking that considered (classical) microfi nance largely immune to mainstream economic and fi nancial crisis.

We should also mention here that the Bolivian experi-ence of 1999 to 2001 (and perhaps the example of Paki-stan in the current crisis) suggests that group lending is more likely to suff er during economic crisis through con-tagion from bad to good borrowers within the group. In the Bolivian example the negative impact of the crisis on group lenders led to them eff ectively abandoning the soli-darity lending model for the more sophisticated individ-ual lending paradigm (BancoSol, for example, today operates largely as a provider of individual micro loans, whereas group credit represented the lion’s share of its loan portfolio prior to the 1999 crisis). It remains to be seen, therefore, whether the current crisis has a similar impact on microfi nance sectors in Pakistan (a development that can be followed through the evolution of at least one major Pakistan MFI) but also on other booming micro-fi nance markets such as India and Mexico.



In particular, the question is whether the contagion/acceleration impact of group lending can be counter-balanced by some of the inherent advantages of groups such as information sharing and reduced transaction costs of lending — and to what extent such advantages may be of more importance in specifi c market contexts. Th ese considerations make for a more complex analysis and render a complete assessment of group versus individual lending models more diffi cult.

A further consideration concerns the contribution of sources of funding to institutional resilience. For example, a study of savings-based (as opposed to donor-dependent or foreign capital-dependent) MFIs operating on the vil-lage-banking model in Bali, Indonesia, suggests that the fi nancial crisis has not had a signifi cant impact and that savings as well as loans have continued to grow (Seibel and Nurcahya, 2009). In contrast to CGAP’s fi ndings, the Bali study fi nds that credit and liquidity risks are not the two biggest risks among savings-based MFIs.

Lessons from microfinanceLeading microfi nance players, however, weathered the storm in Bolivia and in the world more generally through a combination of internal strategies and external coopera-tion. Th is section lists some of the methods used (based on Patten et al., 2001; Rhyne, 2001; Arriola Bonjour, 2003; Navajas et al., 2003; Vogelgesang, 2003; Littlefi eld and Kneiding, 2009; Khan and Ashta, 2011). Our recommendations have been classifi ed into fi ve sections: institutional environment; corporate strategy; human resource management; operations; and credit methodology.

Institutional environmentMicrofi nance crises off er a valuable opportunity to infl u-ence the political agenda and achieve progressive regula-tory reforms. Th e risk of course lies in regulation running amok, as has perhaps been the case in the aftermath of economic and fi nancial crisis in the past.

Th e specifi c danger in microfi nance at this time is for donor-promoted measures that in principle are aimed at protecting borrowers (e.g., initiatives to promote respon-sible and transparent pricing) actually placing MFIs at a competitive disadvantage to far less benign retail lenders and loan sharks; or, even more seriously, encouraging populist propagandists and politicians to promote anti-market tools (such as interest rate caps) with the potential to strangle nascent microfi nance sectors.

Meaningful reforms, however, are those aimed at cur-tailing over-lending — such as the creation of credit bureau to enhance the fl ow of credit information between MFIs. Such information fl ows can avoid the perils of multiple lending to the same borrower by several MFIs, in particular over-indebtedness, without requiring the outright prohibition of multiple loans (which would probably have the negative side eff ect of diminishing com-petition). Legislation to establish a collateral registry for movable goods can be benefi cial provided such new bodies can avoid becoming nests of bureaucratic ineffi ciency and corruption.

Last but not least, crisis off ers a rare opportunity for consolidation in the microfi nance sector, with mergers and acquisitions becoming possible as economic pressures override personal egos and other obstacles.

Corporate strategyIn the case of Bolivia outlined earlier, microfi nance players moved into the gap vacated by mainstream fi nancial inter-mediaries by expanding into new product and geographi-cal market segments. Th is included increased presence in small and medium-sized enterprises, retail and rural credit markets — as well as the establishment of branches in areas of higher social strata and traditional banking areas to capture deposits and off er credits products and other services.

Th e crisis brought about or accelerated the following developments at Bolivian MFIs on the strategy front, which serve as possible recommendations for MFIs elsewhere:



• Higher diversifi cation of credit products• Increased development of non-credit products and ser-

vices, especially savings• Introduction of diff erent savings products providing

incentives to save for family security, emergencies and investment in the business

• Improvements in client service• Greater emphasis on marketing and advertising

Human resource managementSimilar to earlier, a series of measures were pursued in the Bolivian context to enhance the quality of human resources available to the microfi nance sector. Th ese included:• Reformulation of credit staff performance targets

including relaxation of quality parameters and refocus of incentives more narrowly on collection

• Shift from social to economic science education back-grounds (to accommodate move from group to indi-vidual lending)

• More staff with banking experience, especially in man-agement positions

• Increased focus on internal training• Introduction or refi nement of well-designed, results-

based incentive schemes for diff erent staff levels

OperationsMeasures to enhance administrative and operating effi -ciency become increasingly important under pressure from falling loan portfolio yields and rising delinquency. Opportunities for improvement can be found on both the productivity and cost management front.

Productivity improvements often require investment in appropriate management information system (MIS) and other technologies. A streamlined loan process can yield signifi cant benefi ts, enhancing loan productivity and reducing credit risk. For example, a lean and well-defi ned loan approval process can improve decision-making by avoiding the dilution of responsibility, while at the same time boosting effi ciency. Other operational improvements include the standardization of loan agreements and other

aspects of the loan process, including where possible loan conditions and amounts. Seemingly insignifi cant optimi-zation, such as repayments in round sums to avoid the inconvenience of small change, can yield large benefi ts over the large number of operations and installments typical of microfi nance.

Cost management works well when a non-money wasting philosophy fl ows from the top management downwards — as typifi ed by the top managers of leading players in low-cost aviation such as Ryanair and South-west Airlines.

Examples of cost reduction strategies include the introduction of highly decentralized approval systems with delegation of authority and responsibility accompa-nied by manuals to enable standardized decision-making; cost-eff ective staff recruitment and on-the-job training; simplifi ed and easy accounting and record keeping process; and the provision of quick service through the reduction of paper work and bureaucratic procedures.

Credit methodologyLast but not least, changes and improvements in credit and collection methodologies were crucial in the Bolivian and other microfi nance crises, including:• Rescheduling of delinquent loans where the issues

motivating delinquency could be resolved through a restructuring of repayment schedules (e.g., where late payment was due to a timing mismatch between micro-enterprise income fl ows and loan repayments)

• A shift from group to individual lending• Increased emphasis on in-depth fi nancial analysis• Strict verifi cation of credit history and total

indebtedness• Strict scaling of loans abandoned• Automatic renewal of loans to people who repay on

time.• Greater price (i.e., interest rate) diff erentiation accord-

ing to risks and costs• Follow-up mechanisms for delinquency and loan recov-

ery refi ned (e.g., specialized staff for loan recovery;



earlier and greater participation of legal department in loan recovery)

ConclusionSimilar to global crises that follow booms, crises in micro-fi nance are also a victim of their own runaway success. Although to a signifi cant degree insulated from the eff ects of international fi nancial crisis, we have noted some relationships.

First, the fi nancing of larger MFIs has not been severely aff ected by the crisis, with deposits in some coun-tries reducing the dependence on institutional funding and multilateral organizations structuring special credit facilities to ensure that liquidity for the microfi nance sector (at least for the larger players) did not dry-up. Se-condly, the repeat needs of existing micro-entrepreneurs and new loans to previously excluded micro-entrepreneurs have meant that the sector has continued to grow. Th irdly, the fast growth of microfi nance and the continued high reimbursement rates in the past rendered MFIs compla-cent, opening the door for lax practices and failures in credit discipline. Fourthly, the impact of the international crisis has diff ered widely across regions, with local factors rather than the global situation per se driving develop-ments. Finally, NPL ratios have perhaps increased because some micro-entrepreneurs are indeed linked to the formal international economy (i.e., cash crop producers and pro-viders of goods and services for the tourist industry). A global economic crisis increases this vulnerability.

At the same time, we have seen that as microfi nance markets become saturated (especially in urban areas) com-petition is likely to lead to cross-lending and over-indebt-edness: the examples of Bolivia (in the 1999 crisis) and Bosnia and Herzegovina, Morocco, Nicaragua and Paki-stan (in the context of the current crisis) are instructive in this respect. In such a context (and as a pre-crisis avoid-ance strategy), MFIs have to focus particularly on strength-ening credit risk management through appropriate underwriting methodology, loan pricing policies and col-lection processes, consolidating and streamlining back

offi ce and front offi ce operations, increasing communica-tion with staff and investors, as well as enhancing balance sheet management and broader risk management across the institution. We would suggest, furthermore, that pol-icy-makers should renew attention on regulatory struc-tures and institutions such as credit bureaux to provide the required support to fast-evolving microfi nance sectors in emerging markets.

Th e lessons for the broader fi nancial sector are similar to those for the success of MFIs. Painstaking adherence to the basic principles of prudent lending should be fol-lowed in spirit as well as in letter. Funding is one element in the success of a project and bankers must ensure that all the other elements are also present before providing the fi nancing.

References

Armendàriz B, Morduch J. 2005. Th e Economics of Microfi -nance. MIT Press: Cambridge, MA.

Armendàriz de Aghion B. 1993. Analytical issues on LDC debt: a survey. World Economy 16(4): 467.

Arriola Bonjour P. 2003. Microfi nance in Bolivia: History & Current Situation. Inter-American Development Bank (IADB): Washington, DC.

Ashta A. 2009. Microcredit capital fl ows and interest rates: an alternative explanation. Journal of Economic Issues (M.E. Sharpe Inc.) 43(3): 661–683.

Ashta A, Fall NS. 2009. Institutional analysis to understand the growth of microfi nance institutions in West African economic and monetary union. 4th African Microfi nance Conference, Ouagadougou, Burkina Faso; pp. 1–20.

Attuel-Mendes L, Ashta A. 2008. French usury legislation and the development of credit availability for microenterprise. Global Journal of Business Research 2(2): 123–138.

Bernasek A, Stanfi eld JR. 1997. Th e Grameen Bank as progres-sive institutional adjustment. Journal of Economic Issues 31(2): 359–366.

Bhatt N, Shui-Yan T. 2001. Designing group-based microfi -nance programs: some theoretical and policy considerations. International Journal of Public Administration 24(10): 1103–1125.



Bruner RF, Carr SD. 2007. Lessons from the fi nancial crisis of 1907. Journal of Applied Corporate Finance 19(4): 115–124.

Chen G, Rasmussen S, Reille X. 2010. Growth and vulnerabili-ties in microfi nance. In Focus Note 61. CGAP: Washington, DC.

Chowdhury MJA, Ghosh D, Wright RE. 2005. Th e impact of micro-credit on poverty: evidence from Bangladesh. Progress in Development Studies 5(4): 298–309.

Constantinou D, Ashta A. 2010. Comment la microfi nance surmonte-t-elle les crises fi nancières et économiques? Tech-niques Financières et Développement 99: 31–44.

Copestake J, Bhalotra S, Johnson S. 2001. Assessing the impact of microcredit: a Zambian case study. Journal of Development Studies 37(4): 81.

Daru P, Churchill C, Beemsterboer E. 2005. Th e prevention of debt bondage with microfi nance-led services. European Journal of Development Research 17(1): 132–154.

Financial Crisis Inquiry Commission. 2011. Th e Financial Crisis Inquiry Report. Public Aff airs: New York.

Flannery MJ. 1996. Financial crises, payment system problems, and discount window lending. Journal of Money, Credit & Banking 28(4): 804–824.

Gonzales A. 2009. Microfi nance at a Glance — 2008. Microfi -nance Information Exchange: Washington, DC.

Harper M, Rao DSK, Sahu AK. 2008. Development, Divinity, and Dharma: Th e Role of Religion in Development Institutions and Microfi nance. Practical Action Publishing: Rugby.

Khan S, Ashta A. Forthcoming. Cost control in microfi nance: lessons from the ASA case. Cost Management 25(5).

Labrune G. 2010. Le Microcrédit: Une Opportunité Économique Et Sociale? Conseil Économique, Social et Environnemental: Paris.

Littlefi eld E, Kneiding C. 2009. Th e global fi nancial crisis and its impact on microfi nance. CGAP, Focus Note No. 52 (February).

Llanto GM, Badiola JAR. 2009. Th e Impact of the Global Finan-cial Crisis on Rural and Microfi nance in Asia, Discussion Paper No. 2009-24. Philippine Institute of Developing Studies: Makati City.

Marconi R, Mosley P. 2006. Bolivia during the global crisis 1998–2004: towards a ‘macroeconomics of microfi nance’. Journal of International Development 18(2): 237–261.

McGuire PB, Conroy JD. 1998. Eff ects on Microfi nance of the 1997–1998 Asian Financial Crisis. Foundation for Develop-ment Cooperation: Brisbane.

Meagher P. 2002. Microfi nance Regulation in Developing Coun-tries: A Comparative Review of Current Practice. IRIS Center, University of Maryland: College Park, MD.

Navajas S, Conning J, Gonzalez-Vega C. 2003. Lending tech-nologies, competition and consolidation in the market for microfi nance in Bolivia. Journal of International Development 15(6): 747–770.

Patten RH, Rosengard JK, Johnston DE Jr. 2001. Microfi nance success amidst macroeconomic failure: the experience of Bank Rakyat Indonesia during the East Asian crisis. World Develop-ment 29(6): 1057–1069

Reille X, Kneiding C, Rozas D, O’Donohoe AN, and Mariz FRD. 2010. All Eyes on Asset Quality: Microfi nance Global Valuation Survey 2010, Occasional Paper 16. CGAP: Wash-ington, DC.

Rhyne E. 2002. Commercialization and Crisis in Bolivian Microfi nance. In D. Drake and E. Rhyne (eds), Th e Commer-cialization of Microfi nance: Balancing Business and Develop-ment. Kumarian Press: Bloomfi eld, CT. pp. 114–137.

Roodman D, Morduch J. 2009. Th e Impact of Microcredit on the Poor in Bangladesh: Revisiting the Evidence, CGD Working Paper. Center for Global Development: Washington, DC.

Rosenberg R, Gonzalez A, Narain S. 2009. Th e New Moneylend-ers: Are the Poor Being Exploited by High Microcredit Interest Rates?, Occasional Paper 15. CGAP: Washington, DC.

Seibel HD, Nurcahya K. 2009. Growth and Resilience of Savings-based Microfi nance Institutions in the Face of Adversity: Th e Case of the Lembaga Perkreditan Desa in Bali. http://www.mikrofi -nanzwiki.de/sites/default/files/LPD_Bali_-_Growth_and_Resilience_2009.pdf

Seibel HD, Rachmadi A, Kusumayakti D. Forthcoming. Reform, growth and resilience of savings-led commercial microfi nance institutions: the case of the microbanking units of Bank Rakyat Indonesia. Savings and Development 35(3).

Vogelgesang U. 2003. Microfi nance in times of crisis: the eff ects of competition, rising indebtedness, and economic crisis on repayment behavior. World Development 31(12): 2085–2114.

Yunus M. 2003. Banker to the Poor: Micro-lending and the Battle Against World Poverty. Public Aff airs: New York.



BIOGRAPHICAL DETAILS

Dinos Constantinou has been active in banking, microfi nance and management consulting since 1992. He has worked in international and corporate banking with Barclays and Dresdner Bank; strategy consulting with Gemini Consulting (now part of Capgemini); and micro and SME fi nance with IPC Internationale Projekt Consult, where he led major long-term projects for the European Bank for Reconstruction & Development and Inter-American Development Bank. Dinos founded Global Microfi nance Group (a microfi nance investment company) in 2004; and Microfi nance Strategy (a management consulting company specializing in fi nancial services and development fi nance) in 2007. He holds an MBA from IMD International in Lausanne, an MA in European Economic Studies from the College of Europe in Bruges, and a BSc (First Class Honors) in Economics from the University of Bristol. He has edited (with Professor Benoit Leleux) the book From Microfi nance to Small Business Finance — Th e Business Case for Private Capital Investments.

Arvind Ashta holds the Microfi nance Chair of the Burgundy School of Business (Groupe ESC Dijon-Bourgogne), France, fi nanced by Banque Populaire de Bourgogne Franche-Comté. Besides fi nance, control and law, he off ers courses in microfi nance and researches regulatory aspects of microfi nance and advanced technology in microfi nance. He has taught microfi nance as a visiting faculty in Chicago (US), Pforzheim (Germany) and ULB (Belgium). He has a number of publications in international journals. He has recently edited a book on ‘Advanced Technologies for Microfi nance’. He is a member of CERMi.

Correspondence to:Arvind AshtaBurgundy School of Business29 rue Sambin, BP 5060821006 Dijon CedexFranceemail: [email protected]

Documents

Financial crisis: lessons from microfinance