Some pitfalls of the commercialization of microfinance
Robert Lensink
Department of Economics, Econometrics and Finance
Faculty of Economics and Business
University of Groningen and
Development Economics Group,
Wageningen University
email: b.w.lensink@rug.nl
Acknowledgement: The author thanks two anonymous referees for comments on an earlier version, and SNS
Asset Management for providing data.
Abstract
This paper critically discusses the current trend of commercialization of microfinance. It
draws attention to some possible negative effects the commercialization, such as increasing
indebtedness, rising interest rates, and reduced access to credit for the poorest of the poor.
These issues have recently raised ethical debates with respect to the microfinance industry.
The paper also argues that due to the commercialization microfinance institutions are much
more vulnerable for international financial crises. Some preliminary evidence for this is given
by an analysis of the impact of the current economic and financial crisis on microfinance
institutions (MFIs). The results indicate that as a consequence of the crisis, MFIs face
significant negative shifts in performance indicators related to profitability, growth, and
portfolio quality.
1
Introduction
During the past twenty years, microfinance programs have been introduced in many
developing economies. Well-known examples of microfinance institutions include Grameen
Bank, ASA and BRAC in Bangladesh, Banco Sol in Bolivia and Bank Rakyat in Indonesia.
Between December 1997 and December 2007, the number of people who received credit from
microfinance institutions (MFIs) rose from 16.5 to 154.8 million. The number of poorest
clients reported increased from 9 million to more than 106 million during this period
(Armendariz and Morduch, 2010).
The enormous increase in microfinance programs was an immediate response to the public
image that microfinance is about profitable lending to small groups of extremely poor women
who benefit strongly from access to credit. In many conversations on microfinance it has been
argued that the microfinance industry would be able to offer a combination of financial
returns and positive social impact. Hence, the idea that MFIs were able to achieve a so-called
double bottom line received considerable support.
Attention to microfinance and its potential role in reducing poverty increased even further
when Mohammad Yunus received the Nobel Peace Prize in 2006, prompting an almost
euphoric attitude among policy makers and aid organizations about its potential promise.
However, several recent developments challenge the extremely positive view on
microfinance. Empirical testing of the impact of microcredit appears to be very difficult and
controversial (see, for instance, Armendariz and Morduch, 2010, chapter 9). Fortunately, in
the last few years, several new empirical analyses on the impact of microcredit have been
started, using superior methodologies, often based on so-called randomized controlled trials.
Well-known examples include
Karlan and Zinman (2009), who study the impact of
microcredit on investment in Manila, the Philippines, and Banerjee et al. (2009), who focus
on MFIs in the slums of Hyderabad. These studies show mixed results. Most importantly, the
studies were not able to find strong positive effects of microfinance. Another recent study, by
Roodman and Morduch (2009), who attempt to replicate the well-known study by Pitt and
Khandker (1998) on the impact of microfinance in Bangladesh, also could not provide
evidence for a positive impact of microfinance. Bateman (2010) even is much more negative
by arguing that microfinance constitutes a main obstacle to sustainable development. He e.g.
states that the neoliberal microfinance wave mainly resulted in the financing of unproductive
small enterprises, at the expense of the most productive SME sector.
2
Moreover, recently the practices of MFIs are subject to growing ethical concerns (Hudon and
Sandberg, 2012). MFIs have been criticized for their lending practices that led to overindebted
clients. The southern Indian state of Andhra Pradesh even plunged into a microcredit crisis
after the microfinance-induced suicides in 2010. Branches of several large MFIs in Ecuador,
India and Nicaragua have been closed by the authorities. The Nobel laureate Muhammed
Yunes was even accused of stealing 100 million USD from the Grameen Bank. In addition,
the past decade has seen the commercialization of microfinance. As a result of
commercialization, MFIs have increased the interest rates. It is now common to see interest
rates above 50% on a yearly basis. The high rates of interest charged on microloans resulted
in one of the most salient ethical debates with respect to microcredit (for a discussion of fair
interest rates, see e.g. Hudon, 2007 and Hudon and Sandberg, 2011).
Commercial banks now play increasing roles in funding microfinance institutions and help
expand the traditional services offered by MFIs. Whereas once they focused only on
microcredit, in the form of small loans to the poor, today they have shifted from microcredit
to microfinance, including savings, and insurance. The funding situation of MFIs also has
started to change rapidly. Whereas once they relied mainly on private and public donors and
aid organizations, more and more MFIs turn to the capital market for funding.
Some argue that the commercialization of microfinance is a blessing since it both improves
MFI performance and the availability of funds for the poor. Others, such as Bateman (2010),
state that the commercialization of microfinance will even accelerate the negative impacts of
microfinance. Critics fear that the commercialization will encourage MFIs to put shareholders
above the poor they serve. Moreover, the commercialization may lead to a mission drift, such
that MFIs start to concentrate on richer clients, and reduce or even stop lending to the poorest
of the poor. This also raises ethical concerns since in the microfinance literature there is a
discussion whether there is a human right to credit (see e.g. Hudon, 2009). This literature
argues that extreme poverty is a violation of human rights, and that microcredit may help to
alleviate poverty.
In this paper, I will survey and discuss the recent trend of commercialization of microfinance.
I will pay attention to the impact of commercialization on outreach to the poor and the interest
3
rate setting of MFIs, both issues received much attention in ethical debates with respect to
microfinance.
Moreover, I will argue, in line with Gonzalez (2007), that the
commercialization induced an increase in involvement of commercial investors in the
microfinance sector, which implied that MFIs became much more exposed to adverse market
shocks. I will provide some preliminary evidence for this view by showing that the recent
financial crisis had a negative effect on the performance of MFIs.
The rest of the paper is organized as follows. Section 2 discusses the recent wave of
commercialization of microfinance. Section 3 presents an analysis of the impact of the
financial crisis on MFIs. Finally, section 4 concludes the paper.
1. Commercialization of microfinance
1.1. Recent developments
Traditionally MFIs have been funded mainly by subsidies from private and public donors and
aid organizations (for references on subsidies in microfinance, see Chapter 10 in Armendáriz
and Morduch, 2010). Several MFIs also attract savings from their members. However, about
ten years ago, microfinance started to become more commercial, and this commercialization
is increasing very strongly. In the literature, commercialization refers to either the attempt of
MFIs to operate without subsidies, or to the broader process of applying market-based
principles (Armendáriz and Morduch, 2010). In this paper, we use the term commercialization
in the broadest sense. Mainstream financial institutions, including commercial banks and
private and institutional investors, appear more and more interested in the microfinance
market. So-called social investors, who value the social, as well as financial, performance of
their investments, have started to invest. In the course of the global financial crisis, pension
funds also display increasing interest in microfinance, in the search for innovative investment
opportunities with low co-variate risk and higher social returns. In the Netherlands for
instance, the retail bank SNS launched the Institutional Microfinance Fund, almost entirely
funded by Dutch pension funds.
In several developing countries large state banks and private banks have started to provide
microfinance services. In Pakistan, for instance, a number of private commercial banks have
moved into microfinance. In Malaysia, Nepal, and Thailand there are programs stimulating
4
commercial banks to become involved in microfinance. In India the National Bank of
Agriculture and Rural Development (NABARD) recently initiated a program to involve
private banks in microfinance. According to recent studies the growth of microfinance in
India is led by a number of commercial banks such as ICICI, HSBC and ABN AMRO,
together with private venture capital funds and social venture capitalists.
There is also an increase in MFIs for which capital market funding becomes important. The
first example of commercial capitalization of MFIs was the creation of an investment fund
called Profund, which raised $23 million to finance Latin American MFIs. Some
microfinance institutions have even gone public, and became commercial banks. The most
well-known example is Compartamos in Mexico. Compartamos started in 1990 as a standard
MFI by providing joint-liability loans to female borrowers. In 1998 Compartamos became a
regulated financial institution, a so-called Sociedad Financiera de Objeto Limitado (SFOL). In
2002 Compartamos, as one of the first MFIs, issued public debt. In April 2007 Compartamos
went public and transformed itself into a commercial bank. Another example is SKS
microfinance in India. SKS microfinance recently went public, with an initial public offering
(IPO) of $354 million. In the last two years, 40% of all private equity deals in India are due to
MFIs.
Currently, approximately 50 percent of the funding of foreign capital to MFIs is channeled
through specialized Microfinance Investment Vehicles (MIVs). The number of MIVs has
increased rapidly and there are now over 80 in existence MIVs. The main investors in MIVs
are individual investors, among others pension funds (for a discussion, see e.g. Galema,
Lensink and Spierdijk, 2011).
MFIs that traditionally have focused on providing microcredit to the poor in the informal
sector, also have started to recognize that savings and insurance products can be even more
important than providing loans. Diversification represents an essential step in the
development of microfinance. As a result of this shift toward microfinance, MFIs now collect
savings, and provide insurance products. Especially life insurance is a popular product.
Mostly, life insurance is offered as part of a microcredit package, in the form of credit-life
contracts. These contracts ensure that, if somebody dies, the outstanding loans will be paid off
and a fixed payout will be provided to the family. In addition to life insurance products, MFIs
start to develop health insurance plans as well as property and crop insurance. Micro finance
5
institutions even have begun experimenting with pension schemes. These micro pensions
focus on long-term savings, with the aim of providing income security to the aged poor who
have worked in the informal sector and are not covered by formal retirement schemes. The
well-known Grameen Bank in Bangladesh, for instance, introduced the Grameen Pension
Scheme (GPS) in 2001. Several other MFIs and non-governmental organizations (NGOs),
including the Dhan Foundation in India and BRAC in Bangladesh, are now developing micro
pension products for their clients. With their close connections and regular meetings with
low-income workers in the unorganized sector, MFIs could be ideal financial institutions to
channel micro pension products to the poor in informal sectors.
In addition, new banking technology, such as charge cards, ATMs, the use of cell phones and
the internet has begun to enter the microfinance business, helping to reduce costs and improve
the delivery of services. Finally, in many developing countries, induced by the
commercialization of microfinance, and the increased role for micro savings products,
governments started installing regulations to help improving the stability of the microfinance
business.
2.2 Commercialization: a discussion
The commercialization of microfinance has led to severe discussions and ethical debates
among academics. These discussions focus on (1) why are international investors investing in
MFIs? Are MFIs interesting investment opportunities?, (2) What does commercialization and
the focus on efficiency imply for outreach? Is the double bottum line in micorfinance an
utopia? and (3) what are the implications for the interest rates? I will briefly discuss these
issues below.
Investing in microfinance may be interesting for international private investors for several
reasons.
First, there is an increased attention for social responsible entrepreneurship.
Microfinance could be an interesting alternative for private investors who, for altruistic
motives, want to contribute to the fight against poverty. By directly targeting the poor, and
therefore circumventing inefficient/corrupt government structures, microlending could be an
efficient means of doing well. However, as I have argued above, there is little empirical
evidence for a huge poverty reducing effect of microfinance. Moreover, commercial banks
may try to use subsidized loans to the poor as an instrument to enlarge their future profit
making portfolio, based on the assumption that lending to these poor will make them more
6
wealthy clients in the future, who can afford paying market-based interest rates. Second, the
potential size of the microfinance market is still very big. There are currently up to 3 billion
people without access to proper financial services. There are also an estimated 500 million
micro-entrepreneurs worldwide. Thus, there is a largely untapped source of microcredits for
international investment banks that can be pooled together and sold to investors all around the
world. Third, microlending probably gives higher returns than traditional lending. Typically, a
poor farmer or entrepreneur in the developing world generates much higher returns on her
assets than corresponding business in the developed world. Thus, as is argued by some, many
poor can afford to pay much higher lending rates. Some empirical evidence indeed suggests
that interest rates can be much higher for lending to the poor. For instance, moneylenders in
developing countries ask sometimes daily interest rates of 20% (which is more than 5000% a
year!). Fourth, default rates in microfinance are said to be low. The empirical literature on
microfinance lending indeed suggests that default rates in microfinance are very low, often
below five percent. These low default rates are in part caused by innovative MFI lending
techniques, such as group lending, dynamic incentives, and obligatory weekly repayments.
The low default rates, in combination with high lending rates, imply a high risk-adjusted rate.
However, there is also some evidence that default figures are deflated. Moreover, MFIs are
heavily subsidized and the credit quality of average borrowers could fall if the client base is
widened due to the commercialisation. Finally, microfinance also may provide attractive
opportunities for portfolio diversification, because the risk-adjusted returns likely exhibit low
correlations with other assets; that is, many microfinance clients are part of the informal
economy, which is less sensitive to macroeconomic cycles.
The common assumption of reduced exposure of MFIs to financial and economic market
movements (e.g. Hermes and Meesters, 2009; Hermes, Lensink and Meesters, 2010) rests on
several features, including (1) the minimal integration into the formal sector of business
activities undertaken by microfinance clients; (2) high incentives and great loyalty among
microfinance borrowers, which prompt them to repay their loans; (3) the reliance of MFIs on
highly dedicated, socially oriented funding sources, which tend to be less affiliated with
business cycles and can provide MFIs with continuous access to funding (Walter and Krauss,
2009); (4) a combination of short maturities of outstanding loans and long maturities of MFI
funding, which leaves significant room to adjust outstanding loan portfolios in times of tight
liquidity (JP Morgan, 2009); and (5) relatively low fixed costs for both clients’ businesses and
MFIs, which support quick adjustments to changing market conditions.
7
Surprisingly little research undertakes rigorous testing of the assumption that microfinance
provides attractive opportunities for portfolio diversification. Some studies support this
common assumption about reduced exposure to fluctuations in international markets (Galema,
Lensink and Spierdijk, 2011; Walter and Krauss, 2009) and domestic macroeconomic markets
(Gonzalez, 2007). Galema, Lensink and Spierdijk (2011) examine the benefits of investing in
MFIs. They show that it can indeed provide risk-return benefits for institutional investors.
However, they also show that the returns depend heavily on the location of and type of MFI.
Not surprisingly, their analysis suggests that investing in more commercialised MFIs may be
beneficial, but investing in more traditional MFIs and those in Africa do not provide effective
risk–return benefits. Others, however, reject the claim that microfinance is resilient to
macroeconomic developments and find a significant relationship between MFI performance
and domestic economic growth (Ahlin, 2006; Hermes and Meesters, 2009).
Although the literature is not entirely unambiguous, it seems as if investing in microfinance
induced by the commercialization of microfinance
may be interesting for international
investors. However, some hurdles need to be taken. Most mainstream investors deal with
regulated entities that work in a clearly defined legal environment. This implies that MFIs
needs to be transformed from NGOs with weak corporate governance into regulated banks.
Moreover, the macro policy and government regulations have to be adapted to accommodate
commercial microlending. In addition, if international capital markets are to be tapped on a
larger scale, the MFI investors have to be able to hedge the foreign exchange risk that appear
when they borrow/ lend in a foreign currency. However, MFIs are often in countries were
there are no possibilities to reduce exchange rate risk since often the currencies are not
convertible. Finally, lending on subsidized terms may lead to crowding out of commercial
lending.
There is also a discussion on the impact of commercialization on outreach. Some argue that
commercial microfinance, or the for-profit side of the microfinance sector, is necessary to
increase funding for MFIs. In most developing countries there is still a huge amount of
potential borrowers with no access to financial services. Only in India, there are some 120
million families without access to financial services. The microfinance sector needs to scale
up in order to meet this additional demand. Hence, commercialization seems to be needed.
Clearly, a possible advantage of the move of commercial banks into microfinance is that these
8
banks can provide credit to the poor, without being dependent on donor subsidies.
Commercial banks have diversified portfolios, and are able to use profits from lending to
wealthy clients to finance (subsidized) loans to the poor, i.e. cross-subsidization would take
place. Hence, commercialisation may increase the availability of funds and therefore may
improve outreach. Moreover, the commercialisation may lead to more loans with longer
maturities and more diversified funding sources.
However, there are also some concerns. Bateman (2010) is extremely pessimistic about
commercialization of microfinance. In his view, the commercialization goes together with “a
whole host of anti-social and anti-poor developments” (Bateman, 2010, p.153). A potential
disadvantage of the increased involvement of commercial banks is that traditional
microfinance institutions are confronted with increased competition in the market for micro
loans. While increased competition may lead to increased efficiency and stimulate financial
sustainability, it may also reduce the scope for lending to the core poor. Moreover, increased
competition and increased supply of loans may result in higher levels of indebtedness of the
clients as they may take up multiple loans from different sources at the same time. The
commercialization may also lead to a strong increase in interest rates of MFIs, which even
further increases indebtedness of the clients. As I have stated before, the possible increase in
interest rates raised salient ethical discussions. Indeed, due to the commercialization MFIs
have increased the interest rates. It is now common to see interest rates far above 50% on a
yearly basis. Some may argue that high interest are not problematic since the poor are mainly
interested in access to credit, and even with high interest rates they are able to make
profitable investments since marginal returns of scarce capital are very high. However, is this
true? Already for decades, economists debate whether interest rates should be subsidized or
whether the poor can afford high interest rates. In the 1950s and 1960s, most economists
argued that poor people cannot afford to pay high interest rates and therefore need cheap
capital to adopt profitable investments. Consequently, many subsidized credit programs were
developed. These policies came under attack during the 1970s. It was pointed out that
offering “cheap credit” is a wasteful way of encouraging (agricultural) development. Cheap
credit would lead to unproductive use of credit, create opportunities for corruption and
favoritism, and results in discrimination against the poor since typically the better-off farmers
receive the cheap credit. Moreover, low interest rates would discourage savings. Yet, the most
recent empirical literature does not support the idea that the poor do not care about high
interest rates. From the wider economic literature, especially on health inputs, we can learn
9
that demand of the poor is extremely price elastic (Cohen and Dupas, 2010). More
importantly, recent studies based on randomized controlled trials show that the demand for
credit by the poor is also extremely price elastic, even more than that of the not-so-poor
(Dehejia, ET AL., 2011 AND Karlan and Zinnman, 2008). Hence, while high interest rates
are good for profits of the MFIs, they seem to be bad for the poor.
Some even argue that microfinance is the new subprime, especially after recent stories about
loan-shark-style microfinanciers who have driven borrowers to suicide in the Indian state of
Andhra Pradesh. There is not much serious research available on the effect of competition
among microfinance institutions. An exception is a study by McIntosh, De Janvry and
Sadoulet (2005). These authors suggest that wealthier borrowers are likely to benefit from
increasing competition among microfinance institutions, while poorer borrowers will suffer in
terms of lower levels of welfare. In addition, commercialization and the related emphasis on
financial sustainability has raised concerns with respect to the consequences for the outreach
of microfinance, i.e. the number (breadth) and socioeconomic level (depth) of the clients that
are served by MFIs. Since lending to poor borrowers can be very costly, outreach and
sustainability may be conflicting. Cull, Demirgüz-Kunt and Morduch (2007) provide some
evidence for a trade-off between sustainability and outreach. They also suggest that
individual-based MFIs increasingly focus on wealthier clients, a process termed as mission
drift. This may imply that the possible positive effect of comemrcialization in terms of an
increase in loans with longer maturities that are available for the poor (see above) will be
counteracted. Hermes, Lensink and Meesters (2011) also provide evidence on the existence of
the trade-off between sustainability and outreach. Their study strongly suggests that outreach
is negatively related to efficiency of MFIs. This may be bad news in view of the current move
to commercialization of the microfinance industry since commercialization probably induces
a stronger emphasis on efficiency. However, it may also be the case that, due to spill-over
effects, MFIs that strive for efficiency, and score low on outreach to the poor, may ultimately
cause a higher poverty reduction at the macro level than MFIs that score high on outreach
indicators.
The commercialization of microfinance has clearly invited a strong debate among
microfinance practitioners and academics. Some point at positive effects, others are much
more skeptical. The main problem is that both views are not yet backed by serious academic
studies. There are some recent studies available, which suggests that there probably is a trade10
off between commercialization and outreach to the poor. However, most discussions on the
advantages and disadvantages of commercialization of microfinance are based on anecdotal
evidence. More research is needed before the commercialization of microfinance can be
validated.
2.
Microfinance and the financial crisis1
Muhammad Yunus, “the founder” of microfinance, asserted in August 2008 that “despite the
turmoil on financial markets, microfinance still works” (Microcapital, 2008). His claim
reflects the common belief that MFIs exhibit little correlation with formal market movements,
which constitutes one of the main reasons commercial investors have been attracted to such
investments (Deutsche Bank Research, 2007) and why the equity owned by microfinance
institutions deserves a premium compared with traditional finance equity (JP Morgan, 2009).
However, the current financial and economic crisis seems likely to have affected the
performance of MFIs due to the commercialization of microfinance in general and the
increased involvement of commercial investors in the microfinance sector, in particular.
Using a financial dataset containing monthly panel data from 57 MFIs during January 2007August 2010, I investigate whether MFI performance significantly changed due to the
financial crisis. The sample of MFIs is part of the SNS Institutional Microfinance Fund. 2 The
dataset combines monthly data reported by MFIs to developing world markets (DWM) with
the key financial ratios provided in summary reports by DWM to SNS Asset Management
monthly. The profitability measure uses net income divided by total assets, or the return on
assets (ROA), and operational self-sufficiency (OPSELFSUF), which is the extent to which
costs of operations are covered by the income from operations. To measure portfolio quality, I
use four indicators: an early indicator defined as loans with payments behind schedule of less
1
This section is partly based on an earlier analysis of the impact of the financial crisis on MFIs (see Huijsman
and Lensink, 2011).
2
In this sample, larger MFIs (Tier-1) may be overrepresented. To control for the possible sample selection bias,
I adjusted the data set using a post-stratification method. Specifically, after gathering the sample data, the sample
weights are adjusted to make weighted sample moments equal the finite population moments (Carrington et al.,
2000). All observations are adjusted to ensure the sample distribution was equivalent to that of a benchmark data
set of MFIs, using information on a sample of MFIs from CGAP.
11
than 30 days as a proportion of the total loan portfolio (PARKL30), a later indicator reflecting
the proportion of the loan portfolio of non-repaid loans that matured more than 30 days ago
(PAR30), and an indicator that includes restructured and refinanced loans as a proportion of
the total loan portfolio to the proportion of portfolio at risk for more than 30 days
(PAR30RR),3 as well as a final indicator, WRITEOFFS, that is a default indicator of the risk of
default. Monthly growth refers to the outstanding loan portfolio and active clients.
To analyze the impact of the crisis on the MFIs’ financial performance, I tried to identify
structural breaks in the time series of performance indicators over the study period. Fixed
effects regressions allowed the intercept in the regression model to differ cross-sectionally but
not over time (Brooks, 2000). Thus, it is not unambiguous when the crisis started to have
impacts, and I determine the structural breaks endogenously by performing regressions on all
possible breaks (so-called rolling regressions). The model specification is as follows:
𝑦𝑖𝑡 = 𝛼𝑖 + 𝛽𝐶𝑅𝐼𝑆𝐼𝑆(𝑇)
where 𝑦𝑖𝑡 is the observed performance indicator of MFI i in month t, β is the estimated
structural shift in the time series of the performance indicator at month (T + 1), and α is a
constant that represents the average of the performance indicator before month (T + 1). For
each performance indicator, I indentified the CRISIS(T) that best explained the variation of
the performance indicator, according to the model with the highest overall R-within.
After I determined the structural breaks for all the performance indicators of interest, I
investigated whether the impact differed across regions by generating interaction effects
between the crisis dummy and region dummies (CAC, CA, SA, SSEA),4 which produced the
following model:
𝑦𝑖𝑡 = 𝛼𝑖 + 𝛽1*𝐶𝑅𝐼𝑆𝐼𝑆 + 𝛽2* 𝐶𝑅𝐼𝑆𝐼𝑆*𝐶𝐴𝐶 + 𝛽3* 𝐶𝑅𝐼𝑆𝐼𝑆*𝐶𝐴 + 𝛽4 * 𝐶𝑅𝐼𝑆𝐼𝑆*𝑆𝐴 +
𝛽5* 𝐶𝑅𝐼𝑆𝐼𝑆*𝑆𝑆𝐸𝐴
This method provides a more complete picture of the impact on clients’ repayment behavior. If borrowers
cannot to meet their debt obligations on time, MFIs may choose to restructure and refinance loans instead of
defining them as late payments.
4
CAC refers to Central Asia and the Caucasus; CA is Central America and the Caribbean; SA is South America;
and SSEA is South- and South East Asia, each of which take a value of 1 if the MFI is located in that region.
3
12
where 𝛽1 is the estimated shift of the performance indicator after the structural break of MFIs
in Eastern Europe and Russia (EER), and 𝛽2 is the estimated shift of the performance
indicator of MFIs in Central Asia and the Caucasus (CAC) with respect to the estimated shift
of the performance indicator of MFIs in EER. In addition, CRISIS is the CRISIS(T) that the
rolling regressions (Model 1) identified as the best explanation of variation over time in the
different performance indicators (so it may vary across performance indicators).
As we show in Table 1, for all performance indicators, I can identify a significant structural
break, according to the significant coefficients of CRISIS. Although the timing of these breaks
differ somewhat, they occur between November 2008 and April 2009, with the exception of
the write-off ratio, for which the structural break does not occur until August 2009.
Table 1. Results of the financial performance study: Effects of the crisis
(1)
VARIABLES
(2)
(3)
(4)
(5)
(6)
Portfolio at
Portfolio at
Portfolio at
Write-offs
Operational
Return on
Growth in
Loan
risk <30
risk >30 days
risk +
Self-
assets
number of
portfolio
refinanced
sufficiency
(annualized)
active
growth
days
&
(7)
(8)
borrowers
restructured
loans
Timing of the
Nov 08
April 09
April 09
August 09
Jan 09
Jan 09
Nov 08
Nov 08
0.0141***
(6.86e-08)
0.0577***
0.0650***
0.0091***
-0.2228***
-0.0408***
-0.0280***
-0.0442***
(0.00127)
(0.000105)
(0.00287)
(6.79e-06)
(0.000230)
(1.79e-06)
(2.27e-10)
0.0303***
0.0222***
0.0286***
0.0064***
(0.0000)
(0.0000)
(0.0000)
(0.0000)
Observations
893
1544
1703
1213
Number
57
57
57
57
0.155
0.212
0.228
0.079
structural break
CRISIS
Constant
Adjusted
R-
1.2444***
(0.0000)
0.0449***
0.0282***
0.0412***
(0.0000)
(0.0000)
(0.0000)
1777
1819
1373
893
57
57
57
57
0.163
0.120
0.065
0.114
squared
Notes: Robust p-values in parentheses. Fixed-effects regressions.Definition of variables: Portfolio at risk: loans with payments behind
schedule of less than 30 days as a proportion of the total loan portfolio (Portfolio at risk (after 30 days)/Gross loan portfolio); Return on
assets: (net operating income - taxes)/ total assets; Operational self sufficiency = the extent to which costs of operations are covered by the
income from operations = Operating revenue/(financial expense+loan-loss provision expense+operating expense); Write-offs: a default
indicator of the risk of default
*** p < 0.01, ** p < 0.05, * p < 0.1.
The results of the performance study in Table 1 show that the average ROA decreased
significantly after January 2009, from an average level of 4.5% to 0.4%. After January 2009,
13
the average operational self-sufficiency also decreased substantially, from an average of
124% to an average of 102%. Thus, MFIs on average still maintained their ability to cover
operational costs.
The results of the financial performance study show negative effects on all indicators that
measure portfolio quality (see Table 1, columns 1–4). The timing of the structural breaks for
these indicators differ; the earliest pertained to an indicator for credit risk, portfolio at risk <
30 days, which occurred in November 2008. The actual measure of default write-offs did not
exhibit a structural break until August 2009 though. Gonzalez (2007) finds that though the
risk of default may increase during times of weak economic growth, loans eventually can be
recovered. We counter that in times of crisis, this finding does not apply. A significant
adverse impact on default rates does occur.
The results of the financial performance study (Table 1) show that growth suffered significant
negative effects after November 2008. Average monthly growth rates in the loan portfolio
decreased from 4.1% to -0.03%, implying a shrinking loan portfolio. Client growth rates also
were negatively affected. After the structural break in November 2008, the average monthly
growth rates of active clients went down by 2.8 percentage points, resulting on average in
zero growth.
In Table 2, I show that the impact of the crisis differs significantly across regions. With
respect to overall profitability, MFIs from Eastern Europe and Russia are most affected. The
differential impact may reflect that MFIs located in these regions have relatively high
leverage ratios (Loncar et al., 2009), such that when funding costs rise, their profitability may
be more affected than is the profitability of MFIs that depend less on external funding.
Interest payments are fixed obligations, and higher leverage ratios cause MFIs to incur higher
costs when they scale back their operations (Walter and Krauss, 2009).
Table 2: Differences per region
VARIABLES
(1)
(2)
(3)
(4)
(5)
(6)
Portfolio at
Portfolio at
Portfolio at
Write-offs
Operational
risk <30
risk >30
risk +
self-
days
days
refinanced &
sufficiency
(annualized)
restructured
(7)
(8)
Return on
Growth in
Loan portfolio
assets
number of
growth
active
borrowers
14
loans
Timing of the
Nov 08
March 09
March 09
Aug 09
Jan 09
Nov 08
Nov 08
Dec 08
structural break
Crisis
Crises*CAC
Crisis*CA
Crises*SA
Crisis*SSEA
Constant
0.0275***
0.0988*
0.1038**
0.0143***
-0.4161***
-0.0604*
-0.0403***
-0.0474***
(0.000388)
(0.0590)
(0.0272)
(0.00149)
(0.00179)
(0.0809)
(1.34e-05)
(6.90e-05)
-0.0161**
-0.0480
-0.0372
-0.0052
0.1115
0.0033
0.0201**
0.0044
(0.0460)
(0.249)
(0.373)
(0.227)
(0.314)
(0.907)
(0.0374)
(0.699)
-0.0182**
-0.0143
-0.0179
-0.0064
0.2120*
0.0167
0.0138
-0.0070
(0.0241)
(0.809)
(0.685)
(0.165)
(0.0810)
(0.561)
(0.409)
(0.523)
-0.0189**
-0.0633
-0.0644*
-0.0074
0.2329**
0.0304
0.0087
0.0044
(0.0185)
(0.106)
(0.0663)
(0.107)
(0.0247)
(0.260)
(0.332)
(0.744)
-0.0224***
-0.0279
-0.0267
-0.0128***
0.2519**
0.0364
0.0083
0.0188
(0.00417)
(0.478)
(0.452)
(0.00171)
(0.0142)
(0.172)
(0.372)
(0.150)
0.0303***
0.0413***
0.0285***
0.0157
0.0224**
0.0057***
1.2846***
0.0490***
(0.0000)
(0.0000)
(0.0000)
(0.0000)
(0.0000)
(0.0000)
(0.0000)
(0.0000)
Observations
893
1511
1668
1181
1744
1784
1351
Number
57
57
57
57
57
57
57
57
Adjusted R-squared
0.235
0.270
0.273
0.093
0.213
0.138
0.067
0.114
893
Notes: Robust p-values in parentheses. Fixed-effects regressions
*** p < 0.01, ** p < 0.05, * p < 0.1.
The stronger impact also may be related to the relatively high impact on portfolio quality for
MFIs in Eastern Europe and Russia, which causes them to increase their provisions to
anticipate higher future loan losses. The impact on portfolios at risk for less than 30 days,
which is an early signal for credit risk, is strongest in Eastern Europe and Russia, as implied
by the significant and negative coefficient of all the interaction variables. For the other
indicators of credit risk, the negative coefficients of all the interactionvariables (although most
are not significant) also suggest that repayment rates of MFIs in the region Eastern Europ and
Russia has been most affected. Russia has been directly affected by fuel prices and reversals
of foreign capital flows, and its poor economic situation has influenced the entire region
through reduced in trade, remittances, and investments (IMF, 2009a). Moreover, the Eastern
European microfinance sector in general provides relatively large loan sizes, presumably
exposing the assets of these MFIs to more market risk, because larger loans should be more
influenced by regular business cycles (Loncar et al., 2009). The resilience of the portfolio
quality of South American MFIs and the clear negative impact on their loan portfolio growth
rates may be because, even though the crisis certainly hit Latin America, these MFIs
immediately took steps to limit its impact by applying stricter lending policies and increasing
their focus on existing clients (Microrate, 2009). Central America and the Caribbean is part of
Latin America, but these MFIs may feel a more significant impact on their portfolio quality,
because this part of Latin America is located closer to the United States and tends to be more
15
integrated with this economy, which has been in recession since 2007. Moreover, coexisting
conditions, such as political instability and client over-indebtedness, often appear in reports
by MFIs from Central America and the Caribbean as causes for the current deterioration of
repayment rates.
Considering the indicator where restructured and refinanced loans are
included, there is a significant stronger effect for MFIs in Central America and the Caribbean
compared to MFIs in South America.
The results of the financial performance study show that the eventual defaults (WRITEOFFS)
of MFIs in South- and South East Asia are relatively least affected. The dependence of Southand South East Asia on remittances, as a percentage of gross domestic product, is the lowest
of all developing countries (IMF, 2009a), so these countries may enjoy less exposure to
reversals in this income source. Moreover, the economies in this region had begun to revive
already by February 2009 (IMF, 2009c).
4.
Conclusions
This paper provides a discussion of the recent wave of commercialization of microfinance.
The commercialization of microfinance appears to be one of the most important challenges
for the microfinance business in the near future, which raises several ethical debates. The
paper points at some possible positive consequences of commercialization of microfinance,
especially for investors in microfinance. However, the paper also argues that the
commercialization, and the related emphasis on financial sustainability, may have
considerable consequences for the outreach of microfinance. Since lending to poor borrowers
can be very costly, outreach and sustainability are conflicting. Moreover, MFIs increasingly
focus on wealthier clients, a process termed as mission drift. Therefore, there is a trade-off
between sustainability and outreach. MFIs which want to be financial sustainable, without
relying on subsidies, need to commercialize, and
consequently increase interest rates.
However, this will immediately reduce the outreach to the poor. If we really want
microfinance to help the poor, we probably should give up the idea of financial sustainability,
and accept that there is nothing wrong with subsidizing MFIs.
The paper also argues that due to the commercialization of microfinance, MFIs have become
much more vulnerable to international macroeconomic developments. The study provides
some preliminary evidence for this view by showing that the financial crisis considerably
affected MFI performance. The analysis strongly suggests that the microfinance sector is not
16
anymore as disconnected from formal markets as was commonly assumed, prior to the onset
of the current financial crisis. Although microfinance institutions were not involved in the
practices that led to the financial crisis, this study makes clear that MFIs are not insulated
from its impact. The commercialization of microfinance is probably one of the underlying
reasons for this. Fixed effects regressions show that the effects became visible in the fourth
quarter of 2008 and the first quarter of 2009, when MFIs’ profitability and monthly growth
rates in their loan portfolios and client bases began to shift downward, and the first signs of a
negative impact on portfolio quality also emerged. The impact differs highly across regions,
such that MFIs in Eastern Europe and Russia have been most affected, whereas those MFIs in
South America seem least affected.
The long run impact of the commercialization trend in terms of reducing poverty is still
unclear, however. One of the main reasons being a lack of comprehensive studies. The
methodology used in most studies suffers from some important weaknesses. Several studies
are solely descriptive and offer anecdotal evidence without sound empirical backing.
Moreover, the data sets used in most studies are weak and feature mainly case studies,
because cross-country data sets on MFIs have become available only recently. In addition,
most studies pertain to the period before the modern global financial crisis. It is an open
empirical question whether their results are still valid. Further research into the
commercialization of microfinance is therefore highly desirable.
17
REFERENCES
Ahlin, C., Lin, J. (2006). Luck or skill? MFI performance in macroeconomic context.
Working Paper.
Armendariz, B. and J. Morduch (2010). The economics of microfinance: second edition.
London: The MIT press.
Banerjee, A., Duflo, E., Glennerster, R., & Kinnan, C. (2009). The miracle of microfinance?
Evidence from a randomized evaluation. Unpublished paper, Abdul Latif Jameel
Poverty Action Lab and MIT.
Bateman, M. (2010). Why doesn’t microfinance work? The destructive rise of neoliberalism.
London: Zed Books.
Brooks, C. (2000) Introductory Econometrics for Finance. Second edition. Cambridge
University Press, Cambridge, UK
Carrington, W., Eltinge J. , McCue, K. (2000) An economist primer on survey samples.
Centre for Economic Studies Discussion Paper.
Cohen, J, and P. Dupas (2010), “Free Distribution or Cost-Sharing? Evidence from a
Randomized Malaria Prevention Experiment” The Quarterly Journal of Economics,
CXXV, 1, 1-34.
Cull, R., Demirgüç-Kunt, A., & Morduch, J. (2007). Financial performance and outreach: A
global analysis of lending microbanks. The Economic Journal, 117 (1), F107-F133.
Dehejia, R., H. Montgomery, and J. Morduch (2011), “Do Interest Rates Matter?: Credit
Demand in the Dhaka Slums,” Journal of Development Economics, in press.
Deutsche Bank Research (2007) Microfinance an emerging investment opportunity: Uniting
social investment and financial return (available at www.dbresearch.com).
Galema, R., R. Lensink and L. Spierdijk (2011). International diversification and
microfinance. Journal of International Money and Finance, 30, pp. 507-515
Gonzalez, A. (2007) Resilience of microfinance institutions to national macroeconomic
events: An econometric analysis of MFI asset quality, MIX Discussion Paper 1.
Hermes, N., Meesters, A. (2009), The performance of microfinance institutions. Do macro
conditions matter? Working paper.
Hermes, N., Lensink, R., & Meesters, A. (2011). Outreach and efficiency of microfinance
institutions”, World Development, 39, pp. 939-948.
Hudon, M (2007). Fair interest rates when lending to the poor, Ethics and Economics, 5:1-8.
Hudon, M (2009). Should access to credit be a right?, Journal of Business Ethics, 84, pp. 1728.
Hudon, M and J. Sandberg (2011). Towards a theory of fair interest rates on microcredit.
Paper presented at the second European research conference on mcirofiannce, July
2011, Groningen.
Huijsman, S. and R. Lensink (2011). Are microfinance institutions affected by the financial
crisis. Applied Economics, forthcoming.
IMF (2009a), Regional Economic Outlook: The implications of the global financial crisis for
low-income countries (available at www.imf.org).
IMF (2009b), Regional Economic Outlook: Impact of the global financial crisis on subSaharan Africa (available at www.imf.org).
IMF (2009c), Regional Economic Outlook: Asia and Pacific (available at www.imf.org).
J.P. Morgan (2009), Shedding light on microfinance equity valuation: Past and present. JP
Morgan Global Research and CGAP (available at www.jpmorgan.com).
Karlan, D. and J. Zinmann (2008), “Credit Elasticities in Less-Developed Economics:
Implications for Microfinance,”American Economic Review, 98, 3, 1040-1068.
18
Karlan, D., & Zinmann, J. (2009). Expanding microenterprise credit access: Using
randomized supply decisions to estimate the impacts in Manila, Unpublished paper,
Innovations for Poverty Action.
Loncar, D., Novak, C., Ciemil, S. (2009) Global recession and sustainable development: The
case of microfinance industry in Eastern Europe (available at
www.microfinancegateway.org).
McIntosh, C., A. De Janvry and E. Sadoulet, “How competition among microfinance
institutions affects incumbent lenders”, Economic Journal, 115, 2005, pp. 987-1004
Microcapital (2008), Microcapital story: The global financial crisis and microfinance
(available at www.microcapital.org).
Microcredit Summit Campaign (2009), State of the microcredit summit campaign report 2009
(available at www.microcreditsummit.org).
Microrate (2009), Cautious resilience–The impact of the global financial crisis on Latin
American
&
Carribean
microfinance
institutions
(available
at
www.microfinancegateway.org).
Pitt, M.M., & Khandker, S.R. (1998). “The impact of group-based credit programs on poor
households in Bangladesh: Does the gender of participants matter? Journal of Political
Economy, 106 (5), pp. 958–96.
Roodman, D., & Morduch, J. (2009). The impact of microcredit on the poor in Bangladesh:
Revisiting the evidence. Working paper 174, Washington, DC: Center for Global
Development.
Walter I., Krauss, N. (2009), Can microfinance reduce portfolio volatility? Economic
Development and Cultural Change, 58 (1), 85–110.
19