Combining social and financial performance: A paradox?

Is social performance profitable, or at least sustainable? The question may be cynical, but
nevertheless relevant if microfinance is to keep its “promise” of being an economically
viable development tool (Morduch, 1999). The sustainable provision of microfinance
services requires strong financial performance. And yet, sustainability is a huge challenge
for institutions that often lack efficient infrastructure and human resources, and serve
highly vulnerable populations. Considerable subsidies and technical assistance have
allowed microfinance institutions (MFIs) to multiply and grow. Nevertheless, such
external support remains limited. MFIs are expected to cover their operating costs and
even generate profits to finance their growth and attract private investors, whose funds
would allow the sector to scale up (Christen and al., 1994). Transparent financial
reporting is key in this respect, to evaluate, manage and incentivize improvements in
financial performance (Von Pischke, 1996).
Since the 1990s, the concept of financial performance has been subject to lively debate.
Despite diverging perspectives, industry players have gradually reached consensus on the
definition of standard indicators for its evaluation. The Consultative Group to Assist the
Poor (CGAP), a consortium of donor organizations that currently has 33 members,
translated this consensus into a set of guidelines (CGAP, 2003) that have been widely
disseminated. While the emphasis on financial performance has boosted the sector’s level
of professionalism, the focus on profitability has at times led institutions to lose sight of
their social mission (Christen, 2001).
The rapid expansion of microfinance and visible success stories among its clients led
most sector stakeholders to take for granted the social utility of MFIs. This relative
marginalization of social performance assessment resulted in a wealth of information on

the financial aspects of microfinance, but very little on the social side, despite it being
microfinance’s raison d’être (Lapenu and Doligez, 2007).
Starting in the early 2000s, several initiatives emerged to promote the development of
tools to measure and manage social performance, defined as the “effective translation of
social mission into practice” (Hashemi, 2007:3). These tools made it technically possible
for social performance assessment to catch up with financial reporting, perhaps offering
even a way to safeguard against mission drift (Copestake, 2007). The work of these social
performance pioneers, initially centered on a small group of committed MFIs, has
become increasingly mainstream. Many of the sector’s most influential donors, regulators
and networks are now urging all MFIs to go beyond anecdotal evidence and
unsatisfactory proxies to develop a framework for social performance monitoring and
improvement (eMFP, 2008).
There are several reasons for this newfound interest. First, the failure of several
institutions due to massive client dropouts and unexpected surges in delinquency has
made it clear that MFIs are not necessarily offering products adapted to the demand. It is
essential to better understanding clients’ needs and reflect on how to best meet this
demand. Second, the sector’s media exposure in the wake of the 2006 Nobel Peace Prize,
awarded to Muhammad Yunus and the Grameen Bank, and the United Nations’ 2005
Year of Microcredit has caused some critics to raise their voices and challenge
microfinance actors to empirically demonstrate their contribution to development and
poverty alleviation (Duflo, 2010). Third, in countries such as Bolivia, Nicaragua, Ecuador
and Benin, governments are starting to oppose MFIs, consider interest rate caps and
rehabilitate state-owned development banks (Bédécarrats & al., 2012).
As criticism of the sector has increased, so has the expectation that MFIs and even social
investors assess and track social performance. But will it be at the expense of financial
performance? There are contradicting viewpoints regarding the pairing of financial
sustainability and social objectives. Some observers suggest an incompatibility, pointing
to problems of mission drift experienced by MFIs that pursue profitability (Cull et al.,

2009) by insisting on physical collateral, large loans and targeting the better-off
(Christen, 2001). Others emphasize synergy, arguing that social performance improves
mutual trust, client participation and satisfaction, which translates into higher repayment
rates and lower transaction costs (Lapenu, 2007). While these assertions draw on case
studies, the research has not been extensive enough to draw sector-wide conclusions.
Our article brings empirical evidence to this debate, drawing on the main findings of an
in-depth analysis of the relationship between social and financial performance. After
taking stock of the evidence that has fueled conventional wisdom regarding the trade-off
between MFIs’ contribution to development and their financial sustainability, we
highlight the diversity and salient trends of these service providers, according to key
parameters, such as size, age, intervention area, charter type and region. Regression
analysis has been conducted to assess the combination of social and financial
performance, revealing trade-off areas, in particular in terms of individual targeting and
range of traditional services, but also synergy effects linked to social responsibility and
quality of services.

  1. Conventional wisdom based on a partial understanding
    Since 2005, the Social Performance Task Force (SPTF), an international working group,
    has worked to set common standards of social performance for the microfinance sector.
    The SPTF defines social performance along four main dimensions that include 1) serving
    larger numbers of poor and excluded people; 2) delivering high-quality and appropriate
    financial services; 3) creating benefits for clients; and 4) improving the social
    responsibility of MFIs (Hashemi 2007). This notion, at the very heart of microfinance’s
    mandate (“do good”), goes beyond the concept of social responsibility (“do no harm”).
    As the importance of social performance becomes increasingly clear, a growing number
    of scholars are studying its correlation with financial performance (in particular: Cornée,
    2006; Flückinger and Vassilev, 2007; Gutierrez-Nieto et al., 2009; Hermes et al., 2008;
    Cull et al., 2009, Mersland and Strøm, 2010; Engels, 2010). In broad terms, research

findings converge, showing a trade-off between social and financial performance, but
conclude that there is no evidence of mission drift, as neither MFI maturity nor size
appear to have clear effect on social variables1

These studies use sophisticated techniques, but rely on unsatisfying or limited indicators,
such as GLP, average loan balance – occasionally weighted by gross national income per
capita – or number of woman borrowers, which unfortunately do not grasp the full
dimension of social performance. They analyze questions that are fundamental to the
sector, such as the breadth and depth of outreach, but use proxies that ultimately do not
reveal very much in this regard. Furthermore, these questions reflect only one of the
many dimensions of social performance (Armendariz and Szafarz, 2009; Dunford, 2002).
Geographical outreach, adapted services and social responsibility, for example, are
ignored. Moreover, they only account for credit operations, neglecting other aspects of
microfinance. This is understandable, as until recently, insufficient information remained
the main obstacle to reliably assessing the link between social and financial performance.
Another constraining fact is the reliability of the sources. In a majority of studies, data
was culled from the MIX or other self-reported databases and remain therefore largely
unverified. Sound results are simply not easy to come by and impact studies are limited,
costly to replicate and difficult to compare (Copestake, 2003). Fortunately, a new field of
evaluation is emerging, offering new perspectives for analysis.
Dewez and Neisa (2009) conducted a study on synergies and trade-offs between social
and financial performance on the basis of 64 MFIs using the ECHOS© social
performance evaluation by Incofin. Using 43 social performance indicators which
consider outreach, client service and social responsibility, and a financial performance
index combining 48 financial indicators, called the Counter Party Risk Score (CRS),
statistical tests and simple regression analysis reveal a significant positive relationship
between social and financial performance. However, the use of a compound social and a

For an in-depth review of literature with more details on each survey, see the complete study
document on CERISE website:

compound financial indicator makes it difficult for MFIs to apply this information in
order to improve their operations and performance.
Gonzalez (2010) conducted the first econometric research based on a large database, with
data from 208 MFIs in 2008 collected through the MIX’s Social Performance Standard
Reports (SPS). He crossed scores in targeting the poor, non-financial services, training on
social performance, client retention, social responsibility to clients and social
responsibility to staff with the MFIs’ level of productivity, portfolio quality, and
efficiency. Findings reveal efficiency trade-offs for targeting the poorest, for staff
training on social performance and for social responsibility to staff, but also synergies for
productivity and staff training on social performance, social responsibility to staff, as well
as for productivity and efficiency with client retention. Nevertheless, Gonzalez also
indicates the need to control for peer factors which are known to influence the financial
performance of MFIs in order to better understand the different implications of social and
financial performance. Gonzalez’ findings represent a turning point in the sector, not only
because he uses more meaningful proxies for social performance than previous studies,
but also because he uses advanced regression analysis which allows for testing the
aggregate relationship between groups of variables and measuring the relative effect of
each single variable. Yet, sample size is still quite limited and data relies on self-reported

  1. How to assess the social performance of MFIs?
    The Social Performance Indicators tool (SPI) measures to what extent a MFI dedicates
    the means necessary to fulfill its social mission. Developed in 2004 in collaboration with
    a wide range of microfinance practitioners, the SPI collects data on 70 indicators that
    measure the objectives, systems and processes of the four key dimensions of social
    performance as defined by the SPTF. Each dimension is broken down into three criteria
    (see Table 1)

Further details regarding the assessment methodology, and the SPI tool are available on CERISE’s
website (


Table 1: The SPI by CERISE
Dimensions Criteria
D1 Targeting and outreach C1.1 Geographic targeting
C1.2 Individual targeting
C1.3 Pro-poor methodology
D2 Adaptation of services C2.1 Range of traditional services
C2.1 Quality of services
C2.3 Innovative and non-financial services
D3 Benefits to clients C3.1 Economic benefits to clients
C3.2 Client participation
C3.3 Social capital/Client empowerment

D4 Social responsibility C4.1 SR to employees
C4.2 SR to clients
C4.3 SR to the community and the environment
Targeting and outreach (Dimension 1) refers to the MFI’s strategies to reach the poor and
excluded. Targeting can be geographic (C1.1), such as when an institution decides to
operate in isolated, remote and poor areas where often no financial services are available.
It can be individual (C1.2), such as when the MFI purposely selects clients based on
poverty levels or exclusion. It can be methodological (C1.3), such as when services are
designed specifically to reach the poor or excluded.
Adaptation of services (Dimension 2) assesses an institution’s ability to provide products
tailored to client needs. This entails offering a range of financial services (C2.1) of high
quality (C2.2) as well as innovative and non-financial services (C2.3).
Benefits to clients (Dimension 3) are at the heart of the raison d’être for microfinance.
Economic benefits (C3.1) alone justify access to financial services, but need an effort
from the MFIs to track and monitor changes and to implement practices to ensure that the
benefits are geared towards the clients. MFIs may also seek to strengthen social
networks, involving clients in their governance (C3.2) or promoting their empowerment
Social responsibility (Dimension 4) extends to employees through appropriate human
resource policies (C4.1), to clients by guaranteeing respect of consumer protection

principles (C4.2), to the community and the environment by taking care to respect the
culture and context in which the MFI operates (C4.3).

  1. Are socially audited MFIs representative?
    Our analysis draws on data from social audits from 2006 up to 2011, retrieved from 344
    SPI evaluations of 295 different MFIs in 51 countries worldwide with an overall outreach
    to more than 12 million borrowers. Social performance data as well as financial data have
    been reviewed for 84% of the evaluated datasets.
    Figure 1. Comparing the scopes of datasets (CERISE SPI audits, MIX Market and MCS)

CERISE: 189 MFIs (57%)
MIX: 347 MFIs (34%)
MCS: 639 (18%)

CERISE: 84 MFIs (25%)
MIX: 150 MFIs (15%)
MCS: 981 MFIs (28%)
CERISE: 12 MFIs(4%)
MIX: 55 MFIs (5%)
MCS: 87 MFIs (3%)
CERISE: 14 MFIs(4% )
MIX: 189 MFIs (19% )
MCS: 68 MFIs (2% )
CERISE: 32 MFIs (10% )
MIX: 287 MFIs (27% )
MCS: 1723 MFIs (49% )

Source: CERISE database (may 2011), MIX 2009 Benchmarks (oct. 2010) and the Microcredit Summit
Campaign Report 2011 (MSC, 2011).
The majority of evaluated MFIs come from Latin America and the Caribbean, followed
by Sub-Saharan Africa. This is due to the active involvement of microfinance networks
and socially responsible investment funds in these two regions. In comparison to the
distribution of MFIs that report to the MIX Market or the Microcredit Summit Campaign
(MCS), the South and East Asian region and Eastern Europe and Central Asia region are
under-represented in the SPI sample.
Figure 22 shows the distribution of the database according to governance type and scale,
comparing MIX benchmarks and the CERISE Database.


Figure 2: Distribution of MFIs according to governance type and scale

7% Credit

(MFI type) Bank

(MFI type)


(MFI Scale)



(MFI scale)


Source: CERISE Database (May 2011) and MIX benchmarks (Oct. 2010)
When we compare to the MIX financial reporting, there seems to be a larger proportion
of NGOs and Credit Unions that go through social audits than other types of institutions.
Small MFIs also appear to be relatively more eager to implement this kind of

  1. Social performance according to MFI type
    SPI scores from the sample are normally distributed with a median at 58.0% and a mean
    of 57.7%. As shown in Figure 3, the majority of MFIs earn more than half of all possible
    points and while only 5% score less than 35%, suggesting they are barely pursuing a
    social strategy. As the SPI enables a comprehensive assessment of the different
    dimensions of social performance, it is not likely for an MFI to score full points in every
    dimension of the SPI. Results should rather reflect the institution’s self-defined mission
    and strategy.


Figure 3. Plotboxes of MFI SPI results: Distribution of SPI results3

0 %
1 0%
3 0%

Total SPI D1 Targeting
the poor and
D2 Adaptation
of services
D3 Benefits for

D4 Social


Source: CERISE Database (May 2011)
Different institutions prioritize different facets of social performance, depending on their
objectives and context. This is why it is so important to refine analysis beyond the
aggregated score and analyze each dimension against the institution’s strategic priorities.
Likewise, comparing scores is only useful when institutions belong to the same peer

3 This kind of graph is commonly used to represent the dispersion on a variable (see for example CGAP
study referring to interest rates). It can be read like this for the total SPI score: “If we rank 100 MFIs
according to their SPI score, the best scores 95%, the 5% with highest scores obtained at least 80%, the
25% with highest scores obtained at least 67%, half of the MFIs obtained 58%. Only a quarter obrained less
than 49%, one upon 20 scored less than 35% and the lowest score was 18%.”


Figure 4: SPI scores per dimension and per criterion of different charter types.

and outreach

of services

D3. Benefits
to clients

D4. Social

Credit Union

2 0%
6 0%
Targeting C1.2

C1.3 Pro-
poor Methodology

C2.1 Range
of services
C2.2 Quality
of services
and NFS C3.1

C3.2 Client


C4.1 SR to
C4.2 SR to
C4.3 SR to

Source: CERISE Database (May 2011)
Figure 4 compares the dimensions of social performance for MFIs of different
governance types. For profit institutions (banks and NBFIs) generally score lower than
non-profit institutions (NGOs and credit unions). Although the total SPI score is similar
for banks and NGOs, their strengths are in different areas. Banks score well in products
and services (D2) and social responsibility (D4) while NGOs stand out for their proactive
targeting (D1). Credit unions show high scores in benefits to clients (D3). Due to their
targeting focus on rural and agricultural regions, credit unions obtain lower results in
individual targeting. NBFIs score the lowest, appearing to be caught in the middle,
having left targeting strategies to NGOs but with no clear policy concerning the
adaptation of services or social responsibility.4

In short, it seems that small NGOs emerge
as targeting champions, but cannot compete with the range of services and social
responsibility policies of banks.
Peer grouping can be done on other criteria, like location, size, maturity, etc. From a
geographic standpoint, the best scores have been recorded in Asia, particularly for
targeting and outreach (D1). Latin American institutions also score well, especially in

4 Rural banks are excluded due to an insufficient representation in the sample.

adaptation of services (D2) with a wide range of traditional services (C2.1) and with good
quality of services (C2.2). Africa tends to score high in benefits to clients (D3) with
strong results in terms of client participation (C3.2). This result might be due to the
presence of member-based organizations.
In terms of maturity, social performance increases with age. At the beginning, MFIs rely
on a small and committed team and flexible processes. Therefore most of the practices
related to social mission remain informal and are not taken into account by SP
assessments, which evaluate institutionalized processes. Nevertheless, as they grow,
institutions tend to manage only what they can measure and systematize. MFIs wanting to
pursue initial objectives of poverty reduction or development ultimately formalize their
practices. This trend is also observed when comparing MFIs according to their scale
(GLP volume) and outreach (number of borrowers). Large MFIs score better in
adaptation of services (D2) and social responsibility (D4) but are weaker in targeting the
poor and excluded (D1) and benefits to clients (D3).

  1. Links between social and financial performance
    We studied the relationship between social and financial performance by asking
    ourselves: “If an MFI improves its contribution to development and poverty reduction,
    how does this impact its financial performance?”
    To answer this, we used multivariate regression analysis. This involved building models
    to predict financial performance variables, namely productivity, portfolio quality
    (including PAR30 and write-off ratio), operational expense ratio, return on assets and
    operational self sufficiency. The models tested whether these financial performance
    variables were determined by, social performance variables. The latter were either
    standalone social performance indicators, or compound indexes with several social
    performance indicators (i.e. the four dimensions and 12 criteria of the SPI).
    We modeled all possible combinations of variables and controlled for other types of
    variables that may influence directly or indirectly social and financial performance, such

as size, age, profit or not-for-profit status, rural or urban, and target market. Finally, we
selected the most significant and most informative models5

. The following table
summarizes the statistically significant relationships validated with data from 295 MFIs.

5 The Akaike Information Criterion and Schwarz Bayesian Criterion take into account both the
statistical goodness of fit and the number of parameters which have to be estimated to achieve this
particular degree of fit, by imposing a penalty for increasing the number of parameters. Lower values of the
index indicate a better fit and thus the preferred model, which is the one with the fewest parameters that
still provides an adequate fit to the data.


Figure 5. Results from regression analysis of SP and FP

C.1-1 Geographic targeting 
C.1-2 Individual targeting 
9 Women borrowers
C.1-3 Pro-poor methodology
C.2-1 Range of services  
C.2-2 Quality of services 
9 Reasonable interest rate 
9 Client retention
C.2-3 Innovative and NFS 
9 Innovative services 
9 Non-financial services 
C.3-1 Economic benefits  
9 Track changes
9 SP Training + appraisal 
C.3-2 Client participation 
C.3-3 Empowerment
C.4-1 SR to staff 
C.4-2 SR to clients  
9 Avoiding over-indebtedness 
9 Client Protection Policies 
C.4-3 SR community/env. 
significant synergies between
SP and FP Geographic …
significant trade-offs between
SP and FP 9 Women …
Summary of results from social-financial performance analysis

Dimension 2: Adaptation of services
(Borrowers/staff) Financial indicators
Social audit scores per dimension (aggregated indicators)


Scores in single SP indicators
Scores per criteria in the social
Dimension 1: Targeting and outreach

Dimension 3: Benefits for clients

Dimension 4: Social responsibility

Source: CERISE Database (May 2011)

a. Productivity increases with geographic targeting but decreases with service
diversification and prevention of over-indebtedness
Here productivity is measured by the ratio of number of borrowers per staff member.
Several models appear significant when crossing productivity with social variables. The
most informative is the following:

P = 194.571 +118.24(p=0.000)GT -67.027(p=0.086)Range -44.242(p=0.000)Avoiding OI -21.664(p=0.042)TM.

N: 151 Adjusted R2: 0.085
P: Productivity GT: Geographic targeting (C.1.1) Range: Range of services (C2.1) OI: Over-indebtedness TM: Target market


This model shows that a 10% increase in the score for geographic targeting (C1.1)
increases productivity by nearly 12 borrowers per staff, whereas an additional 10% in
range of services (C2.1) will decrease productivity by 6.7 borrowers per staff. MFIs with
compliance systems for avoiding over-indebtedness lose in productivity by 44 borrowers
per employee. If MFIs lower their average loan size by US$ 1, they are able to serve 21
more borrowers per staff member.
 Explaining synergies between social performance and productivity
Regarding geographic targeting, a stronger focus on poor and excluded areas induces
higher staff productivity. As some studies suggest (Hirschland et al., 2008), this is
probably because they allow MFIs to operate in less competitive markets and are often
associated with greater client participation. Participatory models allow MFIs to overcome
some of the operational difficulties inherent to working with low-income populations.
Moreover, analysis of social performance profiles reveals that MFIs that operate in the
most deprived zones often serve the whole local population instead of focusing on the
poorest in the area. This may increase staff capacity to serve a greater number of

MFIs targeting a lower market usually are more productive in terms of clients per
borrower. Smaller loans are easier and faster to disburse as they can be granted without or
with small collaterals. This may also be related to the lending methodology, considering
that small loans are often distributed through groups, which is proven to be less staff
consuming than individual lending. Moreover, they usually serve more clients.
Other models6

indicate that productivity benefits from synergies with lower interest
rates, client participation and empowerment. MFIs with reasonable prices7

serve more
clients per staff. This could be due to their commercial advantage over more expensive
competitors. Regarding client participation, it is obvious that clients involved in
operations and governance share the work burden with staff. Surprisingly, there is no
evidence at this stage that productivity can be improved by socially responsible action
towards employees or with specific training and appraisal on social performance as was
found by Gonzalez (2010).
Interpreting trade-offs with productivity
The fact that a wider range of services (C2.1) is associated with lower productivity is
easily understandable. Indeed, product diversification implies a higher complexity of
internal processes and the multiplication of transaction types keep staff from serving a
large number of borrowers. Nevertheless, the results would be different if we had a
productivity indicator less focused on credit. For example, total clients per staff leads to a
different result, as product diversification implies a larger proportion of clients that are
not borrowers, and in particular savers. As such we do not find any trade-off between
range of services (C2.1) and costs. Social responsibility to clients in general, and active
policies to avoid over-indebtedness in particular, appear to limit the number of
borrowers per staff. This is probably because it reduces disbursement pressure on loan
officers and increases the time spent on assessing reimbursement capacity. Nevertheless,

See CERISE website for the complete study with all the models and technical details.
7 The level of interest rates is analyzed by comparing the cost of funds to the effective interest rates. An
MFI is considered to have a reasonable interest rate if this spread is inferior to 30.

we do not observe effects on portfolio quality, maybe because such client protection
measures are relatively recent.
Moreover, systems implemented to ensure economic benefits to clients, such as impact
studies, staff trainings or appraisal based on SP, etc. are also found to significantly
diminish productivity, possibly because of the burden they imply for the MFI workers.
b. Portfolio quality improves with social responsibility to staff and quality of
For analyzing portfolio quality, a compound variable was created using the sum of the
portfolio at risk at 30 days (PAR30) and the write-off ratio (WOR). The most informative
model built with social performance variables is the following:
Arrears = 0.178 -0.071(p=0.049)Quality -0.073(p=0.092)SRs +2.871E-12

N: 302 Adjusted R2: 0.047

Arrears: PAR + WOR Quality: Quality of services SRs: Social responsibility to staff GLP: Gross loan portfolio
This model means that MFIs reduce WOR and PAR30 by 0.71% when the SPI score for
quality of services (C2.2) increases by 10%. An increase in the score for social
responsibility to staff (C4.1) by 10% also translates into a reduction of arrears of 0.73%.
The scale of the MFI (GLP) serves as a control variable and improves the model’s
informative content, but has no significant direct influence on portfolio quality.
 Explaining synergies between social performance and portfolio quality
Quality of services in general, and reasonable interest rates in particular, appear to
reduce PAR30 and WOR. This might be due to the competitive advantage and higher
customer satisfaction of MFIs that offer lower prices. It could also reinforce clients’
reimbursement capacity, which consequently reduces delinquency and default. Fair
working conditions and training also raise portfolio quality. Such aspects of social
responsibility to staff, which includes training and career opportunities, may encourage
improved portfolio management, better client assessment, and greater employee

commitment. Nevertheless, when controlling for age, we observe that this applies only
for new and young MFIs, not for mature ones.
c. Efficiency increases with improved credits and savings products but reduces with
poor client selection and non-financial services
Efficiency is measured through the operational expense ratio. In our sample, large MFIs
and MFIs with large average loan amounts have the lowest OER. The sample also shows
that rural MFIs are much more cost efficient than urban MFIs, which confirms findings
by Gonzalez (2010). Of the dozen models tested for regression analysis, two appear
particularly significant. The first one is valid for all MFIs and the other one only applies
to large ones.
OER = 0.287 +0.123(p=0.003)IT -0.068(p=0.150)Range -0.210(p=0.000)Quality +0.106(p=0.019)I&NFS -0.082(p=0.09)EB +0.109(p=0.02)SRc -0.092(p=0.049)SRe. N: 344 Adjusted R2: 0.135 BIC: -1007 OER = 0.381 +0.074(p=0.057)IT -0.110(p=0.014)Range -0.182(p=0.000)Quality
+0.083(p=0.048)I&NFS +0.072(p=0.099)SRc -4.659E-10

(p=0.011)GLP -0.092(p=0.000)NPI

-0.035(p=0.082)RI -0.032(p=0.005)TM.

N: 299 Adjusted R2: 0.222 BIC: -973
OER: operational expense ratio IT: Individual targeting Range: Range of services Quality: Quality of services I&NFS:
Innovative and non-financial services EB: Economic benefits for clients SRc: Social responsibility to clients SRe: Social
responsibility to community and environment GLP: Gross loan portfolio NPI: Non-profit institution RI: Rural intervention TM:
Target market (AvLoanPerGNIpc)
The first model means that an increase of 10% in the score for individual targeting
(C1.2), innovative and non-financial services (C2.3), or social responsibility to clients
(C4.2) creates an increase in OER of 1.23%, 1.06%, and 1.09%, respectively. But on the
other hand, a wider range of products (C2.1) and better quality (C2.2) of services, which
improve efficiency by 1.1% and 1.82% for every increment of 10%. A 10% increase in
the SPI score for economic benefits to clients (C3.1) or social responsibility to the
community and the environment (C4.3) reduces costs by respectively 0.82% and 0.92%.
The second model controls for GLP, for profit or non-profit status, rural intervention and
target market. It shows that a 10% score improvements in individual targeting (C1.2),
innovative and non-financial services (C2.3), and social responsibility to clients (C4.2)
cause a loss in efficiency of respectively 0.74%, 0.83%, and 0.72%. Nevertheless, wider

range (C2.1) and better quality (C2.2) of services, which improve efficiency by 1.1% and
1.82% for every increment of 10%. With control variables, we see that OER also declines
when the MFI is larger, not-for-profit and disburses relatively large loans.
 Interpreting the synergies of SP with efficiency
Quality of services seems to reduce operational expenses. This is probably because
quality of services includes reasonable interest rates, client retention and other aspects
that make the MFI more attractive to clients and enhances their retention, therefore
reducing the cost of enrolling new ones. A similar relationship is observed with the range
of traditional (i.e. savings and credit) services, but it doesn’t apply to MFIs serving a
low-end market. Social responsibility towards community and the environment also
improves efficiency. It probably enhances the MFI’s reputation and acceptance in the
community. Moreover, analysis confirms the importance of economies of scale for both
social and financial performance: it seems that the gross loan portfolio volume improves
Promoting economic benefits for clients appears to improve efficiency, but when more
specific variables are included in the analysis, the relationship is not so straightforward.
We observe that this result is related to one specific indicator: the inclusion of social
performance parameters in incentive schemes for staff remuneration (such as outreach to
poor, women or rural clients, client retention, etc.). A higher score in this criterion
implies the existence of such incentive schemes, and these methods are likely to improve
the overall operational performance of the MFI.
Non-profit institutions also happen to have lower expense ratios, but this may be related
to other factors, such as subsidization of some activities, the absence of regulatory
constraints and their inability to take savings. Similarly, MFIs operating in rural areas
are more efficient. This is probably due to the cost reduction strategies they develop, such
as participation (see 5.a), to cope with an inherently high-cost environment.

There is a clear trade-off between efficiency and individual targeting. Institutions that
directly target poor and excluded clients through individual targeting strategies tend to
have higher operational expenses. Such a trade-off is typically explained by several key
factors: Morduch (2000) points out the proportionally higher cost induced by smaller
transaction amounts, while Hashemi and Rosenberg (2006) emphasize the higher risks
and lack of guarantees inherent to this clientele, their reticence to join microfinance
programs and the challenges of providing the non-financial support to this population.
It is also the case with innovative and non-financial services, as additional services are
naturally associated with additional expenses. In terms of social responsibility to clients,
efforts and policies for client protection also induce higher costs. Moreover, it is harder to
target a lower market, as smaller transaction amounts induce proportionally higher
d. The complexity of operational self sufficiency and financial sustainability
Other analyses have been conducted for more complex financial performance indicators,
in particular Operational Self Sufficiency (OSS) and Return on Assets (ROA). Several
significant results stand out, but since they are compound financial variables, they are
determined by a number of factors in addition to social criteria, such as size, target
market, charter type, etc. The relationships must be explained by a combination of
models that are too complex to summarize here. A complete document, available online,
provides a detailed account of these results.
Moreover, several members of the SPTF (Microfinanza, see box below, Incofin, MIX)
are conducting similar research, in order to better understand the complex relationship
between social and financial performance. Other members intend to do the same in the
near future.
“U” shaped relationship between client protection and financial performance
By Micol Guarneri and Lucia Spaggiari, Microfinanza Rating

Hoepner, Liu and Wilson from the University of St Andrews and the University of

performed a multivariate regression analysis on the MicroFinanza Rating

database of 87 Social Ratings9

. Even if more research on larger rating samples will help
to better explain some interactions between social and financial performance, the
interesting results on client protection are already worth sharing with the industry. The
nonlinear model used shows a significant relation between the client protection score
obtained by the MFI in the social ratings and the MFI profitability (ROE), adopting a “U”
shape. A similar “U” parabolic function is found between client protection and
sustainability (FSS). Improving the client protection from a weak to an adequate level is
associated with financial costs, but upgrading the client protection from adequate to good
and very good goes along with higher ROE and FSS. Building client protection systems
from scratch can be costly, but the MFI’s efforts on client protection will pay off once the
MFI has reached the client protection “minimum critical mass” necessary to build the
clients’ loyalty and the government and investors’ trust. The composite nature of the
client protection and financial performance relation is in line with CERISE results

(Combining social and financial performance: a paradox? 2011), where avoiding over-
indebtedness reduces the productivity, while reasonable interest rates increase the

portfolio quality. The social and financial interaction may not always be linear: the
marginal loss in financial performance associated with an increase in the social
performance may reduce for higher levels of social performance, and even convert from
loss to gains, once a certain level of social performance is achieved. Investing in client
protection is in the MFIs best interest not only because the reputation risk needs to be
managed, but also because the financial benefits of client protection are very likely to
outweigh its costs once adequate practices are achieved.

8 Work in progress to be published by the end of 2011: Do microfinance institutions (MFIs) pay for social
responsibility? Evidence from social ratings of MFIs, 2011.
9 Refer to Guarneri, Moauro, Spaggiari “Motivating your BoD to actively promote and deepen the social


Client protection


Thanks to recently developed simple and reliable methods to assess social performance,
we can now evaluate microfinance’s ability to achieve its double or triple bottom line.
Our analysis confirms what many studies have suggested based on incomplete data and
basic analysis: social performance and financial performance are compatible. Therefore,
the double bottom line is no “mission impossible” but can be achieved when trade-offs
and synergies are combined cleverly following a well planned social performance
management strategy.
Individual targeting (i.e. purposely selecting clients based on poverty level or exclusion)
clearly implies higher transaction costs for financial institutions. Nonetheless, this study
brings evidence that with the right strategy and over time, lost efficiency can be regained
through other elements of social and responsible performance. In the end, doing socially
responsible microfinance is neither less efficient nor less profitable. The key is simply to
find the right mix of social performance practices that will ensure financial sustainability.
Firstly, the targeting methodology has to be chosen smartly. In general, individual
targeting is costly due to the work involved in screening out less or non poor potential
borrowers; moreover, it implies limited loan amounts. Nonetheless, individual targeting
might be combined with a wide range and high quality services, which has a positive
effect on efficiency and sustainability. Client retention, in particular, can be improved
when services meet clients’ needs. Participatory structures, too, might be associated with

individual targeting, as they help keep operational costs low. Furthermore, geographic or
pro-poor targeting methodologies represent more financially efficient alternatives.
Secondly, the idea of creating social value through microfinance requires additional
services adapted to the profile of the target market, such as non-financial services and
client protection. Although non-financial services and social responsibility towards
clients are primarily expenses and cause a drop in efficiency as well as sustainability,
synergies are created at the same time. Customer satisfaction and payback capacity are
improved, and these in turn are proven to lead to higher retention rates which in turn
results in better portfolio quality and higher efficiency. Moreover, there is potential for
significant impact in the long run. Larger MFIs should aim at overcoming the trade-off
with non-financial services and social responsibility to clients by focusing on a wide
range of high quality services, which improve efficiency due to fewer drop-outs and
customer satisfaction. Small MFIs, which cannot profit from scale effects, might
implement participatory structures which, on the one hand, lead to higher productivity,
higher ROA and cost reduction; and on the other hand, services that are better adapted to
clients (thanks to client input), and therewith satisfaction, retention, and ultimately
In conclusion, our findings give statistically based evidence that funders should not
ignore MFIs’ investments in social responsibility to clients or staff, in quality
improvements, in non-financial services, or in avoiding over-indebtedness, whether
socially or financially driven. Our analysis indicates the need to go back to the basics.
Providing inclusive, appropriate services in a responsible way that clearly benefits
clients, positively impacts the fundamentals of financial sustainability: productivity,
efficiency and portfolio quality.

Leave a Comment

Your email address will not be published. Required fields are marked *