THE DETERMINANTS OF FINANCIAL PERFORMANCE

OF MICROFINANCE INSTITUTIONS IN MOROCCO:

A PANEL DATA ANALYSIS

LAHCEN EL KHARTI1

Abstract
This paper studies the determinants of financial performance of microfinance institutions
(MFIs) in Morocco. Results show that the Portfolio at Risk (PAR30) and the age of the MFIs
are the main determinants of financial performance of these institutions. Results also indicate
that the outreach of MFIs’ microfinance programs positively affects their financial perform-
ance. Moreover, there is a significant impact of the share of equity in total assets, staff produc-
tivity and the percentage of female clients on MFIs’ financial performance.
Key words: microfinance, financial performance, portfolio at risk, Morocco, 2008 crisis.
JEL: D63, G01, G15, G23, G32.

1. INTRODUCTION

In recent years, the use of microfinance as an innovative tool for poverty

alleviation among people who are economically active but financially con-
strained and vulnerable has gained increasing attention by both policy mak-
ers and regulators in many countries. Indeed, in 2011, microfinance institu-
tions (MFIs) provided microcredit to more than 124 million households liv-
ing in extreme poverty, according to the 2013 State of the Sector Report pub-
lished by the Microcredit Summit Campaign.
In particular, the Moroccan microcredit industry has experienced one of
the most rapid and significant growths seen in the microfinance industry. In
just four years, from 2003 to 2007, MFI loan portfolios multiplied 11 times
and client outreach by four, according to the Microfinance Information Ex-

1 E-mail: [email protected]

27
SAVINGS AND DEVELOPMENT - No 1 - 2014 - XXXVIII

change (MIX)2. These results make the microfinance industry in Morocco as

one of the most extensive and vibrant sector in the Arab world. This success
was largely attributed to the support of the Moroccan Government having
established the Microfinance Law in 1999 and established the Hassan II
Fund to help capitalize the first MFIs, as well as foreign funders and donors
who supported the growth of the sector, and the local talent in Morocco.
This rapid growth soon proved unsustainable and signs of stress resur-
faced in 2007. Nonperforming loans started to rise significantly from one of
the lowest levels in the world, 0.42 percent in 2003 to 1.9 percent in 2007.
From late 2008, the portfolio risk began to increase considerably from an av-
erage less than 1% at sector level between 2003-2007 to over 5% in 2008 and
over 10% in 2009 and 2010. The cost of write offs became one of the major
costs for MFIs, weighing heavily on their profitability and solvency. Conse-
quently, many MFIs were restructured in order to achieve financial sustain-
ability and finance their growth. Indeed, the Moroccan MFIs have to be fi-
nancially sustainable in order to guarantee a large-scale outreach to the poor
on a long-term basis. Currently, there is no evidence about the key determi-
nants of financial performance of Moroccan MFIs. Therefore, this study at-
tempts to fill a gap in the literature by analyzing the main drivers of the fi-
nancial sustainability in the Moroccan microfinance industry.
The empirical literature investigating the financial performance of micro-
finance institutions is scarce and still growing. Hartarska (2005) analyses the
impact of governance on the sustainability of microfinance institutions
(MFIs) in Central and Eastern Europe and the Newly Independent States.
The results indicate that performance-based compensation of managers is
not associated with better-performing MFIs; however, managers’ experience
improves performance. Cull et al. (2007) examine patterns of profitability,
loan repayment, and cost reduction with unusually high-quality data on 124
institutions in 49 countries. Their results show the possibility of earning
profits while serving the poor, but a trade-off emerges between profitability
and serving the poorest. Mersland and Strøm (2009) explore the relationship
between firm performance and corporate governance in MFI using a self-
constructed global dataset on MFIs collected from third-party rating agen-
cies. Results show that financial performance improves with local rather
than international directors, an internal board auditor, and a female CEO.
Mersland et al. (2011) examine how various aspects of international influ-
ence affect MFIs’ financial and social performance. They find that the inter-

2 MIX is a non-profit organization that acts as a business information provider in the micro-
finance sector. (www.mixmarket.org).

28
 L. EL KHARTI - THE DETERMINANTS OF FINANCIAL PERFORMANCE OF MICROFINANCE INSTITUTIONS IN MOROCCO

nationalization of MFIs enhances social performance to a large extent, but

does not enhance financial performance. Tchakoute (2011) investigates the
relationship between the legal status of MFIs and their performance. He
shows that the performance of private corporations is better than that of
NGOs only when portfolio quality is used as an indicator for measuring per-
formance. Ben Soltane (2013) studies the relationship between MFIs gover-
nance quality and repayment performance using data of 250 African MFIs.
He finds a negative link between the governance quality and the percentage
of portfolio at risk. Esubalew et al. (2013) examine the effect of competition
among MFIs on their performance. Their results show that intense competi-
tion is, overall, negatively associated with performance of MFIs.
The present paper contributes to the existing literature by analyzing the
financial performance of Moroccan MFIs using single country data. Results
show that credit volume, productivity ratio and the equity-to-asset ratio im-
proves financial performances. However, the portfolio at risk over 30 days
and the percent of female borrowers decrease financial sustainability.
The remainder of the paper is organized as follows: Section two presents
a literature review that focuses on issues and assumptions on financial per-
formance. Section three describes the methodology and data. Section four re-
ports and discusses estimation results and Section five concludes.

2. BACKGROUND AND HYPOTHESES

There are two main approaches to microfinance in the literature. The first
approach focuses on the social exigency of poverty alleviation (defended by
the Welfarists’ School: Morduch, 2000; Dunford, 1998; Woller et al., 1999 and
Brody et al. 2003). This approach assesses success by how well improves the
immediate welfare of clients. The second focuses on the economic exigency
of institution viability (defended by the Institutionalists’ School: Gonzalez-
Vega, 1993; De Briey, 2005). This approach evaluates success by the progress
of the institution in achieving financial self-sufficiency.
In this study, we were inspired from the Institutionalists’ approach in or-
der to determine the financial performance drivers because the followers of
this school believe that the best method of reaching the great majority of
poor persons without accessing financial services3 is to incorporate microfi-
nance industry in the formal financial system.

3 Nowadays, the share of Moroccan people who access to financial services does not exceed
40% (Honohan, 2008).

29
SAVINGS AND DEVELOPMENT - No 1 - 2014 - XXXVIII

2.1. Financial Performance of MFIs

The financial performance of a MFI can be defined as its ability to cover
its operational and financial costs. The common indicator of the financial
performance is the return on assets (ROA), which reflects the profit margin
as well as the effectiveness of the institution (Bruett, 2005; Cull et al., 2007;
Hartarska, 2005; Lafourcade et al., 2006; Mersland and Strøm, 2009). MFIs of-
fer loans at subsidized rates (below the rate of the market) that reduce the fi-
nancial burden and therefore increase net income. These MFIs are heavily
subsidized and then realize at the same time strong profitability.

2.2. Interest rates and equity

Unlike commercial banks, the compensation structure of lending to MFIs
should include the interest rate as the cost of money, the board of education,
the Monitoring Committee and the advisory committee of recovery. As the
share of MFIs’ equity, MFIs can also increase their lending, either through
debt or accepting larger deposits from their customers. This result leaves to a
further increase of the scope of their work, and their financial sustainability.
Hence, in order to achieve financial sustainability, MFIs should set sufficient-
ly high interest rates to cover its expenses.
Hypothesis 1: The application of interest rates that increases equity
should contribute significantly to the achievement of the financial viability
of MFIs. We expect a positive relationship between the share of equity in as-
sets and the financial performance.

2.3. Quality of the loan portfolio

Group loans with joint and several guarantees are often used as a mecha-
nism to minimize the risk of failure, improve the performance of MFI portfo-
lios, and achieve financial viability (Pitt and Khandker, 1998; Ghatak 1999;
Armendariz de Aghion and Morduch, 2000; Laffont and Nguessam, 2000).
Cull et al. (2007) note that group loans have a positive impact on the quality
of the loan portfolio of MFIs. Indeed, they show that the portfolio risk in-
creases the interest rate applied to most MFIs using individual loans, and
that beyond the threshold of 60%, an increase of these rates is not associated
with long-term profits for the latter, which is not the case for MFIs that rely
on lending methodology to solidarity groups.
Hypothesis 2: The risk reduction in the overall portfolio of the MFI
would profit MFI more meaningful and therefore has a positive impact on
the financial performance of MFIs. We expect a negative relationship be-
tween portfolio risk and financial performance.

30
 L. EL KHARTI - THE DETERMINANTS OF FINANCIAL PERFORMANCE OF MICROFINANCE INSTITUTIONS IN MOROCCO

2.4. Staff productivity

MFIs that affect a portion of the salary to the financial performance lead
to positive expressed satisfaction outcomes (CGAP, 2006). Staffs with proven
technical expertise in the field of microfinance are also a prerequisite for the
selection and monitoring of projects for funding. Furthermore, MFIs should
adopt practices by focusing more on the customer relationship in order to in-
crease their performance management (Churchill, 2000; Schreiner, 2003;
Norell, 2001).
Hypothesis 3: An efficient management through investment in human
and technological resources and financial incentive mechanisms should de-
cline staff costs and increase the productivity of loan officers. Therefore, we
expect a positive relationship between the ratio of personal productivity and
financial performance.
Given that the majority of MFI clients are women, and the experience of
MFIs over the years can help them to refine their selection strategies of clients
and portfolio management at risk, we decided to assess the impact of these ele-
ments on the financial performance of MFIs through the following hypothesis:
Hypothesis 4: The age of MFIs, the outreach of their actions (credit vol-
ume and number of clients) and the percentage of female clients should have
a positive impact on the financial performance of MFIs.

3. METHODOLOGY AND DATA

3.1. Model specification

This paper is focused on Moroccan MFIs’ financial performance – the de-
pendent variable. According to the microfinance literature (Bruett, 2005; Har-
tarska, 2005; Cull et al. 2007; Lafourcade et al., 2006; Mersland and Strøm,
2008 and 2009, Adair and Berguiga, 2010), the return on assets, the opera-
tional self-sufficiency, and the financial self-sufficiency are the main indica-
tors that attempt to capture the complexity of financial performance within
the microfinance industry. In this study, we used return on assets (ROA) and
return on equity (ROE) to proxy financial performance because the micro fi-
nancial intermediation in Morocco is specific4 and MFIs were subsidized.
The ROA is a general indicator that measures the profitability regardless of

4 Morocco has specific legislation on microcredit, which also allows competition between
MFIs to develop in a favorable climate. Indeed, the country has created a unique law in 1999 for
MFIs (NGOs) in the territory. This law sets the maximum amount of loans, the maximum lend-
ing rates and prohibit the collection of deposits.

31
SAVINGS AND DEVELOPMENT - No 1 - 2014 - XXXVIII

the financial structure underlying the institution. Moreover, as Moroccan

MFIs are heavily subsided, the ROE allow us to capture this specificity.
We include seven independent variables, which are commonly used in re-
cent microfinance financial performance research. The first variable is the
portfolio at risk over 30 days (PAR> 30). It measures the percent of the total
loan portfolio that has at least one payment overdue by more than 30 days.
Recall that the functioning of MFIs is characterized by the logic of revolving
credits, that is loanable funds are restored almost completely by credits. The
second variable is the equity-to-asset ratio, which assesses the solvency of
MFIs. It measures the amount of capital required to cover additional unex-
pected losses to ensure that the MFI is well capitalized for potential shocks.
The third variable is the productivity ratio, which measures the number of
active clients compared to the number of loan officers or staff. This ratio
works with the following logic: the higher the ratio is the more productive
the MFI. The remaining independent variables capture institutional and out-
reach indicators, namely the percentage of female borrowers, the size of
credit, the age of MFIs and the number of clients (See table 1).
Based on the above discussion, a model of MFI financial performance can
be estimated following the two specifications (1 and 2):

ROAit = f (Ratioassit + Lncreditit + Lnclientit + Pfemit + Empr_Efit

+ Matureit + Par30it) [1]
ROEit = g (Ratioassit + Lncreditit + Lnclientit + Pfemit + Empr_Efit
+ Matureit + Par30it) [2]

Where i = 1,…; N for each MFI in the panel and t = 1,…; T refers to the time
period. Table 1 presents the variables.

3.2. Data
Table 2 presents some economic and financial indicators for Morocco. In
2009, 8.8% of Moroccan population is considered poor compared with 14.3%
in 2006 – a notable achievement for a country of 32 million people that lacks
significant natural resources. GDP per capita has sharply increased from 2006
to 2009, jumping by an average of more than 7%. This economic growth seems
contribute to the Morocco’s declining poverty rate. Indeed, according to the
Moroccan High Commissariat of Planning, a per capita economic growth of
1% reduces the poverty rate by 2.9 percent. Moreover, MFIs have played an
important role in poverty reduction, due to their activities in local develop-
ment through formal partnerships with the government and local authorities.

32
 L. EL KHARTI - THE DETERMINANTS OF FINANCIAL PERFORMANCE OF MICROFINANCE INSTITUTIONS IN MOROCCO

Table 1: Description of variables

Variables Definitions
Dependant variables
ROA Return on asset ratio = net operating income / total assets
ROE Return on equity ratio = Net Income / Average equity
Independent variables
Ratioass Equity-to-asset ratio = Total equity/total asset
Lncredit Number of Loans in natural logarithm
Lnclient Number of Active Borrowers in natural logarithm
Pfem Percent of female Borrowers = Number of active female borrowers/Adjusted
Number of Active Borrowers
Empr_Ef Productivity ratio = total number of active borrowers / total number of staff
Mature Age of MFIs = number of years functioning as an MFI
Par30 Portfolio at risk > 30 days = (Outstanding balance on arrears over 30 days + total
gross outstanding refinanced (restructured) portfolio) / total gross portfolio.

Table 2: Macroeconomic Indicators of Morocco

Indicators 2006 2009
Population (millions) 31,224 31,276
GDP per capita, PPP 3890$ 4167.5$
Population below national poverty line (2$/ jour) (%) 14,3% 8.8%
Number of MFIs 11 10
Number of active borrowers 1,045,310 915,839

Source: Word bank databank, MIX Microfinance World: 2010 Arab Microfinance Analysis & Bench-
marking Report and High Commissariat of Planning, MDG Report Morocco.

Our analysis is based on a panel of 10 MFIs covering the period 2003-2010.

These are Al Amana, AMOS, AMSSF/MC, Al Karama, ATIL/MC, FBPMC, Za-
koura, FONDEP, INMAA and ARDI. The dimensions of the panel data set are
chosen to include as many MFIs as possible each of them a reasonable time
length of observations. Data came from the Microfinance Information Ex-
change (MIX) database.
According to Morvant-Roux et al. (2014), four organizations dominated
the sector: Al Amana, Zakoura, Fondation Banque Populaire pour le Micro-crédit
(FBPMC), and Fondation pour le Développement Local et le Partenariat
(FONDEP). These four leaders shared 97% of the sector’s loan portfolio in

33
SAVINGS AND DEVELOPMENT - No 1 - 2014 - XXXVIII

the end of 2009, and ranked among North Africa’s biggest microcredit insti-
tutions. Al Amana had issued 45% of all the active loans.
While Table A1 in the appendix reports the means and standard devia-
tions as well as the minimum and maximum, Table A2 in the appendix re-
ports the mutual correlation coefficients for the dependent and explanatory
variables in the econometric analysis, respectively. According to table A1,
Moroccan MFIs have a ROA of 3.44% over the entire period from 2003 to
2010. The difference between Min and Max clearly shows that there are large
differences in profitability among the Moroccan MFIs. The same holds true
for our second main profitability measure, the ROE, which amounts to 4.53%
on average. However, as indicated by the Min values, the 2008 crisis has
largely dropped the profitability of MFIs. The average value of the equity-to-
asset ratio is 0.47. The average PAR ratio (3.33%) is less than the threshold of
5% (CGAP, 1999). This indicates that, on average, the credit portfolio in our
sample is healthy. The mean percentage of female borrowers is nearly two-
thirds (64.51%), ranging from only 31.84% to a 97.98%.
Following Kennedy (2008), correlations need to exceed the limit value of
0.8 to detect collinearity between two variables. All the correlation coeffi-
cients in Table A2 are smaller than 0.8. Then, there is no problem of multi-
collinearity (see table A2 in Appendix).

3.3. Estimation strategy

Between 2003 and 2007, the Moroccan microfinance sector has experi-
enced an amazing growth phase and prosperity, characterized by high-per-
forming institutions and supported not only by the local authorities but also
by international donors. However, since late 2007, the microfinance industry
has been facing a severe economic and managerial crisis since institutional
capacity of MFIs has been enjoyed an unprecedented growth. This has re-
sulted in lax credit policies, obsolete information and management systems,
deficiencies in internal control and weak governance. Therefore, it is essen-
tial to consider the crisis in our model estimation. For this, we divide our
study period into two sub-periods: a sub-period before the crisis (2003-2007),
reflecting the growth of microfinance phase and a sub-period after the crisis
(2008-2012), characterized by a restructuring of the microfinance sector. In
the latter sub-period, there are nine MFIs because Zakoura was taken over in
May 2009 by FBPMC when it proved financially inefficient. For this reason,
the overall sample period is from 2003 to 2010 because data for Zakoura do
not exist after 2010 and we cannot remove Zakoura from the sample data
since it is the second largest MFI in the country.

34
 L. EL KHARTI - THE DETERMINANTS OF FINANCIAL PERFORMANCE OF MICROFINANCE INSTITUTIONS IN MOROCCO

4. EMPIRICAL RESULTS AND DISCUSSION

Table 3 summarizes the results of panel regression for the whole sample
period 2003-2010. The results of the Hausman test (Prob>chi2 > 10%) show
that the random effects model should be chosen over the fixed effects model.

Table 3: Econometric results, period 2003-2010

VARIABLES ROE ROA
Ratioass 0.49 (9.32) 1.18 (1.17)
Lncredit -3.92 (3.68) -1.61*** (0.46)
Lnclient 3.73 (5.12) 2.68*** (0.63)
Pfem -0.005 (0.006) -0.001** (0.0008)
Empr_Ef 0.0005 (0.005) 0.0001 (0.0007)
Mature 6.63 (14.65) 0.014 (1.85)
Par30 -4.65*** (1.37) -0.327** (0.17)
Constante 37.84 (43.79) 1.6 (5.31)
Number of MFIs 10 10
Number of observations 80 80
R-sq: within 0.2 0.27
Hausman test (Prob>chi2) 0.5082 0.4574

*** p<0.01, ** p<0.05, * p<0.1, Standard errors in parentheses.

Results do not support hypothesis 1, which share of equity in assets ratio

is a determinant of Moroccan MFIs’ financial performance since the coeffi-
cient on Ratioass is not significant in any of the two specifications. This find-
ing can be explained by the fact that the volume of MFIs equity is not signifi-
cant enough for the first years of the period studied.
Results indicate that productivity ratio and age did not influence ROA
and ROE of MFIs. These findings are explained by the immaturity of MFI
and the inefficiency of loan officers, respectively. Indeed, in the early period,
the technological management and the management staff were not suffi-
ciently developed to allow loan officers to acquire a proven technical expert-
ise in the field of microfinance.
The coefficient of percent of female borrowers on ROA is negative and
statistically significant at 5%, indicating that a decrease in percent of female
borrowers leads to raising the ROA. This finding is contrary to the hypothe-
sis 4. Recall that microfinance experience has confirmed that women make

35
SAVINGS AND DEVELOPMENT - No 1 - 2014 - XXXVIII

the best borrowers, and it has been confirmed that they repay their loans
more faithfully than men. This negatively impact for the Morocco case is ex-
plained by the payment default of these clients. Indeed, according to CGAP
(2010), multiple lending to the same clients was an aggravating factor of the
sharp rise in non-performing loans.
The strongest result of this study is the support for Hypothesis 2 that MFI
with a higher portfolio at risk had worse sustainability. This finding is com-
parable to that found by Mersland et al. (2011) and Tchakoute T. (2011) but
contrary to that suggested by Mersland and Strøm (2009). The estimated co-
efficients are equal to -4.65 and -0.327 in the ROE and ROA specifiactions, re-
spectively. These results imply that a 1% increase in the PAR30 lowers ROE
and ROA by 4.65% and 0.327%, respectively. The large impact on ROE is ex-
plained by the fact that credits distributed in early periods are primarily
from the equity of MFIs, which originate from donors. This finding suggest
that MFIs with high-risk portfolio would result in limiting revenues from
microcredit and therefore less money to loan, as it was the case for MFI Zak-
oura (see box 1).
Box 1: Zakoura MFI
The Zakoura Foundation has taken micro-credit loans to finance the de-
velopment in rural Morocco with the aim having successful and sustain-
able level. This brings tourists a preserved and credible experience and
uplifts the wider rural communities of Morocco. This community devel-
opment organization, founded in 1995, created the Rural Tourism Pro-
gram in 2003 to extend microenterprises loans of USD 500 to USD 5,000 to
individuals to develop tourism-related products focusing on financial,
cultural and environmental sustainability. Zakoura provides these loans
for the development of accommodation, restaurants, local products and
tourism activities in three rural regions of Morocco. Zakoura was taken
over in 2010 by FBPMC when it proved financially inefficient, due to an
unsustainable PAR.

Moreover, the coefficient of the portfolio at risk over 30 days is the coeffi-
cient of the highest absolute value in regression (4.65). This implies that the
portfolio at risk is the most important determinant of the MFIs’ financial via-
bility. Hence, MFIs should focus on preventive risk management for a port-
folio of credit quality in order to be financially self-sufficient. It appears that
to increase their financial viability, MFIs should favor the proactive risk man-
agement by considering their ability to anticipate problems of credit recov-
ery in order to prevent contagion perverse among borrowers. Indeed, only a

36
 L. EL KHARTI - THE DETERMINANTS OF FINANCIAL PERFORMANCE OF MICROFINANCE INSTITUTIONS IN MOROCCO

small PAR30 is underlying the financial self-sufficiency of the MFIs in the

context of low levels of savings mobilization and narrowing external fund-
ing opportunities. This should push to the professionalization of MFIs in or-
der to demonstrate a relatively efficient management of their credit risk.
The coefficient of the number of client is positive and significant at 1%.
This finding is in line with Hypothesis 4 formulated above. However, the
impact of credit volume (Lncredit) on financial performance is negative and
statistically significant at 1%. This is due to the credit rationing from Moroc-
can MFIs, which is the consequence of the high rate of payment default in
their operations at the end of the period of growth (2003-2007). Indeed, the
MFI did not take into account risk assessment in granting credit during the
period of growth.
To cope with these difficulties and under proper framework for credit
risk management, Moroccan MFIs created a “center of risk” to reduce the
cross-debt, allowing the exchange and a better flow of information on the
borrower’s credit worthiness. In fact, when MFIs begin to compete on cus-
tomers, the default rate rises quickly if they do not have access to a database
showing the appropriate elements of their customer’s behavior. In the ab-
sence of a center of risk, a much greater control MFIs on individual and
group loans with joint liability group members (Conning, 1999; Laffont and
Nguessam, 2000; Ahlin et al., 2007) should play a leading role in some envi-
ronments marked by solidarity of its members as in Morocco. The joint lia-
bility is a guarantee to fight ex ante and ex post against the default risk
(Stiglitz 1990 Ghatak 1999 and Morduch 1999). Therefore, a proper credit risk
management causing a high rate of loan repayment could allow MFIs to in-
crease their average loan portfolio to identical level loads and therefore low-
er their effective interest rates, whose improve their competitiveness.
As indicated in the estimation strategy, we estimate the two specifications
(1 and 2) for the period of growth (2003-2007) and the period of decline
(2008-2012) in order to investigate the determinants of Moroccan MFIs by
taking into account the 2008 crisis. Indeed, in response to this crisis, MFIs are
tightening their credit processes, putting together teams dedicated solely to
loan recovery, and taking judicial action against delinquent borrowers. Table
4 reports the estimation results.
Findings indicate that the coefficient of equity-to-asset ratio is positive
and significant at 5% for the (2008-2012) period. This result highlights the
change in the funding structure of MFIs in increasing their share of equity in
assets, suggesting that MFIs financing capacity is improved.
Results show that the impact of the number of customers and the percent-
age of female on the financial performance of MFIs is positive and statistical-

37
SAVINGS AND DEVELOPMENT - No 1 - 2014 - XXXVIII

Table 4: Econometric results, period 2003-2012

2003-2007 (before crisis)
Variables ROE ROA
Ratioass -0.81 (3.71) 0.22 (1.14)
Lncredit -0.76 (3.10) -1.41 (0.95)
Lnclient 6.98** (4.04) 3.03** (1.24)
Pfem 0.08 (0.14) 0.08** (0.045)
Empr_Ef -0.002 (0.002) -0.0004 (0.0007)
Mature -2.12 (8.65) -0.69 (2.67)
Par30 0.26 (1.30) -0.62 (0.40)
Constante -47.95 (29.93) -7.83 (9.24)
Number of MFIs 10 10
Number of observations 50 50
R-sq: within 0.16 0.04
Hausman test (Prob>chi2) 0.2408 0.2350
2008-2012 (after crisis)
Variables ROE ROA
Ratioass -0.86 (11.26) 6.94** (2.83)
Lncredit 2.02 (1.46) 0.59** (0.36)
Lnclient -6.32** (2.48) -1.55*** (0.53)
Pfem -0.005** (0.002) -0.0005 (0.0004)
Empr_Ef 0.36*** (0.09) 0.08*** (0.018)
Mature 17.46 (12.43) 3.89** (2.26)
Par30 1.03** (0.63) -0.16 (0.18)
Constante -29.46 (25.45) -7.99 (7.07)
Number of MFIs 9 9
Number of observations 45 45
R-sq: within 0.4 0.55
Hausman test (Prob>chi2) 0.1173 0.3661

*** p<0.01, ** p<0.05, * p<0.1, Standard errors are in parentheses. For the period 2008-2012, we select-
ed nine MFIs. Indeed, in May 2009, Zakoura decided to merge with FBPMC.

38
 L. EL KHARTI - THE DETERMINANTS OF FINANCIAL PERFORMANCE OF MICROFINANCE INSTITUTIONS IN MOROCCO

ly significant at 5% during the period 2003-2007. In contrast, the number of

customers and the percentage of female negatively affect the financial per-
formance during the period 2008-2012. This is explained by the fact that
MFIs have considerably slowed growth and reduced the size of their balance
sheets. They also carry out major recovery plans by strengthening their cred-
it methodologies, build teams dedicated exclusively to loan recovery and ini-
tiate legal proceedings against borrowers in arrears. Moreover, they regular-
ly exchange information on their outstanding customer to control cross cred-
it. Consequently, MFIs reduced the risk to a manageable level and, for this
reason, the impact of PAR30 on the ROE is significantly positive after the cri-
sis.
The results also show that age of MFIs and the productivity ratio have a
positive and significant impact on MFIs’financial performance for the period
2008-2012. This result is expected because the Moroccan MFIs mature and
improve their productivity. It seems that financial performance tends to in-
crease with the number of years of existence of the structure (Cull et al.,
2007). Alongside with seniority, MFIs tend to change their relationship with
their users, substituting the customer strategy by the beneficiary strategy,
and giving priority to the establishment of clear rules for structuring an effi-
cient and stable market.

5. CONCLUSION

The objective of this paper is to examine the determinants of Moroccan

MFIs’ financial performance. Indeed, the financial performance is crucial be-
cause the viability of MFIs increases financial services access to the poor and
to all those excluded from traditional financial sector.
Our estimation results indicate that the portfolio at risk over 30 days and
the age of the MFI are the main determinants of financial performance. They
also show that the outreach of MFIs’ programs positively affects financial
self-sufficiency. Moreover, we find a significant impact of the equity-to-asset
ratio, personal productivity, as well as the percentage of female clients on the
financial performance of MFIs.
The Moroccan crisis has shown that risk across the market exists, even in
the case of credit institutions. As competitive microcredit markets often lead
to multiple loans, MFIs are de facto connected and defaults can spread as
fast as the rumors of a client to another. To avoid such a contagion, the Mo-
roccan case provides valuable lessons in governance, infrastructure and mar-
ket supervision.

39
SAVINGS AND DEVELOPMENT - No 1 - 2014 - XXXVIII

In retrospect, we can note that MFIs having performed better governance

are those received technical skills of their members in banking and finance.
They focused on the long-term sustainability, more slowly but more often
than their counterparts whose have significantly increased their risk profile
without changing their management practices of credit risk.
New challenges are already emerging for the traditional Moroccan actor’s
microcredit. In the medium term, finding the right market niche in the global
financial system will be crucial to incorporate an increasingly diversified
market.
An interesting topic for future research would be the broadening of the
scope of this study to encompass other determinants of financial perform-
ance such as subsidies and managerial aspects. Indeed, these data are only
available in the MIX for a few MFIs. It would be interesting to collect these
data during field surveys on Moroccan MFIs.

Acknowledgments

The author is thankful for their very useful comments and suggestions to two
anonymous referees of the journal and professors P. Adair, C. Mayoukou, F. Mourji,
M. Labie and Dr. M. A. Boutabba, as well as the participants in the following: the 29th
conference on Development, Third World Association (ATM) (ERUDITE, University
of Paris-Est Créteil, 2013); the 31th Summer School of social audit (University of
Mons, Belgium, 2013); the 7th International conference: Development and structural
institutional transformations of the North African and Mediterranean economies (Ra-
bat, Morocco, 2013) and 30th Day of the Normandy Economics & Management Doc-
toral School (Rouen, 2013).

References

Adair P. and I. Berguiga, 2010, “Les facteurs déterminants de la performance sociale

et de la performance financière dans la région MENA: Une analyse en coupe in-
stantanée”, Région et développement, Vol. 32, pp. 91-120.
Ahlin C. and R. Townsend, 2007, “Using Repayment Data to Test across Models of
Joint Liability Lending”, Economic Journal, Vol. 117, pp. 11-51.
Armendariz B. and J. Morduch, 2000, “Micro-finance beyond Group Lending”, Eco-
nomics of Transition, Vol. 8, No. 1, pp. 401-420.
Ben Soltane B., 2013, “Governance Quality and MFIs Repayment Performance” Jour-
nal of Economic Cooperation and Development, Vol. 34, No. 1, pp. 47-62.

40
 L. EL KHARTI - THE DETERMINANTS OF FINANCIAL PERFORMANCE OF MICROFINANCE INSTITUTIONS IN MOROCCO

Brody A., Copestake J., Greeley M., Kabeer N. and A. Simanowitz, 2003, “Microfi-
nance, Poverty and Social Performance”. IDS Bulletin, 34.4, Brighton, UK.
Bruett T., 2005, ‘Mesurer la performance des institutions de microfinance. Un cadre
pour la publication, l’analyse et le suivi’, SEEP Network.
CGAP, 2010, “The Rise, Fall, and Recovery of the Microfinance Sector in Morocco”,
Washington, DC.
CGAP, 2006, “La micro-finance et les banques: Le downscaling: une opportunité pour
les banques”, Washington, DC.
Churchill G., 2000, “Banking on Customer Loyalty” Journal of Micro-finance, Vol. 2,
No. 2, pp. 1-21.
Conning J., 1999, “Outreach Sustainability and Leverage in Monitored and Peer-mon-
itoring Lending”, Journal of Development Economic, Vol. 60, No. 1, pp. 229-248.
Cull R., Demirgùç-Kunt A. and J. Morduch, 2007, “Financial Performance and Out-
reach: A Global Analysis of Lending Micro Banks”, Economic Journal, Vol. 117, No.
1, pp. 107-133.
De Briey V., 2005, “Plein feu sur la microfinance en 2005”, Regards Economiques, Vol.
28, No. 1, pp. 1-14.
Dunford C., 1998, “Microfinance: A means to what end?” Global Dialogue on Microfi-
nance and Human Development, Stockholm, Sweden.
Esubalew Assefa, Niels Hermes and Aljar Meesters, 2013, “Competition and the per-
formance of microfinance institutions”, Applied Financial Economics, Vol. 23, No. 9,
pp. 767-782.
Ghatak M., 1999, “Group Lending, Local Information, and Peer Selection”, Journal of
Development Economics, Vol. 60, No. 1, pp. 27-50.
Gonzalez-Vega C., 1993, “From Policies to Technologies, to Organizations: The Evolu-
tion of the Ohio State University Vision of Rural Financial Markets”. Reassessing
Perspectives Washington, pp. 27-28.
Hartarska V., 2005, “Governance and Performance of Microfinance Institutions in
Central and Eastern Europe and the Newly Independent States”, World Develop-
ment, Vol. 33, No. 10, pp. 1627-1643.
Honohan P., 2008, “Cross-country variation in household access to financial services”
Journal of Banking & Finance, Vol. 32, No. 11, pp. 2493-2500.
Kennedy P., 2008, “A Guide to Econometrics”, 6th ed. Blackwell Publishing, Malden, MA.
Labie M., Lelart M. and T. Montalieu, 2010, “Microfinance: le temps de la maturité?”,
Mondes en développement, Vol. 152, No. 4, pp. 7-11.
Lafourcade A. L., Isern J., Mwangi P. and M. Brown, 2006, “Étude sur la portée et les
performances financières des institutions de microfinance en Afrique”, The Micro
Banking Bulletin Vol. 12, No. 1, pp. 3-21.
Laffont J. and J. Nguessam, 2000, “Group Lending with Adverse selection”, European
Economic Review, Vol. 44, No. 4, pp. 773-784.

41
SAVINGS AND DEVELOPMENT - No 1 - 2014 - XXXVIII

MIX Market, 2008, Micro-finance Information Exchange, www.mixmarket.org.

Morduch J., 2000, “The micro-finance schism”, World Development, Vol. 28, No. 4, pp.
617-629.
Mersland R., Randøy T. and R. Strøm, 2011, “The impact of international influence on
microbanks’ performance: A global survey”. International Business Review, Vol. 20,
pp. 163-176.
Mersland R. and R. Strøm, 2009, “Performance and Governance in Microfinance In-
stitutions”, Journal of Banking and Finance, Vol. 33, No. 4, pp. 662-669.
Mersland R. and R. Strøm, 2008, “Performance and trade-offs in microfinance organi-
zations, does ownership matter?”, Journal of International Development, Vol. 20, No.
1, pp. 598-612.
Morvant-Roux S., Guerin I., Roesch M. and J-Y. Moisseron, 2014, “Adding Value to
Randomization with Qualitative Analysis: The Case of Microcredit in Rural Mo-
rocco” World Development, Vol. 56, No. 1, pp. 302–312.
Norell D., 2001, “How to reduce arrears in micro-finance institutions”, Journal of Mi-
cro-finance, Vol. 3, No. 1, pp. 115-130.
Pitt M. and R. Khandker, 1998, “The Impact of group-base de credit programs on
poor households in Bangladesh: Does the gender of participants matter?”, The
Journal of Political Economy, Vol. 106, No. 1, pp. 956-996.
Schreiner M., 2003, “Credit scoring for microfinance: Can it work?”, Journal of Micro-
finance, Vol. 2, No. 2, pp. 105-118.
Stiglitz J. E., 1990, “Peer monitoring and credit markets”, World Bank Economic Review,
Vol. 4, No. 3, pp. 351-366.
Tchakoute T., 2011, “Etude comparative des performances des institutions de microfi-
nance d’afrique sub-saharienne selon leur statut legal”, Annals of Public and Coop-
erative Economics, Vol. 82, No. 1, pp. 63-76.
Woller G. M., Dunford C. and W. Woodworth, 1999 “Where to microfinance?” Inter-
national Journal of Economic Development, Vol. 1, No. 1, pp. 29-64.

42
 L. EL KHARTI - THE DETERMINANTS OF FINANCIAL PERFORMANCE OF MICROFINANCE INSTITUTIONS IN MOROCCO

Appendix
Table A1: Descriptive statistics
Variables Obs. Mean Std. Dev Min Max
ROA 80 3.44 7.13 -19.18 18.37
ROE 80 4.53 49.92 -382.27 114.66
Ratioass 80 0.47 0.63 -0.5 4.29
Lncredit 80 14.92 2.57 9.55 19.39
Lnclient 80 9.93 1.82 6.12 13.06
Pfem 80 64.51 18.22 31.84 97.98
Empr_Ef 80 480.42 1108.2 0.005 6161.88
Mature 80 0.65 0.47 0 1
Par30 80 3.33 4.23 0 17.79

Table A2: Correlation matrix

[1] [2] [3] [4] [5] [6] [7] [8] [9]
[1] ROA 1.0000
[2] ROE 0.4673 1.0000
[3] Ratioass 0.0796 0.0213 1.0000
[4] Lncredit -0.070 -0.054 -0.3657 1.0000
[5] Lnclient 0.2638 0.0819 -0.3417 0.7719 1.0000
[6] Pfem -0.235 -0.109 -0.0454 -0.0302 -0.0900 1.0000
[7] Empr_Eff 0.1696 0.0993 -0.0834 0.2685 0.3722 -0.0574 1.0000
[8] Mature -0.082 -0.033 -0.3104 0.5514 0.4394 0.1162 -0.1026 1.0000
[9] Par30 -0.309 -0.391 0.0544 -0.0464 -0.1999 0.0311 -0.235 0.0368 1.0000

43
SAVINGS AND DEVELOPMENT - No 1 - 2014 - XXXVIII

Résumé
Cet article étudie les déterminants de la performance financière des institutions de
microfinance (IMF) au Maroc. Nos résultats montrent que le portefeuille à risque
(PAR30) et l’âge des IMF sont les éléments les plus déterminants de la performance
financière de ces institutions. Les résultats démontrent également que la portée des
programmes de microfinance des IMF affecte positivement la performance financière.
Par ailleurs, nous trouvons un impact significatif de la part des fonds propres dans
l’actif total, de la productivité des personnels et du pourcentage des femmes parmi les
clients sur la performance financière des IMFs.
Mots-clefs : microfinance, performance financière, portefeuille à risque, Maroc, crise
2008.
JEL : D63, G01, G15, G23, G32.

44