Microfinance institutions envisage offering financial services to underprivileged people who are excluded from formal banking services (
Financial sustainability refers to the likelihood that a business is self-sufficient without any external support (
Recent studies show that microfinance institutions are grappling with financial sustainability owing to increased monitoring costs, competition and growing nonperforming loans. Unlike conventional banks, that have advanced mechanisms of appraising and monitoring borrowers, MFIs have a poor understanding of borrowers’ default risk thus suffers from lower portfolio quality (
Portfolio quality, also referred to as portfolio at risk, is a measures of the share of MFI’s outstanding loan portfolio with more than 30days in arrears (Ayayi&Sene, 2010). Portfolio quality is important because loan portfolio is the key source of risk to institutions engaging in financial intermediation. According to Hermes and Lensink (2007), portfolio quality is part of asset management that put emphasis on decision making by the management. If loan portfolio quality deteriorates, this might lead to financial unsustainability and eventually the collapse of MFIs. Thus, MFIs ought to focus more on their portfolio quality which translates to financial sustainability. In view of the connection between portfolio quality and MFIs financial sustainability, a few researchers have devoted substantial effort to find out the causal relationship; one branch of studies posit a positive causality (Adongo & Stork, 2006) while the other a negative relationship (Bayai & Ikhide, 2018; Tehulu, 2013) which can be explained by contextual and methodological issues. For instance,
2.1. Theoretical Literature: Institutional Theory
This study explores the influence of portfolio quality on MFIs sustainability. According to (
Traditionally, the theory is equally concerned with the organizational ability to conform to the market dynamics (DiMaggio & Powell, 1983;
2.2.1. Portfolio Quality and Financial Sustainability
A firm’s long-term growth and survive is dependent on its financial sustainability. This is so, particularly to MFIs that lend to clients who are viewed as un-bankable, that is high risk borrower. Therefore, MFIs must devise lending techniques that locks out questionable borrowers to avert the risk of default, which could accelerate the deterioration rate of the portfolios hence eroding financial sustainability (
This studies showed that interest earned from loans serves as the main source of income to MFIs (Fernando, 2006;
The main clients of MFIs are financially excluded individuals and micro-enterprises that lack necessary collaterals or reliable financial and accounting information to secure credit. Screening to distinguishing between micro-entrepreneurs and individuals without bias plays a critical role in MFIs sustainability and also hinders the repayment rates (Chowdhury, 2007). Most MFIs apply progressive loans to enhance repayment. Borrowers aim at accessing adequate finance to achieve individual or enterprise growth and to advance their social wellbeing (
The relationship between portfolio quality and MFIs financial sustainability has elicited a lot of research interest among scholars and practitioners, though, their findings are largely debatable. Using unbalanced panel data drawn from 23 microfinance institutions (MFIs) in East Africa from the period 2004 to 2009, Tehulu (2013) found that portfolio quality had a negative and significant effect on financial sustainability of MFIs in Ethiopia. Conversely, Ayayi (2010), who examined the relationship between portfolio quality and financial sustainability with data representing a sample of 101 countries, for period over 1998-2006, the study found that portfolio quality had a positive effect on financial sustainability. 379 MFIs in71 countries for 6 years–from 2003 to 2008. From a similar perspective,
The study used a panel data set of four years drawn from a sample of 98 microfinance institutions. The findings of the study revealed that portfolio quality had a positive effect on performance. Based on the existing literature the study hypothesizes as follows;
Ho: Portfolio quality has no significant influence on MFIs financial sustainability in Kenya.
A research design specifies the plan on collection and data exploration with the intention of combining relevant information for research purpose and procedure. The study adopted an explanatory research design that is quantitative and hypotheses tested by measuring the association between variables using statistical techniques. Further, the study also employed the use of panel data regression model.
3.1. Target Population and Dataset
The target population were all the 52 registered microfinance institutions in Kenya for period 2010-2018. Only 30 MFIs qualified for the study due to their substantial information. Secondary data from the Microfinance Information Exchange (MIX) database on portfolio ratios was extracted using a data collection schedule. The data encompassed panel data which consisted of time series and cross-sections, it was then analyzed using descriptive statistics. Hypotheses were tested using multiple regression analysis. F-statistics was used to test fixed and random effects. Hausman test showed that fixed effect model was the best to explain the association between the variables.
The objective of the study is to establish the effect of portfolio quality of financial sustainability therefore the study had 3 sets of variables. The dependent variable (financial sustainability), which was measured the adjusted operating revenue/adjusted (financial expense + loan loss provision expense + operating expense) (Yaron& Manos, 2007; Henock, 2019). the independent variable (portfolio quality) the study measures portfolio quality by portfolio at risk beyond 30 days (PAR30) which scholars revealed the potential for future losses based on the current performance (Godquin, 2004;
Multiple linear regression model was utilized to investigate the effect in the study. The equation is described as follows;
Where:
FSit = Financial sustainability for MFI i in year t
QLPit=Portfolio quality for MFI i in year t
MFISizeit= Size for MFI i in year t
MFIAgeit= Age for MFI i in year t
α0it = constant
β1it –β3it = coefficients of the regression
εit = error terms
i = MFIs (cross-section dimensions) ranging from 1 to 30
t = Time (time-series dimensions) ranging from 2010 to 2018
Unit root test
The study tested for unit root to establish if the variables were stationary, which is the fundamental assumption of multiple regression analysis. Panel data is said to be stationary if the mean and variance are constant over time (Gujarati, 2004). Non-stationary data leads to a spurious relationship. This study tested unit root using Fisher, Phillips and Peru test. Conventionally, unit root tests are premised on the following hypothesis.
Null hypothesis (Ho): All panels contain a unit root.
The alternative hypothesis (H1): At least one panel is stationary.
Looking at the p-values in Table 1, the null hypothesis was rejected at all conventional significance levels for all the variables of the study, which implied that there was no unit root in the data hence resulting to the independence of means and variances in the data with respect to time.
3.4. Test for Homoskedasticity
The study tested homoskedasticity using White test. The findings indicated that Chi2 (35) was 52.47, the p-value of 0.0592 revealing that the null hypothesis was rejected implying that the assumption of homoskedasticity was not violated. The results were tabulated as shown in Table 2 below.
[Table 2:] White’s test for Homoskedasticity
White’s test for Homoskedasticity
The autocorrelation can be detected using several tests e.gBaltagi-Wu test, Durbin-Watson test and the Breusch-Godfrey test. According to Drukker (2003), these tests employ many assumptions such as individual effects types,need for non-stochastic regressors and inability to work with heteroscedasticity. Wooldridge (2002), further argued that the said limitations can also deal with unbalanced panel data with and without gaps in their observations.The null hypothesis of this test showed no first-order autocorrelation existed in the data. The test statistic reported isF-test with one and 7 degrees of freedom and a value of 6.597. The P-value of the F-test was 0.0671 implying that the F-test was not significant at 5% level. Hence, the hypothesis of first order autocorrelation is supported and the study concludes that residuals are not autocorrelated.
[Table 3:] Wooldridge test for autocorrelation
Wooldridge test for autocorrelation
The tabulation below shows the mean, minimum, maximum and standard deviation of the various variables as used in the model for the period between 2010 and 2018. Based on table 4, financial sustainability mean was .351 with a minimum of -.864, maximum of 4.914 and a standard deviation .931. Whereas, portfolio quality mean was -2.63 with a minimum of -6.91 and a maximum of 2.85. The portfolio standard deviation was 1.39 indicating variability over some time.
[Table 4:] Descriptive Statistics
Descriptive Statistics
This study shows the association of variables to test the nature of their statistical relationships. Table 5 illustrates the correlation matrix of the research variables. The correlation between portfolio quality and financial sustainability was (r= 0.351,) which depicted a positive significance relationship. While, the correlation between financial sustainability and the control variable were as follows MFI age (r=.039, p<0.05) and MFI size (-.271, p<0.05) respectively.
This study shows the association of variables to test the nature of their statistical relationships. Table 5 illustrates the correlation matrix of the research variables. The correlation between portfolio quality and financial sustainability was (r= 0.351,) which depicted a positive significance relationship. While, the correlation between financial sustainability and the control variable were as follows MFI age (r=.039, p<0.05) and MFI size (-.271, p<0.05) respectively.
[Table 5:] Correlation Matrix Results
Correlation Matrix Results
4.1. Effect of Portfolio quality on Financial Sustainability
The results for Generalized Method of Moments(GMM) are shown in table 6 below. Unlike the fixed effect regression and the random effect regression GMM is designed for datasets with many panels and few periods. Besides, this estimation model is not affected by strict exogeneity assumption which is common static panel data techniques. Due to the sample size and the nature of the data, 30 MFIs and panel data for the period over 2010-2018, GMM was the most suitable model for testing the research hypothesis. The study’s hypothesis stated that; Ho portfolio quality has no significant effect on MFIs financial sustainability, however, based on the findings this (β= 0.12; ρ<0.05), the hypothesis was rejected and the study concluded that portfolio quality had a positive and significant effect on MFIs financial sustainability in Kenya.
[Table 6:] Results for regression analysis (GMM)
Results for regression analysis (GMM)
The results conform with those of
Microfinance institutions play a crucial role in ensuring that the financially excluded and underprivileged individuals and their entities access financial services. Despite its impact to an economy, these institutions continue to grapple with worsening portfolio, which has greatly affected their financial sustainability. It’s on this foundation that this study seeks to establish the relationship between portfolio quality and financial sustainability. The study considered 30 MFIs in Kenya using panel data for the period 2010-2018. The findings were positive with a robustly significant relationship. This implies that higher portfolio quality results to more sustainable MFIs. These results also suggested that low default rates lead to better quality of portfolio and improve financial sustainability. Therefore, institutions should exert effort to ensure that they maximize on repayment rates. These finding could be attributed to expansive market access, information sharing, monitoring of portfolio quality, thus impacting positively on financial sustainability.
Business practitioners and shareholders must ensure that MFIs thrive to uphold better portfolios leading to their sustainability through services delivery. Management should craft suitable lending policies to enhance their portfolio through progressive lending, joint liability and make use of information sharing credit bureau (CRB).MFIs should also use of social sanctions to prevent repayment default rates. This article is the first to provide Kenyan empirical evidence on the association between portfolio quality and the financial sustainability. The study findings will be useful to academia and will also provide a reference point for future studies that focus on portfolio quality modeling, especially for small businesses operating in Kenya.