University of Ghana http://ugspace.ug.edu.gh UNIVERSITY OF GHANA DETERMINANTS OF LENDING INTEREST RATE AND LOAN RECOVERY RATES OF MICROFINANCE INSTITUTIONS IN GHANA: A CASE STUDY OF THE ACCRA METROPOLIS BY GIFTY OKUNOR 10637990 A LONG ESSAY PRESENTED TO THE DEPARTMENT OF FINANCE, UNIVERSITY OF GHANA IN PARTIAL FULFILMENT OF REQUIREMENTS FOR THE AWARD OF MASTER OF BUSINESS ADMINISTRATION IN FINANCE MAY, 2019 University of Ghana http://ugspace.ug.edu.gh DECLARATION I hereby declare that this submission is my own work and to the best of my knowledge. It does not contain previously published materials by other researchers nor accepted work for the award of any other degree in the University, except where due acknowledgement has been made in the text. ……………………………………. ………………………….. GIFTY OKUNOR DATE ii University of Ghana http://ugspace.ug.edu.gh CERTIFICATION I hereby certify that this thesis was supervised in accordance with the laid down procedures of the university. …………………………………….. ………………………….. PROFESSOR ANTHONY Q. Q. ABOAGYE DATE iii University of Ghana http://ugspace.ug.edu.gh DEDICATION This research is dedicated to God and to my mother, Selassie Kportufe. iv University of Ghana http://ugspace.ug.edu.gh ACKNOWLEDGEMENT I am grateful to Professor Anthony Q. Q. Aboagye of the University of Ghana Business School (Finance Department) for the tremendous supervision and guidance towards successful completion of my work. I am also thankful to all microfinance institutions and their officers who helped in accessing data for this work. v University of Ghana http://ugspace.ug.edu.gh TABLE OF CONTENT DECLARATION ............................................................................................... ii CERTIFICATION ............................................................................................ iii DEDICATION.................................................................................................. iv ACKNOWLEDGEMENT ................................................................................. v ABSTRACT .................................................................................................. xvii CHAPTER ONE ....................................................................................................... 1 INTRODUCTION ............................................................................................. 1 1.1 Introduction ............................................................................................. 1 1.2 Background of the Study .......................................................................... 1 1.3 Problem Statement ................................................................................... 5 1.4 Objectives of the Study ............................................................................ 6 1.4.1. General Objective ............................................................................. 6 1.4.2. Specific Objectives ........................................................................... 6 1.5 Justification.............................................................................................. 6 1.6 Research Questions and Hypotheses ........................................................ 7 1.6.1. Research Questions ........................................................................... 7 1.6.2. Research Hypothesis ......................................................................... 7 1.7 Significance of the study .......................................................................... 8 1.8 Scope of the Study ................................................................................... 8 1.9 Limitations of the Study ........................................................................... 9 vi University of Ghana http://ugspace.ug.edu.gh 1.10 An overview of the Microfinance Industry in Ghana ............................. 9 1.11 Organization of the Study ................................................................... 13 1.12 Conclusion ......................................................................................... 14 CHAPTER TWO ..................................................................................................... 15 LITERATURE REVIEW ................................................................................. 15 2.1 Introduction ........................................................................................... 15 2.2 Conceptual Framework of Study ............................................................ 15 2.2.1. Debt-Deflation Theory .................................................................... 16 2.2.2. Theory of Multiple Lending ............................................................ 18 2.3 Empirical Framework and Relational Matters in Review ........................ 20 2.4 Determinants of Interest Rates ............................................................... 28 2.4.1. Operational Cost ............................................................................. 28 2.4.2. Credit Risk ...................................................................................... 29 2.4.3. Liquidity Risk ................................................................................. 30 2.5 Minimum Rate ....................................................................................... 32 2.6 Setting of Interest Rate Premium ............................................................ 34 2.7 Conclusion ............................................................................................. 37 CHAPTER THREE ................................................................................................. 39 METHODOLOGY .......................................................................................... 39 3.1 Introduction ........................................................................................... 39 3.2 Study Area ............................................................................................. 39 vii University of Ghana http://ugspace.ug.edu.gh 3.3 Research Design .................................................................................... 41 3.4 Population of the Study .......................................................................... 41 3.5 Sample Size Determination .................................................................... 42 3.6 Sampling Techniques ............................................................................. 43 3.7 Data Collection Tools/ Instruments ........................................................ 43 3.8 Data Collection Procedure...................................................................... 44 3.9 Data Analysis ......................................................................................... 45 3.9.1. Reliability Analysis......................................................................... 45 3.10 Conclusion ......................................................................................... 47 CHAPTER FOUR ................................................................................................... 48 RESULTS AND DISCUSSIONS..................................................................... 48 4.1 Introduction ........................................................................................... 48 4.2 Personal Data ......................................................................................... 48 4.2.1 Educational background .................................................................. 48 4.2.2 Working Experience ........................................................................... 50 4.3 Minimum Rate and Interest Rate Quotations to Customers ..................... 51 4.3.1 Descriptive Analysis ....................................................................... 51 4.3.2 Inferential Statistical Analysis ......................................................... 54 4.4 How Liquidity Risk Affects the Interest Rate ......................................... 56 4.4.1 Descriptive Statistical Analysis ....................................................... 56 4.4.2 Inferential Statistical Analysis ......................................................... 66 viii University of Ghana http://ugspace.ug.edu.gh 4.5 Operational Cost and Its Effect on Loan Recovery ................................. 68 4.5.1 Descriptive Statistical Analysis ....................................................... 68 Inferential Statistical Analysis .......................................................................... 72 4.6 Credit Risk Faced by MFI’s and Its Impact On Recovery Rate. .............. 74 4.6.1 Descriptive Statistical Analysis ....................................................... 74 4.6.2 Inferential Statistical Analysis ......................................................... 78 4.7 Verify Ways Interest Rate Premium are set to Aid in Loan Recovery ..... 81 4.7.1 Descriptive Statistical Analysis ....................................................... 81 4.7.2 Inferential Statistical Analysis ......................................................... 86 4.8 Hypothesis Testing................................................................................. 91 4.9 Conclusion ............................................................................................. 92 CHAPTER FIVE ..................................................................................................... 93 CONCLUSION AND RECOMMENDATIONS .............................................. 93 5.1 Introduction ........................................................................................... 93 5.2 Summary in View .................................................................................. 93 5.3 Conclusion ............................................................................................. 95 5.4 Recommendations .................................................................................. 98 REFERENCES ...................................................................................................... 101 APPENDIX I- DATA SET .................................................................................... 109 ix University of Ghana http://ugspace.ug.edu.gh LIST OF TABLES Table 1.10.1 The Acceptability and Frequency of Borrower Sacrifices………….13 Table 4.2.1.1 Respondent’s educational background ........................................... 49 Table 4.2.2.1 Working experience of Respondents ............................................... 50 Table 4.2.2.2 Frequency distribution of working experience of respondents ...... 50 Table 4.3.1.1 Minimum rate charged on a loan facility ........................................ 51 Table 4.3.1.2 Responses on how often minimum rate is reviewed in each firm .. 52 Table 4.3.1.3 Interest rate quotes to customers’ aid in loan recovery ................. 53 Table 4.3.1.4 How interest rate calculation is done .............................................. 53 Table 4.3.2.1 Correlation on minimum rate charged on loan facility interest rate quotes to customers’ aid in loan recovery ............................................................. 54 Table 4.3.2.2 Paired sample statistics for minimum rate charged on loan facility interest rate quotes to customer’s aid in loan recovery ........................................ 54 Table 4.3.2.3 Paired Sample test for minimum rate charged on loan facility interest rate quotes to customers aid in loan recovery ...................................................... 55 Table 4.4.1.1 Responses on the role liquidity play in calculating interest rate .... 56 Table 4.4.1.2 The quest to stay liquid affects the chase for interest ..................... 56 Table 4.4.1.3 Extent to which interest rate is affected .......................................... 57 Figure 4.4.1.1 Extent to which interest rate is affected by liquidity .................... 57 Table 4.4.1.4 Case processing summary Average Peak month of high recovery * Average peak month of low recovery .................................................................... 58 Table 4.4.1.5 Average Peak month of high recovery * Average peak month of low recovery cross tabulation ...................................................................................... 58 Table 4.4.1.6. Frequency table showing how early liquidity risk are identified .. 61 x University of Ghana http://ugspace.ug.edu.gh Table 4.4.1.7 Number of months in a year high loan recovery is experienced showed in frequencies ............................................................................................ 61 Table 4.4.1.8 Case Processing Summary of How early liquidity risk is identified * How long it averagely takes to recover from liquidity risk .................................. 62 Table 4.4.1.9 How early liquidity risk is identified * How long it averagely takes to recover from liquidity risk Cross tabulation ........................................................ 63 Table 4.4.1.10 Case Processing Summary Group of customers that are affected after normalization * Different rates are quoted to customers during distressed periods .................................................................................................................... 64 Table 4.4.1.11 Group of customers that are affected after normalization * Different rates are quoted to customers during distressed periods Cross tabulation ................................................................................................................................ 65 Table 4.4.2.1 Correlation on the quest to stay liquid affect the chase for interest and extent to which interest rate is affected ......................................................... 66 Table 4.4.2.2 Paired sample statistics for the quest to stay liquid affect the chase for interest and extent to which interest rate is affected ...................................... 67 Table 4.4.2.3 Paired sample test the quest to stay liquid affect the chase for interest and extent to which interest rate is affected ......................................................... 67 Table 4.5.1.1 Responses in frequencies on the composition of operational cost .. 69 Table 4.5.1.2 Operational cost affect interest rate * Proportion of interest rate it covers Cross tabulation ......................................................................................... 69 Table 4.5.1.3 To what extent does operational risk affect recovery * Operational cost in Interest rate calculation affect recovery Cross tabulation ....................... 71 Table 4.5.2.1 Correlation on operational cost in interest rate calculation affect recovery and to what extent does operational risk affect recovery ..................... 72 xi University of Ghana http://ugspace.ug.edu.gh Table 4.5.2.2 Paired sample statistics for operational cost in Interest rate calculation affect recovery and to what extent does operational risk affect recovery ................................................................................................................................ 72 Table 4.5.2.3 Paired sample test for operational cost in Interest rate calculation affect recovery and to what extent does operational risk affect recovery ........... 73 Table 4.6.1.1 Proportion of interest rate to credit risk * Credit risk is faced Cross tabulation ............................................................................................................... 74 Table 4.6.1.2 Frequency on interest rate of customers who do not possess credit risk is affected ........................................................................................................ 75 Table 4.6.1.3 There are credit policies governing disbursement and loan recovery * Personnel’s or committees are in charge of policies Cross tabulation .............. 75 Table 4.6.1.4 Frequency on the extent credit policies used contribute to loan recovery .................................................................................................................. 76 Table 4.6.1.5 Non-compliance that accounts for ineffective credit administration * Non-compliance with credit policies accounts for ineffective credit administration Cross tabulation ..................................................................................................... 77 In Table 4.6.1.5, results show that non-compliance with credit policies accounts for ineffective credit administration. Customer pressure, management pressure, inexperience of implementers are all factors that accounts for the non-compliance of effective credit administration. ................................................................................. 77 Table 4.6.1.6 How often credit policies undergo review * Credit policies reflect macroeconomic factors Cross tabulation .............................................................. 77 Table 4.6.1.7 Responses in Frequency on Macroeconomic Factors Frequencies 78 Table 4.6.2.1 Correlation on proportion of interest rate to credit risk and extent credit policies used contribute to loan recovery ................................................... 79 xii University of Ghana http://ugspace.ug.edu.gh Table 4.6.2.2 Paired sample statistic for proportion of interest rate to credit risk and extent credit policies used contribute to loan recovery ................................. 79 Table 4.6.2.3 Paired sample test for proportion of interest rate to credit risk and extent credit policies used contribute to loan recovery ........................................ 80 Table 4.7.1.1 Interest rate premiums applied depends on credit risk of the customer ................................................................................................................. 81 Table 4.7.1.2 Interest rate applied is same for all customers ............................... 82 Table 4.7.1.3 An approach adopted is not changed till the loan is paid............... 82 Table 4.7.1.4 The approach influences loan repayment ....................................... 82 Table 4.7.1.5 Fixed or flat interest rate approach results in high loan recovery rate ................................................................................................................................ 83 Table 4.7.1.6 Floating rate approach results in high loan recovery rate ............. 83 Table 4.7.1.7 Reducing balance method is used for the calculation of loan repayment schedule ............................................................................................... 83 Table 4.7.1.8 Straight line method is used for the calculation of loan repayment schedule .................................................................................................................. 84 Table 4.7.1.9 There is high loan recovery rate when reducing balance is used ... 84 Table 4.7.1.10 Loan recovery is high when straight line method is used ............. 84 Table 4.7.1.11 Customers opinion is considered when deciding on which method to use ...................................................................................................................... 85 Table 4.7.1.12 Loan repayment schedule is made in consultation with the customer ................................................................................................................................ 85 Table 4.7.1.13 Loan repayments are rescheduled when there is default ............. 85 Table 4.7.1.14 Customers pay back loans when repayment is rescheduled......... 86 Table 4.7.1.15 Customers attitude affect loan repayment .................................... 86 xiii University of Ghana http://ugspace.ug.edu.gh Table 4.7.2.1 Correlation on the approach influences loan repayment and there is high loan recovery rate when reducing balance is used ....................................... 87 Table 4.7.2.2. Paired sample statistic for the approach influences loan repayment and there is high loan recovery rate when reducing balance is used ................... 87 Table 4.7.2.3 Paired sample test for the approach influences loan repayment and there is high loan recovery rate when reducing balance is used .......................... 88 Table 4.7.2.4 Paired sample statistics of paired variables .................................... 89 Table 4.7.2.5 Paired sample correlation of paired variables ................................ 89 Table 4.7.2.6 Paired samples test of paired variables ........................................... 90 xiv University of Ghana http://ugspace.ug.edu.gh TABLE OF FIGURES Figure 1.10.1 The Prevalence of Repayment Struggles among Micro borrowers in Accra, Ghana……………………………………………………………………….. 12 Figure 1.10.2 A Detailed Split of Loan Uses by Micro borrowers……………….12 Figure 2.2.1 Conceptual Framework of the Study ............................................... 16 Figure 2.4.1 Interest Rate Ceiling Effect…………………………………………..32 Figure 2.5.1 Depict Central Bank Base Rate 2016-2018………………………….34 Figure 4.2.1.1. Bar Chart of Educational Background ........................................ 49 Figure 4.2.2.1 Histogram on Working Experience ............................................... 51 Table 4.3.1.1 Minimum rate charged on a loan facility ........................................ 51 Figure 4.3.1.1 Pie Chart on Minimum rate charged on a loan facility ................ 52 Figure 4.4.1.1 Extent to which interest rate is affected by liquidity .................... 57 Figure 4.4.1.2 Bar chart of average peek month of low loan recovery ................ 59 Figure 4.4.1.3 Bar chart of average peek month of high loan recovery ............... 59 Figure 4.4.1.4 Histogram showing the dispersion of average peek month of high loan recovery .......................................................................................................... 60 Figure 4.4.1.5 Pie chart of number of months in a year high loan recovery is experienced ............................................................................................................ 62 Figure 4.4.1.6 Clustered bar chart of how early liquidity risk is identified * how long it averagely takes to recover from liquidity risk .......................................... 64 Figure 4.4.1.7 Clustered bar chart Group of customers that are affected after normalization * Different rates are quoted to customers during distressed period ................................................................................................................................ 66 xv University of Ghana http://ugspace.ug.edu.gh Figure 4.5.1.1 Clustered bar chart showing Operational cost affect interest rate * Proportion of interest rate it covers ...................................................................... 70 Figure 4.5.1.2 To what extent does operational risk affect recovery * Operational cost in Interest rate calculation affect recovery in a clustered bar chart ............ 71 Figure 4.6.1.1 Bar chart in percentages of the extent credit policies used contribute to loan recovery ..................................................................................................... 76 xvi University of Ghana http://ugspace.ug.edu.gh ABSTRACT The study focused critically on issues of determinants of interest rates and loan recovery rate of Microfinance institutions in Accra Metropolis. With a sample size of twenty (20), Microfinance institutions were examined. Random and purposive sampling techniques were used. Sampling was done by selecting microfinance institutions that give credit to only the informal sector or both the informal and the formal sector. The heart of this research was pairwise correlation. Pairwise correlation was done by determining the correlation between the loan recovery rate and some explanatory variable. To do this, a paired sample test and a paired correlational tables were extracted from the data with the use of SPSS. From the results, a correlation exits between liquidity and interest rate but at no significant level. This indicates that liquidity risk has no relationship with interest rate of the MFIs. Credit risk and operational risk both had a strong positive correlation on loan recovery rates. There is a relationship between credit risk and loan recovery of MFI’s at a level of significance = 0.05. This means contribution of credit risk to loan recovery increases with just an increase in the proportion credit risk covers. Operational cost affect loan recovery rate. A correlation of r= 0.588 shows the extent of relationship which indicates a strong positive correlation between operational cost and loan recovery rate. Loan recovery rates does not depend on Interest rate premiums. Thus the way interest rate premiums are set does not affect the loan recovery of microfinance institutions. Also, there is a fairly strong relationship between minimum rate and loan recovery rate. This was shown with a periodic review and a mark-up way of pricing which was majorly accepted by respondents. xvii University of Ghana http://ugspace.ug.edu.gh University of Ghana http://ugspace.ug.edu.gh CHAPTER ONE INTRODUCTION 1.1 Introduction This chapter seeks to introduce the proposed topic delving into the background of the study, underlining the problem statement and highlighting the key objectives of the proposed study. The chapter also seeks to throw more light into the necessity of the research by identifying works of other people in line with these variables. The chapter also includes the questions that need to be answered in the course of the research, the hypothetical views of the researchers and the objectives of the study. In addition, the purpose, scope of the study and some limitations which served as barriers of the study are included in this chapter. 1.2 Background of the Study The concept of microfinance has become popular throughout the world since it has been adopted by many economies as a major strategy in reducing poverty. According to Schreiner, & Colombet (2001), the key goal of the microfinance industry is to make microloans available to low income and poor households. According to Otero (1999), Microfinance is “the provision of financial services to low-income poor and very poor self-employed people”. Ledgerwood (1999) generally considers these to encompass savings and credit as well as other financial services such as insurance and payment services. Ledgerwood (1999) further submits that there is strong evidence that micro lenders are beginning to experiment with other products, including insurance. According to him, Insurance is a product that is likely to be offered more extensively 1 University of Ghana http://ugspace.ug.edu.gh in the future by microfinance institutions, because there is growing demand among their clients for health or loan insurance in case of death or loss of asset. In Bangladesh, Grameen Bank started with the shapening of the modern industry of microfinance in 1976 which was headed by Mohammad Yunus, (Wikepedia, 2009). The introduction of microfinance affected several lives in Bangladesh through the Grameen Bank start-up and shaping the living standard of the people. The report prepared by Grameen Bank (2014) showed that the bank affected 10 members from 1976 with no group as membership and had 20% of female members all within one village with the help of a single branch. After some 33 years in operation, it had a total group membership of 1,253,160 and individual of 7,970,616. The female membership of the bank grew to 97% within 83,458 villages with the help of 2,562 branches. The total amount of loans disbursed by Grameen Bank since its inception is $6.25 billion. Out of this $5.58 billion has repaid with current amount outstanding at $488.90million and a loan recovery rate of 98.28% (Yunus, Turning Beggars Into Entrepreneurs, 2008). From the research of Yunus (2008), Grameen Bank does not require any collateral against its micro-loans and since the bank does not wish to take any borrower to the court of law in case of non-repayment, it does not require the borrowers to sign any legal instrument. The Grameen Bank was able to recover its loans disbursed as quickly as possible. There was the use of some strategies such as Village Phone Programme Struggling members programme, Grameen Housing Loans which actually served as the motivation for the early repayment. There was also the”16 decisions” concept created and used by the bank. In the last decade, the microfinance industry has flourished and expanded in breadth, depth and scope of outreach. This seems to continue unabated, given the large non- 2 University of Ghana http://ugspace.ug.edu.gh served and underserved market. This growth has also changed the risk profile of MFIs (Fernando, 2008). For example, according to Njuguna, Gakure, Waititu, & Katuse (2013), the growth of Microfinance Sector (MFIs) in Kenya is exposed to various risks which originate from both the internal and external environment. The study showed that financial risks threaten MFIs financial viability and long-term sustainability. These findings are parallel to the findings of other studies on MFI risks. A study conducted by Lascelles, (2012) found that the rapid growth of the microfinance sector in recent years has left many microfinance investors and institutions without an adequate understanding of the multitude of risks they face. The high risks associated with the proliferation microfinance operations thus makes debt recovery very crucial for MFIs (Norell, 2010). In an earlier article, Norell (2010) proposed that, “a key to achieving scale and operational and financial self-sufficiency is to reduce the percentage of loans in arrears”. He also indicated that to maintain good portfolio quality, microfinance professionals must understand what causes arrears and how arrears can be reduced. The concept of microfinance has gradually migrated into the African setting. Most African countries have now come to depend on micro lending as a gateway to reduce poverty. In Ghana, the microfinance concept has been well accepted as a tool to provide financial services to more deprived individuals and establishments. According to Bank of Ghana (2014), there is an ongoing need to boost services of the microfinance institutions (MFIs) and strengthen them through training, technical assistance, funding, congenial policies and enabling infrastructure. The support will enable the MFIs to increase outreach and to rapidly increase the availability of financial services to the poor and largely rural population. Ghanaian Microfinance Institutions fall into three 3 University of Ghana http://ugspace.ug.edu.gh main categories, based on their legal status: formal financial institutions, semi-formal financial institutions, and informal financial system (cle). In Ghana, most microfinance institutions face challenges that are within and beyond their control. Financially, some risks are faced by microfinance operators which include credit risk, risk of liquidity and market risk. In an article by Kunateh (2016), it is indicated that the microfinance sector in Ghana is at the verge of collapse and this is largely due to the refusal of beneficiaries of some microcredit schemes including the Microfinance and Small Loans Centre (MASLOC) to pay loans owed the sector. High interest rates have often been associated with very risky activities. From the onset of modern microcredit, interest rates charged by micro lenders have been one of the most controversial dimensions microfinance Institutions have had to contend with. Microfinance lending rates have often been higher than normal bank rates mainly because of obviously high costs associated with lending to and recovering thousands of tiny loans advanced to micro borrowers. In Accra, there is the continuous rapid rise in the number of microfinance institutions that is established. Microfinance is a firm that seeks profit even though it is aimed at reaching to the poor to help reduce poverty. It makes funds available to the needy by offering credit facilities to them to be repaid in a later date. This credit is made available after deposits have been made by some members of the firm. In summary, even though most microfinance institutions in Accra have been in business for some years, the risk element associated with micro financing cannot be eliminated in relation to the repayment of credit facilities made available to the public. This is because the credit opportunities offered by the MFIs very high interest rates and their inability to repay the credit facilities given to them. Also to a large extent. the idea of 4 University of Ghana http://ugspace.ug.edu.gh microfinance has been drifted as some microfinance now offers insurance policies to the public. 1.3 Problem Statement Globally, the inability of borrowers to repay loans taken from microfinance is the risk faced by these institutions. Many have attributed the inability of borrowers to repay MFI loans to high interest rates. According to Quartey (2011), in Ghana there seems to be plenty of disagreement over the level of interest rate charged by microfinance providers because the factors that go into these calculations are not well known. The high risks associated with microfinance operations also makes debt recovery very crucial for MFIs (Norell, 2010). According to (Ghatak, 1999), four key factors determine these rates: the cost of funds, the MFI's operating expenses, loan losses, and profits needed to expand their capital base and fund expected future growth. Studies conducted into microfinance in Ghana have mostly concentrated on the effect of interest rate on loan repayment on Microfinance Institutions. Even at the time when MFIs are finding it very difficult to collect loans which have been given to beneficiaries this area is still receiving much attention. Also, other very few works capture the determinants of interest rates on loan repayments/ default in the Microfinance institutions. All these focuses on the customers and their inability to repay the loans on the basis of high interest rates and in some case their determinants. This creates a serious research gap into the area of interest rate in microfinance institutions of which this study seeks to close. To breech the gap, this study concentrates on specific determinants of interest rates and how it affects loan recovery in terms of the microfinance institutions for liquidity purposes. Do the determinants of interest 5 University of Ghana http://ugspace.ug.edu.gh rates affect loan recovery of Microfinance Institutions? That will be the major purpose of this study. 1.4 Objectives of the Study 1.4.1. General Objective The general objective is to ascertain the impact of the determinants of interest rates on loan recovery rates of MFIs in Ghana. 1.4.2. Specific Objectives The achievement of the general objective will be done through the corresponding achievement of these specific objectives. 1. To determine the effect of base rate on loan recovery rate quotations to customers of MFIs and how is it calculated. 2. To ascertain how liquidity risk affects the interest rate of the MFIs. 3. To evaluate the operational cost and its effects loan recovery rates. 4. To investigate the credit risk faced by the MFIs and its impact on loan recovery. 5. To investigate the ways MFIs set interest rate premiums to aid in loan recovery. 1.5 Justification Microfinance Institutions (MFI) play a critical role in poverty reduction in the Ghanaian economy. Although the MFI industry in Ghana appears to be growing on a tremendous scale, loan recovery remains a major headache for micro lenders. Loan defaults have often been attributed to many factors especially high interest rates. Although much research has been done on the MFI industry in Ghana, there is still a need for better understanding into the operations of MFIs that will improve loan recovery rates. High 6 University of Ghana http://ugspace.ug.edu.gh interest rates continues to dominate in discussions into the causes of loan default. Although, this problem is widely known, much has not been done to study more specifically, whether the determinants of Interest rates affects on loan recovery rates of MFIs in any way. 1.6 Research Questions and Hypotheses 1.6.1. Research Questions In connection with the sampled MFIs, this study will address the following questions. 1. Does the minimum rate affect loan recovery of MFIs and how is it calculated? 2. How does liquidity risk affect the interest rate of the MFIs? 3. What is the effect of operational cost on loan recovery rates? 4. How does credit risk affect loan recovery of MFI’s? 5. To what extent does the ways MFIs set interest rate premiums to aid in loan recovery? 1.6.2. Research Hypothesis The research hypotheses for this study are stated below: Null Hypothesis: (H0) Alternate Hypothesis: (HA) 1. (H01): Minimum rate do not affect loan recovery rate of MFIs. (HA1): Minimum rate affect loan recovery rate of MFIs. 2. (H02): Liquidity risk has no relationship with the interest rate of the MFIs. (HA2): Liquidity risk has a relationship with the interest rate of the MFIs. 3. (H03): Operational cost do not affect loan recovery rate. 7 University of Ghana http://ugspace.ug.edu.gh (HA3): Operational cost affect loan recovery rate. 4. (H04): There is no relationship between credit risk and loan recovery of MFI’s. (HA4): There is a relationship between credit risk and loan recovery of MFI’s. 5. (H05): Loan recovery rate depends on the application of interest premiums by MFIs. (HA5): Loan recovery rate depends on the application of interest premiums by MFIs. 1.7 Significance of the study This research will be of importance to many stakeholders. Many people will directly or indirectly benefit from the findings of the research.  The government and other policy makers will be better informed in instituting favourable policies for MFI operations.  Microfinance Institutions who are in vigorous competition will better understand the impact of determinants of interest rates on their loan recovery operations and may help them manage this area of their operations better. The research will serve as a platform, creating a foundation for other researches to build upon to create awareness and find solutions to the many challenges associated with the MFI industry. 1.8 Scope of the Study This research was conducted among fourteen (20) microfinance institutions (MFIs) in the Accra Metropolitan District. A specific microfinance institution was selected based on the principle that it has operated in the industry for some number of years to pave 8 University of Ghana http://ugspace.ug.edu.gh way for rigorous statistical analysis of their interest rate quotes and what goes into arriving at those rates and how best these affects their various businesses. 1.9 Limitations of the Study The research had certain constraints which had impact on the results that were produced. Information that was required was from credit officers of the microfinance institutions, they did not have enough time to read through the self-drafted questionnaire since they were few in number and had lots of work to finish and this may play a role in the inadequate sample size already indicated. Moreover, most of the officers of the firms felt reluctant to grant simple interviews or answer questions asked them for better understanding of what really goes on in their firms since the fear is that they may disclose information that is sensitive and restricted only for internal use. Again, information that are so vital to this research and should be demanded were not be released. Again, information that were so vital to this research and should be demanded were not be released hence researcher had to use discretion, ranking and ranges to set questions for average data for this research. Finally, due to the distribution of the population which had impact on the sampling, the study could not be done using regression for the analysis since the technique needed a minimum sample for effective running of the analysis. This was considered due to reliability purposes of the study. 1.10 An overview of the Microfinance Industry in Ghana Bank of Ghana (2007) working paper on microfinance evolution indicated that, the sector certainly is not new in Ghana as it has always been common practice for people to save and or take small loans from individuals and groups within the context of self- 9 University of Ghana http://ugspace.ug.edu.gh help in order to engage in small retail businesses or farming ventures. Sketchy evidence suggests that the first credit union in Africa was probably established in Northern Ghana in 1955 by the Canadian Catholic missionaries that were there at the time. However, susu, which is one of the current microfinance schemes in Ghana, is thought to have originated in Nigeria and spread to Ghana from the early 1900s. Over the years, the microfinance sector has thrived and evolved into its current state and thanks to various financial sector policies and programmes such as the provision of subsidized credits, establishment of rural and community banks (RCBs), the liberalization of the financial sector and the promulgation of PNDC Law 328 of 1991, that allowed the establishment of different types of non-bank financial institutions, including savings and loans companies, finance houses, and credit unions etc. Currently, there are three broad types of microfinance institutions operating in Ghana. These include: formal suppliers of microfinance (i.e. Rural and community banks, savings and loans companies, commercial banks) Semi-formal suppliers of microfinance (i.e. credit unions, financial nongovernmental organizations (FNGOs), and cooperatives; Informal suppliers of microfinance (e.g. Susu collectors and clubs, rotating and accumulating savings and credit associations (ROSCAs and ASCAs), traders, moneylenders and other individuals). In terms of the regulatory framework, rural and community banks are currently regulated under the Banking Act 2004 (Act 673), while the Savings and Loans Companies are currently regulated under the Non-Bank Financial Institutions (NBFI) Law 1993 (PNDCL 328). On the other hand, the regulatory framework for credit unions is still being developed to reflect their dual nature as cooperatives and financial institutions. The rest of the players such as FNGOs, ROSCAS, and ASCAs do not have explicit legal and regulatory frameworks, and are largely unregulated. In terms of current policy 10 University of Ghana http://ugspace.ug.edu.gh programmes that affect the Microfinance sub-sector, a number of on-going projects can be cited. These include - the Financial Sector Improvement Project, Financial Sector Strategic Plan (FINSSP), the Rural Financial Services Project (RFSP), the United Nations Development Programme (UNDP) Microfinance Project, the Social Investment Fund (SIF), the Community Based Rural Development Programme (CBRDP), Rural Enterprise Project (REP), and Agricultural Services Investment Project (ASSIP). In an attempt to explain existence of MFIs, Kunateh (2016) provided that the number of microfinance institutions (MFIs) licensed by the Bank of Ghana (BoG) to operate as of March this year stands at 394. They are 344 microfinance companies, 45 money lending companies and five financial non-governmental organisations. He again expressed that this figure depicts a little over 100 percent jump since 2012, an advisor to the Governor of Bank of Ghana and that the three umbrella associations are the Ghana Association of Microfinance, Money Lenders Association of Ghana and Ghana Co-operative Susu Collectors Association. Even with these categorizations aiding micro borrowers to which financial institution, there is still a wide base of over indebtedness on them. Measuring over indebtedness from the customers perspective, Shicks (2011) identified there were repayment struggles. This is shown in the chart below 11 University of Ghana http://ugspace.ug.edu.gh Figure 1.10.1 The Prevalence of Repayment Struggles among Micro borrowers in Accra, Ghana. Source: Schicks, J. (2011). Over-indebtedness of micro borrowers in Ghana an empirical study from a customer protection perspective. Centre for Financial Inclusion p.8 He also identified that loan splitting and acceptable borrower sacrifices are some main indicators to the repayment struggles of micro borrowers in Ghana. Below is a chart and a table respectively depicting the findings. Figure 1.10.2 A Detailed Split of Loan Uses by Micro borrowers. Source: Schicks, J. (2011). Over-indebtedness of micro borrowers in Ghana an empirical study from a customer protection perspective. Centre for Financial Inclusion p.7 12 University of Ghana http://ugspace.ug.edu.gh Table 1.10.1 The Acceptability and Frequency of Borrower Sacrifices. Source: Schicks, J. (2011). Over-indebtedness of micro borrowers in Ghana an empirical study from a customer protection perspective. Centre for Financial Inclusion p.10 1.11 Organization of the Study The study entails five (5) chapters. Chapter one introduces the study by bringing out the background of the study, research problem, justification, objectives, hypotheses and research questions, significant of the study, scope of the study, limitation of the study and organization of the study. Chapter two deals with literature review. Thus, it spells out the variables, theories and empirical evidence of the research. Chapter three contains the research methodology adopted for the research. It outlines the methodology, tool, procedures and techniques for undertaking the secondary and primary data collection, how data was collected and analysed. 13 University of Ghana http://ugspace.ug.edu.gh Chapter four presents the analysis of data gathered and the presentation of the findings. The last chapter which is chapter five also presents the summary of the findings, conclusion and the recommendations for the research. 1.12 Conclusion This chapter generally helped introduce the entire research. It specifically elaborates the background of the study which brings out the history to the research, the problem statement indicating the problems identified to be solved, the bases of the research through the justification, research questions and hypotheses, objectives of the study, significance of the study, scope of the study and the limitation of the study. This chapter also helped with the introduction of the next chapter to the research which carefully examines previous research important to particular aspect of this research. 14 University of Ghana http://ugspace.ug.edu.gh CHAPTER TWO LITERATURE REVIEW 2.1 Introduction This chapter of the study sought to unearth and provide in-depth knowledge regarding the interest rate composition in lending sectors of the wide financial market across the globe. For this course, the researcher sought to review, expose, analyse and provide perspective on existing literature on the ever robust matter of interest rate and loan or credit recovery in the financial landscape especially micro financing. In the quest to fulfilling this, the researcher looked at definitional issues on critical variables contributing to interest rates, conceptual frame work of the study, empirical proves and frame works, and other related matters trending in the interest rate, lending and recovery circus. 2.2 Conceptual Framework of Study Below are the interest rate components the researcher has found significant in establishing their impact on loan recovery. 15 University of Ghana http://ugspace.ug.edu.gh CO N T R OLL DEPENDENT VARIABLES I N G I N D E P E N DENT Interest rate premium BoG- Base Rate Liquidity risk INTEREST LOAN RATE RECOVERY Operational cost Credit risk Figure 2.2.1 Conceptual Framework of the Study 2.2.1. Debt-Deflation Theory Debt-deflation theory (Fisher, 1933) is a theory of economic cycle which holds that recessions and depressions are due to the overall level of debt shrinking (deflating) and in regards the credit cycle is the cause of the economic cycle. This theory was familiar to Keynes (1936) prior to Fisher‟s discussion of it but he found it lacking in comparison to what would become his theory of “Liquidity Preference”. The theory however has enjoyed resurgence of interest since 1980‟s both in mainstream economics and in the heterodox school of the Post-Keynesian economics. Fisher‟s (1933, p. 341.) 16 University of Ghana http://ugspace.ug.edu.gh introduction of these theory came as a result of the greatest depressions in the nineteenth and twentieth centuries. The great booms and depressions were characterised by the two main dominant factors which are; over indebtedness and deflation. At the macroeconomic level, over-indebtedness of a country sets the tone for deflation to easily follow. These two variables create big disturbances in an economy. Wikipedia defined deflation as “the decrease in the general price level of goods and services”. McConnell and Brue (2006) also defined deflation as “the decline in the general price level which is the reverse of inflation”. It occurs when inflation rate fall below zero percent (0%). Fisher (1933) explained that, the issue of over-indebtedness makes debtors unable to repay their debts hence prices will be forced to reduce or fall since many cannot afford as a result of illiquidity. He indicated that one of the starters of economic downturns is the insertion of “indebtedness” which may create room for “deflation” to set in quickly. Chief among the major causes of over-indebtedness Fisher (1933, p. 348) revealed are; new opportunities to invest at big prospective profit as compared to ordinary profits and interest such as through new inventions, new industries, development of new resources and opening of new markets. Many of the MFIs in Ghana have severally been in grips of this theory compounding the over-indebtedness the more. Investors are itching to borrow huge sums of capital or credit with the speculation of new opportunities they have found that will yield them about 100% return per annum or less than that. This Psychological impression according to Fisher held by investors and the credit industry leads to the over indebtedness of 1929. As borrowers are unable to repay their borrowed funds, business activities become slow, employment levels will go down, and prices of other goods and services falls, disposable income are also affected. Over-speculation, over-consumption, over lending 17 University of Ghana http://ugspace.ug.edu.gh without being meticulous is really affecting the credit industry in the West African sub- region and, for that matter, Ghana. The chief starters of over indebtedness leading to the crisis of the 1837 were connected with lucrative investment opportunities from developing the West and Southwest in real estate, cotton, canal building (led by the Eric Canal), steamboats and turnpikes (Fisher, 1933). 2.2.2. Theory of Multiple Lending In most countries, firms tend to borrow from several banks. This applies to more than 85% of European countries (Ongena, 2000). Petersen and Rajan (1994) concludes that “borrowing from multiple lenders increases the price and reduces the availability of credit”. Bennardo, Pagano & Piccado (2008) in their study, argues that actual or potential multiple bank lending can have adverse effects since it induces both borrowers and lenders to behave opportunistically when the banks or creditors‟ rights are not protected and the collateral is volatile. This means that lenders use every opportunity they obtain without considering some credit policies or conditions of borrowers by enquiring some policies adopted by competitors. This opportunistic lending arises when banks are impartially informed on the credit conditions offered by other banks since no bank can observe all outstanding loan offers when deciding whether to accept or deny client‟s credit application. This as a result increases default rate to all banks or lenders. Now, like Fisher (1933) put it that one of the starters of over-indebtedness is the crave for new opportunities to make a prospective 100% profit rather than ordinary profit seem to make way in the multiple lending theory. Multiple lending makes it easy for lenders to lend monies to varieties of clients meaning that they are open to the general public whilst some have also streamlined their clients target making it easier on their part to manage and control the risk of default and capital risk. Borrowers on the other hand have the choice of borrowing from many financial institutions as there are no strict 18 University of Ghana http://ugspace.ug.edu.gh regulations as to where one have to borrow credit from and so borrowers indulge in indiscriminate borrowing which other credit gurus called “credit indiscipline” on the part of the borrowers. Both lenders and borrowers behave opportunistically trying to outsmart competitors in the industry without paying much attention to recovery. Sometimes clients try to borrow from various credit firms without finishing to service their old debts with the various institutions they borrowed from but because the credit industry in Africa and other parts of the world and for that matter Ghana is not much resourced to track client borrowings, they turn out giving money to people who cannot repay and it becomes a burden on their part. In very developed countries it is held that, there is high client credit information tracking such that individuals or firms cannot borrow anyhow as there are management information systems that check whether you borrowed money elsewhere and whether you have finish paying back before the system can allow you to proceed hence creditors are well protected against default and capital risk associated with the lending business. Furthermore, banks lend extra monies to borrowers already in debt in order to achieve their own interest, thus increasing the interest rates and rationing credit. Bennardo, Pagano, & Piccolo, (2013), in their study also spells out the implications of these externalities for credit market equilibrium and investigates how their intensity is affected by information sharing among lenders (credit reporting). They found out that banks deny credit to some applicants and borrowers default strategically when their collateral value is depressed and that there is no information sharing and poor creditor right protection. It means that when their collateral value is not too volatile, information sharing improves credit market performance, then reduces interest rates and default rates. On the other hand, when the value of collateral is very volatile, information sharing induces the credit market to freeze since information sharing allows lenders to 19 University of Ghana http://ugspace.ug.edu.gh better protect themselves against borrowers‟ opportunistic behaviour, leading to the charge of lower interest rates and expand lending. It also enables opportunistic lenders to better target those borrowers to whom they can profitably lend at their competitors‟ expense. Carletti, Cerasi & Daltung (2007), major insight in their paper is to provide a static model, a new explanation for the use of multiple- bank lending as a way to improve monitoring incentives. It emphasizes on bank-firm relationships when monitoring is essential and banks need greater diversification to improve the value of relationships as in the case in small and medium business lending. For the purpose of this study, the researchers looked at how MFIs can adopt some measures to ensure the effective management of multiple lending by using the variables elaborated in the earlier chapter. The researcher also saw the relevance of using these theories as a backbone to consolidating our finding as some of the MFIs in the Accra Metropolitan District do not good attention to the issue of multiple-lending and its associated merits and demerits as well as how to apply it in microfinancing. 2.3 Empirical Framework and Relational Matters in Review The studies conducted unearthed the various works done by previous researchers which provided a convincing ground upon which the research was conducted. The study considered twenty-six (26) works across the globe and at least forty (40) authors writing about interest rates, recovery rates, performance, default risk, profitability and more. Amonoo, Acquah and Asmah (2003) conducted a study on the influence of interest rates as represented by demand for credit on loan repayments; (measured with closeness to borrowers: due to their wide network, monitoring of loans, preference for short-term 20 University of Ghana http://ugspace.ug.edu.gh loans, installment payment systems, grace periods and group lending schemes), with credit demand tested separately using CAMPARI. The study was conducted in Ghana: Central Region (11 districts, NBFIs & BFIs 25, SMEs 50). And as a method of estimation model, Econometric, Regression Analysis (OLS) and Maximum Likelihood Estimation (MLE) Method were engaged to help make a valid conclusion on interest rates estimations on repayment ability of clients of Microfinance institutions. It was found that interest rate and demand for credit were negatively related; thus as interest rates reduce more credit is demanded and the vice versa. The study also showed that a negative relationship existed between interest rates and loan repayment deploying the MLE method. Khieu and Mullineaux (2009) in a somewhat study of recovery rate-measured by loss given default (LDF) given as one less the loss given default (1-d) and loan characteristic, recovery process, borrower characteristic and macroeconomic factors, it was found in their study that Firm leverage before default negatively affects ultimate recoveries, while firm size and also macroeconomic conditions do have a positive impact on recoveries. Further, industry distress does have a negative impact on settlement recovery rates. They deployed a secondary data analysis using Moody’s Ultimate Recovery Database and descriptive, correlation and regression analysis techniques – logistic regression model and method data estimation. Moti el al., (2012) also researched about credit management system measured in terms of credit terms, client appraisal, credit risk control and credit collection. This was conducted in Kenya (Meru town): Sample size of 70 credit officers among 14 MFIs. A descriptive statistics, Chi-square analysis technique was used to represent data. It was found later in the research that Credit terms formulated by MFIs do not affect loan performance but credit terms formulated by credit officers and clients had an effect on performance of loan. 21 University of Ghana http://ugspace.ug.edu.gh Also credit risk and collection policies adopted by MFIs had significant effect on loan performance such that stringent control boost loan performance and less stringent control do not boost loan performance much. Kimando et al. 2012, also in their further research sustainability of MFIs against financial regulations, number of clients served, financial coverage and volume of credit transacted revealed that non-payment of loan by customers, interest rate risk, poor management of the institutions, so much borrowing from the customers and also government policies affect MFIs sustainability since these factors does not often give a positive return. Also, geographical coverage thus the category of the inhabitants in terms of age and other personal factors influences the sustainability of microfinance institutions to a great extent. Gyamfi (2012), in his study on credit risk and loan, risk and creditworthiness showed that small MFIs were more vulnerable to credit risk than the bigger firms. The MFIs had written policies that guided them on credit granting and its related risks but most techniques used for the management of credit risk by MFIs in Accra within the period (2003-2007) were not quite effective. Also, the best technique for determining the credit worthiness of clients was to study their character, savings and cash flow. The study focused on 20 MFIs with descriptive statistics was used for the analysis as a method of data estimation. Arko (2012) closely coming by also delved operations of microfinance institutions; operational cost and credit appraisal techniques, monitoring , diversion of funds, business failure, poor weather conditions, inadequate marketing avenues, wrong timing of credit delivery, high interest rate, and willful default. The study was conduct on Sinapi Aba Trust (35 branches out of 46) Ghana and Histograms, pie charts and line graphs as estimation model. It was revealed that Non-performing loans adversely affected financial performance of SAT in terms 22 University of Ghana http://ugspace.ug.edu.gh of profitability, liquidity and market appeal. Ineffective monitoring of loans accounts for the incidence of NPLs. Peprah (2012) explored into MFIs management and governance face-to-face nominal interest rate, wrong use of loans, insufficient screening and less frequent repayment schedules, late disbursal of loans. The study was conducted in Ghana: Cape Coast. 10 MFIs: purposive and simple random sampling technique given Correlational analysis as mode of data estimation. The results showed that, frequency of repayment is directly associated with client delinquency which is also a correlate of loan default among MFI clients. Causes of loan default among clients are high interest rate, wrong use of loans, insufficient screening and less frequent repayment schedules. There is a bi-directional relationship between loan quality and cost efficiency in MFIs. Ogolla (2012), more so researched on Non-performing loan portfolio measured by environmental conditions, strategic choice, financial performance, short turn-around time, collaboration and partnerships, flexibility in loan repayment, alternative repayment avenues and technological support as against debt recovery rate (as an operational strategy) measured by demand notice, frequent reminders, normative commitment, legal arrangement, outsourcing debt recovery and invoking guarantees in Kenya Nairobi. The study engaged all the 18 branches of NIC Bank in Kenya Through a purposive sample of top executives and a random sample of Bank staff. Given a descriptive statistics; frequency tables, it was found that operational recovery strategy proved helpful in managing nonperforming loan. Nduta (2013) closely following, focused on financial Performance and credit management. The study was also conducted in Kenya: sample size of 59 using a census descriptive survey using descriptive statistics, regression analysis techniques and ANOVA, t-test method of data estimation. It was found that, there is a strong relationship between financial performance of MFIs and credit 23 University of Ghana http://ugspace.ug.edu.gh management thus client appraisal, credit risk control and collection policy. Therefore, it was further concluded that collection policy has the higher effect on the financial performance of MFIs and that a stringent policy is more effective in debt recovery than lenient policy. Munene and Guyo (2013), researched into loan default; thus inability of loan beneficiaries to repay back their loans and business characteristic explained by age, type of business, location and profit. It was found that there is a between the type of business, age of the business, number of employees, business profits and loan repayment default. The study further indicates a strong link between technical training for loan beneficiaries and the performance of businesses among the remote societies. The study was conducted in Imenti North District – Kenya; a sample size of 400 was used deploying a cluster sampling techniques for selection using descriptive statistics to estimate data. Agu and Okoli (2013) came out with a finding that incessant increase in interest rate is a strong and statistically important factor that causes bad debt in Nigeria commercial banks. Also, inadequate close monitoring of the borrowers to ensure proper utilization of funds, lack of adequate knowledge of the loan seeker, failure by commercial banks to give their loan immediate follow up to avoid diversion and poor credit policy administration are some findings. The study examined the impact of interest rate, deposit loan and advances, balances with Central Bank of Nigeria (CBN) on bad debts and doubtful debt (both measured using increase in interest rate on loans and advances). Analysis of Variance (ANOVA) and autoregressive model were applied to validate the result. Jarko, et al (2013) conducted a study into loan default and sector –risk exposure, firm- risk, EU subsidy. The study was done in Europe, Slovakia. Sample size of 700 panel 24 University of Ghana http://ugspace.ug.edu.gh data bank loans was used spanning (2000-2005) using Frim-level panel data analysis deploying descriptive and correlational techniques. Findings from sector-risk perspective indicated that, Agri-food industry does not exhibit high default rate than other sectors. On the firm-risk, highly indebted firms are more likely to default than others and finally, newly introduced subsidy system provides secured source of income hence reducing the probability of default. Gatimu and Mukoma (2014) closely researched in similar study, as they looked at loan default and credit policies, loan recovery procedures, and loan appraisal process. The study used multiple regression, descriptive statistics and correlational analysis technique, descriptive statistics were used in estimating data. It was revealed that credit policies and loan recovery procedures, and loan appraisal process had a significant positive impact on the loan default rate hence MFIs in Kenya have a cause to worry if they have to reduce the loan default rates by considering the aforementioned factors. Fedaseyeu and Hunt (2014) came out with a finding regarding debt collection or loan recovery as against debt collection agencies also known as third party collections. Debt recovery was measured with debt collection agencies, credit market segmentation and asymmetric information whereas Debt collection agencies (measured using dates, number of banks and cost of funds), credit segmentation (measured with customers observation to interest rates and debt collection practices of all banks) and asymmetric information (measured with delegated and undelegated equilibrium. The study showed that Third party debt collection agencies increases credit supply and the aggregate borrower surplus than if the creditors themselves make the attempt to do collection. We also Kamau and Wagoki (2014) coming through with loan recovery and strategic management practices. The study was done in Nairobi Region, Kenya, with sample size of 62 respondents throughout stratified sampling technique and a descriptive statistics 25 University of Ghana http://ugspace.ug.edu.gh and correlational analysis estimation models. It was revealed that, collection strategies have a positive relationship with loan recovery. So how management approaches repayment strategies determine their ability to recover such term facilities. Addae K. (2014), found in his studies that late disbursement of loan business failure, unfavourable payment terms, high interest rate, inadequate loan sizes, unforeseen contingencies, like illness contributed to their default problem. Also poor appraisal, lack of monitoring or improper monitoring, improper client selection, and diversion of funds on the side of the clients resulted in default. This study deployed random sampling of 25 MFIs and 250 clients using descriptive statistics was deployed making using of frequency distribution table. Variables used were Loan Default, Loan Delinquency, Default rate was measured using the high interest rate, inadequate loan sizes, poor appraisal, lack of monitoring, and improper client as against MFIs measured by small loans advancement, client size, savings and investment products. Agbemeva, Nyarko, Adade and Bediako (2016) in their research found that, six (6) factors out of sixteen(16) predictor variables used (Marital status, young and old aged dependents, type of collateral or security, assessment, duration and loan type) were statistically significant in the prediction of loan default payment with a predicted default rate of 86.67%. 548 nontraditional bank customers through stratified sampling of customers from the on time, not on time, default strata were deployed. A binomial logistic regression analysis estimation model was used, Loan Default or Repayment measured using: i. Defaulted: if delayed payment is > 30 𝑑𝑎𝑦𝑠. ii. Non – defaulted: if delayed payment and ≤ 30 days. Explanatory: Age, gender, marital status, type of business, residential status, number of years at residence, dependents, purpose of loan, amount disbursed, date of disbursement, type of collateral (security), guarantor, assessment, officer, duration, loan type. 26 University of Ghana http://ugspace.ug.edu.gh Our empirical review showed that out of the twenty-five (25) authors assessed on the subject matter regarding debt management and recovery rate, sixteen (16) of the authors deployed inferential statistics by making use of regression analysis techniques, econometric analysis, correlation, maximum likelihood estimation methods to actually predict probable cause(s) of default rates and how to enhance recovery of non- performing loans given out to clients. Also six (6) additional authors deployed a mix of descriptive and inferential statistics to arriving at a valid conclusion. The review also revealed that, six (6) authors actually made use of only descriptive statistics in projecting their outcome by looking at measures of dispersion such as; variance analysis, standard deviations, and also measures of central tendencies (mean, mode, median). Very few authors (Fedaseyeu and Hunt 2014, Kebewar 2012, Quarmyne (2014) used some unpopular estimation models to projecting problems regarding debt management and recovery rate and among them are; the common agency and welfare analysis techniques, generalized method of moments, co-integration and vector error correction respectively. It therefore stands to reasons that the credit industry can make use of the inferential statistics to make solid decision to regulate the industry and the MFIs for that matter in their quest to advancing credit to small and medium enterprises. Some of the analysis tools used by the fifteen (15) authors include; simple linear and multiple regression, correlation, group difference analysis (ANOVA, t-test, ordinary least square among others which actually predict the factors that contribute debt issues and the models that can be adopted to salvage the recovery such as credit risk modelling. The descriptive statistics is also a good model of estimation but lack to an extent the ability to predict hence its usage can have very less impact on the subject matter. It is in this regard that we 44 sought to agree with the twenty-one (21) authors that went further to using 27 University of Ghana http://ugspace.ug.edu.gh various inferential techniques coupled with descriptive statistics in solving the recovery rate challenge confronting the microfinance institutions. It is surprising to note that in the review conducted, it appeared that Kenya scored more in research regarding microfinancing which tells us the inertia at which MFIs are rocking the country and this seem not to be different in Ghana. From the empirical works tabulated above, it can be indicated that many variables were used by the researchers and that variables that worked as dependent variables in some cases worked as independent variables for the reason or objective of the research. 2.4 Determinants of Interest Rates 2.4.1. Operational Cost Business dictionary (2019) defines operational cost as the cost per unit of product or service, or the annual cost incurred on a continuous process. This cost does not include capital outlay or cost incurred in design and implementation phase of a new process. Ali & Dipratn (2015) identified three major cost structures of microfinance institutions. Viz; borrowing cost, operational cost and delinquencies cost. From this write up, it is understood that about 70% of MFIs cost structure goes to operational issues whereas borrowing cost is a major item on commercial banks’ balance sheet. This MFIs sector is expected to help alleviate poverty but in their quest to do so, they tend up sinking together with the poor. Could it be that they are not better placed to migrate their clients from such states to a state financial soundness? Today micro financing has tend to be more burdening to the poor ones as MFIs appear to be trading on their head rather than seeking to help manage the poverty among the large unbanked and poor communities. This is felt through high interest charges on loan. MFIs in Ghana today charges between 28 University of Ghana http://ugspace.ug.edu.gh the 5-9% monthly on lend funds to customers with an average lower interest of 60%APR and 108%APR on a high side. The unit cost of maintaining a client’s account translate largely in huge operational expense which is pushed to the consumer to bear and should this become unbearable to the clients they begin to exhibit behaviours that damages the quality of the credit disbursed. The banking industry’s operational efficiency indicators broadly improved as cost to income ratio declined to 84.3 percent in October 2018 from 87.6 percent in October 2017, while cost to total assets ratio declined to 9.9 percent from 12.8 percent during the same comparative period. Similarly, operational cost to total assets declined to 6.4 percent in October 2018 from 7.3 percent in October 2017, while operational cost to gross income ratio increased to 54.1 percent from 50.4 percent in the period under review, due to a slower growth in the industry’s gross income. Overall, the industry was more efficient at minimizing costs relative to assets and income in October 2018 than previous year. 2.4.2. Credit Risk As in the wider banking sector, the microfinance segment could soon see consolidation, as authorities nudge firms to reorganise their loan books and recapitalise. “There has to be consolidation or there will be more collapses in MFIs,” Joe Jackson, director of business development at one non-bank lender, Dalex Finance and Leasing Company, told OBG. “There are around 10 medium players in the microfinance market with the capacity to buy up smaller firms. However, it can be a challenge given the bad status of most firms’ loan books. Companies considering acquiring or merging are looking for institutions that are organised and have strong relationships with their client base.” 29 University of Ghana http://ugspace.ug.edu.gh In July 2015 the BoG mandated that MFIs double their minimum capital requirement to GHS2m by end-2017. Whether this would happen is unclear: in October 2017 the Ghana Microfinance Institutions Network announced that 80% of MFIs would be unable to meet the new requirements by the deadline. Given the segment’s challenges and the dubious reputation of some registered institutions, it can be easy to overlook the fundamental importance of MFIs to Ghana’s small and medium-sized enterprises. “With commercial banks preferring to lend to the government rather than the private sector, the true lenders are non-bank institutions,” Jackson told OBG. “We do not have the luxury of cheap money from the government, so we lend to the country’s small businesses. As mobile money continues to grow and our credit-risk analysis improves, MFIs will remain a key source of finance for the private sector” Oxford Business Report (2019). Credit risk refers to the risk that a certain sum, be it principal or interest of both may not be recovered granting bad credit delivery system connected to several factors. These risks are non-systematic or controllable when good credit systems and recoveries are in place. Generally, consumers are spontaneously unwilling to pay back debts, and as such needed some level of strategies to cause them to respond appropriately. Larger nature of such risk falls in the informal private sector individuals or businesses. Especially no-salary workers who really need much discipline to be able respond to repayment obligations. It’s therefore necessary for MFIs to deal cautiously in credit administration to such low income persons. 2.4.3. Liquidity Risk Liquidity risk is associated with availability of credit to meet the increasing demand for credit and other forms of advances in an economy especially finance industry. According banking sector report by Bank of Ghana (2018); 30 University of Ghana http://ugspace.ug.edu.gh Operational liquidity indicators recorded mixed performance with declines in core liquidity measures while the broad liquidity measures recorded significant increases. The industry’s core liquid assets (comprising cash and due from banks denominated in both domestic and foreign currency and short term bills) to total deposits declined to 37.4 percent in October 2018 from 40.0 percent in October 2017. Similarly, core liquid assets to total assets declined to 23.7 percent from 25.1 percent during the same review period. The industry’s broad liquid assets (comprising core liquid assets and shares and other equities) increased following the issuance of the ESLA bond and the recent GOG bond to offset liabilities of the collapsed banks. The increase translated into higher broad liquidity indicators, namely, broad liquid assets to total deposits of 100.4 percent in October 2018 relative to 86.3 percent in October 2017 and broad liquid assets to total assets rose to 63.6 percent from 54.2 percent. The industry therefore remained adequately liquid to meet its short-term obligations, despite the declines in the core liquidity measures. Liquidity also translates into cost of credit as a result of demand and supply. That is why financial institutions that are highly liquid tend to have relatively lower interest rate compared to those with liquidity challenges. These also send relief to consumers and encourage them to access credit and repayment ease. There are some MFIs that are highly liquid in Ghana who have very attractive interest rate with majority charging very high monthly interest compounding rate and because of interest rate ceiling, and low liquidity consumers have but no choice to borrow form them knowing well they it is highly expensive. This make recovery a challenge to MFIs as low liquidity tends to make interest charge expensive to consumers. 31 University of Ghana http://ugspace.ug.edu.gh Imposing interest rate ceilings leads to three main issues: Shortage in credit supplied to the market; less transparency about the exact cost of the credit in the micro financial institutions; limiting the scope of operation of micro financial institutions and losing the chances of obtaining the loans for underprivileged class in the society. At equilibrium demand and supply of loan meets at an interest rate and should there be no incentive to increase supply lending will not lend at equilibrium price unless there is an inward shift in the supply curve thereby increasing consumption. One can only lend below the equilibrium price due to interest ceiling. Fig. 2.4.1 Interest Rate Ceiling Effect 2.5 Minimum Rate The minimum rate of the MFIs is also the base rate from the Central Bank of Ghana. Base rate plays a critical role in the business of lending across the world that lies in the sole bosom of a country’s financial regulator which the Central Bank (CB). Faure (2013), on Central Banking and Monetary Policy, defined Base rate as a key interest rate (KIR) at which the central bank lends to commercial banks of their reserve with the regulator. KIR has several names such as prime lending rate, repo rate, discount 32 University of Ghana http://ugspace.ug.edu.gh rate, bank rate, and base rate. It is pricing ceiling or maximum rate above which financial institutions cannot lend. It sets the basis upon which financial institutions for that Microfinance firms can provide lending services. This rate is normally set above inflation in order to enhance the store of value benefit of money. Microfinance institutions are by prudential regulation expected to have a certain percentage of their total deposits with the central bank; RR*TD: reserve requirement and total deposit. Should this reserve fall below the threshold, surplus deposit from another institution will be lend to you from the central bank reserve account at a cost or should any Commercial bank or Microfinance needs cash or liquidity, the central bank lends from the reserve accounts at a cost which is referred to as the base rate or prime lending rate. It must be recalled that the central bank is the banker to all banks and microfinance institutions. This rate (Monetary Policy Rate) is reviewed from time to time depending on the economic fundamentals structures by a policy formulation committee known as monetary policy committee (MPC). This rate informs how financial institutions in the business of lending and borrowing should price their rate to final consumers or customers. It is generally known that buying goods at a wholesale price is relatively cheaper than at a retail price and so what retailers end up paying translates higher than wholesalers will be to manufactures or producers. This scenario is not far-fetched when it comes to Central Bank base rate and Commercial banks market quoted rates. Currently inflation rate in Ghana quoted by Ghana Statistical Service stands at 9.6% thus Average Consumer Price (ACP) by the end of December 2018. Policy rate stands at 16% as at January 2019. Real interest rate charged by Bank of Ghana is estimated at 6.4% above inflation hence commercial banks should be borrowing at that rate from the central bank reserve account. According to Fraue (2013) central banks operates two major classes of accounts viz; operational settlements accounts and reserve accounts. It 33 University of Ghana http://ugspace.ug.edu.gh is therefore not financially prudent to lend at the same rate you borrowed to customers. This rates are annual rates. Source: TradingEconomics‖ Bank of Ghana Figure. 2.5.1 Depict Central Bank Base Rate 2016-2018 2.6 Setting of Interest Rate Premium Ali & Dipratn (2015) “posit that for any financial institution wishes to deliver any kind of financial services to low income people or to other clients in the market and for obtaining a permanent source of credit for this institution, it is necessary to be permitted to charge its clients a reasonable interest rate to be able to cover various costs incurred by the institution when carrying out its operations. This is considered to be an essential element so that financial institution can sustain, become self-reliant and operate effectively in the market. When interest rate is charged, dependency on the government subsidies and donors grants is reduced to large extent which means achieving self-independence”. It almost appears that the concept of micro financing to rural and low income people seem challenging on the face of interest rate or interest premiums. This concept as alluded to Prof. Yunus who provided the first loan amounted 34 University of Ghana http://ugspace.ug.edu.gh to US $27 to 42 poor women in jobra poor village of Bangladesh (Yunus, Banker to the Poor, 1967), his prime goal was to provide them with free collateral loans which can be used in the creation of new job opportunities for them that, eventually, will reduce their poverty level. Today the aim of doing so is virtually out of sight. Because MFIs in this dispensation have put on a different cloth of high profitability driven motive, making money out of the poor. The Professor sought to make free money available to people at very little cost or minimal cost not because he is poor, but because he wanted to be good to people out benevolence and not looking at what economic benefit to be derived. If highly solvent or formal institutions who invariably understand the business of finance do heavily default and become bankrupt who do we expect relatively very poor or low incomes earners to respond well to repayment obligations? These call for more policy formulation that can provide solid grounds for financial prosperity among the low income earners. Application of interest charge premiums has also brought more difficulty to repayment abilities that ever. Interest rate charges come in two forms; simple interest and compound interest. Both have advantages and disadvantages, each of these might be fixed or flat, floating or fluctuating. Interest charge on loans and advances can be applied on a straight line basis or reducing balancing basis. With the former, an equal amount is paid across the period of engagement thus both interest and principal. Eg. Given a loan of GHS10,000.00 for 2years at a rate of 19% APR, 19⁄100 ∗ 10,000 1,900.00 = , 158.33 𝑚𝑜𝑛𝑡ℎ𝑙𝑦 𝑟𝑒𝑝𝑎𝑦𝑚𝑒𝑛𝑡 𝑓𝑜𝑟 24𝑚𝑜𝑛𝑡ℎ𝑠, 𝑤ℎ𝑖𝑐ℎ 𝑖𝑛𝑐𝑙𝑢𝑑𝑒𝑠 𝑒𝑞𝑢𝑎𝑙 𝑎𝑚𝑜𝑢𝑛𝑡 𝑜𝑓 𝑝𝑟𝑖𝑛𝑐𝑖𝑝𝑎𝑙 12 35 University of Ghana http://ugspace.ug.edu.gh And interest. For two total interest income is GHS3,800.00 hence in effect the client is expected to GHS13,800.00 at the end of the tenor. When it comes to compound interest charge application, interest can be charged a number of times within a 24hr period across the year. This is a straight line interest premium application. With the reducing balancing repayment application, the client pays the same GHS 158.33 amortised amount but there is a sharp split between interest and principal repayment modality which of course, lies solely on the MFIs to determine how repayment should be done. This is done to so as recover total interest first or principal for the purpose profit reporting which is expected to hit their income statement, and push the principal outstanding to the statement of financial position. So with the above amortised cost, interest repayment can be set in a descending order of repayment whilst the principal climbs slowly in an ascending order which when plot on a diagram can take the shape of a cumulative frequency distribution curve but at every plot of the curve you will still get your GHS158.33. Now because MFIs institutions are scared to tell clients their Annual Percentage Rate (APR), they tend to mention compounding rate which look cheaper to the consumer making it expensive for poor people and low income earners to honour repayment obligation because MFIs have become highly super profit motivated using Prof. Yunus idea to make money out of the poor. This is accounting for huge delinquencies and Non-Performing Asset on the books of microfinance institutions. It is therefore necessary that microfinance institutions consider these things carefully in their quest to profitability. There are key things to note when it comes to interest charge setting, considering commercial banking and central banking perspective; inflation rate, T-bill rate, Base 36 University of Ghana http://ugspace.ug.edu.gh rate; repo, prime, discount etc, interest premium. Currently commercial banking rate averagely 27%APR, now many have linked or thought that as central bank base rate drop to 16% commercial banks should lend just about the same rate but it cannot be the case. As indicated earlier, Bank of Ghana lends to universal banks from reserve accounts and are not free monies as they also lend above inflation rate; average consumer price index as at January 2019 stands at 9% by Ghana Statistical Service. A differential percentage of 7% being the real interest charge is what BoG charges commercial banks per annum but together with inflation cost makes it 16% APR. Now this rate is a wholesale price BoG sells to universal banks as considering credit worthiness. It stands to reason that, any bank lending to a retail consumer must of necessity lend reasonably within that ceiling. Comparing industry average lending rate of 27% less 16% base rate is 11%. Comparing 11% real interest charge and 7% BoG real interest rate, gives a differential of 4%. Now there is room for universal banks to lend up to 16% real rate which will make it 32% APR but that will be unreasonable due to competition. 2.7 Conclusion The variables dominating the reviewed literature explain the need for the use of the variables in conducting the research. Even though some variables were used and dominating and other were not identified as variable within the literatures reviewed, these cases serves as bases for and gaps in which the research can be conducted. Variables including liquidity risk, operational cost, base rate, credit risk and application of interest rate premiums aided in the study and helped gave findings either related or averse to most finding in the review. 37 University of Ghana http://ugspace.ug.edu.gh From the mode of estimation with the dominating variables in the question, the inferential statistics models of analysis and to some extent the descriptive statistics aided with the finding of the research. 38 University of Ghana http://ugspace.ug.edu.gh CHAPTER THREE METHODOLOGY 3.1 Introduction The methodology enumerates coherently, the method, techniques, tools and procedures used in identifying the sample from the population to perform a valid analysis. It also presents the area of study, population of the study and the research design. The steps in determining the sample used in this research are also presented in this chapter. 3.2 Study Area This research had its focus on all registered and licensed microfinance institutions in the Accra Metroplis of the Greater Accra Region as a result of the availability of data needed to make research result as valid as possible. The Accra Metropolitan Assembly (AMA) is one of the twenty-six (26) Metropolitan, Municipal and District Assemblies (MMDAs) in the Greater Accra Region. It was established in 1898 but has gone through several changes in terms of name, size and number of Sub-Metros (Accra Metropolitan Assembly, n.d.). Three newly-created Municipal Assemblies was inaugurated in Accra and indicated as carved out of the Accra Metropolitan Assembly (AMA): Ablekuma Central Municipal Assembly, Ayawaso Central Municipal Assembly and Osu Klottey Municipal Assembly (Accra Metropolitan Assembly, 2019). With these carved out Municipal Assemblies, the number of licensed Microfinance Institutions in AMA is forty-one (41) (Bank of Ghana, 2018). 39 University of Ghana http://ugspace.ug.edu.gh Out of all the registered and licensed microfinance institutions in AMA, twenty (20) out of forty-one (41) were selected for the research as of the time the research was conducted. These MFIs were selected based on the services they rendered to the public. It was held that those microfinance institutions whose focus are on the formal sector are less prone to default or sometimes not at all hence in order to get a better picture of the default situation, it was better to observe those institutions whose operation is distributed between formal and informal sector clienteles respectively. The operation of the selected firms in terms of services rendered cut across people of the informal sector such as the Small and Medium Scale Enterprises (SMEs) and subsequently to the formal sector like individuals living on salary. It was realised that most of the people in the metropolis are into businesses majorly small scale businesses with a few others into petty trading as the rest may be employees of one institution or the other. To venture into any business, the people needed to access some credit facilities in the form of loans from the microfinance institutions since they may not likely to go to the mainstream financial institutions such as the (bank) due to the excessive credit granting requirement they have to face. Also, some of these microfinance institutions have adopted the customer to employee relationship such as the mobile banking style where loan officials or sales agents get to the field to explain to the people especially the petty traders many a time about the flexibilities of policies used in accessing loans and their readiness to grow their businesses in a way of poverty alleviation. 40 University of Ghana http://ugspace.ug.edu.gh 3.3 Research Design The researchers examined the study on determinants of interest rates of the selected microfinance institutions in Accra and the effect it has on recovery of loans disbursed. Variables used were both qualitative and quantitative. Descriptive research design was used to describe what affects interest rate in microfinance institutions and how they contribute to reducing their debt stock. The descriptive design helped present results in the frequency distribution tables, charts and graphs by computing measures of central tendencies; mean, mode, median and measures of dispersion; variance analysis, standard deviation among others. The empirical review revealed that about ten (10) different authors made use of this research design technique. The correlation research design was used in order to evaluate the extent of relationship between and among the variables. Ahiawodzi (2010, p. 164) described this technique as a tool that analyses the relationship among categorical and continuous variables and has no barrier of normality of distribution unlike regression techniques. The study made use of the pairwise correlation technique. 3.4 Population of the Study The population focused on loan officers of all selected microfinance institution which helped in the accessibility and validity purposes of the research. Data was alluded to loan officers of the organisations due to the nature of the variables the study deployed and the type of data needed hence it was appropriate that key persons in the organisation respond to the questions. This was to help the researchers obtain a consistent response within each organisation. There are several methods of estimating population size but then popular among them is the Krejchie and Morgan (1970) table of population and sample size determination table which has being used over the years and due to its 41 University of Ghana http://ugspace.ug.edu.gh reliability and validity, this study adopted its use to estimate the population size. To ensure that the research is reliable, the suitable population was twenty (20) with a corresponding sample size of nineteen (19) according to the Krejchie and Morgan (1970) table. Using this sample size, nineteen (19) officers were selected from the selected microfinance institutions for the study making 100% of the sample size. This was done through purposeful random selection. 3.5 Sample Size Determination As the reliability of this research was of importance to its users, the sample size of the population was determined in order to answer the research questions to the study. Due to relevance, only loan officers was used in the determination of the sample size. There are several methods of determining sample size depending on the area of study the researcher is conducting the study. There are four (4) major ways of finding a suitable sample size as provided by (Singh & Masuku, 2014) and these are; using a census for small population, imitating a sample size of similar study, using published tables and applying formula to calculate a sample size. In this study deployed the published list of microfinance institutions in good standing by Bank of Ghana as at August 2018. Microfinance institutions within the Accra Metropolis in good standing were purposively selected based on the MFIs institutions that provides services to either the informal or both informal and formal sectors. Random selection was done with officers in the MFIs. One officer was selected for each institution sampled. 42 University of Ghana http://ugspace.ug.edu.gh 3.6 Sampling Techniques In relation to the purpose of the study, a sampling technique was chosen. A sampling technique was used in the selection of the various microfinance institutions and the officers of the various institutions. The researcher’s investigation on the number of MFIs in the Accra Metropolitan District showed that, there are forty- one (41) microfinance institutions. Twenty (20) out of these number were selected on the ground that they offered services to both formal and informal sectors because the topic pursued by the researcher was relevant to these selected MFIs as compared to the other institutions that solely dealt with the formal sector where there are salary workers making the issues of default unpopular. The informal sector includes petty trader, small and medium enterprises, agro farmers among others. The researchers used the purposive sampling technique to select the various microfinance institutions to be used for the study. This was to meet the purposive loan default and recovery rate. For reliability purposes in relation to the study area, an officer from each of the twenty (20) institution were selected on a purposive ground to respond appropriately to the questions constituting the whole sample size. 3.7 Data Collection Tools/ Instruments In the data collection, the study used a primary data collection tool particularly, a structured questionnaire. It was preferred to other primary data collection tool such as the interview because of time which served as a limiting factor which was clearly stated in the chapter one. In this regard, open-ended and closed-ended approach to questions were used to collect responds from the various officers of the selected MFIs. In the open-ended approach, participants were allowed to express their opinion regarding the subject matter in writing with spaces provided. With the closed-ended approach to data 43 University of Ghana http://ugspace.ug.edu.gh collection, respondents were limited in terms of expressing their thought but rather options were placed before them to make their choice of answer to the various questions. Some of such categories of closed-ended questions included binary question, continuous, categorical, quantitative and also qualitative type of questions. This helped made analysis easier as compared to the open-ended questions where responds from the various participants had to be categorised before any analysis can be done. The questionnaire was self-designed for the population. Apart from questions which helped in answering the research questions, some personal information was also required of the respondents. The questions were made simple and clear as possible for validity purposes since respondents have limited time to fill out the questionnaire due to their busy schedule. The researchers developed thirty-eight (38) questions which was answered by officers. The self-designed questionnaire for the various MFIs officers had six (6) sections. Section “A” request the respondents to provide background information such as level of education and working experience; Section “B” determined the effect of base rate on interest rate quotations to customers of MFIs and how is it calculated.; Section “C" ascertained how liquidity risk affects the interest rate of the MFIs.; Section “D” evaluated the operational cost and its effects loan recovery rates.; Section “E” investigated the credit risk faced by the MFIs and its impact on loan recovery; Section “F” verified the ways MFIs apply interest rate premiums to aid in loan recovery. 3.8 Data Collection Procedure In the process of collecting the data, questionnaires were administered to respondents personally in institutions where familiarity was already established with respondents. 44 University of Ghana http://ugspace.ug.edu.gh Few friends and colleagues who were trained to understand the questionnaire and how to administer it after permission was granted from the other selected microfinance institutions for the research to be conducted also helped. 3.9 Data Analysis With the data analysis, the data that was collected and analysed using statistical software like IBM Statistical Package for Social Sciences (SPSS, version 23.0). In the attainment of the objectives of this research, the data was collected, coded and entered SPSS. This was used to generate the frequency distribution table to aid in the construction of charts suitable for each objective. The obtainment of central tendency elements like the (mean, mode and median) and measures of dispersion such as variance, standard deviation among others were used for the analysis. Other inferential statistical tests and analysis were also conducted such as; pairwise correlation and paired t-test. The hypothesis testing was also done to affirm the assumptions raised in earlier chapters. 3.9.1. Reliability Analysis The study conducted a reliability test, this was to enhance internal consistency, validity of the survey instrument deployed. The reliability analysis was conducted using a Likert scale, dichotomous variables, ordinal and interval scale variable. For the model of reliability, the Cronbach’s alpha coefficient was used which was pegged on the Mugenda rule of thumb with a lower boundary of (0.6) on a pilot tested survey questionnaire. Objective One: To determine the effect of base rate on interest rate quotations to customers of MFIs and how is it calculated. Quantitative data analysis was used to 45 University of Ghana http://ugspace.ug.edu.gh summarize data that was collected from the respondents and the section “B” of the self- designed questionnaire in tables. The data was sorted, coded and entered into SPSS for the analysis to take place. Descriptive statistics was performed on the data and presentation in forms such as bar charts, histogram , pie chart and graphs to summarise the data. An inference was drawn making use of Pearson’s moment correlation to establish some relationship among the questions and also paired sample test statistics was done to help establish a true hypothesis. Objective Two: To ascertain how liquidity risk affects the interest rate of the MFIs. Data collected from section “C” of questionnaire was used to ascertain the liquidity risk. The analysis of this objective also underwent sorting and entering of the data collected in the analysis software for the analysis to commence. The study started with some descriptive analysis technique tools such as frequency, measure of central tendencies and also measures of variability such as standard deviations, clustered bar charts and pie chart. In addition, inferential statistical analysis techniques pairwise correlation was conducted on the data to explain whether the differences observed in the descriptive analysis were statistically significant or whether they happened by chance or not. Objective Three: To evaluate the operational cost and its effects loan recovery rates. Quantitative data analysis was performed on the data to summarize and present the data collected from the section “D” of the self-designed questionnaire. These were described using frequency tables, multiple response tables and clustered bar chart and for the purpose of decision making, pairwise sample t-test and correlation was conducted on the data. 46 University of Ghana http://ugspace.ug.edu.gh Objective Four: To investigate the credit risk faced by the MFIs and its impact on loan recovery. With the aid of section “E” of the questionnaire that was designed, the correlation between credit risk and loan recovery using SPSS to conduct descriptive analysis and multiple response tables coupled with and pairwise comparison analysis. Objective Five: To verify the ways MFIs set interest rate premiums to aid in loan recovery. Section “F” of the questionnaire verified the application of interest premiums. To achieve this, the data was first cleaned and sorted, coded and entered into a statistical software for the analysis to take effect. A descriptive analysis coupled with inferential analysis such as pairwise correlation was conducted to help in decision making. 3.10 Conclusion The chapter specifically indicated Accra Metropolis as the study area. Also, the population and sample size was determined through sample size determination table and formula computation to confirm the sample size number. The research design, data collection procedure and instruments discussed above were carefully selected. Data presentation and analysis techniques which will be used in the next chapter were introduced. To critically enhance validity and internal consistency of instruments used, a reliability test was done making use of Cronbach’s alpha coefficient as deployed. This chapter set the tone for the researchers to carry out the analysis in the next chapter where the analysis took effect. 47 University of Ghana http://ugspace.ug.edu.gh CHAPTER FOUR RESULTS AND DISCUSSIONS 4.1 Introduction This chapter gives the output of the whole data collected over the period of study. In this chapter the various analysis techniques have been thoroughly elaborated to give adequate meaning to the data collected. The analysis comprised both descriptive and inferential statistics to summarize and present data. Very popular of the descriptive analysis include; frequency distribution table, charts and cross tabulations. That of inferential statistics included some test statistics such as paired sample t-test and pairwise correlational analysis 4.2 Personal Data The personal data analysis comprises the educational attainment and years of working experience of officers who responded to the questions. 4.2.1 Educational background Question (1) in the questionnaire demand educational background of the officers in the various microfinance institution in the Accra Metropolis. This was to help the researchers know the caliber of human resource in the various microfinance institution as a way of tracking their knowledge level when it comes to handling customer default traps coupled with prudent credit administration. 48 University of Ghana http://ugspace.ug.edu.gh Table 12.1.1 Respondent’s educational background Valid Cumulative Frequency Percent Percent Percent Valid Secondary/Technical 2 10.0 10.0 10.0 First degree/HND 14 70.0 70.0 80.0 Post graduate 4 20.0 20.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 As shown in Table 4.2.1.1, it can be seen that out of the twenty (20) respondents, fourteen (14) had their most recent educational qualification in first degree and HND respectively constituting 70% of the total frequency. Respondents with postgraduate qualification were four (4) making 20% of the sample with the remaining 10% constituting two (2) respondents had secondary/technical education. This data is presented in a bar chart below; Figure 4.2.1.1. Bar Chart of Educational Background 49 University of Ghana http://ugspace.ug.edu.gh 4.2.2 Working Experience This data can be found in question two (2) of the questionnaire which required the number of years an officer worked in the organization. Table 12.2.1 Working experience of Respondents N Valid 20 Missing 0 Mean 1.70 Median 2.00 Mode 1 Std. Deviation .733 Minimum 1 Maximum 3 Source: Field Survey Data, April 2019 Table 4.2.2.2 Frequency distribution of working experience of respondents Valid Cumulative Frequency Percent Percent Percent Valid <1-2yrs 9 45.0 45.0 45.0 3-5yrs 8 40.0 40.0 85.0 6-8yrs 3 15.0 15.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 The tables show an account working experience of officers of the MFIs in Accra Metropolis. Out of a total frequency of twenty (20) respondents, nine (9) had <1-2yrs working experience constituting 45% and eight (8) had working experience of 3-5yrs representing 40% with three (3) having a working experience of 6-8yrs constituting 15% of the total frequency. On average each officer had at least two (2) years working experience shown in the statistical table above. The minimum and maximum responses are 1 and 3 respectively. Below is a chart representing the data in the table. 50 University of Ghana http://ugspace.ug.edu.gh Figure 4.2.2.1 Histogram on Working Experience 4.3 Minimum Rate and Interest Rate Quotations to Customers This determine the effect of base rate on interest rate quotations to customers of MFIs and how is it calculated. 4.3.1 Descriptive Analysis Questions (4 and 5) of the questionnaire answers these responses regarding the existence of the minimum rate charged and how frequent the minimum rate is reviewed. Table 4.3.1.1 Minimum rate charged on a loan facility Valid Cumulative Frequency Percent Percent Percent Valid 16-20% 9 45.0 45.0 45.0 21-25% 5 25.0 25.0 70.0 26-30% 2 10.0 10.0 80.0 36-40% 4 20.0 20.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 51 University of Ghana http://ugspace.ug.edu.gh Figure 4.3.1.1 Pie Chart on Minimum rate charged on a loan facility Table 4.3.1.2 Responses on how often minimum rate is reviewed in each firm How often minimum rate is reviewed Frequenc Valid Cumulative y Percent Percent Percent Valid quarterly 7 35.0 38.9 38.9 Semi-annually 2 10.0 11.1 50.0 Annually 6 30.0 33.3 83.3 others, please 3 15.0 16.7 100.0 specify Total 18 90.0 100.0 Missing System 2 10.0 Total 20 100.0 Source: Field Survey Data, April 2019 As shown in Table 4.3.1.1, it can be seen that out of the twenty (20) respondents, 45% will charge 16-20% minimum rate on loan facility given to customers. 25%, 10% and 20% will be charging 21-25%, 26-30% and 36-40% respectively. Table (2) reported on how often the minimum rate is reviewed. Quarterly review respondents were seven (7) which constituted 35% of the total responses. Semi-annual and annual reviews were 10% and 30% respectively. The remaining 15% of the 52 University of Ghana http://ugspace.ug.edu.gh respondents had other periods for review which was not captured in the responses provided. Total respondents for this was eighteen (18). Questions (7 and 8) of the questionnaire had responses to if interest rate quotations to customers’ aid in loan recovery and how its calculated. Table 4.3.1.3 Interest rate quotes to customers’ aid in loan recovery Valid Cumulative Frequency Percent Percent Percent Valid yes 15 75.0 78.9 78.9 no 4 20.0 21.1 100.0 Total 19 95.0 100.0 Missing System 1 5.0 Total 20 100.0 Source: Field Survey Data, April 2019 Table 4.3.1.4 How interest rate calculation is done Frequenc Valid Cumulative y Percent Percent Percent Valid mark up 13 65.0 68.4 68.4 margin 4 20.0 21.1 89.5 others please 2 10.0 10.5 100.0 specify Total 19 95.0 100.0 Missing System 1 5.0 Total 20 100.0 Source: Field Survey Data, April 2019 In Table 4.3.1.3, 75% of the respondents agreed that interest rate quotations aid in loan recovery as 20% were not of this view. Total frequency was nineteen (19). Also in Table 4.2.1.4, mark up way of calculation interest is used by 65% or firms while 20% use margin for the calculation of interest rate. Other ways of calculation were used by two (2) other firms making 10% of the total frequency for these responses. 53 University of Ghana http://ugspace.ug.edu.gh 4.3.2 Inferential Statistical Analysis Using the Pearson’s pairwise correlation coefficient, the Table 4.3.2.1 shows the relationship between these two variables considered in the analysis. Table 4.3.2.1 Correlation on minimum rate charged on loan facility interest rate quotes to customers’ aid in loan recovery Interest rate quotes to Minimum rate customers aid charged on a in loan loan facility recovery Minimum rate charged on Pearson Correlation 1 .511* a loan facility Sig. (2-tailed) .025 N 20 19 Interest rate quotes to Pearson Correlation .511* 1 customers aid in loan Sig. (2-tailed) .025 recovery N 19 19 *. Correlation is significant at the 0.05 level (2-tailed). Source: Field Survey Data, April 2019 Table 4.3.2.2 Paired sample statistics for minimum rate charged on loan facility interest rate quotes to customer’s aid in loan recovery Std. Error Mean N Std. Deviation Mean Pair 1 Minimum rate charged 2.11 19 1.449 .332 on a loan facility Interest rate quotes to customers aid in loan 1.21 19 .419 .096 recovery Source: Field Survey Data, April 2019 54 University of Ghana http://ugspace.ug.edu.gh Table 4.3.2.1 Paired Sample test for minimum rate charged on loan facility interest rate quotes to customers aid in loan recovery Paired Differences 95% Confidence Std. Interval of the Sig. Std. Error Difference (2- Mean Deviation Mean Lower Upper t df tailed) Pair Minimum rate 1 charged on a loan facility - Interest .895 1.286 .295 .275 1.515 3.032 18 .007 rate quotes to customers aid in loan recovery Source: Field Survey Data, April 2019 The results above indicate a fairly strong positive correlation between minimum rate charged on a loan facility and Interest rate quotes to customers aid in loan recovery with a correlation coefficient of 𝑟 = .511 with a (2-tailed test) meaning that as minimum rate charged increases it aids in loan recovery in the MFIs. From Table 4.3.2.2, there is a missing value pairwise on the test variables since 𝑁 ≠ 20. Mean for the minimum rate charged on a loan facility (2.11) is greater than the mean of interest rate quotes to customers aid in loan recovery (1.21). On the average, loan recovery increases ∝= 0.895 ≈ 89.5%. This shows the strength of the correlation in Table 4.3.2.1. Denoted by “Sig. (2-tailed)”, the p-value is 0.007. Meaning there is a 7% chance of getting this results per the population. The p-value consist of a 3.5% chance of finding a difference < - .895 and another 3.5% chance of finding of finding a difference >.895. This is depicted in Table 4.3.2.3. With a t-test of 3.032 and a significance level of 0.05 = 5%, I fail to accept the null hypotheses: Minimum rate do not affect interest rate of MFIs. This is because the P- value on both extreme differences is less than the significance level. Thus 0.035 < 0.05. 55 University of Ghana http://ugspace.ug.edu.gh 4.4 How Liquidity Risk Affects the Interest Rate This ascertain how liquidity risk affects the interest rate among the various microfinance institutions. 4.4.1 Descriptive Statistical Analysis Under this analysis, the researcher deployed the following to describe and present data in line with the objectives and research questions; frequency tables and charts. Table 4.4.1.1 Responses on the role liquidity play in calculating interest rate Cumulative Frequency Percent Valid Percent Percent Valid yes 20 100.0 100.0 100.0 Source: Field Survey Data, April 2019 Table 4.4.1.2 The quest to stay liquid affects the chase for interest Cumulative Frequency Percent Valid Percent Percent Valid increase 13 65.0 65.0 65.0 decrease 7 35.0 35.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 As a step to evaluating liquidity risk and recovery rate, the tables above depict responses received regarding the role of liquidity and the quest to stay liquid thus 100% of the total frequency. All respondent responded that “yes, liquidity play a role in calculating interest rate. Again, 65% of the respondents were of the opinion that the quest to stay liquid increases the chase for interest. 35% of the population had a decreasing opinion about the phenomena. 56 University of Ghana http://ugspace.ug.edu.gh Table 4.4.1.3 Extent to which interest rate is affected Cumulative Frequency Percent Valid Percent Percent Valid 1-10% 11 55.0 55.0 55.0 11-20% 7 35.0 35.0 90.0 41-50%+ 1 5.0 5.0 95.0 51%+ 1 5.0 5.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 The extent to which interest rate is affected was rated by respondents of the questionnaire. 55% rated 1-10% while the 35% was rated by simple minority to be 11- 20%. 41-50% and 51%+ was rated by 5% for each category. Below is a bar chart in percentages. Figure 4.4.1.1 Extent to which interest rate is affected by liquidity 57 University of Ghana http://ugspace.ug.edu.gh Table 4.4.1.4 Case processing summary Average Peak month of high recovery * Average peak month of low recovery Cases Valid Missing Total N Percent N Percent N Percent Average Peak month of high recovery * 16 80.0% 4 20.0% 20 100.0% Average peak month of low recovery Source: Field Survey Data, April 2019 Table 4.4.1.5 Average Peak month of high recovery * Average peak month of low recovery cross tabulation Average peak month of low recovery january june august september december Total Average Peak april 0 0 0 0 1 1 month of high may 0 0 1 0 1 2 recovery june 0 0 1 1 1 3 july 3 0 0 2 0 5 august 0 0 0 0 2 2 september 0 0 0 0 1 1 october 0 0 0 0 1 1 december 0 1 0 0 0 1 Total 3 1 2 3 7 16 Source: Field Survey Data, April 2019 58 University of Ghana http://ugspace.ug.edu.gh Figure 4.4.1.2 Bar chart of average peek month of low loan recovery Figure 4.4.1.3 Bar chart of average peek month of high loan recovery 59 University of Ghana http://ugspace.ug.edu.gh Figure 4.4.1.4 Histogram showing the dispersion of average peek month of high loan recovery The Average peak month of high and low loan recovery was asked in questions 14 and 15 in the questionnaire. Out of the twenty (20) expected respondents, sixteen (16) answered these questions constituting 80% of the total frequency. From the cross tabulation, three (3) respondents acknowledged the firm’s average high loan recovery and average low loan recovery to be July and January respectively depicting the highest combination in the cross table. Some other months of average high loan recovery are April, May, June, August September, October and December while June, August, September and December were picked for average low loan recovery months. For average peak month of low loan recovery, December recorded the highest with 45% of the total valid population followed by January and September with 20% for each month. June recorded the lowest with 5% and August with 15%. The other months were not considered for the question by the respondents. For average peak month of high loan 60 University of Ghana http://ugspace.ug.edu.gh recovery, July recorded the highest with 30% as April, September, October and December recorded the lowest with 5% each. The results for the average peek month of high loan recovery, the results depict a normal distribution as shown in the histogram above. With a mean of 7.13, the population was far from the mean using the standard deviation of 1.996. Table 4.4.1.6. Frequency table showing how early liquidity risk are identified Cumulative Frequency Percent Valid Percent Percent Valid 1week 5 25.0 25.0 25.0 2weeks 1 5.0 5.0 30.0 3weeks 1 5.0 5.0 35.0 4weeks 9 45.0 45.0 80.0 others please specify 4 20.0 20.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 Table 4.4.1.7 Number of months in a year high loan recovery is experienced showed in frequencies Cumulative Frequency Percent Valid Percent Percent Valid one-three months 5 25.0 25.0 25.0 four-six months 9 45.0 45.0 70.0 seven-nine months 5 25.0 25.0 95.0 ten-twelve months 1 5.0 5.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 61 University of Ghana http://ugspace.ug.edu.gh Figure 4.4.1.5 Pie chart of number of months in a year high loan recovery is experienced Most firms take four (4) weeks to identify liquidity while some few others take a week to identify risk as shown in Table 4.3.1.6. In Table 4.4.1.7, frequencies and percentages tell that within a year high loan recovery is experienced averagely for four-six months as indicated by 45% of the total population. 25% of the firm’s experience high loan recovery for one-three months or seven-nine months. Above is a pie chart depicting the results in percentages. Table 4.4.1.8 Case Processing Summary of How early liquidity risk is identified * How long it averagely takes to recover from liquidity risk Cases Valid Missing Total N Percent N Percent N Percent How early liquidity risk is identified * How long it averagely takes to 20 100.0% 0 0.0% 20 100.0% recover from liquidity risk Source: Field Survey Data, April 2019 62 University of Ghana http://ugspace.ug.edu.gh Table 4.4.1.9 How early liquidity risk is identified * How long it averagely takes to recover from liquidity risk Cross tabulation How long it averagely takes to recover from liquidity risk within the first quarter after half year into after a year detection detection of detection Total How early 1week 2 2 1 5 liquidity risk is 2weeks 0 1 0 1 identified 3weeks 0 0 1 1 4weeks 3 2 4 9 others please 0 2 2 4 specify Total 5 7 8 20 Source: Field Survey Data, April 2019 From the case processing summary in Table 4.4.1.8, all respondents answered these two questions. Most respondents are of the view the that it takes their firm 4weeks to identify liquidity risk and they are able to recover from this risk after a year of detection. On the rear side on this analysis, some respondents added that their firms averagely take within the first quarter after detection to recover from liquidity risk yet they take about 4weeks to identify such a risk. In the mist of rear situations, some firms a week to identify risk but at least take first quarter, half year or after a year to recover. Also, other firms were of a varied opinion which was not captured in the options given as to how early risk was identified in their firms. Below is a graphical presentation of the cross tabulation. 63 University of Ghana http://ugspace.ug.edu.gh Figure 4.4.1.6 Clustered bar chart of how early liquidity risk is identified * how long it averagely takes to recover from liquidity risk Table 4.4.1.10 Case Processing Summary Group of customers that are affected after normalization * Different rates are quoted to customers during distressed periods Cases Valid Missing Total N Percent N Percent N Percent 64 University of Ghana http://ugspace.ug.edu.gh Group of customers that are affected after normalization * Different rates are 19 95.0% 1 5.0% 20 100.0% quoted to customers during distressed periods Source: Field Survey Data, April 2019 Table 4.4.1.11 Group of customers that are affected after normalization * Different rates are quoted to customers during distressed periods Cross tabulation Different rates are quoted to customers during distressed periods yes no Total Group of customers that Those with existing 5 2 7 are affected after loan facilities normalization Those who are given credit after 11 1 12 normalization Total 16 3 19 Source: Field Survey Data, April 2019 From this tables, 19 respondents gave answers to the questions asked. Eleven (11) firms agreed that different interest rate is quoted to customers during their distressed periods and that only those who are given credit after the normalization are affected while five (5) of them agreed but after the normalization, only those with existing loan facilities are affected. A clustered bar chart is shown below. 65 University of Ghana http://ugspace.ug.edu.gh Figure 4.4.1.7 Clustered bar chart Group of customers that are affected after normalization * Different rates are quoted to customers during distressed period 4.4.2 Inferential Statistical Analysis Using the Pearson’s pairwise correlation coefficient, the Table 4.4.2.1 shows the relationship between these two variables considered in the analysis. Table 4.4.2.1 Correlation on the quest to stay liquid affect the chase for interest and extent to which interest rate is affected The quest to stay liquid Extent to which affect the chase interest rate is for interest affected The quest to stay liquid Pearson Correlation 1 .348 affect the chase for interest Sig. (2-tailed) .133 N 20 20 Extent to which interest Pearson Correlation .348 1 rate is affected Sig. (2-tailed) .133 N 20 20 Source: Field Survey Data, April 2019 66 University of Ghana http://ugspace.ug.edu.gh Table 4.4.2.2 Paired sample statistics for the quest to stay liquid affect the chase for interest and extent to which interest rate is affected Std. Std. Error Mean N Deviation Mean Pair 1 The quest to stay liquid affect the chase for 1.35 20 .489 .109 interest Extent to which interest 1.80 20 1.361 .304 rate is affected Source: Field Survey Data, April 2019 Table 4.4.2.3 Paired sample test the quest to stay liquid affect the chase for interest and extent to which interest rate is affected Paired Differences 95% Confidence Std. Interval of the Sig. Std. Error Difference (2- Mean Deviation Mean Lower Upper t df tailed) Pair The quest to 1 stay liquid affect the chase for interest - -.450 1.276 .285 -1.047 .147 -1.577 19 .131 Extent to which interest rate is affected Source: Field Survey Data, April 2019 The results above indicate a positive correlation between the quest to stay liquid affect the chase for interest and extent to which interest rate is affected with a correlation coefficient of 𝑟 = .348 with a (2-tailed test) meaning that as the quest to stay liquid increases it increases the effect of liquidity on interest rate in the MFIs. Even though 67 University of Ghana http://ugspace.ug.edu.gh there is a positive correlation, this is not significant and hence liquidity risk cannot be said to be a driving force of interest rate. From Table 4.4.2.2, there is no missing value pairwise on the test variables since 𝑁 = 20. Mean for the quest to stay liquid affect the chase for interest (1.35) is less than the mean of extent to which interest rate is affected (1.80). On the average, interest rate decreases ∝= −0.450 ≈ −45.0%. This shows the strength of the correlation in Table 4.4.2.1. Denoted by “Sig. (2-tailed)”, the p-value is 0.131. Meaning there is a 13.10% chance of getting this results per the population. The p-value consist of a 6.55% chance of finding a difference < - (-.450) and another 6.55% chance of finding of finding a difference > (-.450). This is depicted in Table 4.4.2.3. With a t-test of -1.577 and a significance level of 0.05 = 5%, I accept the null hypotheses: Liquidity risk has no relationship with the interest rate of the MFIs. This is because the P-value on both extreme differences is greater than the significance level. Thus 0.0655 > 0.05. 4.5 Operational Cost and Its Effect on Loan Recovery This evaluate the operational cost and its effects loan recovery rates in the microfinance institutions. 4.5.1 Descriptive Statistical Analysis In order to attain the objective, the following were describe and present data in line with the objectives and research questions: frequency tables, cross tabulations, group tables and clustered bar charts. 68 University of Ghana http://ugspace.ug.edu.gh Table 4.5.1.1 Responses in frequencies on the composition of operational cost Responses Percent of N Percent Cases Administrative cost 19 14.4% 95.0% Salaries 20 15.2% 100.0% Legal fees 11 8.3% 55.0% Rent 19 14.4% 95.0% $ Composition of Operational Office supplies 15 11.4% 75.0% Costa Directors fees 8 6.1% 40.0% Utility 19 14.4% 95.0% Advertisement 10 7.6% 50.0% Insurance Premium 11 8.3% 55.0% Total 132 100.0% 660.0% a. Dichotomy group tabulated at value 1. Source: Field Survey Data, April 2019 Table 4.5.1.1 “Composition of Operational Costa column indicates the what constitutes operational cost in the various MFIs and out of these components salaries emerged the most considered component in the calculation of operational cost. It recorded 100% thus all firms consider salaries for operational cost. This constituted 15.5% of the total frequency. Administrative cost, rent and utility were also highly considered as it constituted 14.4% of the total frequency. This was followed by office supplies with 11.4% then legal fees and insurance premium. Advertisement and Directors fees were the least considered with 7.6% and 6.1% of the total frequency respectively. Table 4.5.1.2 Operational cost affect interest rate * Proportion of interest rate it covers Cross tabulation Proportion of interest rate it covers 1-10% 11-20% 31-40% 41-50% Total Operational cost affect yes 12 2 1 1 16 interest rate no 2 0 0 1 3 Total 14 2 1 2 19 Source: Field Survey Data, April 2019 69 University of Ghana http://ugspace.ug.edu.gh Figure 4.5.1.1 Clustered bar chart showing Operational cost affect interest rate * Proportion of interest rate it covers A total of sixteen (16) firms agrees that operational cost affect interest rate and three (3) disagrees. Out the sixteen (16), twelve (12) of them were of the view that operational cost affect interest rate by 1-10%. While the others were for 11-20%, 31-40% and 41- 50% there was a missing value. 70 University of Ghana http://ugspace.ug.edu.gh Table 4.5.1.3 To what extent does operational risk affect recovery * Operational cost in Interest rate calculation affect recovery Cross tabulation Operational cost in Interest rate calculation affect recovery yes no Total To what extent does 1-10% 5 2 7 operational risk affect 11-20% 5 0 5 recovery 21-30% 1 1 2 51%+ 0 3 3 Total 11 6 17 Source: Field Survey Data, April 2019 Figure 4.5.1.2 To what extent does operational risk affect recovery * Operational cost in Interest rate calculation affect recovery in a clustered bar chart Table 4.5.1.3 shows that operational cost affect loan recovery as eleven (11) responded to that. Six (6) of the respondents disagrees to this. Ten (10) responses shows that operational cost in interest rate calculation affects loan recovery 1-10% and 11-20% (equally distributed). Above is a graph of this analysis. 71 University of Ghana http://ugspace.ug.edu.gh Inferential Statistical Analysis The Pearson’s pairwise correlation coefficient was used for this analysis. The Tables show the relationship between these two variables considered in the analysis. Table 4.5.2.1 Correlation on operational cost in interest rate calculation affect recovery and to what extent does operational risk affect recovery Operational To what extent cost in Interest does rate calculation operational risk affect recovery affect recovery Operational cost in Interest Pearson Correlation 1 .588* rate calculation affect Sig. (2-tailed) .013 recovery N 19 17 To what extent does Pearson Correlation .588* 1 operational risk affect Sig. (2-tailed) .013 recovery N 17 17 *. Correlation is significant at the 0.05 level (2-tailed). Source: Field Survey Data, April 2019 Table 4.5.2.2 Paired sample statistics for operational cost in Interest rate calculation affect recovery and to what extent does operational risk affect recovery Std. Std. Error Mean N Deviation Mean Pair 1 Operational cost in Interest rate calculation 1.35 17 .493 .119 affect recovery To what extent does operational risk affect 2.41 17 1.839 .446 recovery Source: Field Survey Data, April 2019 72 University of Ghana http://ugspace.ug.edu.gh Table 4.5.2.3 Paired sample test for operational cost in Interest rate calculation affect recovery and to what extent does operational risk affect recovery Paired Differences 95% Confidence Std. Interval of the Sig. Std. Error Difference (2- Mean Deviation Mean Lower Upper T df tailed) Pair Operational cost in 1 Interest rate calculation affect - - recovery - To what 1.600 .388 -1.881 -.236 16 .015 1.059 2.729 extent does operational risk affect recovery Source: Field Survey Data, April 2019 The result shows a strong positive correlation between operational cost in interest rate calculation affect recovery and extent to which loan recovery is affected with a correlation coefficient of 𝑟 = .588 with a (2-tailed test) meaning that as the operational cost factored in interest rate calculation affecting loan recovery rises, the extent of the effect of operational risk on loan recovery also rises. From Table 4.5.2.2, there are three (3) missing values pairwise on the test variables since 𝑁 ≠ 20. Mean for operational cost in interest rate calculation affect recovery (1.35) is less than the mean of extent to which interest rate is affected (2.41). On the average, interest rate decreases ∝= −1.059 ≈ −105.9%. This shows the strength of the correlation in Table 4.4.2.1. Denoted by “Sig. (2-tailed)”, the p-value is 0.015. Meaning there is a 1.5% chance of getting this results per the population. The p- value consist of a 0.75% chance of finding a difference < - (-1.059) and another 0.75% chance of finding of finding a difference > (-1.059). This is depicted in Table 4.5.2.3. With a t-test of -2.729 and a significance level of 0.05 = 5%, I fail to accept the null hypotheses: Operational cost do not affect loan recovery rate. This is because the P- 73 University of Ghana http://ugspace.ug.edu.gh value on both extreme differences is greater than the significance level. Thus 0.75 < 0.05. 4.6 Credit Risk Faced by MFI’s and Its Impact On Recovery Rate. The objective investigate the credit risk faced by the MFIs and its impact on loan recovery. 4.6.1 Descriptive Statistical Analysis The analysis used frequency table, cross tabulations, group tables and clustered bar charts to attain the objective. Table 4.6.1.1 Proportion of interest rate to credit risk * Credit risk is faced Cross tabulation Credit risk is faced Yes Total Proportion of interest rate to credit risk 1-10% 10 10 11-20% 4 4 21-30% 2 2 51%+ 3 3 Total 19 19 Source: Field Survey Data, April 2019 From this tabulation, all respondents but for one (1) responded and said yes, credit risk is faced by their firms. In their view,10 firms said1-10% of interest rate is taken by credit risk. 11-20%, 21-30% and 51%+ were proportions covered by interest rate in 4, 2 and 3 firms respectively. 74 University of Ghana http://ugspace.ug.edu.gh Table 4.6.1.2 Frequency on interest rate of customers who do not possess credit risk is affected Cumulative Frequency Percent Valid Percent Percent Valid yes 13 65.0 72.2 72.2 no 5 25.0 27.8 100.0 Total 18 90.0 100.0 Missing System 2 10.0 Total 20 100.0 Source: Field Survey Data, April 2019 From this frequency, interest rate of customers who do not possess credit risk are affected. Table 4.6.1.3 There are credit policies governing disbursement and loan recovery * Personnel’s or committees are in charge of policies Cross tabulation Personnel’s or committees are in charge of policies yes No Total There are credit policies yes governing disbursement and 15 4 19 loan recovery Total 15 4 19 Source: Field Survey Data, April 2019 Table 4.6.1.3 show that there are credit policies governing disbursement and loan recovery and there are personnel’s or committees in charge of these policies. 75 University of Ghana http://ugspace.ug.edu.gh Table 4.6.1.4 Frequency on the extent credit policies used contribute to loan recovery Cumulative Frequency Percent Valid Percent Percent Valid 1-20% 7 35.0 36.8 36.8 21-40% 3 15.0 15.8 52.6 41-60% 6 30.0 31.6 84.2 61-80% 2 10.0 10.5 94.7 81-100% 1 5.0 5.3 100.0 Total 19 95.0 100.0 Missing System 1 5.0 Total 20 100.0 Source: Field Survey Data, April 2019 Credit policies contribute to loan recovery according to Table 4.6.1.4. By simple majority, credit policies contribute to loan recovery by 1-20%. 41-60 contribution to loan recovery was also acknowledged. Other levels of contribution included 21-40%, 61-80% and 81=100%. The bar chart below gives a pictorial view of the levels. Figure 4.6.1.1 Bar chart in percentages of the extent credit policies used contribute to loan recovery 76 University of Ghana http://ugspace.ug.edu.gh Table 4.6.1.5 Non-compliance that accounts for ineffective credit administration * Non-compliance with credit policies accounts for ineffective credit administration Cross tabulation Non-compliance with credit policies accounts for ineffective credit administration yes No Total Non-compliance that Customer 2 0 2 accounts for ineffective pressure credit administration management 1 0 1 pressure Inexperience of 4 1 5 implementers All of the above 12 0 12 Total 19 1 20 Source: Field Survey Data, April 2019 In Table 4.6.1.5, results show that non-compliance with credit policies accounts for ineffective credit administration. Customer pressure, management pressure, inexperience of implementers are all factors that accounts for the non-compliance of effective credit administration. Table 4.6.1.6 How often credit policies undergo review * Credit policies reflect macroeconomic factors Cross tabulation Credit policies reflect macroeconomic factors Yes Total How often credit policies monthly 1 1 undergo review quarterly 3 3 Semi-annually 4 4 Annually 9 9 others, please specify 1 1 Total 18 18 Source: Field Survey Data, April 2019 77 University of Ghana http://ugspace.ug.edu.gh These credit policies are reviewed annually by most firms. Some other firms review them semi- annually and quarterly. Table 4.6.1.7 Responses in Frequency on Macroeconomic Factors Frequencies Responses Percent of N Percent Cases $Macroeconomic_Fact Economic growth 7 15.2% 36.8% orsa Inflation 15 32.6% 78.9% Lending interest rate 19 41.3% 100.0% GDP growth 3 6.5% 15.8% Unemployment 2 4.3% 10.5% Total 46 100.0% 242.1% a. Dichotomy group tabulated at value 1. Source: Field Survey Data, April 2019 Table 4.6.1.2 shows that lending interest is considered by firms when setting credit policies for administrative purposes. It covered 41.3% of the total frequency. Inflation was next considered by firms which covered about 32.6% of the total frequency. With 36.6% from the total responses and 15.2% of the total frequency, economic growth was considered by these firms. GDP growth and unemployment were least considered by MFIs with a 6.5% and 4.3% of total frequency respectively were considered. 4.6.2 Inferential Statistical Analysis The Pearson’s pairwise correlation coefficient was used for this analysis. The Tables show the relationship between these two variables considered in the analysis. 78 University of Ghana http://ugspace.ug.edu.gh Table 4.6.2.1 Correlation on proportion of interest rate to credit risk and extent credit policies used contribute to loan recovery Extent credit Proportion of policies used interest rate to contribute to credit risk loan recovery Proportion of interest rate Pearson Correlation 1 .567* to credit risk Sig. (2-tailed) .014 N 19 18 Extent credit policies used Pearson Correlation .567* 1 contribute to loan recovery Sig. (2-tailed) .014 N 18 19 *. Correlation is significant at the 0.05 level (2-tailed). Source: Field Survey Data, April 2019 Table 4.6.2.2 Paired sample statistic for proportion of interest rate to credit risk and extent credit policies used contribute to loan recovery Std. Std. Error Mean N Deviation Mean Pair 1 Proportion of interest 2.28 18 1.841 .434 rate to credit risk Extent credit policies used contribute to loan 2.28 18 1.274 .300 recovery Source: Field Survey Data, April 2019 79 University of Ghana http://ugspace.ug.edu.gh Table 4.6.2.3 Paired sample test for proportion of interest rate to credit risk and extent credit policies used contribute to loan recovery Paired Differences 95% Confidence Std. Interval of the Sig. Std. Error Difference (2- Mean Deviation Mean Lower Upper t df tailed) Pair Proportion of 1 interest rate to credit risk - Extent credit .000 1.534 .362 -.763 .763 .000 17 1.000 policies used contribute to loan recovery Source: Field Survey Data, April 2019 The result shows a strong positive correlation between Proportion of interest rate to credit risk and Extent credit policies used contribute to loan recovery with a correlation coefficient of 𝑟 = .567 with a (2-tailed test) meaning that as the proportion credit risk covers in interest rate increases, loan recovery is positively contributed to. Even though there is a positive relationship, this is not significant and hence liquidity risk cannot be said to be a driving force of interest rate. From Table 4.6.2.2, there are two (2) missing values pairwise on the test variables since 𝑁 ≠ 20. Mean for Proportion of interest rate to credit risk (2.28) is equal to the mean of extent credit policies used contribute to loan recovery (2.28). this shows that there is a zero (0%) chance of finding this result On the average, interest rate increases ∝= 0.000. This shows the strength of the correlation in Table 4.6.2.1. Denoted by “Sig. (2-tailed)”, the p-value is 1.000. Meaning there is a 0.5% chance of getting this results per the population. The p-value consist of a 0.5% chance of finding a difference < - 0.000 and another 0.75% chance of finding 80 University of Ghana http://ugspace.ug.edu.gh of finding a difference > 0.000 but with the same population mean, there is a 0% chance of getting the same result. This is depicted in Table 4.6.2.3. With a t-test of 0.000 and a significance level of 0.05 = 5%, I fail to accept the null hypotheses: There is no relationship between credit risk and loan recovery of MFI’s. This is because from the test, the contribution to loan recovery increases with just an increase in the proportion credit risk covers. 4.7 Verify Ways Interest Rate Premium are set to Aid in Loan Recovery This objective sought to investigate the ways MFIs set interest rate premiums to aid in loan recovery. This is to enable the MFIs find other ways to impact loan recovery apart from the traditional debt collection hunt of customers. 4.7.1 Descriptive Statistical Analysis Table 4.7.1.1 Interest rate premiums applied depends on credit risk of the customer Cumulative Frequency Percent Valid Percent Percent Valid Strongly Disagree 1 5.0 5.3 5.3 Disagree 1 5.0 5.3 10.5 Uncertain 2 10.0 10.5 21.1 Agree 8 40.0 42.1 63.2 Strongly Agree 7 35.0 36.8 100.0 Total 19 95.0 100.0 Missing System 1 5.0 Total 20 100.0 Source: Field Survey Data, April 2019 81 University of Ghana http://ugspace.ug.edu.gh Table 4.7.1.2 Interest rate applied is same for all customers Cumulative Frequency Percent Valid Percent Percent Valid Strongly Disagree 5 25.0 25.0 25.0 Disagree 7 35.0 35.0 60.0 Uncertain 5 25.0 25.0 85.0 Agree 3 15.0 15.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 Table 4.7.1.3 An approach adopted is not changed till the loan is paid Cumulative Frequency Percent Valid Percent Percent Valid Strongly Disagree 2 10.0 10.0 10.0 Disagree 4 20.0 20.0 30.0 Uncertain 5 25.0 25.0 55.0 Agree 8 40.0 40.0 95.0 Strongly Agree 1 5.0 5.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 Table 4.7.1.4 The approach influences loan repayment Cumulative Frequency Percent Valid Percent Percent Valid Disagree 1 5.0 5.3 5.3 Uncertain 7 35.0 36.8 42.1 Agree 7 35.0 36.8 78.9 Strongly Agree 4 20.0 21.1 100.0 Total 19 95.0 100.0 Missing System 1 5.0 Total 20 100.0 Source: Field Survey Data, April 2019 82 University of Ghana http://ugspace.ug.edu.gh Table 4.7.1.5 Fixed or flat interest rate approach results in high loan recovery rate Cumulative Frequency Percent Valid Percent Percent Valid Strongly Disagree 2 10.0 10.0 10.0 Disagree 8 40.0 40.0 50.0 Uncertain 6 30.0 30.0 80.0 Agree 4 20.0 20.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 Table 4.7.1.6 Floating rate approach results in high loan recovery rate Cumulative Frequency Percent Valid Percent Percent Valid Disagree 5 25.0 25.0 25.0 Uncertain 11 55.0 55.0 80.0 Agree 4 20.0 20.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 Table 4.7.1.7 Reducing balance method is used for the calculation of loan repayment schedule Cumulative Frequency Percent Valid Percent Percent Valid Strongly Disagree 3 15.0 15.0 15.0 Disagree 1 5.0 5.0 20.0 Uncertain 4 20.0 20.0 40.0 Agree 9 45.0 45.0 85.0 Strongly Agree 3 15.0 15.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 83 University of Ghana http://ugspace.ug.edu.gh Table 4.7.1.8 Straight line method is used for the calculation of loan repayment schedule Cumulative Frequency Percent Valid Percent Percent Valid Strongly Disagree 4 20.0 20.0 20.0 Disagree 7 35.0 35.0 55.0 Uncertain 3 15.0 15.0 70.0 Agree 1 5.0 5.0 75.0 Strongly Agree 5 25.0 25.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 Table 4.7.1.9 There is high loan recovery rate when reducing balance is used Cumulative Frequency Percent Valid Percent Percent Valid Strongly Disagree 1 5.0 5.0 5.0 Uncertain 7 35.0 35.0 40.0 Agree 9 45.0 45.0 85.0 Strongly Agree 3 15.0 15.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 Table 4.7.1.10 Loan recovery is high when straight line method is used Cumulative Frequency Percent Valid Percent Percent Valid Strongly Disagree 2 10.0 10.0 10.0 Disagree 6 30.0 30.0 40.0 Uncertain 7 35.0 35.0 75.0 Agree 3 15.0 15.0 90.0 Strongly Agree 2 10.0 10.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 84 University of Ghana http://ugspace.ug.edu.gh Table 4.7.1.11 Customers opinion is considered when deciding on which method to use Cumulative Frequency Percent Valid Percent Percent Valid Strongly Disagree 2 10.0 10.0 10.0 Disagree 9 45.0 45.0 55.0 Uncertain 3 15.0 15.0 70.0 Agree 1 5.0 5.0 75.0 Strongly Agree 5 25.0 25.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 Table 4.7.1.12 Loan repayment schedule is made in consultation with the customer Cumulative Frequency Percent Valid Percent Percent Valid Strongly Disagree 1 5.0 5.0 5.0 Disagree 6 30.0 30.0 35.0 Uncertain 1 5.0 5.0 40.0 Agree 5 25.0 25.0 65.0 Strongly Agree 7 35.0 35.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 Table 4.7.1.13 Loan repayments are rescheduled when there is default Cumulative Frequency Percent Valid Percent Percent Valid Agree 12 60.0 60.0 60.0 Strongly Agree 8 40.0 40.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 85 University of Ghana http://ugspace.ug.edu.gh Table 4.7.1.14 Customers pay back loans when repayment is rescheduled Cumulative Frequency Percent Valid Percent Percent Valid Strongly Disagree 1 5.0 5.0 5.0 Disagree 5 25.0 25.0 30.0 Uncertain 9 45.0 45.0 75.0 Agree 5 25.0 25.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 Table 4.7.1.15 Customers attitude affect loan repayment Cumulative Frequency Percent Valid Percent Percent Valid Disagree 1 5.0 5.0 5.0 Uncertain 2 10.0 10.0 15.0 Agree 3 15.0 15.0 30.0 Strongly Agree 14 70.0 70.0 100.0 Total 20 100.0 100.0 Source: Field Survey Data, April 2019 4.7.2 Inferential Statistical Analysis The Pearson’s pairwise correlation coefficient was used for this analysis. The Tables show the relationship between these two variables considered in the analysis. 86 University of Ghana http://ugspace.ug.edu.gh Table 4.7.2.1 Correlation on the approach influences loan repayment and there is high loan recovery rate when reducing balance is used There is high loan recovery The approach rate when influences loan reducing repayment. balance is used The approach influences Pearson Correlation 1 -.643** loan repayment. Sig. (2-tailed) .003 N 19 19 There is high loan recovery Pearson Correlation -.643** 1 rate when reducing balance Sig. (2-tailed) .003 is used N 19 20 **. Correlation is significant at the 0.01 level (2-tailed). Source: Field Survey Data, April 2019 Table 4.7.2.2. Paired sample statistic for the approach influences loan repayment and there is high loan recovery rate when reducing balance is used Std. Std. Error Mean N Deviation Mean Pair 1 The approach influences 3.74 19 .872 .200 loan repayment. There is high loan recovery rate when 3.58 19 .902 .207 reducing balance is used Source: Field Survey Data, April 2019 From Table 4.7.2.2, there is a missing values pairwise on the test variables since 𝑁 ≠ 20. Mean for the approach influences loan repayment. (3.74) is greater than the mean of there is high loan recovery rate when reducing balance is used (3.58). 87 University of Ghana http://ugspace.ug.edu.gh Table 4.7.2.3 Paired sample test for the approach influences loan repayment and there is high loan recovery rate when reducing balance is used Paired Differences 95% Confidence Std. Interval of the Sig. Std. Error Difference (2- Mean Deviation Mean Lower Upper t df tailed) Pair The approach 1 influences loan repayment. - There is high .158 1.608 .369 -.617 .933 .428 18 .674 loan recovery rate when reducing balance is used Source: Field Survey Data, April 2019 The result shows a strong negative correlation between the approach influences loan repayment. and there is high loan recovery rate when reducing balance is used with a correlation coefficient of 𝑟 = −.643 with a (2-tailed test) meaning that as loan repayment increases due to the approach loan recovery decreases when reducing balance method is used On the average, loan repayment increases at ∝= 0.158. This shows the strength of the correlation in Table 4.7.2.1. Denoted by “Sig. (2-tailed)”, the p-value is .674. Meaning there is a 33.7% chance of getting this results per the population. The p-value consist of a 33.7% chance of finding a difference < .158 and another 33.7% chance of finding of finding a difference > .158. This is depicted in Table 4.7.2.3. With a t-test of .428 and a significance level of 0.05 = 5%, I accept the null hypotheses: Loan recovery rate does not depend on how interest premiums are by MFIs. This is 88 University of Ghana http://ugspace.ug.edu.gh because the P-value on both extreme differences is greater than the significance level. Thus 0.75 > 0.05. Table 4.7.2.4 Paired sample statistics of paired variables Std. Std. Error Mean N Deviation Mean Pair 1 The approach influences 3.74 19 .872 .200 loan repayment. Customers attitude 4.47 19 .905 .208 affect loan repayment Pair 2 The approach influences 3.74 19 .872 .200 loan repayment. Reducing balance method is used for the 3.32 19 1.250 .287 calculation of loan repayment schedule Pair 3 The approach influences 3.74 19 .872 .200 loan repayment. Straight line method is used for the calculation 2.68 19 1.455 .334 of loan repayment schedule Source: Field Survey Data, April 2019 Table 4.7.2.5 Paired sample correlation of paired variables N Correlation Sig. Pair 1 The approach influences loan repayment. & Customers 19 -.185 .448 attitude affect loan repayment Pair 2 The approach influences loan repayment. & Reducing balance method is used for the 19 -.327 .171 calculation of loan repayment schedule Pair 3 The approach influences loan repayment. & Straight line method is used for the 19 .325 .175 calculation of loan repayment schedule Source: Field Survey Data, April 2019 89 University of Ghana http://ugspace.ug.edu.gh Table 4.7.2.6 Paired samples test of paired variables Paired Differences 95% Confidence Std. Interval of the Sig. Std. Error Difference (2- Mean Deviation Mean Lower Upper t df tailed) Pair The 1 approach influences loan - repayment. - -.737 1.368 .314 -1.396 -.078 18 .031 2.348 Customers attitude affect loan repayment Pair The 2 approach influences loan repayment. - Reducing balance .421 1.742 .400 -.419 1.261 1.053 18 .306 method is used for the calculation of loan repayment schedule Pair The 3 approach influences loan repayment. - Straight line 1.053 1.433 .329 .362 1.743 3.203 18 .005 method is used for the calculation of loan repayment schedule Source: Field Survey Data, April 2019 90 University of Ghana http://ugspace.ug.edu.gh 4.8 Hypothesis Testing After the analysis of the data collected from the field, it was identified that all objectives set was achieved hence the corresponding hypothesis to each of these objectives was measured having an impact on the recovery rate of MFIs. 1. (H01): Minimum rate do not affect loan recovery rate of MFIs. (HA1): Minimum rate affect loan recovery rate of MFIs. From the analysis I accept HA1 meaning there is a fairly strong relationship between minimum rate and loan recovery rate. With a periodic review and a mark-up way of pricing, majorly accepted by respondents, the base rate affects the loan recovery rate. Also, 75% of MFIs contacted are of the view that interest rate quotes to customers’ aid in loan recovery 2. (H02): Liquidity risk has no relationship with the interest rate of the MFIs. (HA2): Liquidity risk has a relationship with the interest rate of the MFIs. Here, H02 is accepted indicating liquidity risk has no relationship with interest rate of the MFIs. From this, it was identified the quest to stay liquid has positive impact (increase) on the interest rate (Table 4.4.1.2) but at no significant levels. 3. (H03): Operational cost do not affect loan recovery rate. (HA3): Operational cost affect loan recovery rate. HA3 is accepted since operational cost affect loan recovery rate. At 0.588, the extent of relationship which indicates a strong positive correlation between operational cost and loan recovery rate. 4. (H04): There is no relationship between credit risk and loan recovery of MFI’s. (HA4): There is a relationship between credit risk and loan recovery of MFI’s. 91 University of Ghana http://ugspace.ug.edu.gh Using correlational analysis, the contribution to loan recovery increases with just an increase in the proportion credit risk covers. Due to the positive effect, HA4 is accepted with a level of significance = 0.05. 5. (H05): Loan recovery rate does not depend on how interest premiums are set by MFIs. (HA5): Loan recovery rate depends on how interest premiums are set by MFIs. Here, H05 is accepted. This show that loan recovery rate does not depend on how interest rate premiums are set at a significance level of 0.01. 4.9 Conclusion This chapter analysed responses of personnels’ from microfinance institutions through descriptive and inferential statistics in forms of tables, graphs and charts. The results depicted if certain factors play a role in the loan recovery rate by Microfinance institutions and to what extent these factors affect the recovery rate of the various institutions. This leads to the conclusions drawn from the results and the recommendations thereof. 92 University of Ghana http://ugspace.ug.edu.gh CHAPTER FIVE CONCLUSION AND RECOMMENDATIONS 5.1 Introduction This chapter gives a final out look or outcome of the study undertaken by the researcher in determinants of interest rate and recovery rate among Microfinance institutions in the Accra Metropolis. Conclusions were made on the objectives of the study, outcome of the analysis and also the hypothesis formulated by the researcher. 5.2 Summary in View In summary, research reports are not quick fix exercises as it demands time, energy and money and it’s expected that every researcher must have considered this precursors before embarking on research like this anywhere in the world especially when dealing with finding lasting solutions to everyday socio-economic, business and scientific issues among others in this dispensation and the future world. Many businesses are becoming futuristic today especially; automobile industry in UK and its allies, where they are looking at reducing emissions from vehicles or structuring towards producing zero emission cars, Telcos following suit by buying into the future driving subscribers along, lending or credit has also made navigations into future impacts and these are largely controlled by interest rates friendliness to ordinary borrowers to be able to pay liabilities due them and also improving their business and personal lives. For this course, the researcher took it upon to research into the impact of interest rates determinants on loan recovery of MFIs exploring from the Ghanaian economy perspective. In achieving this ultimate objective, the researcher took major steps in 93 University of Ghana http://ugspace.ug.edu.gh understanding to a larger extent the research objective or subject of study by specifically elaborating the background of the study which brought out the history to the research, the problem statement indicating the problems identified to be solved, the bases of the research through the justification, research questions and hypotheses, objectives of the study, significance of the study, scope of the study and the limitation of the study. As a canon, it was necessary that literature about the subject of research was explored extensively to enable readers or information users to understand the detail necessary if not sufficient of the subject matter and in doing so, the variables that held the research tight were well elaborated to the understanding of the average reader hence variables including liquidity risk, operational cost, base rate, credit risk and application of interest rate premiums aided in the study and helped gave findings either related or averse to most finding in the review. Even though some variables were used and dominating and other were not identified as variable within the literatures reviewed, these cases serve as bases for and gaps in which the research can be conducted. From the mode of estimation with the dominating variables in the question, the inferential statistics models of analysis and to some extent the descriptive statistics aided with the finding of the research gave life to the review. Now it was not enough for the research to end there; in the review stage as it does not give as any conclusion to what we are looking for, hence based on review works the researcher was able to understand trends, and what necessary approach was to be adopted in pursuing research like interest rate and loan recovery activities therefore in expressing the method use in this research, the research used reliable sampling techniques, sample size determination, validity analysis and research design in a way that best helped in the quest in achieving the research objective. In testing for validity of instrument engaged, the Cronbach’s alpha coefficient was a litmus paper deployed 94 University of Ghana http://ugspace.ug.edu.gh to ensure the researcher was getting it right the first time at all time. This stage gave rise to data collection for the purposes of grinding or analysing to make meaning into the variables of study that conveys the best message to the reader or information user. Data analysed were made to exhibit descriptive characters, and inferential characters by getting responses of personnel’s’ from microfinance institutions through descriptive and inferential statistics in forms of tables, graphs and charts. The results depicted if certain factors played a role in the loan recovery rate by Microfinance institutions and to what extent these factors affect the recovery rate of the various institutions. Now it is not always the case that a researcher might meet his/her objective should everything be properly done but in this study journey of the researcher through careful questioning and hypotheses has been able to accept to larger extent the hypotheses surrounding the research and the critical questions duly answered favourably which stands to reason that interest rate determinants have dire effect on loan recoveries and will continue to determine future lending space in either increasing MFIs asset performance or non- performance. 5.3 Conclusion The study determined the effect of base rate on loan recovery of MFIs and how is it calculated. Our descriptive analysis showed that, base rate aided recovery of debts and loans. This was measured by the minimum rate charged, how often minimum rate was reviewed, if interest rate quotes to customers aided in loan recovery and how it was calculated. Mark-up pricing was the used to calculate interest rate by most firms and these quotes aided in loan recovery of the firms. Minimum rate to be charged by most firms was between 16-20% and the rate was reviewed quarterly. The hypothesis test 95 University of Ghana http://ugspace.ug.edu.gh revealed that base rate affects loan recovery rate of MFIs therefore we failed to accept the null hypothesis. The researcher’s objective was duly achieved. Secondly, the study also ascertained how liquidity risk affects the interest rate of the MFIs. In achieving this objective, the researcher enquired the ways proportion and the extent of which liquidity risk could affect interest rate. It was found that liquidity plays a role in calculating interest rate and with this interest rate is affected by 1-10%. The quest to stay liquid increases the chase for interest rate. Firms are able to identify liquidity risk in four (4) weeks and averagely recover after a year but some firms are able to recover half year into detection. Averagely, firms experience high loan recovery in four-six months of each year and peek month of high recovery is July and peek month of low recovery is December. Different rate quotes are given to customers during distressed periods and after normalization, those who then receive credit facilities feel the coldness in the rate quoted. Using the pairwise correlation, the analysis also showed that liquidity has no relationship with interest rate of MFIs. The null hypothesis to that effect was accepted. Thirdly, the study evaluated the operational cost and its effects loan recovery rates and in using the purposive sampling technique in the collection of data it was ascertained personnel’s were majorly with either a first degree or HND qualification and majority of them had between less than a year and two years working experience. The outcome showed that, salaries, administrative cost, rent and utilities are most considered when MFIs were calculating operational cost. Operational cost covers 1-10% of interest rate and when its factored in the calculation it affects loan recovery about 1-10% and 11- 20%. The study therefore showed a strong positive correlation between operational cost in interest rate calculation affect recovery and extent to which loan recovery is affected and I failed to accept the null hypothesis. 96 University of Ghana http://ugspace.ug.edu.gh The study again investigate the credit risk faced by the MFIs and its impact on loan recovery. It was found that, all firms face credit risk and averagely credit risk covers 1- 10% of interest rate. Some customers have proven to be credit worthy hence such quality is factored into interest rate quotes to them. All firms have credit policies to guide them in credit administrations and loam disbursement and the ae personnel’s in charge of these policies to ensure they are strictly followed. Between 1-20% and 41- 60% is how averagely firms rate the contribution of credit policies to loan recovery annually. Even though non-compliance with credit risk result in ineffective administration process, pressure from customers and management as well as inexperience implementers accounts for non-compliance. Credit policies reflect macroeconomic factors especially lending interest rate, inflation and economic growth. Most firms review credit policies annually while some few others do that semi- annually. The hypothesis tested also proved that there is no relationship between credit risk and loan recovery of MFI’s hence the failure to accept the null hypotheses. Finally, the study investigated the ways MFIs set interest rate premiums to aid in loan recovery. Our finding showed that all the Microfinance institutions have a way of applying interest rate premiums for their customers. From the results, interest rate applied depends the credit risk of the customer and the approach adopted influences loan repayment yet it is not changed till the loan is paid. Personnel’s surprisingly are uncertain about the approach the result in high loan recovery. In calculating loan repayment schedule more firms use reducing balance method than the straight line method and this results in high loan recovery rate. In as much as loan repayment sits the customer, they are not considered when deciding on which method to use but are considered for their repayment schedule. Loan rescheduling is done during default but does not determine high loan recovery or not since its highly about the customer’s 97 University of Ghana http://ugspace.ug.edu.gh attitude. From the inferential analysis, the result shows a strong negative correlation between the approach influences loan repayment. and there is high loan recovery rate when reducing balance. The null hypotheses is accepted meaning loan recovery rate does not depend on how interest premiums are by MFIs. 5.4 Recommendations The recommendation as far as MFIs is concerned is to shape the credit industry in Ghana especially the Microfinance Apex body and other associate bodies in the industry to help achieve the paramount objective of poverty alleviation and not compromising it to huge losses as far as lending is concerned. The study invariably adds to knowledge and a threshold for further studies in the area of interest rate and recovery. The study also served as a source of information for the Central Bank of Ghana and the Ministry of Finance and Economic Planning to know current issues regarding non-performing loans and bad loans and how to help the institutions involved tackle the bottle-neck. The first recommendation was on base rate and recovery rate. I suggest that the various MFIs should take a critical look and consider what really matters in the calculation if interest rate like the loan duration, loan amount, risk on loan and the value of the security. Also, minimum rate should be reviewed periodically subject to economic conditions, slow loan recovery, on customer’s request and regulatory adjustments. Firms should consider calculating interest rate by mark-up when margin pricing does not aid in loan recovery. 98 University of Ghana http://ugspace.ug.edu.gh Secondly liquidity risk is should necessarily be identified early in the firms. This helps to reduce the extent of damage it may cost the firm hence reduce the period of recovery if needed. Also, Firms should consider reviewing interest rate of customers with existing loan facility after their distressed period. This will bring down the burden of continuously paying high interest rate which was quoted due to distress. Since giving out loan is the way of making profit even though it’s a diverted concept of the microfinance initiator (Grameen Bank), this will cause customers to have confidence to request for loan facilities since they know they will not be exploited after the period of distress. Again, the rate at which the quest to stay liquid affect interest rate should be at a reducing rate and have a laid down policy on how handle and recover from liquidity risk. Thirdly, firms should be able to review the composition of the calculation of their operational cost. This is because the essence and impact of a component could change over time and less or unnecessarily more attention will be given to it. I suggest firms should have a laid down policy on how operational cost will be calculated and periodic intervals of component review. Furthermore, I recommend that MFIs should put in place measures to ensure effective credit risk administration. The firms must make sure the loan facility is being used for the purpose it was sorted get a good or worthy collateral for the facility taken, provide flexible interest rate to beneficiaries, do proper due diligence before during and after disbursing the loan, there should be accurate documentation of all loan processes, critical periodic review of credit policies, efficient and well-resourced loan recovery department and effective data gathering systems. Finally, MFIs should work at identifying, knowing and understanding what works for loan recovery in their firms. Firms should identify per existing approaches and recovery 99 University of Ghana http://ugspace.ug.edu.gh rate what gives them high loan recovery and as well try new approaches and methods to know which one gives high loan recovery. Firms should also know and understand their customers and as well consider them in decision making regarding repayment and rescheduling. It is important to know that customer’s attitude can make or unmake your firm. 100 University of Ghana http://ugspace.ug.edu.gh REFERENCES (n.d.). Retrieved from cleancookstoves.org/resources_files/ghana-microfinance.docx Accra Metropolitan Assembly. (2019, February 19). News, AMA This Month: Accra Metropolitan Assembly. Retrieved March 4, 2019, from Accra Metropolitan Asembly Web Site: https://ama.gov.gh/news- details.php?n=OHFzNjU3cW80cDUycm9xNDJxNzk0OHA5OW5yMTZzMm 83cDMycnBycg== Accra Metropolitan Assembly. (n.d.). The Assembly, About The Assembly: Accra Metropolitan Assembly. Retrieved March 4, 2019, from Accra Metropolitan Assembly Web site: https://ama.gov.gh/theassembly.php Addae-Korankye, A. (2014). Causes and control of loan default/ delinquency in microfinanceinstitutions in Ghana. American International Journal of Contemporary Research, 4(12), 36-45. Agbemava, E., Nyarko, I. O., A. T., & Bediako, A. K. (2016). Logistic regression analysis of predictors of loan defaults by customers of non-traditional banks in Ghana. European Scientific Journal, 12(1), 172-185. Agu, O. C., & Okoli, B. C. (2013). Credit management and bad debt in Nigeria commercial banks– implication for development. . IOSR Journal of Humanities and Social Science (IOSR-JHSS), 12(3), 47-56. Ali, A., & Dipratn, K. (2015, July). The impact of interest rate ceiling on microfinance industry. Researc Gate. Retrieved from https://www.researchgate.net/publication/281689300_The_Impact_of_Interest _Rate_Ceilings_on_Microfinance_Industry 101 University of Ghana http://ugspace.ug.edu.gh Amonoo, E., Acquah, P. K., & Asmah, E. E. (2003). The impact of interest rates ondemand for credit and loan repayment by the poor and SMEs in Ghana. IFLIP (The Impact of Financial Sector Liberalization on the Poor) Research Paper 03-10. Arko, S. K. (2012). Determining the causes and impact of non-performing loans on the operations of microfinance institutions: A case of Sinapi Aba Trust. (Undergraduate dissertation, Institute of Distance Learning, Kwame Nkrumah University of Science and Technology. Retrieved from http://ir.knust.edu.gh/handle/123456789/4958 Bank of Ghana. (2014). Annual Report 2014. Accra: Bank of Ghana. Retrieved September 2018, from https://www.bog.gov.gh/privatecontent/Publications/Annual_Reports/2014_B ank%20of%20Ghana%20Annual%20Report.pdf Bank of Ghana. (2018). Bank of Ghana Monetary Committee Press Release. Accra. Retrieved from https://bog.gov.gh/privatecontent/MPC_Press_Releases/MPC%20Press%20Re lease%20-%20November%202018.pdf Bank of Ghana. (2018, August). Supervision: Register of Licensed Microfinance Institutions. Retrieved February 12, 2019, from Bank of Ghana Website: ttps://www.bog.gov.gh/privatecontent/Banking_Supervision/List%20of%20Li censed%20Microfinance%20Companies%20as%20at%20August%202018.pd f 102 University of Ghana http://ugspace.ug.edu.gh Bank of Ghana. (2007). Working paper- a note on microfinance in Ghana. Research department. Retrieved December 16, 2018, from http://www.bog.gov.gh/private ontent/.../working paper %20papers/microfinancing Bennardo, A., Pagano, N., & Piccolo, S. (2013). Multiple-Bank Lending, Creditor Rights and information sharing. . Centre for Studies and Economics and Finance: Working paper No.211. Retrieved from https://ideas.repec.org/p/cpr/ceprdp/7186.html Business Dictionary. (2019). Operating cost. Retrieved from Businessdictionary.com: http://www.businessdictionary.com/definition/operating-cost.html Carletti, E., Cerasi, V., & Daltung, S. (2007). Multiple-bank lending: Diversification and free riding in monitoring. Faure, A. (2013). Central Banking & Monetary Policy: An Introduction. bookboon.com. Retrieved December 2018, from http://thuvienso.bvu.edu.vn/handle/TVDHBRVT/15322 Fedaseyeu, V., & Hunt, V. (2014). The economics of debt collection: Enforcement of consumer credit contracts. . Working papers department research. Retrieved from http://www.philadelphiafed.org/research-and-data, ten independence mall, Philadelphia PA 19106-1574 Fernando, N. A. (2008, July). Managing Microfinance Risks: Some Observations and Suggestions. Asian Journal of Agriculture and Development, 4(2). Retrieved from https://www.microfinancegateway.org/sites/default/files/mfg-en-paper- 103 University of Ghana http://ugspace.ug.edu.gh managing-microfinance-risks-some-observations-and-suggestions-jul- 2008.pdf Fisher, I. (1933). The Debt-Deflation Theory of Great Depression. Retrieved from https://fraser.stlouisfed.org/files/docs/meltzer/fisdeb33.pdf Gatimu, E. M., & Mukoma, F. (2014). Assessing institutional factors contributing to loan defaulting in microfinance institutions in Kenya. Journal of Humanities and Social Science, 12(5), 105-123. Ghatak, M. (1999). Group lending, local information and peer selection. Journal of Development economics, 60(1), 27. Glenn, D. I. (1992). Determining Sample Size. Retrieved January 15, 2019, from http://sociology.soc.uoc.gr/socmedia/papageo/metaptyxiakoi/sample_size/sam plesize1.pdf Grameen Bank. (2014). Historical Data Series in USD 1976-2009. Impact of Microfinance: Final Report. Bangladesh: Grameen Bank Communications. Retrieved from arizonajournal.org/wp-content/uploads/2015/10/Greenberg_8- 6 Gyamfi, G. D. (2012). Assessing the effectiveness of credit risk management techniques of microfinance firms in Acccra. Journal of Science and Technology, 32(1), 96-103. Jarko, F., Pave, ,. C., d’Artis, K., & Pokrivcak, J. (2013). Credit constraints: Heterogeneous firms and loan defaults. Annals of economics and finance, 14(1), 53–68. 104 University of Ghana http://ugspace.ug.edu.gh Kamau, G., & Wagoki, J. (2014). Effect of strategic practices on bank loan recovery by commercial banks in Kenya: A case of Family Bank Limited. International Journal of Science and Research, 3(11), 242-248. Kebewar, M. (2012). The effect of debt on corporate profitability Evidence from French service sector. d'Economie et de Gestion, Rue de Blois. Retrieved from http://ssrn.com/abstract=2191075 Keynes, J. M. (1936). The General Theory of Employment, Interest and Money. London: Palgrave Macmillan. Retrieved from http://www.hetwebsite.net/het/texts/keynes/gt/gtcont.htm Khieu, H., & Mullieaux, D. (2009). The determinants of bank loan recovery rates. Journal of Banking & Finance. doi:doi:10.1016/j.jbankfin.2011.10.005 Kimando, L. N., Kihoro, J. M., & Njogu, G. W. (2012). Factors influencing the sustainability of microfinance institutions in Murang'a Municipality. International Journal of Business and Commerce, 1(10), 21-45. Krejcie, R. V., & Morgan, D. W. (1970, September 1). Determining Sample Size for Research Activities. Sage Journals: Educational and Psychological Measurement. doi:https://doi.org/10.1177/001316447003000308 Kunateh, M. A. (2016, March 11). News page: Bad debts kill microfinance sector in Ghana. Retrieved from GhanaDot.com: http://www.ghanadot.com/news.ghanadot.kunate.microfinance.062611.html Lascelles, D. (2012, January 1). Microfinance — A Risky Business: A Time for Strong Leadership. Center for Financial Inclusion at ACCION, 17(1). Retrieved from https://www.centerforfinancialinclusion.org/microfinance-a-risky-business 105 University of Ghana http://ugspace.ug.edu.gh Ledgerwood, J. (1999). Microfinance Handbook: An Institutional and Financial Perspective (Sustainable Banking With the Poor). Washington D.C, U.S.A: The World Bank Group. Retrieved August 2018, from https://openknowledge.worldbank.org/handle/10986/12383 McConnell, R., & Brue, S. P. (2008). Macroeconomic: Introduction to economic growth an instability. McGraw-Hill/Irwin. Moti, H., Masinde, J., Mugenda, G., & Sindani, M. (2012). Effectiveness of creditmanagement system on loan performance: Empirical evidence from microfinance sector in Kenya. International Journal of Business, Humanities and Technology, 2(6), 99-106. Munene, H. N., & Guyo, S. H. (2013). Factors influencing loan repayment default in micro- finance institutions: The experience of Imenti north district, Kenya. International Journal of Applied Science and Technology, 3(3), 80-84. Nduta, R. G. (2013). The effect of credit management on the financial performance of microfinance institutions in Kenya. Masters dissertation, ,University of Nairobi, Kenya. Retrieved from chss.uonbi.ac.ke/sites/default/files/chss Njuguna, J., Gakure, R., Waititu , A., & Katuse , P. (2013). Effects of financial risk management on the growth of microfinance sector in Kenya. Prime Journal of Business Administration and Management (BAM), 3(6), 1. Norell, D. (2010). How to reduce arrears in microfinance institutions. Journal of Finance, 3(1), 116-129. 106 University of Ghana http://ugspace.ug.edu.gh Ogolla, A. B. (2012). Debt recovery as an operational strategy used by NIC Bank to manage non-performing loans portfolio. Retrieved from erepository.uonbi.ac.ke/handle/11295/12565 Ongena, S. S. (2000). Banking Relationships: The performance of financial institutions. Cambridge, UK: Cambridge University Press. Otero, M. (1999). Bringing development back into microfinance. Journal of Microfinance/ ESR Review, 1(1). Retrieved September 19, 2018, from https://scholarsarchive.byu.edu/esr/vol1/iss1/2 Oxford Business Group. (2019). Retrieved from Oxford Business Group.com: https://oxfordbusinessgroup.com/analysis/starting-small-authorities-enforce- higher-standards-make-microfinance-segment-more-reliable Oxford Business Group. (2019). The Report: Myanmar 2019. Retrieved from Oxfordbusinessgroup.com: https://oxfordbusinessgroup.com/myanmar-2019 Peprah, J. A. (2012). Good Money’ chasing ‘Bad Money’ implications for MFIs management and governance in Ghana. Asian Economic and Financial Review, 2(3), pp. 503-517. Petersen, M. A., & Rajan, R. G. (1994). The benefit of lending relationships: Evidence from small business data. Journal of Finance, 49, 1367-1400. Quarmyne, R. C. (2014). Macroeconomic instability and banks’ lending behaviour in Ghana. European Scientific Journal, 10(10), 397-412. Quartey, E. (2011). The effect of interest rate on loan repayment in microfinance institutions- A case study of Tanoah Capital Point Ltd. 107 University of Ghana http://ugspace.ug.edu.gh Schicks, J. (2011). Over-indebtedness of micro borrowers in Ghana an empirical study from a customer protection perspective. Centre for Financial Inclusion. Retrieved January 5, 2019, from https://dipot.ulb.ac.be/dspace/bitstream/.../068fce3f-d18e-479 Schreiner, M., & Colombet, H. H. (2001, September). From Urban to Rural: Lessons for Microfinance from Argentina. Development Policy Review, 19(3), 339-354. doi:https://doi.org/10.1111/1467-7679.00138 Singh, A., & Masuku, M. B. (2014). Sampling techniques & determination of sample size in applied statistics research: An overview. International Journal of Economics, Commerce and Management, 2(11), 1-22. Wikepedia. (2009). History of microfinance. Retrieved from Wikepedia: ttps://en.wikipedia.org/wiki/Microfinance#History_of_microfinance Yamane, T. (1973). Statistics and Introduction Analysis. 3. Retrieved from https://books.google.com.gh/books/about/Statistics_an_introductory_analysis. html?id=dX-HBsrLHMIC&redir_esc=y Yunus, M. (1967). Banker to the Poor. Unites States: PublicAffairs. Yunus, M. (2008, March 1). Turning Beggars Into Entrepreneurs. New Perspectives Quarterly, 25(2). doi:10.1111/j.1540-5842.2008.00990.x 108 University of Ghana http://ugspace.ug.edu.gh APPENDIX I- DATA SET QUESTIONNAIRE UNIVERSITY OF GHANA SCHOOL OF BUSINESS MASTERS OF BUSINESS ADMINISTRATION RESEARCH TOPIC: DETERMINANTS OF INTEREST RATES AND LOAN RECOVERY RATE OF MICROFINANCE INSTITUTIONS IN GHANA: A CASE STUDY OF THE ACCRA METROPOLIS Dear Respondent, This study is aimed at enquiring into whether determinants of interest rate affects loan recovery of Microfinance institutions in Ghana. To help attain this objective, we humbly request you to provide relevant and objective responses to the items on this questionnaire. It will take approximately 20 minutes to be completed. Note that your answers would be treated in strict confidence and used for academic purpose only. Thank you SECTION A - PERSONAL INFORMATION 1. Educational background a. Basic education b. Secondary/Technical c. First degree/HND d. Post graduate 2. How many years of work experience do you have in the firm? a. <1-2yrs b. 3-5yrs c. 6-8yrs d. 9-11yrs 109 University of Ghana http://ugspace.ug.edu.gh SECTION B –MINIMUM RATE AND LOAN RECOVERY RATE (We seek to understand how the minimum rate affects loan recovery rate and its calculations). 3. What factors are considered in calculating interest rate for customer except the base rate? 1. …………………………………................................... 2. …………………………………………………..…….. 3. ………………………………………………………… 4. What is the minimum rate your firm will charge on a loan facility? a. 16-20% b. 21-25% c. 26-30% d. 30- 35% e. 36-40% 5. How often do you review minimum rate? a. Monthly b. Quarterly c. Semi-annually d. Annually e. others please specify………………………………………. 6. What causes a review to be done in your firm? 1. ……………………………………………………………………………. 2. ……………………………………………………………………………. 3. ……………………………………………………………………………. 4. ……………………………………………………………………………. 5. ……………………………………………………………………………. 7. Does the interest rate quotes to customer’s aid in loan recovery? a. Yes b. No. 8. How is the interest rate calculation done? a. Mark up b. Margin c. others please specify………………………………………. 110 University of Ghana http://ugspace.ug.edu.gh SECTION C – HOW LIQUIDITY RISK AFFECTS THE INTEREST RATE (This section seeks to understand how liquidity is measured and the effect it has on interest rate). 9. Does liquidity risk play a role when calculating interest? a. Yes b. No. 10. How does the quest to stay liquid affect the chase of interest rate? Tick as applicable a. Increase b. Decrease 11. To what extent is interest rate affected? Rate in percentage. a. 1-10% b. 11-20% c. 21-30% d. 31-40% f f r r f f e. 41-50% f. 51%+ o o r r 12. How early is liquidity risk identified in your firm? m m a. 1 week o b. 2weeks o c. 3weeks d. 4weeks t t e. others please spmec ify……………m… ……….. 13. How many months in ta year do you experience high loan reco hvery rate in your h t firm? e e h h a. One – three months b. Four - six months l l e e i i c. Seven – nine mlonths l d. Ten – Twelve months s 14. What is the average peiak month of highi recovery in your firm s? t t a…………………s………… s e e 15. What is the average petak month of low trecovery in your firm? d d e a…………………………… e m m d d 16. How long does it take your firm to recover averagely from liquidity risk? i i m m a. Less than a month b. Within the first quarter after detection c c i i c. Half year into detection d. After a year of dertection r c c o 17. Is there a different rater quoted to custom o rers distressed period and when normalized? f f o o a. Yes b. No i i f f f f n n r i 11i1 r a a o n n o n n m a a m University of Ghana http://ugspace.ug.edu.gh 18. Which group of customers get the effect of normalization? a. Those with existing loan facility f b. Those who are given credit after norrmalization f o r m o t m SECTION D- OPERATIONAL COST AND ITS EFFECT ON LOAN RECOVERY h t (This section seeks to find out operational cost effect on loans and what can be done to e enhance recovery) h l e 19. What is the composition of operational cosits in your firm? (Tick as many as l apply) s i a. Administrative t b. Salaries s e c. Legal fees t d. Rent d e e. Office supplies m d f. Directors fees i m g. Utilities c h. Advertisement i i. Insurance premium r c o r 20. Does operational cost affect interest rate? f o a. Yes b. No f f i fr r 21. What proportion of inoterest rate does it covner? i m o a. 1-10% b. 11-20% a c. 21-30%n f d. 31-40%t f h m n a r r e t e. 41-50% l f. 51%+ cf n o o i h s e r c m m t e i e o e t t d l n m i h hm i 112 i s t n e e c s r th s l l o t i University of Ghana http://ugspace.ug.edu.gh 22. In your opinion, do you think operational cost factored in the interest rate calculation have an effect on Loan recovery? a. Yes f b. No f 23. To what extent does operational risk affect loan recovery? r r a. 1-10% b. 11-20% c. 21-30% d. 31-40% f f o o r r e. 41-50%m f f. 51%+ m f o o t t 24. How do your firmr calculate operationral cost? m m ……………………h … ho ……………………o ……………………………………….. t t e e ………………………m ……………………m ……………………………………….. h h l l ………………………t ……………………t ……………………………………….. e e i i ………………………h ……………………h ………………………………………… l l s s e e i i t t l l s s e e i i t t SECTION E – CdR EDIT RISK FACED BY Md FI’s AND ITS IMPACT ON s s RECOVERY RATE. e e m m t t d d (This section seekis to investigate how credit risi k impact on loan recovery rate as far as e e microfinancing is concerned). m m c c d d 25. Do you face credit risk in your firm? i i r r a. Yes m b. No m c c o o 26. If “yes”, what proi portion of interest riate quotation does it carry? r r a.1-10% f b. 11-20% f c. 21-30% d. 31-40% c c f f o o i i r r r r f f e. 41-50%n f. 50%+ n o of o o i i a a 27. Is the interest ratef of a customer whof rdoes not possess credit rmis k affected? m n n n n i io t t a. Yes b. No a a c c 28. How is it affectedn and recalculated? nm h h n n 1…………e …a…………………… e at……………………. e e c c 2…………i ……………………………i………………. n nh l l e e n c ce1 13 n i i i i s s e el s s n n t t i ii t t University of Ghana http://ugspace.ug.edu.gh 3………………………………………………………. 29. Does your firm have credit policies governing disbursement and loan recovery? a. Yes b. No 30. If “yes” are personnel or committee in charge of this policies? a. Yes b. No 31. Can you rate to what extent the credit policies used by the firm averagely contribute to recovery of loans annually? a. 1-20% b. 21-40% c. 41-60% d. 61- 80% e. 81-100% 32. Do you think non-compliance with credit policies accounts for ineffective credit administration? a. Yes b. No. 33. If “yes”, which of the following accounts for that? a. Customer pressure b. Management pressure c. Inexperience of implementers d. All of the above d. Others please specify…………………………………………. 34. Do the credit policies reflect macroeconomic factors? a. Yes [ ] b. No. [ ] 35. If “yes”, which of the following is considered? a. Economic growth b. Inflation c. Lending interest rate d. GDP growth e. Unemployment 36. How often do these credit policies undergo review? a. Monthly b. Quarterly c. Semi-annually d. Annually e. Others please specify………………………………………. 114 University of Ghana http://ugspace.ug.edu.gh 37. What measures should credit firms put in place to ensure effective credit administration? 1. …………………………………................................... 2. …………………………………………………..…….. 3. ………………………………………………………… SECTION F – VERIFY WAYS INTEREST RATE PREMIUM ARE SET TO AID IN LOAN RECOVERY. (Finally, this section seeks to verify ways to apply interest rate premiums to help in the recovery of debt as far as microfinancing is concerned). 38. Please rate to what extent you agree or disagree with each of the following statements. Please circle one response for each statement. Strongly Strongly Disagree Uncertain Agree Disagree Agree a. Interest rate premiums applied 1 2 3 4 5 depends on credit risk of the customer. b. Interest rate applied is same for all 1 2 3 4 5 customers. c. An approach adopted is not changed till the 1 2 3 4 5 loan is paid. d. The approach influences loan 1 2 3 4 5 repayment. e. Fixed or flat interest rate approach results 1 2 3 4 5 in high loan recovery rate. f. Floating rate 1 2 3 4 5 approach results in 115 University of Ghana http://ugspace.ug.edu.gh high loan recovery rate. g. Reducing balance method is used for the calculation of 1 2 3 4 5 loan repayment schedule. h. Straight line method is used for the 1 2 3 4 5 calculation of loan repayment schedule. i. There is high loan recovery rate when 1 2 3 4 5 reducing balance is used. j. Loan recovery is high when straight 1 2 3 4 5 line method is used. k. Customers opinion is considered when 1 2 3 4 5 deciding on which method to use. l. Loan repayment schedule is made in 1 2 3 4 5 consultation with the customer. m Loan repayments are . rescheduled when 1 2 3 4 5 there is default. n. Customers pay back loans when 1 2 3 4 5 repayment is rescheduled. o. Customers attitude affect loan 1 2 3 4 5 repayment. Thank you for your co-operation God bless you. 116