SU+ @ Strathmore 

University Library 
 

  
 

Electronic Theses and Dissertations 

 

 

This work is availed for free and open access by Strathmore University Library.  

It has been accepted for digital distribution by an authorized administrator of SU+ @Strathmore University. 

For more information, please contact library@strathmore.edu 

 

 
2024 

 

Analysis of the drivers of financial 

performance of development financial 

institutions in Kenya. 

 
 

Katsenga, Ruth Mose 
Strathmore Business School  
Strathmore University 
 
 
 
 

 
 

 

Recommended Citation 

Katsenga, R. M. (2024). Analysis of the drivers of financial performance of development financial institutions in 

Kenya [Strathmore University]. http://hdl.handle.net/11071/15587 

 
 
 
 
 

Follow this and additional works at: http://hdl.handle.net/11071/15587 

https://su-plus.strathmore.edu/
https://su-plus.strathmore.edu/
http://hdl.handle.net/11071/2474
mailto:library@strathmore.edu
http://hdl.handle.net/11071/15587
http://hdl.handle.net/11071/15587


 
 

ANALYSIS OF THE DRIVERS OF FINANCIAL PERFORMANCE OF 

DEVELOPMENT FINANCIAL INSTITUTIONS IN KENYA 

 

 

 

                                                      RUTH MOSE KATSENGA 

                                                       REG. No: MDF/112758/18 

 

 

A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE FOR THE AWARD 

OF DEGREE OF MASTER OF SCIENCE IN DEVELOPMENT FINANCE OF 

STRATHMORE UNIVERSITY 

 

 

 

 

                                                        2024 

 

 

 

 

 



  

ii 
 

DECLARATION 

I declare that this thesis is my original work and has not been presented to any other university for 

the award of a degree. Any work done by other people has been duly acknowledged.  To the best 

of my knowledge and belief, the thesis contains no material previously published or written by 

another person.  

 

 

Ruth Mose Katsenga 

 

 2024 

 

APPROVAL 

The thesis of Ruth Mose Katsenga was reviewed and approved by the following: 

 

Dr. Freshia Mugo Waweru,  

PhD Academic Director & Senior Lecturer, (Finance, Derivatives, Leadership and Governance) 

Strathmore University Business School (SBS), 

Strathmore University 

 

 



  

iii 
 

ACKNOWLEDGMENTS 

My Outmost gratitude goes to our Heavenly Father for granting me the wisdom, grace and 

knowledge to undertake this study. My sincere appreciation to my supervisor, Dr. Freshia Waweru 

Mugo who has been instrumental in providing extreme guidance and support in   conducting this 

study and writing this thesis. I also wish to sincerely thank my parents Mr. and Mrs. Katsenga for 

their continuous support and prayers which encouraged me to complete this thesis.  

 

 

 

 

 

 

 

 

 

 

 

 

 

 



  

iv 
 

TABLE OF CONTENTS 

DECLARATION .......................................................................................................................... ii 

ACKNOWLEDGMENTS ........................................................................................................... iii 

TABLE OF CONTENTS ............................................................................................................ iv 

LIST OF TABLES ...................................................................................................................... vii 

LIST OF FIGURES ................................................................................................................... viii 

ABBREVIATIONS AND ACRONYMS .................................................................................... ix 

OPERATIONAL DEFINITION OF KEY TERMS .................................................................. x 

ABSTRACT .................................................................................................................................. xi 

CHAPTER ONE ........................................................................................................................... 1 

INTRODUCTION ........................................................................................................................ 1 

1.1 Background of the Study .................................................................................................. 1 

1.1.1 Drivers of Financial Performance ............................................................................. 5 

1.1.2 Financial Performance .............................................................................................. 7 

1.1.3 Development Finance Institutions in Kenya ............................................................. 9 

1.2 Statement of the Problem ............................................................................................... 10 

1.3  Objectives of the Study ................................................................................................ 112 

1.3.1 General Objective .............................................................................................................. 12 

1.3.2 Specific Objectives of the Study ...................................................................................... 112 

1.4 Research Questions ...................................................................................................... 112 

1.5 Significance of the Study ............................................................................................... 13 

1.6 Scope of the Study.......................................................................................................... 13 

CHAPTER TWO ........................................................................................................................ 15 

LITERATURE REVIEW .......................................................................................................... 15 

2.1 Introduction .................................................................................................................... 15 

2.2 Theoretical Review ........................................................................................................ 15 

2.2.1 Liquidity Preference Theory ............................................................................................ 15 

2.2.2 CAMEL Model .............................................................................................................. 17 

2.3 Empirical Review ........................................................................................................... 18 

2.3.1 Effect of Asset Quality on Financial Performance ................................................. 18 

2.3.2 Effect of Management Efficiency on Financial Performance................................. 20 

2.3.4 Effect of Liquidity Management on Financial Performance .................................. 21 



  

v 
 

2.4 Summary of Research Gaps ........................................................................................... 23 

2.5 Conceptual Framework .................................................................................................. 26 

2.8 Operationalization of Variables ..................................................................................... 28 

2.9 Summary of Literature Review and Research Gaps ...................................................... 28 

CHAPTER THREE .................................................................................................................... 30 

RESEARCH METHODOLOGY .............................................................................................. 30 

3.1 Introduction .................................................................................................................... 30 

3.2 Research Philosophy ...................................................................................................... 30 

3.3 Research Design ............................................................................................................. 30 

3.4 Population and Sampling ............................................................................................... 31 

3.5 Data Collection Techniques ........................................................................................... 31 

3.5.1 Validity of Data....................................................................................................... 32 

3.5.2 Reliability of Data ................................................................................................... 32 

3.6 Research Quality ................................................................................................................ 32 

3.7 Data Analysis ................................................................................................................. 35 

3.8 Ethical Considerations.................................................................................................... 36 

CHAPTER FOUR ....................................................................................................................... 37 

PRESENTATION OF RESEARCH FINDINGS..................................................................... 37 

4.1 Introduction ......................................................................................................................... 37 

4.2 Discriptive Statistics ............................................................................................................ 38 

4.3 Diagnostic Tests .................................................................................................................. 38 

4.3.1 Stationarity Test ............................................................................................................ 38 

4.3.2 Normality Test .............................................................................................................. 39 

4.3.3 Multicollinearity Test ................................................................................................... 40 

4.3.4 Homoscedasticity Test .................................................................................................. 41 

4.3.5 Autocorrelation Test ..................................................................................................... 41 

4.3.6 Hausman Specification Test ......................................................................................... 42 

4.4 Interpretation of Regression Results on Financial Performance ......................................... 43 

4.5 Findings of the analysis of the drivers of financial Performance ........................................ 44 

4.5.1: Effect of Asset Quality on Financial Performance Development Finance Institutions 

in Kenya………………. .......................................................................................................... 45 

4.5.2: Effect of Management Efficiency on Financial Performance Development Finance 

Institutions in Kenya…… ........................................................................................................ 46 



  

vi 
 

4.5.3: Effect of Liquidity Management on Financial Performance Development Finance 

Institutions in Kenya…… ........................................................................................................ 47 

CHAPTER FIVE ........................................................................................................................ 48 

PDISCUSSIONS, CONCLUSIONS AND RECOMMENDATIONS .................................... 48 

5.1 Introduction ......................................................................................................................... 49 

5.2 Discussion of Findings ........................................................................................................ 49 

5.3 Conclusion ........................................................................................................................... 51 

5.4 Policy Recommendations .................................................................................................... 52 

5.5 Contribution to Knowledge ................................................................................................. 52 

5.6 Suggestion for Further Research ......................................................................................... 53 

5.7 Limitations of the Study ...................................................................................................... 53 

REFERENCES ............................................................................................................................ 55 

APPENDICES ............................................................................................................................. 62 

Appendix I: Data Collection Sheet ............................................................................................... 62 

Appendix I: List of Kenyan Development Finance Institutions ................................................... 62 

 

 

 

 

 

 

 

 

 

 

 

 



  

vii 
 

LIST OF TABLES 

Table 2.1: Summary of Research Gaps ........................................................................................ 23 

Table 2.2: Operationalization of Variables ................................................................................... 28 

Table 4.1: Descriptive Statistics ................................................................................................... 37 

Table 4.2: Fisher-type unit root test .............................................................................................. 39 

Table 4.3: Shapiro-Wilk test for Normality .................................................................................. 39 

Table 4.4: Test for Multicollinearity ............................................................................................. 41 

Table 4.5: Breusch-Pagan/Cook-Weisberg test for heteroskedasticity ........................................ 42 

Table 4.6: Autocorrelation test results .......................................................................................... 42 

Table 4.7: Hausman Specification Results ................................................................................... 43 

Table 4.8: Regression Results ....................................................................................................... 44 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



  

viii 
 

LIST OF FIGURES 

Figure 2.1: Conceptual Framework .............................................................................................. 27 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



  

ix 
 

ABBREVIATIONS AND ACRONYMS 

CAMEL – Capital adequacy, Asset quality, Management Efficiency, Earnings and Liquidity 

DFI – Development Finance Institution 

TFC–Tourism Finance Corporation 

AFC–Agricultural Finance Corporation 

ICDC–Industrial Commercial and Development Corporation 

IDB - Industrial Development Bank 

KIE-   Kenya Industrial Estates 

NPL – Non-performing loans 

ROA- Return on Assets 

ROE- Return on Equity 

ROI- Return on Investments 



  

x 
 

OPERATIONAL DEFINITION OF KEY TERMS 

Asset Quality:  A measure on the combination of non-performing loans, disbursed 

loans as well as provision for losses. 

Development Finance Institutions:  These are specialized financial institutions or organizations 

that provide long-term capital, technical assistance, and other forms of 

support to promote economic development and sustainable growth in 

Kenya. 

Financial Performance:  A measure on the use of assets and equity by an organization to 

generate revenues. 

Interest rates: A key monetary policy used by Central banks around the world to 

achieve macroeconomic stability in the economy. They represent the 

economic cost of lending money within commercial banks 

Liquidity Management:  The extent to which organizations cater for their financial obligations 

as the come due in the short run 

Management Efficiency  The capability of the management to deploy its resources efficiently, 

income maximization, reducing operating costs can be measured by 

financial ratios. One of these ratios used to measure management 

efficiency is operating profit to income ratio 

Return on Assets: An accounting ratio that shows the average return on total assets 

employed. 

Return on Equity: An accounting ratio that shows the average return on total investments 

of the shareholders in investments of the shareholders in the form of 

equity 

 

 

 



  

xi 
 

ABSTRACT 

The Development Financial Institutions are a critical nerve Centre to the economic growth of any 

country. The financial performance of Development Finance Institutions in Kenya over the last 

twenty years has not been performing according to the stakeholder expectations. DFI’s in Kenya 

had failed to provide a sustainable long-term finance to the industrial sector and the agricultural 

sector. This was evidenced by credit being allocated on the basis of political and social concerns, 

lack of effective and efficient incentives to collect. Studies on these development institutions have 

remained scanty with those that have attempted having varying outcomes thus making it difficult 

to provide a guide to policy formulation in Kenya.The purpose of the study was to analyse the 

drivers of financial performance of Development Financial Institutions (DFIs) in Kenya. The 

driver of financial performance considered in the investigation included asset quality, management 

efficiency and liquidity management in Kenya. The survey made use of census approach to arrive 

at five Development Finance Institutions employed in the investigation. Relying on information 

of the financial audited reports of these institutions, the data was retrieve spanning over the period 

2012/2013 to 2019/2020. Laying the theoretical foundation for the study was the theoretical 

postulations of the CAMEL model and the Liquidity Preference Theory. The outcomes of the 

investigations were reached owing to the credit accorded to the descriptive and regression 

techniques with the outcomes presented in tables. The outcome uncovered that asset quality is a 

significant and negative driver of Development Finance Institutions’ financial performance; 

management efficiency was unfolded as a positively and significant driver of Kenyan 

Development Finance Institutions financial performance; while liquidity management was 

reported to be a significantly positive driver of Development Finance Institutions financial 

performance in Kenya. Relating to the outcomes, the investigation recommended that the 

management of Development Financial Institutions should strengthen the means through which 

non-performing loans could reduce to boost the financial performance of the institutions. This can 

be done through critical assessment of customers’ credit worthiness to reduce the amount of loans 

that are non-performing in Kenya.



 
 

CHAPTER ONE 

INTRODUCTION 

1.1 Background of the Study 

Development finance institutions (DFIs) have acted as the major mechanism by which 

governments and donors implement economic development strategies. DFIs function as a means 

of directing funds towards priority sectors, frequently with preferential terms, aiming to achieve 

social and economic objectives not fully addressed by market forces alone. Their core mandate is 

to invest in domains where gaps or inefficiencies in the market and institutions leave needs unmet, 

lessening reliance on other financial players (Chin & Gallagher, 2019; Babones et al., 2020; 

Squires et al., 2016; Humphrey & Michaelowa, 2019). By directing capital towards underserved 

areas and populations, DFIs play an important countercyclical and developmental equalizing role 

that supplements—but does not substitute—the private financial system. Their interventions are 

intended to have demonstrative effects that stimulate broader investment and growth over the long 

term. 

DFIs invest in private sector companies and funds with the expectation of earning a financial 

return, as profitability allows DFIs to self-sustain and expand their development work. Likewise, 

DFI shareholders anticipate contributions toward socioeconomic goals like the UN SDGs. Ideally, 

DFIs direct capital toward ventures with highest development upside, dependent on viable private 

opportunities and DFI resource availability (Attridge et al., 2019). 

DFIs thus pursue a double bottom line of profitability and promoting host nation progress. They 

must generate earnings through prudent investing while enabling economic advancement. 

Balancing social and financial returns can be challenging, especially with limitations measuring 

social impacts. DFIs aim to balance these dual directives in optimizing developmental effects 

within constraints of commercial viability. Their dual mandate necessitates strategic compromises 

in balancing sometimes-competing objectives to best maximize long-term development impact 

through sustainable private sector engagement and self-financing. 

 

 

 

 

 



  

2 
 

Development Finance Institutions (DFIs) play a crucial role in fostering economic growth and 

development globally. Owned and controlled by governments, DFIs possess unique attributes such 

as investment opportunities identification, risks assessment, and provide the necessary financial 

and technical support that enable them to serve as catalysts for economic progress (Chin & 

Gallagher, 2019).  These unique attributes as argued by Mishra (2023), collectively enable DFIs 

to bridge the financing gap, mobilize resources, and contribute to sustainable development in 

developing countries. Their ability to provide risk mitigation allows the private sector to engage 

in projects that they might otherwise abandon (Mutunga, 2018).  

DFIs specialize in offering long-term loans and have the capacity to restructure loans to facilitate 

easier repayments in case of defaults (Adesoye, Bolaji, Atanda, & Maliq, 2018). They are often 

endowed with ample capital by governments, thereby affording them the privilege of accessing 

funds designated for technical assistance. Consequently, there is an expectation that they will pass 

on subsidies to beneficiaries in various forms, including extended repayment periods and below-

prime interest rates (Nthiga, 2017). 

Over the past decade, DFIs have gained prominence, providing financial support to both the public 

and private sectors (Bortes, Sinha & Grettve, 2011; Clark, Reed, & Sunderland, 2018). Private 

sector support from DFIs has increased substantially, reaching $33 billion in 2019, representing 

more than a doubling in annual commitments over six years (World Bank, 2020). In 2020, DFI 

support amounted to a quarter of official development assistance (ODA), though it is not typically 

counted as ODA. On average, DFIs contribute between 2% and 12% of total domestic investment 

for the period for which data are available (World Bank, 2021). 

Globally, the asset quality management is a concern because DFI credit does not only affect the 

stability of the DFIs themselves but also on economic stability (Sanyal, Pinchot, Prins, & Visco, 

2016). The significant role of DFIs in developing countries has been a phenomenon that continues 

to be discussed on their role in economic development of a nation by both policy makers and 

academicians (Chin & Gallagher, 2019; Attridge, et al., 2019; Shaikh, Ismail, & Shafiai, 2017; 

Squires, et al., 2016). This continued debates on the role of DFIs is attributed to the critical role 

played in poverty reduction, job creation and infrastructural developemnt of develping 

countries.Aligning the aforementioned view, Chin and Gallagher (2019) observed that the DFIs 



  

3 
 

are sources of capital at a lower cost and helps reduce the dependence on other financial 

institutions. This reduces the risk borne by the private sector in the event of a financial crisis 

(Mutunga, 2018). 

 

In the African context, the lending activities of DFIs are confronted with risks that are primarily 

influenced by the quality of assets (Babones, Åberg, & Hodzi, 2020). This implies that the potential 

for financial losses or non-performing loans is closely tied to the performance and value of the 

assets financed by DFIs in the region. In East Africa, the problem of DFIs asset quality has put 

serious adverse effects on the economy; for example, the Kenyan government has implemented 

various policy measures for management of asset quality and securing confidence in the DFIs 

financial system. These measures include prudential regulations, credit risk assessment, loan 

recovery mechanism, regulatory supervision, risk management framework and corporate 

governance standards (Kipchoge, 2022). In recent times, non-performing loans (NPLs) effect on 

asset quality has been trending and becoming a cause of concern for Development Financial 

Institutions (DFIs) stability in the face of reeling economic downturn (Adesoye, Bolaji, Atanda, & 

maliq, 2018). The effect of default of due debts on DFIs  profitability can be identified with a 

possible DFIs failure, barrier to further lending, reduction in profit level and negative economic 

growth in the society. 

Kenya has outlined its long-term development goals in Vision 2030, which aims to transform the 

country into a middle-income industrialized economy providing high living standards by 2030. 

Regarding the financial sector, the vision is to develop a vibrant and globally competitive industry 

that promotes high savings levels and finances Kenya's investment needs. Achieving this requires 

coordinated efforts from all financial players, including key contributors like Development 

Finance Institutions (DFIs). DFIs are important members of the financial sector that provide long-

term financing and development funding to small and medium enterprises (SMEs) (Lukilah & 

Wekesa, 2021). They also help develop the sector by providing access to long-term credits. 

Initiatives enhancing credit information sharing will further deepen and mature the industry, 

according to (Marbuah et al. 2022). In summary, DFIs are identified as important cogs in realizing 

Kenya's Vision 2030 goals for a robust financial system through facilitating investment and 

business growth. 



  

4 
 

DFIs were established in Kenya to facilitate long term financing for prioritized industries 

supporting industrialization (Nthiga, 2017). Unlike commercial banks which require full collateral, 

DFIs progressively enhance security as projects progress and offer additional benefits such as 

appraisals, monitoring, advisory services (Namusonge, 2004). Despite existing since the 

1960s/70s, poor asset quality management has led most DFIs collapsing and a development 

financing gap remains, raising concern about private sector expansion without suitable long term 

investment funds (Namusonge, 2004; Githua, 2015). Specifically, Githua (2015) identified asset 

quality as a core constraint causing DFIs to significantly underachieve in catalyzing development 

roles. While intended to plug financing gaps, weaknesses in managing loan portfolios seriously 

impeded DFI effectiveness over decades (Nthiga, 2017; Namusonge, 2004; Githua, 2015). This 

highlights asset quality as a persistent problem limiting their developmental impact, underscoring 

need for the current research 

Strong and stable DFIs are important for a country's economic progress, as the performance and 

development of DFIs are closely interconnected with a nation's advancement. DFIs play an 

invaluable role in financial services by channeling funds into productive sectors, intermediating 

the flow of capital from surplus to deficit units, and supporting government's financial and 

economic policies, thus supporting developmental plans. The stability of DFIs in developing 

economies is notably significant, as any distress affects such plans and thereby economic progress. 

DFI stability is thus a prerequisite for sustainable economic growth and resilience against financial 

crises. Similar to other businesses, DFI success is evaluated based on their financial 

performance/profitability and asset quality, which are crucial to long-term viability. Given the 

pivotal role of DFIs, ensuring high-quality assets and strong profitability as measures of 

institutional soundness is essential for continued effective developmental contribution in Kenya. 

(Ombaba, 2013). 

Financial performance and profitability are key benchmarks for success in DFIs as with any 

business (Adesoye et al., 2018). However, rising non-performing loans (NPLs) directly impact 

DFI profitability by diluting returns on assets (ROA), a key performance metric (Mutunga, 2018; 

Chimkono et al., 2016). NPLs negatively affect ROA (Mutunga, 2018) and can lead DFIs incurring 

heavy disposal costs eroding profits (Araka et al., 2018). NPLs also represent opportunity costs as 

the non-interest generating assets could be invested elsewhere to generate earnings (Adesoye et 



  

5 
 

al., 2018). Additionally, DFIs must provision for losses on bad assets, affecting profitability 

(Babones et al., 2020). There are also costs associated with NPL recovery efforts (Babones et al., 

2020). Therefore, higher NPL ratios pose risks to DFI profitability through ROA dilution and 

increased expenses related to provisions, recoveries and disposals. Maintaining sound asset quality 

is important for protecting and enhancing developmental financial institution performance and 

viability. 

Asset quality refers to the level of classified or problematic assets compared to total loans provided 

(Nyarko-Baasi, 2018). For DFIs, asset quality relates to loan quality, measured by non-performing 

loans (NPLs) consisting of past due and restructured loans. Assessing asset quality can reflect a 

DFI's ability to manage productive assets profitably (Githua, 2015). DFI funds are intended to 

generate expected income levels through placement in assets. 

Similar to other financial institutions, the major risk for DFIs is non-performing assets (NPAs), 

which are bad loans where borrowers fail to meet repayment obligations. Nthiga (2017) stressed 

NPAs in loan portfolios undermine operational efficiency, ultimately impacting profitability, 

liquidity and solvency. 

While important for all organizations, asset quality is significantly impactful on DFI profitability 

given their role in financial markets, national economies and development processes. Poor asset 

maintenance can disrupt proper DFI operations and the wider financial system (Wamalwa, 2016). 

Therefore, ensuring strong asset quality is vital for sustained financial and developmental 

contributions by these institutions. 

1.1.1 Drivers of Financial Performance 

Drivers of financial performance can be defined as the key factors or variables that significantly 

influence and shape the financial outcomes and success of a company (Kimani, 2023). These 

drivers as asserted by Trianni, et al (2017) can vary across industries and contexts, but they generally 

encompass various aspects of a firm's operations, strategies, and external environment. The key 

drivers of financial performance include factors that have a significant impact on the activity of 

another entity, such as macro and microeconomic drivers, and business drivers that impact the 

operational and financial results of a business (Issah & Antwi, 2017). These drivers can be 



  

6 
 

quantifiable measurements used to determine, track, and project the economic well-being of a 

business, and they vary significantly by industry (Korhonen et al., 2023).  

Asset quality refers to the risks associated with assets held, like loans, securities, property, and off-

balance sheet items (Arisa, 2018). It assesses credit risk prediction accuracy and management 

effectiveness. 

Asset quality determines an institution's condition based on underwriting standards, credit 

practices, and risk identification (Jebessa, 2020; Gebremeskel, 2021; Belete, 2021). It strongly 

influences performance by impacting interest income and bad debt costs. Higher non-performing 

asset ratios to gross/net assets indicate lower quality, and vice versa (Lamers et al., 2019). 

Financial performance measures how efficiently a firm generates revenues from core operations 

over time, also indicating overall health (Awaysheh et al., 2020). It can be compared within/across 

industries and sectors. Performance improves by actions like optimizing cash flow, divesting 

unneeded assets, revising budgets, cutting expenses, refinancing debt, and analyzing statements 

and indicators (Nzoka, 2015). 

Non-performing loans negatively associate with profitability (Babones et al., 2020). For this study, 

asset quality of DFIs is assessed using the non-performing loans to loan loss provision ratio to 

capture credit risk impacts on financial returns. Effectively gauging asset maintenance is important 

for institutional soundness and sustainability. 

Management efficiency greatly influences firm financial performance by optimizing operations, 

reducing information asymmetry, and facilitating good decisions (Ting et al., 2021). As a complex 

yet pivotal profitability factor, it is challenging to capture management efficiency solely through 

subjective evaluation. Instead, ratios can offer objective measurement, like total asset growth, loan 

growth rate, earnings growth rate, and operational efficiency in managing operating expenses 

(Ongore & Kusa, 2013). 

Specifically, the operating expenses to total assets ratio indicates management quality by reflecting 

cost control abilities (Athanasoglou et al., 2005). Higher efficiency lowers expenses, boosting 

profitability, implying an inverse relationship between the operating expense ratio and profits 



  

7 
 

(Bourke, 1989). Additionally, management efficiency is a key financial performance determinant, 

measured via ratios such as operating expenses to total assets. Efficient management links to 

greater profitability (Guru et al., 2002). For this study, the operating profit to total income ratio 

served as the management efficiency metric to objectively gauge the impact of managerial 

competencies on developmental financial institution performance. 

Liquidity management involves ensuring a company has sufficient cash flow to meet near-term 

obligations through effective working capital strategies (Effiong & Ejabu, 2020). It refers to the 

ability to fulfill financial obligations promptly by keeping optimal capital availability (Kimathi, 

2014). Sound liquidity management relies on timely financial data and is critical for performance 

since it impacts working capital (Chuan-guo et al., 2014). 

As companies mature, liquidity management evolves from basic cash needs to maximizing yield 

across supply chains using full visibility (Coupa, 2023). Different types of liquidity include asset, 

market and accounting liquidity (Islam et al., 2017; Díaz & Escribano, 2020). Practices encompass 

monitoring receivables/payables, optimizing them, and using ratios for assessment (Musiita et al., 

2023).Prior studies examined liquidity via ratios such as current and quick ratios (Khokhar, 2015; 

Bolek, 2013), which consider current assets/liabilities in gauging ability to meet near-term 

obligations (Musiita et al., 2023).Consistent with these liquidity management discussions, the 

current ratio of current assets to current liabilities was adopted as the measure in this study. 

1.1.2 Financial Performance 

Financial performance reflects the extent to which a company achieves its financial objectives over 

time (Ichsan et al., 2021). It is typically evaluated using financial ratios from statements, like 

liquidity, profitability, activity and debt ratios (Devi et al., 2020). Operational performance, 

income sources, and shareholder returns factor into assessments as well (Ichsan et al., 2021). 

Margin growth rates and accounting analysis also provide insights (Silaban, 2017). 

Common financial performance indicators include ROA, NIM and ROE, which measure asset and 

equity-based profitability (Silaban, 2017). Loan concentrations also impact institutional results, 

with quality management linked to better performance (Chege et al., 2019). 



  

8 
 

DFI performance indicates profit sustainability abilities to withstand losses while strengthening 

capital positions (Chege et al., 2019). Persistent losses deplete equity risking investor positions. 

For shareholder value creation, ROE must surpass equity costs. 

ROE divides net income by equity, reflecting returns on shareholder investments. Higher ROE 

signals more internal profit generation abilities (Araka et al., 2018). ROA divides income by assets, 

gauging asset-based profit efficiencies (Babones et al., 2020; Chin & Gallagher, 2019). Higher 

ROA demonstrates more effective resource utilization (Wen, 2010). 

This study uses ROA and ROE as core DFI financial performance metrics in Kenya. They 

encompass various dimensions of institutional soundness critical for ongoing contributions to 

development goals according to San and Heng (2011). 

 

1.1.3 Development Finance Institutions in Kenya 

Development Finance Institutions (DFIs) in Kenya were established to provide long-term 

financing to support industrialization by targeting priority economic sectors. At that time, Kenya's 

financial system was unable to adequately meet the needs of African farmers/business owners or 

supply enough long-term capital for economic development, primarily due to inherent structural 

deficiencies within the system (Mpofu & Sibindi, 2022; Crouzet & Eberly, 2019). As the 

conventional banking sector was ill-equipped to facilitate long-term lending and meet the needs of 

various stakeholders, the government saw a need to establish specialized development banks in 

the form of DFIs to fill gaps in financing and promote growth objectives. The lack of appropriate 

financing mechanisms constrained broader economic progress, contributing to the decision to set 

up DFIs as a strategic solution. 

In response, the Kenyan government deliberately established various DFIs with the mandate of 

channeling long-term financing to the private sector. The first such DFI was the Industrial and 

Commercial Development Corporation (ICDC), founded in 1954, reflecting the strategic 

importance placed on these institutions to fill financing gaps hampering development at the time. 

Their establishment aimed to support industrialization by facilitating access to long-term capital 



  

9 
 

unavailable from traditional banks, demonstrating the developmental roles DFIs were designed to 

fulfill in Kenya's evolving economy and financial architecture. 

Following Kenya's independence, several specialized DFIs were established to support specific 

sectors. These included the Agricultural Finance Corporation (AFC) in 1963, the Industrial 

Development Bank (IDB) in 1963 (later transformed into the Industrial and Commercial 

Development Bank, ICDB), and the Kenya Industrial Estates (KIE) in 1967 (Mutunga, 2018). 

These institutions aimed to provide financial services and support to agriculture, industry, and 

small and medium-sized enterprises (SMEs). The consolidation and restructuring of the DFIs in 

the 1980s and 1990s resulting from changing economic conditions saw the government undertook 

a series of measures to consolidate and restructure DFIs (Chege, Omagwa, & Abdul, 2019). This 

included the merger of ICDC, ICDB, and AFC into the Industrial and Commercial Development 

Corporation (ICDC) in 1986. The ICDC was later transformed into the Industrial and Commercial 

Development Corporation (ICDC) in 1993, which became a holding company for various 

subsidiaries operating in diverse sectors (Sanyal et al., 2016).  

Alongside the consolidation efforts, specialized DFIs were established to cater to specific sectors. 

For example, the Agricultural Finance Corporation (AFC) continued to provide financial services 

to the agricultural sector, while the Export Promotion Council (EPC) focused on supporting 

export-oriented industries (Nganga & Mugo, 2018). The Micro and Small Enterprises Authority 

(MSEA) was also established to provide financial and non-financial support to micro and small 

enterprises. With this, the government introduced new DFIs and reforms in the 2000s and beyond. 

In recent years, Kenya has witnessed the establishment of new DFIs and the introduction of 

reforms to enhance their effectiveness. For example, the Kenya Mortgage Refinance Company 

(KMRC) was established in 2018 to provide long-term funding for affordable housing. 

Additionally, the government launched the Kenya Development Bank (KDB) in 2020, with a 

mandate to provide long-term financing for development projects across various sectors making a 

total of five DFIs in Kenya which include Tourism Finance Corporation, Industrial Commercial 

and Development Corporation, IDB capital, Kenya Industrial Estates and Agricultural Finance 

Corporation with such institutions located in Nairobi (Humphrey & Michaelowa, 2019). While 

Development Finance Institutions (DFIs) have existed in Kenya since the 1960s and undergone 

reforms, there continues to be a sizable gap in development financing, raising concerns about how 



  

10 
 

the private sector can sufficiently expand and grow without appropriate long-term investment 

funding (Mullei & Ng, 1990; Chironga et al., 2021). Even after decades of establishing and 

reforming specialized developmental lending institutions, a notable shortage of financing for 

productive investment persists. This ongoing financing insufficiency poses challenges for 

meaningful private sector advancement, highlighting the importance of DFIs optimizing their 

operations and outreach to more fully meet ongoing needs and fulfill their developmental mandates 

of stimulating broader economic activity and progress through sustainable private sector 

engagement over the long-run. 

DFI investments in Kenya show a clear focus on the finance sector, which receives the lion's share 

of funding at 57% (Marbuah et al., 2022). Energy ranks as the second most supported sector, 

receiving 13% of the funds with an example of collaboration between CDC and Globeleq resulted 

in a noteworthy investment of US$ 66 million, with CDC contributing US$ 50 million and 

Globeleq investing US$ 16 million for the establishment of a solar plant in Malindi, Kenya in 2018 

(Marbuah et al., 2022). Other sectors such as industry, tourism, manufacturing, and transport each 

receive a moderate share of roughly 4% to 5% of DFI investments. Agriculture, though vital, 

represents a relatively smaller portion, accounting for approximately 2% of the DFI flows. It's 

worth noting that even when excluding generic funds, which make up only 3.9% of total DFI funds 

in Kenya, with the overall investment landscape remains largely unchanged (Kiremu, 2020).  

1.2 Statement of the Problem 

Over many years, international donors have concentrated on establishing and reinforcing 

development finance institutions (DFIs). However, numerous such institutions have faced issues 

including borrower delinquencies on loan repayments, high administrative expenses, financial 

insolvency, and reliance on subsidies for ongoing viability. While DFIs were established with the 

aim of catalyzing private sector investment in development priorities, in practice some have 

struggled with operational challenges that undermine self-sufficiency and sustainability objectives. 

Persistent difficulties with asset quality, efficiency, and profitability have prevented certain DFIs 

from fulfilling their developmental mandates and realizing their full potential without ongoing 

donor dependence, bringing into question whether they have optimized mechanisms for generating 

commercial returns and development impact simultaneously in the long run. 



  

11 
 

Over the past two decades, the financial performance of Development Finance Institutions (DFIs) 

in Kenya has failed to meet stakeholder expectations, as reported by the World Bank in 2019. 

Notably, Kenyan DFIs have witnessed a concerning upward trend in non-performing loans 

(NPLs). For instance, non-performing loans increased by 32.4 percent, from Ksh. 11 billion in 

December 2018 to Ksh. 17 billion in December 2019. Furthermore, the ratio of gross non-

performing loans to gross loans inched up from 3.9 percent in December 2019 to 7.9 percent in 

December 2020, according to the Central Bank of Kenya (CBK, 2021). 

These developments have raised questions about the credit policies and overall performance of 

Kenyan DFIs, drawing criticisms from various quarters (Klagge, 2020). The International 

Monetary Fund (IMF) highlighted concerns about DFIs in Kenya failing to provide sustainable 

long-term finance to the industrial and agricultural sectors, with credit allocation being influenced 

by political and social considerations (IMF, 2018). Additional issues include inadequate incentives 

for effective loan collection, pressure to forgive loans, high default rates, elevated credit costs, and 

the absence of efficient appraisal systems (Mutunga F. K., 2018). These challenges persist despite 

various financial reforms aimed at enhancing DFI performance in Kenya (Sporta, 2018). 

In light of these concerns, it becomes imperative to investigate whether asset quality, management 

efficiency, and liquidity management represent the missing factors that hinder the financial 

performance of DFIs in Kenya. If these factors are indeed contributing to the issue, it is crucial to 

understand the extent to which they affect the financial performance of DFIs in Kenya. 

Several studies have analyzed factors influencing financial performance. Gabriel et al. (2019) 

found non-performing loans negatively impact Nigerian bank profitability but focused only on 

NPLs, cash reserves and inflation, lacking contextually and conceptually. 

Kimani (2023) established underwriting risk and solvency determine Kenyan insurance company 

performance but addressed insurance rather than DFI context.Kim et al. (2022) revealed smart 

manufacturing's positive impact on business performance using meta-analysis, differing 

methodologically from this study's descriptive and regression approach specific to DFIs in Kenya. 

Akims (2021) investigated liquidity regulation impact on Kenyan microfinance bank profitability 

but isolated management efficiency's influence, conducted in Kenya but concerning microfinance 

not DFI context. 



  

12 
 

Theoretically, prior studies did not account for liquidity preference theory and CAMEL model 

included in this study framework to better analyze Kenya DFI profitability drivers. 

While informing generally, prior research exhibited gaps conceptually, contextually, 

methodologically and theoretically that this study aimed to address through its Kenyan DFI-

focused analysis empirically testing performance determinants. 

1.3  Objectives of the Study 

1.3.1 General Objective 

The purpose of the study was to analyse the drivers of financial performance of Development 

Financial Institutions in Kenya 

1.3.2 Specific Objectives of the Study 

The specific objectives of the study were to: 

i. To examine the effect of asset quality on financial performance of Development Finance 

Institutions in Kenya 

ii. To analyze the effect of management efficiency on financial performance of Development 

Finance Institutions in Kenya 

iii. To analyze the effect of liquidity management on financial performance of Development 

Finance Institutions in Kenya 

1.4 Research Questions 

The research questions which guided the study included the following: 

i. What effect does asset quality have on the financial performance of Development Finance 

Institutions in Kenya? 

ii. What effect does management effficiency have on the financial performance of 

Development Finance Institutions in Kenya? 

iii. What effect does liquidity management have on the financial performance of Development 

Finance Institutions in Kenya? 

 

 

 

 

 



  

13 
 

1.5 Significance of the Study 

This study on determining the key drivers of financial performance among Development Finance 

Institutions (DFIs) in Kenya holds considerable importance. DFIs serve a vital function in 

furthering economic progress through channeling financial means to various sectors in need of 

support. Nevertheless, a more robust comprehension of the aspects shaping their financial returns 

is still needed. Research investigating the determinants impacting Kenyan DFI profitability would 

add value to current understanding by unveiling the particular elements influencing institutional 

returns within this setting. Such insights would help address existing knowledge gaps and enhance 

appreciation for the contextual nuances surrounding DFI operations locally. Given their 

developmental mandates, optimized performance proves critical to sustaining self-financing 

abilities and development impact over the long-term. This investigation thereby presents 

meaningful potential to furnish more granular perspective guiding improved strategic leadership, 

policy design, and overall viability assessment of these specialized institutions. 

The study carefully selected eight DFIs as representative sample in Kenya. The sample adequately 

captured the diversity of DFIs in terms of size, scope, and operational characteristics. A census 

sampling strategy ensures that the findings would be generalized to the broader population of DFIs 

in Kenya, increasing the study's methodological rigor. Robust statistical techniques help identify 

significant drivers, establish causal relationships, and control for potential confounding variables, 

enhancing the study's methodological soundness. 

The findings of the study would provide valuable insights to DFI managers and policymakers in 

Kenya regarding the factors that drive financial performance. Understanding these drivers would 

help inform strategic decision-making processes, such as resource allocation, investment 

strategies, risk management practices, and operational improvements. The study would guide DFI 

leaders in identifying areas where they need to focus their efforts to enhance financial performance 

and sustainability in Kenya. 

1.6 Scope of the Study 

This study aims to analyze the key drivers of financial performance among Development Finance 

Institutions (DFIs) in Kenya by examining the effects of asset quality, management efficiency and 

liquidity management. Financial performance will be measured using Return on Assets (ROA) and 

Return on Equity (ROE). The unit of analysis is Kenyan DFIs, with focus on five specific 



  

14 
 

institutions: Industrial and Commercial Development Corporation (ICDC), Kenya Industrial 

Estates (KIE), Agricultural Finance Corporation (AFC), Industrial Development Bank Capital 

Limited (IDB), and Tourism Finance Corporation (TFC). The time period covered spans financial 

years 2012/2013 to 2019/2020. Three of the DFIs (TFC, ICDC & IDB Capital) merged in 2021 to 

form Kenya Development Corporation, so the study period ends at 2019/2020. 

 

 

 

 



  

15 
 

CHAPTER TWO 

LITERATURE REVIEW 

2.1 Introduction 

This chapter delves into various theoretical approaches relevant to the study's variables and the 

financial performance of organizations as outlined in the study objectives. Theoretical foundations 

will be developed and utilized to establish relationships among key variables, culminating in the 

formation of a conceptual framework. The chapter will further review empirical research, critique 

existing literature, and identify research gaps in the realm of financial performance drivers. 

2.2 Theoretical Review 

A theoretical review serves as a foundation for understanding the study's variables within a 

universal context. Utilizing multiple theories in both quantitative and qualitative research, as 

advocated by Creswell and Creswell (2017), helps establish robust bases for extracting and 

analyzing the constructs in this study. In this study, two key theories have guided the theoretical 

foundation: the Liquidity Preference Theory and the CAMEL Model. 

2.2.1 Liquidity Preference Theory 

The Liquidity Preference Theory introduced by Keynes challenges the constant money preference 

assumption. It outlines three money-holding motives: transaction needs income-expense bridging; 

precautionary reserves for unforeseen costs; and speculative motive from interest rate anticipation 

(Keynes, 1936). Uchendu (2011) notes it refuted constant preferences, while income velocity 

depends on complex variables. Ankintoye (2000) agrees velocity varies with interest rates, 

preferences, incomes, expenditures, substitutes and non-bank entities' presence. 

Keynes identified transaction motives closing receipt-spending gaps; precautionary reserves for 

unexpected costs; and speculative motives from potential alternative asset losses if rates rise 

(Ogiriki and Andabai, 2014). The transaction motive exists since incomes and needs differ, 

necessitating balances for needs (Gbosi, 2005; Farooque et al., 2007). The transactions motive 

money demand refers to amounts required by institutions for daily operations to facilitate 

transactions, influenced by income amounts, intervals, and spending behaviors. 



  

16 
 

In summary, the theory challenges constant preferences, outlining motives driving variable 

liquidity demand influenced by economic and financial conditions. It guides understanding 

liquidity preferences role on institutional liquid assets and performance. 

Ogiriki and Andabai (2014) define speculative motive as when financial institutions keep cash on 

hand to profit from changes in the price of bonds and securities, as opposed to precautionary 

motive, which is when the institutions want to keep some liquid money on hand to cover any 

unforeseen emergencies, contingencies, or accidents. The amount of revenue, employment, 

pricing, firm turnover, and the typical intervals between receiving income and disbursing cash are 

all elements that affect how the transaction balances change (Parker, 2007).  

Jhingan (2004) agreed with the precautionary demand for money stemming from public desires to 

save for unexpected costs and advantageous purchase opportunities. Demand is influenced by 

arbitrary variables and interest rates according to Jhingan (2004). However, transactions and 

precautionary motives are income elastic but interest inelastic, implying demand declines with 

increasing rates represented by an (L) function. Demand is determined by income (Y) and M1 

levels, where M1 = L1Y (Ogiriki and Andabai, 2014). 

Andabai (2011) noted speculative expectations regarding bond prices or market rates impact 

money demand. Keynes (1936) further posited interest rates are established in money markets, 

with exogenous money sources, though motivations affect money demand and spending levels. 

According to Nzotta (2014), individuals prefer holding wealth as liquid assets if higher future rates 

are anticipated to prevent investment losses associated with anticipated rate rises. 

In summary, demand is driven by diverse motives like precautionary savings and speculation on 

rate changes, influenced by income, interest and uncontrollable factors as per Keynes' liquidity 

preference theory. 

 

 

 

 



  

17 
 

The liquidity preference theory states that individuals and institutions prefer holding liquid assets, 

such as cash or highly liquid securities, rather than illiquid assets. This theory can support the asset 

quality of development finance institutions (DFIs) in Kenya in that the DFIs face various risks in 

their operations, including credit risk, market risk, and liquidity risk. By adhering to the liquidity 

preference theory, DFIs maintains an adequate level of liquid assets that is readily converted into 

cash that helps them mitigate liquidity risk. Having sufficient liquidity ensures that DFIs meets 

their financial obligations, including the repayment of loans , even in times of financial stress or 

unexpected liquidity demands. This, in turn, supports the overall asset quality of DFIs by reducing 

the likelihood of default and improving their ability to meet their financial commitments.  

2.2.2 CAMEL Model 

The 1979 CAMEL framework structures bank examinations through ratios evaluating Capital 

adequacy, Asset quality, Management, Earnings and Liquidity. Widely used in performance/risk 

assessment by researchers, managers and central banks (Dang, 2011). 

Adequate capital ensures absorbing shocks, though optimal levels debate balances shareholders' 

equity costs against regulators' failure-risk objectives (Koch, 1995). This study explores capital 

adequacy's DFI financial performance impact in Kenya.Liquidity management entails effective 

deposit/loan handling. The study investigates its influence on Kenyan DFI returns. 

Efficient spending and operational competence notably affect profitability. High overheads can 

significantly reduce bank profits (Beck and Fuchs, 2014). This study examines expenditure 

management-profitability relationships for Kenyan DFIs.Encompassing growth, loans, earnings, 

management efficiency critically determines financial institution profitability. Exploring this 

driver's connection to performance is a study objective. 

In summary, CAMEL furnishes a tested framework to evaluate capital, assets, management, 

earnings and liquidity drivers of institutional viability through this investigation. 

The CAMEL model is relevant to this study due to its widely used framework for assessing the 

financial performance and soundness of financial institutions such as DFIs. The model provided 

that each component represents a critical aspect of a financial institution's operations. For instance, 

the management quality assesses the effectiveness and competence of the institution's management 

team in guiding its operations and decision-making. Understanding the drivers of management 



  

18 
 

quality within DFIs is crucial for evaluating their overall governance structure, strategic decision-

making processes, risk management practices, and transparency. Examining factors such as 

leadership capabilities, board composition, corporate governance practices, and risk management 

frameworks provides insights into how management quality influences the financial performance 

of DFIs. Additionally, DFIs need to maintain sufficient liquidity to support their lending activities 

and respond to unexpected liquidity demands. Therefore, examining the drivers of liquidity, such 

as funding sources, liquidity management strategies, and regulatory compliance, provides depth 

into how DFIs manage their liquidity and the impact of liquidity on their financial performance in 

Kenya. 

2.3 Empirical Review 

2.3.1 Effect of Asset Quality on Financial Performance 

Asset quality notably influences financial institution profitability as assets such as loans, 

investments and property holdings comprise the primary income sources. Given this, loan portfolio 

quality represents a key determinant as loans typically form the largest proportion of assets. Non-

performing loan ratios serve as useful metrics, with lower ratios indicating healthier portfolios. 

The quality level directly determines profitability level since the greatest risk financial institutions 

face is losses from delinquent loans (Dang, 2011). Therefore, non-performing loan ratios provide 

the strongest proxies for asset quality assessment. Previous research revealed a positive 

relationship between profit rates and asset quality indicators, demonstrating higher profit margins 

accompanying improved asset maintenance (Sangmi & Tabassum, 2010; Ongaki, 2012). While 

crucial, asset quality alone does not dictate overall institutional performance - factors like capital 

adequacy, management competence, earnings and liquidity also exert influence (Kimanzi, 2015). 

Weak asset quality and liquidity specifically constitute the two primary triggers of financial 

institution failures, as evidenced by numerous Kenyan bank collapses in the early 1980s resulting 

from poor portfolio maintenance. 

 

 

 



  

19 
 

According to (Waweru & Kalani, 2009) historical data demonstrates that poor asset quality and 

low liquidity levels have been major contributors to the failure of financial institutions. In Kenya, 

during the early 1980s, 37 banks collapsed due to non-performing loans during the banking crises 

of 1986-1989, 1993-1994, and 1998 (Mwega, 2009).  With insider lending, often to politicians, 

playing a significant role in many of these bank failures (Cheruiyot, 2016). 

Emphasizing the significance of non-performing loans, it becomes evident that credit risk 

management takes precedence over other aspects of DFIs' functioning (Kahindi, 2019). Focusing 

on non-performing loans may divert a significant portion of a DFI's resources towards recovery 

procedures instead of business expansion (Klagge, 2020; Shaikh, Ismail, & Shafiai, 2017). As a 

consequence, NPLs significantly impact the performance and profitability of DFIs (Chin & 

Gallagher, 2019). Furthermore, NPLs can hinder credit expansion for productive purposes (Githua, 

2015), potentially leading DFIs to opt for safer, lower-risk investments that may not be conducive 

to overall economic growth (Shaikh, Ismail, & Shafiai, 2017). Attridge et al. (2019) underscore 

that non-performing assets in loan portfolios can also affect operational efficiency, with 

subsequent implications for profitability, liquidity, and solvency positions of DFIs. 

Gabriel, Victor, and Innocent (2019) conducted an analysis of how asset quality impacts the 

financial performance of commercial banks in Nigeria. Utilizing regression analysis, their study 

revealed a negative and significant relationship between non-performing loans and financial 

performance. It's important to note that this study was conducted in Nigeria, utilizing dynamic 

panel data modeling, while the present study focuses specifically on DFIs in Kenya. 

Budiarto (2021) assessed the impact of asset quality on the financial performance of BPR in 

Central Java, emphasizing the role of empathy credit risk. Their data analysis, employing SEM 

AMOS, showed that credit collectability and non-performing loans had a significant relationship, 

with empathy credit risk moderating this influence. Notably, this study was conducted within BPR 

in Central Java, and the current study examines DFIs in Kenya. 

2.3.2 Effect of Management Efficiency on Financial Performance 

In assessing the financial performance of financial institutions, the efficiency of expenditure 

management is a significant factor. Poor management of expenditures is a primary contributor to 

reduced profitability (Sufian & Chong, 2009). The Cost-Income Ratio (CIR) is commonly used in 



  

20 
 

the banking industry to gauge managerial efficiency, and studies have indicated that local banks 

often exhibit higher CIR compared to their global counterparts, necessitating cost reduction for 

global competitiveness (Mathuva, 2009).  

Beck and Fuchs (2004) examined factors driving high interest margins in Kenyan banks, 

identifying overheads as a key component. Specifically, staff wage costs were relatively higher 

versus peers in Sub-Saharan Africa. However, the link between expenses and profits is complex. 

In less competitive markets affording bank market power, increased costs can be transferred to 

customers through pricing, potentially creating a positive relationship between overhead and 

profitability (Flamini et al., 2009). Indeed, Neceur (2003) found overhead costs positively and 

significantly impacted performance, implying these were sometimes passed on through lower 

deposit rates or higher loan rates. While high costs intuitively reduce earnings, market conditions 

that allow cost-shifting can obscure this dynamic, necessitating empirical investigation of the 

management efficiency-performance nexus in the specific Kenyan DFI context. 

Yadav and Katib (2015) examined the technical efficiency of Malaysia's Development Financial 

Institutions using a Two-Stage DEA Analysis. Their findings suggest that managerial inefficiency 

had a greater role in overall technical inefficiency compared to scale inefficiency. Additionally, 

the study revealed that only specific institutions achieved constant returns to scale during the 

period analyzed. Nevertheless, it's crucial to recognize that this study was conducted in Malaysia, 

using dynamic panel data modeling, while the present research centers on DFIs in Kenya. 

Ngumo, Collins, and David (2017) investigated the variables affecting the financial performance 

of microfinance institutions in Kenya. Their analysis revealed a direct association between the 

financial performance of Kenyan microfinance banks and their operational efficiency. However, 

this study primarily focused on the factors influencing MFBs' financial performance, whereas the 

current study concentrates on factors affecting the financial performance of Development 

Financial Institutions in Kenya. 

Ikapel, Namusonge, and Sakwa (2019) evaluated the effect of financial management on the 

financial performance of Kenyan commercial banks over a ten-year period. Their analysis found 

a strong and positive correlation between managerial efficiency and the financial performance of 

commercial banks. While the study primarily focused on management efficiency, it also identified 



  

21 
 

other crucial factors contributing to commercial banking performance. The present analysis 

expands its focus to encompass Development Financial Institutions in Kenya. 

2.3.4 Effect of Liquidity Management on Financial Performance 

A previous study by Edem (2017) examined the relationship between liquidity management and 

performance of deposit money banks in Nigeria from 1986 to 2011. It analyzed all 24 banks in the 

industry over the specified period using descriptive, correlational and inferential analyses. The 

findings revealed significant connections between liquidity management variables (liquidity and 

cash reserve ratios) and return on equity (ROE), with positive impacts observed. However, a 

negative impact was seen for the loan-to-deposit ratio. Notably, this preceding research only 

considered ROE as the measure of performance, whereas the present study employs both ROA 

and ROE to gauge financial performance. Overall, Edem's (2017) work provided insight into the 

impact of liquidity management on bank profitability in Nigeria. However, the current study 

expands the analysis to a differing context of developmental finance institutions in Kenya, while 

also utilizing a more comprehensive dual-metric of performance comprising ROA alongside ROE. 

Manoj et al. (2018) evaluated performance of major Indian public sector banks using CAMEL 

metrics through descriptive statistics. They analyzed five large state-owned commercial banks, 

identifying liquidity measured by liquid assets to total assets as an important indicator. However, 

their study did not examine the causal relationship between liquidity levels and financial 

performance metrics in DFIs specifically. The current research aims to fill this gap by empirically 

investigating the influence of liquidity on key profitability and other outcome variables in Kenyan 

DFIs beyond just descriptive assessment. While providing insights on CAMEL application and 

liquidity's role, the prior work did not test liquidity management impacts on DFI results. This study 

therefore expands understanding by addressing the causality aspect not explored previously to give 

a more comprehensive view of liquidity's effects on developmental institutions. 

Mugo and Mutsweje (2020) examined how liquidity regulation impacts the financial performance 

of commercial banks in Kenya. Analyzing data from 42 banks, the study revealed that liquidity 

regulation significantly affected the financial performance of commercial banks in Kenya. 

Although their research centered on liquidity and financial performance within the Kenyan 

context, the present study uniquely focuses on Development Financial Institutions in Kenya. 



  

22 
 

Akims, Kiio, Tenai, and Akims (2021) investigated how liquidity affects the profitability of 

Microfinance Banks in Kenya, utilizing agency theory and capital buffer theory. Their panel 

regression analysis indicated that liquidity regulation was not significant in affecting the 

profitability of Microfinance Banks in Kenya. Although the study considered liquidity, capital and 

credit regulations it isolated the significance of management efficiency in affecting the profitability 

of the microfinance banks in Kenya. Furthermore, this study although was conducted in Kenya, it 

differs in contextual areas as the former was tied to microfinance banks while the focus of this 

survey was on the Development Financial Institutions in Kenya. Theoretically, the former survey 

was predicated on the aforementioned theories, however, ignored the relevance of liquidity 

preference theory and the CAMEL model in explaining profitability of the microfinance banks in 

Kenya which this study adopted for its analysis of the drivers of DFIs in Kenya.



  

23 
 

2.4 Summary of Research Gaps 

Table 2.1: Summary of Research Gaps 

S/N Author(s)/Year Title Findings Research Gap 

1 Gabriel, Victor 

and Innocent 

(2019). 

Effect of asset quality on 

the Financial 

Performance of 

Commercial Banks in 

Nigeria. 

The study established that Non-

Performing Loans negatively and 

significantly affected Financial 

Performance of Commercial Banks in 

Nigeria 

The study, however, was 

conducted in Nigeria and utilized 

a dynamic panel data modeling, 

whereas the current study focuses 

on DFIs in Kenya. 

2 Budiarto (2021) The Impact of asset 

quality Towards 

Financial Performance of 

BPR in Central Java, the 

Role of Empathy Credit 

Risk 

Data analysis using SEM AMOS shows 

the result that credit collectability 

(business prospects, debtor 

performance, and ability to pay) have a 

positive significant relationship with 

non-performing loans (NPLs). Non-

performing loans have a negative 

significant effect on the financial 

performance of banks. Empathy credit 

risk moderates the influence of business 

prospects, debtor performance, and 

ability to pay with non-performing 

loans (NPLs). 

The study was, however, 

conducted within BPR in Central 

Java, while the current study 

examines DFIs in Kenya. 



  

24 
 

3 Yadav and Katib 

(2015) 

Technical Efficiency of 

Malaysia’s Development 

Financial Institutions: 

Application of Two-

Stage DEA Analysis 

Results show that managerial 

inefficiency had a greater role in overall 

technical inefficiency compared to scale 

inefficiency. Additionally, the study 

revealed that only specific institutions 

achieved constant returns to scale 

during the period analysed. 

Nevertheless, it's crucial to 

recognize that this study was 

conducted in Malaysia, using 

dynamic panel data modelling, 

while the present research centres 

on DFIs in Kenya. 

4 Kingu, Macha, 

&Gwahula (2018) 

Impact of non-

performing loans on 

bank’s profitability: 

Empirical evidence from 

commercial banks in 

Tanzania. 

The study found that the occurrence of 

non-performing loans is negatively 

associated with the level of profitability 

in commercial banks in Tanzania.  

The study, however, was 

conducted in Tanzania and 

utilized a dynamic panel data 

modeling, whereas the current 

study focuses on DFIs in Kenya. 

 

5 Mohammed, 

Mutur i& 

Samantar (2019) 

Factors affecting loan 

repayment performance 

of banks in Garowe, 

Puntland, Somalia 

There is a considerable relationship 

seen between the performance of loan 

repayment and the characteristics of the 

borrowers in Garowe. There is also a 

significant relationship seen between 

the loan characteristics and loan 

repayment performance as well as a 

significant relationship between the 

The study focused on an Islamic 

bank, but the current study will 

focus on DFIs in Kenya. 

 



  

25 
 

purpose and performance of banks in 

Garowe district. 

6 Anggriani, & 

Muniarty, (2020). 

Influence between Non-

Performing Loans and 

Capital Adequacy Ratio 

both partially and 

simultaneously on the 

Profitability (ROA) of 

PT. Bank Central Asia, 

Tbk 

The results of this study prove that Non-

performing Loans do not affect the 

Return on Assets. However, Capital 

Adequacy Ratio has a significant effect 

on Return on Asset. While 

simultaneously this study proves that 

Non-Performing Loans and Capital 

Adequacy Ratio affect the Return on 

Assets at PT. Bank Central Asia, Tbk 

Anggriani,  and Muniarty only 

applied ROE to assess 

performance while the current 

study will apply both ROA and 

ROE 

Source: Researcher (2022) 



  

26 
 

2.5 Conceptual Framework 

The conceptual framework provides a visual representation of the theoretical relationships between 

variables in a study (Abadalla, 2017). In this research, the conceptual framework outlined the key 

variables and how they relate based on the problem statement. This set the stage for presenting the 

specific research objectives and hypotheses that guided the assessment. The framework 

diagrammatically depicted the study variables and how they would be measured. It conceptually 

linked explanatory factors like asset quality, management efficiency, and liquidity management to 

the financial performance of Development Finance Institutions (DFIs). Asset quality was 

operationalized as the non-performing loans to loan loss provision ratio. Management efficiency 

utilized the operational profit to total income ratio. Liquidity management was measured as the 

current assets to current liabilities ratio. Finally, the dependent variable of financial performance 

was captured using the average of return on assets (ROA) and return on equity (ROE). Overall, 

the conceptual framework provided a concise structure for understanding the theoretical 

relationships between variables under investigation in this study. 

 

 

 

 

 

 

 

 

 

 



  

27 
 

 

 

 

Independent Variables        Dependent Variable 

 

 

                   

 

    Financial Performance 

        

 

 

 

 

 

Figure 2.1: Conceptual Framework. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Asset Quality  

▪ Non-performing loans to loan loss 

provision  

 

Financial Performance 

▪ Return on Assets (ROA) 

▪ Return on Equity (ROE) 

 

Management Efficiency  

▪ Operating Profit to total income  

 

Liquidity Management  

▪ Current assets to current liabilities  

 



  

28 
 

 

 

 

 

2.8 Operationalization of Variables 

Table 2.2: Operationalization of Variables 

Variables Type Measurement Data 

collection 

Tool 

Data Analysis  

Asset Quality 

 

 

Independent 

Variable 

Non-

performing 

loans to loan 

loss provision 

 

Data   

extraction 

form 

▪ Descriptive 

tests 

▪ Panel 

regression 

analysis 

Management 

Efficiency 

 

Independent 

Variable 

Operating 

Profit to total 

income  

Data   

extraction 

form 

▪ Descriptive 

tests 

▪ Panel 

regression 

analysis 

Liquidity 

Management 

 

Independent 

Variable 

Current assets 

to current 

liabilities 

Data   

extraction 

form 

▪ Descriptive 

tests 

▪ Panel 

regression 

analysis 

Financial 

Performance 

Dependent 

Variable 

Average of 

ROA and ROE 

Data   

extraction 

form 

▪ Descriptive 

tests 

▪ Panel 

regression 

analysis 

Source: Researcher (2021) 

2.9 Summary of Literature Review and Research Gaps 



  

29 
 

This section revealed the existence of research gaps, which encompass contextual, conceptual, and 

knowledge-based gaps. Prior studies on macroeconomic and institutional factors and their impact 

on the financial performance of financial institutions primarily focus on developed countries, 

rather than Kenya. These studies yield mixed and inconclusive results, highlighting contextual and 

knowledge gaps. Furthermore, previous research on asset quality and financial institution 

performance fails to consider the specific relationship between asset quality and the financial 

performance of DFIs, representing a conceptual gap. This study aims to address these literature 

gaps analysing the drivers s of financial performance in Kenyan DFIs. 



  

30 
 

CHAPTER THREE 

RESEARCH METHODOLOGY 

3.1 Introduction 

This chapter introduces the research design, population, sampling methods, sample size, data 

collection techniques, data analysis, research quality, and ethical considerations. It presents a 

comprehensive overview of the research approach. 

3.2 Research Philosophy 

In this study, positivism philosophy is employed. This philosophy emphasizes the generation and 

testing of empirically verifiable hypotheses (Basias & Pollalis, 2018). While it is valuable for 

explaining lived experiences, it cannot directly measure unobservable phenomena, making it 

important to focus on the experiences of the study's respondents (Babbie, 2020; Fletcher, 2017). 

“It is problem based and integrates new knowledge with existing knowledge and allows for 

creation of original work or innovative procedures (Synder, 2019; Mays & Pope, 2020).” The 

study aims at describing the lived experience. It focuses on experience that respondents go through. 

However, experience is not observable by an external observer (Attia & Edge, 2017). This implies 

that knowledge generated is not measurable and cannot be tested to allow statistical justification 

of the conclusion (Kornberger, &Mantere, 2020; Portney, 2020). 

 

3.3 Research Design 

The study adopted an explanatory research design to investigate relationships between independent 

variables (asset quality, management efficiency, liquidity management) and the dependent variable 

(DFI financial performance). 

This design examines whether changes in one variable lead to changes in another, controlling for 

other potential influencing factors. It has proven valuable in prior related studies. 

Explanatory research is important for understanding causal relationships between variables, 

achieving objective results, improving business processes, and ensuring reliable, replicable data. 



  

31 
 

By conducting causal investigation, the researcher comprehends dependent variable impacts from 

explanatory factors, leading to informed decision-making regarding Kenyan DFIs. 

The design provides solutions regarding practical data collection and time/cost constraints. It 

allows investigating whether changes in independent variables cause changes in the financial 

performance dependent variable when controlling for other factors. This supports the objective of 

determining influence and causal links between the variables of interest. 

 

3.4 Population and Sampling 

The population consists of all individuals, objects or events that a researcher is interested in 

generalizing results to (Cooper & Schindler, 2014). It is defined by attributes that establish its 

boundaries.This study focused on the target five Kenyan DFIs listed in Appendix II, representing 

the total population. Given the small size, a census approach was deemed suitable relative to 

timeframe and scope (Chege et al., 2019). 

Census sampling examines all population units (Chandran, 2013), which occurred here as the DFIs 

comprised both population and sample owing to manageability (Chimkono et al., 2016; Mugenda 

& Mugenda, 2013). Since the target DFIs were examined comprehensively without exclusion, a 

census method aligned with feasibility and aim of covering the whole population. The limited 

number of DFIs in Kenya precluded sampling, necessitating an all-inclusive census instead for 

population representativeness as per parameters of the research context. 

 

3.5 Data Collection Techniques 

This study utilized secondary data to obtain necessary information relating to the stated objectives. 

Sources like CBK Bank oversight reports, online resources, and DFI reports were reviewed to 

gather pertinent information. Financial statements from the DFIs spanning eight fiscal years 

between 2012/2013 and 2019/2020 were analysed to examine the relationship between 

hypothesized drivers and measured dimensions of financial performance. Relying on pre-existing 

documents minimized resource needs compared to primary data collection. Published records from 

regulatory and institutional entities furnished standardized performance metrics covering an 

adequate timeframe to empirically test proposed linkages through statistical techniques applied to 



  

32 
 

retrospective financial statement information. Secondary data served as a convenient, efficient 

means of gathering ample numerical data required for quantitative analysis. 

The time period chosen for data collection was eight years, from 2012/2013 to 2019/2020. This 

timeframe was selected to establish changes in DFIs over time and use recent data for analysis. 

The period was also important as several regulations for financial institutions had been 

implemented. Major reforms and restructuring of DFIs began in 2013, aimed at improving 

financial performance. Kosikoh (2014) noted a timeframe longer than five years could aid 

calculating ratios of independent and dependent variables across multiple years, enhancing 

analysis. Therefore, data on dependent and independent variables was gathered from financial 

statements of various DFIs using a secondary data collection form. Secondary data incorporated 

assets, equity, liabilities and other financial information necessary for data examination. 

3.5.1 Validity of Data 

 

Validity refers to how accurately a study measures what it set out to measure. They are of 

different types, including content validity which ensures variables fully capture the concept 

being measured. Hair, Money, Page and Samouel (2009) and Fraser et al. (2018) define validity 

similarly as the degree to which a study's elements reliably and consistently represent the real 

meaning of the concepts under investigation. Common forms are content, face and construct 

validity. This study ensured content validity, which refers to the degree to which dimensions and 

items of a measure are relevant to and representative of the target concept. By focusing on 

content validity, the researcher established that the variables used appropriately represented and 

aligned with the concepts being assessed. 

 

3.5.2 Reliability of Data 

Reliability pertains to the consistency of measurement results and the degree to which they are free 

from random error. It indicates how well a study can produce stable and consistent results over 

time. Melcher & Beck (2018) and Bryman & Bell (2011) define reliability similarly as the level 

of accuracy and dependability provided by a measure. Reliability acts as the link between the 

theoretical concept and the observable data. In this study, data reliability was ensured through the 

use of audited financial statements. Audited financial reports are validated by auditors and 



  

33 
 

standardized accounting practices, hence providing assurance the data accurately represents 

underlying information without bias or distortions. The auditing process gave the collected 

secondary data stability and reliability by reducing random errors, establishing its dependability 

for drawing conclusions. Using reliable and consistent data sources enhanced the integrity and 

trustworthiness of findings and interpretations. 

3.6 Research Quality 

The following diagnostic tests were performed to meet the requirements of panel regression 

analysis: 

a) Stationarity Test 

In the context of time series, a stationary process is one where the statistical properties such as the 

mean, variance, and autocovariance are constant over time (Enders, 2014). This is important 

because many time series analysis techniques assume stationarity. However, due to the dynamic 

and time trend as well as nature of data, most appears to be non-stationary thus informing the 

evaluation of the stochastic properties of to avoid spuriosity of the outcomes (Brooks, 2014).  This 

test is guided by 0.05 significance level as threshold. The factors employed in the study were tested 

to determine the presence of unit root and in the event where the factor exhibits unit root, such 

factor was differenced. The test employed Fisher-type test where Augment Dickey-Fuller test was 

run to arrive at the stationarity of the variables in the investigation. The ADF test is designed to 

handle autoregressive structures, allowing for the inclusion of lagged dependent variables in the 

test equation (Kripfganz & Schneider, 2023). This is particularly useful when dealing with time 

series data that exhibit serial correlation or dependence (Fulcher, 2018).  

 

b) Normality Test 

Normality testing was conducted as part of the regression analysis to assess whether the 

assumption of a normal distribution of errors was met. The classical linear regression model 

assumes the errors follow a normal distribution with a mean of zero and constant variance. Al-

Labadi, FazeliAsl and Saberi (2020) also describe this assumption. This study used statistical 

functions in STATA software to test the normality of variable distributions. Specifically, the 

Shapiro-Wilk test was employed as it is suited for smaller sample sizes. As Henze and Visagie 

(2020) explain, a non-significant result above 0.05 from this test indicates the variable is likely 



  

34 
 

normally distributed, whereas a significance below 0.05 means it deviates from normality. By 

running the Shapiro-Wilk normality test at a significance level of 0.05, this study was able to verify 

whether the variables met the normal distribution requirement for regression or if data 

transformations were necessary. 

c) Multicollinearity Test 

 

Multicollinearity refers to strong correlations among independent variables in regression which 

can inflate their standard errors and misidentify insignificant predictors. Mohammaddi (2020) 

defined it. Field notes low multicollinearity poses minimal risk, but higher levels increase type II 

errors and unstable equations as β coefficient errors rise. To test for it, this study generated a 

correlation matrix, variance inflation factors (VIFs), and tolerances as suggested by Field. VIFs 

mathematically indicate 1/(1-R2). Gujarati (2004) views tolerances closer to 1 as evidence of no 

collinearity. Field considers VIFs substantively above 1 suggestive of multicollinearity bias. The 

presence, as defined by a tolerance under 10, means one variable can predict another's outcome. 

Therefore, this study utilized VIFs, tolerances and eigenvalues to gauge possible multicollinearity 

issues, with VIFs determined as 1/(1-R2) and tolerances/VIFs informing decisions about predictor 

independence as per the specified thresholds. This allowed evaluating whether strong relationships 

between independent variables could influence the regression analysis. 

d) Homoscedasticity Test 

Homoscedasticity refers to having equal variance of errors across all independent variable values. 

Gignac and Zajenkowski (2020) note heteroscedasticity distorts results and increases type 1 error 

risks by weakening analysis. To minimize this issue, the study ensured normal data distribution 

and proper regression functional form, as recommended. The Godfrey-Pagan test, which has a 

0.05 threshold, was used to specifically check for heteroscedasticity. When the assumption of 

homoscedasticity is violated and heteroscedasticity is present, it invalidates conclusions drawn 

from regression analysis since the variance is no longer constant. Therefore, this study utilized 

the Godfrey-Pagan test at a 0.05 significance level to empirically verify whether 

heteroscedasticity existed among the regression residuals and if remedial measure were required 

to produce consistent and reliable results. Adhering to these practices helped reduce potential bias 

from incorrect assumptions. 



  

35 
 

e) Autocorrelation Test 

Autocorrelation refers to the correlation between the errors (residuals) of a model across 

consecutive time periods or observations. When autocorrelation is present, the least squares 

(OLS) estimates of regression coefficients become less efficient and can lead to incorrect 

Tstandard errors, t-statistics, and misleading statistical inference (Baltagi, 2011). In time series 

data, autocorrelation can arise due to various factors, such as seasonality, trending, or serial 

dependence. If autocorrelation is not accounted for, it can lead to biased estimates and incorrect 

conclusions (Fomby, Johnson & Hill, 1984). To detect and address autocorrelation, several 

diagnostic tests and remedies are available in regression analysis. Common diagnostic tests 

include the Durbin-Watson test, the Breusch-Godfrey test, and the Ljung-Box test, among others 

(Wooldridge, 2010). Therefore, this study adopted Breusch-Godfrey test due to the lower 

standard deviations and better performance at large autocorrelation levels compared to other 

methods. The presence or absence of autocorrelation was judged based on 0.05 significance level. 

f) Hausman Specification Test 

In panel regression analysis, it becomes imperative to carry out a specification test which serves 

as a basis for adopting a model for estimation. The test is aimed at making a decision between 

adopting fixed effect model or random effect model. The test is based on a hypothesis which 

favours the adoption of the random effect model and alternatively the fixed effect model. The 

decision guiding the acceptance or rejection of the null hypothesis is judged by 0.05 level of 

significance. 

3.7 Data Analysis 

The study utilized a panel data approach to conduct the analysis over an 8 year period from 

2012/2013 to 2019/2020. Panel data is advantageous compared to cross-sectional data as it 

provides more detailed information with a larger dataset (Blundell & Bond, 1998; Hoechle, 2007). 

Specifically, panel data usually yields more observations which improve statistical power and 

precision of estimates, as noted by Cheng Hsiao (2004). The increased data points offer higher 

degrees of freedom while reducing collinearity, thereby enhancing efficiency of econometric 

modeling (Cheng Hsiao, 2004). Another advantage highlighted is the ability to examine economic 

issues more comprehensively than would be possible with pure cross-sectional or time series data 

alone. In summary, panel data techniques were suited for this research given the benefits of control 



  

36 
 

and insight afforded by the longitudinal multi-dimensional dataset spanning both time and entities 

under investigation. Descriptive, correlation and panel regression analyses were employed. The 

panel regression analysis takes the following form: 

FPit = β0+β1AQit+β2MEit+β3LMit + ε 

Key: 

FP = Financial Performance  

AQ = Asset Quality 

ME = Management Efficiency  

LM = Liquidity Management  

β1 - β3 = Regression Coefficients  

i= Firm 

t= Time period 

ε = Error Term 

3.8 Ethical Considerations 

This study sought to contribute new insights into the factors influencing financial performance of 

Development Finance Institutions (DFIs). In doing so, the researcher ensured appropriate and 

ethical conduct of the research as per the standards of Strathmore University in Kenya. An 

important step involved obtaining approval from relevant authorities after successfully defending 

the proposal. Specifically, a research permit was secured from the National Commission for 

Science, Technology and Innovation (NACOSTI) as mandated. Seen through this lens, the study 

aimed to add value to academic knowledge while fully adhering to procedures and guidelines 

governing research activities in the given jurisdictional and institutional contexts 

 

  



  

37 
 

CHAPTER FOUR 

PRESENTATION OF RESEARCH FINDINGS 

4.1 Introduction 

The study examined survey results using both descriptive and panel regression analysis methods. 

findings were discussed in relation to the research objectives. The analysis specifically evaluated 

the impact of drivers on Kenyan Development Finance Institutions' financial performance. 

Drivers were broken down into asset quality, management efficiency and liquidity management 

categories. Analysis then determined how each of these influence the institutions' financial 

performance both descriptively and through regression modelling techniques. 

By taking this two-pronged quantitative approach, the researcher gainfully described trends over 

time and empirically tested relationships between independent and dependent variables of interest. 

This provided a more robust assessment of how the hypothesized drivers affected DFI profitability 

and overall viability over the sample period. 

4.2 Descriptive Statistics 

The descriptive technique was employed to characterize the study data by estimating individual 

parameters underlying the research information. Descriptive statistics encompass both central 

tendency and variability measurements. Standard deviation, variance, minimum/maximum values, 

kurtosis and skewness represent variability measures while mean, median and mode indicate 

central tendency (Kaur et al., 2018). To summarize the inquiry data vis-a-vis standard deviation, 

variance, minimum/maximum element values and their means, descriptive statistics were utilized. 

This allowed depicting trends over time through statistical summaries like averages, dispersions 

from averages and outlier ranges for each variable. Such metrics facilitated analysing overall data 

patterns to contextualize subsequent inferential statistical modelling seeking to identify significant 

relationships between the study variables. 

A summary of the factors' descriptions was presented in Table 4.1. 

Table 4.1 Descriptive Statistics 

Variable Obs Mean Std. Dev. Min Max 

Financial Performance 40 3.175 14.935 -9 90 



  

38 
 

Asset Quality 40 6.579 18.475 0.946 99.408 

Management Efficiency 40 0.040 0.411 -0.754 0.772 

Liquidity Management 40 9.414 13.400 0.037 59.588 

Source:  Study Data (2023)  

The descriptive data in Table 4.1 showed that financial performance had a mean of 3.175 and a 

standard deviation of 14.935. The financial performance score ranges from -9 to 90, with the 

average number falling between those two extremes. This indicates that the Kenyan Development 

Financial Institutions' financial performance differed greatly. As a result, the Kenya’s 

Development Financial Institutions’ financial performance varies largely depending on the 

operational efficiency of the institutions in the utilization of the financial performance drivers. Due 

to the unique context of the study, the sample size is considered adequate. This is in consideration 

of not just the number of firms but also the time scope of the study especially as a census approach 

was utilized. 

4.3 Diagnostic Tests 

The study ran a number of diagnostic tests to ensure that the classical linear regression model's 

(CLRM) hypotheses had not been falsified. Diagnostics may be used to verify the accuracy of the 

results and forecast how the characteristics of development Finance Institutions would affect the 

credit risk of Kenya's DFI’s. Williams, Grajales and Dason (2013) observed that the correction of 

misconceptions about the OLS assumptions entails carefully considering the reasonableness of the 

assumptions in the context of a particular dataset and analysis is an important prerequisite to 

drawing of trustworthy conclusions from the data. According to Kerlinger (1998) any violation of 

the classical linear regression assumptions would cast doubt on the validity of the conclusions 

drawn. This investigation's pertinent diagnostic methods included tests for normalcy, 

multicollinearity, heteroscedasticity, stationarity, and fixed or random effects.  

4.3.1 Stationarity Test 

Since non-stationary series invalidate standard statistical tests because their variance is not 

constant, Asteriou and Hall (2007) claim that stationarity is a fundamental idea underlying time 

series operation. It makes sense to include major time series variables in the regression model if it 

is determined that they are stationary. The Fisher-type test was employed to evaluate if the 

variables were stationary or not. The test made use of the Augmented Dickey-Fuller (ADF) in the 



  

39 
 

evaluation of the stationarity of the variables. The null hypothesis in this test was the unit root of 

the variable. Another explanation was that the variable might not have unit roots. The results of 

the test for stationarity were shown in Table 4.2. 

Table 4.2 Fisher-type unit root test 

Variable Statistic P-value Comment 

Financial Performance 88.082 0.0000  Stationary 

Asset Quality 74.978 0.0000 Stationary 

Management Efficiency 87.110 0.0000 Stationary 

Liquidity Management 23.817 0.0081 Stationary 

Source:  Study Data (2023)  

The findings in Table 4.2 demonstrated that all the variables were stationary, which means they 

lacked unit roots, at the 0.05 level of significance. The results of this test showed that the variables' 

mean and variance were constant throughout the investigation period. The absence of unit roots in 

the study's variables means that there is no possibility of erroneous results. According to Gujarati 

(2003), the significance of the discovery is that the outcomes are valid. 

4.3.2 Normality Test 

A few examples of statistical tests that depend on the premise that the residuals are distributed 

normally or gamma-symmetrically include the t-test, correlation, regression, and analysis of 

variance (González-Estrada & Cosmes, 2019). Hence, a critical requirement in the classical linear 

regression is that the residuals must be normally distributed with zero mean and constant variance 

(Enders, 1995). The populations from which the samples are drawn are thought to have a normal 

distribution. To ascertain if the sample residuals came from a population with a regularly 

distributed population, the normality test of the Shapiro-Wilk test was applied. As a result, when 

the p-value is high and greater than 0.05, the residuals are considered normal; otherwise, they are 

not. The results of the normality tests are shown in Table 4.3. 

Table 4.3 Shapiro-Wilk test for Normality 

Variable Obs W V Z Prob>z 

Financial Performance 40 0.397 23.831 6.673 0.0000 

Asset Quality 40 0.382 23.478