University of Ghana http://ugspace.ug.edu.gh
UNIVERSITY OF GHANA
DETERMINANTS OF FINANCIAL HEALTH OF LIFE INSURANCE
COMPANIES IN GHANA
BY
BENEDICTA BAABA ASARE
(10276392)
THIS THESIS IS SUBMITTED TO THE SCHOOL OF GRADUATE
STUDIES, UNIVERSITY OF GHANA IN PARTIAL FULFILMENT OF
THE REQUIREMENT FOR THE AWARD OF MASTER OF PHILISOPHY
DEGREE IN RISK MANAGEMENT AND INSURANCE
JULY 2015
1
University of Ghana http://ugspace.ug.edu.gh
DECLARATION
CANDIDATE’S DECLARATION
This is to certify that this thesis is the result of my own research work and that no part of it has
been presented for another degree in this university or elsewhere.
Signed: ............................ Date: .....................
Asare Benedicta Baaba (10276392)
(Candidate)
SUPERVISORS’ DECLARATION
We hereby certify that this thesis was prepared from the candidate’s own work and supervised in
accordance with guidelines on supervision of thesis laid down by the University of Ghana.
Signed: ........................... Date: ........................
Dr. Albert Gemegah
(Principal Supervisor)
Signed: ............................ Date: ........................
Dr. Charles Andoh
(Co - Supervisor)
i
University of Ghana http://ugspace.ug.edu.gh
ABSTRACT
This study investigates determinants of financial health of Ghanaian life insurance companies, a
market that has experienced significant change in recent years. It examines the solvency of life
insurance companies and the current system used by the National Insurance Commission (NIC)
to monitor solvency of life insurance companies in Ghana. Secondary data obtained from
financial reports of 16 sampled life insurance companies from 2007 to 2013 were analysed using
the unbalanced panel regression analysis technique. The study used solvency ratio as a proxy for
financial health and assessed the impact of nine predictor variables, both macro and firm specific
variables on it.
Analysis of the data showed that all the predictor variables, with the exception of premium
growth and reforms are significant determinants of financial health. The results also indicated
that aside claims incurred, combined ratio and inflation, all the other predictor variables are
positively related to solvency. Further, analysis indicated a weak to moderate total financial
health of the sampled life insurers during the study. It was revealed an the end of the project that
less than 40% of the registered life companies were strongly solvent every year based on the
categorization of solvency by the study. Therefore, aside the routine checks and ensuring
compliance of insurance regulation by insures which is currently done, the NIC can monitor
these ratios published annually by insurers in addition to moving towards the proposed Risk
Based supervision in order to detect early signs of insurer distress.
ii
University of Ghana http://ugspace.ug.edu.gh
DEDICATION
This work is dedicated to my parents, Mr. Sampson Asare and my mum Mrs. Hilda Asare, and
Evangelist Isaac George Boakye.
iii
University of Ghana http://ugspace.ug.edu.gh
ACKNOWLEDGEMENTS
First, I would like to thank the Almighty God for the divine wisdom and strength to complete
this study. I thank my parents, Mr. Sampson Asare and Mrs. Hilda Asare for their support and
encouragement throughout this degree. My journey through university would not be as enriched
and fulfilling as it has been without you. I thank my sister, Caroline and cousins who were there
to support me, I’m grateful.
This thesis would not have been possible but for the guidance and support from my supervisors
Dr. Albert Gemegah and Dr. Charles Andoh; I say thank you.
I am indebted to my colleagues; your presence provided a source of warmth and hope in
challenging and difficult periods of the research. I would like to express gratitude to Charles
Kwofie and Irene Kumako for their unwavering support and encouragement. Last but not least, I
thank my husband, Dr. Kofi Boakye, for giving me the support and encouragement to go on with
this project even in difficult times.
iv
University of Ghana http://ugspace.ug.edu.gh
TABLE OF CONTENT
DECLARATION.......................................................................................................................................... i
ABSTRACT.................................................................................................................................................. ii
DEDICATION ............................................................................................................................................. iii
ACKNOWLEDGEMENTS ......................................................................................................................... iv
TABLE OF CONTENT.............................................................................................................................. v
LIST OF FIGURES ................................................................................................................................... xi
CHAPTER ONE ......................................................................................................................................... 1
INTRODUCTION....................................................................................................................................... 1
1.1 Background of the Study ..................................................................................................................1
1.2 Research Problem .............................................................................................................................7
1.3 Objectives of the Study ...................................................................................................................10
1.4 Research Questions .........................................................................................................................11
1.5 Significance of the Study ................................................................................................................11
1.6 Scope of the Study ...........................................................................................................................12
1.7 Organisation of the Study ..............................................................................................................12
CHAPTER TWO ......................................................................................................................................13
LITERATURE REVIEW ........................................................................................................................13
2.1 Introduction.....................................................................................................................................13
2.2 Concept of Insurance ......................................................................................................................13
2.3 Context of the Study .......................................................................................................................16
2.3.1 General Overview of Insurance in Ghana ..................................................................................16
2.3.2 Role of the NIC ...........................................................................................................................18
2.3.3 Risk Based Supervision Approach ..............................................................................................20
2.4 Review of Determinants of Financial Soundness of Insurance companies ................................22
2.4.1 Studies in the Ghanaian Context................................................................................................22
2.4.2 Studies in Africa and Other Parts of the World .........................................................................25
2.5 Importance and Contribution of Insurance .................................................................................27
2.6 Financial Health/ Solvency of Insurers .........................................................................................29
2.7 Empirical Findings..........................................................................................................................29
2.8.1 Firm Specific Factors and Financial Health ................................................................................31
v
University of Ghana http://ugspace.ug.edu.gh
2.8.2 Distribution System and Financial Health ..................................................................................35
2.8.3 External/Macroeconomic Determinants of Insurer’s Financial Health .....................................37
2.9 Theoretical Backing ........................................................................................................................38
2.10 Definition and Concept of Regression .........................................................................................39
2.11 Panel Regression ...........................................................................................................................39
2.11.1Fixed Effects Models (FE) ..........................................................................................................40
2.11.2 Random Effects Models (RE)....................................................................................................41
2.11.3 Assumptions for Regression Analysis.......................................................................................42
2.12 Conclusion .....................................................................................................................................43
CHAPTER THREE ..................................................................................................................................44
METHODOLOGY ...................................................................................................................................44
3.1 Introduction.....................................................................................................................................44
3.2 Target Population and Sample ......................................................................................................44
3.3 Sample and Sampling Procedure...................................................................................................45
3.4 Data Collection ................................................................................................................................45
3.5 Model Specification and Description.............................................................................................46
3.5.1 Specified Model .........................................................................................................................46
3.6 Checking for Model Assumptions .................................................................................................47
3.6.1 Checking for linearity .................................................................................................................47
3.6.2 Checking for homoscedasticity ..................................................................................................48
3.7 Measurement of Dependent and Independent Variables ............................................................48
3.7.2 Testing for the Significance of the Model and Its Parameters………………………………………………..53
3.8 Appropriateness of the Methodology ............................................................................................53
3.9 Conclusion .......................................................................................................................................54
CHAPTER FOUR.....................................................................................................................................55
DATA ANALYSIS, INTERPRETATION AND DISCUSSIONS ........................................................55
4.1 Introduction.....................................................................................................................................55
4.2 Descriptive statistics........................................................................................................................55
Table 4.2:  Testing for panel effect ..........................................................................................................58
4.3 Fixed Effect Panel Regression........................................................................................................59
Table 4.3: Unbalanced Fixed effect panel Regression estimates...........................................................59
4.3.1 Diagnostics for the Fixed Effects Panel Regression Model ........................................................60
Table 4.4: Regression statistics (Fixed effect).........................................................................................60
vi
University of Ghana http://ugspace.ug.edu.gh
Table 4.5: ANOVA results (Fixed effect) ................................................................................................60
4.4 Relationship between firm size and financial health ...................................................................61
Table 4.6: Unbalanced Fixed effect panel Regression estimates (including size square) ...................62
Table 4.7: ANOVA results .......................................................................................................................63
Table 4.8: Regression statistics ................................................................................................................63
4.5 Hausman test ...................................................................................................................................64
Table 4.9: Hausman test for best model..................................................................................................64
4.6 Diagnostic Tests for the fixed effects. ............................................................................................65
4.6.1 Testing time-fixed effects. .........................................................................................................65
The null is that no time-fixed effects needed. .............................................................................................65
Table 4.10:  Testing for time effect ..........................................................................................................65
4.6.2 Testing for serial correlation......................................................................................................66
Table 4.11:  Testing for serial correlation...............................................................................................66
4.6.3 Testing for cross-sectional dependence/contemporaneous correlation: using Breusch-Pagan
LM test of independence ....................................................................................................................66
Table 4.12:  Testing for cross-sectional dependence ..............................................................................66
4.6.4 Breusch-Pagan test for Heteroscedaticity .................................................................................67
Table 4.13:  Testing for heteroscedasticity .............................................................................................67
4.7 Categorizations of the solvency ratios ...........................................................................................67
Table 4.14: Categorization of solvency ratios.........................................................................................68
4.8 Discussion of Findings ....................................................................................................................68
4.9 Conclusion .......................................................................................................................................73
CHAPTER FIVE ......................................................................................................................................75
SUMMARY, CONCLUSION AND RECOMMENDATION ...............................................................75
5.1 Introduction.....................................................................................................................................75
5.2 Summary..........................................................................................................................................75
5.3 Conclusions ......................................................................................................................................76
5.4 Potential Areas of Application .......................................................................................................77
5.5 Recommendations ...........................................................................................................................78
5.5.1 Recommendations for Management, Policyholders and Investors...........................................78
5.5.2 Recommendations for Future Research ....................................................................................79
5.6 Limitations of the Study .................................................................................................................79
REFERENCES..........................................................................................................................................81
vii
University of Ghana http://ugspace.ug.edu.gh
APPENDICES ...........................................................................................................................................91
APPENDIX B ............................................................................................................................................92
ORDINARY LEAST SQUARES MODEL.............................................................................................92
Table 1: Unbalanced OLS Regression estimates....................................................................................92
Table 2: ANOVA results (OLS)...............................................................................................................92
Table 3: Regression statistics (OLS)........................................................................................................92
4.6 Random Effect Panel Regression.......................................................................................................93
Table 4: Unbalanced Random Panel Regression ...................................................................................93
Table 6: Regression statistics (Random effect).......................................................................................93
viii
University of Ghana http://ugspace.ug.edu.gh
LIST OF TABLES
Table 1.1: Classification of Control mechanism for Solvency under Insurance Act 724…....6
Table 1.2: Growth indicators of the life insurance industry 2011 to 2012…………………...8
Table 2.1: Categorization of levels of solvency…………………………………………....  30
Table 2.2: Classification of methods of insurance distribution………………………….… 35
Table 3.1: Summary of Dependent and Independent Variables………………………..…. .49
Table 3.2: Signs of Variables Used in Computation……………………………………..… 51
Table 4.1: Descriptive statistics……………………………………………………………. 56
Table 4.2: Testing for panel effect………………………………………………………… 57
Table 4.3: Unbalanced Fixed effect panel Regression estimates…………………………..  58
Table 4.4: Regression Statistics (Fixed effect) ……………………………………….……59
Table 4.5: ANOVA Results (Fixed effect) …………………………………………...….....59
Table 4.6: Unbalanced fixed effect Panel Regression estimates with ………………….61
Table 4.7: ANOVA Results ………………………………………………………..……….62
Table 4.8: Regression statistics ……………………………………………………………. 62
Table 4.9: Hausman test for best model…………………………………………………….63
Table 4.10: Testing for time effect……………………………………………………..……..64
Table 4.11: Testing for serial correlation…………………………………………………….65
ix
University of Ghana http://ugspace.ug.edu.gh
Table 4.12: Testing for cross-sectional dependence………………………………………... 65
Table 4.13: Testing for heteroscedasticity……………………………………………...……66
Table 4.14: Categorization of solvency ratios………………………………………………. 67
Table 1:          Unbalanced OLS regression estimates………………………………………......89
Table 2: ANOVA Results (OLS)…………………………………………………….........89
Table 3: Regression statistics (OLS)………………………………………………...........89
Table 4: Unbalanced Random panel Regression ………………………………………..  90
Table 5: ANOVA Results (Random effect)………………………………………………90
Table 6: Regression statistics (Random effect) …………………………………………..90
x
University of Ghana http://ugspace.ug.edu.gh
LIST OF FIGURES
Figure 1.1: Plot of the Market share of top five insurers by GWP from 2010 to 2012 ……. 3
xi
University of Ghana http://ugspace.ug.edu.gh
LIST OF ABBREVIATIONS
BOG Bank of Ghana
CI Claims Incurred
CML Conditional Maximum Likelihood
COV Covariance
CR Combined Ratio
DF Degrees of Freedom
Fe Fixed Effects
GDP Gross Domestic Product
GLSS Ghana Living Standards Survey
GSS Ghana Statistical Service
GWP Gross Written Premiums
IAIS International Association of Insurance Supervisors
ILO International Labour Organization
IMF International Monetary Fund
IP Investment Performance
KPMG Klynveld Peat Marwick Goerdeler
LAE Loss Adjustment Expense
xii
University of Ghana http://ugspace.ug.edu.gh
LR Liquidity Ratio
MFI Microfinance Institutions
NAIC National Association of Insurance Commissioners
NIC National Insurance Commission
NLS National Longitudinal Surveys
OLS Ordinary Least Squares
OM Operating Margin
PG Premium Growth
PNDC Provisional National Defence Council
Re Random Effects
SIC State Insurance Company
SR Solvency Ratio
STDV Standard Deviation
SWOT Strengths Weaknesses Opportunities and Threats
SZ Firm Size
UML Unconditional Maximum Likelihood
VAR‐VECM Vector Auto Regressive-Vector Error Correlation Model
CI Claims Incurred
xiii
University of Ghana http://ugspace.ug.edu.gh
CML Conditional Maximum Likelihood
COV Covariance
CR Combined Ratio
DF Degrees of Freedom
FE Fixed Effects
GDP Gross Domestic Product
GLSS Ghana Living Standards Survey
GSS Ghana Statistical Service
GWP Gross Written Premiums
IAIS International Association of Insurance Supervisors
ILO International Labour Organization
IMF International Monetary Fund
IP Investment Performance
KPMG Klynveld Peat Marwick Goerdeler
LAE Loss Adjustment Expense
LR Liquidity Ratio
MFI Microfinance Institutions
NAIC National Association of Insurance Commissioners
xiv
University of Ghana http://ugspace.ug.edu.gh
NIC National Insurance Commission
NLS National Longitudinal Surveys
OLS Ordinary Least Squares
OM Operating Margin
PG Premium Growth
PNDC Provisional National Defence Council
Re Random Effects
SIC State Insurance Company
SR Solvency Ratio
STDV Standard Deviation
SWOT Strengths Weaknesses Opportunities and Threats
SZ Firm Size
UML Unconditional Maximum Likelihood
VAR‐VECM Vector Auto Regressive-Vector Error Correlation Model
xv
University of Ghana http://ugspace.ug.edu.gh
CHAPTER ONE
INTRODUCTION
1.1 Background of the Study
The financial sector of an economy mainly consists of a category of stocks containing firms that
provide financial services to commercial and retail customers. This sector includes banks,
investment funds, insurance companies and real estate. The development of the financial sector
in a less developed economy has been the concern of some economists for the past few decades
(Georgantopoulos & Tsamis, 2011). There is a pattern of growing interest in the financial system
of Ghana due to the many structural changes and adjustment it has undergone within the last
decade. Ghana’s financial sector, especially the insurance industry has undergone rapid growth
and major structural transformation over the last decade, which has brought new opportunities
and risks (NIC, 2012). Barros & Obijaku (2007) argued that growth in the industry is muffled
when companies are unable to adapt rapidly to the changing business climate and take optimal
advantage of the opportunities created by recent changes in legislation.
Authorities have been implementing reforms to strengthen the regulatory and supervisory
framework and financial infrastructures (NIC, 2010). According to Herring & Santamero (1996),
although the direct impact of financial institutions on the real economy is relatively minor, their
indirect impact on economic performance is extraordinarily important. An efficient financial
sector reduces the cost and risk of producing and trading goods and services and thus makes an
important contribution to raising the standard of living.
1
University of Ghana http://ugspace.ug.edu.gh
Insurance is one of the key sectors in the economic development of Ghana. The insurance market
is made up of two main stakeholders, the insurance companies who work as sellers of the
insurance products, providing cover for various risks and the intermediaries mainly insurance
agents and brokers. A thriving insurance industry mobilizes savings and allocates credit across
space and time. It provides not only payment services, but also enables firms and households to
cope with economic uncertainties by hedging, pooling, sharing and pricing risks (Huan & Eling,
2013). Due to the many benefits of insurance to individuals and to every economy, measures
must be put in place to ensure their survival in the financial market and to monitor their financial
health in order to prevent insolvencies and protect policyholders in today’s increasingly
competitive markets.
One major challenge of insurers, especially across Africa is the problem of insolvency. A
combination of factors, including firm specific factors and macroeconomic factors like high rate
of inflation, massive devaluation and general poor economic performance of most African
countries can be linked to this problem of insolvency (Chen & Wong, 2004). This reduces the
trust of people in insurance; hence their unwillingness to insure which subsequently leads to poor
economic growth.
Life insurance is an insurance product that is also gradually gaining popularity in Ghana due to
its savings benefits and also because death is inevitable and its time of occurrence is
unpredictable. Many people avoid investing in life insurance simply because they imagine the
costs as being too high for their budgets. Fortunately, there are a wide variety of life insurance
policies available for each individual set of needs and budgets. These policies have low
premiums and payout death benefits for the duration of a specific period. Life assurance
2
University of Ghana http://ugspace.ug.edu.gh
contracts are usually for much longer periods compared to general insurance. Life funds are
therefore accumulated to cater for long-term liabilities.
Currently, there are twenty one registered life insurance companies in Ghana according to the
NIC. While the Life sector experienced a growth in gross premium income of 44% from 2010 to
2011, the non-life sector grew by just 32% (NIC, 2011). In general, by all indicators, the life
sector appears to be growing at a faster rate than the non-life sector. Although a significant
improvement in life insurance growth has been recorded over the past decade, its popularity in
Ghana cannot be compared to that of USA and other developed countries where most people
have life insurance cover.
Figure 1.1 Growth rate of life and non-life insurers from 2008 to 2012 (In Cedis).
450
400
350
300
250
200
150
100
50
0
2008 2009 2010 2011 2012
LIFE NON-LIFE
Source: NIC (2013)
Figure 1.1 provides a summary of the growth rate of life compared to non-life insurers in Ghana
over the period of 2008 to 2012. The graphs show a significant increase in life insurance
premiums over the 5 year period. The sharp increase in non-life premiums from 2010 to 2012
3
University of Ghana http://ugspace.ug.edu.gh
however could be attributed to compulsory insurance of commercial building and Ghana’s oil
find (NIC, 2011).
Powell (2007) established that insurance is not bought but rather sold. It appears that effective
techniques or strategies are not being utilized fully by the Ghanaian insurance companies to
publicize the benefits of the products. Again, claim handling has been reported to be poorly
managed by most Ghanaian insurers over the years.
For every organization to remain financially sound, it has to consider the major factors that
determine its financial performance. Among the major factors that could affect insurers’
financial health in growing economies are firm size, investment performance, underwriting
result, liquidity ratio, combined ratio, and operating margin, premium growth, and growth rate of
surplus (Chen & Wong, 2004). The National Insurance Commission (NIC) is the sole regulator
of insurers in Ghana. Among other responsibilities, it is responsible for maintaining contact and
developing relations with foreign insurance regulators and maintaining international insurance
regulation standards. The NIC also supervises, regulates and controls compliance with the
provisions of the insurance act and regulations made under it and any other enactment relating to
insurance (NIC, 2010). It has the responsibility of ensuring that insurers remain in operation and
monitors them from going insolvent and also to protect the interest of insurance policyholders.
Insolvency occurs when an insurer is not able to meet its financial obligations, or when the
liabilities of an insurer exceed its assets. The current solvency regime by the NIC requires the
assets of non-life insurance companies at any point in time, to exceed their liabilities by at least
10% of net premium income in order to be technically solvent. In case of life companies, assets
must at least equal liabilities (Insurance Act, 2006, Act 724). The International Association of
4
University of Ghana http://ugspace.ug.edu.gh
Insurance Supervisors’ (IAIS) standard on solvency requires a solvency margin to be set to
provide a safety against certain events. In the effort to ensure that Ghanaian insurers are
governed by laws that meet international standards set by the IAIS, the NIC developed a new
solvency framework to replace the current one issued in 2008 which had the same solvency
requirements for life and non-life companies. According to the NIC, the previous framework was
inappropriate because of the differences in operation of life and non-life insurers (NIC, 2010).
The NIC requires both life and non-life insurance companies to have a financial solvency margin
of 50% or the minimum capital requirement whichever is higher and that the solvency margin
shall in no case be less than the minimum capital requirement.
Insurance  regulatory  standards  and  methodologies  are  fast  changing  all  over  the  world.
Methods of supervision of insurers are moving from the compliance and historical toward the
risk-based and the progressive methodologies. Previously, insurance companies were mandated
to comply with various requirements of laws or subordinate legislations. Experience has,
however, shown that mere compliance with rules and legislations does not mean that a company
is and will continue to be financially sound to honour its maturing obligations to policyholders.
This new solvency framework which was to be implemented in January 2012 is in line with
international standards, that of the EU-based regulatory body (Hrechaniuk, Lutz & Talavera,
2007). Unlike the previous solvency framework, the new solvency II framework required an
increase in the minimum level of reserve for insurers to ensure that insurers are financially
healthy. The main aims of the new risk-based supervision approach are to specifically reduce
major risks faced by insurers which include inability to meet financial obligations, unnecessary
delay of claims payment, reduce the risk of incurring loss by policyholders when claim payment
5
University of Ghana http://ugspace.ug.edu.gh
is not paid and to provide a good monitoring system for regulators in order to detect early signs
of insolvency (NIC, 2010).  The new regulation seeks to compel insurers to do business
according to their liquidity ratios - the amount of their assets that could easily be used to defray
debts where necessary.
Due to the immense contribution of insurance to both individuals and firms, as well as its
contribution to economic development as whole, regulators are required to design a system that
can identify insurers which are in or heading towards insolvency. Aside the ratios, solvency
control measures have been put in place to enable the NIC monitor insurers in order to detect
emerging insolvencies before they degenerate into crises. These control mechanisms have been
classified in the form of a traffic light system as seen in table 1.1.
Table 1.1: Solvency Regime under the Insurance Act, 2006, Act 724
Level Margin Corrective Action
Green If asset are 150% or more of Routine Monitoring
liabilities.
If assets are more than 125% Restructure investments
of liabilities but less than
Amber 150% thereof.
Red If assets are at least equal to Enforcement Action/capital
liabilities but less than 125% injection
of liabilities.
Black If liabilities are more than Suspend license with the
assets. possibility of final license
withdrawal and liquidation
Source: Insurance Act (2006)
6
University of Ghana http://ugspace.ug.edu.gh
For every organization to remain financially solid, it has to consider certain major factors that
determine its financial performance. The major factors that significantly affect general insurers’
financial health in growing economies are firm size, investment performance, underwriting
result, liquidity ratio, combined ratio, and operating margin, premium growth, and growth rate of
surplus. Again, some macro determinants of life insurers’ financial health exist (Chen & Wong,
2004). These include interest rate, GDP and competition. The Ghanaian life insurance industry
has experienced keen competition over the past decade as a result of the influx of foreign
companies into the market and the emergence of new firms. Reforms and regulations set in place
by the NIC could also be argued as a contributor of this competition. An example is the recent
increase in minimum reserve level by the NIC from one million US dollars to five million dollars
and the separation of companies to life and non-life.
1.2 Research Problem
According to the world business dictionary, financial health is defined as a way in which to
measure the overall financial aspect of an individual or an organization that includes the amount
of assets they own and how much income they must pay out to cover regular and other expenses.
It is critical to investigate financial status of life insurance companies in Ghana mainly because
operation and management of life and general insurers vary vastly and they are faced with
different risks. In Ghana, the financial stability of the industry is mostly assessed by the NIC
through financial ratios such as the claims ratio, the retention ratio, underwriting profit and
investment income ratios. Aside these ratios used by the NIC to measure performance of
insurers, other factors that affect life insurers’ financial health is firm size (total assets own by a
firm) and age (how long the company has been operating) which are also a major determinants
(Chen & Wong, 2004). Research into how these factors affect insurer performance in Ghana has
7
University of Ghana http://ugspace.ug.edu.gh
not been properly investigated especially in the life insurance sector. Variables used to measure
firm size include total premium, total admitted assets, capital and surplus. Firm size proxies like
market cap and age can be mechanically correlated. Large insurance companies have a lower
probability of becoming insolvent (Cummins, Harrington & Klein, 1995). Regulators are
therefore less likely to liquidate such insurance companies.
Research on the Ghanaian insurance industry exhibit a trend which is quite different, in that,
whereas insurance premiums worldwide declined from 2.5% in 2012 to 1.4% in 2013 (Swiss Re
Sigma, 2013), the Ghanaian insurance market continuously experienced a rise in premiums
(NIC, 2010). This is evident as the industry begun to witness some improvement especially in
2010. It witnessed a 21% reduction in its claims and managerial expenses in 2010. Although
there has been restructuring in the Ghanaian insurance industry and increase in premiums over a
seven year period (2007-2013), it can be observed that some life insurance companies seem to be
more financially “sound” than others.  The factors and determinants underlying this trend are yet
to be ascertained.
The significant growth of the life sector over the past few years is therefore evident and cannot
be overlooked.
Table 1.2: Key Growth indicators of the Life insurance Industry (2011- 2012).
Indicator 2011 (GH₵’ M) 2012 (GH₵’ M) Growth Rate (%)
Premium Income 270.2 355.8 32%
Total Assets 492 669.8 36%
Total Investments 371.1 537.7 45%
Actuarial Liabilities 345.8 464.9 34%
Source: KPMG (2014)
8
University of Ghana http://ugspace.ug.edu.gh
As with most developing countries that have pursued economic and structural reforms, Ghana
has been undergoing a process of financial sector restructuring and transformation as an integral
part of a comprehensive program for some time (BOG, 2007). The financial sector reform is a
critical part of the country’s strategy to achieving emerging market status. In this regard, several
reforms are taking place to enable the financial sector to reach world first class services within a
globalized financial system. Also, the recent increase in the minimum level of reserve for
insurers by the NIC has had significant impact on the financial health of insures. These and many
other restructuring and reform policies in the Ghanaian insurance industry including intense
competition as more insurers enter the market have brought about the need for more research
into the field. There is therefore the need for research into how these structural reforms are
affecting the financial health of insurers.
Insurer insolvency is common in Africa and other developing countries and causes huge
financial loss to economies and policyholders (Barros & Obijiaku, 2007). Quite a large number
of stakeholders including the state go into distress when an insurer is declared insolvent.
Thousands of policyholders suddenly find themselves with some very serious problems
(Cummins, Harrington & Klein, 1995). This calls for the need for periodic monitoring and
evaluation of the financial condition of insurance companies by regulators, investors, and insurer
management. In addition to setting regulatory standards and assessment of annual financial
reports to ensure that insurers are financially sound, the NIC embarks on a number of
supervisory processes ranging from quantitative to qualitative assessments of the insurance
companies (e.g. market analysis report, financial analysis and others, NIC Supervision Dept.,
2007). An insurer seemingly performing financially well could go insolvent at any time if its
financial status is not carefully monitored.
9
University of Ghana http://ugspace.ug.edu.gh
Currently, there are twenty one registered life insurance companies in Ghana (NIC, 2014). Life
insurers in Ghana face challenges that are not very different from those experienced across the
globe. Although a greater percentage of Ghana’s total industry premiums are generated from the
non-life sector, the percentage growth in premiums from life assurers far outweighs that of the
non-life sector. For instance, according to a report by NIC, whereas the non-life sector grew by
23 percent in 2010, the life business recorded a remarkable growth of 52 percent. In other words,
the growth rate of the life sector is more than twice that of the non-life business (NIC, 2010). The
present study focuses only on factors determining the financial well-being of life insurers. This is
because life insurers differ greatly from general insurers in terms of operations, investment
activities, vulnerabilities, and duration of liabilities from general insurers (Brockett, Cooper,
Golden, Rousseau & Wang, 2004). Life insurers play the role of “financial intermediaries” while
general insurers functions as “risk takers.
The growing competition in the insurance industry has exposed it to harmful practices such as
undercutting, unethical underwriting and marketing practices and the excessive dependence on
credit to manage the competition. These problems lead to the inability of insurers to pay claims
and meet other liabilities, which makes clients frustrated when claims are due. These and other
factors bring up the need for the monitoring of financial health of Ghanaian insurers in order to
detect insolvencies.
1.3 Objectives of the Study
The main aim of the study is to assess and highlight the determinants of the financial soundness
of life insurance companies in Ghana. Related to the main aim of the study are the following
specific objectives:
10
University of Ghana http://ugspace.ug.edu.gh
 To determine the variables (both external and internal) that affect the survival and failure
of life insurance firms.
 To investigate the key predictors of financial health of life insurance firms in Ghana.
 To provide an early warning system on the probability of survival and failure which
classify life insurance firms into survived (strong) and failed (moderate, weak and
insolvent).
1.4 Research Questions
The following research questions will be deliberated upon in order to adequately cover the
subject matter of the study. Answers to these questions will fulfill the objectives of the study.
 What are the major determinants of life insurer financial health and is there a systematic
relationship between the firm specific factors (financial ratios) of a Ghanaian life insurer
and its financial health?
 What are the effects of firm specific factors on the financial health of a Ghanaian life
insurer?
 Can the micro and macro determinants of the life insurance industry in Ghana predict
their financial health and serve as an early warning system for the NIC?
1.5 Significance of the Study
This study will  be of value to future research, practice and policy making. It will serve as an
added contribution to existing work of several other researchers who have discussed issues on
the subject area. The findings from this study could also inform the work of the National
Insurance Comission about how reforms are impacting the finances of life insures.
11
University of Ghana http://ugspace.ug.edu.gh
Finally, the study will to contribute to existing knowledge, serving as a build-up on that of
Akotey et al. (2013) and Simpson & Damoah (2009) by looking at how solvency regulation by
the NIC which includes increase the minimum capital requirement of insurers and many others
may have impacted on the financial health of life insurers in Ghana. It can also be a basis for
stakeholders to control the activities of the insurance industry to help grow its financial standing.
1.6 Scope of the Study
This research focused on the life assurance industry in Ghana. The whole study lasted for a one
year period and covers a duration of seven years; from 2007 to 2013. The research examined
financial statements of 16 life insurance companies. This sample was selected based on the fact
that these companies have existed in the industry for at least five years between the period of
2007 to 2013 as used in the study, have reliable data and as such are in the best position to
provide relevant information.
1.7 Organisation of the Study
The present study is orginised into five chapters. Chapter one is the introduction of the study, it
described the research background, research problem, objectives, significance and scope of
study. Chapter two reviews the literature on life insurance. The chapter provides an overview of
the research on  life insurance industry in a global context and its performance in Ghana. Chapter
three outlines the research methodology. It includes the population, sample, data collection,
methods and models. Chapter four entails the analyses and interpretation of results. Chapter five
gives the conclusion, summary of research, discussion and its implications for future research.
12
University of Ghana http://ugspace.ug.edu.gh
CHAPTER TWO
LITERATURE REVIEW
2.1 Introduction
This chapter presents a review of relevant literature on insurer financial health and insolvency
and the factors related to these outcomes. The chapter provides a review of theoretical and
empirical literature on insurance, history of insurance in Ghana, the importance of insurance,
financial health and solvency of insurers in the Ghanaian life insurance industry. A theoretical
review will be done alongside an empirical review concerning these themes.
The review is grouped into twelve main sections. The first section gives a general overview of
insurance and how life insurance operation varies from general insurance. In the second and third
sections, the importance and contribution of insurance to individuals and to economies and the
need for the financial standing of insurers to be monitored are discussed respectively. The next
section give an overview of existing literature. The eighth and ninth sections review key
variables and predictors of financial health of insurance companies and early warning systems
currently used to monitor and prevent future insolvencies. Sections ten and eleven describe the
panel data analysis technique used for the study. The final section gives a conclusion of the
chapter.
2.2 Concept of Insurance
The concept of insurance is an ancient practice, which came into being over 5000 years ago, in
China where piracy was so common that seamen had to develop strategies to reduce losses
(Smith, 2004). As a way of disseminating the risks faced, a number of ships would carry a
portion of another ship's cargo so that if one ship was captured, the entire shipment would not be
lost (Smith, 2004). Insurance was therefore used as a preventive measure against piracy on the
13
University of Ghana http://ugspace.ug.edu.gh
sea. Well-established systems of insurance, which takes into account the level of risk faced by an
individual, the probability of occurrence of the risk and other factors were later developed (Berg,
2010). Several types of insurance were then later designed to cater for different kinds of risks
that individuals and organizations are liable to.
Several definitions of insurance exist. For example, Htay and Salman (2013) defined insurance
as the equitable transfer of the risk of a loss, from one entity to another, in exchange for a
premium, and can be thought of as a guaranteed small loss to prevent a large, possibly
devastating loss. Essvale Corporation Limited (2010), also define insurance as a promise of
compensation for specific potential future losses in exchange for a periodic payment.
Insurance is designed to protect the financial well-being of an individual, company or other
entity in the time of unexpected loss. Schrayer & Rowh (1999) also defined insurance as a
financial arrangement that redistributes the costs of unexpected losses, as well as a contractual
agreement in which one party agrees to compensate another party for losses.
Apparently, the concept of an insurance business is pretty forthright. An insurance firm pools
together premiums that customers pay to offset the risk of loss. This risk of loss can apply to
many different areas, which explains why health, life, property and casualty insurers exist. The
challenging part of being an insurer is properly estimating what future insurance claims will be
and setting premiums at a level that will cover these claims, as well as leave an ample profit for
shareholders.
Outside the above core insurance operations, insurers run and manage investment portfolios. The
funds for these portfolios come from reinvesting profits including earned premiums, where the
premium is kept because no claim occurred during the policy's duration and from premiums
before they get paid out as claims. These operations of insurers imply that they must be
14
University of Ghana http://ugspace.ug.edu.gh
financially sound in other to meet their future obligations to both insureds and shareholders.
They are expected to take wise investment decisions in order to yield profit from the funds they
handle.
Life insurance can simply be explained as a system or an agreement between two parties in
which the insured makes regular payments to an insurance company (insurer) in exchange for
a fixed amount of money that will be paid to the insured when he reaches a certain age  or paid to
someone named by the insured, usually  a family member, when the insured dies. Life insurance
has become an increasingly important type of insurance and an integral part of the financial
sector over the past 40 years, providing a range of financial services for consumers and
becoming a major source of investment in the capital market. Beck & Webb (2003) sampled 13
countries and studied the total assets of life insurance companies in the sample. It was observed
that the total assets of these life companies accounted for 11 percent of GDP between 1980 and
1985 but increased tremendously to 28 percent between 1995 and 1997 for the same countries.
Again, the insurance penetration of life insurance of 19 sampled countries showed a significant
increase from 1.2 percent between 1960 and 1965 to 4.2 percent between 1996 and 2000. These
confirm the increasing contribution of life insurance toward GDP and economic development
over the years. The life insurance industry plays an important role in the economic development
of Ghana in terms of long-term investment, savings mobilization, and contribution to the gross
domestic product (GDP). In general insurance, compensation is usually proportionate to the loss
incurred, whereas in life insurance a fixed sum is usually paid at the end of the contract.
15
University of Ghana http://ugspace.ug.edu.gh
2.3 Context of the Study
2.3.1 General Overview of Insurance in Ghana
Insurance in Ghana has its history dating back to 1924, when the first insurance company was
started by Europeans, to cover only Europeans (Ansah-Adu et al., 2011). The industry in Ghana
has since experienced substantial growth in terms of numbers, life insurance premium, number
of policies and distribution network. The insurance industry has also undergone several reforms
since its inception during pre- independence era. Reforms may be defined as changes in
regulation. It is regulated under the Insurance Law, 1989 (PNDC L227) as amended by the
Insurance (Amendment Laws 1991 and 1993 PNDC L’s 260 and 316) (NIC 2005). The industry
is currently governed by the Insurance Act 2006; Act 724. This Act complies significantly with
the International Association of Insurance Supervisors’ (IAIS) core principles and gives better
regulatory powers to the National Insurance Commission.
In lieu with its objective of complying with international standards (International Association of
Insurance Supervisors (IAIS)), core principles and to provide a level ground operation for all
companies, the Insurance Law, 2006 among other things prohibits the operation of composite
insurance. The Ghanaian insurance system therefore does not support provision of universal type
financial system where insurers can provide composite insurance. All composite insurers were
consequently mandated by the NIC to separate their life and non-life operations into distinct
companies by December, 2007. Theoretical underpinnings of the insurance separation,
consolidation, mergers and acquisition debate in the insurance industry have been grounded on
theories such as the resource based theory which was propounded by Penrose (1959). This theory
hinges on the financial benefits of diversification to a firm.
16
University of Ghana http://ugspace.ug.edu.gh
Over the past few years, the growth in the Ghanaian insurance industry has not been so
spectacular. Gross premium income for non-life business increased from GH¢91.6million in the
year 2005 to GH¢220.7 million in 2009. The growth rate for 2006 was 25% which fell to 23.9%
in 2007. This picked up sharply to 31.6% in 2008 but dropped to 18% in 2009 (NIC, 2009). As
of 2008, insurance penetration in Ghana was 1.6% which is still very low compared to that of
other African countries like South Africa which is 12.6%. However, Ghana’s oil find and the
effective implementation of the compulsory insurance of commercial buildings may have
contributed to a positive growth in 2011 as non-life insurance premiums rose from
GH¢142million in 2007 to GH¢358million in 2011 (NIC, 2011).
Despite all these, growth in the life sector has been consistently high. Ghana’s life insurance
market has witnessed tremendous growth in terms of number of companies, sophistication and
reach. Most Ghanaian companies are now selling universal life products which have a
combination of life cover and investment. A recent financial sector assessment conducted by the
IMF on the insurance industry reported that even though the number of insurance companies in
Ghana grew by more than 200 percent between 2005 and 2009, insurance penetration only
improved slightly from 1.16 percent to about 1.6 percent over the same period (Ghana Life
Insurance Holds Brokers Forum, 2010).
The relatively higher growth rate in the life sector can be attributed to the emergence of
innovative products on the market, the increasing popularity of micro insurance especially credit
life and the utilization of innovative distribution channels. A tremendous growth in the life
assurance market has been recorded over a period of time. The life sector experienced a steady
rise in premium income, increasing from ¢52.1 billion in 2001 to ¢312.5 billion in 2005. This
continuous increase in growth could be an indication of the increase in public awareness and
17
University of Ghana http://ugspace.ug.edu.gh
confidence in the insurance products of the life sector. It could also be attributed to the decrease
in level of interest rates and inflation in the Ghanaian economy over this period. Companies have
developed new universal life products which provide a hedge against inflation and suit the needs
of the insureds (NIC, 2005).
Premium income for life business increased from GH¢31.2million in 2005 to GH¢122.2million
in 2009. The growth rate however dropped from 36.1% to 34% in 2009. But as compared to the
non-life sector, life experienced a much higher growth rate. The growth in the life sector
increased the market share of life business to 35.7% (NIC, 2009). It must be noted that for the
first time life business has more market share than motor insurance which has always been in
lead because it is mandatory by law. The Ghanaian life insurance market is consolidated with the
top five companies accounting for about 80% share in 2012 up from 78% in 2011 (NIC, 2011).
In 2012, the life insurance gross written premium (GWP) stood at GH¢355.8 million, witnessing
a growth of 31.7%. Ghana’s top three life insurers’ performance in 2012 is summarized as
follows; SIC Life Insurance collected GH¢100.3 million representing a premium growth of 39%.
SIC is listed on the Ghana Stock Exchange and owned by government (40%) and others
including institutions and individuals (60%).  Enterprise Life Assurance reported total premium
collection of GHS89.1 million, representing a growth of 45%. GLICO Life Insurance reported a
gross premium of GHS36 million witnessing 4.65% growth.
2.3.2 Role of the NIC
In Ghana, the National Insurance Commission (NIC), which is sole regulator of insurance and
insurance companies was established by the Insurance Act, 2006 (Act 724), which replaced
Insurance Law, 1989 (P.N.D.C.L. 227). It is responsible for licensing, setting standards, and
approving insurance premium and commission rates. The objective of the Commission is to
18
University of Ghana http://ugspace.ug.edu.gh
ensure effective administration, supervision, regulation and control of the business of insurance
in Ghana and to ensure that insurers are financially sound in order to honour their obligations
towards policyholders.
In addition to setting regulatory standards and assessment of annual financial reports to ensure
that insurers are financially sound, the NIC embarks on a number of supervisory processes
ranging from quantitative to qualitative assessments of the insurance companies (e.g. market
analysis report, financial analysis and others, NIC Supervision Dept., 2007).   An insurer
seemingly performing financially well could go insolvent at any time if its financial status is not
carefully monitored. For the above reason, the evaluation of insurers generally involves only
statistical information or a mixture of quantitative and qualitative information (e.g. Swiss Re
Sigma, 2003; Chen & Wong, 2004; Das & Gosh, 2006).  When an insurer becomes insolvent or
financially impaired, state regulators must take appropriate actions such as license revocation,
cease-and-desist orders, and other actions that restrict an insurer’s freedom to do business
(Rejda, 1992).
Another of its major responsibilities is to set in place regulations in order to protect the interest
of the public and that of the industry, this can be associated with the public and private interest
theory of regulation. It protects the interest of the public by setting regulations in place to
monitor the financial health of insurers in order to prevent huge losses and disappointments faced
by insureds (the public) in case of insolvency. An example is evident in the recent regulation by
the NIC which increased the minimum capital of insurers form 1 million to 5 million dollars
(NIC, 2011). As part of its duties to protect the interest of the industry, the NIC put in place the
“no insurance, no cover policy”. Gaffikin (2005) stated that, the rationale for regulation is to
19
University of Ghana http://ugspace.ug.edu.gh
coordinate economic activities by organizing industries and directing the behaviour of people in
an efficient manner, hence insurance businesses are regulated to prevent moral hazard.
Restructuring of insurance industry regulatory systems is a common practice in both developed
and developing countries (Caprio & Klingebiel, 1996). This is due to financial and economic
crisis often encountered (Williamson, 1994; Kroszner, 1999). These restructuring may help
organizations to overcome the problems or challenges of the old law. Regulation is a very
important way of monitoring insurers in order to assess their financial health and prevent future
insolvencies. This is in connection with the main aim of insurance which is to provide reliable
information and good services to stakeholders. The need for regulatory change in the insurance
industry has been addressed by several authors including Batten (1981), who asserts that the
main problem of insurance companies, specifically the life industries is their solvency control
since they operate with a high degree of trust (also Obersteadt, Bruning, Cude, Schwarcz &
Wilkonson, 2013).
In the Ghanaian insurance industry this has been met by requiring new entrants to submit a
business plan and meet the initial capital requirement the Act has stated. Furthermore through
regulation, the NIC mandated all composite insurance to be separated into Life and Non-life.
Demsetz & Lehn (1985) suggested that regulations assist firms to report better financial
performance. The distinct separation of insurance operation in Ghana can be seen to follow the
Conglomeration Hypothesis which hinges on diversification and forbids composite insurance
operation in Ghana (Cummins, Weiss, Xie & Zi, 2010).
2.3.3 Risk Based Supervision Approach
The development of risk-based frameworks follows the pattern of many innovations (Black,
2005b). Risk-based supervision is a new supervisory approach currently being adopted by most
20
University of Ghana http://ugspace.ug.edu.gh
regulatory authorities. Currently, it is in the process of being implemented, by many supervisory
authorities. It is mainly concerned with the capital requirement to ensure the financial stability of
financial organizations like banks and insurers in order to prevent insolvencies. In addition, risk-
based supervision concepts are embedded in the Basel Core principles for effective banking
supervision and are part of the IMF and World Bank's Financial Sector Assessment Programs
(FSAPs) of countries.
Regulators in a number of countries are increasingly developing “risk-based” strategies to
manage their resources, and their reputations as “risk-based regulators” have become much
lauded by regulatory reformers (Black & Baldwin, 2010). The UK for instance has fully
embraced risk-based regulation, at least at the level of exhortation. The extensive certification of
the risk-based regulation or supervision approach in the financial sector of many economies,
together with the experience of regulatory failure, brings to bare the need to consider how risk-
based regulators can accustom the logics of risk analyses to the complex problems and the
dynamics of regulation in practice. Before regulators can decide on adopting this approach, they
must first regulate in a way that is responsive to these five major elements:  regulated firms'
behaviour, attitude, and culture; regulation’s institutional environments, change, regulatory
performance and interactions of regulatory controls (Black & Baldwin, 2010).
In today’s fast moving and interconnected world, along with carrying out on-site and off-site
activities, supervisors need to be forward-looking, and develop plans for intervening early, if a
material problem surfaces in any of the insurance companies. In a quest to protect policyholders
and provide secure treatment on transactions, to upgrade the legal and regulatory framework for
the regulation and supervision of the insurance sector in Ghana, and to work in line with
internationally accepted standards of regulation, the NIC is putting measures in place to adopt
21
University of Ghana http://ugspace.ug.edu.gh
the Risk Based Supervision approach. The Commission has proposed to Government that a new
Insurance Act which follows the Risk Based Supervision Approach.
Considering the inappropriateness of the solvency framework issued in 2008 by NIC, a new
solvency framework with separate solvency requirements for life and non-life companies that
takes into accounts major risks the market faces was implemented in January 2012 (NIC, 2011).
The National Insurance Commission (NIC) Ghana, in the year 2011 in line with its core aim,
decided on a new solvency framework. This framework prohibits insurance companies with
weak financial standing, the ability to undertake bigger business transactions.
2.4 Review of Determinants of Financial Soundness of Insurance companies
This section reviews previous academic research on the financial health of insurance in Ghana
and worldwide. Due to the numerous benefits of insurance to individuals and its contribution to
economies, there has been extensive research on insurer financial health and insolvency
worldwide. Some of these studies focused on the efficiency or solvency of composite insurance
companies, life or non-life. Others focus on methods of evaluating financial health of insures as
well as factors that determine the financial health. The majority of the research has tended to
focus on general insurance because of its short-term nature of operation compared to life
insurance.
2.4.1 Studies in the Ghanaian Context
There is a pattern of growing interest in the financial system of Ghana due to the many structural
changes and adjustment it has undergone within the last decade. According to an IMF report in
May 2011, Ghana’s financial system especially the insurance industry has undergone rapid
growth and major structural transformation over the last decade, which has brought new
opportunities and risks (IMF, 2011). Authorities have been implementing reforms to strengthen
22
University of Ghana http://ugspace.ug.edu.gh
the regulatory and supervisory framework and financial infrastructure (NIC, 2010). Quite a large
number of stakeholders including the state go into distress when an insurer is declared insolvent.
Thousands of policyholders suddenly find themselves with some very serious problems. This
calls for the need for periodic monitoring and evaluation of the financial condition of insurance
companies by regulators, investors, and insurer management.
Unlike the developed countries, the determinants of the financial health of insurance companies
in Ghana has received little attention despite the growth in premiums, the continuous increase in
the number of insurers and the many structural reforms put in place by the NIC. Inadequate
research has been done on the efficiency and determinants of profitability and financial status of
Ghanaian insurers. This study seeks to highlight the determinants of the financial health of life
insurers in Ghana in this era of reforms and growth in premiums and profits in order to enable
the NIC take measures to prevent insurer insolvency and to ascertain the possible reasons for the
low insurance penetration despite the seeming growth in the industry.
Generally, it can be argued that there are not much empirical efficiency studies on the financial
performance of life insurance in Ghana. This lack of research is probably the result of the
unavailability of data and the unpopularity of life insurance in Ghana until recently. There was
also the issue of unethical behavior of insurers concerning sale of insurance products and claims
payment. Kasturi (2006) argued that the performance of insurance companies in financial terms
is normally expressed in net premium earned, profitability from underwriting activities, annual
turnover, return on investment and return on equity. These measures can be classified as profit
performance measures and investment performance measures.
Akotey, Sackey, Amoah & Frimpong-Manso (2013), studied the relationship among three
measures of insurers’ profitability, namely; investment income, underwriting profit and overall
23
University of Ghana http://ugspace.ug.edu.gh
net profit. The study analysed the financial statements of ten sampled life insurers. They
concluded that gross written premiums have a positive relationship with insurers’ sales
profitability, but a negative relationship with investment income. Their study, however, focused
mainly on measuring performance and major determinants of profitability not determinants of
financial health of life insurers as a whole. Another study by Owusu Ansah, Dontwi, Seidu,
Abdulai & Sebil (2010) discussed the efficiencies of insurers using the DEA approach to
evaluate the relative performance of Ghanaian general insurance companies. They employed
debt capital, equity capital and management as inputs used by insurers to produce premium,
claims and investment income. The result from their study showed that Ghanaian general
insurers with higher dimension and market shares tend to have higher efficiencies. Boadi, Antwi
& Lartey (2013) used panel method and ordinary least squared regression to examine secondary
data from financial reports of sixteen insurance firms in Ghana. They concluded that apart from
tangibility which is negatively related, there is a positive relationship between leverage, liquidity
and profitability of insurance firms in Ghana.
Some previous studies have also focused on methods of evaluating insurer solvency. Several
models are adopted worldwide to assess the financial health of insures, but the most suitable
method for a particular economy depends on the nature of the market in which it operates and
some financial characteristics of the industry. According to Locke, S. (2006), Altman developed
a method of measuring a firm’s financial health called the Altman’s Z-Score Method. It is a
multivariate formula to measure the financial health of a company on whether it will enter into
bankruptcy in the coming two years. The Altman’s Z-Score Method, like other methods of
evaluating financial health also makes use of financial ratios of firms. This method uses five
common business ratios: earnings before interest and tax (ebit)/total assets ratio; sales/total assets
24
University of Ghana http://ugspace.ug.edu.gh
ratio; market value of equity/market value of total liabilities; working capital/total asset ratio and
retained earnings/total assets.
Although recommendations as well as conclusions from past studies concerning the financial
health on insurance companies are mixed, results from some studies like Simpson & Damoah
(2009) argue that the model used currently by NIC to assess the financial health of non-life
insurance companies in Ghana is not comprehensive enough to give early warnings to the
industry’s stakeholders and that the CARAMELS model, as proposed by other authors do not fit
the Ghanaian context. The aim of the present study, however, is to investigate the determinants
of financial health and not to evaluate financial health per se.
2.4.2 Studies in Africa and Other Parts of the World
In a global context, insolvencies of insurance companies and other financial institutions have
shown up, this has led to many financial regulators setting up measures to control it. In England
& Wales, for example, measures including the compulsory liquidations and creditors’ voluntary
liquidations and other methods are being applied to curb numerous corporate financial
insolvencies (Pessou & Reenen, 2014). Again, Palande, Shah & Lunawat (2003) studied
insolvencies in the US insurance market and recorded about 300 insolvencies of insurance
companies over a period of 25 years. Also, researchers in the USA identified three major
techniques that were used by the National Association of Insurance Commissioners (NAIC) to
predict insurer insolvencies (Best, 2007).
Insurer insolvency and assessment research has gotten lots of attention from researchers globally
since it is not as straightforward as that of other financial institutions. This is mainly due to the
fact that the unique format of insurance companies’ financials does not lend itself to traditional
25
University of Ghana http://ugspace.ug.edu.gh
financial accounting analysis. Again, assessing financial soundness in the insurance industry is a
complex task since the overall financial position of an insurance company depends on many
factors, some of which are difficult to quantify (Simpson & Damoah, 2009). In line with most
previous research that focused on solvencies using companies’ balance sheets and other financial
characteristics (Pistelis, 2008; Pist elis, 2009; Gatzert & Martin, 2012), this study adopts the
use of financial characteristics to determine the effects of the solvency position of the life
insurance companies. In so doing, the adequacy of the funds a company requires is determined in
order to remain healthy and avoid insolvency.
Most insurance industries in Africa were established during colonial times, the insurance laws
and regulations which provided the framework for regulation and supervision were usually based
on the laws and regulations of the colonial power (Ombudsman for short-term insurance).
Insurance regulation in most African countries has to a large extent been compliance-based,
deriving its powers and protection from the various legislative frameworks of the respective
countries. Various regulatory bodies worldwide have been working hard to achieve effective
regulation but are faced with certain challenges. Ayee, J. (2013) identified the major challenges
faced by regulators on the African continent as; financial dependence of regulatory bodies, weak
financial strength of insurance companies, inadequate expertise and capacity of regulators, lack
of cooperation among regulatory bodies, inadequate disclosure of information by insurers to
stakeholders, lack of strict application of the provisions of the law, emphasis on minimum capital
requirements irrespective of risk levels, cross border transactions by insurance companies and
emerging trend of group structures. However, the challenge that has the most adverse effect on
policyholders and economies has been identified as the problem of weak financial health
(Brockett, Cooper, Golden & Pitaktong, 1994; Browne, Carson & Hoyt, 1999; Grosen &
26
University of Ghana http://ugspace.ug.edu.gh
Jorgensen, 2002). There is therefore the need for more research to be geared towards this
direction. Several other studies including the International Study on Insolvency Systems in the
Middle East and North Africa also confirms that insolvency among insurers is prevalent in North
Africa and the middle East than in developed countries, emphasizing on the need for better
monitoring of insurers in order to predict their financial soundness and suggested compliance
with international standards of regulation as a possible solution (Insol International, 2009).
2.5 Importance and Contribution of Insurance
The role of financial institutions and financial intermediaries in fostering the economic growth
by improving the efficiency of capital accumulation, encouraging savings and eventually
improving the productivity of the economy has been thoroughly established. Recent studies show
that the insurance industry can improve the economic growth through financial intermediation,
risk aversion and generating employment (Ghosh, 2013).
The importance of insurance industries in the financial markets of emerging economies cannot
be overlooked (Haiss & Sümegi, 2008). One major reason why many countries remain poor is
the poor performance of their financial institutions and markets. The main purpose of insurance
industries earlier was to provide the mechanism for risk transfer.  Today, the sector also helps in
channeling funds in an appropriate way to support the business activities in an economy. Levine
(1997) and Haiss & Sümegi (2008) have shown that, insurance contributes to the economic
growth of the country. Insurance provides stability in all emerging economies by enhancing the
finances of countries where it is properly managed.
A well-developed insurance industry enhances economic development by providing long-term
resources for substructure expansion of every economy (Charumathi, 2012; Akotey et al., 2013).
27
University of Ghana http://ugspace.ug.edu.gh
It offers various financial security products to protect individuals and firms, providing a sense of
security for organizations which operate in risky businesses by helping to reduce risks
encountered during business operations (Brown & Churchill, 1999; Bikker & Leuvensteijn,
2008; Rao & Srinivasulu, 2013). Also, net premium earned which is considered as an output in
most insurance efficiency literature are the revenues earned by a company (Mandal & Dastidar,
2014; Huang & Eling, 2013). These revenues earned are the means through which the insurance
sector contributes towards economic development.
Life policies serve as financial support and provide a savings fund for retirement income and
offers a means of accumulating savings while providing for one’s family. A comprehensive life
insurance policy provides financial support to the dependents of an insured during trying
moments and in periods of death. Life insurance also serves as a source of retirement income or
savings package for an insured. Several life policies like the funeral policy and life education
policy have been designed by life insurers to suit the unique needs of its insureds.  According to
Vane (1973), life insurance also serves as a medium through which insured can accumulate
savings while providing for their families.  These savings are pooled by insurance companies and
injected back into the financia‐l bloodstream of the economy in the form of investment.Ghosh (2013) used the VAR VECM model to find out the long run and short run relationship
between life insurance growth and economic growth along with Granger causality test to suggest
any causal relationship. It was observed at the end of his study that there is long term relationship
between life insurance industry and economic development in India, the Granger causality test
suggested that life insurance sector improves the overall economic development in India. Again,
several researchers, using the autoregressive distributed lag (ARDL) bounds approach to
cointegration, have noted a long-run positive relationship between insurance penetration and
28
University of Ghana http://ugspace.ug.edu.gh
economic growth (Pesaran & Shin 1996; Pesaran, Shin & Smith, 2001). This means that funds
mobilized from insurance business have a long run impact on economic growth.
2.6 Financial Health/ Solvency of Insurers
Financial health of an organization in general can be defined as the ability of the business to
utilize its assets in order to generate revenues in an efficient manner. In the insurance industry, it
can be viewed as the ability of an insurer to meet its obligations (pay claims) when due
(solvency) and also manage its internal affairs efficiently without going bankrupt. Bose (2006)
explains the prediction of financial health of a company as similar to the problem of predicting
bankruptcy.
The financial well-being of financial institutions is one of the most rapidly growing streams of
literature and the insurance sector in particular has seen extreme growth in the number of studies.
Review of insurance literature shows extant studies in insurer insolvency. This can be attributed
to the unique format of insurance companies’ financials which does not lend itself to traditional
financial accounting analysis.
2.7 Empirical Findings
The majority of studies that emphasize monitoring or predicting the financial health of insurers
have focused on general or nonlife insurance (Chen, 1999; Shiu, 2004; Hrechaniuk, Lutz &
Talavera, 2007; Simpson & Damoah 2009; Pervan & Pavic, 2010; Malik, 2011; Charumathi,
2012). The major theme across these studies focus on either the financial health, efficiency or the
contribution of insurance to economic development (Barros, Nektarios & Assaf 2010; Bikker &
Leuvensteijn, 2008). In their study to determine the association between firms' soundness and
regulatory policies in the insurance industry, Pasiouras & Gaginis (2013) over the period 2003–
2007 used an accounting based measure, called the Z-score to measure solvency. Their findings
29
University of Ghana http://ugspace.ug.edu.gh
were that, regulations related to both technical provisions and investments have an impact on
soundness of firms.
Due to the importance of insurance to every economy, extensive research has been done on the
determinants of the financial status of insurance companies, both life and non-life in the
developed world (Chen, 1999; Chen & Wong, 2004; Shiu, 2004; Hrechaniuk, Lutz & Talavera,
2007; Pervan & Pavic, 2010; Ahmed, Ahmed & Usman, 2011; Malik, 2011; Charumathi, 2012).
Efficiency studies have been conducted on insurance in Africa and other parts of the world (Hao,
2007; Chen, Powers & Qiu, 2009; Ansah-Adu, Andoh & Abor, 2012). Few studies, however,
have so far focused on life insurance, particularly in Ghana. Likewise other studies have been
conducted on the financial health and efficiency of both life and non-life industries in Asia,
Europe, USA and other countries (Petroni, 2000; Rao & Srinivasulu, 2013; Huang & Eling,
2013; Haiss, & Sümegi, 2008; Ahmed, Ahmed, & Usman, 2011). Little has, however, been done
on the financial health of insurers in Africa and particularly in Ghana.
In a cross-country study, Cummins, Rubio-Misas & Vencappa (2014) provided evidence on the
association between soundness and competition in 10 European Union (EU) life insurance
companies from 1999-2011. No evidence of any improvement in competition over the post-
deregulation period 1999-2011 was found. However, using an accounting base measure (Z-
score) indicated a positive relationship between competition and financial soundness.
Following Eduardo et al. (2014), this study will categorize firms into strong, moderate, weak and
insolvent. For a firm to be financially healthy, it has to be in the strong category and will require
only routine monitoring. The following table details the basics for categorization
30
University of Ghana http://ugspace.ug.edu.gh
Table: 2.1 Categorization of levels of solvency
CATEGORY SOLVENCY RATIO
Strong Greater than 2
Moderate Less than 2
Weak Close to 1
Insolvent Less than 1
Source: Author (2015)
2.8.1 Firm Specific Factors and Financial Health
Based on the resource-based theory of firms, a firm’s probability of survival depends on its
internal characteristics and resources (Thornhill & Amit, 2003). From these resources they
develop distinct capabilities to generate competitive advantage and adapt in market conditions
(Pérez & Castillejo, 2008). Consequently, the theory of public interest stated that the regulators
impose the regulatory framework in a way to prevent the market failure (Domas, 2003). The
financial health of insurers can be influenced by both internal/firm-specific factors and
external/industry-wide factors. Some studies focus on determining the effects of both factors on
the financial stability of insurers (Chen & Wong, 2004; McDonald, 1992). Chen & Wong (2004)
reviewed firm-specific and market factors that affect both life and non-life insurers separately.
The present study, however, focuses on the firm specific determinants of the financial health of
life insurers only.
Most of the firm specific factors that affect the financial soundness of an insurer are financial
ratios. A financial ratio is defined as a comparison between one bit of financial information and
another (Drake, 2013). A financial ratio may also be defined as a relative magnitude of two
selected numerical values taken from an enterprise's financial statements. These ratios are used to
31
University of Ghana http://ugspace.ug.edu.gh
try to evaluate the overall financial condition of an organization. They are mostly obtained from
the balanced sheets or financial statements of firms, in this context the annual reports by the NIC.
Some of the firm specific factors that affect an insurer’s financial health include financial ratios,
surplus growth, operating margin, liquidity ratio and combined ratio.  For the purposes of this
study, certain ratios were selected based on their significance in predicting insolvency from
previous literature including Chen & Wong (2004). The effects of the following ratios estimated
on the financial health of Ghanaian life insurers:
Firm size is mostly measured as the total assets of the insurer and sometimes as the percentage
of market share. BarrNiv & Hershbarger (1990) and Cummins, Harrington & Klein (1995)
studied the effect of firm size on the financial health of organizations. From their work, it was
observed that larger firms or organizations with more assets are less likely to be insolvent. This
is partially due to the fact that regulators are less likely to liquidate large insurers. Hence, larger
firms are likely to be more financially sound than smaller ones. Variables used to measure firm
size include total premium, total admitted assets, capital and surplus.
Operating margin is a firm specific variable that can affect the financial health of insurers. It is
a profitability ratio that measures what percentage of total revenues is made up by operating
income. In other words, it demonstrates how much revenue is left over after all the variable or
operating cost have been paid. A financially sound company is therefore supposed to have a
higher operating margin meaning that it makes enough income from its operating activities to
cater for its costs (liabilities). The operating margin ratio, which is sometimes called operating
profit margin, is the ratio of operating income to net sales. Kramer (1996) employed the use of
operating margin ratio to assess the financial soundness of a company. He found the relation
32
University of Ghana http://ugspace.ug.edu.gh
between operating margin and financial solidity to be positive. BarrNiv & Hershbarger (1990)
found operating margin to be the best single variable that can influence life insurers’ solvency.
Combined ratio is defined by the Economic Times as the sum of incurred losses and operating
expenses measured as a percentage of earned premium (www.economictimes.indiatimes.com,
accessed on 5th February 2015). It is a measure of the profitability of the insurer. It is the sum of
two ratios, one calculated by dividing incurred losses plus loss adjustment expense (LAE) by
earned premiums (the calendar year loss ratio), and the other calculated by dividing all other
expenses by either written or earned premiums. Used in both insurance and reinsurance, a
combined ratio below 100 percent is indicative of an underwriting profit. Combined ratio
measures incurred losses and expenses as a percentage of earned premiums. A ratio above 100%
means the insurance firm is losing money on its insurance operations. Below 100% suggests an
operating profit.
Liquidity ratio provides information on a company's ability to meet its short−term or immediate
obligations using assets that are most readily converted into cash. Assets that may be converted
into cash in a short period of time are referred to as liquid assets; they are listed in financial
statements as current assets. In insurance literature, liquidity ratio measures the capability of an
insurer to pay liabilities when they are due. The major liabilities of an insurer are claim payment
and cost of operation. The sources of liquidity for an insurer are mainly premiums, investment
income and liquidation of assets (Hampton, 1993). According to Dambolena & Khoury (1980),
the stability of the liquidity ratio is a necessary measure of financial solvency. Liquidity ratio is
calculated as a ratio of total assets to total liabilities.
Investment performance which refers to the effectiveness of investment decisions is also a very
significant determinant of financial health of insurers. Profitable investment decisions increase
33
University of Ghana http://ugspace.ug.edu.gh
the finances of an insurer. It has been found to have a negative correlation with solvency rate
(Kramer, 1996). In this study, it will be measured as the investment income. For life insurers,
funds (premiums) are collected in advance of paying benefits and held in reserves until claims
are paid (the intermediation service).
Premiums received by insurers may not be adequate to settle claims filed by insureds, hence
insurers are therefore expected to invest these premiums wisely in order to gain higher returns.
This income generated from the investment helps insurers to offset any incurred losses. Studies
like Yang (2006) assess the impact of investment performance of insurers. Portfolio composition
of life insurance companies should mainly be long term in nature. It is evident that investment
management, portfolio composition and changes overtime is not only important to the survival of
any life assurance scheme but also to the attainment of the corporate mission and vision.
Premium growth is a measure of the rate of market penetration. We expect a growth in
premiums to affect the financial health of an insurer positively, since this signifies better
performance of an insurer. However, empirical results from previous research show that rapid
growth of premium volume is a major cause of insurer insolvency (Kim et al., 1996).
Claims incurred claims are undesirable to insurers, they therefore would like to incur less
claims (Owusu Ansah, 2010). The use of "claims paid" or "losses incurred" as a factor of
financial soundness has attracted criticism because an unexpected up-ward change in losses (due
to an environmental catastrophe or a terrorist attack) would be considered as an increased output
quantity and, therefore, would result in efficiency enhancement of the respective company.
34
University of Ghana http://ugspace.ug.edu.gh
Chen & Wong (2004) also established change in asset mix and change in product mix as firm
specific factors that have an effect on the financial soundness of life insurers due to the nature of
their operation.
2.8.2 Distribution System and Financial Health
Producers tend to get their products on the markets after production through numerous channels.
Distribution is a key determinant of success of all insurance companies. Insurance product
distribution is usually through independent agents, banks or insurance brokers. These insurance
intermediaries are useful both on the demand side and on the supply side. On the demand side,
these intermediaries provide useful information to policyholders about the type of cover of a
particular policy and reduce expenditure on information and transaction costs in insurance
markets (Eckardt & Räthke-Döppner, 2008), hence protecting their interest. On the other hand,
they provide insurers information about the client in order to provide coverage and minimize
adverse selection on the supply side. However, to ascertain the effect of insurance distribution
system on the financial health of insurers, only the supply side is considered. Park, Lee & Kang
(2009) examined the coexistence of multiple distribution systems in property-casualty insurance
industry in the USA from 1990 to 2001 using Stochastic Frontier Analysis. For an insurer to
remain financially sound, it must be able not only to make wise financial decisions, but to take
advantage of the best methods of distribution in order to increase its policyholders. The theory to
back this system of distribution can be linked to the transaction cost theory. This theory
according to Allen (1999) can be explained based on the neoclassical literature on property rights
theory. Bashir, Madhavaiah & Naik (2013) researched on the traditional and modern distribution
channels in India. They categorized distribution systems into traditional and modern as shown in
Table 2.2.
35
University of Ghana http://ugspace.ug.edu.gh
Table: 2.2 Classification of methods of insurance distribution
Traditional Channels Modern Channels Alternative Channels
Individual Agents Bancassurance Telcassurance
Corporate Agent Micro-Insurance Direct Internet Marketing
Broker Shopassurance
Work Site Marketing
Source: Bashir et al. (2013)
Bashir et al. (2013) studied attributes of these channels and also carried SWOT analysis for each
channel. They concluded that the methods of insurance distribution channel adopted by an
insurer have significant impact on its growth and financial wellbeing. In Ghana, insurance
products are distributed through agents, brokers, micro-finance institutions (MFIs) and mobile
networks. Individual agents are the key distribution channel for insurance in Ghana. According
to the NIC website, there are 4,537 licensed agents currently operating in the country. With the
introduction of Bancassurance as a new channel of distribution, significant improvement has
been recorded by both the life and non-life companies who have adopted its use. Looking
forward, corporate agents (banks and microfinance institutions) distribution channel,
bancassurance is expected to witness a high growth, driven by regulatory support and increased
customer base. Bancassurance is defined as an arrangement in which a bank and an insurance
company form a partnership so that the insurance company can sell its products to the bank's
36
University of Ghana http://ugspace.ug.edu.gh
client base. This partnership arrangement is usually profitable for both companies. Assessment of
the financial statement of the life companies in Ghana who have adopted the use of this
distribution system over the years shows an increase in their gross premiums, resulting in an
increase in their market share, which directly affects the financial health of insurers. Operation of
bancassurance in Ghana started in 2007 with few banks like Standard Chartered bank, Unibank
and insurers like Metropolitan insurance and Starlife Assurance but currently involves other
banks and insurers. As at September 2013, there were about 19 bancassurance collaborations
between commercial banks and insurance companies (KPMG, 2014). The choice of marketing
distribution channel, according to Klumpes (2004), can significantly impact on the finances of
insurance firms. The market imperfections hypothesis and the product quality hypothesis have
been used by researchers including Joskow (1973) and Kim et al. (1996) to explain the existence
of these systems.
2.8.3 External/Macroeconomic Determinants of Insurer’s Financial Health
Aside the firm specific determinants of life insurer stability, certain industry wide factors are
observed from previous literature to have an impact. These include reforms, regulations and
inflation.
Inflation is defined by the Economic Times as the percentage change in the value of the
Wholesale Price Index (WPI) on a year-on year basis. Basically, it can be explained as a general
increase in prices and fall in the purchasing value of money. Inflation can affect the value of
money or reserves by insurers over a period of time either negatively or positively. Insurers are
therefore to make wise decisions on investments in order to prevent any negative impact of
inflation on their reserves. Using a panel data aggregated at different frequencies for 68
economies from 1961–2000, Beck and Webb (2003) found that economic indicators—such as
37
University of Ghana http://ugspace.ug.edu.gh
inflation, income per capita, and banking sector development—and religious and institutional
indicators are the most robust predictors of life insurance financial soundness.
Reforms and regulation by insurance regulators can also impact the financial soundness of both
general and life insurers. Reforms may be undertaken owing to financial and economic crisis
encountered (Williamson, 1994; Krozner, 1999).
Regulators may authorize reforms and restructuring in order to help organizations to overcome
the problems or challenges of an old law. This could however turn out to have either a positive or
negative impact on the financial stability of insurers. In Ghana, many structural reforms and
regulations have taken place since the beginning of its operation. As part of the NIC’s efforts to
strengthen the regulation of insurance industry, the Commission introduced a new system of
supervising insurance companies known as the risk-based supervision system, the arrangement
will replace the existing compliance-based system and will make it possible for the regulator to
supervise companies based on the level of risk identified in a particular company. Another
typical example is the increase in minimum capital requirement for insurers by the NIC. This
greatly affected the operating and start-up capital of insurers, especially the newer smaller firms.
In this present study, reform will be captured in the regression as a dummy. Represented as 1 in
the years in which major structural reforms were implemented and zero otherwise.
2.9 Theoretical Backing
Insurance regulation can be seen to be in line with both the public and private interest theories.
The public-interest theory of regulation holds that regulation is designed to benefit the public by
ensuring the security and safety of the public and intervening when the private market fails to
allocate resources properly. In insurance regulation, the public interest theory assumes the
interest of the public is at the heart of regulators hence regulators step into the markets in an
38
University of Ghana http://ugspace.ug.edu.gh
effort to correct market failure or from some political crisis and maximize social welfare
(Kroszner, 2001; Hertog, 2010). The private interest theory also known as the capture theory of
regulation on the other hand states that regulation does not benefit the public interest but rather
individuals in the industry have incentives to implement laws that suit their interest. Regulators
have the aim of also ensuring that insurers remain in operation and that the insureds do not
benefit to the detriment of the insurers. Hence, protecting the private interest as well.
2.10 Definition and Concept of Regression
Regression analysis is a statistical tool for the investigation of relationships between variables.
Usually, the investigator seeks to ascertain the effect of one variable on another as, for example,
the effect of a price increase upon demand, or the effect of changes in the money supply upon the
inflation rate (Skyes, 2008). In this study, regression analysis will be used to determine the
relationship between certain independent variables, mainly financial ratios and the financial
health of life insurers (dependent variable). In other words, regression analysis will be used to
determine how financial ratios can predict the financial health of a life insurer.
2.11 Panel Regression
A panel is a cross-section or group of people or entities that are surveyed periodically over a
given time span. A panel data can also be defined as a cross-sectional data in which economic
entities are observed across time (Ahiawodzi & Sackey, 2010). The data used for this study is
panel data because it employs the process of collecting sample observations from a larger
population over a given time period.
Longitudinal or panel data track the same sample at different points in time. The sample can
consist of individuals, households or establishments; in this study, life insurers. In contrast,
repeated cross-sectional data, which also provides long-term data, gives the same survey to
39
University of Ghana http://ugspace.ug.edu.gh
different samples over time. Longitudinal data allow for the measurement of within-sample
change over time, enable the measurement of the duration of events, and record the timing of
various events.
Panel data have become widely available in both the developed and developing countries. In the
U.S. for example, two of the most prominent panel data sets are the National Longitudinal
Surveys of Labor Market Experience (NLS) and the University of Michigan’s Panel Study of
Income Dynamics (PSID). In a developing country like Ghana, an important source of panel data
is the Ghana Living Standards Survey (GLSS) collected by the Ghana Statistical service.
Panel data usually give a researcher a large number of data points, increasing the degrees of
freedom and reducing the collinearity among explanatory variables hence improving the
efficiency of econometric estimates. More importantly, longitudinal data allow a researcher to
analyze a number of important economic questions that cannot be addressed using cross-
sectional or time-series data sets. Since longitudinal data are collected over time, they also
provide researchers the ability to predict future values of a response for a specific subject. This
type of inference, known as forecasting, which is a special case of prediction. In this study, the
forecasting/predictive property of the longitudinal data used will enable us to determine the
possible financial health of an insurer in future. Two major disadvantages of panel data are
selectivity bias and heterogeneity bias.
2.11.1Fixed Effects Models (FE)
Fixed effects models control for, or partial out, the effects of time-invariant variables with time-
invariant effects. This is true whether the variable is explicitly measured or not. Fixed effect
model can be estimated using statistical techniques like unconditional maximum likelihood
40
University of Ghana http://ugspace.ug.edu.gh
(UML) and conditional maximum likelihood (CML). The estimation of linear regression models
when the effects of omitted individual-specific variables are treated as fixed constants over time.
In general, a fixed effect model could be written as
= μ + +
In panel regressions, we can assume that there is unobserved heterogeneity across individual
entities captured by the error term . If this unobserved individual-specific effects are correlated
with the regressors , then it is a fixed effect model. Where μ is the intercept which is allowed
to vary with time.
2.11.2 Random Effects Models (RE)
In a random effects model, the unobserved variables are assumed to be uncorrelated with (or,
more strongly, statistically independent of) all the observed variables. That is, the individual-
specific effect is a random variable that is uncorrelated with the explanatory variables.
Random effects models can be estimated using Generalized Least Squares (GLS). Under the
rand=om(  ef+fects +mode)l, each individual has the same slope parameters and a composite error termunlike under FE where each individual has the same slope parameters but
different intercept term.
Several considerations affect the choice between a fixed effects and a random effects model.
First, the nature of the variables that have been omitted from the model must be known. A
random effects model is best if there are no omitted variables – or if the omitted variables are
uncorrelated with the explanatory variables that are in the model. This will produce unbiased
estimates of the coefficients, use all the data available, and produce the smallest standard errors.
More likely, however, is that omitted variables produce at least some bias in the estimates.
41
University of Ghana http://ugspace.ug.edu.gh
Again, fixed effects models provide a means for controlling for omitted variable bias if there are
omitted variables which are correlated with the variables in the model. In a fixed-effects model,
subjects serve as their own controls. The idea is that whatever effects the omitted variables have
on the subject at one time, they will also have the same effect at a later time; hence their effects
will be constant, or “fixed.”  However, in order for this to be true, the omitted variables must
have time-invariant (the value of the variable does not change across time) values with time-
invariant effects (the variable has the same effect across time).
Tests for correlation of with independent variables are conducted to decide between random
effect and fixed effect framework.
H0: Cov ( , ) = 0 (no correlation between and ) - RE
H1: Cov ( , ) ≠ 0 (correlation between and ) – FE
represent the adjacent data points, say data for January and February for a particular
variable. For each variable, we want to know if RE and FE consider a particular variable derived
over time to be independently generated or not.
The Hausman test helps to assist in choosing between fixed and random effects.
Rejection of the null favours the fixed effect model.
2.11.3 Assumptions for Regression Analysis
 First, the errors are assumed to have an expected value of zero. This means that on
average the errors balance out.
 Second, it is assumed that the independent variables are non-random.
 Third, there is an assumption that the independent variables are linearly independent.
42
University of Ghana http://ugspace.ug.edu.gh
 Fourth, we assume that the disturbances are homoscedastic: This means that the variance
of the disturbance is the same for each observation.
 Fifth, we assume that the disturbances are not auto correlated: This means disturbances
associated with different observations are uncorrelated.
2.12 Conclusion
The present study departs from previous studies by investigating the determinants of the
financial soundness of life insurers in Ghana. It will make use of financial ratios published in the
annual reports of sampled life companies in Ghana as well as data on inflation from Bank of
Ghana. Again, this study will make use of the unbalanced panel regression analysis method using
data about the annual reports of recently registered life insurers as well as the old ones to
highlight the factors that affect financial stability of life insurers.
43
University of Ghana http://ugspace.ug.edu.gh
CHAPTER THREE
METHODOLOGY
3.1 Introduction
This chapter describes the method for the study. The chapter is divided into nine sections.
Section one gives an introduction to the chapter, section two describes the target population for
the study while section three describes the sample and sampling procedure. Sections four focus
the data used for the study. Section five, six and seven describe the model and the variable used.
Section eight specifies the regression model and section nine talks about the appropriateness of
the methodology used. The final section gives a conclusion of the chapter.
3.2 Target Population and Sample
The population of the study comprises all the 21 life insurance companies registered with the
National Insurance Commission in Ghana as of 2013. The sample is the selected portion of the
population that is being studied. Sixteen companies will be sampled out of the population of 21
based on how long they have been in operation or how long they have been registered with the
NIC. For an unbalanced regression data
≠
where is the number of entities in a year , and is the number of entities in another year.
That is, data for some periods might be missing. For this research, all entities (insurers) are not
observed in all the time periods. Sixteen life companies that were registered with the NIC
between the periods of 2007 to 2013 will be used for the regression model. Panel data analysis
will be employed because the data has both time series and cross sectional dimensions (the same
sample of life insurers will be studied over the period of time).
44
University of Ghana http://ugspace.ug.edu.gh
3.3 Sample and Sampling Procedure
The panel data used study constituted 16 life companies, observed over a seven year period. This
fits the context under which panel data can be employed. Also, an unbalanced panel was used
since some entities (financial ratios) are not observed for all the years. Panel data with missing
observations (unbalanced) have been studied by Wansbeek & Kapteyn (1989) and Baltagi &
Chang (1994). The use of this unbalanced panel data set allows a comprehensive evaluation of
the financial soundness of the Ghanaian life insurance industry, as it also includes firms that
entered or left the market during the sample period. This is because some of the life companies
were not in existence in 2007 where the study starts from. Panel data is best suited for this study
because it allows control for omitted variable bias and permits control for factors that are
unobserved and therefore cannot be included in a multiple regression.
Data on the market share, claims, premiums, combined ratio, operating margin, liquidity ratio
and investment performance of the sampled population were collected and analysed using an
unbalanced panel data analysis technique. As part of reforms to ensure effective operation of
insurers and protection of both the public and insurers, the NIC in 2007 passed a law which
mandated all insurers to distinguish their operation into two separate categories, life and non-life.
Operational data on life insurance solely, can therefore be obtained beginning 2007.
Consequently, data used for this study covers the year period from 2007 to 2013.
3.4 Data Collection
Secondary data comprising the market share, claims, firm size and premiums of the sampled life
companies will be collected and analyzed quantitatively. The data will be sourced from annual
reports and financial statements of life insurance companies from the NIC. Results will be
45
University of Ghana http://ugspace.ug.edu.gh
displayed diagrammatically as tables for easy understanding. This data comprises mainly
secondary data from the annual reports published by the NIC as well as other journals, and write-
ups that might have useful secondary information relevant to the study. Yearly average data on
inflation was also obtained directly from the bank of Ghana.
3.5 Model Specification and Description
3.5.1 Specified Model
The independent variables that are thought to affect the financial health of life insurers include
firm size (FZ), underwriting result (UR), investment performance (IP), liquidity ratio (LR),
premium growth (PG), surplus growth (SG), combined ratio (CR), operating margin (OM), GDP,
reforms and inflation. However, not all the variables were utilized in the regression model.
Variables used were selected according to their significance in determining financial
performance of insurers based on previous research.
This study employs firm size, investment performance, liquidity ratio, premium growth,
operating margin, combined ratio, claims incurred, reforms and inflation as input variables in
line with some of the variables used by Rai (1996); Owusu-Ansah et al. (2010) and Kader et al.
(2014).
It is worthwhile to start with the variables that are thought to have an association with the
response variable. This could be done by selecting variables that are intuitively appealing by
using common sense and experience or based on variables from previous research. At this stage
statistical analysis is not carried out. After a successful completion of the univariate analysis,
variables for the multivariate model are selected. Any variable whose univariate test has p-value
< 0.25 should be considered as likely candidate for the multivariate model. The use of 0.25 as a
46
University of Ghana http://ugspace.ug.edu.gh
screening criterion is based on works by Bendel & Afifi (1997), Mickey & Greenland (1989).
This is because the use of large p-value has the disadvantage of including variables that are of
questionable importance. The procedure for the selection of variables in the regression will be
forward selection (which involves starting with no variables in the model, and then trying out the
variables one by one and including them if they are statistically significant).
Variables for the Study
The study uses solvency ratio as a proxy for financial health, which is the dependent variable.
Again, following Worthington & Hurley (2002), Yang (2006) and Bikker & Leuvensteijn
(2008), this study assumes that, the Ghanaian insurance industry produces two outputs:
premiums and investment income based on the risk-pooling and intermediation services
provided. The intermediation approach developed by Sealey & Lindley (1977) on the other hand
views firms as financial intermediaries that manage a reservoir of assets considered as outputs
for the firm, borrow funds from policyholders which constitute the inputs to be used for
production and thereafter invest the funds.
3.6 Checking for Model Assumptions
To deem a linear regression as appropriate, the underlying assumptions must be examined.
Particularly the assumption of linearity, constant variance (homoscedasticity) and the normality
of the residuals must be examined.
3.6.1 Checking for linearity
The linearity of the regression model can be examined by observing a scatter plot of the
dependent variable against each independent variable. The scatter plot can help identify if a
47
University of Ghana http://ugspace.ug.edu.gh
linear trend exist. If a linear trend does not exist the data can then be transformed to make it
linear or the underlying relationship can be studied.
3.6.2 Checking for homoscedasticity
Homoscedasticity of the residuals implies that the errors have a constant variance. The constancy
of the variance of the dependent variable can be examined using various plots. A plot of the
residuals against any of the independent variables, or against the predicted values can help
unravel this assumption. Lagrange multiplier test for homoscedasticity can also be used to
examine this assumption.
3.7 Measurement of Dependent and Independent Variables
Table 3.1 summarizes the dependent and independent variables and provides how they were
measured for the purposes of this study.
48
University of Ghana http://ugspace.ug.edu.gh
Table 3.1 Summary of Dependent and Independent Variables
VARIABLE MEASUREMENT
SOLVENCY RATIO (SR) =
FIRM SIZE (SZ) Market share =
INVESTMENT PERFORMANCE Measured as investment income
(IP)
LIQUIDITY RATIO (LR) =
PREMIUM GROWTH (PG) Net premium will be used as a proxy for premium
growth.
COMBINED RATIO (CR) +
OPERATING MARGIN (OM) ( + ℎ )
CLAIMS INCURRED (CI) Total claims incurred in a year
INFLATION (INF) End of year value of inflation
REFORMS (RF) 1   if structural reform was implemented in year x
0   otherwise
Source: Author (2015)
49
University of Ghana http://ugspace.ug.edu.gh
To examine whether these variables impact the financial health of life insurers, the following
panel regression model will be employed:
= ++ + + + + + + +
= + + ,
where the dependent variable (FH), is financial health of insurers. Solvency ratio (SR) will be
used as a proxy for financial health.
Table 3.2 shows the expected signs of the variables used in the computation of insurer financial
health.
50
University of Ghana http://ugspace.ug.edu.gh
Table 3.2 Signs of Variables Used in Computation
Variable Expected Sign
Firm size +
Investment Performance +
Liquidity Ratio +
Premium Growth +
Claims incurred -
Combined Ratio -
Operating Margin +
Inflation -
Reforms +
Source: Author (2015)
3.7.1 Discussion of expected signs
Firm size is expected to have a positive relation with financial health. Larger firms are likely to
be more financially sound than smaller ones. This may be due to their relatively high reserve and
also because larger firms have more assets (BarrNiv & Hershbarger, 1990).
Liquidity ratio is expected to show a positive relation with financial health. The higher the ratio,
the easier a firm is able to meet its short term or immediate liabilities using assets that are easily
converted to cash, making it more financially sound.
A positive relation is expected between investment performance and financial health. In this
study, investment performance is measured as income from investment. An insurer with higher
returns on investment (income) is in a better position to settle its liabilities and hence is more
51
University of Ghana http://ugspace.ug.edu.gh
financially sound. The higher the income from investment, the more financially sound the firm
is.
It is expected that growth in premiums will affect the financial health of an insurer positively,
since this signifies better performance of an insurer.
Claims incurred is expected to be negatively related with financial health. The higher the claims
of a firm are, the more liabilities it has, making it less financially sound.
Combined Ratio is a measure of the profitability of the insurer.  It measures incurred losses and
expenses as a percentage of earned premiums. Combined ratio is expected to have a negative
relation with financial health. A ratio above 100% means the insurance firm is losing money on
its insurance operations. Below 100% suggests an operating profit. Therefore, the higher the
ratio, the less financially sound the firm is.
Financial health is expected to have a positive relation with operating margin. Operating margin
shows how much revenue is left over after all the variable or operating cost have been paid. A
financially sound company is therefore supposed to have a higher operating margin meaning that
it makes enough income from its operating activities to cater for its liabilities.
Inflation is expected to be negatively related with financial health. The higher the inflation, the
less financially sound an insurer is as it results in an increase in prices. High inflation results in
higher operating costs making a firm less financially sound. Insurers are therefore to make wise
decisions on investments in order to prevent any negative impact of inflation on their reserves.
Reforms are expected to have a positive relation with financial health. Insurance regulators make
changes to regulations and reforms in order to reduce insurer insolvency and to improve the
performance of insurance companies which then benefits both insurers and insureds.
52
University of Ghana http://ugspace.ug.edu.gh
3.7.2 Testing for the Significance of the Model and Its Parameters
After estimating the coefficients, there is the need to assess the significance of the variables in
the model. This usually involves formulation and testing statistical hypothesis to determine
whether the independent variable in the model is significantly related to the outcome variable.
3.8 Appropriateness of the Methodology
The advantages and limitations of using panel data sets have been discussed by Hsiao (1986) and
Baltagi (2001).
Panel data sets have some limitations which include bias due to sample selection problems. For
the purpose of this study, the sample was selected based on the registered life companies during
the particular years and based on their existence in the insurance market for at least seven years
since 2007. This helps to control for the sampling bias.
Although unbalanced panel data sets is the considered the most appropriate method for this
study, it comes with some limitations. The unbalanced panel is best suited for the study because
of the existence missing data points which arises as a result of late entry of some life insurers
into the industry during the study period. In panel data applications, testing for random
individual effects is of utmost significance. Ignoring these effects lead to huge bias in estimation
(Moulton, 1986). With unbalanced panels, testing for this becomes cumbersome. However, the
Breusch-Pagen test was applied in this study in order to test for these random effects.
The benefits include its ability to control for individual heterogeneity in order to restrict bias in
the individual estimates. Again, considering the informative nature of the data used for this
53
University of Ghana http://ugspace.ug.edu.gh
study, we can obtain more reliable estimates and test more sophisticated models with less
restrictive assumptions using a panel data technique. Furthermore, panel data sets are better able
to estimate effects that are not easily detected using pure cross-sectional data (Baltgai, 1998).
3.9 Conclusion
This chapter has provided an overview of panel data and its basic concepts as well as a detailed
description of the panel regression model that is used in the study. Furthermore, the data
selection procedure, population to be sampled and the advantages and disadvantages of panel
data were discussed. Finally, there was an exposition on the model to be used and the expected
signs of dependent variables (financial ratios) that are believed to influence the financial health
of a life insurer were displayed. The next chapter presents the findings of the study.
54
University of Ghana http://ugspace.ug.edu.gh
CHAPTER FOUR
DATA ANALYSIS, INTERPRETATION AND DISCUSSIONS
4.1 Introduction
This chapter presents the results of the study. The data used for the study consist of yearly data
obtained from the financial reports of sixteen life insurance companies. The variables that were
collected from the financial statements were Solvency Ratio (SR), Liquidity Ratio (LR),
Investment performance (IP), Claims Incurred (CI), Firm Size (SZ), Combined Ratio (CR),
Operating Margin (OM), Reforms (RF) and Premium Growth (PG). Data on inflation (INF) was
also gathered from Bank of Ghana. The chapter begins with a description of each data set using
summary statistics. The chapter then progresses with the ordinary least squares regression of all
the other variables regressed on SR. This is followed by the panel regression which is in two
distinct sections, the fixed effects panel and the random effect panel regression. The Hausman
test was then applied to help choose the most appropriate of the two panel regressions. Lastly,
the Hausman test is conducted and then the diagnostic test for the most appropriate model that is
chosen since panel regression must follow certain assumptions. R software was used for all the
analysis.
4.2 Descriptive statistics
Table 4.1 shows some descriptive statistics of the data used for the study. It displays the mean
and standard deviation of all the nine variables used in the study. For each variable, the mean and
standard deviations are computed collectively on a year-to-year basis in order to unveil the trend.
55
University of Ghana http://ugspace.ug.edu.gh
Table 4.1: Descriptive statistics
Variable Statistic Year
2007 2008 2009 2010 2011 2012 2013
PG Mean (million) 5.76 7.36 8.37 12.37 16.67 21.98 26.67
STDV (million) 5.94 7.58 9.86 0.14 0.22 0.30 0.04
Min (million) 0.30 0.46 0.49 0.33 0.49 0.81 0.05
Max (million) 21.01 27.75 33.87 47.83 16.62 99.67 126.79
SR Mean 2.86 2.17 2.50 2.38 2.80 2.07 2.45
STDV 2.14 1.91 1.40 0.86 3.74 0.55 1.58
Min 0.11 0.26 1.00 1.41 0.97 1.32 1.38
Max 15.85 8.16 6.49 4.74 16.62 3.09 7.10
LR Mean 8.81 9.92 7.68 14.71 6.44 8.31 9.54
STDV 1.80 1.12 1.01 0.24 0.09 0.36 0.37
Min 1.03 0.99 1.02 0.36 0.45 0.40 0.46
Max 3.12 3.38 9.55 3.48 1.19 2.86 3.84
IP Mean (million) 0.69 1.00 1.96 1.90 1.61 4.29 8.94
STDV (million) 0.59 1.09 2.10 1.95 2.00 5.16 1.54
Min (million) 0.10 0.05 0.16 0.02 0.004 0.02 0.24
Max (million) 2.12 4.16 7.45 6.82 6.98 17.97 57.08
CI Mean (million) 1.37 1.74 3.70 4.08 3.45 7.11 9.31
STDV (million) 0.06 0.05 1.32 1.39 1.68 1.93 1.24
Min (million) 0.10 0.08 0.001 0.001 0.013 0.09 0.10
Max (million) 2.03 2.54 5.98 9.26 9.54 11.94 13.20
SZ Mean 0.08 0.08 8.04 0.07 0.06 0.06 0.07
STDV 0.10 0.08 0.41 0.08 0.08 0.10 0.09
Min 0.01 0.005 0.14 0.0005 0.0004 0.001 0.001
Max 0.34 0.30 0.50 0.26 0.268 0.31 0.31
CR Mean 1.36 1.94 4.47 1.91 0.90 5.79 1.17
STDV 0.84 3.54 11.76 3.78 0.59 11.59 0.59
Min 0.03 10.66 0.02 0.000 0.17 0.08 0.299
Max 2.73 0.03 43.49 13.69 2.30 43.02 2.404
OM Mean 0.76 0.26 0.79 0.23 0.12 0.57 0.419
STDV 0.20 0.26 0.19 0.32 0.09 0.26 0.28
Min 0.001 0.001 0.0001 0.01 0.006 0.03 0.002
Max 0.66 0.67 0.81 0.93 0.344 0.64 0.64
INF Yearly Average 10.73 16.46 19.30 10.79 8.73 8.15 11.64
56
University of Ghana http://ugspace.ug.edu.gh
Table 4.1 shows a summary statistics of the data used for this study. Summary statistics are used
to summarize a set of observations in order to communicate the information about the data set as
simply as possible. The table shows the mean, standard deviations, minimum and maximum
values of the variables used in the study across the years. From the table, it is observed that most
of the values were increasing, decreasing or inconsistent over the years. This means that the
financial ratios were not following a uniform trend over the study period, some insurers were
solvent in certain years but insolvent in others. The standard deviations vary widely between the
companies and over time. The large standard deviations can be attributed to the fact that some
life companies are bigger than others in financial terms.
We estimated pooled Ordinary Least squares (OLS) regression for comparison with the panel
regression as shown in Table 1, 2 and 3 in the appendix. The pooled OLS estimator ignores the
panel structure of the data and simply estimates the parameters of the model. In this regard, the
OLS regression does not consider heterogeneity across groups (life companies) or time. It
assumes that the data comes from homogenous entities over time. Since R software was used, the
data was specified as a panel data, with the type of company represented by the data ID and
years as the time parameter.
The significance level of its regression was 5%. Hence, any p-value of the coefficients that is less
than 0.05 represents a significant predictor of solvency. The coefficient of the constant in the
model, LR, CI, IP, SZ, CR, OM and INF are all less than 0.05 hence they are all significant
predictors of solvency except PG and RF. The positive coefficients of PG, LR, IP, OM and SZ
show that they are positively related to solvency. This means that an increase in these variables
associated with the life insurance companies will cause their financial health to increase
positively and vice versa. However, the coefficient of CI, INF and CR are negative, indicating
57
University of Ghana http://ugspace.ug.edu.gh
that they have an inverse relationship solvency. This also means that an increase in these
variables will cause a decrease in the financial health of the life companies.
The ANOVA Table 2 in the appendix displays the overall significance of the model. It shows
whether the overall regression coefficients are significant. The table gives an indication that the
regression model is significant in totality and hence the interpretation of the coefficients makes
sense. This is because the p-value of the F-statistics (0.0005) is less than the level of significance
value of 5% (0.05) hence we reject the null hypothesis that the model is not a good fit. We
therefore conclude on the alternate hypothesis that the model is a good fit of the data.
The value of the adjusted R-square indicates that in considering homogeneity across companies
and over time, 52.99% of the variability in the financial health was explained by the independent
variables. After the OLS estimates, it is laudable to test the data if there exist any panel effect. If
there exists no panel effect, the OLS will be appropriate otherwise we go ahead to estimate the
random and fixed effect model to choose the appropriate one.
Table 4.2 shows the Breusch-Pagan Lagrange multiplier test for panel effect. The null hypothesis
of this test is that there is no panel effect (meaning that OLS is better) and the alternate is that
there is a panel effect, meaning that OLS is not appropriate and hence panel regression must be
used.
Table 4.2:  Testing for panel effect
Chi-square value Df p-value
98.129 1 0.0002
As can be seen, the p-value of 0.0002 is less than 0.05. Here we reject the null hypothesis and
conclude that the OLS is not appropriate and that panel regression analysis should be considered.
58
University of Ghana http://ugspace.ug.edu.gh
That is, there is evidence of significant differences across the life companies, therefore we cannot
run a multiple OLS regression.
4.3 Fixed Effect Panel Regression
This is known as a “fixed effects” regression because it holds constant (fixes) the average effects
of each life company. In the fixed effects model, the individual-specific effect is a random
variable that is allowed to be correlated with the explanatory variables.
Table 4.3: Unbalanced Fixed effect panel Regression estimates
Coefficients Estimate Standard Error t-value Pr(>|t|)
Intercept 3.25E-04 3.02E-05 10.7436 0.000000
PG -3.97E-02 4.23E-02 -0.9378 8.120E-1
LR 2.14E-05 1.24E-06 17.2742 2.129E-2
CI -9.25E-05 8.65E-06 -10.6925 6.281E-3
IP 9.80E-05 5.77E-06 16.9800 6.031E-4
SZ 6.23E-01 4.19E-02 14.8783 8.123E-4
CR -2.00E-01 1.08E-02 -18.4907 0.301E-4
OM 2.21E-05 1.50E-06 14.7333 7.125E-4
INF -3.42E-03 1.20E-04 -28.5000 1.045E-3
RF 4.56E-04 6.11E-04 0.7466 1.236E-1
Table 4.3 shows the fixed effects regression coefficients and their corresponding p-values. Again
the significance level of the regression was 5%, hence any p-value of the coefficients that is less
than 0.05 shows that it is a significant predictor of solvency. The coefficient of the constant in
the model, LR, CI, IP, SZ, CR, OM and INF are all less than 0.05 hence they are all significant
59
University of Ghana http://ugspace.ug.edu.gh
predictors of solvency except PG and RF. The positive coefficients of IP, SZ and OM show that
they are positively related to solvency. This means that an increase in these variables associated
with the life insurance companies will cause their financial health to increase positively and vice
versa. However, the coefficient of PG, CR, INF and CI are negative, indicating that they have an
inverse relationship with solvency ratio. This also means that an increase in these variables will
cause a decrease in the financial health of the life companies.
4.3.1 Diagnostics for the Fixed Effects Panel Regression Model
Table 4.4: Regression statistics (Fixed effect)
R2 Adjusted R2 value
0.7361 0.5418
The R2 is the coefficient of determination. It measures the variability in the dependent variable
that has been explained by the independent variables. The value of R lies between 0 and 1. The
closer the value of R2 to 1 the better the model fits the data. The value of the adjusted R2 shows
that in considering homogeneity across companies and over time, 54.18% of the variability in the
financial health was explained by the independent variables.
Table 4.5: ANOVA results (Fixed effect)
Sum of squares Df F-statistics Pr(>|t|)
Regression 28.23 9 23.246 0.000
Residual 10.12 75
Total 38.35 84
60
University of Ghana http://ugspace.ug.edu.gh
The ANOVA result (Table 4.5) shows the overall significance of the model. It shows whether
the overall regression coefficients are significant. The table gives an indication that the
regression model is significant in totality and hence the interpretation of the coefficients makes
sense. This is because the p-value of the F-statistics (0.000) is less than the level of significance
value of 5% (0.05) hence we reject the null hypothesis that the model is not a good fit and
conclude on the alternate hypothesis that the model is a good fit of the data.
4.4 Relationship between firm size and financial health
From the fixed effect model, it was realised that firm size has a positive significant impact in
financial health of life companies. However, the question of when the positive effect of firm size
on financial health seizes is yet to be ascertained. That is, if a firm increases in size to infinity,
will it continue to positively impact on financial health? Or is there a point at which an increase
in firm size no longer has a positive effect on financial health? To answer this question, a
quadratic model was fitted to financial health (dependent variable), with an extra independent
variable the square of firm size. The result of this regression is displayed in Table 4.7, 4.8 and
4.9. The turning point of this model is then used to determine the maximum firm size of a life
company in relation to it financial health. From the regression, the quadratic model was given as;
= 3.40 − 6−+((92.0.174 −−25)) −−((69..1122 −−61)) ++((95..7400 −−73)) ++((14..8707 −−23))
Where
a= liquidity Ratio
b= claims incurred
c= investment performance
x= size
x2= (firm size)2
d= combined ratio
61
University of Ghana http://ugspace.ug.edu.gh
e= operating margin
f=inflation g=reforms
Now since at the maximum point = 0,
We d1if.8fe0ren−tia2te− th(e2 a∗bo9v.0e 7equ−ati2o)n p=art0ially with respect to x and equate it to zero=
= 1.80 − 2=2 ∗ 9.07 − 20.9923
From the result above we expect that the ratio of a firm’s gross premium to the total gross
premium, which was used to measure market share should be 0.9923 before the positive impact
of size can no longer exist on its financial health. From the data used for this present study, it is
no life company exhibited such a large market share. We therefore conclude that and increase in
firm size has a positive impact on financial health of life companies in Ghana until the ratio of its
gross premium to the total gross premium reaches 0.9923.
Table 4.6: Unbalanced Fixed effect panel Regression estimates (including size square)
Coefficients Estimate Standard Error t-value Pr(>|t|)
Intercept 3.40E-06 3.31E-07 10.2719 2.100E-4
PG -1.80E-06 2.10E-06 -0.8571 6.129E-1
LR 2.14E-05 1.34E-06 15.9701 2.482E-2
CI -6.12E-06 2.50E-07 -24.4800 4.125E-3
IP 9.70E-07 4.30E-08 22.5581 4.013E-2
SZ 1.80E-02 2.10E-03 8.5714 5.131E-4
SZ2 -9.07E-02 1.15E-02 -7.8870 4.205E-3
CR -9.12E-01 1.13E-01 -8.0708 2.106E-2
OM 5.40E-03 3.50E-04 15.4286 3.246E-2
INF -4.77E-03 5.10E-04 -9.3511 1.0034E-2
RF 9.80E-03 2.37E-02 0.4139 4.056E-1
62
University of Ghana http://ugspace.ug.edu.gh
Table 4.7: ANOVA results
Sum of squares Df F-statistics Pr(>|t|)
Regression 34.28 10 25.438 0.000
Residual 11.23 74
Total 45.51 84
The ANOVA result (Table 4.7) shows the overall significance of the model. It shows whether
the overall regression coefficients are significant. The table gives an indication that the
regression model is significant in totality and hence the interpretation of the coefficients makes
sense. This is because the p-value of the F-statistics (0.000) is less than the level of significance
value of 5% (0.05) hence we reject the null hypothesis that the model is not a good fit and
conclude on the alternate hypothesis that the model is a good fit of the data.
Table 4.8: Regression statistics
R2 Adjusted R2 value
0.7532 0.5173
The value of the adjusted R-square shows that in considering homogeneity across companies and
over time, 51.73% of the variability in the financial health was explained by the independent
variables.
Also, Table 4, 5 and 6 in the appendix shows the results of the random effects regression
coefficients and their corresponding p-values. Again the significance level of the regression was
5%, hence any p-value of the coefficients that is less than 0.05 shows that it is a significant
predictor of solvency. The coefficient of the constant in the model, LR, SZ, IP, CI, CR INF and
OM are all less than 0.05 hence they are all significant predictors of solvency except PG and RF.
63
University of Ghana http://ugspace.ug.edu.gh
The positive coefficients of LR, SZ, IP and OM show that they are positively related to solvency.
This means that an increase in these variables associated with the life insurance companies will
cause their financial health to increase positively and vice versa. However, the coefficient of PG,
CI, INF and CR are negative, indicating that they have an inverse relationship solvency. This
also means that an increase in these variables will cause a decrease in the financial health of the
life companies. The interpretation of the coefficients is the average effect of the independent
variable over the dependent variable when the independent variable changes across time and
between entities by one unit.
The ANOVA analysis (Table 5 in the appendix) shows the overall significance of the model. It
shows whether the overall regression coefficients are significant. The table gives an indication
that the regression model is significant in totality and hence the interpretation of the coefficients
makes sense. This is because the p-value of the F-statistics (0.0023) is less than the level of
significance value of 5% (0.05), hence we reject the null hypothesis that the model is not a good
fit and we conclude the alternate hypothesis that the model is a good fit of the data.
The value of the adjusted R-square from Table 3 in the appendix shows that in considering
homogeneity across companies and over time, 52.92% of the variability in the financial health
was explained by the independent variables.
4.5 Hausman test
Table 4.9: Hausman test for best model
Chi-square value p-value
1.012 0.0491
64
University of Ghana http://ugspace.ug.edu.gh
To decide between fixed or random effects we run a Hausman test with the null hypothesis that
the preferred model is random effects versus the alternative to the fixed effects (see (Green,
2008)). It basically tests whether the unique errors are correlated with the regressors, the null
hypothesis states they are not. Since the p-value of the test is less than 0.05 we reject the null
hypothesis. Therefore, we conclude that fixed effect is the preferred model for this study. Hence
our conclusions are now based on the fixed effect model. Consequently the tables and
diagnostics below are all based on the fixed effect model. Below are the diagnostic tests as
applied to fixed effect panel regression model.
4.6 Diagnostic Tests for the fixed effects.
According to the panel regression, it is necessary for the errors to be serially uncorrelated and
also to be homoscedastic (that is the errors need to have a constant variance). It is also relevant to
test for time-fixed effect to see if there is the need to consider the effect of time. The study also
checked for cross-sectional dependence/contemporaneous correlation in the data set.
4.6.1 Testing time-fixed effects.
The null is that no time-fixed effects needed.
Table 4.10:  Testing for time effect
F-value Df p-value
0.1259 8,73 0.9501
If there is a cause to suspect that there are time-specific effects, which affect all individuals in the
same way, then an extended model can be estimated by including a dummy variable for each
time period. This means that if there is the presence of time effect, the model has to be extended
to include the time factor. Now, since the p-value = 0.9501 is greater than 0.05 we conclude on
65
University of Ghana http://ugspace.ug.edu.gh
the null hypothesis that there is no significant time effect. Hence there is no need to consider a
new model that takes into consideration the effect of time.
4.6.2 Testing for serial correlation
Table 4.11:  Testing for serial correlation
Chi-square value Df p-value
0.7124 6 0.2321
Table 4.11 shows the test for serial correlation. One assumption of the panel regression is that the
errors must be serially uncorrelated as it has the tendency of inflating the errors associated with
the model. The table shows the Breuch-Godfrey/Wooldridge test for serial correlation in the
panel model.  Since the p-value is greater than 0.05, it is concluded that there is no serial
correlation in the random effect model.
4.6.3 Testing for cross-sectional dependence/contemporaneous correlation: using Breusch-
Pagan LM test of independence
Table 4.12:  Testing for cross-sectional dependence
Chi-square value Df p-value
0.7321 122 0.2451
The null hypothesis in the B-P/LM tests of independence is that residuals across entities are not
correlated. B-P/LM tests are used to test whether the residuals are correlated across entities.
66
University of Ghana http://ugspace.ug.edu.gh
Cross-sectional dependence can lead to bias in the test results (also called contemporaneous
correlation). The Breusch-Pagan LM test for cross-sectional dependence in panels shows that
there is no cross sectional dependence in the model. The chi-square value of 0.7321 with p-value
of 0.2451 means that there is no cross sectional dependence. This is good for the study since the
bias in the estimates of the model is not inflated by correlated residuals among the different life
companies.
4.6.4 Breusch-Pagan test for Heteroscedaticity
Table 4.13:  Testing for heteroscedasticity
Chi-square value Df p-value
0.9129 9 0.0923
One major assumption of regression is that the errors must be homoscedastic (that is a constant
variance must be assumed). The results show the absence of heteroscedasticity of the Breusch-
Pagan test for heteroscedasticity shows that the errors of the fixed effect model is homoscedastic
since the p-value is greater than 0.05. The null hypothesis of the test is that the errors have equal
variances while the alternate is that the errors are not constant. Hence the assumption of equal
variances is satisfied.
4.7 Categorizations of the solvency ratios
The table below shows the year on year solvency margin of the sixteen life insurance companies
involved in the study. It shows that over the seven year period of the study, there have been 3
cases of insolvency, 20 cases of weak solvency, 31 cases of moderate insolvency and 46 cases of
strong solvency within the study period (from 2007 to 2013).
67
University of Ghana http://ugspace.ug.edu.gh
Table 4.14: Categorization of solvency ratios
Year Solvency margin
Insolvent (%) Weakly Moderately Strongly
solvent (%) solvent (%) solvent (%)
2007 1 (8.3) 2 (16.7) 1 (8.3) 8 (66.7)
2008 1 (7.7) 2 (15.4) 4 (30.8) 6 (46.2)
2009 0 (0.0) 2 (15.4) 4 (30.8) 7 (53.8)
2010 0 (0.0) 2 (13.3) 9 (60.0) 4 (26.7)
2011 1 (6.3) 2 (12.5) 5 (31.3) 8 (50.0)
2012 0 (0.0) 5 (31.3) 4 (25.0) 7 (43.8)
2013 0 (0.0) 5 (33.3) 4 (26.7) 6 (40.0)
Total 3 (3.0) 20 (20.0) 31 (31.0) 46 (46.0)
Source: Author (2015)
The categorizations were made as follows; SR less than 1 means insolvent, SR from 1 to 1.49
means weakly solvent, SR from 1.5 to 1.99 means moderately solvent and SR greater than 2
means strongly solvent.
4.8 Discussion of Findings
This study provides an interesting insight into insurance sector of Ghana, focusing primarily on
life insurance. The study discovered that from annual reports published by the NIC, a trend of
positive growth was recorded in the last decade. Although general insurance has the largest
proportion in Ghana, the life sector has recorded a significant increase in the number of insurers,
gross premium and number of policyholders. The current monitoring procedure of the NIC
which ensures compliance with insurance regulations is not comprehensive enough and does not
serve as a good warning system for detecting future insolvency as previously hypothesized (NIC,
2009).
68
University of Ghana http://ugspace.ug.edu.gh
To begin with, the study addressed both the firm specific and macro factors that affect life
insurer’s financial soundness and used panel regression analysis to assess the impact of some
firm specific factors on a sample of life insurers in Ghana. Solvency Ratio (SR), which was used
as a proxy for Financial health was subdivided into four mutually exhaustive classes such that
the solvency ratio of each life insurer over the years fall within a particular group. The
categorizations were, insolvent, weakly solvent, moderately solvent and strongly solvent. The
results showed that over the period of the study, there were few occasions where some life
insurance companies in Ghana became insolvent. The licenses of these companies were not
revoked by the NIC, but the NIC changed its monitoring approach to help these companies in
business.
In addition, the present study found that most life insurance companies were solvent most of the
time during the study period, based on their financial statements. This conclusion was drawn
based on the study’s categorization of life insurer solvency into insolvent, weakly solvent,
moderately solvent and strongly solvent. It was shown that less than 40% of the sampled life
insurers were between moderately solvent and insolvent, with majority of them being weakly
insolvent, while few were completely insolvent in certain years based on their financial ratios.
This study examined the solvency regulatory framework in Ghana and looked at the factors that
can affect solvency and their impact on financial health. A close look at the data collected and
according to KPMG 2014, there has been a tremendous growth in the Ghanaian life insurance
industry from 16 registered life insurers in 2007 to 21 in 2014.
The following section summarises the main findings of this study and discusses the impacts of
the selected independent variables on the financial health of life insurers. Findings of the effects
of the seven predictor variables on financial health are discussed as follows;
69
University of Ghana http://ugspace.ug.edu.gh
 Firm Size; as shown in the unbalanced fixed effect Table 4.6, firm size (SZ) was
observed to have a positive coefficient, hence a positive relationship with financial
health. That is, larger firms were seen to be more solvent after the analysis. This finding
is in line with existing literature, (Chen & Wong, 2004). Intuitively, a life insurer with a
larger firm size like SIC is unlikely to go insolvent since it has a larger market share and
higher capital. The point at which SZ seizes to have a positive impact on FH was
calculated as 0.9923.
 Investment Performance is also shown in Table 4.4 to have a positive relationship with
financial health. Intuitively, this finding is expected since an insurer that makes wise
investment decisions is expected to gain more interest from investments and hence
improve its financial soundness. Investment is one of the major ways of ensuring the
financial health of insurers. Since life policies are long term in nature, the ability of an
insurer to make wise investment decisions that will yield good returns is very essential in
ensuring that they are able to meet their future obligations as well as operating in an
effective manner.
 Liquidity Ratio, which is computed as the ratio of current assets to current liabilities also
has a positive relation with financial health as expected. Solvency ratio, which was used
as a proxy for measuring financial health is a ratio of total assets to total liabilities.
Consequently, liquidity ratio, which is a subset of the solvency ratio is expected to have a
positive relation. A high value of liquidity ratio, means an insurer is better capable of
meeting its demands and therefore has a low probability of going insolvent.
 Operating Margin also showed a positive relation with financial health as displayed in
Table 4.4. This result is in line with studies by BarNiv & Hershbarger (1990) and Kramer
70
University of Ghana http://ugspace.ug.edu.gh
(1996). The operating margin gives how much revenue is left after all operating costs
have been catered for. A higher value therefore implies that a life insurer is financially
sound since it has more revenue after payment of operating cost.
 Claims Incurred which measures the amount of claims received by an insurer per year
was seen in Table 4.4 to have a negative relationship with financial health as expected.
This is in line with previous literature including Owusu Ansah et al. (2010). An insurer
incurring a smaller amount claims is more financially sound than one with lots of claims
per year.
 Combined Ratio is a measure of profitability showed a negative relationship with
financial health after analysis of the data. This is in line with expectations at the
beginning of the study, which stipulated that an insurer with a smaller percentage (below
100%) to use a smaller amount of money for insurance operations compared to one who’s
combined ratio is above 100%. The life insurers with higher values of combined ratio
were not making an underwriting profit, which meant a loss on insurance operations. This
then resulted in weak financial health or insolvency of those life insurers.
 Premium Growth which shows the growth in premiums collected was shown at the end of
the analysis to be an insignificant predictor of financial health. Its sign after the
regression is also not in line with what was expected at the beginning of the study but is,
however, in line with previous studies (Kim et al., 1996; Chen & Wong, 2004; Luhnen,
2009).
Although this study did not focus on financial performance, the positive relation of investment
performance and negative effect of premium growth to the financial health of Ghanaian life
insurers highlights the differences between financial health and financial performance. In that,
71
University of Ghana http://ugspace.ug.edu.gh
while premium growth is positively related to financial performance, it shows a negative relation
to financial health. Again, investment performance also shows a negative relation to financial
performance (Akotey et al., 2013) but a positive relationship with financial health. Hence a
company exhibiting good financial performance is not necessarily financially sound.
The analysis of the data gathered disagrees with Chen and Wong (2004) conclusion that
combined ratio has a positive relation with solvency. This could be as result of the geographical
differences or difference in the computation of ratios or the data collected. The evidence from
this study is, however, in line with other studies on insurer insolvency including Simpson &
Damoah (2009) who stipulate that the system used by the NIC to monitor insolvency is not
comprehensive enough.
Considering the macro factors discussed in chapter two, the effect of regulation may have an
impact on financial health of life insurers. Major structural reforms took place in 2007, 2009 and
2013 in the Ghanaian insurance industry. As shown in Table 4.1.6, only one life company was
insolvent in 2007 while 8 were strongly solvent, based on the categorization of the present study.
In 2013, however, there was no insolvent life company, 5 were weakly solvent, 4 moderately
solvent and 6 strongly solvent. This shows an improvement in the financial health of the insurers
which can be attributed to the major reforms implemented in those years. Bonin, Hassan &
Watchel (2005) studied the effects of reforms and regulation on the financial soundness of banks
and found a direct relationship. Brissmis, Delis & Papanikolaou (2008) also found a positive
relation between reforms and bank financial performance. The p-value of reform which served as
a control variable in the regression was below 5%, making it insignificant. Hence according to
the study, reform is not a significant determinant of financial health of Ghanaian life insurers.
72
University of Ghana http://ugspace.ug.edu.gh
Nevertheless, inflation which increased significantly from 10.5% in 2006 to 12.7% in 2007 and
to 20.1% in 2009 (KPMG, 2010) is inversely related to investment yield/performance. This
relationship is however only true in the long run. Mishkin (1992) and Khan & Senhadji (2001)
found no empirical evidence of the Fisher effect, which postulates a high correlation between
level of interest rates and inflation in the short run but support the existence of long-run Fisher
effect. Fluctuation of inflation rates as can be observed in the Ghanaian context over the sampled
period would therefore have an impact on financial health. This is in line with the findings of this
study as years with significant differences of inflation levels did not exhibit significant change in
solvency as shown in Table 4.15. For example, from 2007 to 2009, inflation increased from
12.7% to 20.1%, in 2007 one company was insolvent, 2 companies were weakly solvent, 1
company was moderately solvent and 8 companies were observed to be strongly solvent
according to the present study. In 2009, no company was insolvent based on the results of the
study, 2 were weakly solvent, 4 were moderately solvent and 7 were strongly solvent, which
does not show much difference over those year periods. From the regression, inflation showed a
negative relation with financial health. Hence, the higher the inflation, the less financially sound
an insurer becomes. Overall, the macro factors as well as the firm-specific factors can be said to
have impacted on the financial health of the sampled life insurers.
4.9 Conclusion
Thus the main results and findings of the model used in the study are presented. To satisfy the
objectives of this study, descriptive statistics, contingency tables were provided. Also, the main
statistical tool used for the analysis was the multiple ordinary least squares regression, and panel
regression. The unbalanced panel regression was employed in determining the effect of some
73
University of Ghana http://ugspace.ug.edu.gh
financial indicators of the financial health of some selected life companies. Finally, the chapter
discussed the major findings of the study.
74
University of Ghana http://ugspace.ug.edu.gh
CHAPTER FIVE
SUMMARY, CONCLUSION AND RECOMMENDATION
5.1 Introduction
The chapter provides a summary of the findings as well as the conclusions, recommendations
and suggests areas for future research. The main aim of the present study was to investigate
determinants of financial health of life insurance companies in Ghana. The discussion is
organized around the three main objectives of the study.
5.2 Summary
The study began with an overview of insurance in general, its contribution to the economy, life
insurance insolvency and the need to monitor insurers to ensure financial health and protect
policy holders. Since insurer insolvency is quite common, the study reviewed existing literature
in this field and found that monitoring, regulation and proper assessment of financial statements
of insurers by regulators can reduce insurer insolvency by putting measures in place to serve as
an early warning system.
To evaluate the effect and impact of some financial variables on the solvency of the life
companies, a multiple regression and panel regression was adopted. First an ordinary least
squares regression was used without considering the difference in entities (life companies). The
results of this showed that OM, LR, IP and SZ are positively related to solvency. This result is in
line with studies conducted by Chen & Wong (2004), Cummins, Harrington & Klein (1995) and
Yang (2006). This means that an increase in these variables associated with the life insurance
companies will impact positively on financial health and vice versa, while CI, INF and CR can
have a negative impact on the financial health of life insurance companies.
75
University of Ghana http://ugspace.ug.edu.gh
In conducting the panel regression two separate analysis were performed. The fixed effect and
random effect panel regressions were conducted. To choose the more appropriate of the two, the
Hausman test was conducted to choose the best among the two. The result of the Hausman test
showed that the fixed effect panel regression was the most appropriate. Hence all conclusions
and consequences follow from the results of the fixed effect panel regression.
The diagnostic tests for panel regression were tested to ensure model adequacy. The tests
conducted showed that the assumptions of homoscedasticity, serial correlation, cross-sectional
dependence correlation are satisfied.
Although the research only tests the effects of firm-specific variables of insurers on their
solvency, it also offers an insight into the possible effects of other industry-wide variables like
inflation, methods of insurance distribution and reforms and regulation. The study then goes on
to conclude that these macro variables could be significant determinants of an insurer’s financial
soundness. The study therefore still carries important implications for future researchers. The
variables studied may help examine financial health of insurers within different financial
environments.
5.3 Conclusions
Findings from the study show a positive relation between financial health and reforms, from the
years 2007 to 2013, there was a general increase in the number of weakly solvent insurers. This
could be attributed to the increase in the number of life insurers which made it necessary for the
NIC to put in place new reforms and regulations. Inflation, claims incurred and combined ratio
also showed a negative relation to financial health. Findings from the study revealed that a rise in
SZ, LR, IP and OM can increase the solvency ratio of an insurer, hence making it more
76
University of Ghana http://ugspace.ug.edu.gh
financially sound. CI, CR and INF on the other hand, must be low in order to increase financial
soundness.
Careful monitoring of these ratios in the annual financial statements of these companies by the
NIC will go a long way in preventing future insolvencies. This will subsequently protect the
interest of policyholders and serve as a source of financial stability for the Ghanaian economy.
5.4 Potential Areas of Application
This study serves as an insight into areas that influence the decision of the public on taking life
policies, management of life companies and regulators of insurance who work as monitors and
decision makers.
As custodians of insurance and all related matters, regulators must ensure proper monitoring of
insurers, which is key in detecting and preventing future insolvencies. With the intention of
effectively operating life insurance policies to improve the financial sector of the Ghanaian
economy, it is imperative to know the core factors and principles that cause financial distress
among insurers. Studying and monitoring these factors as published by the insurers in their
annual financial statements can lead to better financial performance of life insurers and serve as a
financial booster to the economy as a whole.
Policyholders are also able to make confident decisions about taking up life policies when they
are assured of the financial soundness of the insurers and have no doubt of any uncertainties like
insolvency. Many people are affected aside the state when an insurer becomes insolvent. Due to
the long term nature of life insurance, unlike general insurance, insureds are better able to reap
its savings benefit if the insurers are financially sound and there’s no risk of future insolvency.
77
University of Ghana http://ugspace.ug.edu.gh
Management of the insurance companies are the greatest beneficiaries of this study. Monitoring
and assessing of these determinants of insolvency can make them restructure their operations in
order to prevent bankruptcy. If, for example, the investment profit of a life insurer is not
appreciable enough, management can adopt steps in order to make better investment decisions
which will in turn make the insurer financially sound. Another typical example is to develop
innovative life policies that will attract more customers in order to increase their market share,
which is seen to have a positive relation with financial stability.
5.5 Recommendations
5.5.1 Recommendations for Management, Policyholders and Investors
In order to ensure financial stability of insurers, it is essential to adopt a solvency regulatory
framework that monitors the financial position on a timely basis. This framework should be able
to detect early signs of insolvency and prevent the future bankruptcy among insurers.
Due to the relationship that exists between solvency ratio (SR), premium growth (PG), combined
ratio (CR), operating margin (OM), claims incurred (CI), investment income (IP) and firm size
(SZ) it is highly recommended that life insurance companies pay attention to these indicators as
some of them could have a devastating impact on their solvency and competitiveness in the
business.  It is also necessary that life companies monitor other financial indicators like gross
margin percentage, financial leverage and return on equity, as they could contribute to their
continual stay in business.
To detect early signs of insolvency, the NIC, in addition to move towards the Risk Based
Supervision approach, monitor these financial ratios and intensify field inspections as well as
ensuring compliance with insurance regulation. This can serve as an early warning system in
78
University of Ghana http://ugspace.ug.edu.gh
determining distressed insurers. In order to achieve better overall performance, policy makers of
the insurance industry must also put in place measures to ensure better corporate performance of
the existing companies to make the selling of life insurance appealing to consumers.
5.5.2 Recommendations for Future Research
The following areas could be addressed in future research in order to increase robustness and
applicability of the model: Alignment of the dimensions, completeness of the model and
coverage of the dimensions. The effect of other financial variables on the solvency of life
insurance companies can also considered using other statistical tools.
It is also recommended that future researchers assess the robustness of the new solvency
regulatory system which is yet to be put in place by the NIC. This system is expected to follow a
Risk Based Supervision Approach. Again, future researchers can compare the robustness of the
solvency regulatory framework currently in use by the NIC and the Risk Based Supervision
method and see which one fits more in the Ghanaian context in preventing insolvency.
5.6 Limitations of the Study
Notwithstanding the benefits of this study, there are certain limitations and drawbacks which did
not guarantee a hundred percent accuracy of findings. Some of these limitations include the cost
of undertaking the study in terms of moving about to obtain some data that was not readily
available on the financial statements from some of the registered Life Insurance Companies
mainly due to security reasons and for the confidential nature of some of the data required. It is
for this reason that the NIC has mandated all insurers to ensure that their annual financial
statements conform to international standards. This will make a routine assessment of their
financial characteristics easier.
79
University of Ghana http://ugspace.ug.edu.gh
Finally, the topic was limited in terms of the time range. Due to the broad nature of the industry,
a longer duration would have been ideal for accurate information to be derived but this, however,
was impossible because data on life insurance solely in Ghana are time bound, starting from
2007.
80
University of Ghana http://ugspace.ug.edu.gh
REFERENCES
Ahiawodzi, A. K., & Sackey, F. G. (2010). Determinants of Credit Rationing to the Private
Sector in Ghana. African Journal of Business & Economic Research, 5(1), 23-49.
Ahmed, N., Ahmed, Z., & Usman, A. (2011). Determinants of performance: a case of life
Insurance sector of Pakistan. International Research Journal of Finance and Economics, 61,
123-128.
Akotey, J. O., Sackey, F. G., Amoah, L., & Manso, R. F. (2013). The financial performance of
life insurance companies in Ghana. . Journal of Risk Finance, 14(3), 286-302.
Akotey, J., Sackey, F., Amoah, L., & Frimpong Manso, R. (2013). The financial performance of
life insurance companies in Ghana. The Journal of Risk Finance, 14, 286-302.
Allen, D. W. (1999). Transaction costs. Encyclopedia of law and economics, available at:
www2.bren.ucsb.edu/~glibecap/Allentranscosts.pdf.
Anders Grosen, A., & Jørgensen, P. L. (2002). Life Insurance Liabilities at Market Value: An
Analysis of Insolvency Risk, Bonus Policy, and Regulatory Intervention Rules in a Barrier
Option Framework. Journal of Risk and Insurance, 69(1), 63-91.
Ansah-Adu, K., Andoh, C., & Abor, J. (2012). Evaluating the cost efficiency of insurance
companies in Ghana. Journal of Risk Finance, 13(1), 61-76.
Ayee, J. (2013). Public Administrators under Democratic Governance in Ghana. International
Journal of Public Administration, 36(2), 440-452.
Baltagi, B. H., & Chang, Y. J. (1994). Incomplete panels: A comparative study of alternative
estimators for the unbalanced one-way error component regression model. Journal of
Econometrics, 62, 67-89.
Bank of Ghana (BOG). Economic and Financial Report Bank of Ghana. (2007). Available at:
https://bog.gov.gh/privatecontent/Monetary_Policy/2007/Dec/inflation outlook . Accessed on: 12
February 2015.
BarrNiv, R., & Hershbarger (1990). Classifying Financial Distress in the Insurance Industry.
Journal of Risk and Insurance, 57, 110-136.
Barros, C. P., Nektarios, M., & Assaf, A. (2010). Efficiency in the Greek insurance industry.
European Journal of Operational Research, 205(2), 431-436.
Barros, P., & Obijiaku, E. L. (2007). Technical Efficiency of Nigerian Insurance Companies.
Department of Economics, Institute for Economics and Business Administration (ISEG),
Technical University of Lisbon. Working Papers No. 18. [Accessed on 15 February 2015].
Bashir, I. J., Naik, C., & Madhavaiah, R. (2012). Potential Business Applications of Quick
Response (QR) Codes. Prajnan, XLI(4), available at www.researchgate.net.
Bashir, S. I., Madhavaiah, C., & Naik, J. R. (2013). Critical Analysis of Traditional and Modern
Insurance Distribution Channels in India. The Journal of Insurance Institute of India.
81
University of Ghana http://ugspace.ug.edu.gh
Batten, M. (1981). Insurance: A turbulent time for the industry. The Chartered Accountant in
Australia, 51(11), 66-68.
Beck, T., & Webb, I. (2003). Economic, Demographic, and Institutional Determinants of Life
Insurance Consumption across Countries. Oxford Journals, 17(1) [Accessed on 25 May
2015].
Berg, H. (2010). Risk Management: Procedures, Methods and Experiences. Salzgitter, Germany,
1.
Best A.M., (2007). Best Ratings and Analysis. Available at www.ambest.com.
Bikker J. A., & Leuvensteijn, M. V., (2008). Competition and efficiency in the Dutch life
insurance industry. Applied Economics, 40, 2063-2084.
Bikker, J. A., & van Leuvensteijn, M. (2008). Competition and efficiency in the Dutch life
insurance industry. Applied Economics, 40(16), 2063-2084. doi:
10.1080/00036840600949298
Black, J. & Baldwin, R. (2010). Really Responsive Risk-Based Regulation. Journal compilation
The University of Denver/Colorado Seminary, 32(2), 181-213.
Black, J. (2005). The Emergence of Risk-Based Regulation and the New Public Risk
Boadi, E. K., Antwi, S., & Lartey, C. V (2013). Determinants of the Profitability of Insurance
Firms in Ghana. International Journal of Business and Social Research, 3(2).
Boadu, F., Dwomo-Fukuo, E., Boakye, J., & Frimpong, A. (2014). Assessing the Life Insurance
Industry in Ghana. European Journal of Business and Management, 6(21).
Bonin, J. P., Hassan, I., & Watchel, P. (2005). Bank performance, efficiency and ownership in
transition countries. Journal of Banking and Finance, 29(1), 31-53.
Bose, I. (2006). Deciding the financial health of dot-coms using rough sets. Information and
management, 43(7), 835-846.
Brissmis, S. N., Delis, M., & Papanikolaou, N. (2008). Exploring the nexus between banking
sector reform and performance: Evidence from newly acceded EU countries. Journal of Banking
and Finance, 32(12),2674-2683.
Brockett, P.L., Cooper, W.W., Golden, L.L., & Pitaktong, U. (1994). A Neural Method for
Obtaining an Early Warning of Insurer Insolvency. The Journal of Risk and Insurance. 61(3),
402-424.
Brockett, P.L., Cooper, W.W., Golden, L.L., Rousseau, J.J., & Wang, Y. (2004). Evaluating
solvency versus efficiency performance and different forms of organization and marketing in
US property-liability insurance companies. European Journal of Operational Research,
154(2), 492-514.
82
University of Ghana http://ugspace.ug.edu.gh
Brown, W., & Churchill, C. (1999). Providing insurance to low-income households: Part 1: a
primer on insurance principles and products, Working Paper, Microenterprise Best Practices
Project.
Browne, M., & Kihong, K. (1993). An International Analysis of Life Insurance Demand. The
Journal of Risk and Insurance, 60, 616-634.
Browne, M.J., Carson, J.M., & Hoyt, R.E. (1999). Economic and Market Predictors of
Insolvencies in the Life-Health Insurance Industry. The Journal of Risk and Insurance, 66(4),
643-659.
Bruderl, J. (2005). Panel Data Analysis. University of Mannheim, Germany. [Accessed on 12
April 2015].
Caprio, G. J., & Klingbiel, D. (1996). Dealing with Bank Insolvencies: Cross Country
Experience. Policy Research Department Working Paper. The World Bank.
Charumathi, B., (2012). On the determinants of profitability of Indian life insurers – an empirical
study”. Proceedings of the World Congress on Engineering I.
Chen, B., Powers, M.R., & Qiu, J. (2009). Life-insurance efficiency in China: a comparison of
foreign and domestic firms. China & World Economy, 17(6), 43-63.
Chen, R., & Wong, K. A. (2004). The Determinants of Financial Health of Asian Insurance
Companies. Journal of Risk and Insurance, 71(3), 469-499.
Chen, T. (1999). Pricing Private Health Insurance Products in China. Retrieved from
http://www.actuaries.org/IAAHS/Colloquia/Dresden/Tao%20 paper.pdf.
Claessens, S., & Laeven, L. (2004). What Drives Bank Competition? Some nternational
Evidence. Journal of Money, Credit, and Banking, 36, Part 2, 563-584.
Cummins, J. D., Rubio-Misas, M., & Vencappa, D. (2014). Competition, efficiency and
soundness in European life insurance markets. CAREFIN Working Paper/ May 2014.
Cummins, J. D., Weiss, M. A., Xie, X., & Zi, H. (2010). Economies of scope in financial
services: A DEA efficiency analysis of the US insurance industry. Journal of Banking &
Finance, 34(7), 1525-1539. doi: http://dx.doi.org/10.1016/j.jbankfin.2010.02.025
Cummins, J.D., & Weiss, M.A. (2000). Analyzing firm performance in the insurance industry
using frontier efficiency and productivity methods, in G. Dionne (ed) Handbook of
Insurance, Boston, MA: Kluwer Academic Publishers.
Cummins, J.D., Harrington, S.E., & Klein, R. (1995). Insolvency experience, risk based capital,
and prompt corrective action in property-liability insurance. Journal of Banking and Finance,
19, 511-527.
83
University of Ghana http://ugspace.ug.edu.gh
Cummins, J.D., Tennyson, S., & Weiss, M.A. (1998). Consolidation and efficiency in the US life
insurance industry. Journal of Banking and Finance, 23, 325-357.
Dambolena, I. G. & Khoury, J. S (1980). Ratio Stability and Corporate Failure. Journal of
Finance, 35(4), 1017-1026.
Das, A., & Ghosh, S. (2006). Financial deregulation and efficiency: An empirical analysis of
Indian banks during the post reform period. Review of Financial Economics, 15, 193-221.
Demsetz, H., & Lehn, K. (1985). The Structure of Corporate Ownership: Causes and
Consequences. Journal of Political Economy, 91(6).
Demsetz, H., & Lehn, K. (1985). The structure of corporate ownership: Causes and
consequences. Journal of political economy, 93(6), 1155-1177.
Domas, M. H (2003). The Public Interest Theory of Regulation: Non-Existence or
Misinterpretation? European Journal of Law and Economics, 15(2), 165-194.
Drake, P.P. (2013). Financial Ratio Analysis. James Madison University.
Eckardt, M., & Räthke-Döppner, S. (2008). The Quality of Insurance Intermediary Services-
Empirical Evidence for Germany. Andrássy Working Paper Series No. XXV.
Eduardo, K., Kimura, H., Basso, L. F., & Santos, D. F (2014). Innovation efforts and
performances of Brazilian firms. Journal of Business Research, 67(4), 527-535.
Eling, M., & Luhnen, M. (2010). Efficiency in the international insurance industry: A cross-
country comparison. Journal of Banking and Finance, 34, 1497-1509.
Essvale Corporation Limited (2010). www.safaribooksonline.com/library/view. [ Accessed on 3
may 2015].
Fama, D., & Jensen, M.C. (1983). Agency problems and residual claims. Journal of Law and
Economics, 26, 327-352.
Gaffikin, M., (2005). Regulation as Accounting Theory. Accounting and Finance Working
Papers, 50. [ Accessed on 2 march 2015].
Gatzert, N., & Martin, M. (2012). Quantifying credit and market risk under Solvency II:
Standard approach versus internal model. Insurance: Mathematics and Economics, 51(3),
649-666.
Georgantopoulos, A. G., & Tsamis, A. D. (2011). The Macroeconomic Effects of Budget
Deficits in Greece: A VAR-VECM Approach. International Research Journal of Finance and
Economics, 79, ISSN 1450-2887.
Ghana insurance law, act 724 (2006), allAfrica.com.
Ghana Statistical Service. (2013). Three-cycle Labour Force Report. [Accessed 3rd March 2015].
84
University of Ghana http://ugspace.ug.edu.gh
Ghosh, A. (2013). Does life insurance activity promote economic development in India: an
empirical analysis? Journal of Asia Business Studies, 7(1), 31-43.
Graphic (2011). NIC Wants Solvency Law to Deny Weak Insurers Bigger Business. Retrieved
from: http://www.modernghana.com/news/361443/1, Date Retrieved: [Accessed 8th July
2014].
Green (2008). Introduction to Regression Analysis. Third Edition, Singapore year book of
statistics. [Accessed on 12 February 2015].
Gregg, D.W. (1973). Life and health insurance handbook. McGraw-Hill/Irwin Publishers.
Grosen, A., & Jorgensen, P, L. (2002). Life Insurance Liabilities at Market Value: An Analysis
of Insolvency Risk, Bonus Policy, and Regulatory Intervention Rules in a Barrier Option
Framework. Journal of Risk and Insurance, 69(1), 63-91.
Guha-Khasnobis, B., & Ahuja, R. (2004). Extending formal insurance to the informal economy
workers in India. Emerald group publication.
Haiss, P., & Sümegi, K. (2008). The relationship between insurance and economic growth in
Europe: A theoretical and empirical analysis, 35, 405-431.
Hampton, J. J. (1993). Financial Management of Insurance Companies. (Amacom, NY:
American Management Association)
Hao, J. C. J., & Chou, L. Y. (2005). The estimation of efficiency for life insurance industry: The
case in Taiwan. Journal of Asian Economics, 16, 847-860.
Hao, J.C.J. (2007). Efficiency test on Taiwan’s life insurance industry using x efficiency
approach. Information and Management Sciences, 18(1), 37-48.
Herring, R.J., & Santomero, A.M. (2009). The role of the financial sector in economic
performance. The Wharton School, University of Pennsylvania, PA.
Hertog, J. A. (2010). Review of economic theories of regulation. Discussion Paper Series /
Tjalling C. Koopmans Research Institute, 10(18).
Hrechaniuk, B., Lutz, S., & Talavera, O. (2007). Do the Determinants of Insurers’ Performance
in “Old” EU and the “New” EU and Outside?
Hsiao, C. (2014). Analysis of Panel Data. Third Edition.
Htay, N. S., & Salman S. A. (2013). Viability of Islamic Insurance (Takaful) in India: SWOT
Analysis Approach. Review of European Studies, 5(4).
Huang, W., & Eling, M. (2013). An efficiency comparison of the non-life insurance industry in
the BRIC countries. European Journal of Operational Research, 226, 577-591.
85
University of Ghana http://ugspace.ug.edu.gh
IAIS (2003). Insurance core principles and methodology. International Association of Insurance
Supervisors, Singapore.
Implications for the capitalization of bank and non-bank financial institutions, Bank of Ghana
Consultation Paper October 2007.
IMF (2011). The Global Competitiveness Report. Available at: www.citeseerx.ist.psu. [Accessed
on 13 March 2015].
Insurance Act 2006. Available at: www.nic.org. [Accessed on 11 March 2015].
Jin, P. S., Lee, H., & Kang, B. (2009).The insurance distribution systems and efficiency in the
property-casualty insurance industry. Managerial Finance, 35(8), 670-681.
Joskow, P. L. (1973). Cartels, competition and regulation in the property liability insurance
industry. The Bell Journal of Economics, 4 (2), 375–427.
Kasturi, M. (2006). Improving call centre agent performance: A UK‐India study based on the
agents’ point of view. International Journal of Service Industry Management, 17(2), 136-
157.
Khan, M. S., & Senhadji, A. S. (2001). Financial Development and Economic Growth: A review
and new evidence. Journal of African Economies, 12(2), 89-110.
Kim, W., Mayers, D., & Smith, C. W. (1996). On the Choice of Insurance Distribution Systems.
Journal of Risk and Insurance, 63(2), 207-227.
Klumpes, P. J. (2004). Performance Benchmarking in Financial Services: Evidence from the UK
Life Insurance Industry. The Journal of Business, 77(2), 257-273.
KPMG (2014). Review of the Ghanaian insurance industry. [Accessed on 08 March 2015].
Kramer, B. (1996). An ordered Logit Model for the Evaluation of Dutch Non-Life Insurance
Companies. De Economist, 144, 79-91.
Kroszner, R. (1999). Is the Financial System Politically Independent?: Perspectives on the
Political Economy of Banking and Financial Regulation. Citeseer.
Kroszner, R. S. (1999). The Role of Private Regulation in Maintaining Global Financial
Stability. Cato Journal, 18(3), pp. 355-61.
Kroszner, R. S. (2001). The Motivations behind Banking Reform. Booth School of Business,
University of Chicago; National Bureau of Economic Research. Regulation, 24(2).
Leuvensteijin, M., Bikker, J. A., Van Rixtel, A. R. J. M., & Sorensen, C. K. (2011). A new
approach to measuring competition in the loan markets of the euro area. Applied Economics,
43, 3155-3167.
Levine, R. (1997). Financial Development and Economic Growth: Views and Agenda. Journal
of Economic Literature, XXXV, 688-726.
86
University of Ghana http://ugspace.ug.edu.gh
Locke, S. (2006). Regional Development Policy and Small Medium Enterprise Financial
Performance, ALTMAN 1968. Australasian Journal of Regional Studies, 12(2), 191-204.
Luhnen, M. (2009). Determinants of Efficiency and Productivity in German Property-Liability
Insurance: Evidence for 1995-2006. The Geneva Papers, 34, 483-505.
Malik, H. (2011). Determinants of Insurance Companies Profitability: An Analysis of Insurance
Sector of Pakistan. Academic Research International, 1(3), 315-321.
Management in the United Kingdom Reader in Law and member of the ESRC Centre for the
Analysis of Risk and Regulation, London School of Economics and Political Science. Available
at http://s3.amazonaws.com/academia.edu.documents/.
Mandal, S., & Dastidar, S. J. (2014). A DEA-investigation of efficiency of the Indian general
insurance during recession. Journal of Advances in Management Research, 11(1), 115-36.
Markides, C. C., & Williamson, P. J. (1994). Related diversification, core competences and
corporate performance. Strategic Management Journal, 15(2), 149-165.
McCullagh, P. (2002). What is a Statistical Model? The Annals of Statistics, 30(5), 1225-1310.
McDonald, J. B (1992). Identifying Financial Distress in the Insurance Industry: A Synthesis of
Methodological and Empirical Issues. Journal of Risk and Insurance, 59(4), 543-573.
Mishkin, F. S. (1992). Is the Fisher Effect for real? Journal of Monetary Economics, 30(2), 195-
215.
NIC (2005). The 2011 Annual Report, National Insurance Commission, available at:
www.nic.gh.org [Accessed on 25 May 2015].
NIC Supervision Department Report (2007). Available at http:www.nic.org [Accessed 3 March
2015].
NIC (2009). The 2011 Annual Report, National Insurance Commission, available at:
www.nic.gh.org.
NIC (2010). Annual Report, National Insurance Commission, available at: www.nic.gh.org.
NIC (2011). Annual Report, National Insurance Commission, available at: www.nic.gh.org.
NIC (2012). Life Insurance Companies. National Insurance Commission. Accra: Ghana.
Available at http://www.nicgh.org/live/en/?pg=127&pp=95 [Accessed on 04 March 2015].
NIC (2014). Life Insurance Companies. National Insurance Commission. Accra: Ghana.
Available at http://www.nicgh.org/live/en/?pg=127&pp=95 [Accessed on 10 July 2015].
Obersteadt, A., Bruning, C., Cude, B., Schwarcz, D., & Wilkonson, J. (2013). State of the Life
Insurance Industry: Implications of Industry Trends. National association of Insurance
Commissioners, CIPR Study.
87
University of Ghana http://ugspace.ug.edu.gh
Outreville, J. (1996). Life Insurance Markets in Developing Countries. The Journal of Risk and
Insurance, 63, 263-278.
Owusu-Ansah, E., Dontwi, I. K., Seidu, B., Abudulai, G., & Sebil, C. (2010). Technical
efficiencies of Ghanaian general insurers. American Journal of Social and Management
Sciences, 1(1).
Palande, P. S., Shah, R. S., & Lunawat, M. L. (2003). Insurance in India: Changing Policies and
Emerging Opportunities.
Park, J., Lee, S., & Bin Kang, H. ( 2009). The insurance distribution systems and efficiency in
the property-casualty insurance industry. . Managerial Finance, 35(8), 670-681.
Pasiouras, F., & Gaganis, C. (2013). Financial supervision regimes and bank efficiency:
International evidence. Journal of Banking and Finance, 37(12), 5463-5475.
Penrose, E. (1959). The theory of the growth of firms, Oxford: Basil Blackwell.
Perez, S. E., & Castillejo,  J. (2008). The Resource-Based Theory of the Firm and Firm Survival.
Small Business Economics, 30(3), 231-249.
Pervan, M., & Pavić, K.T. (2010). Determinants of Insurance Companies’ Profitability in
Croatia. The Business Review Cambridge, 16(1), 231-238.
Pervan, M., Curak, M., & Marijanovic, I. (2012). Dynamic Panel Analysis of Bosnia and
Herzegovina Insurance Companies’ Profitability.  Recent Researches in Business and
Economics, ISBN: 978-1-61804-102-9.
Pesaran, H. M., & Shin, Y. (1996). Cointegration and speed of convergence to equilibrium.
Journal of Econometrics, 71(1-2), 117-143.
Pesaran, H. M., Shin, Y., & Smith R. J. (2001). Bounds testing approaches to the analysis of
level relationships. Journal of Applied Econometrics, 16(3), 289-326.
Pessoa, J. P., & Reenen, V, J. (2014). The UK Productivity and Jobs Puzzle: Does the Answer
Lie in Wage Flexibility? The Economic Journal, 124(57), 433-452.
Petroni, A. (2000). The future of insurance industry in Italy: determinants of competitiveness in
the 2000s. Futures, 32, 417-434.
Pigou, A. C. (1932). The economics of welfare
Pistelis, G. (2008). An Overview on Solvency Supervision, Regulations and Prediction of
Insolvency. Belgian actuarial bulletin, 8(1).
Pistelis, G. (2008). Solvency supervision based on a total balance sheet approach. Journal of
Computational and Applied Mathematics, 233(1), 83-96.
88
University of Ghana http://ugspace.ug.edu.gh
Pistelis, J. (2009). Solvency Supervision Based on a Total Balance Sheet Approach. Journal of
Computational and Applied Mathematics, 233(1), 83-96.
Powell, L.S. (2007). Internal Versus External Capital Markets in the Insurance Industry: The
Role of Reinsurance. Journal of Financial Research, 31(2), 173-188.
Rao, M.S., & Srinivasulu, R. (2013). Contribution of Insurance Sector to Growth and
Development of the Indian Economy. Journal of Business and Management, 7(4), 45-52.
Rejda, G.  E. (1992).  Principles of Risk Management and Insurance.  Herper Collins Publishers,
New York.
Schrayer, R. M. & Rowh, M. (1999). Opportunities in Insurance Careers, NTC Publishing
Group, USA.
Sealey C. W., & Lindley, J. T. (1997). Inputs, Outputs, and a Theory of Production and Cost at
Depository Financial Institutions. Journal of Finance, 32(4), 1251-1256.
Sebiyam, M.  (2005).  The  impact  of  Insurance  on  Transport  Business  in  Ghana:  A  case
Study  in Sunyani Municipality. (Unpublished thesis). Sunyani Polytechnic, Sunyani, Ghana.
Shiu, Y.M. (2004). Determinants of United Kingdom general insurance company performance.
British Actuarial Journal, 10, 1079 -1110.
Simpson, S. N. Y., & Damoah, O. B.O. (2009). An Evaluation of Financial Health of Non-Life
Insurance Companies from Developing Countries: The Case of Ghana. 21st Australian
Finance and Banking Conference 2008 Paper. Available at SSRN:
http://ssrn.com/abstract=1138338 or http://dx.doi.org/10.2139/ssrn.1138338 [Accessed on 25
May 2015].
Skyes (2008). Fundamentals of Regression, McGrawhill Publishers.
Smith, K. (2004). The History of Life Insurance, Edward Elgar Publishers.
Swiss Re Publications SWISS RE SIGMA. (2013). World insurance in 2012: Progressing on the
long and winding road to recovery, Sigma Magazine [Accessed on 25 May 2015].
Swiss Re, http://media.swissre.com/documents/sigma3_2013_en.pdf [Accessed on 25 May
2015].
Swiss Re. (2014). World insurance in 2013: Steering world recovery’ Sigma No.3, Zurich: Swss
Re Publications, http://media.swissre.com/documents/sigma3_2014_en.pdfs [Accessed on 25
May 2015].
89
University of Ghana http://ugspace.ug.edu.gh
The Political Economy of Policy Reform edited by John Williamson (1994). Third Edition.
Thornhill, S., & Amit, R. (2003). Learning about Failure: Bankruptcy, Firm age and the
Resource-Based View. Organisation Science, 14(5), 497-509.
Vane, D. (1973). Life insurance market in Developing Countries. Journal of Risk and Insurance,
31, 391-401.
Wessels, B. (2009). International Insolvency Law. Available at:
Williamson, J. (1994). The Political Economy of Policy Reform. %I Institute for International
Economics, %U https://books.google.com.gh/books?id=cm-3mnvjEl8C.
Williamson, O. (1975). Markets and hierarchies: Analysis and antitrust implication. The Free
Press, New York, NY.
Worthington, A. C., & Hurley, E. V. (2002). Cost Efficiency in Australian General Insurers: A
Non-Parametric Approach. The British Accounting Review, 34(2), 89-108.
Yang, Z. (2006). A two-stage DEA model to evaluate the overall performance of Canadian life
and health insurance companies. Mathematical and Computer Modelling, 43, 910-919.
90
University of Ghana http://ugspace.ug.edu.gh
APPENDICES
APPENDIX A: R-CODES
install.packages("plm")
library(plm)
# For installing the panel regression packages in R
pdata<-plm.data(mydata, index=c("Company","Year"))
# This makes the R-program identify the data as panel, with the panel ID being the insurance co
mpany
pooling<-plm(SR~PG+LR+CI+IP+SZ+CR+OM,data=pdata, model="pooling")
summary(pooling)
# This specifies the regression model without assuming any panel
fixed<-plm(SR~PG+LR+CI+IP+SZ+CR+OM,data=pdata, model="within")
summary(fixed)
# This specifies the fixed effect model
random<-plm(SR~PG+LR+CI+IP+SZ+CR+OM,data=pdata, model="between")
summary(random)
# This specifies the random effect model
Hausman test
phtest(fixed,random) # For testing the best of fixed and random effect model
Testing for time-fixed effects
fixed.time<-plm(SR~PG+LR+CI+IP+SZ+CR+OM+factor(Year),data=pdata, model="within")
summary(fixed.time)
Testing for cross-sectional dependence correlation: using Breusch-Pagan LM test of
independence
install.packages("lmtest")
library(lmtest)
pcdtest(fixed, test = c("lm"))
Testing for serial correlation
pbgtest(random)
91
University of Ghana http://ugspace.ug.edu.gh
APPENDIX B
ORDINARY LEAST SQUARES MODEL
Table 1: Unbalanced OLS Regression estimates
Coefficients Estimate Standard Error t-value Pr(>|t|)
Intercept 3.54E+00 3.29E-01 10.77019 1.125E-4
PG 7.87E-07 2.47E-06 0.31911 7.015E+1
LR 1.29E-01 1.11E-02 11.68084 8.934E-4
CI -4.36E-05 2.80E-06 -15.59868 6.530E-4
IP 1.41E-03 9.48E-05 14.87908 9.832E-4
SZ 9.43E-06 5.30E-07 17.80219 8.571E-4
CR -4.29E-01 3.65E-02 -11.76051 5.620E-4
OM 1.95E-01 1.19E-02 16.36525 7.370E-3
INF -9.87E-03 4.53E-04 -21.79470 2.311E-4
RF 9.23E-04 9.85E-04 0.93696 8.721E-1
Table 2: ANOVA results (OLS)
Sum of squares Df F-statistics Pr(>|t|)
Regression 47.13 9 26.80 0.0005
Residual 17.60 90
Total 64.73 99
Table 3: Regression statistics (OLS)
R2 Adjusted R2 value
0.7280 0.5299
92
University of Ghana http://ugspace.ug.edu.gh
RANDOM EFFECT MODEL
4.6 Random Effect Panel Regression
Table 4: Unbalanced Random Panel Regression
Coefficien Estimate Standard t-value Pr(>|t|)
t Error
Intercept 1.32E+00 7.62E-02 17.2841 1.2341E-2
PG -2.04E-07 5.80E-07 -0.3521 9.9021E-1
LR 2.93E-01 2.01E-02 14.5771 9.3084E-3
CI -2.84E-05 1.82E-06 -15.6088 3.7128E-3
IP 4.12E-06 3.30E-07 12.4848 3.4981E-4
SZ 2.90E-07 1.78E-08 16.2949 3.6120E-5
CR -9.91E-01 8.27E-02 -11.9831 2.2401E-3
OM 9.48E-01 5.99E-02 15.8280 3.2347E-2
INF -4.56E-03 4.13E-04 -11.0412 2.0012E-3
RF 4.22E-04 5.89E-04 0.7170 2.0241E-1
Table 5: ANOVA results (Random effect)
Sum of squares Df F-statistics Pr(>|t|)
Regression 29.19 9 21.642 0.0023
Residual 11.24 75
Total 40.43 84
Table 6: Regression statistics (Random effect)
R2 Adjusted R2 value
0.7220 0.5213
93