The application of demographic projections to predict changes in life and non-life insurance penetration rates

Transcription

1 The application of demographic projections to predict changes in life and non-life insurance penetration rates By Christine van Heerden and Lance Moroney Presented at the Actuarial Society of South Africa s 2016 Convention November 2016, Cape Town International Convention Centre ABSTRACT The current penetration rates of life and non-life insurance in South Africa, Kenya, Ghana, Zambia and Nigeria are considered along with influencing factors. The influence of economic, regulatory, industry, cultural and demographic factors on current penetration levels are considered. Given the projected levels of population growth in the selected countries, the paper explores how a wider set of demographic factors will be affected by this, and how this will affect future insurance demand. Publicly available population projections are used for the purpose of projecting a wider range of demographic factors including: disposable income levels per person, spread of income levels, unemployment, dependency ratios, and poverty levels. This paper does not attempt to accurately quantify future penetration rates and all of the demographic factors considered, but rather explores limitations to insurance growth imposed by demographic factors of the selected countries. Finally, key challenges to improving insurance penetration in these markets are considered. KEYWORDS Africa; insurance; penetration; demographics CONTACT DETAILS Christine van Heerden, QED Actuaries & Consultants (Pty) Ltd, Sandton Christine.van.Heerden@qedactuarial.co.za 331

2 332 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES 1. introduction and background 1.1 Study Objectives This study aims to explore the impact of demographic projections on insurance penetration in coming decades. Life and non-life insurance are considered in this paper A substantial level of population growth is expected across many countries in Africa over the coming decades. The impact of projected population growth in the selected countries on dependency ratios and other demographic factors are considered which in turn should impact the level of affluence and affordability of insurance Projections over three decades are considered starting from The study does not attempt to accurately quantify future penetration rates, and all of the demographic factors considered, but rather explores limitations to insurance growth imposed by demographic factors of the selected countries The countries considered in this paper are South Africa, Ghana, Kenya, Nigeria and Zambia. These were selected due to the significance of their insurance industries for actuaries practising in South Africa and the rest of Africa. The approach and methodology taken in this study can be applied to other countries in Africa. 1.2 Study Approach and Methodology The study begins by considering economic, regulatory, industry, cultural and demographic factors that are expected to influence overall and product-specific penetration in any country. Global values for a selection of the factors are also shown. Current data on the chosen factors are then shown for each of the five countries included in the study. The influence of the demographic and other factors on insurance penetration levels is discussed for each country Population, together with a wider set of demographic factors, is then projected for each country. It is necessary to apply assumptions for economic growth to indicate future levels of affluence, and therefore affordability of insurance. This is followed by a discussion of the influence of the projected data on future insurance penetration. Finally, key challenges to improving insurance penetration in these markets are considered. 2. Influencing factors 2.1 The influence of economic, regulatory, industry, cultural and demographic factors on current penetration levels were considered for each country. This section details the data considered in each of these categories, as well as the rationale for

3 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES 333 choosing the data to be considered. In all cases, the data considered was chosen as it was expected to influence the insurance penetration rates, either of the insurance market as a whole, or of a particular line of business. Penetration rates are shown at the end of the section for each country. We have considered 2013 as this is the most recent year where penetration data are available for each of the countries considered. 2.2 Economy The gross domestic product (GDP) of each country was considered in US dollars, as well as the real GDP growth percentage and GDP per person in US dollars. As at 20 September 2016, the rand dollar exchange rate was R13.99 per US$ The GDP is an indicator of the health and size of the country s economy. The World Bank (2016d) defines GDP as follows: GDP is the sum of gross value added by all resident producers in the economy plus any product taxes and minus any subsidies not included in the value of the products. It is calculated without making deductions for depreciation of fabricated assets or for depletion and degradation of natural resources A healthy economy results in low unemployment rates and wage increases (Investopedia, 2016b). The GDP of the world was US$ billion in 2013, real GDP growth was 2.40% and real GDP growth per person was 1.16% (World Bank, 2016d). The GDP per person in 2013 was US$ (World Bank, 2016d) The GDP in dollars is influenced by the real growth in GDP, inflation and exchange rate fluctuations. When considering past GDP, all of these factors will impact the level of the GDP. However, projections of the GDP are done in real terms, based on the 2013 GDP, thus removing the impact of inflation and exchange rate fluctuations in the projections GDP is broken down by sector of the economy as at 2013 for each country. Industries are defined based on the International Standard Industrial Classification of All Economic Activities, Revision 3.1 (ISIC) (World Bank, 2016d). Agriculture, industry, and services are considered, as well as manufacturing, which is a subset of industry. Full details of how these classifications are determined are shown in Appendix A In 2013, 3.98% of the world s GDP was from agriculture and 27.94% was from industry. The remaining 68.08% of GDP was from services (World Bank, 2016d). Table 1 shows the composition of the World s GDP from 2010 to 2013.

4 334 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Table 1 World GDP by sector (World Bank, 2016d) Sector Agriculture 3.9% 3.9% 3.9% 4.0% Industry 28.5% 28.8% 28.4% 27.9% of which manufacturing 16.8% 16.5% 16.3% 16.3% Services 67.6% 67.3% 67.7% 68.1% Where a country s GDP is concentrated in one sector or division the country is exposed to a decrease in global demand for products from that sector resulting in a large decrease in the GDP. A spread of sectors generating GDP results in a more stable outlook for GDP The gross national income (GNI) is defined by the World Bank (2016d) as follows: GNI (formerly GNP) is the sum of value added by all resident producers [in one calendar year] plus any product taxes (less subsidies) not included in the valuation of output plus net receipts of primary income (compensation of employees and property income) from abroad We consider the GNI per country as at 2013 in US dollars, the GNI per person and the GNI per employed person. The world GNI in 2013 was US$ billion and the GNI per person was US$ (World Bank, 2016d). The global GNI per employed person in 2013 was US$ The World Bank (2016d) has published definitions for four income levels, namely: low income (L), lower-middle income (LM), upper-middle income (UM), and high income (H). These levels are based on the income earned by a person per year as at The scale is updated by the World Bank each year. Table 2 shows the definition of each of the income levels as at Table 2 Definition of income categories Income category Income L Up to US$1 045 LM US$1 046 US$4 125 UM US$4 126 US$ H Greater than US$ The GNI per person for each country is used to classify the income level of the countries.

5 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES The wealth distribution in each country is considered by showing the share of wealth held by each quintile of the population. For each quintile we show the average GNI per capita. The GNI per person for each quintile is used to classify each quintile of the population by income For each country the Gini index, which measures the extent to which the distribution of income among individuals in an economy deviates from a perfectly equal distribution, is considered. A Gini index closer to 100 represents greater skewness in the income distribution, while an index of zero represents a perfectly equal distribution of income. The Gini index is not calculated for the world as a whole. 2.3 Regulatory Environment The regulatory environment of a country is a key influence in the insurance market. A competent regulator supported by sound legislation will instil confidence in the insurance industry, which will increase the demand for insurance. Regulators may also make some forms of insurance compulsory, which results in a higher penetration of those insurance products. Restrictions may be imposed on insurance products and companies by the regulator. This may influence the insurance market in different ways depending on the regulation For each country the regulator of life and non-life insurers is stated and the date of establishing the regulatory body is provided. A list of compulsory insurance products is also provided. 2.4 Insurance Industry The insurance industry may include nationwide funds or insurers, where membership is compulsory. Where nationwide funds exist to provide some level of insurance cover, the cover provided by insurers may supplement and complement that offered nationally. Alternatively, insurers will not offer these types of products, as the public already have access to these products In some markets, life insurance products are used as savings vehicles. This results in an increase in the life insurance penetration rates. For each country the use of life insurance products for savings purposes is described The number of life and non-life insurers in each industry is also provided. In some cases, we also consider the number of composite insurers. 2.5 Culture The propensity which a population has to save will impact the demand for insurance products. The rate of savings as a percentage of GDP was considered as an indicator of a population s propensity to save, where savings are calculated as the

6 336 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES national income, less consumption plus net transfers. This will be partly impacted by the level of wealth and disposable income of the population, as well as regulations in the country. However, it is a readily available measure to compare each of the five countries. Savings as a percentage of GDP for the world was 22.53% in 2013 (World Bank, 2016d) The corruption perception index is a measure of how corrupt a country s public sector is perceived to be. The score is on a scale of 0 to 100, where 0 means the country is highly corrupt and 100 means that the country is perceived as very clean (Transparency International, 2016). The corruption perception index was used as a proxy for fraud in each country. High levels of fraud create distrust in insurers, which result in a lower level of demand for insurance. In 2013 the average of all countries corruption perception index scores was 43. The average corruption perception index score for the countries considered in this paper was The literacy rate, which is defined as the percentage of people aged 15 years and older who can both read and write with understanding a short simple statement about their everyday life, was 85% for the world in For males the literacy rate was 89% and the literacy rate for females was 81% in 2010 (World Bank, 2016a). Given the complexity of even the simplest insurance products, and that the terms and conditions of the products are recorded in writing, it is unlikely that illiterate individuals will have a strong desire to purchase insurance products. 2.6 Demographics The population pyramid as at 2013 is shown for each of the five countries, along with key demographic information Figure 1 shows four stages of population pyramids The wide base of Stage 1 shows a high birth rate. There is a rapid decline in each upward age because of high death rates. The high death rates result in a short life expectancy, which is shown by the narrow top of the pyramid. The high birth and death rates result in a stationary population size. Figure 1 Four stages of population pyramids (Lawrence, 2016)

7 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Stage 2 also has a wide base, indicating high birth rates. However, the rate of decline as ages increase is more gradual than that in Stage 1, indicating a lower mortality rate and a higher life expectancy. The high birth rate and average mortality result in an increasing population size Stage 3 shows a late expanding population. The birth rate has gradually decreased, and the mortality rate is low. This results in a greater proportion of the population being older In Stage 4 the birth rate and mortality rate are low, resulting in a base which is narrower than other parts of the pyramid When interpreting population pyramids a wide base indicates a high birth rate and a narrow base indicates a decreasing birth rate. Straight or near vertical sides show a low death rate and a concave slope suggests a high death rate. Bulges in the slope indicate high immigration rates. Indents in the slope indicate emigration or age or sex specific deaths Figure 2 shows the population pyramid of the world as at The world s population in 2013 was 7.1 billion people (World Bank, 2016d). A slight bulge can be seen in the 20-to-29-year age bands. The size of the working population and the number of people who depend on them impacts the disposable income, which impacts the demand for insurance. Employment statistics are therefore also considered. Figure 2 Population pyramid for the world in 2013 (World Bank, 2016d)

8 338 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES The unemployment rate, as measured by the International Labour Organisation (ILO), and defined as the share of labour force that is without work but available for and seeking employment, was 6.0% in 2013 (World Bank, 2016d) The dependency ratio is the ratio of dependants, who are people younger than 15 years or older than 64 years, to the working age population (World Bank, 2016a). For the world the dependency ratio was 0.54 in 2013 (World Bank, 2016a) The ratio of the number of non-working people in a population, which includes dependants as well as those who are of working age but unemployed, to the employed population, we refer to as the employment dependency ratio. The world employmentto-population ratio, which is the proportion of the country s population aged 15 years and older which is employed, as measured by the ILO, was 59.6% in 2013 (World Bank, 2016d). This implies that there were 3.1 billion employed people in the world and hence 4.0 billion unemployed people (including people of non-working age), which equates to one employed person supporting 1.28 unemployed people The actual fertility rate per wealth quintile of the population, as well as the wanted and ideal fertility rate per quintile, is shown for each country. The actual fertility rate is defined as the number of children that would be born to a woman if she were to live to the end of her childbearing years and bear children in accordance with age-specific fertility rates observed in the three years preceding the effective date. The wanted fertility rate is an estimate of what the total fertility rate would be if all unwanted births were avoided and is thus a subset of the actual fertility rate. The ideal fertility rate is the mean ideal number of children for all women (World Bank, 2016b) The actual fertility rate gives an indication of the number of child dependants per wealth quintile. The ideal fertility rate shows cultural differences, where some wealth quintiles show a preference for more or less children. The difference between the actual and wanted fertility rate indicates whether the actual fertility rate may be because of a lack of family planning or contraceptive availability The ideal fertility rate may in some cases be higher than the actual fertility rate as some women are unable to bear children In general, developing countries have higher birth rates because of parents wanting children for labour, to take care of them in their old age, to continue the family name, for prestige, as children are regarded as net contributors to family income and to replace children who have died (Geography for Nature Lovers, 2016) Developed countries have lower birth rates because of children being costly, the government taking care of people in their old age through pensions and health

9 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES 339 services, more women wanting a career and a higher status, widespread use of family planning, less need to replace children because of low infant mortality and urbanisation being associated with social change and a decline in traditional beliefs and customs (Geography for Nature Lovers, 2016) Fertility rates by wealth quintile are not available for the world. 2.7 Penetration Rates The penetration rates for life and non-life insurance are provided for each country, by product type where available. Penetration rates are the gross insurance premiums written as a percentage of a country s GDP. In 2013 the global life insurance penetration was 3.39% and the non-life insurance penetration was 2.74% (Swiss Re, 2016). 3. South Africa 3.1 An overview of the South African market as at 2013 followed by commentary on the penetration levels is provided in this section. 3.2 Economy The GDP of South Africa in 2013 was US$366 billion. The GDP per person in 2013 was US$ Real GDP growth in 2013 was 2.21%, while the real growth in GDP per person was 0.61%, both of which are lower than the world s GDP growth in 2013 (World Bank, 2016d). Table 3 shows the GDP by sector for South Africa from 2010 to Table 3 South African GDP by sector (World Bank, 2016d) Sector Agriculture 2.6% 2.5% 2.4% 2.3% Industry 30.2% 29.9% 29.7% 29.9% of which manufacturing 14.4% 13.3% 13.1% 13.2% Services 67.2% 67.6% 67.9% 67.8% Services contribute the most to the GDP, and the portion of GDP from to services has grown from 2010 to Agriculture has the smallest contribution to GDP, and has fallen from 2.6% of GDP in 2010 to 2.3% of GDP in The GDP from industry is greater than the world average, while agriculture and services are lower than the world average South Africa s GNI in 2013 was US$356 billion. The GNI per person in 2013 was US$7 400, which places South Africa in the upper-middle income group (World Bank, 2016d). In South Africa in 2013 the GNI per employed person was US$

10 340 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Table 4 shows the GNI per person from 2009 to 2013 in South Africa. Table 4 South African GNI per person 2009 to 2013 (World Bank, 2016d) Year GNI per person Income category 2009 US$5 910 UM 2010 US$6 250 UM 2011 US$7 050 UM 2012 US$7 640 UM 2013 US$7 400 UM South Africa s GNI per person has equated to South Africa being classed as an upper-middle income country for the five years from 2009 to The GNI per person increased from 2009 to 2012 and then dropped from 2012 to This is the only country considered in this paper where the GNI per person has decreased from the previous year Table 5 shows the distribution of income between the population of South Africa in 2011, from the World Bank (2016c), where the population has been divided into wealth quintiles. The 20% of the population with the highest income receive 68.95% of South Africa s wealth. The GNI per person for each of the quintiles, as at 2013, is also shown, assuming that the income distribution remains constant from 2011 to The income category for each quintile is also shown. The quintiles are referred to as Q1 to Q5, where Q1 is the lowest earning quintile and Q5 is the highest earning quintile. Table 5 Distribution of wealth between the South African population Population Percentage of income share 2011 GNI per person 2013 Income category 2013 Q1 2.47% US$828 L Q2 4.71% US$1 579 LM Q3 7.97% US$2 672 LM Q % US$5 331 UM Q % US$ H The Gini index for South Africa in 2011 was 63.4 (World Bank, 2016c). The Gini index, as well as the split of wealth shown in Table 5, indicate that there is an inequitable spread of income in South Africa Of South Africa s population, 40% are upper-middle or higher income and would likely consider purchasing insurance products. The other 60% fall in the lower-middle

11 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES 341 and low income categories. This portion of the population has a constrained income, which needs to meet many expenses, hence they are unlikely to consider purchasing insurance products, as there is no income available for these purchases. 3.3 Regulatory Environment Life and non-life insurers are regulated by the Financial Services Board (FSB) which was founded more than 20 years ago (Financial Services Board, 2016). The FSB also regulates pension funds (Financial Services Board, 2016). Medical schemes are separately regulated by the Council for Medical Schemes (Council for Medical Schemes, 2016). Life and non-life insurers may offer medical products, provided they meet certain conditions, the most notable being that the products may not indemnify the policyholders and benefits must be fixed amounts It is compulsory for businesses to have public liability insurance in place, to cover accidents which may occur at the business (Sky Insurance, 2016). 3.4 Insurance Industry There are four acts in South Africa which resulted in the formation of bodies which offer nation-wide industry cover. These are the Road Accident Fund (RAF) Act, the Unemployment Insurance Act, the Compensation for Occupational Injuries and Diseases Act (COIDA) and the Occupational Diseases in Mines and Works Act (ODMWA) The Road Accident Fund (2016) states that the RAF provides the following cover: The RAF provides compulsory cover to all users of South African roads, citizens and foreigners, against injuries sustained or death arising from accidents involving motor vehicles within the borders of South Africa. This cover is in the form of indemnity insurance to persons who cause the accident, as well as personal injury and death insurance to victims of motor vehicle accidents and their families The RAF is funded through a fuel levy, expressed as a number of cents per litre, which is collected on all petrol and diesel sold in South Africa (Sky Insurance, 2016) The Department of Labour (2016b) states the following regarding the Unemployment Insurance Fund (UIF): The Unemployment Insurance Act provides protection to workers who become unemployed. It prescribes claiming unemployment benefits for unemployment, maternity benefits, illness benefits, adoption benefits and dependents benefits Employees and employers each contribute 1% of the employee s salary, subject to an upper threshold, to the UIF. Membership to the UIF is compulsory for all employees employed in the formal sector as well as domestic workers who are employed for more

12 342 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES than 24 hours a month with a particular employer (Unemployment Insurance Act, 2001) COIDA and ODMWA allow for compulsory workmen s compensation. While COIDA and ODMWA differ by who is covered and the level of benefits, together they ensure that they provide workmen s compensation. The Department of Labour (2016a) states the following regarding COIDA: To provide for compensation for disablement caused by occupational injuries or diseases sustained or contracted by employees in the course of their employment, or for death resulting from such injuries or diseases; and to provide for matters connected therewith Acts Online (2016) states the following regarding ODMWA: To consolidate and amend the law relating to the payment of compensation in respect of certain diseases contracted by persons employed in mines and works and matters incidental thereto Life insurers include a savings component in many of their products and life products are thus used for savings purposes as well as insurance purposes. In 2013 there were 74 and 97 primary life and non-life insurers registered with the FSB respectively (Financial Services Board, 2013a; 2013b). 3.5 Culture Savings as a percentage of GDP in South Africa in 2013 was 14.35%. This is significantly lower than the world s savings as a percentage of GDP of 22.53% (World Bank, 2016d) South Africa s corruption perception index in 2013 was 42 (Transparency International, 2016), only slightly worse than the average of 43 for all countries in the world The South African literacy rate was 94% in The literacy rate was 95% and 93% for males and females respectively (World Bank, 2016a). South Africa has the highest literacy rate of the countries considered in this paper, exceeding the world literacy rate of 85%. Furthermore, the gap between the male and female literacy rate is 2%, which is the smallest gap of the countries considered in this paper, and lower than the world gap of 8% In South Africa, a high level of importance is placed on funerals, which results in a large amount of money being spent on funerals (O Keeffe & Swartz, 2016; Case et al., 2013; Roth, n.d.). This may be expected to increase the demand for funeral cover. However, there are informal insurers and burial societies, which offer funeral cover,

13 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES 343 often by providing goods for the funeral as well as cash (Roth, n.d.). Some residents prefer these informal societies and insurers (Roth, n.d.), hence the penetration of funeral insurance may not be as high as initially anticipated. 3.6 Demographics Figure 3 shows the population pyramid for South Africa in 2013, based on data from the World Bank (2016a). The population in South Africa in 2013 was 53 million people (World Bank, 2016a) The pyramid has a bulge in the 15-to-34-year age range. This is most likely because of immigration from neighbouring countries of people of a young working age who are seeking employment. From 2008 to 2012 the average net migration rate to South Africa was people over the five-year period (World Bank, 2016d). The average net migration from 1993 to 2012 was 3.2 million people (World Bank, 2016d). Net migration is the number of immigrants less the number of emigrants. From age 35 years the population gradually decreases. This indicates a nation which is transitioning from a developing country to a developed country. The convex sides of the pyramid indicate a low death rate. The dependency ratio in South Africa in 2013 was The unemployment rate was 24.6% in 2013 (World Bank, 2016d). In comparison to the world unemployment rate of 6.0%, South Africa s unemployment is high. Of those employed, 9.3% are in vulnerable employment, which is defined as unpaid family workers and own-account workers (World Bank, 2016d). Figure 3 Population pyramid for South Africa in 2013

14 344 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES The employment-to-population ratio was 39.3% in 2013 (World Bank, 2016d). Recall that the employment-to-population ratio is the proportion of the country s population aged 15 years and older which is employed. In South Africa in 2013 there were 37.3 million people aged 15 years and older (World Bank, 2016d). This implies that there were 14.6 million employed people, which equates to an employeddependency ratio of In 2011 there were 17.9 million people in South Africa who were living on less than $3.10 a day, based on 2011 purchasing parity power (PPP). Of these, 8.5 million were living on less than $1.90 per day, based on 2011 PPP (World Bank, 2016c) Figure 4 shows the actual, wanted and ideal fertility rate by wealth quintile for South Africa effective as at There is a clear trend across the actual, wanted and ideal fertility rates, where Q1 has the highest fertility rate and this decreases to Q5 which has the lowest fertility rate. The difference between the ideal number of children for Q1 of 3.5 and the ideal number of children for Q5 of 2.4 is 1.1. However, the actual fertility rate has a much larger difference, where the actual fertility rate for Q1 is 4.8 children per woman and only 1.9 children per woman in Q5. The actual fertility rate for Q5 is below the ideal rate, while for Q1 the actual fertility rate exceeds the ideal rate The high actual fertility rate for Q1 spreads the income of the working population in this wealth quintile among more people, as there are a greater proportion of dependants in this quintile compared to the other quintiles. Figure 4 South African fertility rates

15 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Penetration Rates Insurance penetration rates have been determined based on data from Financial Services Board (2013a), Financial Services Board (2013b) and the World Bank (2016d) The life insurance penetration rate in South Africa in 2013 was 10.57% and the non-life insurance penetration was 2.67%. Table 6 and Table 7 show the split of the penetration rates by line of business for life and non-life insurance respectively. This is ordered in descending penetration level. Table 6 South African life penetration rates by line of business Line of business Penetration rate Fund business 5.04% Life 4.82% Sinking fund 0.24% Disability 0.18% Assistance 0.17% Health 0.12% Total 10.57% Health business has the lowest penetration rate of the life business lines. This is likely because of a high portion of the population making use of medical schemes, which are separate to the life and non-life insurers Life business includes retirement annuities, which are used to provide a pension in retirement Fund business has the highest penetration. This high penetration is likely because of individuals making use of these products for savings purposes. The South African population places a high level of trust in insurance companies, and hence life insurance products are used for investment purposes, primarily because of their simplicity for the client and the returns provided, in comparison to other investments Motor insurance is not compulsory in South Africa. However, it is the highest penetrating non-life insurance product. Many cars bought in South Africa are financed with loans, which require insurance for the vehicle. Furthermore, the premiums are paid monthly and are thus affordable for most. These are likely to be the key influencers on why motor insurance has the highest penetration in the South African non-life market.

16 346 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Table 7 South African non-life penetration rates by line of business Line of business Penetration rate Motor 1.11% Property 0.91% Accident and health 0.18% Liability 0.11% Miscellaneous 0.11% Engineering 0.09% Guarantee 0.08% Transportation 0.08% Total 2.67% 4. Ghana 4.1 An overview of the Ghanaian market as at 2013 followed by commentary on the penetration levels is provided in this section. 4.2 Economy The GDP of Ghana in 2013 was US$48 billion. The GDP per person in 2013 was US$ Real GDP growth in 2013 was 7.31%, while GDP growth per person was 4.77%, both significantly greater than the world GDP growth of 2.40% and 1.16% per person (World Bank, 2016d) Table 8 shows the GDP by sector for Ghana from 2010 to Table 8 Ghanaian GDP by sector (World Bank, 2016d) Sector Agriculture 30.8% 26.0% 23.6% 23.2% Industry 19.8% 26.2% 28.9% 28.7% of which manufacturing 7.0% 7.1% 6.0% 5.5% Services 49.4% 47.8% 47.5% 48.1% Services contribute the most to the GDP. The portion of GDP from industry is increasing, while the proportion of GDP from agriculture dropped from 30.8% in 2010 to 23.2% in The level of GDP from agriculture is significantly greater than that of the world where agriculture is only 4.0% of GDP. Services are much lower than the world average, while industry is comparable with the world average.

17 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Ghana s GNI in 2013 was US$46 billion. The GNI per person in 2013 was US$1 740, which places Ghana in the lower-middle income group (World Bank, 2016d). In Ghana in 2013, the GNI per employed person was US$ Table 9 shows the GNI per person from 2009 to 2013 for Ghana. The income categorisation based on the GNI per person is also shown. Table 9 Ghanaian GNI per person 2009 to 2013 (World Bank, 2016d) Year GNI per person Income category 2009 US$1 210 LM 2010 US$1 260 LM 2011 US$1 410 LM 2012 US$1 570 LM 2013 US$1 740 LM Ghana s GNI per person has equated to Ghana being classed as a lower-middle income country for the five years from 2009 to The GNI per person increased from 2009 to Table 10 shows the distribution of income between the population in 2005, from the World Bank (2016c), where the population has been divided into wealth quintiles. The 20% of the population with the highest income receive 48.59% of Ghana s wealth. The GNI per person for each of the quintiles, as at 2013, is also shown, assuming that the income distribution remains constant from 2005 to The income category for each population section is also shown. Table 10 Distribution of wealth between the Ghanaian population Population Percentage of income share GNI per person Income category Q1 5.24% US$465 L Q2 9.89% US$878 L Q % US$1 299 LM Q % US$1 922 LM Q % US$4 313 UM Of Ghana s population 20% are classified as upper-middle earners and would likely consider purchasing insurance products. The other 80% fall in the lower-middle and low income categories. This portion of the population has a constrained income,

18 348 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES which needs to meet many expenses, hence they are unlikely to consider purchasing insurance products, as there is no income available for these purchases The Gini index for Ghana in 2005 was 42.8 (World Bank, 2016c). 4.3 Regulatory Environment The National Insurance Commission regulates life and non-life insurers in Ghana and was established under the Insurance Law 1989 (National Insurance Commission, 2016c). Pension funds are regulated by the National Pensions Regulatory Authority (National Pensions Regulatory Authority, 2016b), although some life insurers offer pension and annuity products (National Insurance Commission, 2016d) In Ghana, it is compulsory for commercial buildings under construction to be insured against injury, death or damage to property of any workman on site, or member of the public ( Act firmly on compulsory insurance, 2015; Insurance Act, 2006). All commercial buildings must be insured against damage to the building as well as third-party cover (Insurance Act, 2006) Third-party motor insurance is compulsory for anyone who uses a vehicle on a public road (National Insurance Commission, 2016b) Clinical trials liability insurance, professional indemnity for insurance brokers and loss adjustors, transit bonds for goods intended for transshipment, liability against oil pollution of Ghanaian waters by vessel owners and marine imports, and aviation liability are also compulsory insurances (Alexander Forbes, 2016a) The Commissioner of Insurance set up a committee in 2015 to recommend classes of insurance which should be made compulsory. The Commissioner of Insurance said that compulsory insurances are required to compensate innocent third parties who may suffer injury or damage to property ( NIC advocates, 2015). 4.4 Insurance Industry The Ghanaian Insurance Act, 2006, makes provision for a Fire Service Maintenance Fund, Motor Compensation Fund and Client Rescue Fund The Fire Service Maintenance Fund provides funds and equipment to state institutions for the purpose of fighting fires. This is funded through a percentage of the gross premiums received by direct insurers for commercial buildings products (Insurance Act, 2006) The Motor Compensation Fund compensates individuals who suffer injury or death through a motor accident and are unable to claim compensation from an

19 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES 349 insurance company. This is funded through a percentage of the contributions received for the issue of a sticker which indicates that a vehicle is insured (Insurance Act, 2006) The Client Rescue Fund was established to compensate clients of insurance companies who go bankrupt. This is funded through a percentage of the gross premiums received by each direct insurer (Insurance Act, 2006) The pension structure is a three-tiered system. The first tier is compulsory for all employees, while the second tier is compulsory for all employees but is privately managed. The third tier consists of voluntary provident funds and pension schemes (National Pensions Regulatory Authority, 2016a) Ghana has a National Health Insurance Scheme which provides healthcare to the population. Formal sector employees and vulnerable people do not pay a premium for the services, while employees in the informal sector pay a premium to belong to the scheme. Formal sector employees contribute to the Social Security and National Insurance Trust. Of these contributions, 2.5% is allocated to the National Health Insurance Scheme (National Health Insurance Scheme, 2016a) In 2013, there were 18 and 25 life and non-life insurers respectively (National Insurance Commission, 2013) Some ordinary life business includes a savings component. 4.5 Culture Savings as a percentage of GDP in Ghana in 2013 was 13.65% which is lower than the world average of 22.53% (World Bank, 2016d) Ghana s corruption perception index was 46 in 2013 (Transparency International, 2016), slightly better the average of 43 for all countries in the world The Ghanaian literacy rate was 71% in The literacy rate was 78% and 65% for males and females respectively (World Bank, 2016a). The literacy rate for Ghana is below the world s literacy rate of 85%, and the gap between male and female literacy rates of 13% is greater than the world s gap of 8% In Ghana, the financial needs of an individual increase after retirement, because of family engagements, family leadership and other social responsibilities. For this reason, the income replacement ratio required in retirement is greater than in other countries (National Pensions Regulatory Authority, 2016a).

20 350 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES 4.6 Demographics Figure 5 shows the population pyramid for Ghana in 2013, based on data from the World Bank (2016a). The population in Ghana in 2013 was 26 million people (World Bank, 2016a) The wide base and narrow top of the population pyramid indicate an increasing population with a high number of child dependants and a low life expectancy. This is characteristic of a developing country. The dependency ratio in Ghana in 2013 was The unemployment rate was 1.8% in 2013 (World Bank, 2016d). This is very low in comparison to the world average of 6.0%. However, in 2010, 76.8% of those employed were classified as being in vulnerable employment The employment-to-population ratio was 68.1% in 2013 and there were 16.0 million people aged 15 years and older in 2013 (World Bank, 2016d). This implies that there were 10.9 million employed people, which equates to an employed dependency ratio of In 2005 there were 10.5 million people in Ghana who were living on less than $3.10 a day, based on 2011 PPP. Of these, 5.4 million were living on less than $1.90 per day, based on 2011 PPP (World Bank, 2016c) Figure 6 shows the actual, wanted and ideal fertility rate for Ghana by wealth quintile. The actual fertility rate is effective as at 2011, while the wanted and ideal fertility rates are as at Figure 5 Population pyramid for Ghana in 2013

21 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES The same downward pattern is observed as in South Africa, where the fertility rates are highest for Q1 and decrease to Q5 which has the lowest rates. The ideal and wanted number of children shows a jump from Q2 to Q1. The actual number of children shows a steady decrease from Q1 to Q On average, women in Q1 have 3.1 more children than those in Q5. While this is in line with what women in Q1 wanted and find ideal, it results in greater dilution of the income for individuals in the poorer quintiles. 4.7 Penetration Rates Insurance penetration rates have been determined based on data from the National Insurance Commission (2013) and the World Bank (2016d) The life insurance penetration rate in Ghana in 2013 was 0.50% and the non-life insurance penetration was 0.62%. The low levels of penetration are likely because of the low income levels in Ghana. Only the top quintile has a per person GNI which places them in the upper-middle income class. For the other four quintiles we do not expect that they would be interested in purchasing insurance because of their limited income The main lines of business for life insurance in Ghana are universal life, funeral, whole life, endowment, term and group life, where individual life is the predominant line. Table 11 shows the split of the penetration rates by line of business for non-life insurance. This is ordered in descending penetration level. Figure 6 Ghanaian fertility rates

22 352 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Table 11 Ghanaian non-life penetration rates by line of business Line of business Penetration rate Motor 0.28% Fire, burglary and property 0.14% General accident 0.07% Marine hull, cargo and aviator 0.05% Engineering 0.03% Bonds 0.02% Liability 0.02% Other 0.01% Total 0.62% Other includes travel insurance. General accident includes workmen s compensation and personal accident (National Insurance Commission, 2013) Recall that in Ghana, third-party motor insurance and commercial property insurance are compulsory. This is likely why motor and fire, burglary and property insurance have the highest levels of penetration in Ghana. The other insurance classes listed in Table 11 are likely to be predominantly purchased by companies rather than individuals Healthcare does not appear in the list of classes of non-life penetration. As Ghana has a National Health Insurance Scheme which provides healthcare to the population the demand for this insurance is low. 5. Kenya 5.1 An overview of the Kenyan market as at 2013 followed by commentary on the penetration levels is provided in this section. 5.2 Economy The GDP of Kenya in 2013 was US$55 billion. The GDP per person in 2013 was US$ Real GDP growth in 2013 was 5.69%, while growth in GDP per person was 2.91%, both above the world average (World Bank, 2016d) Table 12 shows the GDP per sector from 2010 to 2013 for Kenya Services contribute the most to the GDP, although the portion of GDP from services fluctuated from 2010 to The portion of GDP from agriculture has increased since 2010, while the portion from industry has decreased slightly, although

23 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES 353 fluctuating from 2010 to However, a higher portion of Kenya s GDP is from agriculture than seen in the world where only 4.0% of GDP is from agriculture. Table 12 Kenyan GDP by sector (World Bank, 2016d) Sector Agriculture 27.8% 29.3% 29.1% 29.5% Industry 20.8% 21.0% 20.7% 19.9% of which manufacturing 12.6% 13.1% 12.3% 11.7% Services 51.4% 49.7% 50.2% 50.6% Kenya s GNI in 2013 was US$55 billion. The GNI per person in 2013 was US$1 180, which places Kenya in the lower-middle income group (World Bank, 2016d). In Kenya in 2013 the GNI per employed person was US$ Table 13 shows the GNI per person in Kenya from 2009 to 2013, as well as the resulting income category in which the GNI per person placed Kenya. Table 13 Kenyan GNI per person 2009 to 2013 (World Bank, 2016d) Year GNI per person Income category 2009 US$930 L 2010 US$1 000 L 2011 US$1 040 LM 2012 US$1 090 LM 2013 US$1 180 LM Kenya s GNI per person has equated to Kenya being classed as a low income country for 2009 and 2010 and a lower-middle income country thereafter. The GNI per person increased from 2009 to Table 14 shows the distribution of income between the population in 2005, from the World Bank (2016c), where the population has been divided into wealth quintiles. The 20% of the population with the highest income receive 54.13% of Kenya s wealth. The GNI per person for each of the quintiles, as at 2013, is also shown, assuming that the income distribution remains constant from 2005 to The income categorisation for each quintile is also shown None of Kenya s population quintiles fall in the upper-middle or higher income earners bands. Q4 and Q5 fall in the lower-middle income category, while Q1, Q2 and Q3 are in the low income category. The majority of the population has a constrained

24 354 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES income, which needs to meet many expenses, hence they are unlikely to consider purchasing insurance products, as there is no income available for these purchases The Gini index for Kenya in 2005 was 48.5 (World Bank, 2016c). Table 14 Distribution of wealth between the Kenyan population Population Percentage of income share GNI per person Income category Q1 4.68% US$293 L Q2 8.65% US$542 L Q % US$804 L Q % US$1 236 LM Q % US$3 393 LM 5.3 Regulatory Environment Life and non-life insurers in Kenya are regulated by the Insurance Regulatory Authority. The Insurance Regulatory Authority followed the then Office of the Commissioner of Insurance which came into existence with the enactment of the Insurance Act, CAP 487 in Prior to this, insurance regulation in Kenya was based on the United Kingdom s legislation under the Companies Act 1960 (Insurance Regulatory Authority, 2016a). Pension funds are regulated by the Retirement Benefits Authority (Retirement Benefits Authority, 2016) Third-party motor insurance is compulsory in Kenya (Insurance Regulatory Authority, 2016b; Blossom Insurance Brokers, 2016; Alexander Forbes, 2016b). Workmen s compensation is compulsory (Alexander Forbes, 2016b; Work Injury Benefits Act, 2007), unless the company maintains a security which is held to compensate employees in line with the Work Injury Benefits Act (Work Injury Benefits Act, 2007). Also compulsory is professional indemnity insurance for brokers, aviation third-party liability insurance for bodily injury, and oil pollution by vessels (Alexander Forbes, 2016b). 5.4 Insurance Industry Kenya has a National Hospital Insurance Fund which is compulsory for all employees in formal employment. Premiums vary, based on the income of the individual. Insurers offer supplementary cover such as outpatient benefits as well as greater inpatient benefits (National Hospital Insurance Fund, 2016) The National Social Security Fund provides basic financial security benefits to members on retirement. Membership is mandatory for all employees in the formal and informal sectors (National Social Security Fund, 2016a; 2016b).

25 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Some life insurance products include savings components In 2013 there were 12 life insurers and 24 non-life insurers registered with the Insurance Regulatory Authority. There were also 12 registered composite insurers (Insurance Regulatory Authority, 2014). 5.5 Culture Savings as a percentage of GDP in Kenya in 2013 was 9.76% which is much lower than the world average of 22.53% (World Bank, 2016d) Kenya s corruption perception index was 27 in 2013 (Transparency International, 2016) which is significantly worse than the average of 43 for all countries in the world The Kenyan literacy rate was 72% in The literacy rate was 78% and 67% for males and females respectively (World Bank, 2016a). The Kenyan literacy rate is similar to that of Ghana, with the gap between male and female literacy rates of 11% slightly lower than the gap of 13% in Ghana. 5.6 Demographics Figure 7 shows the population pyramid for Kenya in 2013, based on data from the World Bank (2016a). The population in Kenya in 2013 was 44 million people (World Bank, 2016a). The triangular shape, with a wide base and low top indicate an increasing population. There is a high proportion of child dependants and a low life expectancy. The dependency ratio was 0.82 in Figure 7 Population pyramid for Kenya in 2013

26 356 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES The unemployment rate was 9.1% in 2013 (World Bank, 2016d). This is slightly higher than the global unemployment rate of 6.0%. In 1999, the most recent figure available, 63.40% of those employed were in vulnerable employment (World Bank, 2016d) The employment-to-population ratio was 61.1% and there were 25.2 million people aged 15 years and older in 2013 (World Bank, 2016d). This implies that there were 15.4 million employed people, which equates an employed dependency ratio of In 2005 there were 20.8 million people in Kenya who were living on less than $3.10 a day, based on 2011 PPP. Of these, 11.9 million were living on less than $1.90 per day, based on 2011 PPP (World Bank, 2016c) Figure 8 shows the actual, wanted and ideal fertility rate by wealth quintile for Kenya as at There is a large difference between the actual fertility rate and the wanted and ideal fertility rates in Kenya, particularly among the lower earning quintiles. On average women in Q1 have 4.1 more children than women in Q5. For families in Q1 and Q5 with two parents, where all parents earn the same amount, the difference in the number of children implies that families in Q5 receive 83% more income per person than those in Q1. Figure 8 Kenyan fertility rates

27 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Penetration Rates Insurance penetration rates have been determined based on data from Insurance Regulatory Authority (2014) and the World Bank (2016d) The life insurance penetration rate in Kenya in 2013 was 0.94% and the non-life insurance penetration was 1.79%. Tables 15 and 16 show the split of the penetration rates by line of business for life and non-life insurance respectively. This is ordered in descending penetration level. Table 15 Kenyan life penetration rates by line of business Line of business Penetration rate Ordinary life 0.37% Pensions 0.34% Group life 0.23% Total 0.94% Table 16 Kenyan non-life penetration rates by line of business Line of business Penetration rate Medical 0.44% Motor commercial 0.41% Motor private 0.29% Fire industrial 0.16% Workmen s compensation 0.10% Theft 0.07% Personal accident 0.07% Engineering 0.06% Marine 0.06% Miscellaneous 0.05% Liability 0.03% Fire domestic 0.03% Aviation 0.02% Total 1.79% Recall that third-party motor insurance and workmen s compensation insurance are compulsory in Kenya, which explains the higher penetration of these non-life classes. Medical insurance is supplementary cover to the National Hospital Insurance Fund and is classified as non-life business.

28 358 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES 6. Nigeria 6.1 An overview of the Nigerian market as at 2013 followed by commentary on the penetration levels is provided in this section. 6.2 Economy The GDP of Nigeria in 2013 was US$515 billion. The GDP per person in 2013 was US$ Real GDP growth in 2013 was 5.39%, while growth in the GDP per person was 2.60%, both higher than the world GDP growth (World Bank, 2016d) Table 17 shows the GDP per sector from 2010 to 2013 for Nigeria. Table 17 Nigerian GDP by sector (World Bank, 2016d) Sector Agriculture 23.9% 22.3% 22.1% 21.0% Industry 25.3% 28.3% 27.3% 26.0% of which manufacturing 6.6% 7.2% 7.8% 9.0% Services 50.8% 49.4% 50.6% 53.0% Services contribute the most to the GDP. The portion of GDP from agriculture and industry both decreased from 2012 to The proportion of the GDP from manufacturing has increased year on year from 2010 to Nigeria s GNI in 2013 was US$489 billion. The GNI per person in 2013 was US$2 680, which places Nigeria in the lower-middle income group (World Bank, 2016d). In Nigeria in 2013 the GNI per employed person was US$ Table 18 shows the GNI per person and resulting income category for Nigeria from 2009 to Nigeria s GNI per person has equated to Nigeria being classed as a lower-middle income country from 2009 to The GNI per person increased from 2009 to Table 18 Nigerian GNI per person 2009 to 2013 (World Bank, 2016d) Year GNI per person Income category 2009 US$1 160 LM 2010 US$1 460 LM 2011 US$1 720 LM 2012 US$2 470 LM 2013 US$2 680 LM

29 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Table 19 shows the distribution of income between the population in 2009 (World Bank, 2016c), where the population has been divided into wealth quintiles. The 20% of the population with the highest income receive 48.98% of Nigeria s wealth. The GNI per person for each of the quintiles, as at 2013, is also shown, assuming that the income distribution remains constant. The income categorisation for each quintile is also shown. Table 19 Distribution of wealth between the Nigerian population Population Percentage of income share GNI per person Income category Q1 5.37% US$760 L Q2 9.71% US$1 374 LM Q % US$2 033 LM Q % US$3 055 LM Q % US$6 934 UM Of Nigeria s population 20% are categorised as upper-middle income and would likely consider purchasing insurance products. The other 80% fall in the lower-middle and low income categories. This portion of the population has a constrained income, which needs to meet many expenses, hence they are unlikely to consider purchasing insurance products, as there is no income available for these purchases The Gini index for Nigeria in 2009 was 43.0 (World Bank, 2016c). 6.3 Regulatory Environment The National Insurance Commission (NAICOM) regulates life and non-life insurers in Nigeria. The National Insurance Commission derives its powers from the National Insurance Act 1997 and the Insurance Act 2003 (National Insurance Commission, 2016a). Pension products are separately regulated by the National Pension Commission (PenCom, 2016) Builders liability, occupiers liability, employers liability, motor third-party insurance and healthcare professional indemnity insurance are compulsory in Nigeria (Insurance Consumers Association of Nigeria, 2016; Nigerian Finder, 2016). There are two forms of employers liability. The first is a form of group life cover and is mandated by the Pension Reform Act The second is a form of workmen s compensation and is mandated by the Workmen s Compensation Act (Insurance Consumers Association of Nigeria, 2016).

30 360 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Professional indemnity for brokers, aviation and construction insurance, and marine cargo imports insurance are also compulsory in Nigeria (Alexander Forbes, 2016c). 6.4 Insurance Industry A number of life insurers include a savings component in their products. Life products are used for savings purposes as well as insurance purposes. In 2013 there were 23 and 41 life and non-life insurers (National Insurance Commission, 2014) A National Health Insurance Scheme exists (National Health Insurance Scheme, 2016b) which covers those employed through compulsory premiums of 15% of employees basic salaries, of which the employer pays 10% and the employee pays 5%. Cover for the poor, elderly, veterans and disabled is free while participants in the informal sector are expected to make monthly contributions (Joint Learning Network, 2016). 6.5 Culture Savings as a percentage of GDP in Nigeria in 2013 was 19.83% (World Bank, 2016d). This is near the savings as a percentage of GDP for the world of 22.53% Nigeria s corruption perception index was 25 in 2013 (Transparency International, 2016), significantly worse than the average of 43 for all countries in the world The Nigerian literacy rate was 51% in The literacy rate was 61% and 41% for males and females respectively (World Bank, 2016a). Nigeria had the lowest literacy rates of the countries considered, as well as the largest gap, along with Zambia, of 20% between male and female literacy Table 20 shows the split of religions which the population follows, in descending order. In Nigeria the majority of the population are Muslim. Islam prohibits the purchase of conventional insurance products (About Religion, 2016). However, Muslims may use takaful, commonly referred to as Islamic insurance, which operates on a mutual basis, rather than having shareholders (Investopedia, 2016a). Table 20 Split of the religions followed in Nigeria (The World Factbook, 2016) Religion Percentage of population Muslim 50.0% Christian 40.0% Indigenous beliefs 10.0%

31 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Demographics Figure 9 shows the population pyramid for Nigeria in 2013, based on data from the World Bank (2016a). The population in Nigeria in 2013 was 173 million people (World Bank, 2016a) The pyramid shows a population which rapidly decreases with age. There is a high ratio of child dependants and a high mortality. The Nigerian dependency ratio was 0.88 in The unemployment rate was 7.5% in 2013 (World Bank, 2016d). No figures are available for the proportion of the population in vulnerable employment The employment-to-population ratio was 51.8% and there were 96.6 million people aged 15 years and older in 2013 (World Bank, 2016d). This implies that there were 50.0 million employed people, which equates an employed dependency ratio of There were million people in 2009 in Nigeria who were living on less than $3.10 a day, based on 2011 PPP. Of these, 83.0 million were living on less than $1.90 per day, based on 2011 PPP (World Bank, 2016c) Figure 10 shows the actual, wanted and ideal fertility rate by wealth quintile for Nigeria as at Figure 9 Population pyramid for Nigeria in 2013

32 362 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Figure 10 Nigerian fertility rates Nigeria is the only country in this study where the ideal fertility rate is greater than the actual fertility rate for all wealth quintiles. Furthermore, the ideal number of children per quintile are higher than the ideal number of children for the respective quintiles for each of the other countries considered The difference between the actual and wanted fertility rates in Nigeria is small, indicating that the high actual fertility rate is not a consequence of lack of access to family planning measures The high fertility rates will result in rapid growth of the population, as well as create challenges for the income level per person, as the income earned by a working person will be split among a large number of dependants. 6.7 Penetration Rates Insurance penetration rates have been determined based on data from National Insurance Commission (2014) and the World Bank (2016d) The life insurance penetration rate in Nigeria in 2013 was 0.10% and the nonlife insurance penetration rate was 0.22%. These are the lowest penetration rates for the five countries considered in this paper, despite having the second highest GNI per person after South Africa. This may be because of the large Muslim population, which may not be adequately catered for with the current insurance products. In 2013 NAICOM issued Operational Guidelines 2013 Takaful Insurance Operation. Furthermore, of the study, Nigeria has the worst corruption index score, with an index score of 25, which may indicate a distrust of insurers, resulting in an unwillingness to enter insurance contracts.

33 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES In 2014, 52% of the life business in Nigeria was in respect of group life business, 33% was in respect of individual life business and 15% was in respect of group pensions (National Insurance Commission, 2015) Table 21 shows the split of the penetration rates by line of business for non-life insurance. This is ordered in descending penetration level. Table 21 Nigerian non-life penetration rates by line of business Line of business Penetration rate Oil and gas 0.07% Motor 0.06% General accident 0.03% Fire 0.03% Marine 0.02% Miscellaneous 0.01% Total 0.22% The majority of Nigeria s economy is built on the export of oil and gas, which is why the largest penetration is in respect of oil and gas insurance. Although motor third-party insurance is compulsory in Nigeria, the penetration is low. 7. Zambia 7.1 An overview of the Zambian market as at 2013 followed by commentary on the penetration levels is provided in this section. 7.2 Economy The GDP of Zambia in 2013 was US$28 billion. The GDP per person in 2013 was US$ Real GDP growth in 2013 was 5.13%, while growth in GDP per person was 1.96%, both greater than the world averages (World Bank, 2016d) Table 22 shows the GDP per sector from 2010 to 2013 for Zambia. Table 22 Zambian GDP by sector (World Bank, 2016d) Sector Agriculture 10.5% 10.6% 10.1% 9.1% Industry 35.5% 37.5% 34.8% 35.7% of which manufacturing 8.4% 8.2% 7.7% 6.6% Services 54.0% 51.9% 55.1% 55.2%

34 364 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES The proportion of GDP from services increased from 54.0% in 2010 to 55.2% in The contribution to GDP from agriculture decreased from 10.5% to 9.1% from 2010 to The proportion of GDP from industry increased slightly from 2010 to 2013, while the portion from manufacturing decreased over the same period Zambia s GNI in 2013 was US$27 billion. The GNI per person in 2013 was US$1 730, which places Zambia in the lower-middle income group (World Bank, 2016d). In Zambia in 2013 the GNI per employed person was US$ Table 23 shows the Zambian GNI per person from 2009 to 2013, as well as the income category which this resulted in Zambia being classified as. Table 23 Zambian GNI per person 2009 to 2013 (World Bank, 2016d) Year GNI per person Income category 2009 US$1 260 LM 2010 US$1 310 LM 2011 US$1 400 LM 2012 US$1 650 LM 2013 US$1 730 LM Zambia s GNI per person has equated to Zambia being classed as a lower-middle income country from 2009 to The GNI per person increased from 2009 to Table 24 shows the distribution of income between the population in 2010, from the World Bank (2016c), where the population has been divided into wealth quintiles. The 20% of the population with the highest income receive 61.05% of Zambia s wealth. The GNI per person for each of the quintiles, as at 2013, is also shown, assuming that the income distribution remains constant. The income categorisation for each quintile is also shown. Table 24 Distribution of wealth between the Zambian population Population Percentage of income share 2010 GNI per person 2013 Income category 2013 Q1 3.81% US$341 L Q2 6.76% US$605 L Q % US$939 L Q % US$1 601 LM Q % US$5 464 UM

35 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Of Zambia s population, 20% are classified as upper-middle earners and would likely consider purchasing insurance products. The other 80% fall in the lower-middle and low income categories. This portion of the population has a constrained income, which needs to meet many expenses, hence they are unlikely to consider purchasing insurance products, as there is no income available for these purchases The Gini index for Zambia in 2010 was 55.6 (World Bank, 2016c). 7.3 Regulatory Environment Life and non-life insurers are regulated by the Pensions and Insurance Authority in Zambia, which also regulates pension funds and micro insurance. The Pensions and Insurance Authority derives its mandate from the Pension Scheme Regulation Act No 28 of 1996 and the Insurance Act No 27 of 1997 (Pensions and Insurance Authority, 2016) In Zambia, motor third-party liability insurance, professional indemnity insurance for insurance brokers, and aviation third-party liability insurance are compulsory. Workmen s compensation is also compulsory. However, this is purchased through the Workers Compensation Board (Alexander Forbes, 2016d). 7.4 Insurance Industry The Workers Compensation Fund provides compensation to employees on death, total disablement, partial disablement and temporary disablement. Employers pay contributions to the Workers Compensation Fund, based on the earnings of their employees, assessed risks of the work place, cost of compensation payable and requirements of the Workers Compensation Fund (International Labour Organisation, 2016) Life insurance products are used as savings vehicles in Zambia, in addition to providing protection against risk In 2013, there were 8 and 15 life and non-life insurers registered with the Pensions and Insurance Authority respectively (Pensions and Insurance Authority, 2013) Zambia has Universal Health Insurance which covers formal sector employees. All beneficiaries pay a flat fee which is subsidised by the Zambian government (Health Market Innovations, 2016). 7.5 Culture Savings as a percentage of GDP in Zambia in 2010 was 31.43% (World Bank, 2016d). Figures for 2013 are not available. This is greater than the world s savings as a percentage of GDP of 22.53%.

36 366 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Zambia s corruption perception index was 38 in 2013 (Transparency International, 2016), worse than the average of 43 for all countries in the world The Zambian literacy rate was 61% in The literacy rate was 72% and 52% for males and females respectively (World Bank, 2016a). The Zambian literacy rates are significantly lower than those in Ghana and Kenya, particularly for females. The gap between male and female literacy rates are 20%, which is the same as the gap seen in Nigeria. 7.6 Demographics Figure 11 shows the population pyramid for Zambia in 2013, based on data from the World Bank (2016a). The population in Zambia in 2013 was 15 million people (World Bank, 2016a) The wide base and narrow top indicate a growing population. There are a high number of child dependants, as well as a low life expectancy. This pyramid indicates a rapidly growing population. The dependency ratio was 0.97 in The unemployment rate was 13.1% in 2013 (World Bank, 2016d). Of those employed, 79.0% were in vulnerable employment in 2012 (World Bank, 2016d) The employment-to-population ratio was 68.9% and there were 8.2 million people aged 15 years and older in 2013 (World Bank, 2016d). This implies that there were 5.6 million employed people, which equates to an employed dependency ratio of Figure 11 Population pyramid for Zambia in 2013

37 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES There were 11.0 million people in 2010 in Zambia who were living on less than $3.10 a day, based on 2011 PPP. Of these, 9.0 million were living on less than $1.90 per day, based on 2011 PPP (World Bank, 2016c) Figure 12 shows the actual and wanted fertility rates for Zambia as at 2014 as well as the ideal fertility rate as at 2007 by wealth quintile. The ideal fertility rates are fairly similar across the quintiles, while the actual and wanted fertility rates show a greater spread across the quintiles. On average, women in Q1 have 4.1 more children than women in Q5, resulting in fewer financial resources per person in Q1 than in Q5, even if each working person in Q1 earned the same as a working person in Q Penetration Rates Insurance penetration rates have been determined based on data from the Pensions and Insurance Authority (2013) and the World Bank (2016d) The life insurance penetration rate in Zambia in 2013 was 0.77% and the non-life insurance penetration was 0.74%. The life insurance penetration consists of long-term insurance (0.31%) and pensions (0.46%). These penetration rates are at a similar level to those in Ghana, but much lower than the world average Individual life business is the predominant line of life business written in Zambia. Life business written in Zambia includes term assurance which includes funeral cover, endowments in the form of education policies and with-profits policies, credit life, whole of life, annuities, group life, medical insurance, and unit linked with profit funds. Figure 12 Zambian fertility rates

38 368 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES The penetration rates for non-life business are not publicly available. Our experience in the Zambian market indicates that motor and property insurance make up approximately half of the non-life insurance premiums with other lines of business making up the other half. There is no medical insurance penetration in Zambia. 8. Observations 8.1 Ghana, Kenya, Nigeria and Zambia are developing economies and by global standards are characterised by low income levels and widespread poverty. Insurance penetration in these countries reflects a low level of affordability. The majority of the population of these countries (at least 80%) have a constrained income and are unlikely to consider purchasing insurance products. 8.2 To some extent the insurance penetration levels appear to be correlated with literacy rates. For example, South Africa has the highest literacy rates of the countries considered, and is the only country considered where the literacy rate exceeds the world s literacy rate. South Africa also has the highest level of insurance penetration. Conversely, Nigeria has the lowest literacy rates and also has the lowest level of insurance penetration, despite having the highest GNI per person after South Africa. However, Zambia is a bit of an outlier, where the insurance penetration in Zambia is between that of Kenya and Ghana but the literacy rate is significantly below the literacy rates in Kenya and Ghana. 8.3 Kenya is somewhat of an anomaly amongst the developing countries considered in this paper as it has relatively higher levels of insurance penetration, especially for non-life, despite having the lowest GNI per person of the countries considered. In 2013, higher penetration levels of 1.79% for non-life insurance in Kenya are contributed by medical insurance at 0.43% and motor insurance at 0.70% (third-party motor insurance being compulsory). 8.4 South Africa is a country in the process of transitioning from a developing to developed economy. A higher proportion are able to afford insurance. The highest earning quintile of the population are classified at a high income level and the second highest quintile are an upper-middle income category. These two highest earning quintiles of the population, where affordability of insurance is indicated, share 85% of the income of the population. 8.5 All of the countries considered in this paper have relatively high dependency ratios compared to world averages. The following sections of this paper explore the impact of population projections on dependency ratios and income levels. As the countries transition from developing to more developed economies, some reduction in dependency ratios is anticipated. This may result in higher per person income and increase the affordability of insurance.

39 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Projections introduction 9.1 Introduction For each of the countries key demographic indicators were projected, based on the World Bank s population projections, as well as the GNI, in order to assess the potential for insurance penetration growth in these markets in the future. The projections also provided information as to some of the challenges which must be overcome in order to increase insurance penetration in these markets. 9.2 Demographics A projection of the population pyramids and demographic indicators are provided for each country. The demographic indicators include the population size, the size of the working population, the dependency ratio and employed-dependency ratio. The population size projections are as provided by the World Bank. The size of the working population, dependency ratio and employed-dependency ratio have been projected for this paper, based on the World Bank s population projections Projections of the proportion split of the working population, child dependants and elderly are shown. The working population are all people between ages 15 and 65. Child dependants are people younger than 15 and the elderly are those aged 65 years and older We start with the employment-to-population ratio for ages 15 years and upwards as at 2013 for each country. Recall from Section 2 that this is the proportion of the country s population aged 15 years and older which is employed. Based on the World Bank s population projections for the population aged 15 years and upwards, we determine the number of jobs which must be created each year in order for the employment-to-population ratio to remain constant. This assumes that the jobs which are created are sustainable jobs and not temporary employment or are the net number of jobs created. 9.3 Gross national income Schmidt (2016) states the following regarding GNI: Gross National Income (GNI) is GDP plus income paid into the country by other countries for such things as interest and dividends (less similar payments paid out to other countries) Given that GNI and GDP are correlated, as well as the lack of data available on GNI projections, we have projected the GNI based on the GDP increases indicated by the International Monetary Fund s World Economic Outlook Database for each country. GDP increases are provided to year 2021, after which we assume that the increase in 2021 will apply going forward.

40 370 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES We recognise that the projections from 2022 onwards may be too high given that GDP growth is expected to slow down as the countries transition from developing to developed countries. Thus, the level of GNI shown in the projections may be considered to be optimistic The projected GNI and populations are combined to project the real GNI per person. The GNI for each of the quintiles is shown, assuming that the income distribution remains constant going forward. 10. South African Projections 10.1 Demographics The South African population pyramid, as at 1993, 2003 and 2013, as well as the projected pyramids for 2023, 2033, and 2043 are shown in Figure 13, as provided by the World Bank Over 50 years the shape of South Africa s population pyramid changes significantly. The pyramid in 1993 is one of a developing country which has an expanding population. The South African population grows each year from 1993 to 2043; however, the rate of growth gradually decreases. By 2043 the pyramid is projected to have more of a bell shape. The birth rate is declining which will result in a gradual decrease in the size of the population. However, the increased longevity, indicated by the wider top of the pyramid, will mean the decline in birth rates does not reduce the population s size for a while Table 25 shows key past South African demographic indicators as at 1993, 2003 and 2013 as well as the projected demographic indicators as at 2023, 2033 and The World Bank provides the projections of the population in five-year age bands. The other demographics in the table above are derived from the World Bank projections Projections for the employed population assume the same employment-topopulation ratio as observed in 2013 for the population aged 15 years and older From 2013 to 2043 the pyramids show growth of the South African population, predominantly in the ages of 30 onwards. The working population as a proportion of the total population shows a significant increase in size, as does the elderly population, while the proportion of children decreases. The growth in the working population of 41% from 2013 to 2043, compared to population growth of 34% over the same period, results in a decrease in the dependency ratio. With everything else equal, a lower dependency ratio should result in greater disposable income.

41 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES 371 Table 25 South African key demographics Demographic factor Population 37.5 million 46.1 million 53.2 million 60.1 million 66.2 million 71.1 million Working population 22.3 million 28.6 million 34.6 million 40.1 million 44.9 million 48.9 million Employment-to-population ratio 41% 40% 39% 39% 39% 39% Employed population 9.6 million 12.3 million 14.6 million 17.2 million 19.7 million 21.7 million Dependency ratio Employed dependency ratio Percentage older than 65 years 3% 4% 5% 6% 8% 9% Percentage between 15 and 65 years 60% 62% 65% 67% 67% 68% Percentage younger than 15 years 37% 34% 30% 27% 25% 23% Figure 13 South African population pyramids

42 372 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES However, the growing working population is expected to place pressure on employment rates, as the number of people seeking a job increases. Figure 14 shows the absolute and percentage increase in the number of jobs in South Africa from 1992 to Thereafter, the absolute and percentage increase in the number of jobs, which is required to maintain the employment-to-population ratio of 39%, is shown from 2014 to The actual figures are to the left of the solid black bar, while the projected figures are to the right of the solid black bar In the past, the net number of jobs created each year has been volatile. A steady but low level of jobs will need to be created each year in South Africa in order to ensure that the employment-to-population ratio remains at the 2013 level of 39.3%. Any improvement in the unemployment ratio implies a higher level of job creation. A downward trend in projected absolute values in Figure 14 indicates that this level of job creation may be achievable Gross national income Table 26 shows the real increases applied to the GNI from 2014 to 2043 for South Africa, which are based on the GDP growth projections shown in the International Monetary Fund s World Economic Outlook Database. The growth estimates for South Africa start after 2014, thus the growth shown for 2014 is actual growth Table 27 shows GNI projections, based on the increases shown in Table 26. The GNI per person, based on the population increases above, are also provided, as well as the GNI per person for the wealth quintiles in South Africa, assuming that the distribution of wealth in South Africa remains constant. Figure 14 Employment growth in South Africa

43 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES 373 Table 26 South African real GDP growth projections Year Increase % % % % % % % 2021 onwards 2.40% Table 27 GNI projections for South Africa GNI US$356.4 billion US$429.0 billion US$543.8 billion US$689.3 billion GNI per person US$7 400 US$7 135 US$8 208 US$9 694 GNI per employed person US$ US$ US$ US$ GNI per person in Q5 US$ US$ US$ US$ GNI per person in Q4 US$5 331 US$5 673 US$6 526 US$7 707 GNI per person in Q3 US$2 672 US$2 843 US$3 271 US$3 863 GNI per person in Q2 US$1 579 US$1 680 US$1 933 US$2 283 GNI per person in Q1 US$828 US$881 US$1 014 US$ The GNI per person in 2013 is as per the World Bank (2016d). However, if the GNI per person is calculated as the GNI divided by the population, the resulting GNI per person is US$ This calculation is used to determine the GNI per person in each quintile throughout the table, as well as the GNI per person for years 2023, 2033 and Figure 15 shows the GNI per person in each of the quintiles. Reading from the lowest level of the graph to the highest, the various tints indicate the lower income, lower-middle income, upper-middle and higher income categories respectively Currently Q1 has a GNI per person which places them in the low income category. Q2 and Q3 are in the lower-middle category. Q4 and Q5 are in the uppermiddle and high income categories respectively. Based on the projections these income groups will remain the same until 2036 when Q1 will move to the lower-middle

44 374 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Figure 15 GNI per person in South Africa income category. In 2047 Q3 shifts from the lower-middle to upper-middle income group. From % of the population will be in the upper-middle income category or higher, and are expected to be able to afford insurance products Penetration Rates Of the countries considered, South Africa had the highest insurance penetration rate as well as the highest GNI per person. South Africa was also the only country to be categorised as an upper-middle income country in South Africa had the highest unemployment rate of the countries considered in However, of those in employment, South Africa had the lowest vulnerable employment rate The population is projected to grow by 34% from 2013 to 2043 and the working population is projected to grow by 41% over the same period. The employed dependency ratio will decrease slightly from 2.63 in 2013 to 2.28 in Real growth in GNI is projected to remain below 2.50% going forward. We expect that the population in Q5 have already purchased a significant level of insurance, while some of those in Q4 will have purchased insurance. Growth in the penetration rates is projected to be from Q4, as the income levels increase. Only in 2047 does Q3 shift from the lower-middle income category to the upper-middle category A decrease in dependency ratio contributes to improvements in disposable income and anticipated insurance usage.

45 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Ghanaian Projections 11.1 Demographics The Ghanaian population pyramid, as at 1993, 2003 and 2013, as well as the projected pyramids for 2023, 2033, and 2043 are shown in Figure 16, as provided by the World Bank The change in Ghana s population pyramid s shape from 1993 to 2043 is not as drastic as that of South Africa. In 1993 the pyramid had a wide base, concave sides and a narrow apex. This indicates high birth rates, coupled with high mortality rates. In 2013 the shape is similar to that of 1993 although the population is larger and the slopes are not as concave. By 2023 the steeper slope across the lowest two age bands indicates that birth rates have decreased slightly. The projected 2043 pyramid continues to indicate population growth, although the birth rate is decreasing and Figure 16 Ghanian population pyramids

46 376 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES the sides are getting steeper, indicating that the country is moving towards a late expanding population Table 28 shows key past Ghanaian demographic indicators as at 1993, 2003 and 2013 as well as the projected demographic indicators as at 2023, 2033 and Table 28 Ghanaian key demographics Demographic factor Population 15.9 million 20.3 million 26.2 million 32.4 million 38.8 million 45.5 million Working population 8.6 million 11.4 million 15.1 million 19.2 million 24.2 million 28.9 million Employment-to-population ratio 65% 66% 68% 68% 68% 68% Employed population 5.9 million 7.9 million 10.9 million 13.8 million 17.6 million 21.4 million Dependency ratio Employed dependency ratio Percentage older than 65 years 3% 3% 3% 4% 4% 5% Percentage between 15 and 65 years 54% 56% 58% 59% 63% 64% Percentage younger than 15 years 43% 41% 39% 37% 33% 31% The World Bank provides the projections of the population in five-year age bands. The other demographics in the table above are derived from the World Bank projections Projections for the employed population assume the same employment-topopulation ratio as observed in The Ghanaian population is projected to increase by 74% from 2013 to 2043, while the working population will increase by 92%. This increase will be driven by an increase in the working age population, with the elderly population also increasing marginally. The proportion of the population below 15 years of age is projected to decrease from 39% in 2013 to 31% in The strong growth in the working population results in a decrease in the dependency ratio from 0.74 in 2013 to 0.57 in The dependency ratio has decreased from 1993 to However, the ability of the working population to support the elderly and children relies on the ability to earn an income, which depends on job creation. Figure 17 shows the absolute and percentage increase in the number of jobs in Ghana from 1992 to Thereafter the absolute and percentage increase in the number of jobs, which is required to maintain the employment-to-population

47 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES 377 Figure 17 Employment growth in Ghana ratio of 68%, is shown from 2013 to The actual figures are to the left of the solid black bar, while the projected figures are to the right of the solid black bar The growth in employment has fluctuated from 1992 to The percentage increase in jobs required appears reasonable. However, the absolute increase in jobs required is significantly higher than the average increase in net jobs from 1992 to Ghana is thus at risk of the unemployment rate rising as the number of new jobs created may fail to keep up with the increase in the working population Gross National Income Table 29 shows the real increases applied to the GNI from 2014 to 2043 for Ghana, which are based on the GDP growth projections shown in the International Monetary Fund s World Economic Outlook Database. The growth estimates for Ghana start after 2014, thus the growth shown for 2014 is actual growth. Table 29 Ghanaian real GDP growth projections Year Increase % % % % % % % 2021 onwards 4.56%

48 378 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Table 30 shows GNI projections, based on the increases shown in Table 29 above. The GNI per person, based on the population increases above, are also provided, as well as the GNI per person for the quintiles in Ghana, assuming that the distribution of wealth in Ghana remains constant. Table 30 GNI projections for Ghana GNI US$46.5 billion US$76.2 billion US$119.1 billion US$186.0 billion GNI per person US$1 740 US$2 352 US$3 068 US$4 091 GNI per employed person US$4 271 US$5 506 US$6 769 US$8 707 GNI per person in Q5 US$4 313 US$5 713 US$7 452 US$9 936 GNI per person in Q4 US$1 922 US$2 546 US$3 321 US$4 428 GNI per person in Q3 US$1 299 US$1 720 US$2 244 US$2 992 GNI per person in Q2 US$878 US$1 163 US$1 517 US$2 023 GNI per person in Q1 US$465 US$616 US$804 US$ The GNI per person in 2013 is as per the World Bank (2016d). However, if the GNI per person is calculated as the GNI divided by the population, the resulting GNI per person is US$ Figure 18 shows the GNI per person in each of the quintiles. Currently Q1 and Q2 have a GNI per person which places them in the low income category. Q3 and Q4 are in the lower-middle category. Q5 is in the upper-middle income category. Based on the projections, these income groups will remain the same until 2019 when Q2 will move from the low category to the lower-middle category. In 2041 Q4 is projected to Figure 18 GNI per person in Ghana

49 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES 379 shift from the lower-middle category to the upper-middle category and in 2043 Q1 is projected to shift from the lower to the lower-middle category. From % of the population will be in the upper-middle income category, and are expected to be able to afford insurance products Penetration Rates The dependency ratio in Ghana has decreased from 1993 to 2013, and is further projected to decrease to 2043, because of the increase in the proportion of the population aged 15 years and older The projected decrease in the employed dependency ratio from 1.41 in 2013 to 1.13 in 2043 is based on the assumption that the new jobs required, as depicted in Figure 17, can be met. Given the number of jobs created in the past, it appears challenging to meet this job creation target in the future, which may result in an increase of the employed dependency ratio Projected growth in Ghana s GNI is relatively high, with the lowest level of GNI growth projected to be 3.49% in This is characteristic of developing countries, which tend to have higher GDP growth than developed countries. The projected GNI growth outstrips the projected population growth, resulting in projected increases in the GNI per person The distribution of income in Ghana is the most equitable of the countries considered in this paper, as indicated by the Gini index as well as the distribution of income between the population. The GNI per person per year ranges from US$465 for Q1 to US$4 313 for Q5 in These are expected to increase going forward. However, for a long time the four lower quintiles will remain in the lower-middle income category. There is a spread of income in each quintile, with some in the same quintile receiving a higher income than others. In Q5 we expect the higher earning population to have some demand for insurance while those on the lower end will have little to no demand for insurance products The strong projected GNI growth is likely to create an opportunity for insurance penetration to increase in Ghana. However, this is constrained by the high dependency ratio. Furthermore, it does not appear that there is a large portion of the population which has a demand for insurance products, given their income levels. The portion of the market which has a demand for insurance products is expected to increase slowly over the next thirty years, because of the GNI increases, and we expect insurance penetration to increase in line with the increase in the market for insurance Initially, we expect the majority of the insurance penetration increases to be from compulsory insurance classes as businesses expand and more individuals can

50 380 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES afford to purchase private vehicles. Thereafter, we may expect to see an increase in more luxury insurance products such as various life products. 12. Kenyan Projections 12.1 Demographics The Kenya population pyramid, as at 1993, 2003 and 2013, as well as the projected pyramids for 2023, 2033, and 2043 are shown in Figure 19, as provided by the World Bank The 1993 population pyramid has a wide base with concave slopes, indicating that the birth and mortality rates were high. The narrow top of the pyramid further shows the high mortality rates which were experienced in Kenya. The slope in 2013 is still concave, although not to the same extent, and with a slight bulge in the middle. By 2043 the projected population pyramid shows high birth rates, improved mortality and a growing population. Figure 19 Kenyan population pyramids

51 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Table 31 shows key past Kenyan demographic indicators as at 1993, 2003 and 2013 as well as the projected demographic indicators as at 2023, 2033 and Table 31 Kenyan key demographics Demographic factor Population 25.8 million 33.6 million 43.7 million 56.0 million 69.7 million 84.7 million Working population 12.8 million 18.1 million 24.0 million 32.2 million 42.3 million 52.8 million Employed-to-population ratio 66% 60% 61% 61% 61% 61% Employed population 8.9 million 11.4 million 15.4 million 20.8 million 27.5 million 34.8 million Dependency ratio Employed dependency ratio Percentage older than 65 years 3% 3% 3% 3% 4% 5% Percentage between 15 and 65 years 49% 54% 55% 58% 61% 62% Percentage younger than 15 years 48% 43% 42% 39% 35% 33% The World Bank provides the projections of the population in five-year age bands. The other demographics in Table 31 are derived from the World Bank projections Projections for the employed population assume the same employment-topopulation ratio as observed in The Kenyan population is projected to increase by 94% from 2013 to Furthermore, the projected 2043 population pyramid is one which indicates that the population is expected to continue growing in the future. This growth occurs across all ages; however, the portion of the working population is projected to increase from 55% of the total population in 2013 to 62% in As with the other countries, the larger working population reduces the dependency ratio, which should have a positive impact on the household income, provided that jobs can be created to match the growing working population. Figure 20 shows the absolute and percentage increase in the number of jobs in Kenya from 1992 to Thereafter the absolute and percentage increase in the number of jobs, which is required to maintain the employment-to-population ratio of 61%, is shown from 2014 to The actual figures are to the left of the solid black bar, while the projected figures are to the right of the solid black bar. The growth in net employment in absolute terms is high, although if the net absolute trend from 2006 to 2012 continues, the net employment growth should exceed that required to maintain the employment-topopulation ratio of 2013.

52 382 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Figure 20 Employment growth in Kenya 12.2 Gross National Income Table 32 shows the real increases applied to the GNI from 2014 to 2043 for Kenya, which are based on the GDP growth shown in the International Monetary Fund s World Economic Outlook Database. The growth estimates for Kenya start after 2014, thus the growth shown for 2014 is actual growth. Table 32 Kenyan real GDP growth projections Year Increase % % % % % % % 2021 onwards 6.48% Table 33 shows GNI projections, based on the increases shown in Table 32. The GNI per person, based on the population increases above, are also provided, as well as the GNI per person for the population quintiles in Kenya, assuming that the distribution of wealth in Kenya remains constant The GNI per person in 2013 is as per the World Bank (2016d). However, if the GNI per person is calculated as the GNI divided by the population, the resulting GNI per person is US$ Figure 21 shows the GNI per person in each of the population quintiles.

53 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES 383 Table 33 GNI projections for Kenya GNI US$54.8 billion US$99.9 billion US$187.2 billion US$350.7 billion GNI per person US$1 180 US$1 784 US$2 686 US$4 142 GNI per employed person US$3 557 US$4 795 US$6 816 US$ GNI per person in Q5 US$3 393 US$4 828 US$7 269 US$ GNI per person in Q4 US$1 236 US$1 759 US$2 649 US$4 084 GNI per person in Q3 US$804 US$1 144 US$1 722 US$2 655 GNI per person in Q2 US$542 US$772 US$1 162 US$1 791 GNI per person in Q1 US$293 US$417 US$629 US$ Currently the lowest three quintiles are classified as low income, while Q4 and Q5 are in the lower-middle category. In 2019 Q5 is projected to move from the lower-middle category to the upper-middle category. Q3 is projected to move from the lower category to the lower-middle category in In 2031 Q2 is projected to move up to the lower-middle category. In 2044 Q4 is projected to move to the uppermiddle category and Q1 is projected to move to the lower-middle category in Q5 is projected to move to the high income category in Q4 and Q5 are the only quintiles which we expect to be able to afford insurance in 2050, both being in the upper-middle income categories or higher from 2044 onwards. 2.3 Penetration Rates The working population is increasing as a proportion of the total population, and hence the dependency ratio is projected to decrease from 0.82 in 2013 to 0.60 in Figure 21 GNI per person in Kenya

54 384 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES A decrease in the employed population dependency ratio from 1.84 in 2013 to 1.44 in 2043 is projected In order to maintain a constant employment-to-population ratio, a high level of steady job growth is required from 2014 to If this level of job growth can not be achieved, the employed dependency ratio may increase The projected GNI growth for Kenya is in excess of 5.00% each year. The growth in GNI is greater than the population growth, resulting in a projected increase in GNI per person. The GNI per person in 2013 was US$1 180, placing Kenya in the bottom end of the lower-middle income category. The three lowest quintiles are in the lower income category, while the other two quintiles are in the lower-middle categories. This indicates that the majority of people will not have income levels which result in a demand for insurance. Rather, the low income levels result in individuals trying to ensure that basic needs are met The projections of the GNI per person by quintile show that, from 2013 to 2050, the majority of the population will have low income levels which place them in the low and lower-middle income categories. Insurance products can likely only be sold to Q5, which only moves into the upper-middle category from Initially, we expect that only a small portion of this quintile will have any demand for insurance Given the high projected levels of GNI growth, we do predict that insurance penetration will increase in Kenya. However, we predict that the increase in penetration will be gradual, as more of the population becomes more affluent. 13. Nigerian Projections 13.1 Demographics The Nigerian population pyramids, as at 1993, 2003 and 2013, as well as the projected pyramids for 2023, 2033, and 2043 are shown in Figure 22, as provided by the World Bank The population shows significant growth from 1993 to 2013, although the shape of the pyramid stays similar. The concave sides indicate a high birth and mortality rate. The population is projected to continue to grow significantly until The population pyramid in 2043 has more steep sides, indicating an improving mortality rate Table 34 shows key past Nigerian demographic indicators as at 1993, 2003 and 2013 as well as the projected demographic indicators as at 2023, 2033 and The World Bank provides projections of the population in five-year age bands. The other demographics in the table are derived from the World Bank projections.

55 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES 385 Figure 22 Nigerian population pyramids Projections for the employed population assume the same employment-topopulation ratio as observed in The Nigerian population is projected to double from 2013 to 2043, and the population pyramid as at 2043 shows that the population will continue to increase thereafter. Although this increase will be across all ages, the predominant increase is in the working population, which is projected to increase from 53% of the total population in 2013 to 60% in Unlike the other countries seen so far, the portion of the population aged 65 and older remains constant from 2013 to 2043, which may indicate that the mortality rate is not improving over this time. Rather, the birth rate is decreasing slightly, resulting in a smaller portion of the population being younger than 15 years of age.

56 386 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Table 34 Nigerian key demographics Demographic factor Population million million million million million million Working population 54.2 million 71.1 million 91.8 million million million million Employed-topopulation ratio 53% 51% 52% 52% 52% 52% Employed population 30.1 million 38.1 million 50.0 million 66.2 million 87.7 million million Dependency ratio Employed dependency ratio Percentage older than 65 years 3% 3% 3% 3% 3% 3% Percentage between 15 and 65 years 52% 53% 53% 54% 57% 60% Percentage younger than 15 years 45% 44% 44% 43% 40% 37% Figure 23 shows the absolute and percentage increase in the number of jobs in Nigeria from 1992 to Thereafter, the absolute and percentage increase in the number of jobs, which is required to maintain the employment-to-population ratio of 52%, is shown from 2014 to The actual figures are to the left of the solid black bar, while the projected figures are to the right of the solid black bar The percentage growth rate required to maintain the employment-topopulation ratio is in line with the growth achieved from 2006 to However, the Figure 23 Employment growth in Nigeria

57 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES 387 absolute increase in jobs is much greater than the employment growth in the past ten years. This large increase in jobs may be difficult for Nigeria to create and sustain, which will result in decreased employment rates in Nigeria Gross National Income Table 35 shows the real increases applied to the GNI from 2014 to 2043 for Nigeria, which are based on the GDP growth projections shown in the International Monetary Fund s World Economic Outlook Database. The growth estimates for Nigeria start after 2015, thus the growth shown for 2014 and 2015 is actual growth. Table 35 Nigerian real GDP growth projections Year Increase % % % % % % % 2021 onwards 4.02% Table 36 shows GNI projections, based on the increases shown in Table 35. The GNI per person, based on the population increases above, are also provided, as well as the GNI per person for the population quintiles in Nigeria, assuming that the distribution of wealth in Nigeria remains constant. Table 36 GNI projections for Nigeria GNI US$489.2 billion US$711.6 billion US$ billion US$ billion GNI per person US$2 680 US$3 197 US$3 754 US$4 494 GNI per employed person US$9 776 US$ US$ US$ GNI per person in Q5 US$6 934 US$7 830 US$9 914 US$ GNI per person in Q4 US$3 055 US$3 449 US$4 050 US$4 849 GNI per person in Q3 US$2 033 US$2 295 US$2 695 US$3 227 GNI per person in Q2 US$1 374 US$1 552 US$1 822 US$2 182 GNI per person in Q1 US$760 US$858 US$1 008 US$1 207

58 388 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES The GNI per person in 2013 is as per the World Bank (2016d). However, if the GNI per person is calculated as the GNI divided by the population, the resulting GNI per person is US$ Figure 24 shows the GNI per person in each of the quintiles In 2013, Q1 of the Nigerian population was classified as low income. Q5 was classified as upper-middle and the remaining three quintiles were lower-middle. In 2035, Q4 is projected to move to the upper-middle category and, in 2036, Q1 is projected to move to the lower-middle category. From 2035, the top two quintiles will be upper-middle income and are expected to be able to afford insurance products, while the other three quintiles are projected to be in the lower-middle income categories and thus we expect insurance to be beyond the financial means of these individuals Penetration Rates Despite having the second highest GNI per person in 2013 after South Africa, Nigeria has the lowest insurance penetration of the countries considered. This may be because of the large Muslim population in Nigeria who do not purchase conventional insurance. However, the introduction of takaful to the Nigerian market began in 2005 when African Alliance Insurance first introduced takaful to Nigeria (Ismail, 2015) and should thus cater for the Muslim population A more likely reason for the low penetration in Nigeria may be the skewness of income. As at 2009, 69% of the Nigerian population were living on less than $3.10 (2011 PPP) per day and 48% (2011 PPP) were living on less than $1.90 per day (World Bank, 2016c). Figure 24 GNI per person in Nigeria

59 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES The population growth rate in Nigeria is high, although the rate of growth is projected to decrease slowly. Furthermore, a high level of job creation is required to maintain the employment-to-population ratio of If this ratio can be maintained, the employed dependency ratio is projected to decrease from 2.45 in 2013 to 2.08 in The dependency ratios are high and this is impacted by the highest fertility rates of all the countries considered The high levels of wealth inequity in Nigeria result in the majority of the population remaining in the low and lower-middle income categories for most of the projection period. Only Q5 is in the upper-middle income category, and Q4 is projected to move to the upper-middle category in Based on these projections, we do not expect the insurance penetration to increase significantly over the short term. We project that penetration will only start to increase significantly when the wealth of the first four quintiles increases. 14. Zambian Projections 14.1 Demographics The Zambian population pyramid, as at 1993, 2003 and 2013, as well as the projected pyramids for 2023, 2033, and 2043 are shown in Figure 25, as provided by the World Bank The population pyramids from 1993 to 2013 are similar, characterised by concave slopes and a population which is increasing. The population increases significantly from 2013 to 2043 and the slopes of the pyramid gradually become less concave, indicating an improving mortality rate Table 37 shows key past Zambian demographic indicators as at 1993, 2003 and 2013 as well as the projected demographic indicators as at 2023, 2033 and The World Bank provides the projections of the population in five-year age bands. The other demographics in the table above are derived from the World Bank projections Projections for the employed population assume the same employment-topopulation ratio as observed in The Zambian population is projected to more than double from 2013 to This follows an increase in the population size of 73% from 1993 to While growth has occurred across all age bands, the predominant growth is in the working population, which, as a proportion of total population is projected, to increase from 51% in 2013 to 57% in The elderly population as a proportion of the total population is projected to increase slightly from 3% in 2013 to 4% in 2043, while the

60 390 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES Figure 25 Zambian population pyramids child population is projected to decrease from 46% of the population in 2013 to 39% in The projected growth of the working population, of 167% from 2013 to 2043, may place pressure on employment rates, as it may be difficult for the level of job growth to match the increase in the working population. Figure 26 shows the absolute and percentage increase in the number of jobs in Zambia from 1992 to Thereafter, the absolute and percentage increase in the number of jobs, which is required to maintain the employment-to-population ratio of 69%, is shown from 2014 to The actual figures are to the left of the solid black bar, while the projected figures are to the right of the solid black bar.

61 C VAN HEERDEN & L MORONEY DEMOGRAPHIC PROJECTIONS TO PREDICT CHANGES IN LIFE AND NON-LIFE INSURANCE PENETRATION RATES 391 Table 37 Zambian key demographics Demographic factor Population 8.8 million 11.4 million 15.2 million 20.4 million 27.6 million 36.1 million Working population 4.4 million 5.8 million 7.7 million 11.0 million 15.3 million 20.7 million Employment-to-population ratio 64% 68% 69% 69% 69% 69% Employed population 3.0 million 4.1 million 5.6 million 8.0 million 11.1 million 15.2 million Dependency ratio Employed dependency ratio Percentage older than 65 years 3% 3% 3% 3% 3% 4% Percentage between 15 and 65 years 50% 50% 51% 53% 55% 57% Percentage younger than 15 years 47% 47% 46% 44% 42% 39% The growth of net jobs from 1992 to 2013 has been volatile in Zambia. The absolute growth in jobs required to maintain the employment-to-population ratio of 68.90% in 2013 shows an upward trend and a growth in jobs which is greater than the average net number of jobs created in the past Gross National Income Table 38 shows the real increases applied to the GNI from 2014 to 2043 for Zambia, which are based on the GDP growth projections shown in the International Monetary Fund s World Economic Outlook Database. All the increases shown in Table 38 are estimates. Figure 26 Employment growth in Zambia