Micro-Insurance in the Context of Social Protection Thankom Arun,1 Susan Steiner 2 1 University of Central Lancashire 2 University of Manchester October 2008 [email protected] [email protected] BWPI Working Paper 55 Brooks World Poverty Institute ISBN : 978-1-906518-54-7 Creating and sharing knowledge to help end poverty www.manchester.ac.uk/bwpi Abstract This paper aims to understand the role of micro-insurance as an element of social protection. It outlines the current status of micro-insurance provision in Ghana and Sri Lanka, two countries with very different socio-cultural backgrounds. It concludes that both countries are unlikely to extend their social security systems to the entire population in the short to medium term, making private micro-insurance initiatives essential mechanisms to help people reduce their vulnerability. Keywords: Micro-insurance, Social protection, Ghana, Sri Lanka Thankom Arun is Reader in International Finance at the Lancashire Business School, University of Central Lancashire; Honorary Senior Fellow, University of Manchester; and Research Associate, Brooks World Poverty Institute, University of Manchester. Susan Steiner is PostDoctoral Research Fellow at Brooks World Poverty Institute, University of Manchester; and at the German Institute of Global and Area Studies, Hamburg. 2 1. Introduction All households in developing countries, whether better or worse off, are exposed to a variety of risks, such as illness, disability, death, unemployment, crop failure, natural catastrophes, or crime. Low-income households, however, are less able to prevent and mitigate risks than others; and in the case of shocks, they are less able to cope with the consequences (Churchill, 2006). They are therefore more vulnerable to risks, i.e. they are more likely to experience a significant decline in wellbeing when a shock occurs (World Bank, 2001). Social protection measures, which include ‘all interventions from public, private, voluntary organisations and social networks, to support communities, households and individuals, in their efforts to prevent, manage and overcome vulnerability’ (CPRC, 2008, 53), are thus essential to prevent people from falling (deeper) into poverty (Jacquier et al., 2006). One such measure is micro-insurance, which promises to be an effective strategy for people, both currently poor and non-poor, to mitigate risk and reduce their vulnerability to shocks (Dercon et al. 2008; Siegel et al. 2001). Micro-insurance is the ‘protection of low-income people against specific perils in exchange for regular premium payments proportionate to the likelihood and cost of the risk involved’ (Churchill, 2006, 12). In equivalence with regular insurance, micro- insurance serves as an instrument to isolate fluctuations in consumption from fluctuations in income and wealth (consumption smoothing). The central underlying principle is the pooling of risks, which implies that financial contributions are collected from the members of an insurance scheme, and the loss of one individual is spread among all members in case of risk occurrence. The main difference between micro- insurance and regular insurance is that the first is specifically targeted at low-income people, who have limited financial resources and often irregular income flows. Thus, the product design is adapted to these people’s needs and financial capabilities. In many countries around the globe, microfinance institutions (MFIs), cooperatives, (rural) banks, service providers, commercial insurance companies, as well as informal groups, have started to provide the low-income population with micro-insurance. On the one hand, this has happened in response to the fact that poorer segments of society generally do not have access to formal insurance mechanisms, provided by either the state or private insurers, and instead rely on imperfect informal insurance. In many developing, particularly the poorest, countries, public social security systems cover employees of the formal sector, civil servants and the military but not informal sector workers, who make up the majority of the population. Market-based insurance is often non-existent, and where it exists, it is only accessible to the better-off, as premiums are beyond low-income people’s capacity to pay. On the other hand, the emergence and expansion of micro-insurance were encouraged by the now extensive experience and 3 widespread success with the provision of loans and savings products to the poor. In fact, many micro-insurance products are closely linked to MFIs, partly because existing networks make it less costly to deliver new products, and partly because these institutions have started to tie their loans to insurance against the death of the borrower. Even though there is a large and still growing literature on social protection and its many facets (Barrientos and Hulme, 2008, and the contributions therein), the question of micro-insurance in the context of social protection has so far received only limited attention from the international research community (with the exception of e.g. Jacquier et al., 2006; Dercon et al., 2008). Micro-insurance schemes can be crucial components of more comprehensive social protection systems. For instance: 1) micro-insurance schemes may achieve redistribution through internal cross- subsidies or by channelling public subsidies to their members; 2) micro-insurance schemes may have a significant socio-economic impact on members and non-members; and 3) micro-insurance schemes can play a role in the empowerment and participation of their members (Jacquier et al., 2006) This paper aims to understand the nature of micro-insurance products and their relation to social protection systems, policies and instruments. Since it appears highly interesting to investigate the interrelation between social protection and micro-insurance in countries with completely different socio-cultural backgrounds, we focus on Ghana and Sri Lanka. The remainder of this paper is organised as follows. In the following section, we provide a short overview of micro-insurance provision worldwide. We then illustrate the prevailing micro-insurance institutions and products in our two country cases – Ghana and Sri Lanka. Finally, we relate the experience with micro-insurance in these two countries to the overall social protection goal. 2. Micro-insurance provision in developing countries The global outreach of formal micro-insurance still appears to be rather limited, as a recent study of the MicroInsurance Centre – a consultancy for micro-insurance – shows (Roth et al., 2007). Out of the 100 poorest countries in the world, micro-insurance could be identified in 77. The number of micro-insured people was estimated to amount to 78 million, which is not a particularly high number, given that China and India are both among the 77 countries with micro-insurance. Due to the high population numbers in these two countries, the Asian region accounts for 86 percent of the global outreach of micro-insurance. Nevertheless, only 2.7 percent of the poor population in Asia were covered by micro-insurance; and the coverage of the poor in Africa and Latin America was 0.3 percent and 7.8 percent, respectively. 4 This low coverage is partly due to the fact that insurance markets in most developing countries are still in their infancy, which can be illustrated in terms of penetration (premiums in percent of GDP) and density (premiums per capita). For example, out of 87 analysed countries worldwide, South Africa (rank 32) and Namibia (rank 44) are the only sub-Saharan African countries that rank relatively high in terms of density (Swiss Re, 2007). Angola (rank 74), Kenya (rank 82) and Nigeria (rank 86) are positioned at the lower end of the ranking. In terms of penetration, the picture looks better, though not satisfying for all countries (South Africa rank 2; Namibia rank 16; Kenya rank 53; Angola rank 60; Nigeria rank 84). Micro-insurance providers worldwide currently offer four types of insurance: life, health, accidental death and disability, and property insurance (Roth et al., 2007). Sixty-four million people are covered by life insurance, 41 million by accident and disability insurance, 36 million by property insurance, and 35 million by health insurance. Clearly, life insurance is the most widely distributed product. Reasons for this are manifold. Life insurance is one of the most demanded forms of cover (together with health insurance), it is easy to price, resistant to problems of fraud and moral hazard, and independent from the existence of other infrastructure, such as clinics and hospitals. Although health insurance is a highly demanded insurance policy as well, the actual availability is well below three percent, even in the best cases. Property insurance (including home, crop, weather index, livestock and other possessions insurance) is not very widespread either; only 0.7 percent of poor people are covered by property insurance. On the one hand, many micro-insurers, particularly non-profit insurers, ignore property insurance, as demand for property insurance is significantly lower than demand for life and health insurance. On the other hand, underwritings and claims validations are very difficult in the property policy. Compared to other micro-insurance types, the number of accidental death and disability insurance products is much smaller; the largest numbers of products and numbers of lives covered can be found in South Asia. One major obstacle in the micro-insurance business is the relatively low level of experience of the target group with formal insurance. Many people do not understand the concept of insurance, let alone the terms and conditions of a contract, and are reluctant to pay in advance for a service they may not ever receive (Cohen and Sebstad, 2006). In East Africa, people have been found to confuse insurance with savings, and to believe that they must use the service for which they pay premiums, resulting in unnecessary visits to doctors’ practices (Millinga, 2002). It thus appears that education on insurance, or, as some people call it, promoting financial literacy (e.g. Cohen and Sebstad, 2006) is one of the crucial areas where micro-insurance providers as well as donors should engage, in order to make micro-insurance a viable enterprise. 5 2. Micro-insurance: Experience from Ghana and Sri Lanka Despite the comprehensive Ghana Poverty Reduction Strategy (GPRS II), risk management for the poor does not play a significant role in Ghana, in either public or private sector policies. In Ghana as in other countries the insurance industry has mostly focused on the upper income market. An important difference in Ghana is that there is recognition by insurers that they need to move down market in order to increase the scale of their activities. (CARE International, 2004, 9) Nevertheless, there is currently only one commercial insurer providing micro-insurance – the Gemini Life Insurance Company (GLICO). Between 2001 and 2004, CARE International, in cooperation with GLICO, designed the so-called Anidaso Policy (Anidaso = ‘hope after grieving’). The policy is specifically targeted at low-income people, both in urban and rural areas. After the product development and trial phase, GLICO took full responsibility for the product. To our knowledge, donors have not been engaged in the provision of the Anidaso Policy, or any other formal micro-insurance product, since then. The policy offers term life assurance up to age 60, accident benefits (income protection with total/partial temporary/permanent disability benefit lumped together) and in-hospitalisation benefits (calculated per each day spent in hospital) for the policyholder, the spouse, and up to four children. Contributions towards purchase of annuity upon maturity of the policy (which serves as a savings scheme) can be added, on a voluntary basis. Average premiums range from two to four Cedi, or more if the savings component is chosen. GLICO currently cooperates with 26 rural banks and MFIs all over the country, who are charged with the distribution of the policy. These institutions appoint one of their staff members to act as the Personal Insurance Advisor, being responsible for the sale and after-sales service of the policy. In November 2007, GLICO counted about 14,000 policyholders. Given that Ghana’s total population is about 19 million and the poverty headcount 28 percent, this number of micro-insured people is still quite low. The Anidaso Policy does not include a health insurance element; yet, the public National Health Insurance Scheme (NHIS) caters for this need. The NHIS is not specifically targeted at low-income people and is hence not a micro-insurance mechanism in the narrow sense, but it provides benefits to the entire population. It was launched in 2004 and replaced the cash-and-carry healthcare system, whereby patients were required to pay for treatment at the time of seeking it. This system is widely known to have resulted in a decline of health services utilisation (Waddington and Enyimayew, 1989). In contrast to this earlier scheme, the NHIS provides free-of-charge medical care at public hospitals, recognised private hospitals and health centres for contributors, their dependents and indigent people. Contributors to the scheme are both formal sector employees, who pay 6 into the Social Security and National Insurance Trust (SSNIT) (making up 12 percent of all members in June 2007), and informal sector workers, who can join the scheme voluntarily (24 percent of all members). The contributions of these two groups subsidise the healthcare for their dependents, i.e. children and elderly persons, SSNIT pensioners as well as indigents (together 64 percent of all members). In June 2007, the number of NHIS members amounted to 8.2 million; or approximately 38 percent of Ghana’s total population. Given that the scheme had only been initiated three years earlier, this is a considerable coverage rate. It shows that the NHIS has been well received, especially among informal sector workers who had hitherto often done without using health services, as a result of lacking resources and insurance alternatives. Yet, a recent study found that membership is still lower in the lower socio-economic quintiles, limiting the desired effect of the scheme in terms of equitable access to healthcare (Asante and Aikins, 2008). Beside these two formal micro-insurance schemes, informal insurance mechanisms are very widespread in Ghana. Since funerals are an important and expensive social event in Ghana (de Witte, 2003), it is not surprising that there are numerous funeral societies (CARE International, 2004). Osei-Akoto (2004) has investigated the existence of other informal groups and their support in terms of insurance provision to their members. He found that between 30 and 55 percent of adults in West Gonja and Nkoranza districts join informal groups of different kinds, such as occupation-based associations, religious- based groups, women groups, savings and credit associations, or hometown and tribal- based associations. These groups obviously provide a wide range of services, but members can expect funeral donations and support for healthcare expenses from most of them. There are four different approaches in Sri Lanka to assist the poor in risk management. They are: 1) government agencies, which provide social security schemes and pension for low-income earners; 2) Samurdhi Authority, another government agency, which provides income support along with social security benefits; 3) private insurance companies; and 4) MFIs and community-based organisations (CBOs), which provide micro- insurance. The Sri Lankan government spent three percent of its GDP on social protection in the year 2004. This social protection system comprises three main components: employment protection and promotion; safety nets; and social security/insurance programmes. Sri Lanka provides basic protection of core labour standards to formal sector workers. The Sri Lankan safety net system comprises the Samurdhi programme 7 and disability payments. The Sri Lankan social security system can be divided into two groups: formal and informal. Both formal and informal insurance systems provide pension, disability, survivors and unemployment insurance. However, there is a large gap between the two sectors in the context of coverage. Two publicly sponsored programmes offer social security for informal sector workers: Samurdhi, and the Agriculture and Agrarian Insurance system. Under the Samurdhi Act, introduced in 1995, families whose monthly income is less than Rs1,500 are eligible for receiving Samurdhi benefits. Around two million families in Sri Lanka are currently receiving Samurdhi benefits. The monthly income supplement provided by this programme varies from Rs 250 to Rs1,000, depending on the size of the family and its income level. Families receiving between Rs500 and Rs1,000, i.e. the highest three levels of income supplements, have compulsory deduction made for savings (Rs100), insurance (Rs30) and a housing lottery (Rs10). A person who is older than 65 years is not eligible for the insurance scheme. Samurdhi provides a death benefit of Rs5,000, Rs2,000 on birth of the first child, and Rs1,000 for subsequent births. In addition, Rs3,000 on marriage, and hospitalisation benefits of Rs50/day, for a maximum 30 days, are available. Table 1: Largest formal micro-insurance providers in Ghana and Sri Lanka Institution Insurance Ghana Gemini Life Insurance Company (GLICO) ‘Anidaso policy’ – life, hospitalisation, accident, voluntary pension National Health Insurance Scheme (NHIS) Health Sri Lanka Sarvodaya Economic Enterprises Life, loan protection, funeral, hospital cash Development Services Ltd (SEEDS) benefits Yasiru Mutual Provident Society Death, health, hospitalisation Women’s Development Federation – Death, marriage, childbirth, hospitalisation Hambantota and surgical operations All Lanka Mutual Assurance Organisation Life savings, accident, health, funeral (ALMAO) Source: Authors’ illustration The Agriculture and Agrarian Insurance Board is the other government agency which provides social security for low-income people. It has two insurance schemes: pension, death and disability insurance; and crop and livestock insurance. On the other hand, informal sector workers participate in funeral aid societies, which can be introduced as quasi formal groups. The ADB (2006) has proved that most Sri Lankans are willing to buy funeral benefits. The funeral aid society concept was the origin for micro-insurance 8 schemes in Sri Lanka. These societies provide assistance in case of death. This assistance is not limited to the member of the society, but applies to other family members as well. There is no standard type of assistance and it may vary from one society to another. Recently, a few organisations have started micro-insurance schemes in Sri Lanka on this foundation. For example, in 1991 the All Lanka Mutual Assurance Organisation (ALMAO), which is one of the apex micro-insurance schemes in Sri Lanka, was started by seven cooperating funeral aid societies. Private commercial companies are not serious service providers of micro-insurance in Sri Lanka. At the outset, commercial companies entered the micro-insurance field with insurance products that are linked to microfinance programmes, such as the micro- insurance programme of Gamipupuduwa (‘village awakening’) conducted by Hatton National Bank (HNB). Commercial insurance companies are not interested in offering life insurance to low-income groups. The problem is that there is public apathy towards contract life insurance, because the high inflation rate in Sri Lanka has rendered the maturity value of term life insurance policies almost insignificant. Even those who had the misfortune of contracting life insurance policies either allow them to lapse or opt for premature surrender value. This is the dilemma facing the life insurance business in Sri Lanka. Only a few private companies are dealing with the sale of micro-insurance life products. The penetration of life insurance is therefore very low. For example, HNB Assurance provides micro-insurance products which cover life benefits. Heyleys AIG Insurance Company has entered into the micro-insurance market with a personal accident policy for low-income workers, such as estate workers, farmers and fishermen. In addition, Ceylinco Insurance Company provides three micro-insurance products: ‘Ceylinco Grameen Govi Udana’, ‘Ceylinco Govi Rakshanaya’ and ‘Ceylinco Livestock Insurance’ for farmers, and one micro-insurance product named ‘Grameen Devara Udana’, for fishermen. Amana Takaful Insurance (ATL) has introduced a micro takaful policy named ‘Navodaya’ (‘dawn of a new era’), catering to low-income groups. This product is targeted at factory blue-collar workers, workers on tea estates and in the apparel sector, labourers, and other self-employed workers. This product appears to be the first branded micro takaful product to be marketed in the world. There are no regulatory supplies for micro-insurance in Sri Lanka, similar to many other countries. Most of the micro-insurance providers have registered under the Society Act. Micro-insurance emerged as a service support activity in the field of microfinance and it was mainly concerned with loan protection insurance, e.g. Sarvodaya Economic Enterprises Development Services Ltd (SEEDS). Presently, it has developed to provide a wide range of insurance cover, with linkage to products such as health and property insurance. MFIs have strengths in providing micro-insurance to their clients in Sri Lanka, as in most other developing countries. This strength comes from the fact that MFIs have an institutional framework which is closer to people, and they have generally gained people’s trust. For example, in 1996, ALMAO merged with the insurance section of Thrift 9 and Credit Societies (Sanasa Federation). Sanasa Federation is a well established MFI in Sri Lanka (ILO, 2004b). However, as a delivery channel of micro-insurance in Sri Lanka, we can identify some weaknesses of MFIs. These are: lack of actuarial and other insurance knowhow; the risk of venturing into a new and complex field of financial services before they are soundly established; and the danger that the insurance concept may divert the MFIs from their original role. The considerable capital base needed for insurance can be identified as another drawback of MFIs as a delivery channel. For example, the financial requirement for a long-term life insurance is Rs25 million, and for general insurance, Rs50million. There is very limited donor support for Sri Lankan micro-insurance providers. Only one micro-insurance scheme, Yasiru Mutual Provident Society, receives significant technical assistance from a donor agency named RoboBank Foundation, via the RoboBank Groups’ insurance company – Interpolis Re (ILO, 2004b). They have supported Yasiru since 1997 and their support is limited to the Yasiru Mutual Fund. Even though the World Bank does not support micro-insurance directly, it appears to be the promoter of various MFIs that are dealing with micro-insurance or social security, e.g. Kandy Women’s Association and Hambantota Women’s Association. On the other hand, the United States Agency for International Development (USAID) supports the Women’s Development Federation (WDF) in Hambantota, but with a focus on microfinance. Even though the Asian Development Bank’s Rural Finance Sector Development Programme generally mentioned ‘insurance’ under the assessment of financial needs of the rural population, it does not explicitly plan to support micro-insurance. 4. Implications for social protection In both Ghana and Sri Lanka, there is an increasing tendency to provide formal micro- insurance, with the number of providers being clearly higher in Sri Lanka. Public as well as private institutions engage in the field of micro-insurance, though possibly for differing reasons. As clearly stated above, Ghanaian commercial insurers regard micro-insurance to be a way to expand their market reach and secure future profits; this does not necessarily hold true in Sri Lanka, as the example of the first takaful micro-insurance shows. The public initiatives in both countries are, rather, attempts to extend the existing social security systems and include more people in social insurance programmes. However, as exemplified by Ghana’s NHIS, the outreach of public schemes is likely to remain limited to a certain percentage of the population and, more importantly, biased towards more affluent groups. Private – both commercial and non-profit – initiatives are therefore essential mechanisms ‘to support communities, households and individuals, in their efforts to prevent, manage and overcome vulnerability’ (CPRC, 2008, 53), and hence to provide social protection to the people. In neither Ghana nor Sri Lanka is the state likely to expand its social security system to the entire population in the medium 10