«Protecting the poor A microinsurance compendium Edited by Craig Churchill Protecting the poor A microinsurance compendium Protecting the poor A ...»
This chapter first summarizes the main types of health microinsurance providers and then analyses their relative effectiveness in meeting the needs of the low-income market over the long term.
1 Institutional options All insurers must satisfy the basic value proposition, namely that they reduce the long-term cost of the risk for the insured. An additional requirement, which is specific to microinsurance, is that the type of organization should function effectively within an environment of low premiums. As discussed in Chapter 2.1, such a situation might lead to severe rationing of benefits and, when coupled with a broad variety of insurance needs for the heterogeneous
low-income market, result in product fragmentation to suit many small groups of clients. To reduce the long-term cost of risk, the insurer has to aggregate many individual risk profiles that have different statistical distributions and that can be diversified over time. While some types of microinsurance providers can fulfil the requirement for large numbers more easily than others, they might suffer from other weaknesses in the business process.
The typology presented in this chapter considers four main providers of health microinsurance: 1) licensed insurers operating the “partner-agent” model, 2) the charitable insurance model, 3) healthcare providers that also operate health insurance and 4) the mutual model discussed in the previous chapter.2 This typology results from distinguishing organizations along two dimensions: a) the primary motivation for entering the market, since this motivation influences the design of the business process and hence the product, and b) the entity bearing most of the risk of losses, as depicted in Figure
26. The description of the organizational models that follows contains a discussion of their advantages and disadvantages in fulfilling business-process functions, and their effectiveness in minimizing (or not) conflicts of interests within the system. The analysis also takes into account certain differences, such as governance mechanisms.
2 The authors recognize that there are certainly other institutional arrangements as well, but the typology described here emerges most clearly from the case studies. Furthermore, one could divide these four provider types into sub-categories, some of which would have “hybrid” characteristics of more than one provider.
Institutional options for delivering health microinsurance 403
1.1 The partner-agent model As described in Chapter 4.2, under the partner-agent model the relationship between the policyholder and an insurance company (“the partner”) is facilitated by an intermediary (“the agent”) such as an NGO, a microfinance institution or any other organization with close contacts to the target group.
Owing to the regulatory conditions discussed in Chapter 5.2, examples of the
partner-agent health microinsurance model are common in India, including:
– VimoSEWA and ICICI Lombard – Shepherd and United India Insurance Company (UIIC) – Karuna Trust and National Insurance Company (NIC) The insurance company is responsible for all decisions affecting product manufacturing, sales, servicing and maintenance of long-term sustainability, i.e. it carries the risk. Although it may consult the agent organization when designing a product, the insurer maintains control over the strategic operations that define the risk transfer mechanism.
The agent deals with sales and product-servicing within the boundaries of the products that the insurance company is allowed to sell, and at commissions that meet the regulatory limits or are agreed on with the partner.
Agents have better knowledge of (and ties to) the target market, but their primary role is to represent the insurer to the clients. This is an area where conflicts of interest might arise, as the agent organizations usually regard themselves as advocates for their clients, and might feel uncomfortable communicating the insurance company’s position.
Consider the case of a claim settlement procedure, where the agent needs to defend the insurer’s position to its clients. If a conflict arises over whether a claim is valid and should be paid, the agent might need to agree with one side, running the risk of alienating the other. Usually, its position as an agent of the insurer means having to side with the latter, and communicate the rejection of a claim to the client. If such cases occur frequently, agents might find their reputation in the community damaged and the community’s trust in them – the very attribute that attracted the insurer to the agent – will diminish or be lost. Therefore, in practice, agents such as VimoSEWA occasionally cover claims from their own coffers if they feel that the claim rejection is unjustified.
404 Institutional options Another potential conflict arises with adjustments to the premium levels.
For example, after BAIF’s claims ratio had exceeded 100 per cent, UIIC decided to increase the premium charged to BAIF’s insured clients by about 80 per cent. Unable to justify such a rise to its clients, BAIF decided to turn its insurance scheme into a mutual.3 As illustrated in Figure 27a, neither clients nor healthcare providers have direct input into the production process, and bear no responsibility for longterm sustainability. The agent’s role is usually also confined to sales and aftersales service, although the latter is sometimes dealt with by the insurer directly or through a third-party administrator (TPA). For example, the Arogya Raksha Yojana scheme near Bangalore, India is linked up with ICICI Lombard for health insurance, and has contracted a TPA for administration (Figure 27b).
In the partner-agent arrangement, each side can benefit from the comparative advantages of the other, but a couple of inherent problems often remain unresolved. An insurance company is usually interested in selling a predesigned product (often a scaled-down version of its products for the formal sector). This type of product is easier for the company to monitor and does not need to be priced anew. Some insurance regulators also require new products to be registered and few companies are willing to do this for every agent/community; usually, only large agent organizations have the negotiating power to push for a tailor-made product.
This lack of flexibility in product design is particularly important because, as discussed in Chapter 2.1, product features are more likely to influence adoption among the target population in health insurance than in life or property insurance. However, this problem can be solved in the partner-agent model. Due to its proximity to the target group, the agent should be well placed to explore the actual demand, while the insurer can use its actuaries to turn the demand into a well-priced product. This was the arrangement for Karuna Trust, which engaged in a detailed demand analysis before linking up with NIC. Although the benefits demanded and the price negotiated caused a severe headache for NIC’s actuaries, the insurer was willing to pilot this scheme. Similarly, Shepherd and UIIC designed a benefit package together making use of their respective competencies.
3 BAIF is an NGO working on rural development in India. Coming from cattle breeding, it subsequently expanded its activities to a broad variety of services and now provides microfinance and, in one pilot area, life and health microinsurance.
Institutional options for delivering health microinsurance 405
For health microinsurance, the agents’ real comparative advantage (and hence their source of attractiveness for the insurer) is highlighted in the sales process. Insurers, which often lack a relationship of trust and access, both physical and psychological, to potential clients, rely on the agent’s proximity to the market and the trust built up over the years through the agent’s other operations. However, market penetration is one thing, but complete transparency another: clients quickly realize that there is little incentive for them to provide information about their health status or about a neighbour who they know is withholding information, and so the flow of information in both directions is incomplete in the partner-agent model. This constitutes an increased risk to insurance companies for which their shareholders (logically) expect to be compensated by increased returns (invariably, ceteris paribus, leading to higher premiums). Higher premiums in turn result in clients’ increased demand for “value for money” and thus amplify moral hazard (again a higher risk for the insurer). Thus, a vicious cycle of dysfunction can evolve which may cause the opportunities inherent in this model to be squandered. For as long as risk and returns are not balanced from the insurer’s perspective, there will be no incentive to enter the market in a meaningful manner.
This incentive problem is amplified when it comes to product-servicing and claim verification. The insurance company may expect the agent to verify the claims, and if so hopes that the strong ties of the agent with the target groups will ensure a good flow of information. However, as in any commercial insurance scheme, clients have no incentive to provide information that will benefit the insurance company at their (or their neighbour’s) expense.
Clients might even consider it legitimate to cheat a large company in a distant city following the logic: “we are poor and they are rich, so they can pay.” This manifestation of the “them and us” paradigm implies an attachment to certain networks, norms and trust at the community/client level – which captures the essence of the social capital concept – at least from the perspective of Putnam (1995) and his followers.4 As insurance companies experience this problem with clients from every market segment, they establish monitoring mechanisms for verifying claims.
However, these mechanisms are costly, and in the context of microinsurance may be prohibitively expensive to the point where affordability for the poor 4 The concept of social capital has been the subject of much interdisciplinary examination over the decade following Robert Putnam’s 1995 article “Bowling alone: America’s declining social capital”. It is generally accepted that if authors wish to use the term, they should define how they will use it. While it is not within the scope of this book to develop a definition of social capital in the context of health insurance for the poor, references to some useful reviews of the subject by Farr (2004), Manski (2000), Portes (1998), Sobel (2002) and Woolcock (1998) are included in the bibliography.
Institutional options for delivering health microinsurance 407 could be jeopardized. Furthermore, attempting to solve this problem according to the logic of the traditional business process for high-net-worth clients is the hallmark of scaled-down commercial insurance products – hardly an innovative health microinsurance solution.
To solve this problem, VimoSEWA trained its people in claims investigation techniques, established a claims committee with appropriate expertise and persuaded its insurance partner to allow it to adjudicate claims. However, more can be done: synchronizing the clients’ incentives with the incentives of the insurance company (e.g. through profit-sharing arrangements) modifies the business process in such a way that the problem might not arise in the first place, as clients would then have an increased incentive to keep information flowing (perhaps not about themselves but about others who are cheating the system).
A similar (though not identical) application has been extensively documented in the related field of microfinance, whereby mechanisms (notably joint liability and contingent renewal) have been put in place to use the power of communities to compensate for the information advantage customers had over the lender (Van Bastelaer, 2000). This is an excellent example of social capital at work – replacing traditional, more formal (and costly) means of evaluating creditworthiness used by commercial banks with peer pressure and character-based lending (DeFilippis, 2001).
The main point is that a true sense of ownership and “buy-in” among the clients (preferably through leveraging communities’ social capital) is indispensable for a successful health microinsurance scheme, and might be even more important than in other business areas of corporate insurance companies.