Risk-pooling--necessary but not sufficient?

The search for a ‘‘good’’ health system is constant. Even in countries where comprehensively organized national health systems have existed for over a century, the debate on how best to design them to ensure the effective financing, organization and provision of care is as lively and controversial as ever. During the last quarter of the 20th century many countries proposed a wide variety of sweeping reforms, some of which were adopted and refined, and others rejected. That period can be viewed as one in which a ‘‘natural experiment’’ was conducted on a global scale. It has yielded surprisingly little, however, in the way of real, evidence-based understanding of what works well. Much of current research on health policy seeks to discover quantitative relationships that can explain the achievement of health systems in relation to their design and the wider context within which they operate. One such relationship which is already quite well understood concerns the role played by risk-pooling, or insurance (1). There is a growing consensus that, other things being equal, systems in which the degree of risk-pooling is greater achieve more. Risk-pooling is beneficial because health care costs are generally unpredictable and sometimes high. People cannot reliably forecast when they will fall ill and need to make use of health services. When it happens, the costs of those services can be significant. Risk-pooling increases the likelihood that those who need health care will be able to obtain it in an affordable and timely manner. It allows resources to be transferred from the healthy to the sick. From the viewpoint of individuals and households, contributions during times of good health can be used to meet health care costs in the event of illness. In many cases, pooling also contributes towards redistributive goals by making those with higher incomes contribute more in order to subsidize the poor. Community-based health insurance or ‘‘micro-insurance’’(2) is one approach to realizing the benefits of risk-pooling in countries without the organizational capacity to operate formal schemes on a national level. Risk-pooling, on any scale, has its downside, however. Clients who are insured and therefore do not have to pay the full cost of services may be inclined to over-use those services, while providers may be happy to let them do so because it enables them to earn more (moral hazard). In addition, those who are more likely to require care have a stronger incentive to join a voluntary risk-pooling scheme (adverse selection). Such problems could be reduced in micro-insurance schemes since their small scale and community focus might provide informal safeguards against them. Smallness in its turn is a mixed blessing though. As Dror explains in this issue of the Bulletin (pp. 672–678), the smaller a scheme is the harder it is for it to remain solvent. That is a statistical reality which cannot be avoided. He suggests reinsurance as an approach to help micro-insurance schemes overcome the threat of financial failure. Just as households can pool their resources to help reduce fluctuation in their health spending, so can insurance schemes themselves. Through reinsurance, schemes that have faced fewer demands on their funds than anticipated can indirectly subsidize those that have faced more. The potential benefits of reinsurance in countries where micro-insurers operate are clear. Unfortunately, however, reinsurance also faces the twin difficulties of moral hazard and adverse selection. Micro-insurance schemes that reinsure may have weaker incentives to contain the costs of the benefits they pay out; and schemes that are more likely to face budget blow-outs in any given period are more likely to choose to reinsure. Reinsurance is attractive because it expands the size of the risk pool, and allows greater scope for costs to be shared. In some circumstances, however, it may be preferable simply to establish schemes with larger risk pools from the outset. Examples already exist of district-based health insurance (3), targeting tens of thousands of people rather than a small community. Such larger schemes can also benefit from economies of scale in administration and transaction costs. Alternatively it may be feasible to encourage progressive scaling up of microinsurance schemes, possibly through a planned programme of mergers, to form ever larger risk pools. Such an approach could allow infrastructure and other capabilities to develop over time, and may eventually lead to the emergence of a single regional (or national) risk pool which many would consider to be the ideal. In addition, attempts to employ riskpooling will fail if they do not have the confidence of the people they seek to serve. Any pooling scheme will be judged by its results, and if problems of access or affordability persist people will quickly become cynical. Appropriate arrangements are needed to oversee the scheme so as to ensure that the benefits of risk-pooling are in fact realized. More fundamentally, those who pay contributions when they are healthy must know that, if and when they need care, the pooling scheme will still be there to help them. The failure of an insurer can be as catastrophic for those affected as the collapse of a bank. Countries may thus need to regulate insurers and micro-insurers to ensure financial probity in the same way that they regulate the banking industry. The importance of oversight and regulation for effective risk-pooling should not be overlooked. Similarly, other initiatives to improve health system performance — whether by increasing the efficiency of service providers, building up human resources or enhancing access to traditional healing — will struggle where government is weak in its role as protector of the public’s interests. Success in that role, defined by the World Health Report 2000 — health systems: improving performance as stewardship, is undoubtedly a key factor in determining levels of health system achievement. In contrast to the concept of pooling risk, however, little evidence currently exists as to what constitutes effective stewardship, how to bring it about and themechanisms bywhich it has an impact on achievement. Therein lies another key challenge in understanding the complex relationships that link policy and performance. n