1.1.         Budgets and resource scarcity

This report starts from the assumption that we are operating under conditions of resource scarcity and fixed budgets. The budgets in question are typically government budgets, themselves funded through tax revenue.[1] Studies may make the case for the expansion or reduction of budgets. This report, however, is only concerned with the dilemmas policy makers face when confronted with a fixed budget. There is thus one important disclaimer to make: certain moral imperatives, such as fostering equality of opportunity for all persons, are not discussed. Often, they are good arguments for the expansion of a given budget, or the creation of new budgets with specific purposes. Limited and fixed budgets, however, call for effective allocation of resources.

Similarly, this report does not consider whether some of the shortcomings of the policies in question give grounds to creating new, separate and dedicated budgets to tackle the problems that cannot be tackled by existing budgets. The question under consideration is what system of resource allocation will generate the most overall ‘good,’ and which one will satisfy ‘needs-claims’ in the most effective way possible.

Finally, this report does not consider the issue of unlimited budgets. Unlimited budgets exist only in theory, and even societies that allocate ‘unlimited budgets’ to some policy areas are confronted with the realities of resource scarcity. In healthcare, for example, the ‘unlimited budgets’ of the Soviet Union resulted in every Soviet citizen being fully covered in theory – to the full extent of his or her needs – but in practice, this was at the cost of sometimes interminable waiting lists.[2]

1.2.         Needs that are rare

This report is concerned with high-cost special needs (HCSNs hereafter) in education and healthcare. These two policy areas were chosen because they are very different and yet conjure similar dilemmas in the case of HCSNs. Insights generated by this report are thus less likely to be policy area-specific, and hopefully can be extrapolated to apply to other policy areas.

The special needs in question are rare in terms of statistical frequency. There is no strict definition for rarity. A rare disease in one part of the world can be widespread in another. Statistical definitions of rare diseases, however, range from 1 in 1,000 to 1 in 200,000.[3]

An example of this is the disease known as ‘Jumping Frenchmen of Maine,’ which affects people’s reflexes at the rate of 1 in 200,000, or 20 to 21 people in New Zealand.[4] Some disabilities that are relevant in an educational context are often less rare, but still affect a very small percentage of the population. Non-verbal Learning Disabilities (NVLD hereafter) are found in 10% of children with learning disabilities, which themselves represent 4 to 5% of the school-age population, or 1 in 200 to 250. Other learning disabilities with genetic roots are even rarer.[5]

1.3.         Needs that are high-cost

Meeting these special needs sometimes requires exceptional levels of spending to be provided for. Sometimes, meeting those needs will in fact not be costly at all. The need in question would then not be classified as a ‘high-cost’ special need. This report is concerned with special HCSNs where rarity is a problem because it precludes the emergence of economies of scale. In the case of pharmaceuticals that are designed for small patient populations, start-up costs per individual are very high. This can push the cost of providing for those populations beyond what a budget may allow, or beyond a spending threshold above which a medicine will be deemed to provide poor value for money.[6] In the case of education, meeting certain needs may require for example developing teacher training programmes or changing the design of educational facilities that will only be used by a very small proportion of the population.

A mathematical representation of this is in the case of funding for a specific drug would be:

 B/T = A

where B is the projected budget, T the total population the budget is concerned with, and A is the average spending per patient projected in the budget. In cases where decisions had to be based on A, we could say that if:

 X/P > A

where X is the start-up cost of a drug, and P the total target patient population for that drug, then the drug should not be funded by the government. Indeed, where P = 500,000, the start-up cost per patient will be X/500,000, which is likely to be closer to A than, say, X/50. The prima facie case for this is a fairness claim: in cases of overspending on some pharmaceuticals, others will not get funded and some patients will lose out.

1.4.         Policy problems: 3 factors in decision-making

Decisions are, of course, not made in such a simplistic and mechanistic fashion. This would conflict with powerful ideas we hold about how human beings should be treated. For example, they should not be made to bear the consequences of sheer bad luck. There are also more sophisticated views on the matter: democratic politics is rife with ‘rights-claims’ and ‘needs-claims’ for equal education and healthcare for all, no matter the cost. Some people find the idea of making decisions based on a utilitarian framework is morally reprehensible, however decision-makers have to take into account the comparative efficiency of different policies. Finally, spending decisions can have positive spill-over effects in other areas. These three issues are detailed below.

1.4.1.        Utilitarianism

Broadly, utilitarianism is interested in the maximisation of ‘utility,’ defined variously as happiness, pleasure, welfare or even ‘good.’ The claim here is that resources should be allocated to maximise welfare, i.e. for the ‘greatest good of the greatest number.’[7] The main relevant objection here is that it is very difficult to measure ‘utility.’ If we count it as ‘preference-satisfaction,’ the problem seems endless, as preference is a subjective measure, and people’s preferences can be adaptable, or be the result of cognitive biases.

There are, however, ways around this problem for the two policy issues under consideration. In the medical world, practitioners resort to the concept of a ‘QALY,’ or Quality-Adjusted Life Year. In the educational world, standardised achievement tests such as literacy and numeracy tests can be used as proxies for gains in education. Both these measurement methods have their detractors, of course. After all, the ‘quality’ of a life year (or to be precise, here the extra utility per life year derived from some medical operation) is to some extent a subjective concept.[8] They serve, however, to measure the effectiveness of spending: other things equal, it stands to reason that we should prioritise spending for policies that achieve greater QALY gains, or greater gains on literacy and numeracy tests.

1.4.2.        Rights and needs

The language of ‘rights’ is very powerful. It is often invoked to defend policy demands made on decision-making bodies. An argument that is often made is that, if people indeed have equal rights to education and healthcare, then failing to provide them with resources on a needs basis (rather than an ‘equal distribution’ basis) constitutes a breach of their rights and is detrimental to their human dignity.[9]

Some learning disabilities such as the Mathematics Learning Disability have been shown to affect boys to a greater extent than girls.[10] Similarly, rare diseases can affect one sex only and the non-provision of resources to address these specific HCSNs can lead to accusations of gender discrimination. This is the claim made in the case of the campaign for PHARMAC funding for herceptin, a drug used to fight breast cancer.[11] Likewise, agencies like New Zealand’s PHARMAC are discriminating against certain disease states, claims a report on high-cost pharmaceuticals.[12] However, rights-claims often compete against other rights-claims: in other words, there can be conflicts of rights, which in practice will mean that sometimes, some rights will be overridden.

1.4.3.        Other benefits

Pharmaceuticals. The key benefit resulting from funding high-cost pharmaceuticals is the impact this has on pharmaceutical R&D. By subsidising the cost of a drug for patients, the government increases the size of the market for that drug. This makes it easier for pharmaceutical companies to recoup their investment. This argument is not a trivial one, as many important scientific discoveries are made by chance, or at least are unanticipated. Medical discoveries of use to the broader population might be made while developing high-cost pharmaceuticals, and failing to fund these high-cost pharmaceuticals might lead to fewer such discoveries.[13]

Education. Spreading the educational net wide undoubtedly increases inclusiveness in a community. Other things equal, the outcomes are things that are intrinsically good, such as greater equality, greater autonomy for the individuals concerned, etc. It is often difficult for policy makers to quantify the benefits of increased inclusiveness, however. In the case of learning disabilities, linking specific policies with specific outcomes can be difficult: policies are not implemented in a vacuum, and correlation (let alone causation) can be difficult to establish.

1.5.         Knowledge gap

To sum up, there are two relevant moral tensions. In the first place, a choice has to be made between providing for the HCSNs of some persons and the opportunity costs to those denied beneficial healthcare or educational resources that are spent instead of meeting very few people’s needs at very high cost. In the second place, decisions have to be made in the knowledge that current measurement methods are flawed and do not capture the whole range of benefits that a funding decision might provide.

Decisions can be made on a case-by-case basis. There are important drawbacks in defaulting to this position, however. Case-by-case decision-making is likely to be less consistent and transparent, and there is less of a benchmark to hold decision-makers accountable.[14] This report argues that a morally sound way of making this kind of decision is based on an extrapolation of Joel Feinberg’s ‘garrison threshold’ thought experiment.[15]

Briefly, the argument holds that funding HCSNs is morally permissible, and other things equal, should be done, when meeting a given HCSN has an otherwise negligible impact on the total budget and imposes negligible opportunity costs on others. When funding that HCSN leads to high opportunity costs, defined as a significantly reduced figure for A (or the average spending per individual projected in a given budget – see subsection 2, ‘Needs that are high-cost’ above), then this form of spending will be problematic.