«Edited by ANNE MASON Research Fellow, Centre for Health Economics University of York and ADRIAN TOWSE Director, Ofﬁce of Health Economics Radcliffe ...»
Williams A (1986) Health economics: the cheerful face of the dismal science? In Williams AH (ed.) Health and Economics: Proceedings of Section F (Economics) of
the British Association for the Advancement of Science, Bristol, 1986. Houndmills:
Macmillan, pp. 1–11.
Williams A. (1988a) Priority setting in public and private health care: a guide through the ideological jungle, Journal of Health Economics, 7, 173–183.
Williams A (1988b) Ethics and efﬁciency in the provision of health care. In: Bell JM and Mendus S (eds) Philosophy and Medical Welfare. (Royal Institute of Philosophy Lecture Series 23, Supplement to Philosophy). Cambridge: Cambridge University Press, pp. 111–26.
Williams A (1991a) Some methodological issues in the use of cost-beneﬁt analysis in health care. Osaka Economic Papers 40: 63–9.
Williams A (1991b) Is the QALY a technical solution to a political problem? Of course not! International Journal of Health Services 21: 365–9.
Williams A (1992) Cost-effectiveness analysis: is it ethical? Journal of Medical Ethics 18: 7–11.
Williams A (1993a) Cost-beneﬁt analysis: applied welfare economics or general decision aid. In: Williams A and Giardina E (eds) Efﬁciency in the Public Sector.
Aldershot: Edward Elgar, pp. 65–79.
Williams A (1993b) Some methodological issues in the use of cost-beneﬁt analysis in health care. In El-Agraa AM (ed.) Public and International Economics. New York: St Martin’s Press, pp. 33–4.
Williams A (1994) Economics, society and health care ethics. In: Gillon R (ed.) Principles of Health Care Ethics. Chichester: John Wiley and Sons, pp. 829–42.
Williams A (1997) Intergenerational equity: an exploration of the ‘fair innings’ argument. Health Economics 6: 117–32.
Williams A (1998a) QALYs and ethics: a health economist’s perspective. Social Science and Medicine 43: 1795–804.
Williams A (1998b) If we are going to get a fair innings, someone will need to keep the
74 THE IDEAS AND INFLUENCE OF ALAN WILLIAMS
score! In: Barer M and Stoddart G (eds) Health, Health Care and Health Economics:
perspectives on distribution. Chichester: John Wiley and Son, pp. 319–30.
Discussion of Anthony Culyer’s paper:
‘Resource allocation in health care: Alan Williams’ decision maker, the authority, and Pareto’... Adrian Towse Tony’s paper is intended to take us to a better understanding of the relevance of welfarist and extra-welfarist approaches to rationing the provision of health care. I follow Tony in also characterising these approaches as the Paretian and decision-maker approaches respectively.
In making the statement ‘What is the question?’ – THAT is the answer!’, Tony suggests that the choice of approach to address a rationing issue depends on what question we are trying to answer (and from what perspective). I agree and return to this later.
My starting point is that Tony (like most extra-welfarists) is not opposed in principle to using markets or indeed to welfarist analysis in general. He argues, rather, that an extra-welfarist approach is more comprehensive because preferences and utility are considered alongside other factors that impact on well-being, and distributional issues can also be addressed. This comprehensiveness is particularly relevant to health care, because health is a ‘different’ element of well-being. Hence he prefers to address healthcare rationing issues using an extra-welfarist approach.
He has three main criticisms of welfarism as an approach to resource
allocation in health:
1 the focus on individual utility derived from preferences over goods and services gives a very narrow view of what matters to people. In health care it is ability to function and psychic state (which contribute to health status) and the length of life that matter. The welfarist toolkit is not helpful in this context 2 the application of willingness to pay techniques – arguably the only part of
7576 THE IDEAS AND INFLUENCE OF ALAN WILLIAMS
the welfarist toolkit that is relevant to obtaining an understanding as to the values individuals put on health or other outcomes provided by the delivery of health care – is of limited value. This is because preferences have to be backed by income to count, and because the technique cannot value ‘nonpreference elements’ such as the rule of rescue. The requirement for income to enter the equation raises major equity issues in using this approach in health care 3 the use in the standard welfarist approach of a requirement for Paretian improvement (or indeed a Kaldor–Hicks improvement) with no interpersonal comparison of utility, or even a requirement for gainers to compensate losers, makes it very restrictive. Interpersonal comparisons are essential for decision making about resource allocation in health care.
By contrast, extra-welfarism is pragmatic in several important respects:
1 the economist can focus on the objectives of the decision maker. There is no mystery as to where the objective function comes from. It is imposed. It is not the task of the economist to derive it from individual preferences (or indeed to fail as Arrow’s Impossibility Theorem tells us) 2 the decision maker can have all sorts of interesting things in their objective function – they don’t have to be based on, or limited to, individual utilities derived from preferences over goods and services. This, in principle, enables extra-welfarism to provide a more comprehensive view of well-being 3 economists can ﬁnd innovative (and eclectic) ways of measuring the impact of interventions on the objective function, enabling the decision maker to understand whether (say) action A or action B will better help achieve their objective given a resource constraint 4 decision making can be informed by results of deliberative mechanisms, i.e. individuals discussing issues and collectively reaching a view as to their preferences between social outcomes (e.g. which groups of patients should get access to treatment).
This contrast gives rise to a number of questions which I consider brieﬂy in
the rest of my paper:
1 which is the more useful ‘general theory’ of resource allocation?
2 is health special?
3 what is the role for preference elicitation in extra-welfarism?
4 is extra-welfarism as pragmatic and eclectic as implied – and indeed welfarism as incomplete?
5 does welfarism have nothing to offer on distributional matters?
6 is there a role for welfarism in healthcare resource allocation?
DISCUSSION OF ‘RESOURCE ALLOCATION IN HEALTH CARE’ 77WHICH IS THE ‘GENERAL THEORY?’ In step (f), Tony sets out the limitations of the welfarist reliance on preferences over goods and services. To paraphrase, humankind cannot live by preferences alone – to focus on goods and services is ‘commodity fetishism’. It is helpful to draw on Sen’s framework for thinking about well-being (or, in Sen’s terms, the ‘standard of living’) (Sen, 1987a, 1987b). This is summarised in Figure 7.1.
FIGURE 7.1 Utilities, functionings, capabilities and their sources.
Adapted from Muellbauer J (1987) Professor Sen on the Standard of Living. In: Hawthorn G (ed.) The Standard of Living. Cambridge: Cambridge University Press. Used and adapted with permission of Cambridge University Press.
What really determines peoples’ well-being is their capabilities, functioning and psychic state. Goods and services play an important part of this. But so do many other things including their physical environment and their social and family relationships. The extra-welfarist framework is helpful in addressing public policy issues, including the measurement of ‘national income’ and resource allocation in public services such as health care. But it is not clear that its all-embracing nature offers any conceptual or practical advantages in analysing markets for most goods and services. As Tony recognises, welfarists might argue that theirs is the general theory and that the utility framework can deal with these other factors – for example, individuals derive utility from the environment (indeed the welfarist approach of contingent valuation has played a major role in the development of environmental economics to support policy making (e.g. Arrow et al., 1993)). But when we apply this to (say) family relationships Tony is right to suggest that we are better off looking at the direct beneﬁts that people derive. Moreover, it remains unclear whether welfarism can address equity issues (see below). Neither welfarism nor extra-welfarism can therefore be seen as an approach that is universally helpful. It is clear, however, that for most goods and services the welfarist approach sufﬁces and provides powerful analytical tools. For some key social goods (I would include health, education and justice) we are more interested in the impact on people’s functioning and well-being. Society prefers to make these goods directly accessible rather than provide income and let individuals decide for
78 THE IDEAS AND INFLUENCE OF ALAN WILLIAMSthemselves whether to consume them. Here extra-welfarism is more helpful.
Many ‘non-market’ activities (including social and personal relationships) may also ﬁt better within this approach.
IS HEALTH SPECIAL?
Health is clearly special in two senses:
● there is a lot of uncertainty in the market as well as market failure on both the demand and supply sides of the market (e.g. imperfect agency, information asymmetry and suppliers with market power). There is also strong societal concern about the impact of income inequalities on access to health care. None of these issues is unique to health care, but the combination is unusual (indeed health care may be one of the few markets where all of the standard concerns about the ability of a market to allocate efﬁciently may apply) and has led to substantial government intervention in the regulation, funding and supply of health care ● health is fundamental to human well-being and so access to healthcare is an essential good. It ﬁts the Sen model. We are interested in physical functioning and psychic state. And we want to provide access to services that will improve physical and mental health – not to provide people with money with which they can choose whether to buy health care or other goods and services.
Welfarism helps us understand and analyse the market failure in health care but is less useful than extra-welfarism as a framework for resource allocation, i.e. in providing tools to address concerns about access and about the impact of health care on health status.
WHAT IS THE ROLE FOR PREFERENCE ELICITATION IN EXTRAWELFARISM?
In health care, preferences are expressed when people choose insurance policies in both private markets and social insurance systems with competing insurers; within the tax-based NHS by patients and carers when making individual treatment decisions jointly with doctors; and by choice of provider, either when choosing a GP practice or hospital (under Patient Choice). These preferences may be shaped by supply-side issues (e.g. a patient may opt for a hospital with a shorter waiting list) and therefore help determine resource allocation issues. These are not preferences that usually interest an extrawelfarist. Their concern is primarily around health gain.
The interesting issue is therefore whether valuations (of impact on individuals) in public policy making are:
● postulated by the decision maker – the decision-making approach. This can be benign dictatorship, although, following Sugden (see Chapter 2),
DISCUSSION OF ‘RESOURCE ALLOCATION IN HEALTH CARE’ 79it makes more sense to think of decision makers as accountable to an electorate inferred from market behaviour (transactions) or attempts to elicit ● preferences (through experiments). This is the welfarist or Paretian approach.
There is nothing in the extra-welfarist approach to stop the decision maker being informed by market behaviour or attempts to elicit preferences. Indeed one might hope that in allocating healthcare resources the decision maker wants to understand the preferences of the individuals on the menu of competing combinations of services the decision maker has to choose between.
However, extra-welfarists seem to be ambiguous about the role of preference elicitation.
On the one hand they:
● strongly favour the use of health status instruments to establish the physical functioning and mental state of individuals, i.e. what can they do/feel (although we can note Sen’s opposition to self-reported morbidity because – like income-based preference measures – it can be distorted by poverty (Sen, 2002)) ● support the valuation of those health status measures using individual preference-elicitation instruments grounded in decision theory (irrespective of whether patients or the public are being asked to value a health state, they are being asked to answer how it impacts/would impact on them).
On the other hand, they:
● are opposed to the use of willingness-to-pay techniques, primarily because it introduces the bias of an income constraint, but also because it moves away from measuring health towards measuring utility. Many extra-welfarists (but not all and Tony may be an exception here) have particular problems with combining non-health outcomes (such as the impact on patients of reduced waiting times – over and above any health gain from being treated earlier – an improved hospital environment, travel times, and the quality of hospital food) with health outcomes.
This is because money valuations of these other elements are needed, which are usually obtained by the use of willingness-to-pay techniques.