Welfarism and Extra-Welfarism

Welfarism and extra-welfarism are alternative normative economic frameworks for ranking resource allocations. By normative, we mean economic analysis intended to answer questions such as, What ‘ought’ we to do? or Which resource allocation is the best? or Is policy A preferred to policy B? Normative analysis unavoidably rests on value judgments regarding, for example, as to what constitutes a benefit. Normative economic analysis contrasts with positive economic analysis in that, by answering questions such as, What will be the effect of policy A on the allocation of resources among the members of the society?, it attempts to describe what will happen without making any judgment as to the goodness or desirability of the predicted effects.

Welfarism is one element of the welfare-economic framework that dominates normative analysis in economics. It dictates that the only information relevant for ranking alternative allocations of resources is the utilities attained by the individuals in a society.

Extra-welfarism, in contrast, argues that normative economic analysis should be based on a larger set of information than simply the utilities attained by individuals in the society. Different variants of extra-welfarism emphasize different types of information to either supplant or supplement utility information.

As noted, normative economic analysis is conducted to rank-order policy options. Each policy generates a particular allocation of goods and services within the society and an associated distribution of well-being among the members of the society. The policies considered can be quite narrowly defined and generate relatively circumscribed effects, such as

would be the case when comparing alternative dosages of a drug used to treat a relatively rare, minor ailment, or they can be broadly defined with wide-ranging, profound effects on resource allocation, such as would be the case when comparing alternative systems of health-care finance, which may affect labor markets, income distribution, health-care consumption, and other economic activities. In either case, the term ‘resource allocation’ refers simply to who gets what in society, including health-care goods and services, other government goods and services, private consumption goods and services, and so forth. Hereafter, we use the terms ‘policy’ and the associated resource allocation created by a policy interchangeably.

Welfarism And The Welfare-Economic Framework

The mainstream welfare-economic framework is built on four central tenets: utility-maximization, individual sovereignty, consequentialism, and welfarism.

Utility-maximization refers to the behavioral assumption that individuals choose rationally: given a set of choice options, an individual chooses the most preferred option among them according to the defined notions of consistency. Utility has been interpreted variously in the history of economics and continues to have multiple interpretations. The two dominant interpretations are hedonistic utility and preference-based utility. Hedonistic utility, which derives from classical utilitarianism, equates utility with the pleasure, happiness, or satisfaction that an individual derives from a good or an activity. It is a psychological construct and an individual is assumed to choose and act so as to maximize his utility. In contrast, the preference-based definition of utility eschews a psychological interpretation and defines utility as a function that represents a preference ordering: those goods or activities that an individual chooses are assigned a higher utility value than those that are not chosen. Utility therefore simply represents preferences it makes no assumptions as to the reasons why one thing is preferred to another. Modern microeconomic theory adopts this preference-based interpretation of utility. At times, some have argued that utility is defined only over goods and services, but in modern economics, utility can be defined over goods, services, activities, and virtually any phenomenon on which an individual can express a preference.

Individual sovereignty (sometimes referred to as ‘consumer sovereignty’) is the maxim that individuals are the best judges of their own welfare. Any judgment regarding an individual’s welfare should be based on the person’s own assessment. Individual sovereignty explicitly rejects paternalism – the notion that a third party may know better than the individual what is best for himself or herself. It dictates that individual preferences be respected in the process of evaluation.

Consequentialism holds that any policy must be judged exclusively in terms of the resulting, or consequent, effects. The motivation for or intention behind the policy does not matter; ethical imperatives such as duties, rights, and obligations do not matter. All that matters is the effects that flow from the policy.

As already noted, welfarism holds that the goodness of any resource allocation is to be judged solely on the basis of the utilities attained by the affected individuals. No other aspect of the situation matters. Together, these four tenets require that any policy be judged solely in terms of the resulting utilities achieved by individuals, as assessed by the individuals themselves.

The final ranking also depends on the criterion used to rank alternative policies based on this utility information. Early Neoclassical Welfare Economic Theory assumed (like classical utilitarianism) that utility was cardinally measurable and interpersonally comparable, and defined the best allocation as the one that maximized total utility (i.e., the sum of individual utilities) for the population. Modern welfare economics assumes that utility is ordinally measurable and noninterpersonally comparable, and it replaces the utilitymaximizing criterion with the Pareto criteria: an allocation of resources is judged to be Pareto optimal if it is not possible to reallocate resources so as to increase one person’s utility without decreasing another person’s utility. Although many economists prefer the Pareto criterion to the sum-maximizing criterion because it makes less restrictive assumptions regarding utility and it makes weaker ethical assumptions, the Pareto criterion suffers some important limitations for applied welfare analysis. For a given set of resources, each of the many possible allocations can be Pareto optimal: the Pareto criterion does not necessarily lead to full ranking that identifies a single allocation as best. Besides, as nearly all real-world policy changes hurt at least one individual, organization or group in a society, strict application of the Pareto criterion leads to policy paralysis in which no policy can be judged better than the status quo.

These limitations of the Pareto criterion led to the development of the potential Pareto criterion (also call the hypothetical compensation test), which states that one allocation is preferred to another if the gains to the winners are sufficiently large to enable them to (hypothetically) compensate the losers while are still leaving the winners better off. Crucially, however, compensation does not have to be paid to the losers, so the losers are in fact still worse off. The potential Pareto criterion is the basis for much normative economic policy analysis.

Neither the sum-maximizing criteria of classical utilitarianism nor the Pareto criterion are sensitive to how utility is distributed among the members of a society. Welfare economics has tried to incorporate distributional concerns through the concept of a social welfare function. A social welfare function maintains the welfarist assumption that rankings depend solely on the utilities attained by the members of the society, but includes a measure of the society’s aversion to inequality. Hence, even if the total utility is the same under both the policies, and if the society is averse to inequality the policy under which utility is distributed more equally among individuals would be preferred.

Welfare-economic analysis has been importantly shaped by the two fundamental theorems of welfare economics. The first theorem states that a well-functioning market (where this has a specific meaning) leads to a Pareto optimal allocation of resources. The second theorem states that, given the right initial distribution of income in society, any Pareto optimal allocation (including those judged to be equitable) can be achieved through market allocation. When combined with the claim that questions of distribution are fundamentally political rather than economic, the second theorem has led economists to focus on issues of efficiency, effectively separating the analysis of efficiency from the analysis of distribution. In addition, given that any allocation can, in principle, be achieved through a system of well-functioning markets, market-based allocation serves as the reference standard for judging efficiency. Hence, within the welfare-economic framework government intervention in a market can be justified by equity concerns (e.g., income inequality) or by market failure, a situation in which deficiencies in the market cause market allocation to be inefficient. The latter, however, has dominated economic analyses on the role of government policy. As a corollary, within this framework the objective of any corrective public policy is to achieve the allocation that would have resulted from a well-functioning market.

Empirical Welfare Analysis

Using this welfare-economic framework as a guide to empirical normative analysis presents a number of challenges. The methodology of applied welfare analysis is called cost–benefit analysis. The goal of cost–benefit analysis is to determine whether the adoption of a policy will be more efficient than a specified alternative policy, where efficiency is defined by the potential Pareto criterion. Within cost–benefit analysis, utility is measured using a money metric. Benefit to a member of a society is defined as the amount of money a person is willing to pay for the effect achieved by a policy (e.g., improved health). As both benefits and costs are measured in monetary units, a policy is deemed efficient (relative to the alternative against which it is being compared) if the net benefit (incremental benefits–incremental costs) is positive. The goal of cost–benefit analysis is to mimic for government policies the allocation of resources that would have resulted from a well-functioning market.


At its most general sense, extra-welfarists argue that normative assessment should be based on a wider set of information than solely the utilities attained by individuals. Although utility information may be relevant, it is insufficient and additional, or extra, information should be incorporated into the analysis. Extra-welfarism is not strongly prescriptive as to what this additional information is; indeed, many extrawelfarists argue that the relevant information depends on the context. As extra-welfarism is, in part, a reaction against the dominant welfarist paradigm in economics, it tends to be more eclectic as different extra-welfarist writers have reacted against different elements of welfarism and proposed different ways of addressing welfarism’s limitations. As a result, it is perhaps easier to articulate what extra-welfarism is not than what it is. Still, although extra-welfarism continues to develop within heath economics, it embodies a sufficient set of foundational principles and ideas as to define a coherent normative framework distinct from welfarism (Hurley, 2000; Brouwer et al., 2008).

Extra-welfarist ideas have existed within health economics since long before the term ‘extra-welfarism’ was coined. Much of the work of extra-welfarists in health economics has been to integrate these ideas into a framework with deeper conceptual foundations. The work of Culyer (1990) – who coined the term ‘extra-welfarism’ – has been particularly important in this respect. These extra-welfarist tendencies in health economics arose for a number of reasons:

  • Health economists have traditionally worked closely with noneconomist health professionals, including clinicians, epidemiologists, bioethicists, decision scientists, and others not schooled in welfare economics and who found some of its elements unpalatable. In particular, a primary outcome of health-care interventions is improved health, and notably, ‘lives saved.’ Many noneconomists objected to assigning a monetary value to lives saved. This gave rise to the use of cost-effectiveness analysis rather than cost–benefit analysis in the evaluation of health-care interventions. Cost-effectiveness measures outcomes in natural units (e.g., life-years gained and cases detected) and refrains from assigning a monetary value (or indeed any explicit social value) to the health gains produced by a health intervention.
  • Government health policy commonly rejects the welfare-economic view that the desired allocation of health-care resources is the one that would follow from market forces based on people’s willingness to pay. The explicitly stated objective of many governments is to improve population health based on allocation of health-care resources according to ‘need’, regardless of a person’s ability to pay or willingness to pay. This requires that access to health care be independent of a person’s income or wealth. Even if the society could somehow get the distribution of income ‘correct’ (as envisioned in most of the welfare-economic analysis), it would be very difficult to achieve this objective through market-based allocation of health care using prices. Instead, the objective requires allocation based on nonmarket principles.
  • Health economics has been heavily influenced by the ‘decision-maker’ approach to cost–benefit analysis, which rejects the individualistic, welfarist foundations of traditional cost–benefit analysis. The decision-maker approach instead emphasizes the objectives (and weights) of high-level policy decision makers. This approach has strong affinities with the view within decision sciences that the role of the analyst is to assist a decision-maker in achieving the decision-makers stated goals. Combined with the point above regarding the stated objectives of most of the health policies, this perspective led health economists to develop methods of evaluation in which health – rather than utility – is the primary measure of benefit.
  • The demand for health care derives from the demand for health. This derived demand for health care means that health care is an input into the production of health, which permits normative assessments of health-care consumption using supply-side notions of efficiency, unlike other sectors of the economy in which individuals demand goods for their direct effects on utility. This underlying production relationship enables a third-party analyst to use evidence regarding the impact of a health-care service on health (generated, for instance, by clinical studies) to assess the efficiency of health-care consumption: it cannot be efficient to consume a service known to be ineffective in producing health. This perspective on normative assessment in the health sector reinforces a focus on health as the main outcome of interest.

Extra-welfarism in health has been shaped importantly by the work of the Nobel-Prize winning economist Amartya Sen. Sen broke sharply with welfarism and has developed an alternative framework based on the concepts of human functionings and capabilities (e.g., Sen, 1999). Sen argued that welfarism was an insufficient basis for normative economic analysis because utility focuses too narrowly on people’s mental and emotional reactions to circumstances and not enough on what they can achieve with their material and other resources. Welfarism, for instance, suffers from the problem of adaptation. A person born into poverty, who adjusts his/her life expectations to conform to his/her limited life possibilities may, as measured by utility, be better off than a well-off person given every advantage and opportunity in life but whose expectations exceeded what his/her was able to realize and who therefore ends up disappointed. Welfarism also either ignores nonutility aspects of a situation, such as whether basic human rights are being violated, or, to the extent that they are captured by the analysis, they enter only through the metric of utility. In the end, Sen argues that neither utility – whether defined in traditional hedonistic terms as desire-fulfillment or pleasure or in modern terms as preference satisfaction – nor a person’s material possessions can serve as a proper basis for assessing alternative social policies.

Sen argued that, instead, evaluation should focus on functionings and capabilities. The central ideas of his approach can be briefly summarized as follows. He begins with commodities – goods and services – that can be under a person’s command and which have characteristics: a bicycle is a commodity whose salient characteristic is that it can provide transportation; health care is a commodity whose salient characteristic is the potential to improve a person’s health. Commodities and their characteristics do not depend on features of the person who possess them. A functioning is what a person succeeds in doing with the commodities at his or her command; it is what the person manages to do or to be. Bicycling is a functioning. The functionings possible for a person depend on the commodities at the command of a person and features of the person and their environment. A quadrapalegic with a standard bicycle cannot achieve the functioning of bicycling. Functionings range from the trivial (enjoy ice cream) to the profound (form meaningful relationships with others). The set of all possible functionings for a person, which Sen calls their capabilities, depends on their feasible set of commodities available and, given their personal characteristics and social institutions, the feasible set of means available to them to transform commodities into functionings. Alternative states of the world should be evaluated based on the people’s capabilities in that world; that is, the functionings open to them. Sen emphasizes a nonconsequentialist perspective that the evaluation be based not on the functionings a person actually achieves, as these may depend on personal choices not of concern to normative assessment, but rather on what was possible for the person – the life prospects that were available to them given the social arrangements and their place in society. Sen’s framework is commonly referred to as the ‘capabilities’ approach within extra-welfarism.

As noted, Culyer was one of the earliest health economists to recognize the potential for adapting Sen’s ideas to the health sector. The concept of need, for example, is widely used by health professionals, health scientists, and, historically by health economists, though it does not fit easily into the individualistic, welfarist framework of traditional welfare economics in which people are characterized only by utility; and mainstream economics has in fact been hostile to the concept, seeing it largely as an attempt by some to gain privileged status for their preferences. Yet, the concept of need for health care fits easily into the capabilities approach wherein health care is a commodity that is essential for (and needed by) an ill-person to become healthy and thereby be able to realize many different types of functionings.

Within a broad welfarist framework, functionings and capabilities could, in principle, be valued using a metric of utility. Indeed, among welfarists who believe that preferences extend over any aspect of life, there is no concern of extrawelfarism that could not, in principle, be captured through preferences. The point of extra-welfarism, however, is that preferences are the wrong metric by which to value such outcomes. The extra-welfarist criticism of utility is not that it can only be applied to a limited range of outcomes; rather the criticism is that it provides a limited valuation of all those outcomes to which it is applied.

Though Sen’s capabilities approach is a particularly influential source of extra-welfarist thought in health economics, it is only one variant of extra-welfarism. Brouwer et al. (2008) represents perhaps the most comprehensive attempt to articulate the scope and definition of extra-welfarism in the health sector, and in particular ways in which extra-welfarism differs from the welfarist normative framework. They emphasize four ways in which extra-welfarism differs from the welfarist framework: (1) extra-welfarism permits the use of outcomes other than utility; (2) extra-welfarism permits the use of sources of valuation other than the affected individuals; (3) extra-welfarism permits the weighting of outcomes (whether utility or other) according to the principles that need not be preference-based; and (4) extra-welfarism permits interpersonal comparisons of well-being in a variety of dimensions, thus enabling movement beyond Paretian economics. Below we summarize some of the main elements of extrawelfarism in the health sector, discussing these and other points related to extra-welfarism.

Extra-welfarism emphasizes the use of nonutility information in normative economic analysis. Conceptually, extrawelfarism does not reject a possible role for utility information – it simply argues that utility information alone is insufficient. The predominant nonutility outcome of interest within health economics is health status. The focus on health derives from a number of factors – the influence of the decision-maker approach and the fact that health policy-makers emphasize health as the outcome of interest; the fact that health is observable and, within certain bounds, interpersonally comparable; and the fact that health can be integrated into Sen’s capability approach. Extra-welfarism can also accommodate other types of nonutility information such as rights, dignity, and other ethical concerns. Indeed, it places no restriction on the type of nonutility information that can, in principle, be included in an analysis. Some have argued, however, that a shortcoming of how extra-welfarism in the health sector has developed in practice is a near-exclusive focus on health as the outcome of interest in evaluation. To the extent that this is true, extra-welfarism has simply substituted one restrictive definition of the outcome space – utility – with another – health.

Health status can be measured in numerous ways depending on the specific question under consideration, and the emphasis on health as the outcome of concern has spurred economists to develop health measures with desirable properties. The dominant health-related outcome measure for the evaluation of health programs, services, and technologies is the quality-adjusted life-year (QALY). The QALY illustrates well that extra-welfarists do not completely reject the use of utility information and utility-based methods, but that they often use utility information in ways not consistent with welfare economics. The QALY is designed to capture the effect of a health intervention on both the quantity and quality of life-years. It is defined simply as the quality-weighted sum of life-years: ∑iQiYi, where Yi is the number of years spent in health state i and Qi is a quality weight for health state i that takes on a value between 0 and 1, where the weight for full health is 1.0 and the weight for being dead is 0. The quality weights used to construct a QALY measure are often constructed using methods derived from utility theory; that is, the quality weights underlying the QALY are elicited using the methods drawn from utility theory.

This fact has led to considerable confusion and debate regarding the interpretation of a QALY, and this debate reflects important issues in the debate on welfarism and extrawelfarism. The confusion stems from the fact that a QALY can be interpreted differently depending on the auxiliary assumptions one is willing to make. Under certain assumptions regarding the nature of people’s utility functions, a QALY can be interpreted as a measure of individual utility or preferences. These assumptions are restrictive, however, and do not accord well with evidence on the actual structure of people’s utility functions. Hence, welfarists – for whom utility is the outcome of interest – have criticized the QALY because it does not represent well people’s preferences over health states. Extrawelfarists, however, have argued this criticism is misguided: although preference information is used in the construction of a QALY, the QALY is not intended as a measure of preferences but instead it is a measure of health, or perhaps as more commonly stated, health-related quality of life. Hence, the fact that QALYs do not accurately map preferences over health states is not a weakness. Finally, some extra-welfarists have explored the potential for interpreting the QALY as a measure of the value of an individual’s capability set. Doing so requires making the rather strong assumption that the QALY can represent all of an individual’s well-being rather than making the more traditional assumption that QALY represent only one component of an individual’s well-being (the health-related component). Regardless of the specific interpretation one may adopt, the point is that the QALY illustrates one way extra-welfarists integrate utility information in unconventional ways, and, in addition, that one cannot understand certain debates in health economics without being attuned to subtle distinctions between welfarist and extra-welfarist approaches.

Unlike welfare economics, extra-welfarism permits sources of valuation other than the individuals affected by a policy. Extra-welfarism is comfortable with certain types of paternalism. Acting as agents for the members of the society, decision-makers are not assumed to act in the way they think individuals would act, but rather as they think individuals ought to act. The potential sources of value are many. The valuation could be based on empirical information drawn from the general public, but it may include information other than preferences, it may even include preferences but aggregated in ways inconsistent with individualistic welfareeconomic analysis, it may gather information in ways inconsistent with preference measurement such as deliberative process and consultations of various types, or it could be responses to questions quite different from preference information. Valuation could be reasoned argument that has no basis in empirical measurement.

Extra-welfarism and welfarism differ importantly in a number of ways pertaining to the aggregation of costs and benefits across individuals. Extra-welfarism is more accommodating of interpersonal comparisons of well-being. Sen has argued in particular for frameworks that admit more possibilities than the two extremes of complete rejection of interpersonal comparability (as in modern welfare economics) and the assumption of full interpersonal comparability (as in classical utilitarianism). He emphasizes the scope for limited degrees of comparability that allow for at least partial orderings of the different possible states of the world. Such partial orderings are all that the analysis requires in many situations. The shift to nonutility outcomes such as health also provides greater scope for assuming degrees of interpersonal comparability. Again, even if one recognizes that it is not possible to conclude which of the two people have greater health in all situations, it often is possible to make relatively uncontested judgments regarding relative levels of health across individuals. The same is true for pivotal concepts such as the need for health care.

Extra-welfarism provides greater scope than welfarism for integrating equity concern. Both the Pareto criterion of analytical welfare economics and the net-benefit criterion on cost–benefit analysis are completely insensitive to distributional concerns. Extra-welfarism, however, allows for the integration of nonunitary equity weights that reflect the fact that society may value health gains among certain groups in society more than others. The equity weights constitute a type of nonutility information that may be derived from a number of sources. Some examples include differential weighting by age (greater weight to the young), by one’s role in society (e.g., greater weight to caregivers, especially those caring for children), by the extent to which one is responsible for one’s ill-health (lesser weight to those who are at least in part responsible for their illhealth), or by baseline health status (greater weight to those with more severe health conditions). The use of such weights also implies breaking with the welfare-economic assumption of anonymity, which holds that no characteristic of an individual matters for evaluation except their value with respect to the outcome of interest.

Extra-welfarism is also more accommodating of equity concerns beyond distributional equity, such as procedural equity. Procedural equity emphasizes fairness in the process by which resources are allocated, respect for the rights of individuals, and related matters that may not bear on the actual final distribution of resources. Although welfarist approaches have developed the idea of ‘process utility’ to capture some of these types of concerns, such an approach values them only to the extent that they affect utility, whereas many argue that aspects of procedural equity lie outside the logic of a consequential calculus.

Empirical evidence consistently demonstrates that among the general public concerns regarding equity weigh heavily in judgments on the allocation of health-care resources. People are willing to reduce the total amount of health produced in order to achieve a more equal distribution of health among the members of the society. Moreover, although extra-welfarism can readily accommodate such attitudes, some analysts have been critical of the fact that, as a pragmatic matter, a good deal of extra-welfarist evaluation in health has adopted a health-maximization criterion: that is, the best option is that which maximizes the amount of health produced, in a manner analogous to the classical utilitarian approach within welfarist economics. As with the concern regarding a near-exclusive focus on health in the outcomes space, the issue is not what extrawelfarism embodies in principle, but rather what happens on the ground in actual practice.

Extra-Welfarism In Empirical Normative Analysis

There is no single ‘extra-welfarist’ approach to empirical normative analysis. In the area of health technology assessment, extra-welfarist ideas are incorporated into cost-effectiveness and cost–utility analyses. In the normative analysis of broader health system policies extra-welfarist ideas manifest themselves in a focus on health as the outcome of interest rather than willingness to pay as measured by the demand curve for health-care services. Two brief examples below illustrate how selected aspects of extra-welfarism translate into empirical normative analysis, and especially how these aspects differ from the welfare-economic approach. The first illustrates, in the context of health technology assessment, how extrawelfarism draws on sources of valuation distinct from the individualistic approach of welfarism. The second illustrates how a focus on health rather than willingness to pay can change the normative conclusions regarding the optimal policy to combat insurance-induced moral hazard in the health services market.

Sources Of Valuation In Extra-Welfarism

Consider the economic evaluation of a new health-care technology the only effect of which is to reduce the probability of dying within a given period. Under the welfare-economic framework, these additional life-years would be valued according to the amount each affected individual was willing to pay to reduce the chance of death, so some (e.g., the wealthy) would be willing to pay more than others (e.g., the poor). The new technology would be deemed efficient if the net benefit was positive. If, instead, a cost-effectiveness analysis (CEA) is conducted, the analyst refrains from placing a value on the additional life-years produced by the new technology. The efficiency of the intervention is expressed using the incremental cost-effectiveness ratio (ICER), which indicates the additional cost incurred per life-year gained. Although CEA allows the analyst to avoid placing a monetary value on a lifeyear gained, if the results of the CEA are to be used as the basis for an adoption decision, the decision-maker must decide whether the additional cost per life-year is above or below the amount the society is willing to pay-that is, are the extra benefits worth the extra costs. The extra-welfarist CEA bases the decision on the social value of the health benefit (e.g., life-year gained), which may be set by the decision-maker, by community consultation, or by some other process. Although, as noted above in the discussion of equity weights, the social value of the health gain may differ across the groups of individuals in a society, the social value does not depend on the individuals’ own valuation of the health gain.

The Evaluation Of Policies To Combat Moral Hazard

The analytic importance of using health rather than willingness to pay as a measure of social value can be illustrated with an example adapted from Reinhardt (1998). Consider two families with identical preferences and full insurance for physician visits, the Chens, who are wealthy, and the Smiths, who are poor. Each family has just had a baby. Baby Chen is healthy but Baby Smith is sickly. With full insurance, the Smiths demand nine physician visits during the period and the Chens demand six. To combat moral hazard, the insurer seeks to reduce total visits by five. To do so, it imposes a copayment equal to $15 per visit. Under cost-sharing, both families demand five visits per period, for a reduction of five visits overall (from 15 to 10). Within the standard welfareeconomic analysis, the reductions of four and one, respectively, for the Smiths and the Chens are the optimal way to reduce total visits by five because it imposes the lowest welfare loss as measured by willingness to pay. However, because Baby Smith is sickly while Baby Chen is healthy, the marginal health gain of a physician visit is always greater for the Smiths than for the Chens. If we measure benefit by the health effects rather than willingness to pay, then to reduce overall visits by five in a way that minimizes health effects, Baby Chen’s visits should be reduced from six to two whereas the Baby Smith’s are reduced from nine to eight. Such a reduction cannot be achieved by a single user-charge policy. Consequently, this extra-welfarist analysis indicates that moral hazard should be combated by an alternative policy that can selectively reduce the visits that have the least health gain.

Current Issues In Extra-Welfarism

Extra-welfarism continues to develop in health economics, and we highlight just a few areas of the ongoing development. One active area of work seeks to translate Sen’s capability approach into a practical, empirical method. Capability sets, as noted above, include not only the functionings a person actually achieves, but all the possible functionings a person could have achieved. Hence, a major challenge is how to measure these potential functionings that were not achieved by an individual, and in particular to distinguish those that were available but not chosen from those that were not available due to some social or economic barrier in society. A second area focuses on the relationship between extrawelfarism and welfarism. Brouwer et al. (2008) argue that welfarism is simply a special case of extra-welfarism. Much of the literature has tended to focus on the two polar cases – welfarism which limits the outcome space to utility – and extra-welfarist approaches that ignores utility in the outcome space (even if it has used elements of utility in constructing indices such as the QALY). Less well developed is systematic thinking regarding the rich middle ground that admits both utility and nonutility outcomes, and in particular, principles that can guide the role of each in an analysis. The nature of the outcome information included, for instance, might depend on the level of aggregation at which a decision is being made. At high levels of aggregation, such as a central ministry allocating resources to regional authorities responsible for their respective populations, the concepts of need and health likely dominate considerations of possible differences in preferences. However, in the design of programs at the local level, the role for preferences may be larger as the program has more direct dealings with those receiving services. Similarly, it is natural to ascribe a larger role for preferences (especially over process aspects of care delivery) in those contexts in which differences in health and other outcomes are small. Finally, there is ongoing tension between the emphasis in extra-welfarism on diversity, flexibility, and adaptability of methods and standardization of methods both to increase comparability across studies and to improve quality in those aspects for which there may be generally recognized ‘better’ methods (e.g., of certain aspects of measurement). This tension is an inherent feature of extra-welfarism. Only by attending carefully to its demands can extra-welfarism avoid the pitfalls of, on one hand, lacking sufficient shared, core principles required for a coherent, distinct vision and, on the other, a type of standardized rigidity against which extrawelfarism is in part a reaction.


  1. Brouwer, B. F., Culyer, A. J., van Exel, N. J. and Rutten, F. (2008). Welfarism vs. extra-welfarism. Journal of Health Economics 27(2), 325–338.
  2. Culyer, A. J. (1990). Commodities, characteristics of commodities, characteristics of people, utilities, and the quality of life. In Baldwin, S., Godfrey, C. and Propper, C. (eds.) Quality of life: Perspectives and policies, pp. 9–27. London: Routledge.
  3. Hurley, J. (2000). An overview of the normative economics of the health sector. In Culyer, A. J. and Newhouse, J. P. (eds.) Handbook of health economics, pp. 55–118. Amsterdam: Elsevier Science B.V.
  4. Reinhardt, U. (1998). Abstracting from distributional effects, this policy is efficient. In Barer, M., Getzen, T. and Stoddart, G. (eds.) Health, health care and health economics: Perspectives on distribution, pp. 1–53. Toronto: John Wiley and Sons.
  5. Sen, A. (1999). Commodities and capabilities. Oxford: Oxford University Press.
  6. Anand, P. and Hees, M. (2006). Capabilities and achievements: An empirical study. Journal of Socio-Economics 35, 268–284.
  7. Boadway, R. and Bruce, N. (1984). Welfare economics. Oxford: Basil Blackwell.
  8. Coast, J., Smith, R. and Lorgelly, P. (2008). Welfarism, extra-welfarism and capability: The spread of ideas in health economics. Social Science and Medicine 67, 1190–1198.
  9. Cookson, R. (2005). QALYs and the capability approach. Health Economics 14, 817–829.
  10. Culyer, A. J. (1989). The normative economics of health care finance and provision. Oxford Review of Economic Policy 5(1), 34–58.
  11. Hurley, J. (1998). Welfarism, extra-welfarism and evaluative economic analysis in the health sector. In Barer, M., Getzen, T. and Stoddart, G. (eds.) Health, health care and health economics: Perspectives on distribution, pp. 373–396. Toronto: John Wiley and Sons.
  12. Sen, A. (1979). Personal utilities and public judgments: Or what’s wrong with welfare economics. The Economic Journal 89(355), 537–558.
Theory of System Level Efficiency in Health Care