Welfare Economics

Topic 10. The theoretical foundations of cost-effectiveness analysis*

Key reading: see reading list + references in these slides.

Today’s lecture:

  • Critiques of welfarism
  • Alternative paradigms:
  • Non-welfarism (Sen)
  • Extra-welfarism (Culyer)
  • Communitarianism (Mooney)
  • Critiques of extra welfarism
  • The consistency of SWO between welfarist & extra welfarist approaches.

* CEA is interpreted here as including both CUA and CEA – i.e. any non- CBA (and thus non-welfarist) approach to economic evaluation.

  1. The welfarist (i.e. welfare economics) approach to SWO: some characteristics
  • Individualism: a ranking of social states is achieved by the effect of each state on individuals’ utilities.
  • Utility maximisation: by each individual/household.

→ precludes ‘counterpreferential’ choices i.e. individuals might contribute to some common good which might result in lowering their own utility. [Mooney 2002]

  • Consequentialism: processes are not valued in themselves, but are instrumental towards some outcome. Sources of utility restricted to consumption of goods/services.
  1. Welfarism: Sen’s critique, and ‘non-welfarism’
  • Individuals are different in terms of their ability to convert commodities into wellbeing. Even if goods were allocated equally across individuals, this would not necessarily lead to equal utility.
  • Characteristics of people:
  • Functionings: ‘what he or she manages to be’
  • Capabilities: ‘freedoms; opportunities to choose’
  • Sen on utility:
  • Physical condition neglect/ valuation neglect: ‘realistic desires’.
  • Measurement taken too far (“ ‘waiting for toto’ may not be a cunning strategy in a practical exercise…”)
  1. Extra welfarism (EW)
  • Culyer: the objective function to be maximised shifts from utility to health.
  • Maximisation of health, subject to the budget constraint, the basis upon which states are ordered.

→ dispute over what ‘extra-‘ as opposed to ‘non-welfarism’ actually means. [see Tsuchiya and Williams 2001].

→ both involve a rejection of social choices based on individual utilities; EW in theory is consistent with a wide range of quite dissimilar approaches.

e.g. why choose ‘health’ as the maximand?

→ based on an ‘exernal judgement’.

e.g., that health is agreed to be the principal output of health care services.

  • In principle, EW admits a wide range non-utility arguments other than health (e.g. bearing on the characteristics of people) into the social ranking.
  • Its practical interpretation in CEA/CUA has evolved in a rather specific way.
  1. Critiques of extra welfarism

[see Birch and Donaldson 2002; Mooney 2002; Tsuchiya and Williams 2001]

  • The assumption that health is the only argument in the objective function is only one of many possible assumptions that might be defined as EW.
  • Mooney [2002]: contends that both W and EW share common problems

(a) individualism: individuals’ preferences determine weights

(b) consequentialism (narrowly so under EW; e.g. excludes ‘process of care’ utility)

(c) cannot accommodate differences between groups e.g. in valuation, or ‘constructs’, of health.

Mooney’s point (a) could be debated. Tsuchiya and Williams (2001) argue that:

  • The values of individuals who are gaining/losing QALYs is not the key feature of EW.
  • Indeed individual’s values (of those affected) are generally rejected in favour of population values.
  • The values used in QALYs are population-based and applied uniformly across all affected people regardless of their individual valuations.

[Tsuchiya & Williams 2001]

Question for discussion:

What does efficiency mean under the non-welfarist and extra-welfarist paradigms?

  • Not clear where Sen’s critique leaves us vis a vis rankings of states
  • Is ‘efficiency’ (max. QALYs) under EW consistent with Pareto efficiency?

In one sense (e.g. study question 13) it is.

But…

  • ‘better’/’worse off’ from whose perspective?
  • Does Pareto require QALYs to be based on the valuation of states by the individuals concerned?
  1. Communitarianism [Mooney 2002]

Communities, as well as individuals, are relevant – and community values may be different from the aggregation of individual values.

Is ‘society’ more than a collection of individuals?

→ Suggests eliciting community preferences and allocating resources in manner that respects differences in valuations between groups.

  • But what/who defines what the relevant ‘communities’ are?
  • The focus is on equity; but what does ‘efficiency’ mean in this context? (see related debate in HE between Sculpher & Gafni and Robinson & Parkin on sub-group analysis)
  1. Welfarism and extra welfarism: a bridge too far?
  • Maximisation of QALYs subject to a health budget reveals an implicit valuation of a QALY

→ if an option ‘accepted’ its ICER is the minimum value placed on a QALY by decision makers, given a fixed budget.

Might this differ from society’s willingness to pay to gain a QALY?

  • If a constant WTP for QALYs existed, this would enable the results from CUA to be re-expressed in CBA.
  • But requires very restrictive conditions on utility functions for equivalence to be established.

Recommended reading on the relationship between the theoretic foundations of CBA & CEA

Dolan,P., Edlin, E. (2002) Is it really possible to build a bridge between CBA and CEA? JHE 21:827-843

Birch S., Donaldson C. (2002) Valuing the benefits and costs of health care programmes: where’s the extra in extra welfarism? Soc Sci Med 56(5):1121-1133.

Tsuchiya, A., Williams A. (2001) Welfare economics and economic evaluation. Ch 2 in: Drummond and McGuire.