Grupos de Investigación

Social Choice and the Capability Approach

  • Flávio Comim (Universidade Federal do Rio Grande do Sul (UFRGS) y Centre of Development Studies, University of Cambridge, Reino Unido)
Amartya Sen does not systematically appeal to the Capability Approach in his empirical works. From “Poverty and Famines” (1981) to “An Uncertain Glory” (2013) (with Jean Drèze) , it is difficult to find explicit references to a single approach when Sen examines issues such as poverty, health, education and inequality. When it comes to the “Capability Approach” (CA) the most certain thing is that it refers to the choice of informational spaces in assessing social possibilities. This suggests that the CA is only part of a broader framework of analysis that can be found elsewhere.
Indeed, in several opportunities  Sen (e.g. 2002, 2009) has manifestly expressed that his overarching framework of analysis is Social Choice Theory (SCT). The CA is strategically key for Sen’s SCT in broadening the informational basis used in social evaluations  but it does not seem to provide the same range of tools and analytical structure that comes with SCT. It is indeed remarkable how the links between the CA and SCT have not been further explored by the CA secondary literature.
In one of the very few notable exceptions in this literature, Qizilbash (2007) examines the links between the CA and SCT trying to explore their differences and complementarities. He doesn’t seem to disagree with the above remark that the relevance of the CA is mostly for providing broader informational spaces for evaluation (that he calls a ‘thin view’). However, he tends to see SCT as a field of application of the CA based on public reasoning (that he calls a ‘thick view’). This in itself is not a problem but it raises the question about why does everyone continue to call the CA ‘an approach’, if its field of application is given by a different approach? Indeed, essential items for any evaluative approach, such as selection of weights, seem to be within Sen’s thick view, as argued by Qizilbash (2007: 176). As much as his interpretation provides an articulated and consistent view of Sen’s CA and SCT it leads to an overlap of approaches when for instance he concludes that (2007: 185), “Sen’s thick view [he referred to it earlier as an approach] thus provides a framework [an approach?] for applying the capability approach…”. One could try to scrutinize the conceptual differences between ‘view’, ‘perspective’, ‘approach’ and ‘framework’ but they are bound to have little practical relevance.
Sen (2009: 232) clarifies that the CA “does not, on its own, propose any specific formula about how that information [on capabilities] may be used” and in addition (p.232-233) that “it does not lay down any blueprint for how to deal with conflicts between, say, aggregative and distributive considerations”. It is clear that given Sen’s pluralism, he wishes to avoid imposing specific recommendations about how societies should establish their priorities and define their social policies. But at the same time, it seems that the core of an evaluative exercise is not addressed by the CA, what raises again the same question about to what extent shall we call it ‘an approach’. 
Within the context of SCT, the broadening of informational spaces (for instance, by using capability spaces) is important as a possible way for circumventing Arrow’s (1951) impossibility theorem. The problem goes back to Lionel Robbins’ critique of interpersonal comparisons of utility and how Pareto optimality has been used to bridge this gap within contemporary welfare economics. As Sen (1996: 55) has put it, “the most likely route of escaping the Arrow dilemma in making social welfare judgments lies not in any marginal modification of one of the Arrow axioms, but in the general direction of enriching the informational input into the analysis”. In particular, by using ‘capabilities’ that are objective (in contrast to subjective utilities) Sen can assemble different types of interpersonally comparable information. 
However, his argument is not about capabilities per se but about pluralism given that the mechanical use of a single formula, even if based on capabilities, would be similarly inadequate. 
But what could be said of SCT? Most importantly, what are the main features of Sen’s SCT that would be relevant for examining human development issues? To what extent SCT’s analytical structure can provide an approach to the use of capability spaces in social evaluations? Answering these questions would put in evidence the use of different forms of rankings in Sen’s work. As much as the role of the capability ‘approach’ seems indispensable for Sen’s broadening of informational spaces in social welfare evaluations, it has a limited function and much more needs to be said about how SCT provides an overarching framework for social evaluations. 

© 2011 |Tel: +598 24131007 | Gonzalo Ramírez 1926 | C.P. 11200 | Montevideo - Uruguay