A sensible notion of social welfare does not come without some sort of interpersonally comparable well-being. On the other hand, collective choice rules can be designed to improve the chance that a democracy achieves good things for its citizens. But without being more specific in stipulating what constitutes social welfare, our measuring rod for identifying “good things” disappears just when we would wish to use it. Further, merely noting that interpersonal comparisons of individual well-being are needed to gain traction over social welfare doesn’t tell us how to make the comparisons. Without more details, how can we tie our notions of collective choice to social welfare? How are we to get around the problems Arrow identified? That is the focus of this chapter.

TO AGGREGATE OR NOT TO AGGREGATE: A FALSE DILEMMA

Arrow’s negative finding stems from his insistence that what is being aggregated are either individual preferences or something very similar to that: an indicator of well-being with ordinal properties of consistency but no basis for interpersonal comparisons. This is a relatively unfortunate, if realistic, assessment of the problem of social choice. After all, voting rules only aggregate some shadow of the rankings of proposals by voters in some way. To make a social choice, the rules aggregate very limited individual information. Any conception of social welfare must utilize a more complex mix of inputs than voting rules utilize.