Working Paper: NBER ID: w29389
Authors: Sandro Ambuehl; B. Douglas Bernheim
Abstract: Collective decision making requires preference aggregation even if no ideal aggregation method exists (Arrow, 1950). We investigate how individuals think groups should aggregate members' ordinal preferences—that is, how they interpret "the will of the people." Our experiment elicits revealed attitudes toward ordinal preference aggregation and classifies subjects according to the rules they implicitly deploy. Majoritarianism is rare while rules that promote compromise are common. People evaluate relative sacrifice by inferring cardinal utility from ordinal ranks. Cluster analysis reveals that our classification encompasses all important aggregation rules. Aggregation methods exhibit stability across domains and across countries with divergent traditions.
Keywords: preference aggregation; social choice; ordinal preferences; majoritarianism; Borda rule
JEL Codes: C91; D7
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| use of scoring rules (C52) | specific outcomes in preference aggregation (D79) |
| choice of aggregation method (C43) | outcomes of preference aggregation (D79) |
| type of scoring rule used (C52) | aggregation outcomes (C43) |
| method of preference aggregation (D79) | perceived intensity of preferences (D11) |
| cultural context (Z10) | choice of aggregation rules (C43) |