Ballot Propositions and Information Costs: Direct Democracy and the Fatigued Voter

Abstract

Here b denotes the benefit from having your .candidate win, p the probability of your vote deciding the outcome, and c the costs of voting (Downs 1957). By and large this formulation has considered b and c at the systemic level; ease of registration, competitiveness of the election, literacy requirements and many similar factors have all been used to assess levels of turnout (Rosenstone and Wolfinger 1978; Powell 1980). In terms of this literature, then, c refers to a set of costs imposed on all voters within a given political jurisdiction. Once a voter has arrived at the voting booth the costs of actually marking a ballot would seem to be trivial relative to the costs of actually turning out to vote. This last point, however, is not true for all types of ballot. Many ballots require much more of voters than simply turning out and marking a single preference for one office. Under preferential systems or where a large number of elective offices and/or propositions are on the ballot, the decisions facing voters are quite complex. Between 1974 and 1988, for example, in addition to a wide range of elective offices, Californians faced an average of 13 propositions at each election. As can be seen from Figure 1, California's voters express some dissatisfaction with this process. Nearly one-third of voters gave responses which suggested that propositions made too many demands upon them. Responses which noted that there were too many propositions or that they were too confusing (i.e., that too many demands were being made) far outweighed responses which expressed a lack of faith in the efficacy of the proposition process (e.g. that special interests dominate the process). The question here is whether we can represent such

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 112 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.