Suchergebnisse

Abstract. The argument that people underestimate the costs of providing public goods, and overestimate the benefits of government programmes, has received widespread attention in the political science literature. According to this argument the (misinformed) citizen demands greater amounts of public sector goods than if he or she had been completely informed. Democratic competition leads political parties to adjust to the perceived demands of the voters rather than to the actual interests of the electorate. Therefore, the supply of public goods and services tends to be greater than the social optimum. The present article tests two propositions related to this argument using survey data. First, an analysis of a nationwide survey from Norway (1990) indicates a widespread misperception of government costs, both the unit costs of particular services and the aggregate costs of the public sector. However, the study reveals no sign of a general underestimation of costs. Second, the survey implies that while the demand for government services declines as the unit cost increases, the price elasticity of government services seems to be small. The (mis‐)perceived tax‐price seems to have little impact on demand compared to the (mis‐)perceived costs levied on society as a whole. Even if people did under‐rate government costs, the impact on demand would be marginal. Thus, empirical evidence provides little or no support for the hypothesis of fiscal illusions.

In an influential study, Matthew Gentzkow found that the introduction of TV in the United States caused a major drop in voter turnout. In contrast, the current analysis shows that public broadcasting TV can increase political participation. Detailed data on the rollout of television in Norway in the 1960s and 1970s are combined with municipality-level data on voter turnout over a period of four decades. The date of access to TV signals was mostly a side effect of geography, a feature that is used to identify causal effects. Additional analyses exploit individual-level panel data from three successive election studies. The new TV medium instantly became a major source of political information. It triggered political interest and caused a modest, but statistically significant, increase in voter turnout.