Gerber: Legislative response to the threat of popular initiatives

Gerber. 1996. Legislative response to the threat of popular initiatives. American Journal of Political Science 40:99-128.

In Brief

Legislatures respond to the threat of initiatives, even if initiatives are not frequently used. This leads to policy that is closer to the median voter (in spatial dimensions) in initiative states than policy in non-initiative states. Gerber's model is similar to Romer and Rosenthal's setter-vetoer model.

Caveats: The locations of the preferences of the median voter, the legislature (taken as unitary actor) and costs of initiatives impact their influence across states. See Figures 2-4c (p. 108-111) for variants of the preference and cost distributions.

See also Lascher et al. (1996) and Lupia and Matsusaka (2004).

Data

Analysis of parental consent and notification legislation across the states. 1988-90-92 Pooled Senate Election studies using small random samples (Ns range from 55-107, mean 83) from each of the 50 states. Used mean (not median...more gradations with mean) of responses to poll about preference for parental consent. Only used those who reported voting. Why only use voters? The theory of pressure on legislators comes from the public's preference point, not the voters' preference point.

Evidence: Fancy equations p. 118 and 122. Used logit regressions, results on p. 119 and multinomial p. 123. Plug all data into a mass regression; find that the more preference for parental consent, the more likely it is a state will have such a law. Even stronger results for initiative states, supporting the theory. The greater the number of signature required for ballot qualification the less likely it is a law will exist, further supporting the cost part of the theory.