Employing a commonly‐used method of creating a continuous income variable from categorical data, we obtain results from a fiscal survey that reveal a strong nonmonotonic effect of income on the willingness to pay additional taxes for state expenditures on education and public aid. The existence of income‐demand schedules that are U‐ or inverted U‐shaped casts doubt on the appropriateness of assuming that the median income voter is decisive. After investigating the sensitivity of our results to different income measures, we suggest that fiscal surveys should be designed to provide sufficiently detailed information about respondents’incomes, especially for high‐income respondents.
|Original language||English (US)|
|Number of pages||16|
|Journal||Economics & Politics|
|State||Published - Mar 1995|