Voter demand for fluoridated water: a tale of two c(av)ities

Abstract

Government fluoridation of public water systems to promote dental health has long been advocated by many health associations, based on the findings of mainstream scientific studies. Despite this, fluoridation remains a controversial issue. Some in the public are antithetical to the science behind fluoridation while others view it as an infringement on individual choice. Voting data from referendums in two of the six largest US cities without fluoridated water (Portland, Oregon, and Wichita, Kansas) are used to examine the factors driving voter demand for and against fluoridation. Although regression analysis reveals differences between the cities, a strong commonality is greater support for fluoride coming from voting precincts with higher concentrations of college graduates. Additionally, even though advocates often laud water fluoridation as a relatively inexpensive way to extend dental health benefits to all children (regardless of income levels), presence of children in households surprisingly does not appear to translate into voter support. Lastly, after controlling for socio-economic factors, results suggest that opposition to fluoridation does not appear to come from the political centre, but from the libertarian right and environmental left.

Keywords:

• fluoridation
• referendums
• public water
• dental health

JEL Classification:

• D72
• I18

Notes

¹  Fluoridation has also recently met voter resistance in Australian and Canadian municipalities (Kliff, 2013).

²  Portland voted on fluoridation five times since 1956. In 1978, Portland voted for fluoride, but did not add it to the water supply before voters repealed the decision in 1980. A 1964 referendum in Wichita overturned its city commission vote to fluoridate its water. We were unable to obtain detailed data on these earlier votes.

³  Results are unaffected if concurrent party votes are substituted.

⁴  There is high correlation between %Poverty and %Children for Wichita78 (r = 0.42) not found in the other samples.

⁵  Being born out of state does not necessarily indicate that an individual was exposed to fluoridation, nor does being born in-state necessitate being born in Wichita or Portland.

⁶  In 1978, the Republican Party in Kansas was much more moderate than today. President Gerald Ford, the Republican candidate, was also from the party’s moderate wing. Dropping %Poverty washes out %Republican in Wichita78; it becomes negative and significant for Portland and Wichita12.

⁷  For Wichita78, %Libertarian and Conservative includes the Libertarian, American, American Independent and Prohibition party votes. For Wichita12, only the Libertarian vote, and for Portland, the votes of the Libertarian Party and two branches of the Constitution Party.

⁸  This is not quite an apples to apples comparison as the party variables are based on different elections. To test for structural change, the Wichita samples were pooled. Additional variables were a Wichita12 dummy and Wichita12 interacted with the previous regressors. The only significant new variables were the Wichita12 interactions with %Children and the party variables. F-Tests indicated that the socio-economic interactions were jointly significant, as were the party interactions (p-values = 0.03, 0.006).