Graduation Year


Document Type

Open Access Senior Thesis

Degree Name

Bachelor of Arts


Environmental Analysis

Reader 1

Bowman Cutter

Reader 2

Char Miller

Terms of Use & License Information

Terms of Use for work posted in Scholarship@Claremont.

Rights Information

© 2015 William S Lewis


In California, voters frequently face ballot propositions dealing directly or indirectly with environmental protection. Records of these votes provide powerful evidence of the character of voters’ demand and willingness-to-pay for environmental public goods (e.g., air quality, watershed ecosystem services, parks and recreation), and have been used in past environmental econometrics research to produce aggregated income and price effect estimates. Using neighborhood-level voting records on seven environmental-related ballot propositions in California between 2002 and 2010, this econometric study investigates the nature of voters’ demand for environmental public goods, focusing on the effect of household income on pro-environment voting. Unlike previous studies, this study uses geographically weighted regression (GWR) to determine how estimates vary across the historically, culturally, and politically diverse state of California. Preliminary statewide results from an ordinary least-squares regression model suggest that demand decreases with voter income, and that this negative income effect is strongest among lower-income households. However, GWR results suggest that the magnitude, and even the sign, of income effects varies regionally. The San Francisco Bay Area, in particular, stands out as anomalous from the statewide model estimates: in this region, wealthier households are more likely than lower-income households to support environmental propositions, ceteris paribus. This finding is consistent across all propositions studied, which include water bonds, State Parks funding, and the California High-Speed Rail program, among others. GWR results suggest that political geography and regional culture determines the way in which income (as well as education and other factors) affects voters’ support of environmental propositions.