Local Growth Control at the Ballot Box: Real Effects or Symbolic Politics?

ABSTRACT:

Growth control regulations are pervasive in local jurisdictions throughout the United States; yet there is still much uncertainty about their effectiveness in slowing down or halting growth. Moreover, there is considerable debate over whether there are unintended (or sometimes intended) exclusionary consequences that disproportionately affect minority and low-income populations. Employing multiple regression analyses, this study examines the effects of growth control ballot measures, adopted by voters, on housing growth and sociodemographic change in local jurisdictions. The findings from the multiple regression analyses reveal that cities in which growth controls were adopted at the ballot box do have slower rates of housing growth. There is also evidence that ballot box growth controls reduce growth in Hispanic and lower-income populations. Overall, the results from this study suggest that the adoption of ballot box growth controls is not merely “symbolic politics,” but has real measurable consequences on housing growth. Unfortunately, growth controls adopted by the ballot box may also contribute to the sociospatial segregation of cities by race/ethnicity and income.

Notes

1 Regulation is measured as an additive score of all regulations relating to rent controls, land use and zoning, infrastructure, and building and subdivision codes. Malpezzi (1996) developed a more sophisticated measure of regulation using factor analysis, but found that the factor scores had high correlations with the simple additive scales. Therefore, he decided to only report the results for the simple additive scales.

2 If the CAR database did not contain sufficient detail to determine whether it was a growth control measure, then the sample ballots from these measures were also requested from the City Clerk’s offices.

3 Upon requesting the sample ballots, it was realized that there were five duplicate measures that were on the list and, therefore, these were omitted. Other reasons for the lack of retrieval of ballot measures involved City and County Clerks’ staff not being able to find copies of sample ballots or not having sufficient staff members to locate these documents. In general, City and County clerks’ staff were helpful.

4 The inclusion of baseline variables corresponding to dependent variables allows for interpretation of the dependent variable as a rate of growth.

5 A minimum 1-year lag period is used between the adoption of the growth control measure and change variables because prior research has indicated that there are noticeable effects of growth control policies after 1–2 years (Levine, 1999).

6 All other jurisdictions outside of the San Francisco Bay Area, Los Angeles Region, and Central Valley are included in this category.

7 For all variables that were not normally distributed (e.g., positively skewed distributions), the natural log of the values for these variables were used in the multiple regression analyses. These variables are signified by (LN). The only variable that was not logged, even though there was evidence of positive skewness, is employment change. This was done because the transformed variable did not improve the fit of the model or significantly alter the results. Therefore, this variable was left in its original form.

8 Bivariate analyses of all independent variables were conducted to test for multicollinearity. Correlation coefficients for the bivariate analyses did not exceed .70, which is the conventional threshold that indicates that multicollinearity may be present in the regression models.

9 The decision to use a dummy variable instead of a continuous variable was due to the lack of variation among cities and the number of growth control ballot measures that were adopted. The majority of cities only adopted one ballot measure with a few adopting more than one.