ABSTRACT
Homeowner associations (HOAs), as a form of private residential governance, have grown exponentially in the United States, representing a profound transformation in urban governance. This paper examines the fiscal correlation between local municipal governments and private residential governments, and specifically, the extent to which the proliferation of HOAs in a municipality is related to municipal expenditures and revenues. Administrative data of the Florida Department of Business and Professional Regulation are employed with spatial econometric methods for accounting for potential inter-jurisdictional spillover effects. Empirical evidence suggests that a greater presence of HOAs within a local municipality is associated with the municipality’s declining public expenditures and revenues for providing public services to local residents.
Disclosure statement
No potential conflict of interest was reported by the author.
Supplementary material
Supplemental data for this article can be accessed here.
Notes
1. For more information on Census Geocoder, please see https://www.census.gov/geo/maps-data/data/geocoder.html.
2. Actual number of observations for each model specification may vary and thus differ from 170 due to missing data.
3. Please refer to Koop (Citation2003) for a more detailed and complete introduction of Bayesian econometrics, and to LeSage and Pace (Citation2009) for introduction of the Bayesian MCMC estimation method and its applications.
4. There is also a growing consensus that spatially lagged dependent variables in SAR should be properly instrumented. For a review of discussions and IV strategies, please see Ferraresi (Citation2019) and Gibbons and Overman (Citation2012).
5. The summary of the survey findings can be found from the website, http://nhts.ornl.gov/2009/pub/stt.pdf.
6. Given the page limit, estimation results of spatial models using the weighting matrix of a 15-mile bandwidth and employing the inverse economic distance of per capita personal income are presented in the appendix.
7. In the SDEM 2SLS models, λ shows 5% significance in the models with Tot_Exp, CR_Exp, Tax_Rev, and PT_Rev as dependent variables, 10% significance in the model with PS_Exp, Tr_Exp, and TotRev as dependent variables.
Additional information
Notes on contributors
Shaoming Cheng
Shaoming Cheng is an associate professor in the Department of Public Policy and Administration at Florida International University. His research interests centre on entrepreneurship and small business development policy as well as regional economic health, performance, and development.
Hai (David) Guo
Hai (David) Guo is an associate professor in the Department of Public Policy and Administration at Florida International University. His research focuses on state and local public finance, budgeting, and financial management.