Skip to main content

Advertisement

Log in

The impact of land use regulation across the conditional distribution of home prices: an application of quantile regression for group-level treatments

  • Original Paper
  • Published:
The Annals of Regional Science Aims and scope Submit manuscript

Abstract

By increasing housing prices, land use regulations can have positive impacts among homeowners, but they can also have negative impacts on the availability of affordable housing. We examined heterogeneity in the price impacts of land use regulation across the conditional house price distribution; this heterogeneity may ameliorate or exacerbate the impacts of land use regulation on affordable housing. Our results suggest that the distribution of price impacts across the conditional house price distribution is relatively uniform. Results suggest that land use regulations both constrain housing supply and induce within housing market migration that spreads price impacts throughout the region.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2

Similar content being viewed by others

Notes

  1. See Ihlanfeldt (2007), Quigley and Raphael (2005) and Gyourko (2009) for more comprehensive surveys on the consequences of housing supply constraints.

  2. While Cunningham (2007) studied housing price volatility, the other two papers focused on the level of housing price.

  3. Malpezzi (1996) and Mayer and Somerville (2000) have used an early version of Wharton data to study the effect of regulation on construction and housing prices across metropolitan areas.

  4. Though most evidence comes from the USA due to data availability and significant variation in land use regulations across communities, evidence for similar relationships in England exists (Hilber and Vermeulen 2016).

  5. The lagged variables included proportion of adults possessing a college degree, proportion of residents over the age of 55, proportion white, proportion black, proportion of households who owned their home and average household income.

  6. Quigley et al. (2008) used preexisting measures of the political predisposition in each city as instruments. Saiz (2010) used religion and local government financial structure variables as instruments. Hilber and Vermeulen (2016) used exogenous variation from a policy reform, vote shares and historical density as instruments to identify the endogenous constraints—measures.

  7. Communities are defined as a Census Place.

  8. The Bartik shock is a measure for labor demand, named after Bartik (1991), and popularized in Blanchard and Katz (1992). It has since been employed by many papers including Diamond (2016), Notowidigdo (2013) and Saks (2008). See Goldsmith-Pinkham et al. (2018) for a literature view of Bartik shock.

  9. The classification of industries is based on the three digital Census industry codes.

  10. The difference in sample sizes was small: 3.1 million houses (ACS 1% sample) and 3.25 million parcels (Zillow annual sample). Zillow’s sample included houses from all counties in the ACS sample with some additional counties as well. Zillow’s median sale price was slightly lower than ACS. The median house size across the two sources was identical, and ACS properties had a slightly older median age. (Median year built in the ACS sample was 1975 compared to 1982 for the Zillow sample.)

  11. Coefficient estimates for block group characteristics should be interpreted with caution. A robust literature discusses the correspondence between these characteristics and other unmeasured local characteristics (Bayer et al. 2017; Beracha et al. 2016; Black 1999). For our purposes, these variables merely serve as a means of broadly controlling for local conditions in order to isolate the community fixed effect.

  12. In each quantile, we failed to reject the null hypothesis that the specified endogenous regressor WRLURI can actually be treated as exogenous. The test statistic is distributed as chi-squared with degrees of freedom equal to the number of regressors tested. In this case, the degree of freedom is one. The test statistics for each sequential quantile with p value in parenthesis are as follows: 2.98 (0.08); 3.68 (0.06); 2.28 (0.13); 2.05 (0.15); 2.1 (0.15); 1.62 (0.2); 1.23 (0.27); 1.63 (0.2); 2.61 (0.11).

References

Download references

Acknowledgements

Data are provided by Zillow through the Zillow Transaction and Assessment Dataset (ZTRAX). More information on accessing the data can be found at http://www.zillow.com/ztrax. The results and opinions are those of the author(s) and do not reflect the position of Zillow Group.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Tammy Leonard.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional ailiations.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Leonard, T., Yang, X. & Zhang, L. The impact of land use regulation across the conditional distribution of home prices: an application of quantile regression for group-level treatments. Ann Reg Sci 66, 655–676 (2021). https://doi.org/10.1007/s00168-020-01032-z

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00168-020-01032-z

JEL Classification

Navigation