An externality of groundwater depletion: land subsidence and residential property prices in Phoenix, Arizona

This article was originally published with errors. This version has been corrected. Please see Corrigendum (http://dx.doi.org/10.1080/21606544.2016.1240754).

ABSTRACT

One of the main physical effects of the depletion of aquifers is land subsidence – the lowering of the land-surface elevation as a result of groundwater overdraft. A second effect is the development of earth fissures as a result of the horizontal movement of sediments during subsidence. To determine the value of these effects we investigated the impact of land subsidence and earth fissures on residential property values in Maricopa County, Arizona. Using 82,716 arms-length property sales between 2004 and 2010, we estimated a fixed effects hedonic price model. We found that existing and future land subsidence, and earth fissures had a negative impact on the property values. The mean value of properties located in land subsidence features was lower than those located outside land subsidence features, and the disamenity associated with earth fissures was largest for properties located in land subsidence features.

KEYWORDS:

• Land subsidence
• earth fissure
• hedonic price model
• disamenity impact
• Maricopa County

View correction statement:

Corrigendum

Acknowledgments

I acknowledge Brian D. Conway, a hydrologist at Arizona Department of Water Resources, who generously provided GIS data on land subsidence. Also, I dedicate this paper to my daughter, Emmanuelle Yoo.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

This article was originally published with errors. This version has been corrected. Please see Corrigendum (http://dx.doi.org/10.1080/21606544.2016.1240754).

1. A housing unit is owner-occupied if the owner or co-owner lives in the unit, even if it is mortgaged or not fully paid for. The unit is “owned by you or someone in this household with a mortgage or load” if it is being purchased with a mortgage or some other debt arrangement such as a deed of trust, trust deed, contract to purchase, land contract, or purchase agreement. The homeownership rate is computed by dividing the number of owner-occupied housing units by the number of occupied housing units or households.

2. The vast majority of hedonic studies have used linear, semi-log, and log–log models to avoid the risk of omitted variable bias. We tested model performance of linear, semi-log, and log–log models, in terms of adjusted R² and found semi-log model yields the largest R².

3. Roads include all primary, secondary, local neighbourhood, and rural roads, city streets, vehicular trails, ramps, service drives, alleys, parking lot roads, private roads for service vehicles, bike paths or trails, etc.

4. The estimated coefficients on city dummy variables from OLS model with fixed effects are available upon request from the authors.

5. Using 2010 Census housing unit shape file on a block-level, we calculated the total number of housing units in West Valley land subsidence feature using zonal statistics on ArcGIS Desktop 10.3. The total number of housing units in West Valley land subsidence feature in 2010 was 174,531. The mean reduction in housing prices in the affected area was $24,570.