Impact of Land Subsidence on Housing Sale Values: Evidence from the San Joaquin Valley, California

Mehdi Nemati; Michelle Sneed; Ariel Dinar

doi:10.3368/le.102.1.092324-0083R

Article Figures & Data

Figures

Tables
Additional Files

Download figure
Open in new tab
Download powerpoint
Figure 1
Principle of Effective Stress and Its Relationship to Groundwater Levels, Aquifer-System Compaction, and Land Subsidence
Source: Galloway, Jones, and Ingebritsen (1999).
Note: Subsidence of the land surface is a result of a decrease in pore-fluid pressure (ρ) and resulting increase in effective stress (σe) in fine-grained material under conditions of total stress (σT) in a one-dimensional, fluid-saturated geologic medium.
Download figure
Open in new tab
Download powerpoint
Figure 2
Land Subsidence and Groundwater Levels in California, 2005–2014: (a) Mendota; (b) Madera
Sources: Luhdorff and Scalmanini Consulting Engineers and the US Geological Survey for well data; University Navigation Satellite Timing and Ranging Consortium for continuous global positioning system (CGPS) data.
Note: CGPS station P304 and well 13S/15E-31J6 show groundwater-level decline and subsidence during drought periods near Mendota. CGPS station P307 and well 12S/17E-2J1 show groundwater-level decline and subsidence during drought periods and nondrought periods near Madera. Shaded periods represent calendar years affected by increased pumping.

View popup

Table 1

Number of Counties and Areas Affected by Land Subsidence in the San Joaquin Valley, California, 2015–2021

	No. of Counties with Subsidence	Total Subsidence Area (km²)	Area Under Low Subsidence (km²)^a	Area Under High Subsidence (km²)^b
2015	7	11,100	8,497	2,603
2016	7	6,652	6,629	23
2017	8	8,260	7,167	1,093
2018	7	6,280	6,271	9
2019	7	10,885	10,075	810
2020	7	10,936	9,119	1,817
2021	7	12,262	9,895	2,367

Note: Subsidence occurred in parts of the counties, not entire counties.
↵^a Low subsidence is < 0.18 m per year.
↵^b High subsidence is ≥ to 0.18 m per year.

View popup

Table 2

Observable Variables for Housing Sale in the San Joaquin Valley, California

	Full Sample (SJV)		No Subsidence		Low Subsidence^a		High Subsidence^b
	Mean	SD	Mean	SD	Mean	SD	Mean	SD
Sales price ($1,000 and 2021 dollars)	319	188	324	183	281	212	240	223
% agriculture (5 km)	0.07	1.73	0.06	1.54	0.17	2.84	0.13	1.95
Groundwater dependent	15.48	36.17	14.88	35.59	19.60	39.70	22.08	41.49
Distance to highway (km)	1.99	2.15	2.07	2.23	1.40	1.38	1.37	0.90
Distance to city (km)	0.85	4.03	0.90	4.22	0.54	2.18	0.51	2.33
% agriculture employed	8.05	9.79	7.59	9.34	10.82	12.12	15.51	11.26
Temperature (°C)	17.79	1.09	17.78	1.17	17.80	0.30	18.14	0.21
Precipitation (mm)	266.89	112.83	272.19	116.77	234.78	67.52	181.58	36.11
Age (years)	37.69	25.38	38.02	25.45	35.06	24.35	37.65	27.18
No. of bedrooms	3.21	0.93	3.23	0.88	3.04	1.25	3.05	0.97
No. of baths	2.02	0.82	2.02	0.81	1.97	0.88	1.89	0.61
Story	1.10	0.53	1.11	0.52	1.07	0.56	1.05	0.66
Pool (%)	16.86	37.44	16.74	37.33	18.28	38.65	14.25	34.96
Observations	275,122		241,894		29,234		3,994

Note: This table reports sample means, with sample standard deviations.
↵^a Low subsidence is < 0.18 m per year.
↵^b High subsidence is ≥ to 0.18 m per year.

View popup

Table 3

Log Sale Price on Subsidence in the San Joaquin Valley, California: Fixed Effects

	(1)	(2)	(3)	(4)	(5)
Subsidence	−0.065^***	−0.073^***	−0.024^***	−0.025^***
Subsidence	(0.005)	(0.005)	(0.004)	(0.004)
Low subsidence (< 0.18 m/year)					−0.024^***
Low subsidence (< 0.18 m/year)					(0.004)
High subsidence (≥ 0.18 m/year)					−0.058^***
High subsidence (≥ 0.18 m/year)					(0.008)
Year fixed effects	Yes	No	No	No	No
County fixed effects	Yes	No	No	No	No
Sub-basin fixed effects	Yes	No	No	No	No
County-year fixed effects	No	Yes	Yes	Yes	Yes
Sub-basin-year fixed effects	No	Yes	Yes	Yes	Yes
Housing characteristics	No	No	Yes	Yes	Yes
Other controls	No	No	Yes	Yes	Yes
Weather controls	No	No	No	Yes	Yes
% agriculture employed	No	No	No	Yes	Yes
Average sales value	278,713	278,713	278,713	278,713	278,713
Observations	275,122	275,122	275,122	275,122	275,122
R-squared	0.205	0.209	0.621	0.635	0.635

Note: Each column represents a separate regression using a different combination of controls and fixed effects. The dependent variable in all regressions is the log sale price. Standard errors are in parentheses and clustered at the zip code level with 1,000 bootstrap repetitions. The number of zip codes is 256. Housing characteristics include the year built, number of bedrooms, bathrooms, stories, and the presence of a pool. Other controls include the percentage of agriculture (5 km) and the distance to the closest highway and city (km). Weather variables include mean annual temperature and total precipitation. We include the square terms of temperature and precipitation to account for the nonlinear relationship between home values and weather variables.
↵*** p < 0.10.

View popup

Table 4

Log Sale Price on Subsidence in the San Joaquin Valley, California: Matching

	City	Sub-basin	County
	(1)	(2)	(3)
Subsidence	−0.032^***	−0.027^***	−0.021^***
Subsidence	(0.008)	(0.007)	(0.006)
County-year fixed effects	Yes	Yes	Yes
Sub-basin-year fixed effects	Yes	Yes	Yes
Housing characteristics	Yes	Yes	Yes
Other controls	Yes	Yes	Yes
Weather controls	Yes	Yes	Yes
% agriculture employed	Yes	Yes	Yes
Average sales value ($1,000)	242,708	246,540	235,260
Observations	14,234	24,508	32,668
R-squared	0.527	0.556	0.556

Note: Each column represents a separate regression that limits treated and comparison units to the same geography. The dependent variable in all regressions is the log sale price. Standard errors are in parentheses and clustered at the zip code level with 1,000 bootstrap repetitions. The number of zip codes is 256. Housing characteristics include the year built, number of bedrooms, bathrooms, stories, and the presence of a pool. Other controls include the percentage of agriculture (5 km) and the distance to the closest highway and city (km). Weather variables include mean annual temperature and total precipitation. We include the square terms of temperature and precipitation to account for the nonlinear relationship between home values and weather variables.
↵*** p < 0.10.

View popup

Table 5

Log Sale Price on Subsidence in the San Joaquin Valley, California: Fixed Effects with Repeat-Sales Data

	(1)	(2)	(3)
Subsidence	−0.027^***	−0.031^***	−0.032^***
Subsidence	(0.002)	(0.002)	(0.002)
Year fixed effects	Yes	No	No
County fixed effects	Yes	No	No
Sub-basin fixed effects	Yes	No	No
County-year fixed effects	No	Yes	Yes
Sub-basin-year fixed effects	No	Yes	Yes
House fixed effects	Yes	Yes	Yes
Weather controls	No	No	Yes
Average sales value ($1,000)	304,191	304,191	304,191
Observations	84,540	84,540	84,540
R-squared	0.731	0.741	0.741

Note: Each column in each panel represents a separate regression using a different combination of controls and fixed effects. The dependent variable in all regressions is the log sale price. Standard errors are in parentheses and clustered at the zip code level with 1,000 bootstrap repetitions. The number of zip codes is 218. Weather variables include mean annual temperature and total precipitation. We include the square terms of temperature and precipitation to account for the nonlinear relationship between home values and weather variables.
↵*** p < 0.10.

View popup

Table 6

Log Sale Price on Land Subsidence (LS) in the San Joaquin Valley, California: Heterogeneous Impacts

	High Early LS Areas	Low Early LS Areas	High-Education Areas
	(1)	(2)	(3)
Subsidence	0.002	−0.024^{^***}	−0.052^{^***}
Subsidence	(0.026)	(0.004)	(0.004)
County-year fixed effects	Yes	Yes	Yes
Sub-basin-year fixed effects	Yes	Yes	Yes
Housing characteristics	Yes	Yes	Yes
Other controls	Yes	Yes	Yes
Weather controls	Yes	Yes	Yes
% agriculture employed	Yes	Yes	Yes
Average sales value ($1,000)	192,276	273,641	325,918
Observations	4,106	238,713	49,564
R-squared	0.576	0.634	0.652

Note: Each column represents a separate regression on a different subsample: (1) areas with substantial early subsidence, (2) areas with low early LS exposure, and (3) areas with a higher share of educated households. The dependent variable in all regressions is the log sale price. Standard errors are in parentheses and clustered at the zip code level with 1,000 bootstrap repetitions. The number of zip codes is 256. Housing characteristics include the year built, number of bedrooms, bathrooms, stories, and the presence of a pool. Other controls include the percentage of agriculture (5 km) and the distance to the closest highway and city (km). Weather variables include mean annual temperature and total precipitation. We include the square terms of temperature and precipitation to account for the nonlinear relationship between home values and weather variables.
↵*** p < 0.10.

View popup
Table 7
Average and Aggregate Impacts from Land Subsidence in the San Joaquin Valley, California
Average Impacts ($1,000) Aggregate Impacts ($1,000)
Subsidence 6.967 $1,442,448
Low subsidence (< 0.18 m/year) 6.689 $1,052,113
High subsidence (≥ 0.18 m/year) 16.165 $804,209