TABLE A3

Additional Robustness Checks for the Spatial Error Model

(a)(b)(c)(d)(e)(f)(g)
Distance to nearest city center×Within 10 miles−0.0086***
(0.0013)		−0.0103***
(0.0012)		−0.0094***
(0.0013)		−0.0096***
(0.0012)		−0.0099***
(0.0013)		−0.0093***
(0.0012)		−0.0106***
(0.0014)
Distance to nearest city center×Within 10 miles×Post-2008 dummy0.0046*
(0.0026)		0.0022
(0.0027)		0.0003
(0.0027)		0.0045*
(0.0025)		0.0048*
(0.0025)		0.0046*
(0.0025)		0.0052**
(0.0025)
Distance to nearest city center×Beyond 10 miles−0.0076***
(0.0012)		−0.0083***
(0.0011)		−0.0073***
(0.0012)		−0.0083***
(0.0011)		−0.0086***
(0.0012)		−0.0080***
(0.0011)		−0.0096***
(0.0013)
Distance to nearest city center×Beyond 10 miles×Post-2008 dummy0.0067***
(0.0025)		0.0028
(0.0024)		0.0002
(0.0015)		0.0064***
(0.0024)		0.0063***
(0.0024)		0.0064***
(0.0024)		0.0066***
(0.0024)
Incremental distance to second-nearest city center−0.0036***
(0.0007)		−0.0043***
(0.0007)		−0.0044***
(0.0008)		−0.0044***
(0.0007)		−0.0045***
(0.0007)		−0.0042***
(0.0007)		−0.0049***
(0.0008)
Incremental distance to second-nearest city center×Post-2008 dummy0.0038**
(0.0017)		0.0010
(0.0016)		0.0004
(0.0010)		0.0035**
(0.0016)		0.0034**
(0.0016)		0.0036**
(0.0016)		0.0030*
(0.0016)
Urban population within 25 miles0.0003***
(0.0000)		0.0003***
(0.0000)		0.0003***
(0.0000)		0.0003***
(0.0000)		0.0003***
(0.0000)		0.0003***
(0.0000)		0.0002***
(0.0000)
Urban population within 25 miles×Post-2008 dummy1.00E–04
(0.0001)		−3.64E–04***
(0.0001)		−1.54E–04**
(0.0001)		1.56E–04
(0.0001)		1.58E–04
(0.0001)		1.33E–04
(0.0001)		1.60E–04
(0.0001)
Gravity index2.08E–05***
(0.0000)		2.29E–05***
(0.0000)		2.56E–05***
(0.0000)		2.55E–05***
(0.0000)		2.47E–05***
(0.0000)		2.54E–05***
(0.0000)		2.43E–05***
(0.0000)
Gravity index×Post-2008 dummy−2.14E–05***
(0.0000)		−2.28E–05***
(0.0000)		−2.55E–05***
(0.0000)		−2.61E–05***
(0.0000)		−2.50E–05***
(0.0000)		−2.60E–05***
(0.0000)		−2.50E–05***
(0.0000)
Building area % of parcel0.1790***
(0.0542)		0.0307
(0.0540)		0.0402
(0.0498)		0.1511**
(0.0521)		0.1437***
(0.0520)		0.1534***
(0.0521)		0.1498***
(0.0517)
Distance to highway on-ramp−0.0063*
(0.0035)		−0.0076**
(0.0033)		−0.0069**
(0.0031)		−0.0052
(0.0034)		−0.0054
(0.0035)		−0.0056
(0.0033)		−0.0044
(0.0036)
Distance to railway station0.0053
(0.0036)		0.0021
(0.0034)		0.0004
(0.0032)		0.0036
(0.0036)		0.0039
(0.0036)		0.0043
(0.0035)		0.0039
(0.0036)
Prices with assessed building valuesYes
The post period is 2008 onlyYes
Shifting the year of change to 2005Yes
Spatial weights matrix2nn2nn2nnScalar-normalizedInverse-distance1nn4nn
Spatial autocorrelation coefficient0.1255***
(0.0102)		0.1459***
(0.0105)		0.1454***
(0.0099)		0.0673***
(0.0051)		0.1814***
(0.0118)		0.0844***
(0.0077)		0.2827***
(0.0148)
AIC20,69217,03519,18919,87819,82119,93119,711
Root mean squared error0.65170.58490.58360.62640.62400.62870.6197
Number of observations10,37810,37810,37810,37810,37810,37810,378

  • Note: In all models the dependent variable is the log of per-acre agricultural land prices without structures, except for column (a). In column (d), we use a scalar-normalized spatial weights matrix instead of a row-standardized one, following the suggestions of Kelejian and Prucha (2010), in which the scalar is the minimum of the maximum row sums and maximum column sums of the input weights. In column (e), we use an inverse-distance spatial weights matrix rather than a k-nearest-neighbor one. 1nn, 2nn, and 6nn denote a 1-nearest-neighbor, 2-nearest- neighbor, and 6-nearest-neighbor spatial weights matrix, respectively. The estimated standard errors of the spatial autocorrelation coefficients are in parentheses. AIC, Akaike information criterion.

  • *, **, and *** indicate the coefficient is significant at the 10%, 5%, and 1% levels, respectively.