Skip to main content
. Author manuscript; available in PMC: 2019 Oct 18.
Published in final edited form as: Soc Sci Med. 2018 Dec 4;220:411–420. doi: 10.1016/j.socscimed.2018.12.004

Table A2.

County-Level Determinants of Years of NRPS Roll-out

Variables Number of years since initial roll-out in China
(1) (2) (3)
county demographic characteristics
Proportion of residents aged > 60 −1.0912
(1.4380)
−1.7857
(1.6366)
0.7350
(2.5520)
Proportion of residents aged 45-59 2.6473
(1.6512)
2.2985
(1.7937)
3.3090
(2.7339)
Population with local Hukou (10,000) 0.0008
(0.0009)
0.0009
(0.0013)
0.0012
(0.0013)
Autonomous county, state or region −0.0682
(0.0853)
−0.0708
(0.0855)
−0.0867
(0.1997)
county economic and social characteristics
National poverty-stricken county −0.0567
(0.0565)
0.0429
(0.0923)
GDP per capita (10,000) 0.0129
(0.0261)
0.0467
(0.0406)
Proportion of primary industry added value in GDP −0.0049
(0.0045)
−0.0077
(0.0055)
Net revenue per capita of the local government (10,000) 0.3892
(0.4447)
−0.3473
(0.5908)
Number of beds per 10,000 people in hospitals and orphanages −0.0023
(0.0015)
−0.0014
(0.0030)
Key politician (party secretary) profile
Born in this municipality −0.2310
(0.1629)
Age 50-53 0.2303
(0.1591)
Age 54-57 0.1247
(0.1556)
Age above 58 0.2120
(0.1759)
Male 0.3716
(0.2371)
Minority −0.0995
(0.1598)
highest degree = M.A 0.0864
(0.1059)
highest degree = PhD 0.0400
(0.1237)
Graduated from the Party School −0.0156
(0.0972)
Major in agriculture 0.1191
(0.1389)
major_medicine −0.0058
(0.1598)
Major in humanities / social science 0.1490
(0.1343)
Major in science or technology 0.1289
(0.1038)
Province FE Yes Yes Yes
N 2011 2009 1928
Adjusted R-sq 0.1171 0.0976 0.1901

Data sources: same as Table A1

Notes: [1] Column 1 presents results with county demographic characteristics. Column 2 adds other economic and social characteristics. Party secretary background variables are further included in Column 3. [2] Constants are omitted to save space. [3] ***, ** and * represent statistical significance at the 1%, 5% and 10% levels, respectively. Standard errors are reported in the parentheses. [4] All the standard errors are clustered at the county level.