# RECORD 28afe169f4e2851e590b8152147dbd5bb4ee90f0
# TITLE: Quality of Life in Chinese Cities
# YEAR: 2021.0  VENUE: Social Science Research Network  DOI: 10.2139/ssrn.3764334
# AUTHORS: Tiechou Shi; Wenzhang Zhu; Shi-He Fu
# PAGES: 39  CANDIDATE TABLE PAGES (1-based): [27, 30, 32]
# ABSTRACT: Abstract The Rosen-Roback spatial equilibrium theory states that cross-city variations in wages and housing prices reflect urban residents' willingness to pay for urban amenities or quality of life. This paper is the first to quantify and rank the quality of life in Chinese cities based on the Rosen-Roback model. Using the 2005 1% Population Intercensus Survey data, we estimate the wage and housing hedonic models. The coefficients of urban amenity variables in both hedonic models are considered the implicit prices of amenities and are used as the weights to compute the quality of life for each prefecture-level city in China. In general, provincial capital cities and cities with nice weather, good air quality, and accessible public transit have high quality of life. We also find that urban quality of life is positively associated with the subjective well-being of urban residents.


===== PAGE 1 (score 1, numeric-paren 0, table-words 0, var-words 1) =====

Munich Personal RePEc Archive
Quality of Life in Chinese Cities
Shi, Tie and Zhu, Wenzhang and Fu, Shihe
Xiamen University, Xiamen University, Xiamen University
12 January 2021
Online at https://mpra.ub.uni-muenchen.de/105266/
MPRA Paper No. 105266, posted 19 Feb 2021 06:32 UTC


===== PAGE 27 (score 94, numeric-paren 67, table-words 8, var-words 11) =====

26 
 
 
Table 2 Results from hedonic models 
 
(1) 
(2) 
(3) 
Variables 
Housing price 
Housing rent 
Monthly income 
Temperature index 
-0.001 
-0.005 
0.007*** 
 
(0.00) 
(0.00) 
(0.00) 
Mean monthly precipitation 
0.001 
0.002*** 
0.002*** 
 
(0.00) 
(0.00) 
(0.00) 
Mean daily sunshine duration 
0.056*** 
0.078*** 
0.025** 
 
(0.02) 
(0.02) 
(0.01) 
Annual mean concentration of PM2.5 
-0.000 
-0.002 
0.001** 
 
(0.00) 
(0.00) 
(0.00) 
Green coverage ratio 
0.002 
0.002 
0.003** 
 
(0.00) 
(0.00) 
(0.00) 
Coastal city 
0.178*** 
0.019 
0.138*** 
 
(0.04) 
(0.05) 
(0.03) 
Population density 
-0.002 
-0.003 
-0.003*** 
 
(0.00) 
(0.00) 
(0.00) 
Population size 
-0.001*** 
-0.000 
-0.000 
 
(0.00) 
(0.00) 
(0.00) 
Number of theatres 
0.055*** 
0.108*** 
0.032* 
 
(0.02) 
(0.03) 
(0.02) 
Number of hospital bed 
0.001 
0.003*** 
0.000 
 
(0.00) 
(0.00) 
(0.00) 
Teacher-student ratio in high schools 
0.001 
-0.003 
0.001 
 
(0.00) 
(0.00) 
(0.00) 
Number of higher education institutions 
0.001 
0.001 
0.002** 
 
(0.00) 
(0.00) 
(0.00) 
Road area ratio 
-0.003 
0.006 
0.003 
 
(0.00) 
(0.00) 
(0.00) 
Number of airline routes 
0.003*** 
0.002* 
0.001** 
 
(0.00) 
(0.00) 
(0.00) 
Capital city 
0.163*** 
0.107* 
0.029 
 
(0.06) 
(0.06) 
(0.03) 
R2 (full model) 
0.734 
0.542 
0.616 
R2 (excluding amenities) 
0.702 
0.511 
0.589 
Obs. 
421660 
227594 
1157385 
Note. The dependent variables in Columns (1)-(3) are ln(housing price), ln(housing rents), and 
ln(wage), respectively. Standard errors are in the parentheses and clustered at the city level. “*”, 
“**”, “***” denote significance at the 10%, 5%, and 1% levels, respectively. The housing hedonic 
models in Columns (1) and (2) also include housing attributes and group fixed effects as listed 
in Table A1; the wage hedonic model in Column (3) also include worker attributes and 
demographic group fixed effects as listed in Table A2.  
 
 


===== PAGE 28 (score 12, numeric-paren 0, table-words 1, var-words 10) =====

27 
 
 
Table 3: Compensating differentials and implicit prices for amenities (CNY) 
 
(1) 
(2) 
(3) 
Amenity 
Compensating 
differential: housing 
Compensating 
differential: Wage 
Implicit price  
Temperature index  
-4.38 
-52.512 
-56.90 
Mean monthly precipitation 
4.38 
-15.01 
-10.63 
Mean daily sunshine duration 
245.28 
-187.56 
57.72 
Annual mean concentration of PM2.5 
0 
-7.50 
-7.50 
Green coverage ratio 
8.76 
-22.51 
-13.75 
Coastal city 
779.64 
-1035.34 
-255.70 
Population density  
-8.76 
22.51 
13.75 
Number of theatres 
240.9 
-240.08 
0.82 
Number of hospital bed 
4.38 
0 
4.38 
Teacher-student ratio in high schools 
4.38 
-7.50 
-3.12 
Number of higher education institutions 
4.38 
-15.01 
-10.63 
Road area ratio 
-13.14 
-22.51 
-35.65 
Number of airline routes 
13.14 
-7.50 
5.64 
Capital city 
713.94 
-217.57 
496.37 
Note. Column (1) = Monthly housing expenditure (CNY365)×𝛽2×12; Column (2) = (-1)×
Average monthly income (CNY625.20)×𝛾2×1.83×12; Column (3) = Column (1) + Column 
(2). 𝛽2 and 𝛾2 are coefficients in Models (8) and (9).  
 
 


===== PAGE 30 (score 35, numeric-paren 21, table-words 6, var-words 2) =====

29 
 
 
 
Table 5 Correlation matrix for QOL index and its components 
QOL indexes 
Climate 
Environmental quality 
Urban condition 
Environmental quality 
0.154** 
 
 
 
(0.009) 
 
 
Urban condition 
0.009 
-0.162*** 
 
 
(0.875) 
(0.006) 
 
Overall 
0.244*** 
0.027 
0.960*** 
 
(0.000) 
(0.573) 
(0.000) 
Note. “*”, “**”, and “***” denote statistical significance at the 10%, 5%, and 1% levels, 
respectively. p values are in the parentheses.  
 
 
Table 6 Subjective well-being and quality of life in cities 
 
(1) 
(2) 
(3) 
(4) 
(5) 
QOL 
0.052** 
 
 
 
 
 
(0.02) 
 
 
 
 
QOL: Climate 
 
0.580*** 
 
 
0.625*** 
 
 
(0.20) 
 
 
(0.20) 
QOL: Environmental quality 
 
 
-0.513 
 
-0.593 
 
 
 
(0.36) 
 
(0.36) 
QOL: Urban condition 
 
 
 
0.043** 
0.028 
 
 
 
 
(0.02) 
(0.02) 
R2 
0.016 
0.054 
0.018 
0.011 
0.085 
adj. R2 
0.008 
0.045 
0.009 
0.002 
0.061 
N 
117 
117 
117 
117 
117 
Note. The dependent variable is subjective well-being of cities, which are the coefficients of city 
dummy variables from regressing individual self-reported life satisfaction on city dummy 
variables and a set of individual demographic traits using the 2010 CFPS data. The quality of 
life index has been divided by 10,000. Robust standard errors are in the parentheses. “*”, “**”, 
and “***” denote statistical significance at the 10%, 5%, and 1% levels, respectively. 
 
 
 


===== PAGE 31 (score 2, numeric-paren 0, table-words 1, var-words 0) =====

30 
 
Table 7: Comparison between our ranking and the ranking by The Research Center 
for Urban Quality of Life 
 
City 
 
 
Ranking by authors 
 
Ranking by RCUQL 
Average rank between 2011 and 2018 
Ranking based on subjective 
satisfaction index 
Ranking based on 
objective index 
Shanghai 
1 
19 
6 
Beijing 
2 
24 
2 
Guangzhou 
3 
22 
3 
Kunming 
4 
24 
11 
Shenzhen 
5 
31 
4 
Chengdu 
6 
10 
22 
Xi’an 
7 
23 
8 
Haikou 
8 
12 
27 
Xiamen 
9 
14 
13 
Urumqi 
10 
19 
23 
Chongqing 
11 
10 
32 
Shenyang 
12 
16 
11 
Dalian 
13 
20 
19 
Qingdao 
14 
10 
20 
Changsha 
15 
22 
16 
Hangzhou 
16 
6 
8 
Nanjing 
17 
18 
3 
Harbin 
18 
21 
28 
Wuhan 
19 
21 
11 
Guiyang 
20 
25 
18 
Zhengzhou 
21 
19 
32 
Nanning 
22 
23 
28 
Yinchuan 
23 
15 
14 
Xining 
24 
11 
33 
Fuzhou 
25 
16 
26 
Shijiazhuang 
26 
16 
20 
Taiyuan 
27 
24 
24 
Lanzhou 
28 
25 
28 
Huhehaote 
29 
23 
9 
Ningbo 
30 
7 
13 
Jinan 
31 
9 
21 
Hefei 
32 
9 
22 
Changchun 
33 
10 
15 
Nanchang 
34 
23 
27 
Tianjin 
35 
16 
23 
 Note. The ranking by the subjective satisfaction index in 2018 is not provided by the RCUQL. 
 
 


===== PAGE 32 (score 24, numeric-paren 21, table-words 1, var-words 1) =====

31 
 
 
Table 8 Spearman’s rank correlation coefficients between our and the RCUQL’s 
rankings for 35 cities  
 
(1) 
(2) 
 
Ranking based on  
subjective satisfaction index 
Ranking based on  
objective index 
2011年 
-0.075 
0.047 
 
(0.694) 
(0.807) 
2012年 
-0.230 
0.382** 
 
(0.221) 
(0.037) 
2013年 
-0.303 
0.292 
 
(0.104) 
(0.117) 
2014年 
-0.202 
0.401** 
 
(0.285) 
(0.028) 
2015年 
0.192 
0.471*** 
 
(0.308) 
(0.009)  
2016年 
0.037 
0.445** 
 
(0.846) 
(0.014)  
2017年 
0.023 
0.455** 
 
(0.903) 
(0.012) 
2018年 
 
0.523*** 
 
 
(0.003) 
Note. The ranking by the subjective satisfaction index in 2018 is not provided by RCUQL. P values are in 
the parentheses. “*”, “**”, and “***” denote statistical significance at the 10%, 5%, and 1% levels. 
 
 


===== PAGE 33 (score 13, numeric-paren 0, table-words 1, var-words 11) =====

32 
 
Online Appendix: 
  
Table A1 Housing attributes of owner-occupied units (housing price equation) 
Variable 
Description 
Mean 
Standard 
deviation   
Number of rooms 
 
2.473 
0.917 
Floor area (sq. 
meter) 
 
79.763 
36.431 
Building age 
year 2005 minus completed year 
13.382 
10.554 
Residential  
Dummy variable, 1= houses for both residential 
and business 
0.015 
0.121 
Building type 
Dummy variable, 1= storied building 
0.877 
0.328 
Structure 
Dummy variable, 1= reinforced-concrete 
0.448 
0.497 
Tap water 
Dummy variable, 1= with tap water 
0.947 
0.224 
Kitchen 
Dummy variable, 1= with kitchen 
0.983 
0.128 
Cooking fuel 
Dummy variable, 1= with gas/electricity 
0.862 
0.345 
Toilet 
Dummy variable, 1=with toilet 
0.940 
0.237 
Bathroom 
Dummy variable, 1= with bathroom 
0.745 
0.436 
Commodity 
housing 
Dummy variable, 1= commodity housing 
0.401 
0.490 
Public housing 
Dummy variable, 1= new public housing 
0.150 
0.357 
Old public 
housing 
Dummy variable, 1= old public housing 
0.449 
0.497 
Sample size: 421,660. We create the following categorical variables from the data: urban status 
(city, county, town); building structure (reinforced-concrete, mixed, other); number of floors (one, 
two to six, seven or more floors); building age (less than 5, 5~10, 10~20, more than 20 years); 
floor area (less than 60, 60~75, 75~90, more than 90 square meters). Combining these categories, 
we create 3×3×3×4×4=432 groups for all housing units and include these group fixed effects in 
the housing hedonic model.