# RECORD 2075796388c43cd5aedc64728dd832af194aa20d
# TITLE: Effects of air quality on house prices: evidence from China’s Huai River Policy
# YEAR: 2020.0  VENUE: nan  DOI: 10.1080/00779954.2020.1827014
# AUTHORS: Xinghua Liu; Qiang Li; S. Chand; K. Sharpe
# PAGES: 16  CANDIDATE TABLE PAGES (1-based): [7, 10, 12, 13, 14]
# ABSTRACT: We employ a regression discontinuity (RD) design to estimate the impact of air pollution on house prices across a river that demarcates regions with and without coal-fired heating resulting from the Huai River Policy. This policy was decreed by the Chinese government in the 1950s and mandated the burning of coal for indoor heating at subsidised prices north of the Huai River. Employing quasi-experimental variation in particulate matter of 10 micrometres or less in aerodynamic diameter (PM10) generated by this arbitrary policy and a regression discontinuity (RD) design based on distance from Huai River, we estimate the local average treatment effect (LATE) to provide new evidence on the capitalisation of PM10 air pollution into house values. By using panel data covering 30 large cities on either side of the river for the period 2006–2015, we found that 1 µg/m3 (micrograms per cubic metre) reduction in average PM10 is associated with an approximately 1% increase in house prices. The results are robust to using parametric and nonparametric estimation methods, adjustment to a rich set of covariates, and using a subsample excluding first-tier cities.


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

Effects of air quality on house prices: evidence from China’s
Huai River Policy
Author:
Liu, X; Li, Q; Chand, S; Sharpe, K
Publication details:
New Zealand Economic Papers
v. 55
Chapter No. 1
pp. 52 - 65
0077-9954 (ISSN); 1943-4863 (ISSN)
Publication Date:
2021-01-01
Publisher DOI:
https://doi.org/10.1080/00779954.2020.1827014
License:
https://creativecommons.org/licenses/by-nc-nd/4.0/
Link to license to see what you are allowed to do with this resource.
Downloaded from http://hdl.handle.net/1959.4/unsworks_72524 in https://
unsworks.unsw.edu.au on 2026-10-20


===== PAGE 7 (score 16, numeric-paren 0, table-words 3, var-words 7) =====

NEW ZEALAND ECONOMIC PAPERS
5
Figure 2. The comparison of PM10 and house prices in northern and southern Chinese cities.
Notes: This is a provincial map of provinces of mainland China as well as Taiwan Province. The PM10 levels and house prices in
the capital city of each province are shown in the map. The 30 cities are matched with the provinces in which they are located
except for Tibet, Taiwan, Hong Kong and Macau due to unavailability of data. The unit for PM10 and house prices is μg/m3 and
yuan/m2, respectively. Data on PM10 are from China Environment yearbook and data on house prices are from China Real Estate
Statistical Yearbook.
Table 1. Summary statistics for key variables.
Variable
Observations
Mean
Standard deviation
Min
Max
PM10 (μg/m3)
300
102.99
31.14
30
305
Real house prices(yuan/m2)
300
5916.31
4243.11
1541
49,659.82
Per capita real GDP(10,000 yuan)
300
6.537922
5.270577
1.221441
70.41043
Real wage(yuan)
297
45,112.18
27,029.21
16,911
39,8203.3
Fiscal surplus (100 million yuan)
299
–152.18
248.85
–2315.93
601.83
Population density(people/km2)
299
1617.66
1373.97
223.31
11,449.3
Per capita floor space (m2)
300
8.06
4.32
0.97
24.81
1 (Coastal city)
300
0.13
0.34
0
1
Students enrolled at higher education
300
433,557.1
225,334.1
10,280
1,043,221
No. of hospitals owned per 10,000 people
299
0.53
0.42
0.23
4.70
1(North of Huai River)
300
0.5
0.50
0
1
Latitude distance from Huai River
300
0.57
7.06
-13.7
15.08
Notes: The entries are calculated across 30 cities during the 2006–2015 period.
north are generally exposed to substantially higher level of PM10 concentrations compared to those
in the south. This is consistent with the claim that air pollution is more severe in northern Chinese
cities. However, the figure fails to reveal some of the systematic patterns of the spatial distribution of
housing price across the river line.
Table 1 reports summary statistics for the key variables in this analysis. China’s mean value for
PM10 concentrations was 102.99 μg/m3 over the period 2006–2015. For the purpose of comparison,
we calculate the US PM10 concentrations for the same period based on data reported by the United
States Environmental Protection Agency (EPA). The figure is 68.58 μg/m3 for the US, making China’s
PM10 concentrations larger by a factor of 1.5 to US levels.


===== PAGE 8 (score 5, numeric-paren 0, table-words 1, var-words 3) =====

6
X. LIU ET AL.
4. Econometric model
We use both the conventional and the RD approach to estimate the association between PM10 and
house prices. The basic idea and structure of this section follows closely that of Chen et al. (2013)
and Ebenstein et al. (2017). The conventional strategy uses the city fixed effects regression model to
estimate the following equation:
HPit = β0 + β1PMit + Xit + γi + εit
(1)
where i denotes a city and t indexes a year, and
n
i=1
γi = 0. PMit is the PM10 concentration in city i
in year t, Xit is a vector of observed covariates used to control for other determinants of house prices
other than air quality, γi are city fixed effects, and εit is an error term. The outcome variable is the
natural logarithm of the average residential house prices in city i in year t. β1 is the coefficient of
interest for Equation (1).
The estimation of Equation (1) faces problems of endogeneity, with the risk that the estimates of β1
are biased. Recall that the validity of the conventional approach depends on the assumption that the
adjustment of covariates rules out all the confounding factors. Although we have controlled for a set
of observed characteristics for which the requisite data is available, unobserved determinants of house
prices may also co-vary with PMit. The use of panel data analysis allows us to eliminate the effects
of any time invariant unobservables by inclusion of city fixed effects, but this is far from sufficient to
address potential problems of endogeneity. Measurement error of pollution concentrations is a serious
risk to potential endogeneity as there is evidence that air quality readings tend to be underreported
by the officials. Consequently, the estimates of the impact of air quality on the price of housing from
Equation (1), our parameter of interest, may be biased.
The second approach leverages the regression discontinuity (RD) design where we exploit the
exogenous variation caused by the Huai River Policy to estimate its impact on both PM10 concentra-
tions and house prices. The development of RD identification is based on the idea that the treatment
status is determined by a continuous variable crossing a cut-off due to the imposition of some arbi-
trary rule. This feature provides a local randomisation for estimating treatment effects. RD design has
emerged almost simultaneously in a wide range of disciplines but has only recently become prevalent
in applied econometrics. We exploit an RD design that has a break in the spatial dimension, whereby
a treatment is applied along a geographical boundary. This is done using two different approaches.
First, we use a sharp regression discontinuity (sharp RD) design to test any discontinuity in PM10 and
house prices across the river. Under this approach, the Huai River Policy is treated as a treatment vari-
able and its impacts on the two endogenous variables are estimated. In this case, the treatment status
is a deterministic and discontinuous function of the running variable (the degree distance north of
the Huai River boundary), making the estimates a sharp RD design. Second, we use a fuzzy regres-
sion discontinuity (fuzzy RD) design to estimate the impact of PM10 on house prices. A fuzzy RD
is where the treatment status is no longer deterministically related to the threshold-crossing rule. In
our context, the PM10 pollutant exists on both sides of the river but the probability of PM10 expo-
sure increases considerably across the river, making our study a fuzzy RD in nature. Consequently,
the discontinuity becomes an instrumental variable for PM10. It is worth noting that the validity of
our RD design rests on the assumption that any unobserved determinants of PM10 and house prices
change smoothly as they cross the river line. If this identification assumption is true, adjustment for
flexible polynomial terms in degrees latitude north of the river (i.e. local linear regressions (LLR))
will eliminate all sources of bias and hence allow us to draw causal inferences.
Of course, one may argue that heating facilities might increase housing construction costs and
therefore raise house prices in Northern China. Since we hypothesise that, holding all else constant,
the housing price will be higher for cities just to the south of the river, our estimations in this paper
are expected to be downwardly biased.


===== PAGE 10 (score 42, numeric-paren 17, table-words 8, var-words 5) =====

8
X. LIU ET AL.
5. Empirical results
Table 2 reports the summary statistics for several key determinants (PM10 and control variables) of
house prices and lends support to the validity of our RD design. The first two columns report the
means and the standard derivations in cities located north and south of the Huai River. Column [3]
presents the mean difference between the south and the north along with the standard errors.
The table shows that PM10 exposure differs significantly among northern and southern cities.
Concentrations of sulphur dioxide (SO2) and nitrogen dioxide (NO2) are also reported in the China
Environmental Yearbook. Indeed, Table 2 demonstrates that the differences among these pollutants
are also statistically significant among southern and northern cities. This difference is reassuring since
sulphur dioxide and nitrogen dioxide are lighter and therefore can travel further than PM10, which is
our measure of environmental dis-amenity on house prices caused by living north of the river.
Table 2 also suggests that the mean value for both students enrolled at higher education institutions
and the number of hospitals owned per 10,000 people are statistically different across northern and
southern cities. To gather evidence on the validity of the RD estimates, we check for discontinuities in
these two variables that may confound the results. Figure A1 in the appendix shows the relationship
between degrees north of the Huai River and these two control variables. The figure indicates that
these control variables do not show significant jumps at the cut-off, suggesting no breaks around the
cut-off.
5.1. Estimates of the effect of the Huai River Policy on PM10 and house prices
First, we graphically show the effect of the Huai River Policy on air pollution in Figure 3. The dots in
Figure 3 represent the mean of PM10 concentrations falling within each bin. The fitted values repre-
sented by the solid line are from a kernel weighted 2nd-order polynomial regression fit of the PM10
on the score. All control and treated observations are included in the control and treated fit, respec-
tively. The plot reveals a positive jump at the cut-off: the PM10 concentration seems to be higher in
cities north of the river than those to the south.
Figure 4 is a scatterplot of house prices of various cities against their degree distance north of the
river. A simple linear regression was also run, and we plotted the fitted values on top of the scatterplot.
The slight monotonic decline of the fitted line indicates a negative correlation between distance in
latitude from the Huai River and the house prices.
In addition to the graphical evidence, Table 3 reports the empirical results from estimating Equa-
tions (2a) and (2b). All the estimates are adjusted for the full set of covariates reported in the data
sources section. Column [1] reports the fixed effect estimates of the coefficient on a dummy variable
Table 2. Differences in key determinants of house prices at the Huai River.
Variable
North
South
Difference in means
PM10 (μg/m3)
119.4133
86.56
−32.85333***(3.058324)
SO2 (mg/m3)
0.0508667
0.0339867
−0.01688***(0.0020688)
NO2 (mg/m3)
0.04354
0.0416533
−.00018867*(0.0013518)
Real House Price (yuan/m2)
5442.61
6390.02
−947.41**(487.70)
Real House Price (log)
8.43
8.62
0.19*** (0.06)
Per capita real GDP(10,000 yuan)
6.449367
6.626477
0.17711(0.6095278)
Real wage(yuan)
45047.3
45,176.63
129.3319(3142.108)
Fiscal surplus (100 million yuan)
−148.8867
−155.4967
−6.610044(28.82844)
Population density(people/km2)
1652.998
1582.079
−70.91901(159.132)
Per capita floor space (m2)
7.937815
8.188631
0.2508162(0.4990873)
Students enrolled at higher education
379,014.2
488,100.1
109,085.9*** (25,285.29)
No. of hospitals per 10,000 people
0.598874
0.4537101
−0.1451639*** (0.0477961)
***Significant at the 1% level.
** Significant at the 5% level.
* Significant at the 10% level.


===== PAGE 11 (score 3, numeric-paren 0, table-words 1, var-words 1) =====

NEW ZEALAND ECONOMIC PAPERS
9
Figure 3. Fitted value from RD regression of PM10 exposure on distance from the Huai River.
Figure 4. A linear fit from regression of housing price on distance from the Huai River.
indicating whether it is located to the north of the river after controlling for a third-order polyno-
mial term in distance from the river as well as its interaction with the north river dummy using the
full sample. Columns [2] and [3] apply the parametric and non-parametric RD approach to estimate
these two equations. Column [2] uses the subsample within 7° latitude of the Huai River which was
imposed manually as our discontinuity sample. Conversely, in Column [3], the estimates are based on
the local linear regression with a triangular kernel and optimal bandwidth (h) selected by the method
prescribed by Imbens and Kalyanaraman (2012).
Overall, Table 3 presents statistical evidence of a considerable rise in PM10 and a decrease in hous-
ing price at the Huai River divide, which is similar to what we have presented in Figures 3 and 4. At the


===== PAGE 12 (score 29, numeric-paren 16, table-words 5, var-words 2) =====

10
X. LIU ET AL.
Table 3. RD estimates of the impact of the Huai River Policy.
Outcome
[1]
[2]
[3]
PM10
19.1157**
41.251***
47.47836*
(8.64)
(7.7153)
(28.10 87)
Housing price (log)
–0.272**
−0.42026***
–0.4306474***
(0.107)
(0.05896)
(0.1575027)
RD type
–
polynomial
Local Linear Regression(LLR)
Polynomial function
Third
Linear
Sample
All
7°
***Significant at the 1% level.
** Significant at the 5% level.
* Significant at the 10% level.
Table 4. Comparing fixed effect and RD estimates of PM10’s impact on house prices.
Outcome
[1]
[2]
[3]
Housing price (log)
–0.001
–0.0101651*
–0.0083*
(0.0008)
(0.0058542)
(0.00467)
Estimation method
Fixed effect
IV
IV
RD Type
–
Local Linear Regression(LLR)
Local Linear Regression(LLR)
Sample
All
7°
***Significant at the 1% level.
** Significant at the 5% level.
* Significant at the 10% level.
threshold, the PM10 increase by 19 μg/m3 as indicated by conventional method, whereas in the RD
design, the increase is substantially larger, by 41–47 μg/m3, respectively. In comparison, the log term
of house prices declines as we move across the border by 27% and 42% for the conventional and para-
metric RD approach. The estimates using nonparametric approach are of slightly larger magnitude:
the estimated decline in log (house prices) of cities north of the river is 43%.
5.2. Estimates of the effect of PM10 on house prices
Table 4 reports the estimated effect of PM10 on house prices based on a variety of approaches. Col-
umn [1] reports the results from application of the conventional fixed effect method, which is used as
a baseline model to compare the estimates with those based on RD and 2SLS approaches. Specifically,
the fixed effect estimator of the coefficient of PM10 exposure on house prices is statistically insignifi-
cant and marginal in size: a 1 unit increase in PM10 is correlated with a 0.1% increase in housing price,
but as noted above the parameter estimate is insignificantly different from zero. Additionally, fuzzy
RD using a restricted sample based on the data-driven approach and bandwidth selected manually
are presented in Columns [2] and [3], respectively. They are statistically significant at 10% significant
level. Overall, the results indicate that the fuzzy RD approach delivers a larger coefficient in magni-
tude compared to conventional estimates. This lends support to the proposition that omitted variables
may have attenuated the estimate of the critical parameter, and the true value of clean air.
6. Robustness
We examine the robustness of our results by (i) using placebo cut-off lines; and (ii) testing whether our
results were affected qualitatively by the decisions made in our paper along several dimensions, such
as adjustment for covariates, sample selection, functional forms of our models, and kernel weighting
methods.
First, we repeat the estimation of fuzzy RD using placebo cut-offs. While we know the Huai
River line is the true cut-off of interest due to the indoor heating policy in the 1950s, we consider


===== PAGE 13 (score 28, numeric-paren 14, table-words 6, var-words 1) =====

NEW ZEALAND ECONOMIC PAPERS
11
Figure 5. RD placebo tests on the impact of PM10 on house prices.
Table 5. Comparing the fixed effect and regression discontinuity estimates of PM10’s impact on house prices.
2SLS Polynomial RD
(within 7 degrees)
Fixed effect
Fixed effect (within
7 degrees)
Local Linear Regressions
[1]
[2]
[3]
[4]
[5]
[6]
[7]
[8]
House prices
–0.001
–0.0011
–0.0197***
–0.0083*
–0.0204***
–0.0087*
–0.0237***
–0.0116*
(0.0008)
(0.0008)
(0.00557)
(0.00467)
(0.00582)
(0.0047)
(0.00757)
(0.0066)
Obs.
294
200
196
200
196
200
196
Controls
Yes
Yes
No
Yes
No
Yes
No
Yes
Kernel
Triangle
Triangle
Epanech.
Epanech.
Uniform
Uniform
***Significant at the 1% level.
** Significant at the 5% level.
* Significant at the 10% level.
demarcations close to the Huai River as the cut-offs for the placebo tests. We redraw a window of
four displacements at 0.5°-latitude intervals north and south of the Huai River across China as well
as at the actual Huai River (which is equivalent to the 0° displacement) and re-estimate our key coef-
ficient of interest. The results from these falsification exercises are shown in Figure 5. According to
Figure 5, the coefficient is only statistically significant at the actual Huai River. This suggests that the
previous findings reported in this paper are a result of the Huai River Policy rather than an artefact
of the application of the RD approach to the data.
In addition, we examine the robustness of our results based on tests from several dimensions,
including adjustment for covariates, functional forms of our models, and kernel weighting methods.
In Table 5, we expand the results of Table 4 by including a full set of parametric estimates with and
without adjusting for the covariates, and local linear estimates using a variety of kernel weighting
methods. The entries in columns [1] and [4] are identical to those in Table 4, which are adjusted
using the full set of covariates and a linear polynomial in distance to the Huai River. Columns [3],
[5] and [7] report the estimated coefficients excluding the covariates. Using a triangular kernel, the
entries in column [4] match the local linear results in Table 4, while estimates using an Epanechnikov
and uniform kernel are presented in columns [5] to [8], respectively.


===== PAGE 14 (score 32, numeric-paren 15, table-words 5, var-words 6) =====

12
X. LIU ET AL.
Table 6. Robustness checks of choice of functional form for latitude.
R¸estricted sample (within 7 degrees)
Linear
Quadratic
Cubic
Quartic
Quintic
[1]
[2]
[3]
[4]
[5]
Housing price
–0.00833*
–0.0081**
–0.01204***
0.08037***
−0.01249***
(0.00467)
(0.00333)
(0.00284)
(0.01939)
(0.00081)
***Significant at the 1% level.
** Significant at the 5% level.
* Significant at the 10% level.
Table 7. Comparing fixed effect and RD estimates of PM10’s impact on house prices (subsample excluding first-tier cities).
Outcome
[1]
[2]
[3]
Housing price (log)
–0.0010686
–0.0063145*
–0.00743*
(0.0007496)
(0.0036558)
(0.00489)
Estimation method
Fixed effect
IV
IV
RD Type
-
Local Linear Regression(LLR)
Local Linear Regression(LLR)
Sample
All
7°
Moreover, we report estimates from a number of specifications to check if the results are sensi-
tive to the order of the polynomial used. In Table 6, we adopt alternative approaches to implement
the parametric RD design by using various functional forms, varying from a linear polynomial inter-
acted with a North dummy to a quartic polynomial subsequently as one moves from left to right. We
consider the restricted sample of cities within 7 degrees latitude of the Huai River in columns [1] to
[5]. For example, columns [1], [2] and [3] include degree distance to the Huai River, its square, and its
cube, and each term’s interaction with a North dummy in the discontinuity sample. The results show
that except for column [4], the figures are all qualitatively similar to those presented in the baseline
specifications.
Noticeably, two out of three first-tier cities of our sample (Shanghai, Guangzhou and Beijing) are
located south of the Huai river (see Appendix for the location of our sample cities relative to the Huai
river), and they all have extremely high house prices. This may affect the major finding of the paper,
i.e. the negative correlation between air quality and housing price. To account for this big city effect
which may drive our findings, we conduct the same analysis as the baseline using a subsample that
excludes Beijing, Shanghai and Guangzhou. The results are presented in Table 7. As can be seen, the
results remain qualitatively similar to those of Table 4. This may, to some extent, reduce the concern
about the big city effect.
7. Conclusion
Having good estimates of the value of clean air is critical to the cost–benefit analysis of policy inter-
vention regarding environmental protection. In this paper, we draw on a unique intervention with
respect to the use of coal-fired heating amongst households in China in order to estimate the elastic-
ity of house prices with respect to concentrations of PM10. The Huai River Policy generated a stark
difference in air quality in southern and northern China. Our analysis exploits the structure of this
unique Chinese Huai River Policy divide to provide new evidence of the extent to which amenities
from the level of pollution is reflected in house prices.
We employ a regression discontinuity (RD) design to estimate the impact of air pollution on house
prices across a river that demarcates regions with and without coal-fired heating resulting from the
Huai River Policy. This policy – which was decreed by the Chinese government in the 1950s – enabled
the burning of coal at subsidised prices for indoor heating only to the north of the Huai River. Using a
quasi-experimental variation in particulate matter of 10 micrometres or less in aerodynamic diameter
(PM10) and a regression discontinuity (RD) design based on distance from Huai River, we estimate


===== PAGE 15 (score 14, numeric-paren 0, table-words 0, var-words 14) =====

NEW ZEALAND ECONOMIC PAPERS
13
the local average treatment effect (LATE) of PM10 air pollution on house prices. By using panel data
of 30 large cities on either side of the river for the period 2006–2015, we found that 1 μg/m3 reduction
in average PM10 results in a 1% increase in the price of housing.
The findings also confirm the importance of instrumenting for local air pollution to reduce prob-
lems of endogeneity. Our econometric analysis reveals that conventional fixed effect regressions tend
to underestimate the value of clean air. Accounting for problems of endogeneity improves the esti-
mates, providing statistically significant parameter estimates of the effect of PM10 on house prices.
These estimates of the economic value of clean air are robust to adjustment in covariates and are less
sensitive to model specification.
Extrapolation from US-based studies suggests that the costs of poor air are substantial in terms of
health effects and welfare in general. This paper shows that the value of clean air in China exceeded
that for the US. A 1 μg/m3 reduction in TSPs results in a 0.2–0.4% increase in mean housing values
in the US (Chay & Greenstone, 2005), compared to an approximately 1% increase in housing price in
China based on our estimates. Our findings therefore suggest that the value of clean air may increase
with the level of pollution.
Notes
1. Total suspended particles (TSP) measures the mass concentration of particulate matter in the air. Within TSP, PM10
stands for particles with a diameter of 10 μm or less, and PM2.5 stands for those with a diameter of 2.5 μm or less.
Particulates that are 10 μm or greater are filtered and generally do not enter the lungs. Particulates smaller than
10 μm are likely to enter the lungs. Particulate matter that is smaller than 2.5 μm (PM2.5) can enter into the Alveoli
where gas exchange occurs. Throughout the world, ambient monitoring now focuses on PM10 and PM2.5.
2. We exclude Lhasa (capital city of Tibet) in our sample and restrict the sample to the year 2006–2015 due to
availability of certain covariates.
Disclosure statement
No potential conflict of interest was reported by the author(s).
ORCID
Satish Chand
http://orcid.org/0000-0003-3524-1878
References
Almond, D., Chen, Y., Greenstone, M., & Li, H. (2009). Winter heating or clean air? Unintended impacts of China’s
Huai River Policy. American Economic Review, 99, 184–190.
Ambrey, C. L., Fleming, C. M., & Chan, A. Y.-C. (2014). Estimating the cost of air pollution in South East Queensland:
An application of the life satisfaction non-market valuation approach. Ecological Economics, 97, 172–181.
Angrist, J. D., & Lavy, V. (1999). Using Maimonides’ rule to estimate the effect of class size on scholastic achievement.
The Quarterly Journal of Economics, 114, 533–575.
Angrist, J. D., & Pischke, J.-S. (2008). Mostly harmless econometrics: An empiricist’s companion. Princeton, NJ: Princeton
University Press.
Bayer, P., Keohane, N., & Timmins, C. (2009). Migration and hedonic valuation: The case of air quality. Journal of
Environmental Economics and Management, 58, 1–14.
Besley, T., & Mueller, H. (2012). Estimating the peace dividend: The impact of violence on house prices in Northern
Ireland. American Economic Review, 102, 810–833.
Black, S. E. (1999). Do better schools matter? Parental valuation of elementary education. The Quarterly Journal of
Economics, 114, 577–599.
Brunekreef, B., & Holgate, S. T. (2002). Air pollution and health. The Lancet, 360, 1233–1242.
Chay, K. Y., & Greenstone, M. (2005). Does air quality matter? Evidence from the housing market. Journal of Political
Economy, 113, 376–424.
Chen, Y., Ebenstein, A., Greenstone, M., & H, L. I. (2013). Evidence on the impact of sustained exposure to air pol-
lution on life expectancy from China’s Huai River Policy. Proceedings of the National Academy of Sciences, 110,
12936–12941.