Housing prices and carbon emissions: a dynamic panel threshold model of 60 Chinese cities

ABSTRACT

In the past two decades, both housing prices and carbon emissions have risen dramatically in China. However, it is unclear whether there is a nexus between them. This study employs a dynamic panel threshold regression model using residents' income levels as the threshold variable to investigate the relationship between housing prices and per capita carbon emissions. By considering 60 Chinese housing price-monitored cities, a threshold effect of income on carbon emissions was identified with a threshold value of 2.448, which divided the full sample into two regimes to further analyse the impacts of housing prices on per capita carbon emissions at each income level. The empirical results show that housing prices have a slight but positive effect on per capita carbon emissions both at two income levels. Specifically, a 10% change in housing prices can lead to a change in the same direction in per capita carbon emissions by 0.7% and 0.2% at lower and higher average income levels, respectively. However, in small cities, limited population may contribute to the negative impact of housing prices on carbon emissions. Furthermore, the results also confirm that the urbanization rate is an important factor of per capita carbon emissions.

KEYWORDS:

• Carbon emissions
• per capita disposable income
• threshold regression model
• urbanization
• housing prices

JEL CLASSIFICATION:

• A12
• C33

Acknowledgments

Financial supports from the National Natural Science Foundation of China (Grant nos.71822403 and 31961143006) and the Hubei Provincial Outstanding Youth Foundation of Natural Science (Grant no.2019CFA089) are gratefully acknowledged.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

¹ In China, the house price refers to the average sale price of commercial housing.

² The original data of housing prices are from National Bureau of Statistics and are then adjusted in 2010 constant prices. The detailed process can be found in Section 3.3.

³ In this study, it is assumed that related industries are mainly located in a local city.

⁴ The specific expression of the weight matrix can be seen in Seo and Shin (2016).

⁵ lnP represents the natural logarithm of total population (Unit: 10 thousand person).

Additional information

Funding

This work was supported by the National Natural Science Foundation of China [31961143006,71822403] and the Hubei Provincial Outstanding Youth Foundation of Natural Science [2019CFA089].