Page 74 - Read Online

Hu et al. Carbon Footprints 2024;3:16 https://dx.doi.org/10.20517/cf.2024.28 Page 7 of 16

distributions. The interaction variable TREAT is set up as follows: if a city establishes a relevant industrial
i,t
park, TREAT is 1; otherwise, it is 0. Furthermore, to ensure the reasonableness of the analysis results, this
i,t
study defines that if industrial parks are established in July and later in a year, then the time is labeled the
next year; if the parks are established before July, then the time is labeled the current year. According to the
“Statistical Table of Administrative Divisions of the People’s Republic of China” published by the Ministry
of Civil Affairs of the People’s Republic of China in 2022, there are a total of 333 prefecture-level
administrative units in the Chinese mainland. For completeness of the data for analysis, this paper selects
204 prefecture-level cities, and the study period is from 1999 to 2019. For the list of parks, the ETDZ data
come from the website of the Ministry of Commerce of the People’s Republic of China , the EIDP data
[29]
come from the website of the Ministry of Ecology and Environment of the People’s Republic of China [30-33] ,
and the LCIP data come from the website of the Ministry of Industry and Information Technology of the
People’s Republic of China [34,35] .

Control variables
Several control variables are included, based on literature on carbon emissions. Specifically, drawing on
Chen et al., energy utilization efficiency (GDP per unit of energy consumption in CNY/ton of coal
equivalent) is included . Average investment level (ratio of total fixed asset investment to GDP), industrial
[36]
development (logarithm of total industrial output value), economic development (GDP per capita), and
population density (resident population to area ratio) are also considered. The average investment level is
expressed as the ratio of total investment in fixed assets of a city to its GDP, industrial development is
represented using the logarithm of the total industrial output value, and economic development is indicated
by the GDP per capita, with the data obtained from the Statistical Yearbook of China’s Cities (1999-2019).
Population density is given by the ratio of the resident population of prefecture-level cities to the area of
prefecture-level cities, where the data on the resident population come from the China City Statistical
Yearbook and the data on the area of cities from the CEIC database.

Intermediary variables
Building on previous research [22,25] , mediators usually involve energy and technology development. This
study considers urban industrial structure, a variable that may influence city carbon emissions due to
varying industry-specific intensities. The mediator variable is defined as the ratio of secondary to tertiary
industry GDP, and the data are from the China City Statistical Yearbooks.

Data processing
For the 333 prefecture-level cities in mainland China, we selected 204 cities based on the availability and
completeness of data over the study period (1999-2019). These 204 cities exhibit a reasonably balanced
distribution across different regions of China, making it possible to capture the geographical and economic
diversity necessary for a comprehensive analysis. We checked that the selected cities exhibited consistency
between variables, including carbon emissions, industrial park establishment, and other control variables,
such as GDP and population density. The selected cities include those with established national-level
industrial parks, which are the focus of this study. Overall, the sample includes a diverse range of cities in
terms of geography, economic development, and industrial structure, which would allow a comprehensive
evaluation of how industrial parks affect carbon emissions.

The time frame of 1999 to 2019 was used, aiming to cover a sufficiently long period for analyzing the effects
of the various types of industrial parks (ETDZ, EIDP, LCIP), as these parks were introduced and/or
established at different times. This period provides a potentially adequate window for observing both the
short-term and long-term effects of industrial park policies on carbon emissions.