A Study on the Impact of China’s Prefabricated Building Policy on the Carbon Reduction Benefits of China’s Construction Industry Based on a Difference-in-Differences Method

Abstract

The construction industry is a significant contributor to carbon emissions in China. To effectively meet the “dual carbon” targets, several provincial regions within the country started to implement policies promoting prefabricated buildings. This study examines data from 18 provinces in China over the period from 2012 to 2021, treating the introduction of prefabricated building policies as a quasi-natural experiment. Utilizing the difference-in-differences methodology, this research assesses the impact of these policies on the carbon emission performance of China’s construction sector and evaluates the robustness of the findings. The results indicate that the prefabricated building policies positively influenced the carbon emission efficiency of the construction industry. Specifically, these policies enhance carbon emission efficiency by increasing labor productivity, optimizing the allocation of mechanical resources, and improving the utilization rate of building materials. Additionally, the effectiveness of these policies is positively correlated with the level of regional technological innovation, environmental protection efforts, and the advancement of energy structure optimization. The study concludes with several policy recommendations aimed at further enhancing the effectiveness of prefabricated building policies.

1. Introduction

Since the beginning of the 21st century, technological advancements and industrial development led to significant increases in greenhouse gas emissions, exacerbating the global greenhouse effect and resulting in frequent extreme weather events and natural disasters [1]. In response to these challenges, global attention to environmental issues intensified. In 1992, the United Nations General Assembly adopted the United Nations Framework Convention on Climate Change (UNFCCC), which aimed to reduce greenhouse gas emissions to mitigate dangerous disruptions to the climate system caused by human activities. In 2005, the Kyoto Protocol came into force, setting binding emission reduction targets for participating countries. This was followed by the Paris Agreement, which took effect in 2016, focusing on global cooperation to address climate change and limit global warming. At the end of 2020, China’s National Development and Reform Commission (NDRC) released the Action Plan for Carbon Peaking and Carbon Neutrality (2021–2025), which outlined the national goals for achieving peak carbon emissions and carbon neutrality for the first time. During the 75th session of the United Nations General Assembly, China pledged to peak carbon emissions by 2030 and achieve carbon neutrality by 2060, setting an ambitious “dual carbon” goal to completely eliminate or offset carbon dioxide emissions.

The Chinese government has long been committed to implementing carbon reduction policies. Starting in 2005, China introduced a series of new energy development plans focused on expanding clean energy sources such as wind, solar, and nuclear power. These initiatives aimed to increase the share of clean energy in the energy mix and decrease reliance on fossil fuels [2]. In 2011, China began experimenting with carbon emissions trading by establishing pilot markets in cities such as Beijing and Shanghai. This marked the country’s initial foray into carbon trading mechanisms. By the end of 2020, China had officially launched a nationwide carbon emissions trading market. This market encompasses a wide range of sectors, including electricity, heat, cement, chemicals, metals, petrochemicals, automobiles, and public buildings [3]. These policies significantly advanced China’s carbon reduction efforts, demonstrating the country’s commitment to mitigating climate change and transitioning towards a more sustainable energy system [4].

The carbon emissions of China’s construction industry in 2020 were approximately 850 million tons of carbon dioxide, accounting for over 40% of the country’s total carbon emissions [5]. Whether the Chinese construction industry can effectively reduce energy consumption became one of the main constraints on whether China can achieve the “dual carbon” goals on schedule. With the increase in China’s urbanization rate, the construction industry is expected to continue to grow to a certain extent, which will bring huge obstacles to the smooth implementation of China’s future carbon reduction plans [6,7]. Prefabricated buildings are recognized as one of the effective means to alleviate China’s current carbon emission contradiction due to their low cost, low energy consumption, and reduced material waste characteristics [8,9,10]. In 2016, the State Council of China released the “Opinions of the State Council on Promoting High-Quality Development of the Construction Industry and Building Materials Industry”. This document directed provincial authorities to provide subsidies for projects that adhere to certification standards for prefabricated green buildings. Additionally, it emphasized the promotion of technological and process innovations related to prefabricated components within the construction sector [11]. In the same year, the Ministry of Housing and Urban-Rural Development of China issued the “Action Plan for the Development of the Prefabricated Building Industry (2016–2020)”. This plan aimed to expedite the application of prefabricated materials in construction projects and set a target for the use of prefabricated technology in new buildings to exceed 30% within ten years Additionally, the Chinese government established 30 demonstration pilot cities and initiated 119 demonstration projects [12]. which clarified the guiding ideology, overall goals, and key tasks for the development of prefabricated buildings in China in the next five years and pointed out the direction for the development of the industry. In the process of promoting prefabricated buildings, whether it can effectively promote the diversification of China’s construction industry’s energy structure is also one of the possible entry points for promoting China’s low-carbon transformation and development [13,14]. Therefore, this study evaluates the carbon reduction benefits of China’s prefabricated building policies by analyzing the energy efficiency of the construction industry across various regions before and after the implementation of these policies.

This article treats the implementation of prefabricated building policies as a quasi-natural experiment and employs the difference-in-differences (DID) method to assess the impact of these policies on carbon emissions associated with the output value of the construction industry. Utilize the propensity score matching with difference-in-differences (PSM-DID) methodology to conduct a further validation of the benchmark results. Additionally, a randomized repeated experiment was conducted to perform a placebo test. Mechanism analysis was carried out to identify the factors influencing the carbon emissions related to prefabricated building policies, explore the policy’s action mechanisms, and investigate strategies to enhance the policy’s effectiveness. The study also offers policy recommendations aimed at providing guidance for policy implementation and promotion.

2. Literature Review

The prefabricated construction industry chain encompasses a supply chain network that integrates various entities such as prefabricated component production factories, design units, transportation services, and construction firms. This chain is centered around the factory production of building components and includes upstream activities such as development and design, as well as downstream processes such as transportation and assembly. Internationally, countries such as the United States, the United Kingdom, Japan, Sweden, and Singapore are recognized as leaders in advancing the development of prefabricated buildings [15,16,17]. In foreign countries, where prefabricated construction was established for a longer period, the industry achieved maturity in terms of its industrial chain, policy frameworks, and practical application of components. This maturity led to notable successes in addressing labor shortages and reducing building carbon emissions.

The factory production mode of prefabricated buildings is conducive to the rational allocation of resources and can achieve dual carbon reduction goals. Energy is considered one of the factors that have the greatest impact on buildings, and process innovation in the manufacturing stage of prefabricated components has a significant effect on improving energy efficiency [18,19]. The results of the literature review and investigation by Zhang et al. indicate that the use of prefabricated construction technology in China reduced the use of wood templates in construction equipment by 87% and water consumption by 70% [20]. Compared to traditional cast-in-place buildings, the production process of prefabricated building components can reduce hidden carbon emissions by 15.6% and operational carbon emissions by 3.2% [21,22,23], reducing energy consumption by 20.49% [24,25,26]. Liu and Ying found through their research on a typical prefabricated building in Shanghai that adopting prefabricated building technology can effectively reduce the use of building materials and the generation of solid waste [27]. By establishing a calculation model for energy consumption and waste flow, Chen et al. found that the use of prefabricated processes reduced material usage and energy consumption by 20%, as well as the production of greenhouse gases, such as carbon dioxide, by 20% [24]. The above results indicate that prefabricated construction technology has a significant effect on reducing environmental benefits such as energy consumption, carbon emissions, and even acidic gases and water in the construction industry.

At present, research on carbon emission measurement methods for the entire industry is mainly divided into two categories [28,29,30]: One type is based on the production end, which only considers the carbon emissions of the product production point (including exports) and no longer considers the product usage end. The other type is based on the consumer end, which is exactly opposite to the production end, that is, only considering the carbon emissions of the product’s consumer end (including exports), without considering the production end [31,32]. Currently, what is more widely recognized in academia is the consumption-based carbon emission calculation, such as Zhang et al., who used the consumption-based method to complete the carbon emission accounting for Chinese cities [33]. Steininger et al. argued through a comparative study of two methods that consumer-based carbon emission policies are more conducive to improving the cost-effectiveness and fairness of industries [31]. Guan et al. believe that consumer-based carbon emission accounting can effectively alleviate global climate issues [34]. The traditional research methods for carbon emissions in the construction industry mainly include process analysis, the input–output method, and mixed method [35,36]. Process analysis is based on the energy consumption and other characteristics of the entire life cycle of buildings, combined with carbon emission factors based on the Intergovernmental Panel on Climate Change (IPCC) and the China Life Cycle Database (CLCD) [22]. We can accurately calculate the carbon emissions value throughout the entire lifecycle of the building [37,38]. Regarding the output method, some scholars used input–output analysis to study the carbon emissions and energy consumption of buildings in Ireland, and found that infrastructure generates approximately 11.7% of carbon emissions [39]. The mixed rule is a method that combines process analysis and input–output methods to produce results. Considering the complexity of the construction industry involving multiple industrial sectors and the need for research, this article adopts a combined approach based on consumer end and construction process analysis to study the carbon emission accounting of prefabricated construction industry. This means that the entire construction industry process, including materials, manpower, transportation, and other aspects, consumes a large amount of energy such as oil, electricity, and coal, which are all accounted for.

The research methods for calculating energy consumption are also becoming more mature, mainly divided into single factor energy efficiency (SFEE) and total factor energy efficiency [40,41]. Single factor measurement considers energy itself as the sole input factor [42], and all factor energy efficiency measures energy efficiency by considering the actual energy consumption of the entire building [43]. The current mainstream analysis of total factor energy consumption is the economic heat index, which refers to the ratio of economic output to energy consumption of an industry as an indicator of energy efficiency. The method of total factor energy efficiency analysis needs to fully consider various outputs and inputs of the industry, including not only energy consumption, but also many industrial factors such as capital and labor [44]. At present, experts in the industry have different opinions on the accurate selection of input and output factors, and consensus is yet to be formed, which is still in the exploratory stage [45]. Compared to this, the measurement process of single-factor energy efficiency is more direct and simple. Therefore, this article adopts the ratio of economic output to energy consumption in the single factor energy efficiency research method as the method to measure the energy efficiency of the construction industry.

In the field of analyzing factors affecting carbon emissions, Li and Liu et al. believe that energy, building materials, and machinery are the main sources of carbon emissions in the construction industry [46]. Agi, Diabat et al. divided the factors that affect carbon emissions in the construction industry into five major factors: government, market, technology, society and culture, and supply chain coordination, and emphasized the importance of policies [47,48,49]. Sun Shao nan et al. divided the influencing factors into five aspects when considering the effect of prefabricated carbon reduction: design planning, building materials, energy use, building environment, and construction organization [50]. Liang et al. identified urbanization, per capita GDP, technology and equipment ratio, energy consumption structure, innovation support, environmental regulation, industrial contribution rate, and industrial concentration as external factors affecting carbon emissions in the construction industry [4]. Du, Zhou, and others believe that the mechanical equipment rate and technological innovation of construction enterprises are important factors affecting building carbon emissions [51]. In addition, as prefabricated buildings are an emerging technology, the introduction of relevant policies will to some extent promote technological innovation in relevant regions, and technological innovation will also have a significant impact on building carbon emissions. Therefore, this study considers regional technological innovation as one of the factors that may affect carbon emissions in the construction industry.

It is evident from the above results that while current research methods and systems for studying carbon emissions in the construction industry are relatively comprehensive, there remains a significant gap in the systematic evaluation of the overall effectiveness of China’s prefabricated building policy system. This study introduces several key innovations compared to prior research: (1) Provincial-Level Systematic Evaluation: It is the first comprehensive assessment of the overall energy consumption impacts of prefabricated building policies on the construction industry from a provincial perspective. This approach addresses the existing gap in the quantitative evaluation of these policies’ environmental benefits in China. (2) Advanced Methodology: By employing the difference-in-differences (DID) model and utilizing provincial panel data, the study effectively minimizes the impact of randomness and micro-level specificity. This enhances the objectivity of the results, providing a more reliable basis for policy reference. (3) Confirmed Benefits and In-Depth Analysis: The study validates the positive effects of prefabricated building policies and offers a systematic analysis of the influencing factors. These insights contribute practical recommendations for the further refinement and promotion of these policies, offering valuable guidance for decision-makers.

3. Methods and Data

The research framework of this article is shown in Figure 1.

3.1. Difference-in-Differences Model

The difference-in-differences (DID) model is employed in this study to assess the impact of prefabricated building policies on energy consumption and carbon emissions in the construction industry. By treating these policies as a quasi-natural experiment, the study compares provinces that implemented the policies earlier (e.g., Jiangsu, Zhejiang, Shandong, Hubei, and Hebei) with those that adopted them later or not at all (e.g., Hunan, Shanxi, Henan). This approach helps isolate the effect of the prefabricated building policies from other factors, excluding regions influenced by overlapping carbon trading policies. For prefabricated building policies and carbon emissions in the construction industry, the double difference model of this article is constructed as follows:

$${\mathrm{A}}_{\mathrm{i}\mathrm{t}}={\mathsf{\beta }}_0+{\mathsf{\beta }}_1{\mathrm{M}}_{it}\ast {\mathrm{T}\mathrm{i}\mathrm{m}\mathrm{e}}_t+\sum {\mathsf{\beta }}_j{X}_{it}+{\mathsf{\mu }}_i+{\mathsf{\gamma }}_t+{\mathsf{\epsilon }}_{it}.$$

(1)

Among them, ${\mathrm{A}}_{\mathrm{i}\mathrm{t}}$ represents the natural logarithm of the carbon emission figures of the construction industry for each group of samples. β₀ represents a constant, β₁ and β_j represent the efficiency coefficient of prefabricated building policies, and M_it is the dummy variable we constructed to represent the fixed effects of provinces. If a province implements prefabricated building policies, M_it = 1 represents the experimental group. If the province has not implemented a prefabricated building policy, M_it = 0, indicating a control group. ${\mathrm{T}\mathrm{i}\mathrm{m}\mathrm{e}}_t$ is the time dummy variable we constructed to represent time-fixed effects. When time is greater than or equal to 2016, ${\mathrm{T}\mathrm{i}\mathrm{m}\mathrm{e}}_t=1$; and when time is less than 2016, ${\mathrm{T}\mathrm{i}\mathrm{m}\mathrm{e}}_t$ = 0. The result of ${\mathrm{M}}_{it}\ast {\mathrm{T}\mathrm{i}\mathrm{m}\mathrm{e}}_t$ indicates whether the region implemented a prefabricated building policy. ${\mathsf{\mu }}_i$ is used to represent fixed effects in provinces. ${\mathsf{\gamma }}_t$ is used to represent fixed time effects. ${\mathsf{\epsilon }}_{it}$ represents the random error term. β₁ is the core factor for studying whether prefabricated building policies can promote carbon emissions reduction in the construction industry. ${\mathrm{A}}_{\mathrm{i}\mathrm{t}}$ indicates the net effect of prefabricated building policies on carbon emissions in the construction industry.

3.2. Mechanism Analysis

Due to the fact that prefabricated construction technology itself is a new technology directly aimed at the construction industry, the promotion of prefabricated construction policies inevitably promotes technological innovation and mechanical performance updates in the construction industry. At the same time, technological innovation and mechanical performance updates will have a huge impact on the reduction in carbon emissions in the construction industry. Therefore, when analyzing the carbon reduction benefits of prefabricated building policies on the construction industry, we need to consider whether some influencing factors are spillover effects of prefabricated building policies; that is, whether some factors are intermediary factors between prefabricated building policies and the carbon reduction benefits of the construction industry. In order to conduct relevant mechanism analysis, we constructed a mechanism analysis model as follows:

$${\mathrm{A}}_{it}={\mathrm{e}}_0+{\mathrm{e}}_1{treat}_i{period}_t+\sum {\mathrm{e}}_j{X}_{it}+{\mathsf{\mu }}_i+{\mathsf{\gamma }}_t+{\mathsf{\epsilon }}_{it}$$

(2)

$${\mathrm{W}}_{it}={\mathrm{e}}_0+{\mathrm{e}}_2{treat}_i{period}_t+\sum {\mathrm{e}}_j{X}_{it}+{\mathsf{\mu }}_i+{\mathsf{\gamma }}_t+{\mathsf{\epsilon }}_{it}$$

$${\mathrm{A}}_{it}={\mathrm{e}}_0+{\mathrm{e}}_3{treat}_i{period}_t+{\mathrm{e}}_4{\mathrm{W}}_{it}+\sum {\mathrm{e}}_j{X}_{it}+{\mathsf{\mu }}_i+{\mathsf{\gamma }}_t+{\mathsf{\epsilon }}_{it}.$$

Among them, ${\mathrm{W}}_{it}$ is an intermediate variable, and other symbols are consistent with those in model (1), which should be checked as follows: In the first step, if it is significant, it means that the prefabricated building policy has a significant impact, and the verification in the second step will be carried out; otherwise, the program will terminate. The second step is to verify whether the prefabricated building policy has an impact based on its significance. If there is an impact, proceed to the third step; otherwise, terminate. The third step is to determine whether it is a mediating variable based on its significance.

3.3. Variables and Date

3.3.1. Dependent Variable

The dependent variable Ait refers to the natural logarithm of carbon emissions per unit output value of the construction industry in each province, and its calculation formula is as follows [52]:

$${\mathrm{A}}_{it}=\ln ({\displaystyle \frac{EF}{AB\times \sum a_m{b}_m}}).$$

(3)

Among them, $a_m$ represents the various energy consumptions of the construction industry in each province. These data are sourced from the regional energy balance table of each province in the “China Energy Statistical Yearbook”, which mainly includes coal, oil products, electricity, etc. The construction industry part mainly includes coal, oil products, electricity, etc. The coefficient $b_m$ from the “General Rules for Comprehensive Energy Consumption Calculation” (GRCCEC, GB/T 2589-2020) is used to convert various forms of energy consumption into standard coal for measurement [53]. AB represents the standard coal carbon emission factor coefficient released by the Energy Research Institute of the National Development and Reform Commission of China, while EF represents the total output value of the construction industry in each region. These data are from the China Construction Industry Statistical Yearbook (CSYC).

3.3.2. Core Explanatory Variables

The core explanatory variable is ${\mathrm{M}}_{it}\ast {\mathrm{T}\mathrm{i}\mathrm{m}\mathrm{e}}_t$. If ${\mathrm{M}}_{it}\ast {\mathrm{T}\mathrm{i}\mathrm{m}\mathrm{e}}_t$ = 1, it means that province i will start promoting prefabricated building policies after time t in 2016. If ${\mathrm{M}}_{it}\ast {\mathrm{T}\mathrm{i}\mathrm{m}\mathrm{e}}_t$ = 0, it indicates that time t was before 2016, or province i did not promote prefabricated building policies after time t. If the coefficient of the variable ${\mathrm{M}}_{it}\ast {\mathrm{T}\mathrm{i}\mathrm{m}\mathrm{e}}_t$ is positive and significant, as it indicates that the prefabricated construction policy can promote the carbon emissions of the construction industry output value ${\mathrm{A}}_{it}$ in province i.

3.3.3. Mechanism Variables

Mechanism variables are key to understanding how prefabricated building policies influence carbon emissions in the construction industry. This study examines three such variables: labor productivity, mechanical power rate, and material utilization rate.

Labor productivity: This variable measures the efficiency and effectiveness of labor in construction enterprises. It is calculated as the ratio of the total output value of provincial-level construction enterprises to the number of workers. Data for this measure are sourced from the China Construction Industry Statistical Yearbook (CSYC). By evaluating labor productivity, the study aims to gauge the impact of prefabricated building policies on the management level and efficiency improvements within construction firms.

Mechanical power rate: This variable assesses the extent of mechanical power used in construction processes. An increase in mechanical power usage could imply a shift towards more automated and efficient construction techniques, which can influence carbon emissions.

Material utilization rate: This variable examines how effectively construction materials are used. Higher material utilization rates suggest reduced waste and more efficient use of resources, which can contribute to lower carbon emissions.

By analyzing these mechanism variables, the study aims to understand how prefabricated building policies drive changes in labor efficiency, technological adoption, and material usage, and how these changes, in turn, affect carbon emissions in the construction sector [54]. The mechanical power equipment rate is suitable for measuring the impact of policies on the use of machinery by construction enterprises. The data are sourced from the average mechanical power equipment rate of construction enterprises in various provinces in the China Construction Industry Statistical Yearbook (CSYC) [54]. The carbon emissions of building materials are an important component of the carbon emissions throughout the building lifecycle [55]. The material utilization rate is used to measure the promotion of prefabricated construction technology and the improvement of material utilization efficiency through the use of prefabricated components. The data are sourced from the ratio of the total output value of the construction industry in each province to the cement consumption in the China Construction Industry Statistical Yearbook (CSYC).

3.3.4. Control Variable

The selection of independent variables for other possible factors that may affect carbon emissions in the construction industry in this article is determined based on the correlation of the dependent variables and previous research. This article selects energy structure, technological innovation, per capita GDP of each province, research and development expenditure, environmental protection efforts, and technological equipment rate as control variables. The energy structure, as a technological factor, is an important factor affecting the carbon emission intensity of the construction industry. The optimization of the energy structure will to some extent promote the reduction in carbon emissions in the construction industry [56]. The energy structure obtained in this article is used to measure the promoting effect of policies on the reduction in total energy consumption in the construction industry. In this study, the natural logarithm of the ratio of energy consumption in the construction industry to total energy in each province is used as a measure [52]. The data are sourced from the China Energy Statistical Yearbook (CESY) and the energy balance tables of various provincial statistical yearbooks, and are converted into standard coal for relevant calculations using the general principles for comprehensive energy consumption calculation. Technological innovation defines the ability of a region to create new products or change existing ones, and the promotion of new policies will promote industry innovation efforts [57]. Therefore, it may also be a factor affecting carbon emissions in the construction industry. This study measures the proportion of regional R&D expenditure to GDP, representing the regional level of technological innovation. The data are sourced from the China Science and Technology Statistical Yearbook (CTSY) [4]. The improvement of economic conditions and social quality may also be important factors affecting carbon emissions. Therefore, this study takes per capita GDP and environmental protection intensity as control variables, and environmental protection intensity is measured by the proportion of energy-saving and environmental protection funds in each province to the general public expenditure of the government. The above data are, respectively, from the China Statistical Yearbook (CSY). With the improvement of construction technology in construction enterprises, the widespread use of mechanical equipment by enterprises is also a major factor that may affect carbon emissions in the construction industry. The data are sourced from the per capita technical equipment rate in the China Construction Industry Statistical Yearbook (CSYC). The relevant variable descriptions are shown in

Table 1. Description of variables.

Symbol	Variable	Indicator Description	Source
${\mathrm{A}}_{it}$	Carbon emissions from the construction industry	The natural logarithm of the ratio of total output value of the construction industry to carbon emissions.	CSYC&CEYC
LP	Labor productivity	Per capita labor productivity in the construction industry.	CSYC
MPT	Mechanical power rate	Per capita mechanical power equipment rate in the construction industry.	CSYC
ES	Energy-resource structure	Energy consumption and total energy consumption percentage in the construction industry.	CEYC
TI	Technological innovation	R&D expenditure as a percentage of GDP.	CSY
TER	Technical equipment rate	Mechanical equipment allocation rate of construction enterprises.	CSYC
PGDP	Per capita GDP	The ratio of GDP to population in each province over the years.	CSY
EPE	Environmental protection efforts	The natural logarithm of the ratio of environmental protection to total fiscal expenditure in each province.	CSY
MUR	Material utilization rate	The ratio of the total output value of the construction industry in each province to the amount of cement used.	CSYC

:

The research data in this article come from panel data from 18 provinces from 2011 to 2021. Divide 18 provincial-level regions into 5 experimental groups and 13 control groups. Based on the timing and characteristics of the promotion of prefabricated building policies in various provinces, this study sets the start year of the prefabricated building policy as 2016. The description of relevant statistical variables is as follows in

Table 2. Variable Description Statistics.

	SYMBOL	Quantity	Average	Std. Dev	MIN	MAX
Dependent variable	${\mathrm{A}}_{it}$	198	11.4248	0.5267	10.5185	12.9933
Mechanism variables	LP	198	6.1621	2.4471	2.4000	13.3300
	MPT	198	0.8898	0.3322	0.0673	1.7247
	MUR	198	−4.2887	0.5021	−5.4474	−3.2088
Control variable	ES	198	1.4480	0.6381	0.4300	2.9500
	TI	198	4.9572	3.0710	1.6024	37.3533
	PGDP	198	−3.5805	0.3074	−4.4407	−2.7907
	EPE	198	1.2300	0.8118	0.3152	10.8588
	MUR	198	11.4248	0.5267	10.5185	12.9933

:

4. Experimental Analysis

4.1. Benchmark Regression Results

We will evaluate the effects of prefabricated building policies by inputting relevant data into model (1), and the calculation results are shown in

Table 3. Impact of prefabricated building policies on carbon emissions in the construction industry.

Ait	(1)	(2)	(3)	(4)	(5)	(6)
$\mathrm{Policy}\mathrm{variables}({\mathrm{M}}_{\mathrm{i}\mathrm{t}}\ast {\mathrm{T}\mathrm{i}\mathrm{m}\mathrm{e}}_{\mathrm{t}})$	0.462 ***	0.346 ***	0.372 ***	0.333 ***	0.321 ***	0.311 ***
	(0.0705)	(0.0764)	(0.0784)	(0.0743)	(0.0742)	(0.0735)
ES		−0.312 ***	−0.306 ***	−0.297 ***	−0.293 ***	−0.308 ***
		(0.0917)	(0.0915)	(0.0862)	(0.0858)	(0.0851)
TI			−0.164	−0.189 *	−0.193 *	−0.162
			(0.119)	(0.112)	(0.111)	(0.111)
PGDP				0.0300 ***	0.0304 ***	0.0286 ***
				(0.00636)	(0.00633)	(0.00631)
EPE					0.177 *	0.180 *
					(0.103)	(0.102)
MPT						−0.0445 **
						(0.0203)
Province FE	Y	Y	Y	Y	Y	Y
Year FE	Y	Y	Y	Y	Y	Y
Constant	11.27 ***	9.869 ***	10.09 ***	10.01 ***	10.66 ***	10.63 ***
	(0.0521)	(0.414)	(0.444)	(0.418)	(0.560)	(0.554)
N	198
R2	0.246	0.295	0.303	0.385	0.396	0.413

* p < 0.10, ** p < 0.05, *** p < 0.01.

.

We used a difference-in-differences method to fix the year and region in both directions, so that the standard deviations were all concentrated in the province. From the chart results, we can see that the coefficient of the time and space interaction term in column (1) is positive, and the coefficients in columns (1) to (6) are always positive. With the continuous increase in control variables, they all passed the significance test at the 1% level, indicating that the implementation of the prefabricated building policy significantly improved the carbon emissions level of the construction industry in each experimental group province. The significance coefficient of column (6) is 0311, indicating that the implementation of the prefabricated building policy has a significant improvement effect on the total output value of the construction industry created by each ton of carbon dioxide emissions. This is consistent with the conclusion of many scholars, such as Chen, that the promotion of prefabricated building technology can effectively reduce energy consumption in the construction industry by reducing greenhouse gas production by about 20% [24].

4.2. Parallel Trend Test

To ensure the validity of empirical analysis using a difference-in-differences method, it is crucial that the control group and experimental group exhibit similar trends before the policy intervention. The analysis of carbon emission efficiency trends in the two sample groups, as shown in Figure 2, reveals the following:

Pre-policy trend consistency: Prior to 2016, the carbon emission efficiency trends in both sample groups were consistent, suggesting that any observed differences post-2016 can be attributed to the policy rather than pre-existing disparities. This consistency supports the parallel trend assumption necessary for the double difference method.

Post-policy divergence: Since 2016, the trend lines of the two sample groups diverged significantly, indicating a notable policy impact. The policy’s dynamic effect began to show significant differences between the groups from 2016 onwards, confirming that the policy had a measurable impact on carbon emission efficiency.

Increasing policy impact: After 2018, the dynamic effect of the policies continued to increase, which suggests that the benefits of the prefabricated building policies became more pronounced over time. This trend is likely due to the gradual improvement in the prefabricated building industry chain and the maturing technology within construction enterprises.

Overall, the parallel trend test results validate the use of the difference-in-differences method for analyzing the policy’s impact, and the increasing policy effects underscore the importance of ongoing development and expansion in prefabricated building practices.

5. Robustness Testing

5.1. Placebo Testing

To address research limitations and mitigate the impact of randomness on carbon emission efficiency related to prefabricated building policies, this study employs placebo testing. Placebo testing involves randomly selecting 5 regions from a pool of 18 provincial regions, conducting 500 iterations of the experiment, and evaluating the credibility of the experimental results based on the distribution and significance of these iterations. If the majority of placebo results do not show significant deviations from the original findings and lack statistical significance, it supports the reliability of the original quasi-natural experiment’s conclusions.

Figure 3 shows the placebo test results, which are clustered around 0 and have a significant deviation from the original experimental results in terms of p value and are not statistically significant. This indicates that the experimental results of the experimental group and the control group are inevitable and can be referenced.

5.2. Heterogeneity Analysis

To further assess the robustness of the experimental results, this study employed sev samples, and quantile analysis of core variables. Given the potential influence of outliers on the significance of the results, these methodologies were implemented to enhance the accuracy and reliability of the findings. The robustness test results are detailed in

Table 4. Robustness test for excluding outliers and outlier times.

	(1)	(2)	(3)	(4)	(5)	(6)
	Winsor	Winsor	Heteroskedasticity − Robust + Standard + Error	Clustering Standard Error	Delete 2021	Delete Hubei and Jiangsu
Policy variables (Mit ∗ Timet)	0.461 ***	0.243 ***	0.329 **	0.329 ***	0.292 ***	0.316 ***
	(0.0704)	(0.0732)	(0.143)	(0.0694)	(0.0760)	(0.0873)
ES		−0.321 ***	−0.313 *	−0.313 ***	−0.294 ***	−0.241 ***
		(0.0821)	(0.171)	(0.0846)	(0.0938)	(0.0922)
TI		−0.175	−0.182	−0.182 *	−0.169	−0.161
		(0.107)	(0.209)	(0.108)	(0.125)	(0.117)
PGDP		0.0982 ***	0.0284 ***	0.0284 ***	0.0276 ***	0.0280 ***
		(0.0178)	(0.00384)	(0.00561)	(0.00640)	(0.00658)
EPE		0.169 *	0.180	0.180 **	0.165	0.188 *
		(0.0997)	(0.141)	(0.0860)	(0.109)	(0.108)
MPT		−0.102 **	−0.0434 **	−0.0434 **	−0.0450 **	−0.0441 **
		(0.0422)	(0.0168)	(0.0178)	(0.0205)	(0.0211)
Prvince FE	Y	Y	Y	Y	Y	Y
Year FE	Y	Y	Y	Y	Y	Y
Constant term	11.27 ***	10.41 ***	10.63 ***	10.63 ***	10.66 ***	10.84 ***
	(0.0521)	(0.541)	(0.996)	(0.526)	(0.604)	(0.588)
N	198
R2	0.246	0.451	0.416	0.883	0.395	0.346

* p < 0.10, ** p < 0.05, *** p < 0.01.

. Columns (1) and (2) present the regression results. Following the inclusion of control variables, the coefficient for the core explanatory variable, as adjusted using the Winsor2 algorithm, decreased from 0.461 to 0.243. The introduction of the carbon trading rights policy by the Chinese government in mid-2021 may have influenced the assessment of the benefits associated with the prefabricated building policy. To address this, data from 2021 were excluded from the analysis, and the robustness test was executed. The coefficient for the policy variable remained approximately 0.292, suggesting that the core findings are consistent. Furthermore, the robustness of the results is supported by the minimal impact of the robustness checks, which may be due to the delayed effects of the carbon trading policies. Specifically, the overlap between the implementation of the prefabricated building policy and the pilot period for the carbon trading rights policy is particularly evident in Hubei and Jiangsu provinces, which adopted the prefabricated building policy earlier. To address this, data from Hubei and Jiangsu were excluded from the robustness analysis, resulting in a coefficient of 0.316 for the core variable. This suggests that while the carbon trading rights policy pilot had some impact on evaluating the benefits of the prefabricated building policy, the results remain within a manageable range. Additionally, the clustering standard error and heteroskedasticity-robust standard error reported in this study are both 0.329, confirming the statistical significance of the findings. Thus, the benchmark test results demonstrate robustness in the analysis.

To test for heterogeneity, the article uses the quantile method with a 15% scale value. The analysis, detailed in

Table 5. Quantile heterogeneity test.

	(1)	(2)	(3)	(4)
	15%	45%	75%	90%
Policy variables (Mit ∗ Timet)	0.0398	0.188 **	0.171 *	0.211 ***
	(0.0543)	(0.0933)	(0.0908)	(0.0296)
ES	−0.463 ***	−0.452 ***	−0.544 ***	−0.581 ***
	(0.0629)	(0.108)	(0.105)	(0.0343)
TI	−0.00251	−0.171	−0.313 **	−0.281 ***
	(0.0821)	(0.141)	(0.137)	(0.0447)
PGDP	0.0400 ***	0.0322 ***	0.0323 ***	0.0313 ***
	(0.00466)	(0.00800)	(0.00779)	(0.00254)
EPE	−0.0584	0.0571	0.167	0.0868 **
	(0.0753)	(0.129)	(0.126)	(0.0410)
MPT	0.00815	−0.0142	−0.0763 ***	−0.0896 ***
	(0.0150)	(0.0257)	(0.0250)	(0.00815)
Prvince FE	Y	Y	Y	Y
Year FE	Y	Y	Y	Y
	9.487 ***	10.42 ***	10.91 ***	10.49 ***
	(0.446)	(0.767)	(0.746)	(0.243)
R2	0.6987	0.6727	0.7421	0.7978

* p < 0.10, ** p < 0.05, *** p < 0.01.

, reveals that as the carbon emission efficiency of prefabricated building policies improves, the positive impact of policy variables on this efficiency becomes more pronounced. This effect could be linked to advancements in prefabricated building technology, improved process management by companies, and the increasing expertise of industry workers.

5.3. PSM-DID

Given the substantial variations in economic and industrial development across different provinces, this study employed the propensity score matching with difference-in-differences (PSM-DID) method to assess the robustness of the experimental results. To achieve this, relevant variables were selected as covariates for proximity matching, ensuring that both the experimental and control groups had comparable covariates. Subsequently, balance tests were conducted to verify this equivalence. As illustrated in

Table 6. Balance test for PSM.

Variable	Matched	Treated	Control	%Bias	|Bias|	t	p	V(T)/V(C)
ES	U	−4.348	−4.2659	−18		−1.03	0.304	0.35 *
	M	−4.2537	−4.2299	−5.2	71.1	−0.21	0.833	0.48 *
TI	U	2.0704	1.2086	169.3		10.68	0	1.01
	M	1.8221	1.7749	9.3	94.5	0.42	0.673	0.93
PGDP	U	6.7556	4.2655	90.5		5.47	0	0.67
	M	5.4958	4.6755	29.8	67.1	1.54	0.129	0.27 *
EPE	U	−3.5305	−3.5997	22.4		1.42	0.156	1.07
	M	−3.5001	−3.3946	−34.1	−52.5	−1.38	0.172	1.82
MPT	U	1.1826	1.2482	−9.2		−0.51	0.612	0.19 *
	M	1.2909	1.257	4.8	48.2	0.21	0.831	0.25 *

* p < 0.10.

, the analysis revealed a significant reduction in the deviation between the two sets of variables. Consequently, these adjustments led to more robust and convincing results.

Following the completion of the propensity score matching (PSM) balance test, the study proceeded with correlation calculations on the two datasets, with the results presented in

Table 7. The estimation results of PSM-DID.

	(1)	(2)
Policy variables (Mit ∗ Timet)	0.5511 ***	0.2410 **
	(0.1386)	(0.1274)
ES		−1.1137 ***
		(0.2134)
TI		0.0483
		(0.2291)
PGDP		0.0221 ***
		(0.0071)
EPE		−0.0432
		(0.2062)
MPT		−0.0358
		(0.0227)
Prvince FE	Y	Y
Year FE	Y	Y
	11.354 ***	6.1727 ***
	(0.0973)	(1.2579)
R2	0.2394	0.5980

** p < 0.05, *** p < 0.01.

. The findings depicted in Table 7 indicate that, irrespective of the inclusion of control variables, the prefabricated building policy exerts a positive impact on carbon emissions within the construction industry. This stability in the experimental results underscores the robustness of the observed effects.

6. Mechanism Variables Analysis

Following the benchmark regression and robustness testing, the study confirms the benefits of prefabricated building policies on carbon emission efficiency and validates the experimental results. The next phase involves examining how these policies affect carbon emission efficiency through mechanism analysis. This analysis will categorize carbon emissions in construction into labor, materials, and machinery. Using Model (2), the study will assess labor productivity, material utilization rates, and mechanical power rates as mediating variables to explore their roles in the relationship between the policy and carbon emission efficiency.

Table 8. Mechanism Verification.

	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
	LP			MPT			MUR
Policy variables (Mit ∗ Timet)	0.793 ***	0.305 ***	0.432 ***	0.658	0.319 ***	0.470 ***	0.360 ***	0.274 ***	0.438 ***
	(0.241)	(0.0761)	(0.0744)	(0.483)	(0.0740)	(0.0708)	(0.0862)	(0.0770)	(0.0742)
ES	−0.0587	−0.308 ***		−0.544	−0.314 ***		0.0143	−0.310 ***
	(0.279)	(0.0853)		(0.559)	(0.0854)		(0.0997)	(0.0847)
TI	−0.0371	−0.162		−1.619**	−0.180		−0.0455	−0.157
	(0.364)	(0.111)		(0.730)	(0.113)		(0.130)	(0.111)
PGDP	0.0551 ***	0.0282 ***		−0.0704 *	0.0278 ***		0.00732	0.0279 ***
	(0.0207)	(0.00646)		(0.0415)	(0.00636)		(0.00739)	(0.00630)
EPE	0.550	0.175 *		0.837	0.189 *		−0.412 ***	0.222 **
	(0.334)	(0.103)		(0.670)	(0.102)		(0.119)	(0.105)
MPT	0.0608	−0.0450 **		0.272 **	−0.0414 **		0.0315	−0.0478 **
	(0.0664)	(0.0204)		(0.133)	(0.0205)		(0.0237)	(0.0203)
Mediating variables		0.00811	0.0322		−0.0114	−0.0142		0.103	0.0738
		(0.0239)	(0.0258)		(0.0119)	(0.0127)		(0.0663)	(0.0713)
Bootstrap algorithm	[0.0892, 0.4568]			[0.0758, 0.4870]			[0.0931, 0.4553]
Prvince FE	Y	Y	Y	Y	Y	Y	Y	Y	Y
Year FE	Y	Y	Y	Y	Y	Y	Y	Y	Y
	4.801 ***	10.59 ***	11.16 ***	9.052 **	10.73 ***	11.36 ***	−0.791	10.71 ***	11.22 ***
	(1.817)	(0.567)	(0.103)	(3.641)	(0.564)	(0.0986)	(0.649)	(0.554)	(0.0684)
N	198
R2	0.420	0.414	0.253	0.288	0.417	0.252	0.438	0.422	0.251

* p < 0.10, ** p < 0.05, *** p < 0.01.

mechanism analysis shows that the core explanatory variable’s coefficient for labor productivity is positive and significant at the 1% level in both columns (1) and (2), demonstrating that the prefabricated construction policy enhances labor efficiency. The bootstrap algorithm’s parameter interval [0.0892–0.4568] supports the confidence in the coefficient of 0.432 in column (3) in Table 8, confirming that labor productivity significantly mediates the relationship between the policy and carbon emission efficiency. Thus, the prefabricated building policy improves carbon emission efficiency by enhancing labor productivity in the construction industry.

In column (4) of Table 8, the coefficient for the mechanical power rate is positive and significant at the 1% level, indicating that prefabricated construction policies enhance the mechanical power rate in construction enterprises. This aligns with the benchmark regression’s finding that a higher per capita technical equipment rate negatively correlates with carbon emission efficiency. The use of prefabricated components appears to increase machinery utilization and encourages the updating of construction equipment, leading to better resource allocation. The coefficient for mechanical power rate as an intermediary variable in column (6) is 0.470 within the confidence interval of [0.0758, 0.4870]. This confirms that prefabricated construction policies contribute to the optimal allocation of mechanical resources, thereby improving carbon emission efficiency.

In columns (7) and (8) of Table 8 the core explanatory variable coefficients for material utilization are positive and significant at the 1% level. This suggests that prefabricated building policies enhance the material utilization rate in construction enterprises. Prefabricated construction shifts component production from sites to factories, where skilled workers operate in controlled environments, leading to more efficient production, reduced energy consumption, less waste, and fewer defects. The coefficient in column (9) is 0.438, which falls within the bootstrap algorithm’s confidence interval [0.0931, 0.4553]. Thus, it is evident that prefabricated building policies improve material utilization efficiency, contributing to better carbon emission efficiency in buildings.

7. Discussion on Variable Factors

7.1. Discussion on Core Variables

Through benchmark regression analysis of the core variables and subsequent robustness verification, we find that the implementation of prefabricated building policies significantly affects carbon emissions in the construction industry within the experimental group. This finding corroborates the conclusions of various scholars, including Chen et al., who argue that the promotion of prefabricated building technology effectively reduces energy consumption in the construction sector, thereby alleviating carbon emissions in China’s construction industry [24]. Additionally, further investigation into parallel trends reveals that as the implementation of prefabricated building policies deepens, the associated benefits continue to strengthen. This enhancement is likely attributed to the progressive improvement of the prefabricated building industry chain and the maturation of the construction enterprises’ prefabricated building technologies. Consequently, the broader the scope of prefabricated building policy implementation, the more seamless the flow of industry and technology, resulting in a greater impact on the carbon emission efficiency of the construction industry.

7.2. Discussion on Control Variable

In addition to focusing on prefabricated building policies as the core variable, we also examined other variables that might influence these policies, while controlling for them in the research process. If the significance test coefficients for per capita GDP, environmental protection efforts, and technological innovation are all positive, it suggests that these factors have a promoting effect on the carbon emission efficiency of the construction industry. This emphasizes the important influence of social concepts, including economic factors, on the carbon emission efficiency of the construction industry, which is consistent with the conclusion emphasized by Agi, Diabat et al. on the influence of government, society, culture, and other aspects [47,48,49]. The greater the intensity of technological innovation, the faster the industrial upgrading it brings. On the one hand, it can promote the improvement of production efficiency, on the other hand, it can promote the use and promotion of renewable energy and improve the energy utilization efficiency of the construction industry.

On the contrary, the significance test of energy structure and technological equipment rate is negative, indicating a negative correlation between the above two factors and carbon emissions in the construction industry. The lower the proportion of energy consumption in the construction industry to the total energy consumption in society, the higher the carbon emission efficiency of the construction industry. This study uses the consumption of standard coal as a variable to measure the energy structure of the construction industry, and traditional energy itself is one of the important sources of carbon emissions. Therefore, the prefabricated building policy can effectively improve the energy consumption structure of China’s construction industry. One possible entry point for further policy promotion is to promote the use of renewable energy instead of traditional energy by prefabricated components, material production enterprises, and prefabricated component production enterprises. The negative correlation effect of the improvement of technological equipment rate on the carbon emission efficiency of the construction industry indicates that there is currently a certain lag in the management of mechanical equipment, updates, and elimination of new and old equipment in Chinese construction enterprises, which has a certain degree of negative impact on carbon emission efficiency.

7.3. Discussion on Mechanism Variables

Labor productivity: Mechanistic analysis confirms that labor productivity is the primary channel through which prefabricated building policies influence carbon emission efficiency. This finding aligns with the conclusions of Bo Li and Shu et al., who identify labor quality as a significant factor limiting improvements in carbon emission efficiency within the construction industry [54]. Over the past few decades, China’s construction industry experienced significant growth, which led to an inadequacy in scientific training systems for technical workers and a shortage of skilled industrial professionals. This resulted in suboptimal project management. The introduction of prefabricated building policies, on one hand, spurred the growth of prefabricated component enterprises, facilitated the skill development of industrial workers within these enterprises, and enhanced production efficiency. On the other hand, by shifting some construction tasks to prefabricated component production facilities, construction enterprises capitalized on the advanced skills of industrial workers and the refined management systems of these factories. This shift notably reduced material and energy losses during construction, thereby improving efficiency and effectively lowering carbon emissions during the building phase.

Mechanical power rate: The mechanical power rate was shown to serve as an intermediary pathway through which prefabricated building policies impact carbon emissions in the construction industry. This view is consistent with Qiu et al.’s assertion that optimizing the allocation of mechanical resources and phasing out outdated machinery can effectively reduce carbon emissions [58]. Prefabricated building technology, which involves transporting pre-manufactured components to the construction site for assembly, employs mechanical tools differently than traditional construction methods. This shift encourages construction companies to modernize their machinery, replacing outdated, energy-inefficient equipment and thus enhancing the prefabricated parts market [45]. Additionally, it drives prefabricated component manufacturers to upgrade their production lines with more energy-efficient machinery, reducing equipment idleness and minimizing mechanical resource wastage.

Material utilization rate: Material utilization rate is widely recognized by scholars as a crucial factor in mitigating carbon emissions in the construction industry [45,50]. The mechanism variable analysis in this article further highlights the significant impact of prefabricated building policies on enhancing material utilization. Compared to traditional construction methods, the production process of prefabricated components utilizes more green and low-carbon materials, improving the reuse rate of production templates and reducing wooden template waste [50]. Additionally, the specialized, process-oriented factory production minimizes defects and material loss while facilitating waste recycling. Favorable storage conditions in prefabricated parts production also contribute to reduced component wear and material loss.

8. Conclusions and Policy Recommendations

8.1. Research Conclusions

The study aims to assess the carbon emission efficiency of China’s construction industry by analyzing the total output value per ton of carbon dioxide emitted. Utilizing various methods, including the double difference method, parallel trend test, benchmark regression, robustness and heterogeneity analysis, and mechanism analysis, the research explores the impacts of prefabricated building policies. The key conclusions are:

Improvement in carbon emission efficiency: The implementation of prefabricated building policies significantly enhanced the carbon emission efficiency of China’s construction industry. As the industrial chain matures, the benefits of these policies are expanding.

Policy focus areas: the prefabricated construction policy positively influences carbon emission efficiency through improvements in labor productivity, mechanical power rate, and material utilization rate.

Effects of control variables: In addition to the core variables, regional technological innovation, per capita GDP, and environmental protection also positively impact carbon emission efficiency. Conversely, energy structure and the technology assembly rate of construction enterprises negatively affect carbon emission efficiency.

8.2. Research Limitations

This study adopts a provincial-level perspective on the construction industry and relies primarily on macro panel data. Given the varying priorities placed on economic development and environmental policies across different provinces, as well as their fluctuations over time, there may be discrepancies in the implementation and promotion of prefabricated building policies across different years. To maintain the coherence and feasibility of the research, this study idealized these changing factors. Consequently, further investigation is recommended to more precisely evaluate the carbon emission benefits of prefabricated buildings, considering these temporal and spatial variations. Secondly, there is a current lack of precise quantitative measurement systems for assessing the impact of various variables on carbon emissions in the construction industry. Developing a comprehensive indicator system to evaluate the effects of these variables on carbon emissions may be an area for future research.

8.3. Policy Recommendations

The policy recommendations based on the research findings are as follows: (1) Sustainability and delays: Recognize that the benefits of prefabricated building policies may experience delays. Governments should ensure the sustainability of policy implementation, focus on the maturity of the industrial chain, and consider the time and capital costs associated with technology. (2) Technological innovation: Promote technological research and innovation by establishing special funds for new prefabricated building technologies. Offer innovation subsidies and tax incentives to encourage enterprises to invest in R&D, enhance technological innovation, and accelerate the industry’s maturation. Implement policies to improve material utilization and recycling in prefabricated construction, encourage the use of high-performance, eco-friendly materials, and establish standards for material usage and waste reduction. (3) Energy structure adjustment: Introduce policies to adjust the local energy structure by replacing traditional energy sources with clean energy and diversifying energy sources. Enforce restrictions on excessive carbon emissions, establish a coal consumption cap with gradual reductions, and support the technological transformation of prefabricated construction enterprises to enhance energy efficiency. (4) Performance evaluation and management: Develop performance evaluation systems to improve construction enterprise management, provide systematic training for construction workers, and boost labor productivity. Implement incentive mechanisms to phase out inefficient machinery, reduce wasteful equipment, and adopt efficient construction techniques to optimize resource allocation and enhance mechanical power efficiency.