The Belt and Road (B&R) Initiative, launched by the Chinese government in 2015, has become a significant global infrastructure project involving numerous countries. Simultaneously, the importance of green development (GD) – a paradigm emphasizing sustainable economic growth and environmental protection – has grown internationally. The construction industry, a cornerstone of national economies, is known for its high energy consumption, waste generation, and substantial carbon dioxide (CO2) emissions. This study addresses a critical gap in the literature by examining the impact of the B&R policy on the GD level of construction enterprises in China. Previous research has explored the B&R Initiative's economic and environmental consequences, but its specific influence on the sustainability performance of construction firms remains understudied. This research aims to fill this gap by investigating the mechanism through which the B&R policy affects the GD level of construction enterprises, focusing on CO2 emission intensity as a key indicator of green development within the industry. Understanding this relationship is crucial for developing effective policies that promote both infrastructure development and environmental sustainability. The Chinese government's commitment to a 'green B&R' further underscores the importance of this research, emphasizing the integration of GD principles into the Initiative's implementation. The study's findings will contribute to both the econometrics of sustainable development in the construction industry and inform policy recommendations for building a truly green B&R.

The literature review examines existing research on the GD of construction enterprises and the assessment of the B&R policy. Studies on the GD of construction enterprises focus on influencing factors (governmental behavior, industry scale, natural resources, urban development, R&D investment), methods for promoting GD (business models, green production, public environmental needs), and assessment of GD levels (indicator systems, information entropy models). Regarding B&R policy assessment, research has utilized various approaches, such as gravity models for international trade analysis, assessments of Chinese direct investment in overseas countries, and econometric studies on the relationship between financial development and the ecological environment in B&R Initiative countries. However, there's a scarcity of research directly evaluating the B&R policy's impact on the GD of construction enterprises. This research draws upon the existing literature to establish a theoretical framework and methodological approach for its empirical investigation.

This study employs a quasi-natural experimental design using the differences-in-differences (DID) and propensity score matching (PSM-DID) methods to assess the B&R policy's impact on the GD of construction enterprises. The study utilizes panel data for 28 Chinese provincial administrative regions from 2010 to 2020. The CO2 emission intensity of construction enterprises is used as the primary indicator of GD. The sample is divided into an intervention group (17 key B&R regions) and a control group (remaining regions). 2015, the year of the B&R policy's official implementation, serves as the policy intervention node. A fixed effects model is constructed with CO2 emission intensity as the dependent variable. The baseline model includes a treatment dummy, a time dummy, and an interaction term representing the policy effect. Subsequently, control variables are added to the model to account for other factors that may influence GD. These control variables include R&D investment, sewage treatment investment, labor productivity, education level (measured by undergraduate enrollment), and regional development level (measured by GDP per capita). A parallel trend test is conducted to ensure the validity of the DID approach. PSM is employed to address potential selection bias and improve the comparability of the treatment and control groups. Statistical analysis is performed using Stata 16.0 to estimate the model's coefficients and assess the significance of the policy effect and control variables. The data is logged after multiplication to facilitate comparison of non-dummy variable coefficients.

The DID regression analysis reveals that the B&R policy has a significant negative effect on CO2 emission intensity in the planned key regions, indicating a positive impact on GD. The coefficient of the interaction term (TREATED × TIME) is significantly negative at the 1% level, suggesting a reduction in CO2 emission intensity of 0.346 units after policy implementation. R&D investment shows a significant negative relationship with CO2 emission intensity, suggesting that increased R&D promotes GD. Conversely, regional GDP per capita, undergraduate enrollment, and sewage treatment investment exhibit positive relationships with CO2 emission intensity, indicating that increased economic activity, higher education levels and higher sewage treatment investment may have a negative impact on GD, at least in the short term. Labor productivity is not statistically significant in this model. The PSM-DID analysis, conducted to mitigate selection bias, confirms the findings of the DID analysis, showing a consistently significant negative effect of the B&R policy on CO2 emission intensity in the planned key regions. The results are robust across different model specifications.

The findings demonstrate that the B&R policy has positively influenced the GD level of construction enterprises in its planned key regions, primarily by curbing CO2 emissions. This supports the hypothesis that the policy promotes green development. The influence of control variables aligns with existing literature in some respects, but also reveals complexities. While R&D promotes GD, increased regional economic activity, higher education, and investment in sewage treatment appear to initially hinder GD. The latter finding may reflect the trade-off between economic development and environmental protection in the short term, suggesting the need for targeted policies to mitigate this. The study provides valuable insights for policymakers in China and other countries involved in the B&R Initiative. The positive effect of the policy on GD, particularly when coupled with increased R&D investment, indicates that strategic planning and investment in green technologies can promote sustainable infrastructure development. The study's limitations, such as the exclusion of certain regions due to data unavailability, need to be considered when interpreting the results.

This study contributes to the understanding of the B&R Initiative's environmental impact, specifically its influence on the GD of construction enterprises. The key finding is the positive effect of the policy on GD levels in its planned key regions, primarily through a reduction in CO2 emission intensity. Further research could investigate the long-term effects of the B&R policy on GD, explore the influence of other factors (market maturity, enterprise willingness to adopt GD practices), and employ more sophisticated econometric techniques to account for potential endogeneity issues and other complexities.

This study has some limitations. Data limitations restricted the analysis to 28 Chinese provinces, excluding others due to data unavailability. The study focuses on CO2 emission intensity as the primary measure of GD, potentially overlooking other important aspects of green development. The model does not fully account for potential unobserved factors that could influence the relationship between the B&R policy and GD. Future research should address these limitations through data expansion, inclusion of alternative GD indicators, and more complex econometric techniques to increase the study's robustness and generalizability.