The 2030 Agenda for Sustainable Development prioritizes poverty eradication. While progress was made before the COVID-19 pandemic, the global poverty reduction has stalled due to the pandemic, climate change, and conflicts. China's successful "Targeted Poverty Alleviation Strategy" provides a model. The Belt and Road Initiative (BRI), proposed by China in 2013, aims to foster economic partnerships and cooperation with participating countries. President Xi Jinping later proposed the BRI as a pathway towards poverty alleviation. Many BRI countries are developing nations facing significant poverty challenges. This study addresses the crucial question: Does the BRI effectively contribute to poverty reduction in these countries? The study's significance lies in informing the ongoing implementation of the BRI and contributing to global poverty governance, offering insights into whether the BRI's economic benefits translate into tangible improvements for low-income populations.

Existing literature on the BRI primarily focuses on its economic impacts, examining its effects on China's economy (industrial development, enterprise competition, improved export quality, reduced financing constraints), and on BRI countries (economic growth, regional integration, infrastructure improvements, urban development). Studies highlight the BRI's positive influence on economic growth and regional economic integration, while also acknowledging potential drawbacks like increased competition among local enterprises and environmental concerns. While some research links the BRI to the UN Sustainable Development Goals (SDGs), particularly SDG1 (poverty eradication), a definitive causal relationship between the BRI and poverty reduction, along with a detailed explanation of the underlying mechanisms, remains largely unexplored. This article aims to fill this gap.

This study employs a difference-in-differences (DID) approach to analyze the causal relationship between BRI participation and poverty reduction. The study utilizes a 15-year panel dataset (2005-2019) from 151 countries, with BRI participating countries forming the treatment group and non-participating countries serving as the control group. The year 2013, when the BRI was launched, marks the policy juncture. The dependent variable is the poverty headcount ratio (moderate poverty line of $3.2 per person per day in 2011 PPP). The model controls for various economic (GDP per capita, GDP growth rate, physical capital, natural resource endowment, agricultural output value), social (urbanization level, labor force participation rate, population density, dependency ratio), and institutional quality factors. Heterogeneity analysis is conducted to examine the impact across different geographical locations (neighboring vs. non-neighboring countries), income groups (low, lower-middle, upper-middle), and BRI types (Land Silk Road Economic Belt vs. 21st-Century Maritime Silk Road). A mediation analysis is undertaken to explore the mechanisms through which the BRI impacts poverty reduction, focusing on the 'Five Cooperation Priorities' (policy consultation, trade promotion, infrastructure connectivity, financial cooperation, people-to-people exchanges). The mediating variables include unimpeded trade (multilateral and bilateral trade volumes), financial integration (number of commercial bank branches, FDI inflows, greenfield investment projects), facilities connectivity (mobile subscriptions, internet usage), people-to-people bonds (number of Confucius Institutes), and policy coordination (number of official visits, technical cooperation exchanges). Stepwise regression and the Sobel test are employed to assess the mediating effects.

The study's key findings are: 1. Joining the BRI significantly reduces the poverty headcount ratio in participating countries compared to non-participating countries. This effect is more pronounced in the long term. 2. The BRI's impact on poverty reduction varies across countries. The effect is stronger in countries geographically closer to China and in lower-middle-income countries. The Maritime Silk Road shows a more significant poverty reduction effect than the Land Silk Road. 3. Mediation analysis reveals that the BRI contributes to poverty reduction through several mechanisms: a) **Unimpeded Trade:** Increased multilateral and bilateral trade volumes with China positively affect poverty reduction. b) **Financial Integration:** Increased commercial bank branches and greenfield investment projects from China contribute positively to poverty reduction. c) **Facilities Connectivity:** Improved mobile subscription penetration positively impacts poverty reduction, although the effect of internet usage is less clear. d) **Policy Coordination:** Increased official visits and technical cooperation exchanges are associated with poverty reduction, indicating the importance of knowledge sharing and technological transfer. e) People-to-people bonds (Confucius Institutes) show limited direct impact on poverty reduction, potentially reflecting the long-term nature of educational investments.

The findings support the hypothesis that the BRI contributes to poverty reduction in participating countries. The varying impacts across different countries highlight the importance of geographical proximity and income level in determining the effectiveness of BRI initiatives. The mediation analysis sheds light on the pathways through which the BRI influences poverty, emphasizing the importance of trade, finance, infrastructure development, and knowledge exchange. These findings provide valuable insights into the effectiveness of large-scale infrastructure and development initiatives and offer policy implications for poverty reduction efforts in developing countries.

This study demonstrates a significant positive effect of the BRI on poverty reduction in participating countries, although this effect varies across different contexts. The findings highlight the multifaceted nature of the BRI's impact, underscoring the importance of trade, finance, infrastructure, and knowledge transfer. Future research could investigate the BRI's impact in the context of the COVID-19 pandemic and utilize time-varying DID models to capture the varied entry times of participating countries. Further investigation into the long-term impacts of various BRI initiatives is also needed.

The study's limitations include the use of a standard DID model with a fixed policy cutoff point (2013), potentially overlooking the staggered entry times of BRI countries. The study's analysis period (2005-2019) predates the COVID-19 pandemic and recent geopolitical shifts, limiting the generalizability of the findings to the post-pandemic period. Data availability also posed challenges, leading to the exclusion of some countries from the analysis.