Impact of the Belt and Road Initiative on Poverty Reduction in Countries along the Route

Poverty reduction is an important driving force for the global realization of the United Nations 2030 Sustainable Development Goals and the building of a global community with a shared future for mankind. is paper takes the Belt and Road Initiative (BRI) as a quasi-natural experiment. Based on the panel data of 134 countries in the world from 2000 to 2018, this paper uses the dierence-in-dierence (DID) model to examine the impact of the BRI on poverty reduction in countries along the route and to study its internal mechanism from the dual perspectives of trade openness and investment openness. e results show that (1) the BRI can signicantly reduce the incidence of poverty with the impact coecient stable at −0.26, and the empirical conclusion has passed the parallel test, placebo test, and instrumental variable test; (2) the mediating eect test model shows that the BRI can achieve poverty reduction through the dual openness of trade and investment, and the intermediary eect of trade openness is greater than investment openness; and (3) from the perspective of heterogeneity analysis, geographically, the BRI has a slightly higher role in promoting poverty reduction in landlocked countries than in coastal countries. In terms of economic location, the eect of poverty reduction has a certain “pro-poverty” characteristic, that is, the BRI promotes poverty reduction in lowand middle-income countries far more than other types of income countries. erefore, we believe that the continuous deepening of the BRI high-quality construction and the strengthening of cooperation among countries along the route will play a key role in promoting the international cycle of trade, investment, and other factors, as well as the cause of poverty reduction in the region and the world.


Introduction
Poverty reduction is an important driving force for the global realization of the United Nations 2030 sustainable development goals and the construction of community of a shared future for mankind. In recent years, the global poverty issue has been signi cantly alleviated. According to the data released by the World Bank, the proportion of the global extreme poverty population (measured by the daily cost of living less than USD 1.90 per person (purchasing power in 2011)) has decreased from 35.9% in 1990 to 9.2% in 2017; in other words, more than 1.2 billion people have got rid of extreme poverty in the almost past 30 years (Data source: Poverty and shared prosperity 2020: Reversals of Fortune released by the World Bank). As the largest developing country in the world, China has always attached great importance to poverty alleviation, which has helped 770 million rural poor population (over 70% of the targeted global poverty population) completely shake o absolute poverty (Data source: Poverty Alleviation: China's Experience and Contribution issued by China's State Council Information O ce), adding its own mark in the world's history of poverty reduction e orts. Nevertheless, due to the impact and constraints of natural geography, social culture, economic development, and other factors, poverty reduction progress varies greatly across the globe: Poverty issue in South Asia and sub-Saharan Africa is still very prominent. In 2015, the total number of extreme poverty population in these two regions accounted for more than 85% of the world poverty population. In addition (Data source: Poverty and Shared Prosperity 2018: Piecing Together Poverty Puzzle released by the World Bank), a sudden pandemic in 2019 has not only seriously hindered the global economic development but also slowed the process of global poverty reduction. Data show that there would be another new 97 million population living in extreme poverty globally in the year of 2020 (Data source: Updated estimates of the impact of COVID-19 on global poverty: Turning the corner on the pandemic in 2021? released by the World Bank). erefore, from the global perspective, the poverty reduction task is faced with challenges and difficulties and there is still a long way to go before achieving the 2030 poverty reduction goals as scheduled.
When Chinese President Xi Jinping visited Central Asia and Southeast Asian in 2013, he raised the initiatives of jointly building the Silk Road Economic Belt and the 21st-Century Maritime Silk Road (hereinafter referred to as the Belt and Road Initiative (BRI)), which have attracted close attention and extensive discussion from all over the world. In 2015, the National Development and Reform Commission of China, the Ministry of Foreign Affairs, and the Ministry of Commerce jointly issued Vision and Actions on Jointly Building Silk Road Economic Belt and 21st-Century Maritime Silk Road (hereinafter referred to as the Vision and Action).
e Vision and Action defined the principles, framework, cooperation priorities, and cooperation mechanisms, hence officially marking the start of implementation of the BRI. In fact, the establishment of the BRI is not to rebuild a Eurasian international trade network of the ancient silk routes but to create an open, shared, and inclusive international cooperation platform based on its cultural connotation [1]. From the perspective of the geographical region, the BRI covers a vast territory, involving China, Mongolia, Russia, Southeast Asia, South Asia, Central Asia, West Asia, North Africa, Central and Eastern Europe, and other regions, which consists of 39% of global land area, 31% of global GDP, and 62% of global population [2]. However, as observed from the indicators such as infrastructure conditions, agricultural modernization, industrialization, and per capita GDP, most countries along the BRI are still developing countries. erefore, how to effectively alleviate poverty is an important problem facing those regions.
With the continuous improvement and popularity of the BRI, China will actively integrate into the economic and social development of countries along the BRI and show the responsibility as a great power in regional poverty reduction. For one thing, since the adoption of reform and opening-up in 1978, China has accumulated rich experience in poverty alleviation and thus been able to provide Chinese wisdom through policies and mechanisms for countries along the BRI to decrease poverty population. For another, the enhancing infrastructure connectivity of the participating countries leads to more frequent exchanges and cooperation between China and countries along the BRI in terms of trade and investment [3], hence bringing the economic openness of the participating countries into a new level and reducing the uncertainties of development [4] and building confidence in global economic recovery [5]. As can be seen, the BRI will promote the orderly flow of economic factors in the vast Eurasian region and restructure trade and investment networks.
In fact, since the BRI was put forward in 2013, the Chinese government has been actively strengthening institutional cooperation with countries along the route, adhering to the spirit of openness, and carrying out wider, higher-level, and deeper-level regional cooperation [6]. By the end of July 2022, China had signed more than 200 cooperation documents with 149 countries and 32 international organizations (Data source: https://www.mfa.gov.cn/web/fyrbt_673021/ jzhsl_673025/202208/t20220818_10745415.shtml).
Of course, among the many cooperation documents, the most eye-catching is that China and the EU have fully reached the China-EU Comprehensive Investment Agreement (CAI). CAI benchmarks against international high-level economic and trade rules and is a balanced, high-level, mutually beneficial agreement [7][8][9], which will not only but will also strongly stimulate the recovery of the world economy in the post-pandemic era, promote the liberalization and facilitation of global trade and investment, and make important contributions to building an open world economy [10][11][12]. Relevant data indicates that in terms of trade, the total import and export volume between China and other countries along the BRI increased from USD 1.04 trillion to USD 1. 34 [13], "Chinese-style neocolonialism" [14], or "China's debt trap" [15], whose purpose is to expand the scope of forces and seize natural resources. en, will the promotion of the BRI have a significant impact on poverty reduction in countries along the route? If will, what is the mechanism behind it? As the BRI has enhanced trade openness and investment openness, is it possible to facilitate regional poverty reduction through the openness of these factor? Answers to these questions will help to ensure a higher-quality development of the BRI and promote the sustainable development of countries along the route.
Based on the analysis above, this paper conducts a quasinatural experiment. Based on the panel data of 134 countries worldwide from 2000 to 2018, this paper uses the differencein-difference method to examine the impact of the BRI on poverty reduction in countries along the route. Meanwhile, the mechanism of the poverty reduction effect of the BRI is further tested from the perspectives of trade openness and investment openness. is paper includes the following parts: e first part is the introduction, the second is the literature review and 2 Discrete Dynamics in Nature and Society research hypothesis, the third includes research methodology and data, the fourth is the empirical results analysis, and the last part is the conclusion and policy implications.

Literature Review and Hypothesis
e literature closely related to this paper involves two aspects. e first aspect is the impact of economic openness on poverty reduction. Poverty is a comprehensive problem formed by the interaction of multiple factors in a specific space-time context, to which worldwide governments and scholars have been committed to finding solutions. One of the solutions that advocates the further enhancement of trade openness and investment openness has become a key factor in achieving national poverty reduction [16][17][18].
In terms of the impact of trade openness on poverty reduction, Bhagwati and Srinivasan [19] first explored the relationship between the two, arguing that trade openness can stimulate national economic growth and thus effectively reduce regional poverty. Later, Dollar and Kraay [20], Tsai and Huang, [21] and other scholars generally followed this track for empirical analysis. Jenkins [22] believed that the income distribution effect brought by trade openness can help achieve effective poverty alleviation. In addition, some scholars have, by taking multiple perspectives into account, more systematically explored the mechanism of the impact of trade openness on poverty reduction. For example, Winters et al. [23] believed that trade openness could have a positive impact on poverty reduction through economic growth, family and market, wages and employment, and government income and expenditure. Heo and Doanh [24] took Vietnam as an example and found that trade openness could significantly reduce poverty in Vietnam through economic growth, enterprises, markets, and government. At present, as the issue of poverty has been studied even more extensively and intensively, some papers reveal the trade openness impact of poverty reduction and its mechanism from the perspective of multi-dimensional poverty. However, there are also scholars who believe that trade openness may exacerbate market competition, resulting in a sharp rise in unemployment among low-skilled workers and a gradual widening of the "Matthew Effect" between the rich and poor [25], and as a result, regional poverty will not be alleviated in a long-term period.
In terms of the impact of investment openness on poverty reduction, most scholars hold that foreign direct investment (FDI) has a significantly positive impact on poverty reduction in host countries. For example, based on panel data of 12 Vietnam provinces from 1992 to 2002, Hung [26] found that every 10% increase in foreign direct investment will lead to a 0.5% decrease in the poor population. Jalilian and Weiss [27], based on the study on the data of ASEAN country from 1997 to 2007, found that FDI could significantly increase income for the bottom 20% of the population. Later, Zhang [28], Nunnenkamp et al. [29], and Khan et al. [30] drew the similar conclusions when studying countries such as China, Bolivia, and Pakistan. e second aspect is that the implementation of the BRI has promoted the trade openness and investment openness.
Guided by the principle of "extensive consultation, joint contribution and shared benefits," China has conducted a series of trade and investment cooperation with countries participating in the BRI, which will better meet the development needs of countries along the BRI and further promote economic growth [31]. In terms of trade openness, most empirical literature shows that the BRI has continuously enhanced transport connectivity to countries along the route. For example, the China-Europe Railway Express will reduce transportation costs and trade time to a large extent [32], thereby further promoting trade openness and growth in countries along the BRI [33]. From the perspective of the whole areas covered in the BRI, Soyres et al. [34] found that the transportation infrastructure under the BRI will have a positive impact on the transportation costs and time of countries along the route by reducing 1.7%-3.2% and 1.5%-2.8%, respectively. On this basis, Baniya et al. [35] further analyzed how shorter trade time would affect bilateral trade and found that the BRI increased trade flows among participating countries by 4.1% and benefited time-sensitive products most. From the perspective of different regions along the BRI, Herrero and Xu [36] believed that the BRI strengthened freight links with European countries through the China-Europe Railway Express. is trans-continental railway transportation has the dual advantages of lower transportation cost and shorter trade time, thus playing an important role in promoting the development of trade between China and Europe. In addition, studies involving Southeast Asia [37] and South Asia [38] also show that the proposal of the BRI can significantly promote trade growth. erefore, under the BRI's cooperation framework, the continuous strengthening of trade links among the route will help countries achieve their goals of exploring endowment and comparative advantages, increasing employment of the poor, facilitating business transactions, and increasing income.
In terms of investment openness, many scholars have also conducted empirical studies on the impact of the BRI on FDI in countries along the route. He and Cao [39] constructed a bilateral FDI network of countries along the BRI and found that the advent of the initiative led to an expanding FDI network as well as closer investment links. Chen and Lin [40] discovered that the BRI will strengthen the cross-border transport network connection of countries along the route and reshape the spatial organization of crossborder investment by diverting 3% of the total FDI flows to participating countries.
To sum up, the BRI has been believed as an open and winwin cooperation platform and promotes the orderly flow and openness of international factors such as trade and investment by strengthening the connectivity of transportation infrastructure with countries along the route to achieve effective poverty reduction and high-quality development in the region. Figure 1 reflects the mechanism of the BRI on poverty reduction in countries along the route. e following to-betested research hypotheses are put forward.

Hypothesis 1.
e implementation of the BRI has a significant positive impact on poverty reduction in countries along the route.

Hypothesis 2.
e BRI can reduce poverty in countries along the route through trade openness and investment openness.

Model Building.
is paper proposes the BRI as a quasi-experiment, and difference-in-difference (DID) is used to evaluate its impact on poverty reduction in countries along the route. Learning from the practice of Kong et al. [41], we construct a two-way fixed effect DID model. e model can be expressed as where Poverty it represents the incidence of poverty of country i in year t and post t represents the dummy variable of the policy time effect. Since the BRI was put forward in 2013, 2013 is taken as the first year of policy advancement. Hence, post t is equal to 1 in 2013 and after the BRI, otherwise it is equal to 0. Treat i represents the dummy variable of the processing group. If the country belongs to the BRI, then Treat i is equal to 1, otherwise Treat i � 0. α 1 is the coefficient of the core independent variable. If the value is significantly positive, it means that the proposal of the BRI is conducive to reducing the incidence of poverty in countries along the route. X it denotes all control variables. c i is the country fixed effect. σ t is the year fixed effect. ε it is the error term.

Variables Selection.
Explained variable: incidence of poverty (Poverty). In consideration of data availability and comparability between countries, the incidence of poverty in this paper is based on the latest international standard of extreme poverty, published by the World Bank in 2015, that is, the proportion of the population living on less than USD 1.90 per person per day (purchasing power in 2011). e greater the indicator, the higher the incidence of poverty in the country. Core variable: In defining whether a country is involved in the BRI (post t ), this paper is based on the development priorities cited in the Vision and Actions on Jointly Building Silk Road Economic Belt and 21st-Century Maritime Silk Road and the scope defined by the People's Daily Online and the BRI Research Center of the Chinese Academy of Social Sciences, setting the scope of the BRI involves 66 countries including China. However, there is a lack of data in 12 countries including Afghanistan, the United Arab Emirates, Bahrain, Brunei, Bhutan, Cyprus, Kuwait, Laos, Maldives, Oman, Qatar, and Yemen, and 54 of them are selected as countries along the BRI. Meanwhile, 80 countries not along the BRI are selected as the control group. e list and classification of those countries are shown in Table 1.
Control variables: On the basis of existing empirical literature, the level of economic development (Economy), industrial structure (Structure), urbanization (Urban), transport development (Transport), education development (Education), and medical development (Medical) is screened out.
Economy: Most literature shows that the level of economic development has a significant negative impact on regional poverty. For one thing, the improvement of economic development has prospered the market and increased employment opportunities, making more jobs (especially non-agricultural jobs) available to the regional poverty population, thereby alleviating income poverty. For another, with the rapid development of regional economy, the government has increased financial expenditure on infrastructure, education, health care, and social security, thereby improving the living and production capacity of the poor population and alleviating the overall poverty in the region. e significant negative impact of economic development level on regional poverty is expressed by per capita GDP in this paper.
Structure: Ravallion and Datt [42] found that tertiary industry had the greatest impact on poverty reduction, followed by the secondary industry and the primary industry with the weakest impact. We conclude that the influence mechanism of industrial structure upgrading on poverty alleviation involved two aspects. First, the optimization and upgrading of industrial structure will transfer factors such as labor, capital, and science and technology innovation to higher industrial sectors, thus further promoting economic growth and reducing the incidence of poverty. Second, the upgrading of industrial structure will optimize the employment structure to a higher level in that the regional poor population will seek employment in the secondary and tertiary industries and get paid higher, thus reducing the incidence of poverty. In summary, this paper argues that the level of industrial structure has a significant negative impact on regional poverty, which is expressed by the ratio of the added value of the tertiary industry to that of the secondary industry.
Urban: ere is still no unified conclusion about the impact of urbanization level on regional poverty. On the one hand, rapid urbanization can directly promote the transfer of rural population to urban areas, thereby increasing non-agricultural employment opportunities for rural labor population and alleviating regional poverty. At the same time, urbanization can promote economic development, industrial structure upgrading, and agricultural production efficiency, thereby indirectly alleviating poverty. On the other hand, due to the influx of a large number of rural labor force into the city, some production factors in rural areas will gather in urban areas, resulting in sluggish rural economic development. In   Discrete Dynamics in Nature and Society addition, the migration of rural people to cities leads to accelerated urbanization, which will not only put strains on current resources but also produce serious ecological environment problems, even more highlighting urban poverty problems. erefore, with both positive and negative effects existing, the urbanization level often has a nonlinear impact on regional poverty, which is expressed by the proportion of urban population in the total population in this paper. Transport: Due to a lack of port and aviation data at the national level, this paper focuses on the impact of roads and railways on global poverty reduction. As the two important types of transportation infrastructure vary greatly in mileage, volume, and flexibility, they are discussed in a separate way. To eliminate the impact brought by land areas in different countries, this paper selects the density of highway network (Road) and the density of railway network (Rail) as specific measurement indicators.
Education and Medical: Education and medical care are closely related to people's lives. Education is an effective way to improve the quality of human capital and a fundamental way to block intergenerational poverty, which is expressed by the public education expenditure per capita in this paper. Medical care is to improve the health level of human capital and is conducive to strengthening the medical and health protection of the poor, which is expressed by the public health expenditure per capita in this paper. e descriptive results of variables are shown in Table 2.
3.2. Data. Socio economic statistics are from multiple sources. GDP, per capita GDP, total population, employed population, added value of primary, secondary, and tertiary industries, total import and export trade volume, urban population, land area, total railway mileage, public education expenditure and medical expenditure, incidence of poverty, and other data are from the World Bank's public database; the total road mileage is from the World Road Statistics published by the International Road Federation (IRF), the International Statistical Yearbook compiled by the National Bureau of Statistics, and the World Affairs Almanac published by the World Affairs Press, and the stock data of foreign direct investment is from the database of the United Nations Conference on Trade and Development (UNCTAD). Since the March of the 21 st century, countries around the world have gradually increased their attention to poverty issue and implemented various poverty reduction strategies. erefore, this paper sets the time of the chosen sample from 2000 to 2018 and includes 134 countries, 54 of which belong to countries along the BRI. As observed from the whole sample, the total GDP of the sample in 2018 accounts for 97.58% of the global GDP, and the total population accounts for 99.16% of the global population.

Overall Regression Analysis.
Based on the macroeconomic data of 134 countries worldwide from 2000 to 2018, the DID model of the two-way fixed effect is used to empirically test the impact of the BRI on poverty reduction. To make comparisons, this paper adds control variables in turn, as shown in Model 1-Model 8 of Table 3. e coefficient of did in Model 1 is −0.439 and passes the 1% significance test, while the coefficient of did in Model 2-Model 8 decreases significantly and stabilizes at around −0.26 after adding control variables. is fully demonstrated that the BRI could significantly reduce the incidence of poverty of countries along the BRI, but this poverty reduction effect could be overestimated without adding control variables. us, hypothesis 1 is proved. e regression results of control variables show that the level of economic development, industrial structure, urbanization, road development, education development, and medical care development contributes to poverty reduction, while the level of railway development has no significant impact on the incidence of poverty.

Parallel Trend Test.
e premise of the DID model is that before a policy event occurs, the change trends in the treatment group and control group should be consistent. According to the research methods of Beck et al. [43], this paper selects the three years before and after the implementation of the BRI in 2013 to further observe the change trends of the treatment group and the control group. If DID is not obvious before the implementation of the BRI, there will be no significant difference between the treatment group and the control group, and thus the parallel trend hypothesis will be proved. 6 Discrete Dynamics in Nature and Society As shown in Table 4, after adding control variables in turn, all regression results of DID before 2013 in Model 1-Model 7 do not pass the significance test. In addition, in 2013 and beyond, all regression coefficients of DID pass the significance test, and the influence coefficient is significantly smaller than that before the implementation of the BRI (Figure 2), which fully showed that the development trends of the treatment group and the control group are consistent before 2013, while the development trends of the two groups are quite different after the implementation of the BRI. erefore, the sample passes the parallel trend test.

Placebo Test.
Referring to the research method of Cai et al. [44], the placebo test is conducted by randomly selecting the same number of countries along the BRI from the full sample. erefore, this paper randomly selects 54   Note. * , * * , and * * * indicate that statistics are significant at the 10%, 5%, and 1% level of significance, respectively.   Note. * , * * , and * * * indicate that statistics are significant at the 10%, 5%, and 1% level of significance, respectively. Discrete Dynamics in Nature and Society countries from the full sample of 134 countries as the treatment group and considers other countries as the control group. e year of policy implementation is set at 2013. Since the processing group is randomly generated, the regression coefficient of the core explanatory variable did would not be likely to pass the significance test. To avoid errors in the estimation results caused by small probability events, 500 repeated tests are conducted for the above process. Figure 3 reflects the estimated coefficient of 500 randomly selected treatment groups and the distribution map of the corresponding p value. It can be clearly found therefrom that the mean value of the regression coefficient is close to zero, at −0.003, and most p values do not pass the test level of 1%. At the same time, the vertical line in Figure 3 represents the actual estimated coefficient of the sample, which is obviously abnormal. Overall, the missing variables do not cause any errors to the empirical results in this paper. Among the countries along the BRI, China and India face the toughest poverty issue as a result of a large population scale. Since the turn of the new century, the two countries have been attaching higher importance to poverty reduction and contributed a great deal to the wider efforts of global poverty reduction. To prevent the interference of these two abnormal samples on the results of this paper, the samples of these two countries are deleted, and the remaining countries along the BRI are estimated. If the coefficient of did is still significantly negative, the robustness of the empirical results of this paper would be further proved. According to Model 1-Model 4 in Table 5, the coefficient of did is significantly negative whether China or India is deleted or not. At the same time, the regression coefficients of did in Model 2 and Model 4 are −0.228 and −0.252, respectively, while the coefficient of did in the total sample above is −0.255, indicating that compared with the absence of India, the absence of China will significantly reduce the impact of the BRI on poverty reduction in countries along the route.

Endogeneity Test.
Due to the non-randomness of the Chinese government's choice of countries to implement the BRI, this may lead to the estimation of the impact of the implementation of the BRI on the incidence of poverty to be affected by endogenous issues. In order to deal with the endogenous problem, this paper further uses the instrumental variable method to estimate. Drawing lessons from the "Six-Dimensional eory of Culture" (Six-Dimensional eory of Culture includes Long-Term Orientation/Short-Term Orientation), Indulgence/Constraint, Right Distance Index, Individualism/Collectivism, Masculinism/Feminism and Uncertainty Avoidance) proposed by Hofstede [45], the cultural distance (we use the model constructed by Kogut and Singh [46] to measure the cultural distance) between countries is selected as an instrumental variable. e rationality of this instrumental variable is as follows: e smaller the cultural distance between a country and China, the stronger the cultural identity of the two countries, and the more likely it is to join the BRI, that is, to meet the relevance conditions; in addition, cultural differences are caused by differences in the cultural and historical backgrounds, social systems, and inheritance of the respective nations, which are relatively stable and do not change much over time, so they do not have a direct impact on the incidence of poverty in each country in the sample period. We use the two-stage least squares (2SLS) model to identify this endogenous problem. Model 1 in Table 6 is the regression result of the first stage: e estimated coefficient of iv is significantly positive at the 1% level, indicating that the greater the cultural distance from China, the lower the possibility of joining the BRI, that is, the correlation condition is met. At the same time, the F statistic is greater than the critical value of 10, indicating that there is no weak instrumental variable problem; Model 2 in Table 6 is the regression result of the second stage: e estimated coefficient of did is significantly negative, which shows that after considering the endogeneity problem, the BRI still significantly reduces the incidence of poverty in countries along the route.

Mechanism Analysis.
Based on the above analysis, this paper argues that the BRI can have a positive impact on poverty reduction by elevating trade openness and investment openness of countries along the route to a higher level.  Discrete Dynamics in Nature and Society erefore, referring to the research results of Tang et al. [47], a mediation effect model is constructed to examine the mechanism of the BRI on poverty reduction. e test model for hypothesis 2 is as follows: lnPoverty it � α 0 + α 11 did it + α 12 lnX it + c 11 + σ 11 + ε 11 , lnMed it � β 0 + β 11 did it + β 12 lnX it + c 12 + σ 12 + ε 12 , lnPoverty it � c 0 + c 11 did it + c 12 lnMed it + c 13 lnX it where Med it represents the intermediary variables, involving the trade openness level (Trade) and investment openness level (FDI), which are expressed as the proportion of the total import and export trade in the GDP and the total foreign investment in GDP of each country. e mediation effect of did it is equal to β 11 × c 12 . e mediation effect accounted for the total effect proportion is equal to (β 11 × c 12 )/(β 11 × c 12 + c 11 ). Table 3 reports that the BRI has a significant negative impact on the incidence of poverty in countries along the route, that is, the estimated coefficient α 11 is significantly negative, indicating that the mediation effect test can be carried out. Model 1 and Model 2 of Table 7 report whether the BRI will affect poverty reduction in countries along the route through trade openness. In Model 1, β 11 is significantly positive, indicating that the BRI does promote the level of trade openness of countries along the route; furthermore, when trade openness is introduced into the poverty reduction equation, both c 11 and c 12 in Model 2 pass the significance test, so the Sobel test is unnecessary. Overall, the mediation effect of trade openness is -0.013, accounting for 5.098% of the total effect, which shows that the BRI can affect poverty reduction in countries along the route through trade openness. Hypothesis 2 is partly proved. Table 7 report whether the BRI will affect poverty reduction in countries along the BRI through investment openness. In Model 1, β 11 is not significant, indicating that the BRI has no significant impact on the investment openness of countries along the route; furthermore, by introducing investment openness into the poverty reduction equation, both c 11 and c 12 in Model 3 pass the significance test and Sobel test. From this, it can be found that the mediating effect of investment openness is −0.003, accounting for 1.176% of the total effect, which indicates that the BRI can affect poverty reduction in countries along the route through investment openness. Hypothesis 2 is finally proved.

Heterogeneity Analysis.
Heterogeneity analysis is conducted mainly from two dimensions: First, according to the geographical location, countries are separated into coastal countries and landlocked countries. Second, according to the classification of national income level by the World Bank, countries are separated into high-income countries (per capita national income over USD 12,376 or higher), middle-high-income countries (per capita national income between USD 3,996 and USD 12,375), middle-low-income countries (per capita national income between USD 1,026 and USD 3,995), and low-income countries (per capita national income about USD 1,025 or lower). e results of heterogeneity analysis are shown in Table 8.
In Table 8, the regression coefficients of did in Model 1 and Model 2 are −0.383 and −0.229, respectively, which pass the 1% significance test, indicating that the implementation of the BRI has a positive impact on poverty reduction in landlocked countries and coastal countries along the route. Specifically, the BRI has a stronger impact on poverty reduction in landlocked countries than in coastal countries, which means that the opening of China-Europe Railway Express has widen the openness of landlocked countries and provided them with greater economic development potential.
In Table 8, the regression coefficients of did in Model 3-Model 6 are 0.150, −0.038, −0.524, and −0.259, respectively, with only Model 4 failing to pass the significance test, indicating that the implementation of the BRI has a positive impact on poverty reduction in low-and middle-income countries and low-income countries, and yet has a negative impact on high-income countries. In addition, its impact on middle-and high-income countries was not obvious. Note. * * * indicates that statistics is significant at the 1% level of significance. Note. * * * indicates that statistics is significant at the 1% level of significance.
Discrete Dynamics in Nature and Society erefore, the impact of BRI on poverty reduction is propoor. For middle-and low-income countries and low-income countries along the BRI, to meet their own development needs, they are eager to join the BRI to obtain development opportunities. At the same time, the BRI strengthens the connectivity of transportation infrastructure in developing countries along the BRI, upgrades the level of openness, and transforms their resource advantages into economic advantages, thus achieving prominent poverty reduction results. For high-income countries, absolute poverty (the quantity of poverty) has been solved, but relative poverty (the quality of poverty) should be focused and alleviated. erefore, under the BRI, they should make an effective use of international factor resources to achieve high-quality economic development and improve the welfare of people. For middle-and high-income countries, there is only one step to join in the rank of high-income countries, and yet they tend to fall into the middle-income trap. In addition, when some middle-and high-income countries optimize the development of industrial structure, their social employment and people's living standards may be affected to some extent, thus they should also take advantage of the BRI to achieve their further development.

Conclusions and Policy Implications
ere are still more than 700 million extremely poor people in the world, and a considerable part of them live along the BRI. How to effectively lift themselves out of poverty has become a challenge for most developing countries along the BRI. As pointed out in the Joint Communique of the Roundtable Summit of the Belt and Road Forum for International Cooperation signed at the first Belt and Road Forum for International Cooperation (BRF), "all countries, especially developing countries, still face common challenges such as eradicating poverty, promoting inclusive and sustained economic growth, and achieving sustainable development." e joint construction of the BRI is an open international cooperation platform, and its many concepts and practical measures have provided practical opportunities for promoting poverty reduction and governance in the countries along the BRI. In this context, it is of great theoretical and practical significance to assess the impact of the BRI on poverty reduction in countries along the route. Based on the panel data of 134 countries around the world from 2000 to 2018, we propose the 2013 BRI as a quasi-natural experiment, using the DID model and mediation effect model to examine the impact and mechanism of the BRI on poverty reduction. e main conclusions are as follows: First, the study finds that the BRI has an impact on poverty reduction in countries along the route, with the impact coefficient stable at −0.26, and the empirical conclusion has passed the parallel test, placebo test, and instrumental variable test. Second, mechanistic analysis results show that the BRI can reduce the incidence of poverty in countries along the route by promoting trade and investment openness. Finally, heterogeneity analysis shows that the BRI has a slightly greater role in promoting poverty reduction in inland countries along the route than in coastal countries. In addition, the BRI has a positive impact on the poverty reduction performance of low-and middle-income countries and low-income countries, a negative impact on high-income countries, and a non-significant impact on middleand high-income countries.  Note. * and * * * indicate that statistics are significant at the 10% and 1% level of significance, respectively.
Based on the above conclusions, we put forward policy implications. First, the BRI originated in China, but the opportunities and results are shared by all countries. At present, we should let go of the prejudice against China. e BRI is not a "Chinese version of the Marshall Plan," "Chinese-style neocolonialism," or "China's debt trap." It is an inclusive, open, and sharing platform, and it is also a goaloriented development governance model, which differs from traditional standardized governance and is more inclusive. e degree of recognition of the BRI is a prerequisite for the effective and long-term service of the development of the areas along the route.
Second, facility connectivity is the foundation for cooperation and development of countries along the BRI. From the current point of view, the BRI is faced with a series of obstacles to transportation infrastructure, such as sparse transport network layout, imperfect port facilities, and complicated customs clearance procedures. erefore, it is difficult to promote the orderly flow of factors such as trade, investment, and personnel in the countries along the route so that the intermediary effect of trade and investment in this paper is relatively small. Under this situation, on the one hand, it is necessary to establish a more complete and sound infrastructure cooperation mechanism, further optimize the customs clearance process, improve the efficiency of customs clearance, and reduce the obstacles that hinder the cross-border flow of factors; on the other hand, transport technology facilities in the countries along the route, especially in Central Asia, West Asia, and other regions, need more efforts in planning layout, engineering construction, speed-up and upgrading, etc.
Finally, there are significant differences in the stages of poverty reduction in different types of countries along the BRI, so there should be significant differences in development methods. For most developing countries along the route, which are still in the initial stage of poverty reduction, they are solving the problem of absolute poverty. erefore, more emphasis should be placed on the development of economic scale and the improvement of people's living standards. For developed countries, they are in the late stage of poverty reduction and alleviate the problem of relative poverty. erefore, more emphasis should be placed on high-quality economic development and the improvement of people's welfare.

Data Availability
All data used to support the findings of the study are available from the corresponding author upon request.

Conflicts of Interest
e authors declare no conflicts of interest.