In this chapter we focus on the intended and unintended side effects of Chinese development finance, which we address from two different angles, both at the country level and at subnational scales. In the first part of this chapter, we investigate whether and to what extent Chinese-financed aid and debt affect recipient countries’ propensity for civil conflict and environmental degradation. We then turn to the question of how Chinese-funded projects might affect the quality of governance in recipient countries and regions. China claims to follow a policy of non-interference in the domestic affairs of sovereign governments, which implies that its allocation decisions are made without considering the quality of governance, so that Chinese funds might prop up rogue regimes and delay much-needed governance reforms. The second part of this chapter turns to the popular claim that significant financial support from China impairs the effectiveness of aid from Western donors and lenders. Specifically, we investigate whether the effects of World Bank aid differ in countries or subnational regions that receive large volumes of Chinese support compared to other recipients.

This chapter reviews the first six years of implementation of the Belt and Road Initiative (BRI), a major Chinese foreign policy initiative introduced in 2013. The authors explain how China’s transition from benefactor to banker, in con- junction with its push for expanded influence on the global stage, led to the adoption of the BRI. They then consider whether and why China might choose to ‘multilateralize’ the BRI. The authors conclude that if Beijing wants to multilateralize the BRI, it will need to either comply with—or help redesign— international development finance rules and standards. At the same time, the establishment of an inclusive and revitalized development finance regime does not rest solely on the shoulders of Beijing. If OECD-DAC and multilateral donors and creditors wish to avert a crisis of confidence and relevance, they will need to rewrite international development finance rules and norms in ways that accommodate Beijing’s interests and more effectively account for the preferences of low-income and middle-income countries.

This chapter investigates the motivations of recipient governments who seek out Chinese development projects. We examine how host country leaders shape the ways that Chinese development finance is allocated across subnational jurisdictions. First, we evaluate whether China’s allocation of aid and debt within countries flows to areas with higher levels of socioeconomic need. Second, we explore whether political leaders manipulate incoming financial flows from China to advance their own political interests. We do so by examining whether and when funds from Beijing favor the home provinces of political leaders. Finally, we compare and contrast the ways in which Chinese and World Bank development projects are subnationally distributed. To do so, we assign latitude and longitude coordinates to the specific locations where China implemented its de- velopment projects.Our findings indicate that Chinese development finance does not go to the geographic areas within recipient countries where it is most needed: much of it ends up in wealthier provinces. In addition, Chinese development projects favor politically privileged jurisdictions: the home provinces of political leaders receive substantially more Chinese development finance in countries with competitive elections, and even more at election time. This is a problem from a development perspective because the average home province of a political leader is significantly wealthier than the rest of the country.

This chapter introduces the central argument of the book: that China’s 21st- century transition from a “benefactor” to a “banker” has had far-reaching im- pacts in low-income and middle-income countries that are not yet widely understood. Beijing’s growing use of debt rather than aid to bankroll big-ticket infrastructure projects has created new opportunities for developing countries to achieve rapid socioeconomic gains, but it has also introduced major risks, including corruption, conflict, and environmental degradation. Some countries are more effective than others at managing these risks and rewards. This chapter “zooms in” on two countries—Sri Lanka and Tanzania—to illustrate the tension between efficacy and safety confronted by developing countries banking on Chinese development finance. It also provides a roadmap for the rest of the book.

The dataset introduced in this book includes thousands of Chinese government- financed development projects. This chapter begins to analyze these data at the cross-national level and addresses a basic question: Which types of projects does China finance around the world? We provide a detailed overview of the allocation of Chinese development finance based on key variables such as destination countries, flow types (such as grants, loans, technical assistance, debt forgiveness, or sectors). The dataset enables us to distinguish between Chinese-financed aid and debt, and this distinction reveals that China’s donor-to-banker shift occurred in the 2000s following the implementation of the “Going Out” strategy. More generally, this chapter uses our new dataset to demonstrate the value of separating out aid and debt projects and show how different countries have different experiences receiving Chinese projects. In providing readers with an aerial global view of the “known universe” of China’s international development finance projects since 2000, the chapter also situates China’s development finance in the context of that of other major donors.

This chapter examines the effectiveness of Chinese development finance. At the recipient-country level, we test the impact of Chinese development finance on economic growth, infant mortality, and the spatial concentration of economic activity. We then move below the country level and investigate the economic development effects of China’s development finance at the subnational level using luminosity data at fine spatial resolution (in addition to infant mortality and spatial concentration). We disentangle differences between Chinese aid and debt and compare these effects to those of World Bank funding. In addition, this chapter analyzes whether the motivational forces that shape the provision of Chinese development finance affect downstream development outcomes in recipient countries and regions. The empirical evidence presented in the chapter shows that, irrespective of political bias, Chinese aid and debt improve socio- economic outcomes at both national and subnational scales. However, these impacts vary significantly across jurisdictions. We also find that socio-economic impacts of Chinese development projects are comparable, if not superior, to those generated by the World Bank.

In this book, we introduced and analyzed the 1.0 version of AidData’s global dataset of Chinese ODA- and OOF-financed projects. Shortly after we completed our manuscript (Chapters 1–9), we collaborated with AidData to publish a new (2.0) version of the dataset based upon an improved version of the TUFF methodology (Custer et al. 2021). The 2.0 dataset, which was published in September 2021, captures 13,427 Chinese ODA- and OOF-financed projects worth US$843.1 billion across 165 countries. These projects were financed by more than 330 Chinese government institutions and state-owned entities. The dataset covers projects approved (i.e., officially committed) between 2000 and 2017—and implemented between 2000 and 2021—in every low-income, lower-middle income, and upper-middle income country and territory across Africa, Asia, Oceania, the Middle East, Latin America and the Caribbean, and Central and Eastern Europe.

In this chapter, we use the new data to investigate which factors influence the allocation of Chinese development finance across countries, and how these motivations compare to those of traditional donors and creditors such as the World Bank. We specifically test the claim that Beijing is primarily using its economic largesse to cement alliances with political leaders, secure access to natural re- sources, and create exclusive commercial opportunities for Chinese firms. We run statistical models to examine the cross-country allocation of Chinese aid and debt, arguing that these two flows are motivated by different strategic objectives. Our results show that Chinese aid and debt are means to different ends. Like Western donors, Beijing allocates aid in response to levels of poverty and uses aid to reward countries that adopt its foreign policy positions and punish those that do not.Contrary to the conventional wisdom, Beijing is no more likely than major Western donors to provide aid to corrupt or authoritarian regimes. None of these findings are consistent with the notion that China is a “rogue donor” compared to its Western peers. However, Beijing’s loans issued at or near market rates are more likely to go to corrupt and authoritarian countries. The conflation of aid and credit seems to be at the heart of the confusion: Western politicians and journalists use “rogue donor” as a shorthand term for all of China’s overseas development projects, but many of these projects are not aid.

Social science research on the aims and impacts of Chinese development finance remains in its infancy because Beijing shrouds its overseas portfolio of grants and loans in secrecy. This chapter introduces the Tracking Underreported Financial Flows (TUFF) methodology that the authors have developed to assemble a comprehensive dataset of Chinese aid and debt-financed development projects around the globe. It also provides an overview of previous attempts to quantify Chinese development finance, and explains how the authors’ methods and data are different from those of others. This chapter also tests whether an alternative approach—field-based data collection—might yield more useful and reliable re- sults. Drawing upon evidence from a “ground-truthing” exercise in Uganda and South Africa, the authors demonstrate that field-based and TUFF-based data collection methods produce similar results. However, the TUFF methodology is less vulnerable to detection bias and more readily scalable than field-based data collection.