Profit or Power: Determinants of Chinese Financing to the Developing World

ABSTRACT

Since the launch of its ‘going out’ strategy in 1999, China has become the largest bilateral sovereign lender to the developing world. This article argues that, since Xi Jinping came to power in 2012, China has used overseas financing to developing countries to consolidate a strategical wedge between certain borrowing countries and the United States, China’s main geopolitical rival. Using a Heckman two-stage estimation, this article shows how China, albeit providing financing to a wide array of developing nations, lends more and on more favourable terms to countries that are farther away from the United States in terms of foreign policy preferences. Statistical analyses on a sample of 148 developing countries from 2000 to 2017 provide support for the hypotheses.

Introduction

China committed $841 billion in loans to 135 developing countries between 2000 and 2017.¹ Of this amount, $426 billion were committed from Xi Jinping’s coming to power—first as Secretary-General of the Chinese Communist Party (CCP) in 2012, then as president of the People’s Republic of China (PRC) in 2013. Under Xi, China has loaned $64 billion to Russia, $37 billion to Venezuela and $10 billion to Iran (Figure 1). Other developing countries—many of doubtful creditworthiness or besieged by international sanctions—also found in China a financier willing to lend tens of billions of dollars without imposing the strict economic policy conditionality that Western governments and multilateral institutions normally require. Why does China lend tens of billions of dollars to regimes that run a high risk of never repaying it?

Figure 1. Top ten recipients of Chinese financing under Xi Jinping.

Note: Author’s elaboration with AidData (Custer and others, 2021). All amounts are in billions of constant 2017 US dollars.

This article argues that, since Xi’s accession as PRC leader, China has offered financing in greater volumes to developing countries that are the least aligned with the United States in terms of foreign policy, in pursuit of a reward-based wedging strategy.² China uses the most cost-effective means at its disposal, the vast reserves of foreign currency accumulated from exports, as an economic inducement to maintain a separation between the developing countries and the United States, China’s main geopolitical rival. This strategic motivation, thus far neglected by the literature, is complementary to the commercial motivations identified by the scholarship on China’s overseas lending programme.

The literature has given different answers to the question of why China lends so much to uncreditworthy borrowers. Some authors argue that China is a commercial lender as any other, and lends badly due to lack of experience and sophistication as a lender. As a relative newcomer to the big lenders club, China would be making the same mistakes those lenders made in the past.³ This view of China as a ‘naïve lender’ is hard to square with the overall economic competence of the Chinese bureaucracy, which has not only presided over China’s impressive economic growth since the late 1970s, but also has successfully weathered several global and domestic economic crises. Unlike the ‘naïve lender’ theory, this article argues that China intentionally lends to risky borrowers.

Other authors see a more sinister motive in China’s lending to risky borrowers: China would be saddling poor countries with unpayable debt in order to take control of critical infrastructure given as collateral for those loans—the so-called ‘debt-trap diplomacy’ hypothesis.⁴ This explanation has been countered by authors who have found no evidence of a deliberate Chinese strategy to entangle other countries in debt—as a matter of fact, China often attempts to renegotiate bad debts to avoid the seizure of collateral assets.⁵ Unlike the ‘debt trap’ hypothesis, this article does not claim China wants its loans to go unpaid.

This article makes two theory contributions to the literature on China’s overseas development financing.⁶ First, while other studies show that geopolitical alignment with China does not drive an increase in the value of Chinese loans⁷; this article argues that China’s foreign policy aims to consolidate existing divergences from the United States. International status is a relational measure, and China’s relative standing is enhanced by lowering that of its main rival. By appearing to buy the ‘neutrality’ of third countries, China is contributing to raise its relative status. The measures of foreign policy affinity considered in the literature so far have overlooked this nuance, which may have resulted in the inaccurate conclusion that geopolitics does not matter for China’s overseas lending programme.

Second, this paper argues that leadership matters. China’s geoeconomic strategy is grounded on the ‘going-out’ strategy launched under Jiang Zemin (1993–2003) and continued by Hu Jintao (2003–2013), but the implementation under Xi differs markedly. While increased lending to ‘anti-American’ regimes was not clear before Xi came to power, after his accession it has become evident (Figure 2).

Figure 2. Plot of loans/total by UNGA. Lower values of UNGA indicate more similarity with the United States’ voting behaviour at the UNGA.

Empirically, this article proposes a different way of looking into Chinese lending data. Most studies of Chinese lending have followed the distinction established by the OECD Development Assistance Committee (OECD-DAC) between concessional and non-concessional loans. This distinction is unhelpful and potentially misleading when it comes to China. China does not abide by the OECD-DAC’s rules. Instead, the concessionality of Chinese overseas financing must be considered as existing on a spectrum. In addition, this article argues that each country has a ‘price tag’: it costs less money to buy strategic influence over Burundi, which had a GDP of $2 billion in 2012, than over Brazil, with a 2012 GDP of $2.4 trillion.⁸ Burundi and Brazil, nonetheless, have the same one vote in international organisations such as the United Nations.

This article is organised as follows: the next section reviews the literature on Chinese overseas financing, situating it within the greater debate on Chinese economic statecraft, and focusing on China’s geopolitical competition with the United States. The following section sets out hypotheses and the theory that underpins them, after which comes a description of the research design and a discussion of the results, robustness checks and alternative explanations. The last section concludes.

Literature Review

Two strands of scholarship are relevant for the argument developed in this paper: first, the study of cost-effective geoeconomic strategies and how China applies them within its economic statecraft; and second, the literature on the determinants of Chinese overseas financing. Baldwin’s (2020) work on geoeconomics emphasises that, as rational actors, countries will choose the least costly means available to achieve geopolitical objectives through economic instruments.⁹ There is evidence that China deliberately targets smaller, more vulnerable—and therefore, less costly—peripheral countries in its geoeconomic pursuits.¹⁰ For example, MacDonald shows that China has offered economic inducements to small countries in the Caribbean to secure their support against the United States in international forums at a relatively low cost.¹¹

Domestic channels facilitate the connection between China’s foreign policy preferences and its economic actions. The CCP, which wields control over the commanding heights of the Chinese economy and can mobilize state and private resources to achieve its geopolitical ends. Reilly, for instance, shows that to secure support, a foreign policy initiative must appeal not only to strategic and diplomatic constituencies within China’s bureaucracy but also to actors prioritising economic goals, and that China’s banking system plays a pivotal role in such arrangements.¹² As government-owned institutions, China’s biggest lenders such as the China Development Bank and China Eximbank direct capital towards politically desirable projects and borrowers. These banks’ mandates extend beyond loan profitability; they also serve political and geopolitical objectives.¹³ Under Xi Jinping’s leadership, CCP committees within privately owned companies have directly influenced decisions regarding foreign operations. This enables the party to impose its priorities on state-owned firms directly.¹⁴

Officials with a political remit are involved in the origination and approval of Chinese sovereign loans. Often, these loans are initiated through requests from politicians and high-level executive officials in borrowing countries. These requests are handled by its embassies in the borrowing countries, and processed by China’s Ministry of Commerce before being assigned to a Chinese policy bank or lending agency.¹⁵

While the literature on China’s economic statecraft establishes a clear link between power and profit, whether geopolitical factors determine Chinese overseas financing is less clear. There is some agreement that concessional financing, such as grants or loans on favourable terms, known as official development assistance (ODA) in the OECD-DAC jargon, often accompanies policy concessions, while non-concessional loans, categorized as other official flows (OOF), are less motivated by geopolitical gain.¹⁶ Other studies also link foreign aid, i.r., concessional transfers, to foreign policy similarity.¹⁷ At the same time, it has been found that China’s aid is small compared to traditional donors’, while China’s commercial lending surpasses that of other sovereign creditors.¹⁸

Empirical studies of Chinese lending tend to focus on the demand side, and find few factors to be significantly associated with Chinese loans. Some countries turn to China because it does not impose strict economic policy or governance conditions and is more lenient towards unreliable payment practices.¹⁹ Thus, countries with budget deficits, low credit ratings, or suffering from economic mismanagement find China an attractive source of financing. Kaplan, for example, notes a correlation between Latin American borrowers’ budget deficits and their overseas debt owed to Chinese policy banks, and that countries with lower levels of institutionalisation are more likely to comply with the ‘buy Chinese’ requirement since they can easily set aside public disclosure or procurement rules and renege on international commitments without facing significant consequences.²⁰ Furthermore, several studies note a link between Chinese financing and local corruption on the borrowers’ side.²¹ Finally, scholars have found that certain macroeconomic conditions, such as GDP per capita in the borrowing country,²² global commodity prices, global growth rates, and U.S. interest rates,²³ can influence Chinese overseas lending.

The literature provides valuable insights into the economic and political dynamics of Chinese overseas lending, but it overlooks how China’s geopolitical priorities also affect its lending behaviour. What is puzzling about China’s pattern of lending is not that it lends out the funds it has accumulated—as one should expect that it would –, but that it lends so much and so cheaply to countries from which other lenders stay away. Given China’s competition with the United States, which has intensified under Xi Jinping, and the significant control the CCP exercises over economic actors, it is unlikely that Chinese rulers would disregard its geostrategic considerations when financing projects in the developing world. Taking a comprehensive view of Chinese financing, considering both concessional and non-concessional loans, and comparing the years since Xi became the PRC’s paramount leader with the preceding years, reveal the importance of geopolitical factors in China’s lending decisions.

Theory: Market and Might

China’s Wedging Strategy

China has pursued a wedging strategy to enhance its status relative to its main geopolitical rival, the United States. This strategy builds on initiatives launched by Xi Jinping’s predecessors, but has become discernible following Xi’s accession to the office of CCP Secretary-General in late 2012. China has relied on economic inducements—larger loans and better terms—to keep developing countries distanced from the United States. The most efficient instrument for China to pursue this strategy is to offer financing cheaply to more vulnerable developing countries, in particular those countries that have the most difficulty accessing credit in the traditional markets and that already display a measure of anti-Americanism in their foreign policy (Figure 1). That has the additional advantage of putting to use the vast foreign currency reserves that China has accumulated due to its export-led development model since the beginning of the reform and opening up era. With its wedging strategy, China hopes to consolidate its status as a great power and near-peer to the United States and make the international system safer for its authoritarian regime.²⁴

Wedging is a ‘divide-and-balance’ fragmentation strategy that consists of the use by a state of instruments of power—including economic instruments—to dissuade a target state from aligning with a rival and thus prevent the formation of an opposing coalition.²⁵ By wedging, a state tries to turn the target—a potential opponent—into a neutral party or an ally at a lower cost than that of engaging in military action or alliance building. Wedging can be used to reinforce a target state’s neutrality.²⁶

Wedging is thus a cost-effective strategy that neutralises the lesser threat represented by the target as compared to the rival state or degrades the target’s level of cooperation with that rival. The cost consideration is especially relevant when comparing reward-based to coercive wedging: a state is likely to offer inducements as opposed to threats when it has enough reward power to offer more than its rival in order to attract its target.²⁷ Recent scholarship has explored how China seeks to wedge against the United States’ influence, especially in China’s near-abroad.²⁸

China prefers wedging through economic inducements for developing countries as a strategy because, as a still-developing authoritarian country, China suffers from an ‘inability to win durable partners among the advanced democracies’²⁹ and lags behind the United States in the other areas, notably the military and ideational realms.³⁰ Attracting developed countries that are allied with the United States is a tall order for China: neither is China’s state-run development model, based on comparatively cheap labour, export-oriented manufacturing and massive infrastructure investment, attractive to countries that are already rich, nor is its autocratic (and nominally communist) regime attractive to liberal democracies. China’s unique political economy and ‘the way in which government and business relations have been structured cannot be readily replicated in other countries’.³¹ Similarly, ‘flipping’ a developing country that is closely aligned with the United States is expensive, and outright impossible in some cases.

For China, which must put its vast reserves of foreign exchange to use, it is more efficient and less risky to reward poorer countries that already show some level of divergence from the United States in their foreign policies by offering them financing than to engage in a military confrontation or soft-power contest with the United States. Targeting the ‘weakest links’ is the least costly way of implementing China’s wedging strategy and gaining ground relatively to the United States. Thus, Chinese lending should target especially vulnerable developing countries: those that have more difficulty in accessing traditional markets because they have been sanctioned by Western governments, failed to implement effective corruption controls, defaulted on their debt, or just mismanaged their economy. To be clear, this does not mean that Chinese lending is entirely uncommercial. Countries such as Russia, Venezuela and Iran are not only governed by anti-American regimes, but they are also important energy suppliers of crude oil, natural gas and petroleum products to China. However, this paper shows that, even controlling for the effect of oil exports and trade with those countries, the amounts China lends to them stand out from countries that are more aligned politically with the United States.

In the past decade or so, China has become a more vocal actor in the United Nations, exercising its veto in the Security Council more often, and asserting more vehemently its vision of a world of sovereign states in which human rights and foreign intervention in domestic affairs have no place.³² China has sought to gain influence over the United Nations system via staffing appointments,³³ and natural resources imports from poorer states,³⁴ often as a way to oppose the United States’ preferred policies.

China’s strategic goal to assert itself as a near peer to the United States complements its economic goal to put its reserves to productive use: ‘one of the most distinctive aspects of China’s economic statecraft [is] its pursuit of multiple goals through a single initiative’.³⁵ Accordingly, Chinese lending is guided by the economic objectives that motivated its ‘going out’ strategy, i.e. diversifying reserve holdings, exporting industrial and construction overcapacity, and seeking access to natural resources, but also by the strategical imperative of competing with the United States for status and influence, in particular in the developing world.

The Importance of the Leader

A leader’s preferences over the foreign policy are especially important in authoritarian regimes.³⁶ Xi Jinping has followed some policies launched by his predecessors such the ‘going out’ strategy and overseas lending programme initiated by Jiang Zemin and deepened by Hu Jintao. In several aspects, Jiang and Hu set the stage for Xi’s articulation of a more assertive Chinese foreign policy.³⁷ Nonetheless, Xi has reshaped Chinese foreign policy to return China to what he views as its historically central role—a claim this paper will test empirically.³⁸ Acting on a strategic opportunity to return China to great power status first identified in 2002, Xi ‘has charted a new roadmap for China’s foreign policy that is primarily concerned about its own national interests and status goal’.³⁹

Xi’s accession to power in late 2012 is crucial to determine when China intensified the pursuit of the wedging geoeconomic strategy. While much of the logic has been inherited from his predecessors, under Xi, the geopolitical element in China’s overseas lending programme (rewarding countries to maintain a distance from the United States) has become discernible. Xi abandoned China’s decades-old low-profile strategy and set out to ‘strive for achievement’ of a new global order organised around Sinocentric initiatives.⁴⁰ As a Marxist-Leninist and thus a historic materialist, Xi views America’s—and the West’s—eventual decline and China’s ascent as an inevitable result of dialectic processes, and believes in unleashing the mobilisational power of the Chinese state led by the CCP as a vanguard party to accelerate that outcome.⁴¹

Since coming to power, Xi has forced a shift in China’s foreign relations rhetotic and advertised his intention to forge a ‘new type of international relations’,⁴² and articulated political slogans such as the ‘Chinese dream’, ‘the great rejuvenation of the Chinese nation’ and a ‘community of shared destiny for all humankind’, which all point to an emerging Sinocentric international order. These slogans’ essence is none other than making China rich and strong, and respected as a rule maker in the world stage.⁴³ Chinese diplomatic and academic discourse reflects an assertive and self-confident nation, now ‘striving for achievement’ and ready to lead that ‘community of shared destiny’. The rhetorical change has been complemented by a flurry of new international initiatives and organisations led by China, launched under Xi albeit in some cases following a logic set out by his predecessors. The Belt and Road Initiative (BRI), the Asian Infrastructure Investment Bank, the New Development Bank, the Global Development Initiative, the Global Security Initiative and the Global Civilisation Initiative are some examples.

The BRI, in particular, has much in common with China’s overseas lending programme—not least that the latter finances much of the former. The BRI has been called ‘a top-level design on which President Xi has staked his personal legacy’,⁴⁴ backed by ideas that ‘can be understood as the latest development of a “Sino-centric” world order conceptual apparatus’.⁴⁵ Just as one of the justifications behind China’s lending is to avoid excess reserves entering the domestic economy, the BRI also has domestic economic justifications, mainly exporting excess capacity in China’s construction and industrial sectors.⁴⁶ Ultimately, the economic and political success of these initiatives reinforces the CCP legitimacy domestically and promotes China’s influence internationally, counterbalancing what China sees as United States’ efforts to contain its rise.⁴⁷

Interpreting Chinese Lending

This article posits that Chinese lending should be considered on a continuum of degree of concessionality, and not according to the unhelpful dichotomy between ODA and OOF. In the case of China, a strict separation between aid and commercial loans runs into obstacles, given the lack of comparability between Western notions of ODA and OOF and the way China lends, and the lack of transparency involving Chinese overseas financing. First, China’s overseas financing is hardly comparable to most other governments‘.⁴⁸ Research shows that ‘China’s official financing is less concessional than World Bank financing in comparable settings; however, nearly all Chinese loans have some degree of concessionality’.⁴⁹ Chinese concessional loans are less concessional than those granted by the World Bank, for example, and Chinese commercial loans are less commercial than those offered by Western lenders.⁵⁰ This ambiguity between concessional and non-concessional transactions is ‘unavoidable, even intentional’ as ‘the Chinese government encourages its agencies and commercial entities to closely mix and combine foreign aid, direct investment, service contracts, labor cooperation, foreign trade and export’.⁵¹ Second, China does not publish much data on its lending—about 50% is estimated to be unreported,⁵² leading to considerable guessing on the exact recipients, amounts, and terms of Chinese foreign loans.⁵³

Thus, China offers little in terms of highly concessional aid, but a lot in terms of loans under more favourable terms—in particular lower interest rates—than what prevails in the commercial market. By trying to fit China into the parameters of the OECD-DAC rules, the literature overlooks a large portion of Chinese loans. Furthermore, the concessional element is reflected not only in lower interest rates, but also in the fact that many of the countries that borrow from China would not qualify to any credit by Western or multilateral lenders due to their dire macroeconomic situation and geopolitical isolation.

Finally, this article argues that Chinese loans must be considered in terms of their relevance to the borrower. China needs to spend less to buy influence over Burundi than to influence Brazil. In absolute amounts, China still makes larger loans to bigger countries and exporters of natural resources, mainly oil (see Figure 1). Since Chinese loans are expected to influence a target’s behaviour only if the loans are economically important to that target, one must consider how much of its external borrowing needs the target is getting from China—in other words, what the ‘price tag’ of the borrower is. Lending relatively more to those ‘weakest links’, which in many international organisations have the same voting rights as richer countries, brings China the greatest gains at the lowest cost. By targeting those needier countries, China has found a way of achieving its strategic objective of establishing itself as a near peer to the United States that costs less than the levels of ODA practiced by developed economies and is more effective than non-economic alternatives such as pure diplomacy or military action.

Therefore, it is expected that since Xi Jinping took power China has lent more and on more favourable terms to developing countries that have less affinity with the United States’ preferred foreign policy positions. The hypotheses can be formulated as follows:

H1:

Since Xi Jinping came to power, China lends more to countries that are more distant from the United States in terms of foreign policy preferences.

It is expected that China will lend relatively larger amounts to countries that display the least geopolitical affinity with the United States, its main rival. More specifically, it is expected that the proportion of those countries’ overall debt owed to China will be larger than that of countries that are more closely aligned with the United States.

H2:

Since Xi Jinping came to power, China lends more favourably to countries that are more distant from the United States in terms of foreign policy preferences.

The terms of Chinese loans should be more favourable to countries that display the least affinity with the United States in terms of foreign policy preferences. More specifically, it is expected that the interest rate charged under those loans will be lower than that China charges countries that are more closely aligned with the United States.

Empirical Strategy

To test hypothesis 1 – China lends more to developing countries less aligned with the United States—the dependent variable is the proportion of a country’s indebtedness that is owed to China. To determine a country’s ‘price tag’, the annual US dollar amount of loans committed by China to each borrowing country is divided by the total US dollar amount of loans owed to external creditors that year (the variable is labelled Loans/Total). If China wants to influence a target, it must be important to the borrower when compared to the alternatives, i.e. all creditors combined. The dependent variable to test hypothesis 2 is the annual average interest rate charged by Chinese lenders to a country on all loans committed in a year (Interest).

The AidData database tracks underreported financial flows covering aid and loans granted by the Chinese government and its affiliated entities to 148 developing countries from 2000 to 2017. It contains information about 3,103 loans, identifying transaction amounts in 2,757 cases and interest rates in 1,659 cases. These data have some limitations as China does not publish granular data on its lending. AidData’s tracking methodology achieves more than other similar efforts, but it is still not a complete picture. Given these gaps, one should approach the results of any study on China’s overseas financing with caution. On the other hand, one would expect that countries that have lower levels of transparency would be more prone to not reporting or underreporting loan data. If that is the case, as countries with lower transparency tend to be less aligned with the United States,⁵⁴ higher levels of reporting would most likely reinforce the results of this paper, in particular increase the coefficients for Loans/Total in the results related to hypothesis 1.

The main independent variable, foreign policy divergence from the United States, is measured as the distance between a country’s voting affinity at the United Nations General Assembly and the United States in a given year (UNGA). This ‘ideal point distance’ comes from Voeten and others.⁵⁵ Votes in the UNGA form the basis for a large body of studies on states’ foreign policy preferences, as they provide ‘comparable and observable actions taken by many countries at set points in time’.⁵⁶ Every country has a record for all the years it has occupied a seat in the UNGA, thus making the ideal point distance between two countries especially helpful in assessing how two countries converge or diverge in the way they manifest their foreign policy preferences. A lower ideal point distance denotes more proximity in those preferences.

As controls, the models include factors that the literature has identified as influencing Chinese overseas financing. In terms of a borrowing country’s political and institutional characteristics, the models control for regime type (Polity), level of corruption (Corruption), regulatory quality (Regulatory), whether it recognises Taiwan (Taiwan), whether it has signed a memorandum of understanding to join the BRI (BRI), and whether it occupies a seat in the UNSC (UNSC).⁵⁷ Regarding the borrower country’s economic characteristics, the models control for GDP per capita (GDP), overall foreign indebtedness (Debt), fiscal balance (Fiscal), trade openness (Open), trade with China (China trade), crude oil exports (Oil) and natural resources rents (Rents).⁵⁸ Finally, models include two global macroeconomic variables shown to affect the level of Chinese lending to Latin American countries: global GDP growth rates (Growth) and interest rates in the United States (US interest).⁵⁹ Table A1 in the Appendix shows a description of each variable.

Regressions are estimated using Heckman selection models, which reduce selection bias by first estimating a selection function followed by a linear regression equation.⁶⁰ Heckman models are, however, sensitive to the choice of variables in the selection function.⁶¹ To determine the most appropriate model specification for the selection equations, a random-effects logit model⁶² is first run to assess the likelihood of a country i receiving a loan from China in year t, where UNGAxXJP is an interaction term between UNGA in year t-1 and the dummy XJP; and the control variables for country i comprise Polity, Corruption, Regulatory, BRI, Taiwan, UNSC, GDP, Debt, Fiscal, Open, China trade, Oil, and Rents, as well as the global macroeconomic variables Growth and US interest. The variables that yield significant results in the logit model – UNGA, Polity, Corruption, Taiwan, China trade and Oil – are included in the selection equation. Table A2 in the Appendix reports the results for the logit analysis.

Findings

Table 1 reports the results for hypothesis 1. In column (1), which shows the second-stage estimates for H1 (dependent variable Loans/Total), the coefficient associated with UNGA is negative but does not achieve statistical significance. The coefficient associated with XJP is negative and statistically significant. However, in light of H1, it is not expected that under Xi Jinping China will generally lend more to all countries, but more to those countries less aligned with the United States. The coefficient associated⁶³ with XJP needs to be interpreted as the loans to be expected when a country is fully aligned with the United States, other variables held constant. H1 however expects a positive association between UNGA and Loans/Total after Xi came to power. Consistent with this expectation, both the interaction term UNGAxXJP and its joint effect with the interaction term UNGAxXJP are positive and statistically significant.

Table 1. Chinese loans as a share of a country’s overall foreign debt and mean interest rate on Chinese loans (2000–2017)

	(1)	(2)	(3)	(4)
Dependent variable	Loans/Total	Loans/Total	Interest	Interest
		(Selection)		(Selection)
UNGA	−0.0023	0.1650**	−0.0081**	0.1903***
	(0.0052)	(0.0660)	(0.0032)	(0.0689)
XJP	−0.0655**		0.0144	−0.0040
	(0.0285)		(0.0176)	(0.0902)
UNGA x XJP	0.0257***		−0.0034
	(0.0089)		(0.0054)
Regulatory	−0.0137**		−0.0136***
	(0.0053)		(0.0025)
BRI	0.0063		−0.0005
	(0.0095)		(0.0047)
UNSC	−0.0036		0.0054
	(0.0085)		(0.0043)
GDP	−0.0036		0.0099***
	(0.0025)		(0.0013)
Debt	−0.0003***		−0.0000
	(0.0001)		(0.0000)
Fiscal	0.0001		−0.0006*
	(0.0006)		(0.0003)
Open	0.0002**		−0.0001**
	(0.0001)		(0.0000)
Rents	0.0006**		−0.0004***
	(0.0003)		(0.0001)
Growth	−0.0028*		−0.0005
	(0.0017)		(0.0008)
US interest	−0.0038*		0.0031***
	(0.0021)		(0.0012)
Polity		0.0343***		0.0290***
		(0.0083)		(0.0085)
Corruption		−0.9153***		−0.4820***
		(0.0720)		(0.0732)
Taiwan		−2.3819***		−1.7049***
		(0.1634)		(0.2425)
China trade		0.0623***		0.1354***
		(0.0211)		(0.0223)
Oil		−0.0258***		−0.0129***
		(0.0050)		(0.0049)
ΒUNGA + ΒUNGAxXPJ	0.0235*** (.0077)		−0.0115*** (0.0048)
Observations	1,322^63	1,322	1,327	1,327
Chi2	85.72	85.72	118.9	118.9

Heckman selection models. Selection equation results in columns (2) and (4).

Standard errors in parentheses. Statistical significance indicated as ***p < 0.01, **p < 0.05, *p < 0.1.

Substantively, this means that one standard deviation increase in distance from the United States in terms of UNGA voting affinity is associated with an increase of 2.35 percentage points in the share of loans from China (effect of Β_UNGA + Β_UNGAxXPJ)—in relation with loans from other external creditors, holding other factors constant—under Xi (Figure 2). This magnitude is important given that in the period under analysis, the sample mean is 4.7% of a country’s total external debt owed to China (about $669 million constant 2017 US dollars), which would increase by 50% if the average country in the sample moves one standard deviation away from the United States. These results provide support for H1.

The coefficients associated Regulatory and Debt are statistically significant and negative, while Open and Rents are significant and positive. Other controls do not achieve statistical significance—including BRI, which may come as a surprise. This may be due to the fact that many of the countries that have signed up to the BRI did so from 2018. While there is evidence that Chinese lending continued strong during 2018 and 2019, coming to an abrupt slowdown during the COVID-19 pandemic and failing to pick up afterwards, those two years are not covered in the sample.

Although not central to the theory, the selection equation (column 2) shows interesting results: Polity, Corruption, Taiwan and Oil are significant. Consistently with the literature, China lends less, or simply does not lend, to countries that recognise Taiwan. Countries that depend more on natural resources borrow more from China. This is probably because many Chinese loans are backed by exports of commodities as collateral.⁶⁴ Intriguingly, the coefficient associated with Polity is positive and significant, suggesting that China lends relatively more to more democratic countries. Similarly, China seems to lend less, albeit at lower rates, to countries with worse corruption controls and worse regulatory quality. This differs from the literature,⁶⁵ but is probably explained by the fact that, once other variables are controlled for (especially UNGA voting behaviour), China lends more to relatively more developed countries.

Columns (3) and (4), testing hypothesis 2, show a slightly different picture. The coefficient associated with UNGA is negative and statistically significantly associated with Interest, indicating that this negative relationship existed prior to Xi coming to power. The interaction term UNGAxXJP is negative as expected, but fails to meet statistical significance. However, the joint effect of UNGA and the interaction term UNGAxXJP generates a negative and statistically significant result, which provides support for H2. In general, countries that are farther away from the United States in terms of UNGA voting behaviour have paid lower interest rates to China since 2000. A country in the analysed sample paid a mean interest rate to China of 2.82% and on average would pay 1.11% points less (effect of Β_UNGA + Β_UNGAxXPJ)—a reduction of almost 40% – for one standard deviation away from the United States’ ideal voting point.

Beyond the statistical results, the connection between Chinese loans and foreign policy alignment is also reflected in politicians’ and officials’ rhetoric, on the Chinese and borrowers’ side. The geopolitical competition between China and the United States is often referred to by officials from China and borrowing countries when signing new loan agreements or cancelling existing debt. In 2006, as China made its first large cash-for-oil loan to Venezuela, the late Hugo Chávez declared that his country and China were forming ‘a strategic alliance with the strength of the Great Wall’ against American hegemonism.⁶⁶ Visiting Beijing to ask for a fresh loan of $5bn in 2018, Venezuelan president Nicolás Maduro saluted Mao Zedong’s role in ushering in a ‘multipolar twenty-first century’ and thanked China for its role in keeping Venezuela ‘on its feet’ while it faced a ‘financial blockade’ by the United States.⁶⁷ Visiting Addis Ababa in 2023 to announce the partial cancellation of Ethiopia’s debts to China,⁶⁸ Chinese foreign minister Qin Gang made a point of taking the usual verbal jibes at the United States and the American-led international order, vowing to ‘oppose hegemonic, high-handed, bullying, and racially discriminatory acts, jointly uphold genuine multilateralism, promote democracy in international relations, (…) and work together to make the global governance system more just and equitable’.⁶⁹ Similarly, when announcing the cancellation of 23 loans to 17 African countries in 2022, Qin’s predecessor Wang Yi made a point of condemning ‘the West’s “zero-sum Cold War mentality”’ and ‘major-country rivalry for geopolitical gains’⁷⁰ – thinly disguised jabs at the United States.

Robustness Checks

Heckman models are sensitive to the choice of variables in the selection equation.⁷¹ As robustness checks, the models are re-estimated following, for the selection equations, variables found to be significant in other studies. Following Dreher and others,⁷² selection models include Corruption, Taiwan, GDP, China trade and Oil. Following Kaplan,⁷³ another set of models includes Regulatory, GDP, Debt, Growth and US interest in the selection equations. Results, reported in Table A4 in the Appendix, are substantively similar to the main models.

In the sample, there are seven observations where Loans/Total exceeded 0.8 – that is, years in which a country owed more to China than 80% of its overall foreign indebtedness. Table A5 in the Appendix shows that the results are robust to the exclusion of these outliers.

To ensure these models are capturing the strategic divergence from the United States and not affinity with China, I re-estimate the models replacing UNGA with UNGA China, the voting ideal point distance from China. In these models, the coefficient associated with UNGA China in the selection equation is negative and statistically significant, indicating countries that are farther away from China get less money (see Table A6 in the Appendix), but in the main model, the direction of UNGA China is positive. This seems counterintuitive, but can be explained by some countries which, while still distant from the United States, also also displayed a higher-than-average distance from China in the period between 2013 and 2017. In particular, Algeria, Bolivia, Burundi, Cuba, Iran, Kosovo, Nicaragua, Syria, Venezuela and Zimbabwe received from China loans higher than the mean (measured as a percentage of their total foreign debt) in one or several years since Xi Jinping came to power, despite also not being particularly aligned with China (UNGA China > 1). One of the aspects most of those countries share is their anti-American (UNGA > 3.5) or anti-Western regimes. This reinforces the notion that, rather than buying allegiance, China is wedging by consolidating existing positions and preserving the strategic wedge between those regimes and the United States.

Finally, models in Table A7 address concerns regarding reverse causality: are Chinese loans causing recipient countries to distance themselves from the United States, or is China rewarding countries that display less affinity with its geopolitical rival? The H1 Heckman model is reversed and reestimated as a linear regression using UNGA as the dependent variable and Loans/Total as the main independent variable, lagged one, two and three years. In these models, the coefficients for Loans/Total do not achieve statistical significance in any of the lags, either as a standalone variable or in interaction with XJP.

Alternative Explanations

Possible alternative explanations to Chinese unusual lending patterns are that poorer countries prefer to borrow from China given the latter’s lack of policy conditionality, and that China lends more to countries from which it imports or expects to import hydrocarbons. Those plausible explanations are accounted for by controlling for GDP and Oil in the Heckman models testing hypotheses 1. The borrower’s GDP per capita is statistically insignificant for Loans/Total (Table 1) as it is in the logit mode (Table A2). Hydrocarbon exports is statistically significant but negative. On the other hand, Rents is significant for both loan amounts and interest rate (albeit with small coefficients), indicating that China lends more and at lower interest rates to countries that are more dependent on natural resources. The results are sensitive to the model specification: in the robustness checks, GDP and Oil gain significance, while Rents loses a bit but remains (Table A4).

Countries which are in dire financial straits but are perceived to be politically aligned with the United States offer additional support for this idea—note that these are rare cases, since most countries highly aligned with the United States’ UNGA voting are rich liberal democracies. Iraq, occupied by the United States between 2003 and 2011 and seen as politically aligned with its former occupier, borrowed only US$ 54 million from China between 2013 and 2017, despite being an oil exporter and among the countries with the worst fiscal balance in the world in that period.⁷⁴ In contrast, Iraq’s neighbour Iran borrowed US$ 10 billion from China in the same period. Similarly,Timor-Leste, with serious fiscal problems and starving for cash, is comparatively well aligned with the United States in terms of UNGA voting behaviour. Timor-Leste received only US$ 10 m in Chinese loans between 2003 and 2017.

An important caveat is that this article does not argue that that China only lends to countries that are distant from the United States in terms of foreign policy preferences, or that only geostrategic considerations explain lending. This article’s argument is that, in addition to economic justifications, China seeks to obtain a relative gain in international status by creating an incentive for other countries to maintain low levels of alignment with the United States. It is this combination of market and might seeking that is the essence of China’s economic statecraft.

Conclusion

This study has shown that China targets countries that are farther away from the United States in terms of foreign policy preferences and, by providing them more and cheaper credit, allows those countries to maintain foreign policies that diverge from American preferences. Since Xi Jinping’s accession in late 2012, China has lent more money to regimes that are less aligned with the United States in terms of voting behaviour in the UNGA. China also lends to those countries at lower interest rates.

China seeks a more prominent role on the international stage. Status is relational, and China’s quest for international prominence depends on it gaining status in relation to its rivals, the most important of which is the United States. Thus, China uses economic inducements—loans to developing countries—to achieve greater status internationally by rewarding regimes that display less affinity with the United States so they can maintain that separation.

China seems to be killing two birds with one stone. Since Xi came to power, China has used the least costly means at its disposal—vast foreign exchange reserves accumulated as a result of its successful export-oriented policies—in order to assert itself internationally and consolidate a wedge between developing countries and the United States. By doing this, China is seeking both profit and power. It is in the economic realm that China has the best competitive advantage in relation to the United States, making China’s geoeconomic strategy the path available to it at the lowest relative cost. By reducing the United States’ relative standing in the developing world, China is boosting its own position—a potential change in the pattern of hegemony also welcome by at least some of the borrowing countries.

This article has highlighted Xi’s accession as paramount leader of the PRC has been a key driver to this policy, even if Xi’s predecessors set the stage for some of his geoeconomic policies. While there is a large body of literature on leader personality and behaviour influences in politics, there is surprisingly little that focuses on the influence of leaders of authoritarian regimes over foreign policy—recent work by Zhao⁷⁵ being one notable exception. This is an area with potential for future research.

It is beyond the scope of this article to determine whether China has been successful in consolidating this geopolitical wedge between recipients of its overseas financing programme and the United States. While developing countries seem to have diverged slightly from the United States in their UNGA voting behaviour since Xi’s accession, the models used in this paper do not support a direct attribution of that trend to the size of Chinese loans relative to the borrowers’ overall foreign indebtedness. However, recent scholarship continues to investigate the geopolitical effect of Chinese foreign lending,⁷⁶ and more can be done in that area.

Finally, Chinese overseas lending seems to have contracted since 2019, and the BRI has been scaled back, but there are no indications that Chinese global aspirations have diminished. If that is the case, how does China intend to fund its campaign to make friends and influence countries now that its appetite to finance the construction of infrastructure in the developing world has waned? Future research may want to look into what other geoeconomic tools the PRC is using to advance its ambitions.

Acknowledgments

The author wishes to thank Carolina Garriga, Han Dorussen, Ximena Velasco Guachalla, Allyson Benton, Sarah Shair-Rosenfield, Diana Zhu, Zenobia Chan, Nicolás Fernández, and the participants in the 10^th European Workshops in International Studies for comments received.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

¹ Samantha Custer and others, Tracking Chinese Development Finance: An Application of AidData’s TUFF 2.0 Methodology (AidData at William & Mary, 2021).

² Timothy W. Crawford, ‘Preventing Enemy Coalitions: How Wedge Strategies Shape Power Politics’, International Security 35, (2011), pp. 155; Yasuhiro Izumikawa, ‘To Coerce or Reward? Theorizing Wedge Strategies in Alliance Politics’, Security Studies 22, (2013), pp. 498; Stacie E. Goddard and Daniel H. Nexon, ‘The Dynamics of Global Power Politics: A Framework for Analysis’, Journal of Global Security Studies 1, (2016), p. 4; Kristen Hopewell, ‘Balancing, Threats, and Wedges in International Political Economy: The Origins and Impact of the Sino-Indian Alliance at the WTO’, Journal of Contemporary China 32, (2023), p. 369.

³ Matt Ferchen, ‘China, Venezuela, and the Illusion of Debt-Trap Diplomacy’. https://carnegietsinghua.org/2018/08/16/china-venezuela-and-illusion-of-debt-trap-diplomacy-pub−77,089.

⁴ Brahma Chellaney, ‘China’s Debt-Trap Diplomacy’, Project Syndicate. Accessed December 21, 2022. https://www.project-syndicate.org/commentary/china-one-belt-one-road-loans-debt-by-brahma-chellaney-2017–01.

⁵ Deborah Brautigam, ‘A critical look at Chinese “debt-trap diplomacy”: the rise of a meme’, Area Development and Policy 5, (2020), p. 1.

⁶ Axel Dreher and others, Banking on Beijing: The Aims and Impacts of China’s Overseas Development Program (Cambridge University Press, 2022); Stephen B. Kaplan, Globalizing Patient Capital: The Political Economy of Chinese Finance in the Americas (Cambridge University Press, 2021); Lukas Wellner and others, Can Aid Buy Foreign Public Support? Evidence from Chinese Development Finance (CESifo Working Papers, 2022); Zhongshu Li and others, ‘Pushing out or pulling in? The determinants of Chinese energy finance in developing countries’, Energy Research & Social Science 86, (2022), p. 102, p. 441.

⁷ Kaplan, Globalizing Patient Capital: The Political Economy of Chinese Finance in the Americas; Axel Dreher and others, ‘Apples and Dragon Fruits: The Determinants of Aid and Other Forms of State Financing from China to Africa’, International Studies Quarterly 62, (2018), p. 182. This finding has recently been challenged by Raess and others (2023).

⁸ IMF, World Economic Outlook Database (2022).

⁹ David A. Baldwin, Economic Statecraft: New Edition (Princeton University Press, 2020).

¹⁰ Barry Naughton and Briana Boland, CCP Inc.: The Reshaping of China’s State Capitalist System, 2023).

¹¹ Scott B. MacDonald, The New Cold War, China, and the Caribbean: Economic Statecraft, China and Strategic Realignments (Palgrave Macmillan Cham, 2023).

¹² James Reilly, Orchestration: China’s Economic Statecraft Across Asia and EuropeChina’s Economic Statecraft Across Asia and Europe (2021).

¹³ William J. Norris, ‘China’s Post-Cold War Economic Statecraft: A Periodization’, Journal of Current Chinese Affairs 50, (2021), p. 294.

¹⁴ Naughton and Boland, CCP Inc.: The Reshaping of China’s State Capitalist System 10.

¹⁵ Dreher and others, Banking on Beijing: The Aims and Impacts of China’s Overseas Development Program.

¹⁶ Dreher and others, ‘Apples and Dragon Fruits: The Determinants of Aid and Other Forms of State Financing from China to Africa’.

¹⁷ Gustavo A. Flores-Macías and Sarah E. Kreps, ‘The Foreign Policy Consequences of Trade: China’s Commercial Relations with Africa and Latin America, 1992–2006’, The Journal of Politics 75, (2013), pp. 357-362; Georg Strüver, ‘What Friends are Made of: Bilateral Linkages and Domestic Drivers of Foreign Policy Alignment with China’, Foreign Policy Analysis 12, (2016), pp. 170-20.

¹⁸ Between 2000 and 2014, Chinese loans were, on average, more than twice as large as World Bank loans ($307 million versus $148 million), while the average grant element of a Chinese loan is only a fraction of the World Bank’s ($9 million versus $44 million). Overall, the World Bank’s loan portfolio is approximately 30% more concessional than China’s. (Scott Morris, Brad Parks and Alysha Gardner, Chinese and World Bank Lending Terms: A Systematic Comparison Across 157 Countries and 15 Years, 2020). At the same time, developing countries owed more to China in 2017 ($370 billion in their estimate) than to the governments of the 22 members of the Paris Club of developed sovereign creditors combined ($246 billion) (Sebastian Horn, Carmen M. Reinhart and Christoph Trebesch, China’s Overseas Lending (NBER Working Paper Series, 2019).

¹⁹ Kaplan, Globalizing Patient Capital: The Political Economy of Chinese Finance in the Americas.

²⁰ Similarly, Li and others find that a substantial portion of Chinese policy bank loans to developing countries’ electric power sectors is tied to the purchase of Chinese goods and services. Li and others, ‘Pushing out or pulling in? The determinants of Chinese energy finance in developing countries’.

²¹ Dreher and others, Banking on Beijing: The Aims and Impacts of China’s Overseas Development Program; Ann-Sofie Isaksson and Andreas Kotsadam, ‘Chinese aid and local corruption’, Journal of Public Economics 159, (2018), p. 146; Andreas Kern, Bernhard Reinsberg and Patrick E. Shea, IMF Programs, Chinese Lending, and the Political Economy of Leader Survival. Working Paper #118, 2022).

²² Dreher and others, ‘Apples and Dragon Fruits: The Determinants of Aid and Other Forms of State Financing from China to Africa’; Kaplan, Globalizing Patient Capital: The Political Economy of Chinese Finance in the Americas.

²³ Kaplan, Globalizing Patient Capital: The Political Economy of Chinese Finance in the Americas.

²⁴ Jessica Chen Weiss, ‘A World Safe for Autocracy?’, Foreign Affairs 98, (2019), p. 92; Elizabeth Economy, The World According to China (Polity Press, 2022); Gang Ding, ‘Why say the Russia-Ukraine conflict changed European views of China? (为什么说俄乌冲突改变了欧洲人的中国观?)’ (2023 Macro Situation Annual Forum and the 3rd Forum on the International Influence of China’s Think Tanks).

²⁵ Crawford, ‘Preventing Enemy Coalitions: How Wedge Strategies Shape Power Politics’.

²⁶ Khang X. Vu, ‘External Coercion, Internal Accommodation: China’s Wedge Strategies Towards the Vietnam-United States Partnership, 2013–2022’, Journal of Contemporary China (2023), p. 1.

²⁷ Izumikawa, ‘To Coerce or Reward? Theorizing Wedge Strategies in Alliance Politics’; Crawford, ‘Preventing Enemy Coalitions: How Wedge Strategies Shape Power Politics’; Goddard and Nexon, ‘The Dynamics of Global Power Politics: A Framework for Analysis’.

²⁸ Ruonan Liu, ‘Reinforcing Wedging: Assessing China’s Southeast Asia Policy in the Context of Indo-Pacific Strategy’, China Review 23, (2023), p. 277; Andrew D. Taffer, ‘Threat and opportunity: Chinese wedging in the Senkaku/Diaoyu dispute’, Asian Security 16, (2020), p. 157; Vu, ‘External Coercion, Internal Accommodation: China’s Wedge Strategies Towards the Vietnam-United States Partnership, 2013–2022’.

²⁹ Ian Johnson, ‘Has China Lost Europe? How Beijing’s Economic Missteps and Support for Russia Soured European Leaders’ Foreign Affairs. Accessed June 10, 2022. https://www.foreignaffairs.com/articles/china/2022-06-10/has-china-lost-europe.

³⁰ Reilly, Orchestration: China’s Economic Statecraft Across Asia and EuropeChina’s Economic Statecraft Across Asia and Europe.

³¹ Barry Naughton, ‘China’s Distinctive System: can it be a model for others?’, Journal of Contemporary China 19, (2010), p. 437.

³² Courtney J. Fung and Shing-hon Lam, Mixed report card: China’s influence at the United Nations, 2022.

³³ Courtney J. Fung and Shing-hon Lam, ‘Staffing the United Nations: China’s motivations and prospects’, International Affairs 97, (2021), p. 1143.

³⁴ Yi Che, Xiaoyu He and Yan Zhang, ‘Natural resource exports and African countries’ voting behaviour in the United Nations: Evidence from the economic rise of China’, Canadian Journal of Economics/Revue canadienne d’économique 54, (2021), p. 712.

³⁵ Reilly, Orchestration: China’s Economic Statecraft Across Asia and EuropeChina’s Economic Statecraft Across Asia and Europe.

³⁶ ‘[W]hile leaders matter in all political systems, they mater more in totalitarian and authoritarian systems that allow for the propensity of leaders’ ambitions’ (Suisheng Zhao, The Dragon Roars Back: Transformational Leaders and Dynamics of Chinese Foreign Policy (Stanford University Press, 2022) 5). See also Maryann E. Gallagher and Susan H. Allen, ‘Presidential Personality: Not Just a Nuisance1’, Foreign Policy Analysis 10, (2014), p. 1.

³⁷ Dan Blumenthal, ‘China’s Steps Backward Began Under Hu Jintao: Beijing’s new aggression and ideological reaction started well before Xi Jinping’. Foreign Policy https://foreignpolicy.com/2020/06/04/china-xi-jingping-hu-jintao-aggression-ideology/; Yong Deng, China’s Strategic Opportunity: Change and Revisionism in Chinese Foreign Policy (Cambridge University Press, 2022).

³⁸ Zhao, The Dragon Roars Back: Transformational Leaders and Dynamics of Chinese Foreign Policy.

³⁹ Deng, China’s Strategic Opportunity: Change and Revisionism in Chinese Foreign Policy , pp. 15–16.

⁴⁰ Xuetong Yan, ‘From Keeping a Low Profile to Striving for Achievement’, The Chinese Journal of International Politics 7, (2014), p. 153.

⁴¹ Kevin Rudd, ‘The World According to Xi Jinping: What China’s Ideologue in Chief Really Believes’, Foreign Affairs 101, (2022), p. 8.

⁴² Graham Allison, The Thucydides Trap: Are the U.S. and China Headed for War? (2015).

⁴³ Feng Zhang, ‘The Xi Jinping Doctrine of China’s International Relations’, Asia Policy 14, (2019), p. 7.

⁴⁴ Suisheng Zhao, ‘China’s Belt-Road Initiative as the Signature of President Xi Jinping Diplomacy: Easier Said than Done’, Journal of Contemporary China 29, (2020), p. 319.

⁴⁵ Ray Silvius, ‘China’s Belt and Road Initiative as Nascent World Order Structure and Concept? Between Sino-Centering and Sino-Deflecting’, Journal of Contemporary China 30, (2021), p. 314.

⁴⁶ Hongyi Lai, ‘The Rationale and Effects of China’s Belt and Road Initiative: Reducing Vulnerabilities in Domestic Political Economy’, Journal of Contemporary China 30, (2021), p. 330.

⁴⁷ Weifeng Zhou and Mario Esteban, ‘Beyond Balancing: China’s approach towards the Belt and Road Initiative’, Journal of Contemporary China 27, (2018), p. 487.

⁴⁸ Sebastian Horn, Carmen M. Reinhart and Christoph Trebesch, ‘China’s overseas lending’, Journal of International Economics 133, (2021), p. 103, p. 539. See also Damian Raess, Wanlin Ren and Patrick Wagner, ‘Hidden Strings Attached? Chinese (Commercially Oriented) Foreign Aid and International Political Alignment’, Foreign Policy Analysis 18, (2022), p. 1.

⁴⁹ Morris, Parks and Gardner, Chinese and World Bank Lending Terms: A Systematic Comparison Across 157 Countries and 15 Years.

⁵⁰ Anna Gelpern and others, How China Lends: A Rare Look into 100 Debt Contracts with Foreign Governments, 2021; Morris, Parks and Gardner, Chinese and World Bank Lending Terms: A Systematic Comparison Across 157 Countries and 15 Years.

⁵¹ Robert A. Blair, Robert Marty and Philip Roessler, ‘Foreign Aid and Soft Power: Great Power Competition in Africa in the Early Twenty-first Century’, British Journal of Political Science 1, (2021), p. 12.

⁵² Sebastian Horn, Carmen M. Reinhart and Christoph Trebesch, ‘How Much Money Does the World Owe China?’ Harvard Business Review. Accessed February 26, 2020. https://hbr.org/2020/02/how-much-money-does-the-world-owe-china.

⁵³ Morris, Parks and Gardner, Chinese and World Bank Lending Terms: A Systematic Comparison Across 157 Countries and 15 Years; Gelpern and others, How China Lends: A Rare Look into 100 Debt Contracts with Foreign Governments.

⁵⁴ The correlation between UNGA and the HRV transparency index is −0.65, which indicates that countries less aligned with the United States in terms of UNGA voting are less transparent. James R. Hollyer, B. Peter Rosendorff and James Raymond Vreeland, ‘Measuring Transparency’, Political Analysis 22, (2017), p. 413.

⁵⁵ Erik Voeten, Anton Strezhnev and Michael Bailey, United Nations General Assembly Voting Data (Harvard Dataverse, 2021).

⁵⁶ Michael A. Bailey, Anton Strezhnev and Erik Voeten, ‘Estimating Dynamic State Preferences from United Nations Voting Data’, Journal of Conflict Resolution 61, (2017), pp. 430-431.

⁵⁷ Dreher and others, ‘Apples and Dragon Fruits: The Determinants of Aid and Other Forms of State Financing from China to Africa’; Dreher and others, Banking on Beijing: The Aims and Impacts of China’s Overseas Development Program; Isaksson and Kotsadam, ‘Chinese aid and local corruption’; Kaplan, Globalizing Patient Capital: The Political Economy of Chinese Finance in the Americas.

⁵⁸ Axel Dreher, Peter Nunnenkamp and Rainer Thiele, ‘Does US aid buy UN general assembly votes? A disaggregated analysis’, Public Choice 136, (2008), p. 139; Dreher and others, ‘Apples and Dragon Fruits: The Determinants of Aid and Other Forms of State Financing from China to Africa’; Kaplan, Globalizing Patient Capital: The Political Economy of Chinese Finance in the Americas; Li and others, ‘Pushing out or pulling in? The determinants of Chinese energy finance in developing countries’; Flores-Macías and Kreps, ‘The Foreign Policy Consequences of Trade: China’s Commercial Relations with Africa and Latin America, 1992–2006’; Scott L. Kastner, ‘Buying Influence? Assessing the Political Effects of China’s International Trade’, The Journal of Conflict Resolution 60, (2016), p. 980.

⁵⁹ Kaplan, Globalizing Patient Capital: The Political Economy of Chinese Finance in the Americas.

⁶⁰ James J. Heckman, ‘Sample Selection Bias as a Specification Error’, Econometrica 47, (1979), p. 153. Heckman models have also been shown to correct for missingness even when a substantial proportion of data is missing, an important challenge when investigating Chinese financing, given the lack of transparency of China’s overseas lending and strict confidentiality clauses in its loan agreements. It is possible that data are not missing randomly. The missingness may be related to identified variables reflecting a country’s institutional capability such as Corruption or Regulatory, or another unspecified variable, as countries with stronger capabilities are more likely to report their external borrowings. See Siaka Koné and others, ‘Heckman-type selection models to obtain unbiased estimates with missing measures outcome: theoretical considerations and an application to missing birth weight data’, BMC Medical Research Methodology 19, (2019).

⁶¹ Derek C. Briggs, ‘Causal Inference and the Heckman Model’, Journal of Educational and Behavioral Statistics 29, (2004), p. 397.

⁶² Random effects are recommended after Hausman test is run on the model.

⁶³ Table A3 presents a list of countries and years observed for the Heckman model for H1.

⁶⁴ Horn, Reinhart and Trebesch, ‘How Much Money Does the World Owe China?’.

⁶⁵ Kaplan, Globalizing Patient Capital: The Political Economy of Chinese Finance in the Americas.

⁶⁶ Jonathan Watts, ‘Chávez says China deal “great wall” against US’, The Guardian. London. https://www.theguardian.com/world/2006/aug/25/venezuela.china.

⁶⁷ ‘Venezuela aumenta su deuda millonaria con China’ Radio Televisión Martí. Accessed January 12, 2023. https://www.radiotelevisionmarti.com/a/venezuela-aumenta-su-deuda-millonaria-con-china/210355.html.

⁶⁸ Charlie Mitchell, ‘Debt write-off shows China means business in Africa’, The Times. London. Accessed January 12, 2023. https://www.thetimes.co.uk/article/qin-gang-lures-ethiopia-with-huge-debt-write-off-of-13–7-bn-loan-dpkn3c6cd.

⁶⁹ Jianghui Dong, ‘Chinese FM expounds on developing China-Africa relations’, Xinhua. Accessed January 12, 2023. https://english.news.cn/20230112/07e9077fd7d1427da39b21a3f16697b2/c.html.

⁷⁰ Ben Norton, ‘China forgives 23 loans for 17 African countries, expands “win-win” trade and infrastructure projects’, Global Economy. Accessed January 12, 2023. http://geopoliticaleconomy.com/2022/08/20/china-forgives-debt-africa/.

⁷¹ Briggs, ‘Causal Inference and the Heckman Model’.

⁷² Dreher and others, ‘Apples and Dragon Fruits: The Determinants of Aid and Other Forms of State Financing from China to Africa’.

⁷³ Kaplan, Globalizing Patient Capital: The Political Economy of Chinese Finance in the Americas.

⁷⁴ M. Ayhan Kose and others, ‘A cross-country database of fiscal space’, Journal of International Money and Finance 128, (2022), p. 102, p. 682.

⁷⁵ Zhao, The Dragon Roars Back: Transformational Leaders and Dynamics of Chinese Foreign Policy.

⁷⁶ Raess, Ren and Wagner, ‘Hidden Strings Attached? Chinese (Commercially Oriented) Foreign Aid and International Political Alignment’.