This symposium contribution draws on newly identified contract texts and other primary sources to analyse transaction design patterns in secured overseas lending by Chinese institutions. Together with its companion quantitative study of 620 collateralized debt transactions between Chinese lenders and public sector borrowers in emerging market and developing economies (EMDEs), it offers a rare window into some of the relationships and practices that animate debates about geopolitics and finance. Collateralization is a regular feature of Chinese lending in EMDEs. Contrary to stereotype, Chinese lenders are far more likely to use control over liquid assets—revenue streams and bank accounts—than ports or mineral deposits to secure risky public and publicly guaranteed (PPG) debt. We trace the design, provenance, and functions of secured credit in the most common transaction structure in the dataset, and examine two variations on the theme: a railway construction project in Kenya and a road construction program in Ghana. In the case studies and in the companion dataset, we find that Chinese lenders adapt and scale market-standard tools, experimenting with legal techniques to maximise recovery, broaden and deepen the debtor–creditor relationship. Transactions that give creditors the ability to block or capture foreign currency cash flows are commonplace in our dataset and have roots in 19th and 20th century sovereign debt practice. Design experiments for risk mitigation can give creditors economically and politically significant control over export revenue flows in vulnerable borrowing countries. They can also set off competition for flow control among lenders, which can delay or derail crisis resolution when the borrower falls into debt distress. Complex collateralized debt structures obscure governance problems and misaligned incentives and undercut borrower agency. Their apparent prevalence, and the fact that they have gone unaddressed for decades, point to significant gaps in the legacy international financial architecture.

Publication 

This paper is the first comprehensive analysis of the secured lending practices of Chinese creditors in emerging market and developing economies (EMDEs). We present a new dataset and detailed case studies of collateralized public and publicly guaranteed (PPG) loans from Chinese state-owned institutions in EMDEs between 2000 and 2021. Almost half of China's total PPG loan portfolio in EMDEs is effectively collateralized—amounting to $420 billion in collateralized debt across 57 countries. We document that Chinese lenders use techniques adapted from export and project finance to build multi-layered legal safety nets, which help ensure that risky EMDE loans will be repaid. As security, they use liquid, easily accessible assets, such as cash in bank accounts located in China. They rarely take infrastructure project assets as collateral, but often rely for repayment on established commodity revenue streams unrelated to the project. Typically, EMDE governments and state-owned enterprises commit to route foreign currency proceeds from commodity sales through bank accounts controlled by the lender. The cash balances in these accounts can be very large; in low-income, commodity-exporting countries, they average more than 20% of annual PPG debt service to all external creditors. The same revenue source can secure multiple successive borrowings over many years. Our findings reveal a previously undocumented pattern of revenue ring-fencing, where a significant share of commodity export receipts never reaches the exporting countries. Revenues routed overseas secure priority repayment for the creditor; they remain out of public sight and largely beyond the borrower’s reach until the secured debts are repaid. These findings raise new concerns about debt transparency, fiscal management, fiscal autonomy, and the quality of macroeconomic surveillance, particularly in commodity-exporting EMDEs.

PIIE Working Paper | Data

Preliminary abstract:

This paper studies how collateral affects the performance and restructuring of sovereign debt. I exploit newly assembled loan-level data on Chinese official lending, where collateralization is unusually widespread and often takes the form of creditor control over dedicated revenue streams and bank accounts. I examine whether collateral reduces the incidence of repayment distress and how it shapes outcomes once distress occurs. Preliminary results suggest that collateral does not prevent distress, but it matters for its resolution: collateralized loans are more likely to be resolved without formal restructuring and, conditional on restructuring, tend to generate smaller creditor losses. The findings highlight collateral as an important enforcement technology in sovereign lending and shed light on how contractual seniority can shape bargaining between sovereign borrowers and creditors.

Preliminary abstract:

This project studies the persistence of political connections after regime change using German firms connected to the Nazi regime. We examine how the collapse of the regime affected firm performance and whether differences in postwar outcomes can be explained by the persistence of managerial, organizational, and physical capital. The project sheds light on which components of politically connected firms’ advantage are regime-specific and which survive major political discontinuities.

In this brief research note, we introduce the 2.0 version of the How China Lends (HCL 2.0) dataset. Our new and expanded dataset comprises 371 debt contracts between 20 Chinese state-owned creditors and 155 borrowers in 60 countries signed between 1990 and 2025. All of these contracts can be accessed through an online repository at http://china-contracts.aiddata.org. We use the updated data to revisit key findings from our 2021 study and confirm that they continue to hold in the larger sample. In particular, we confirm the extensive use of collateralization, borrower confidentiality clauses, and “No Paris Club” clauses.

Policy Brief | Data