The Cost of Capital in Developing Countries
This paper analyzes a unique dataset of 845 infrastructure transactions across 66 developing countries to reveal that borrowing costs often defy traditional sovereign-risk models, with private projects sometimes borrowing below sovereign rates and investor confidence significantly lowering costs over time, suggesting that policies catalyzing initial market entry could effectively reduce the cost of capital.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Money has a price, just like any other commodity. In the world of finance, this price is called the cost of capital. It is the interest rate a company or a government must pay to borrow money to build something new, such as a power plant, a road, or a school. For wealthy nations, this price is usually low because lenders feel confident they will get their money back. For developing nations, however, the price is often much higher. This difference is not just a number on a spreadsheet; it is a barrier that decides whether a country can build the clean energy systems it needs to fight climate change or simply provide electricity to its people. When the cost of borrowing is too high, projects that would be profitable in a stable economy become impossible to finance in a developing one.
For years, experts have tried to understand exactly how high these prices are and what drives them. Because direct data from actual bank loans in these countries is rare, researchers have relied on estimates, surveys, and indirect clues like government bond ratings. These methods provide a general picture, but they often miss the specific details of how real money changes hands. Without seeing the actual numbers from real deals, it is difficult to know if policies designed to lower borrowing costs are working or if the standard ways of measuring risk are even accurate.
A new study by researchers from Harvard and Oxford changes this picture by assembling the first large-scale collection of real transaction data from the developing world. The team gathered information on 845 specific infrastructure deals, ranging from solar farms to mining operations, across 66 different countries. They spoke directly to 19 different investors, including banks and investment funds, to get the actual interest rates these projects paid. This is not a guess or a model; it is a record of money actually lent and borrowed between 2009 and 2026.
The researchers found that the cost of borrowing varies wildly from one country to another, but not always in the way people expect. They discovered that the standard tools used to measure a country's risk, such as its credit rating or the interest rate on its government bonds, explain very little about the actual rates private companies pay. In fact, nearly half of the projects in their dataset borrowed money at a lower rate than their own country's government. This suggests that a project's specific details, like who is building it and where it is located, matter more than the broad label of the country itself.
One of the most striking discoveries in the data is what the authors call the "investor confidence effect." The study shows that when an investor lends money to a project in a country for the first time, they charge a high interest rate. However, as that same investor completes more deals in that specific country, the price they charge drops significantly. By the time an investor has made five deals in a country, the interest rate they offer can be more than two percentage points lower than for their very first deal. This drop is so large that it is bigger than the difference in borrowing costs between the most expensive region in the study, Sub-Saharan Africa, and the cheapest, Southeast and East Asia.
The data suggests that this drop in price happens because investors become more comfortable with a market over time. They learn how local regulations work, they get to know the local partners, and they gain confidence that the projects will succeed. This learning process lowers the risk they perceive, and they pass those savings on in the form of lower interest rates. The study also found that development finance institutions, which are often expected to offer cheaper loans to help developing nations, actually charged slightly higher rates than private investors in this dataset, though the difference was not statistically significant.
The researchers are careful to note that their work describes a pattern rather than proving a cause. They cannot say for certain that the experience of making a deal is what causes the lower rates, as it is possible that investors who are already good at lending in a country simply choose to make more deals there. However, the pattern is strong and consistent across different types of investors and sectors. The study implies that if policymakers can help investors make their first few deals in a new market, perhaps by sharing some of the risk, it could trigger this cycle of learning and lower rates. This could lead to a permanent reduction in the cost of capital for that country, unlocking investment that would otherwise remain out of reach.
This research provides a clearer view of the financial landscape in the developing world, moving beyond broad estimates to the reality of individual transactions. It highlights that the high cost of capital is not just a fixed feature of geography but a dynamic price that can change as relationships between investors and markets grow. By understanding that familiarity breeds lower costs, there is a new path forward for helping developing nations build the infrastructure they need for a stable and prosperous future.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.