BLOG POST

Publish What You Lend

Just how bad is the debt crisis in developing countries? We know that it’s bad—most African countries are currently spending more on repayments than on health or education. But we don’t know exactly how bad because some lending is never actually made public. There may be as much as $1 trillion in “hidden debt” according to one recent estimate. This suggests one relatively simple policy change.

Today’s debt crisis is harder to solve than the last one, 30 years ago, as there are more lenders involved (Figure 1). This limits what traditional donor governments can do. But there are still concrete steps they can take to improve the transparency of both public and private lending. Whilst this won’t solve the debt crisis, it would definitely help.

Figure 1: Total debt service by African countries, as a share of government revenues

Note: Data from the World Bank IDS and IMF WEO (with additional interpolation by Gethin, 2024)

In 2021, according to a World Bank report, nearly half of all low-income countries had published no recent data on sovereign debt. This lack of data leads to estimates of debt burden which differ by as much as 30 percent of GDP, depending on who is reporting.

Low-income countries should, of course, do a better job of publishing their own debt contracts. Citizens pay for government borrowing through taxes, so they have a right to know what they are being signed up for. Greater transparency may also give lenders greater confidence, and lower interest rates. 

But regardless of what those borrowing governments do, rich countries like the UK can also play a role.

Would greater transparency really improve borrowing terms?

Evidence suggests that when borrowing countries improve their own transparency, borrowing costs can fall. A 2008 IMF study by Rachel Glennerster and Yongseok Shin found that publishing improved IMF data on the economy and financial regulation in borrowing countries led to reduced borrowing costs. Similarly, increased transparency on the private financial sector and on fiscal policy are associated with cheaper borrowing. 

More recently, there is a hitherto unexamined natural experiment in Cameroon. Between 2016 and 2020, Cameroon was the only developing country to publish all of its loan contracts in their entirety online (as noted by our former colleague Scott Morris and co-authors in their paper “How China Lends”). Did Cameroon’s unilateral increase in loan transparency reduce its borrowing rates? We only have a small number of data points here so take this analysis with a grain of salt. But we can compare trends in average borrowing rates for Cameroon with other African countries, before and after Cameroon began publishing their loan contracts online. The result is that we can see that Cameroon did closely track other African countries, until around 2016 when they started publishing loan contracts. At which point a roughly two percentage point gap opens up. This gap then shrinks again in 2020 when the Cameroon debt contract website went offline (Figure 2). This difference is statistically significant in a simple difference-in-difference regression framework. Alongside the reduction in borrowing costs, there was also a reduction in borrowing volumes. Full contract transparency may therefore have led to better rates reducing perceived risks related to corruption, governance, and hidden liabilities, or it might have simply averted the most expensive loans from being signed at all. The improvement in transparency was also associated with an improvement in Cameroon’s score on the World Bank’s rating of its fiscal policy (CPIA score), and in its fiscal balance. 

Figure 2: Borrowing costs in Cameroon were lower than other African countries in the 5 years they had full loan contract transparency 

Note: Data on fiscal balance from Ayhan Khose et al 2022. All other data from World Bank International Debt Statistics (IDS) and CPIA database; DT.INR.PRVT; DT.INR.DPPG; DT.COM.PRVT.CD; DT.COM.OFFT.CD; IQ.CPA.FISP.XQ. Dotted vertical lines indicate 2016 (when the internet webarchive identifies the first occurrence of the debt contract website on the Ministry of Economy and Planning website http://dad.minepat.gov.cm/) and 2020 (when the website was last online).

So, what should rich countries do?

First, donors should get their own house in order, and make sure they fully report on bilateral lending. Paris Club donors do well on transparency after a borrower is in debt distress, but that is too late. G7 countries committed in 2021 to publishing information on a loan-by-loan basis, but the UK is so far the only country to have done so. Non-Paris Club donors are generally less transparent.

Second, donors can support borrower governments to improve their own transparency. They can provide a range of technical assistance to low- and middle-income countries, developing better standard reporting templates, providing embedded advisors, as well as supporting scrutiny by parliaments and civil society (these suggestions are elaborated in the USAID Debt Transparency Monitor 2022).

A third approach is to focus instead on private lenders from rich countries. Most private lending takes the form of publicly-traded bonds, for which information is inherently public too. It is private loan contracts that are not routinely made public. The OECD debt transparency initiative (2022) runs a voluntary scheme, gathering data from lenders on debt owed by low-income country governments, but this is under-used. Similarly, UN Trade and Development (UNCTAD) has developed a set of principles for responsible lending to sovereign governments, but these are again entirely advisory. A stronger approach would require lenders to register new loan agreements, if they want to have any legal recourse in case of non-repayment. This would make cases like the Mozambique tuna bond scandal—in which London-based Credit Suisse bankers paid bribes to corrupt officials, committing the government to $2 billion in guarantees—impossible. These officials did not have the appropriate authorization to make these guarantees, and yet the default responsibility falls on the government of Mozambique. If Credit Suisse had been required to make a public notification of the agreement before any funds were transferred, the problem could have been avoided. Enforcement of such a system would require a legal change that would nullify any agreement that was not publicly notified (for more detail on the practicalities of such a proposal see Connelly 2021). This is a proposal supported by the Debt Justice Campaign and 50 cross-party members of the UK parliament in a 2019 letter.

Improving transparency around international debt is no silver bullet. But it would help to avoid some of the worst excesses, like the Mozambique scandal, and should also lower overall borrowing costs, as the case of Cameroon suggests. They say sunlight is the best disinfectant. In sovereign lending, it may also be a surprisingly effective interest rate reducer.


To walk the talk on transparency, you can find the data and code underlying the analysis on Cameroon here. Thanks to Justin Sandefur, Clemence Landers, and Tom Hart for helpful comments.

Disclaimer

CGD blog posts reflect the views of the authors, drawing on prior research and experience in their areas of expertise. CGD is a nonpartisan, independent organization and does not take institutional positions.