Kenya’s $2.25 billion did not build any roads or hospitals. Nairobi used the money to buy back bonds maturing in 2028 and 2032, paying yields of 8.1% and 8.95%.
From a liability-management perspective, this makes sense: it pushes back a wall of maturities and reduces refinancing risk. Yet a government that collects taxes in shillings locked in a rate of nearly 9% in dollars. A little earlier, in the same wave, Côte d’Ivoire issued 15-year bonds at 5.39%. Two African countries, separated by more than three percentage points. That spread says more about the “Africa premium” than any continental average.
The debate over the premium has long been divided into two camps. The African Peer Review Mechanism, the UN Economic Commission for Africa, the UN Development Programme and Afreximbank argue that Africa pays a penalty because of bias. A 2023 study commissioned by the UNDP estimated that fairer ratings could save up to $74.5 billion a year, while the African Finance Corporation has referred to a “bias premium.”
On the other side is a 2023 IMF working paper, which finds that the premium disappears once financial-sector depth, fiscal transparency, the size of the formal economy and institutional quality are taken into account. Moritz Krämer, a former chief ratings analyst at S&P, goes further. In his view, African countries were overrated because African issuers defaulted slightly more often than other issuers with the same ratings.
In my view, both sides are half right. The IMF’s finding does not prove that the premium is fair. It shows what the premium is linked to: narrow domestic capital markets, poor data and debt denominated in a currency the government cannot print. All of this can be fixed, but attacks on rating agencies will fix nothing. The figures for Kenya and Côte d’Ivoire point to the same conclusion: investors clearly assess African borrowers individually.
Some of the demand came from investors looking to move out of U.S. assets just as borrowing costs fell. Africa had nothing to do with that. S&P describes borrowing by governments in the region as procyclical and dependent on commodity prices. As early as March, it warned that war in the Middle East could derail borrowing plans if central banks started raising rates. The agency expects the 27 African countries it rates to borrow $155 billion on commercial terms this year, up from $140 billion in 2025. Windows opened by global sentiment close on the same sentiment—and rarely for everyone at once.
This is where the issuance boom runs up against the history of restructurings. Kenya, Côte d’Ivoire and Gabon have access to the market. Senegal is still negotiating with the IMF after hidden debt shut it out of the market, while Ethiopia has yet to reach a deal with bondholders. This year, market access has returned to a select group of African issuers; the rest are waiting.
South Africa’s first infrastructure bonds raised $693 million, with orders more than twice that amount. Afreximbank spread its funding across “samurai” and “panda” bonds, then returned to the dollar market in July with a $1.5 billion issue. Senegal shows what this can look like: the IMF counts its total-return swaps as external debt, while borrowing in the regional market accounts for nearly 58% of debt-service costs in the revised 2026 budget. Whatever these instruments may have saved, their main effect has been to make Senegal’s debt harder to see.
The African Union’s own answer is the African Credit Rating Agency, AfCRA. After a series of failed attempts, it was scheduled to launch in October 2025. It could give the market more data and help investors get to know the region better. The real test will be trust. If investors suspect the agency of being too generous with its ratings, they simply will not pay attention to them.
What should finance ministries do while the window is open? Use it to extend maturities and repay expensive debt, as Kenya did, rather than spend the money on day-to-day expenses. Every Eurobond issue should be accompanied by efforts to develop local-currency markets; a reliable way out of the dollar premium is to need fewer dollars. Most importantly, publish all debt in full, including swaps, collateralized loans and regional bonds. Of all the factors driving the premium, transparency is the only lever entirely in governments’ hands—and it costs less than waiting for an upgrade.
Author: PhD candidate and teaching assistant at the Department of World Economy and International Finance, Financial University under the Government of the Russian Federation, Rafael Mulenga