Consider an illustration set out in July by a treasury specialist at the Africa Finance Corporation. A Nigerian power project raises dollars at 5 percent for five years, and the sponsor’s model celebrates it as the cheapest money on the table. Then someone asks what the loan costs in naira, the currency in which every customer will pay. Converting the dollars through a swap adds about 20 percent a year, pushing the fully hedged cost into the mid-twenties. The 5 percent was never the price. It was the price of a risk that someone else was about to carry. Few models show that second number
By Chidi Nwafor
Consider an illustration set out in July by a treasury specialist at the Africa Finance Corporation. A Nigerian power project raises dollars at 5 percent for five years, and the sponsor’s model celebrates it as the cheapest money on the table. Then someone asks what the loan costs in naira, the currency in which every customer will pay. Converting the dollars through a swap adds about 20 percent a year, pushing the fully hedged cost into the mid-twenties. The 5 percent was never the price. It was the price of a risk that someone else was about to carry. Few models show that second number.
In my last article I argued that infrastructure needs a payer who can be named, a price that covers cost, a payment that can be seen and an arrangement that outlasts an election. Even then, one problem remains. The money collected arrives in naira, cedis or shillings, while much of the money lent must be repaid in dollars or euros. The Missing Middle of Currency is the gap between the currency in which infrastructure earns and the currency in which it is financed. Nobody owns it.
The mismatch is structural, not accidental. TCX, the currency hedging fund set up by development finance institutions, states in its own materials that over 90 percent of development finance loans are made in hard currency. It describes depreciation as the major driver of a rise of more than 50 percent in Africa’s debt between 2013 and 2020. These are the fund’s own figures, and they come from a provider with an interest in the problem, but the direction is not in dispute.
The arithmetic is unforgiving. Take a purely illustrative case: a project borrows $10 million when a dollar buys N1,000 and expects annual revenue of N10 billion to service it. If the rate moves to N1,500, the same debt now consumes N15 billion in naira terms, a 50 percent heavier burden, while nothing about the plant, the customers or the management has changed.
Nigeria supplied the proof. After the central bank unified the foreign exchange windows in June 2023, the naira fell 55 percent over the year to about N1,043 to the dollar at the official market, the third-worst performance among the 151 currencies Bloomberg tracks. A tariff set in naira cannot follow a move of that size, and a dollar loan does not wait for it to try.
In power, the damage runs through the whole chain. Gas is priced in dollars. So are much of the plant and equipment, and the loans behind several privatised generators. Revenue is in naira and, as I showed last time, only partly collected. The exchange rate is among the triggers written into Nigeria’s multi-year tariff order for review, but a trigger is not a payment. A currency clause is only as good as the political will to apply it, which returns the argument to permanence. Between a tariff review and a devaluation, someone carries the difference: the generator, the gas supplier or the lender. The sector’s legacy debt to generators, about N4 trillion before verification and about N3.3 trillion after, is the accumulated residue of many such gaps, currency among them.
The Minister of Power recognised part of this in February 2024, when he proposed that domestic gas be paid for in naira, calling it a key step toward stability. The instinct was right: align the currency of the input with the currency of the revenue.
Hedging is the usual answer, and it is not free. The Africa Finance Corporation specialist breaks the cost into three layers. A liquidity premium arises in a shallow market with few quoting banks. A credit premium compensates for counterparty risk, which development institutions currently carry deal by deal. And there is the gap between naira and dollar interest rates, which narrows only as inflation and fundamentals converge. That last layer, he notes, is one no Nigerian institution is mandated to bear.
Progress is real. TCX has provided over $17 billion of hedging since 2007, including $4 billion across 31 African countries. The African Development Bank approved a $25 million equity investment in it in September 2025, and the European Union signed a €150 million guarantee with it in 2025. These are the right instruments. But a hedge prices the gap; it does not close it. Lenders understand this. They respond by shortening tenors, demanding guarantees or walking away, which is how a currency problem becomes a capital problem.
Why does this persist? Because each party behaves rationally within its mandate. A development lender funds itself in hard currency and therefore lends in it. A sponsor racing to financial close takes the cheaper headline rate and assumes the exchange rate will behave. A regulator sets a tariff in local currency and treats devaluation as an event rather than a design variable. A central bank manages the currency, not the infrastructure sector’s exposure to it. Like any unbooked subsidy, the risk lands on whoever cannot refuse: the offtaker’s creditors, the generator or the customer.
The limits matter. Currency is not the binding constraint where revenue is itself dollar-linked, as with some export-oriented projects or industrial customers who earn in hard currency. And local-currency lending does not rescue a project whose payer does not pay; a naira loan to an insolvent utility is still a bad loan. Nor does a hedge cure a tariff set below cost; currency and price are separate failures, and they compound. The honest test is whether revenue and debt service sit in the same currency. The argument applies where revenue is local, costs and capital are not, and nobody has decided who bears the difference.
Where it applies, the responsibilities are clear. Development finance institutions should lend more in local currency, or fund the hedge as part of the loan rather than leave sponsors to find it. Regulators should write currency adjustment into tariff formulas that are transparent and rule-based, so that pass-through is a mechanism rather than a negotiation. Central banks and market regulators should decide who carries the long-dated interest differential, and say so. Governments should keep aligning the currency of domestic inputs with that of domestic revenue. Local capital markets matter most of all: The N501 billion bond issued in January and the N728.9 billion issued in August under the government’s power sector debt reduction programme show that long-dated naira paper can be placed, though under state sponsorship, which is not the same as a market for private projects. And sponsors and investors should price the hedged cost, not the headline rate, and treat an unhedged dollar loan against naira revenue as a bet on the exchange rate, not a financing.
The treasury officer’s question deserves a place beside the credit analyst’s: what does this loan cost in the currency we earn? A project with a named payer and a cost-reflective price is still a wager if its debt is in the wrong currency. That is the Missing Middle of Currency. Yet even a project with matched currency and a sound revenue chain rests on one more signature: that of a public counterparty with the authority and the capacity to honour it. That is a question for another day.
Nwafor is founder and lead strategist at De-Lazuli Consult, an advisory practice focused on energy transition, project finance & Investment readiness, DFI engagement and infrastructure investment facilitation. He writes from Abuja.
