Restructuring gets the summits; debt management gets the training workshops. Yet in a low-income country, the debt office is crisis-prevention and early-warning infrastructure, not a narrow technical function – and a fundamental institutional reform is overdue.
All four G20 Common Framework applicants (Chad, Ethiopia, Ghana and Zambia) had passed through the International Monetary Fund and World Bank Heavily Indebted Poor Countries initiative, with relief between 2004 and 2015. Each returned to distress.
The prescriptions repeat after every round: consolidate, reform, sharpen the debt sustainability analysis and publish more. All necessary, but insufficient. If distress keeps recurring under the same institutional design, the design is part of the diagnosis.
The missing reform has an address: the debt office – largely seen as the hinge between the fiscal and financial systems and the sole authority overseeing the consolidated perimeter, from the borrowing plan to guarantees. It did not cause the shocks, but every shock crossed its desk, becoming rollover risk, bank exposure and a late warning. Weak design then widens each passage.
Hence, the reform must be institutional: statute elevating the office, explicit guidance written for these countries and accountability for outcomes within five years. The evidence is drawn principally from African economies; the institutional structures are not.
The load on the hinge
A market-access debt manager is a price-taker in one continuous market; a low-income debt manager works two disconnected ones: abroad, an episodic window too fragile to anchor a financing strategy, and at home, where regulation, liquidity needs and scarce alternatives make sovereign paper the default asset and the banking system the sovereign’s captive marginal lender.
The sovereign-bank nexus is growing faster in sub-Saharan Africa than anywhere; in the West African Economic and Monetary Union, government debt is 43% of banks’ assets. Some 41% of new debt in low-income countries in 2025 was short-term. Ghana restructured in 2023 and transitioned to weekly Treasury bill auctions; by late 2025, its domestic market debt averaged below three months.
In much of Africa, the frontier is not optimising the curve but extending it beyond 90 days. The sovereign funds themselves through three channels: banks; non-bank pools, with over 80% of the continent’s $2.1tn in institutional capital sitting in government treasuries, and obligations that never reach an auction, including guarantees, state-enterprise borrowing, and arrears. Call it ‘a shadow deficit’. In Chad, the largest commercial loan (over 40% of the portfolio) was not recorded in the state’s recording system.
Any arrangement that encumbers future public resources belongs inside the perimeter, whatever its label. No one assembled the whole: domestic debt sat in fragments until the African Debt Database built the first harmonised picture in 2025, drawing on official sources that no one had joined.
Three tasks
What can be done? First, steward the sovereign balance sheet, not just the debt portfolio: manage liabilities against assets, align currencies with revenues, integrate cash with issuance, and project the shadow deficit as it accumulates rather than discovering it only when it has crystallised. This is the minimum condition for solvency management.
Second, build the investor base as the central policy variable. Shifting paper from captive banks to captive pension funds diversifies names, not risk, and what is needed are holders that price duration and absorb losses. Tanzania and Uganda have lengthened their curves. The route off the 90-day treadmill runs through the investor base, not around it.
Third, watch and warn. The Debt Sustainability Analysis is an annual photograph; the auction book is a live monitor, registering stress as it forms – in Ghana, one auction in five has gone under-subscribed since the 2023 restructuring. Yet no one is obliged to act on what it shows.
The debt office, therefore, warrants a statutory duty to warn, not merely an authority to borrow, built on three elements: protected reporting lines beyond the finance ministry to a legislature equipped to act on them, pre-committed distress indicators, and graduated disclosure. Pre-commitment converts warning from an act of courage into an act of compliance – early warning without early panic. Transparency is the cheapest credit enhancement a poor country can buy.
The call
While the three tasks define the function, the three deliverables below would implement it, ideally codified in a new set of public debt management guidelines for low-income countries.
- Statute: make the debt office an economic agent of the state, with its own objective, board, perimeter authority defined by economic substance, not instrument form, and a warning function no minister can silence. It confers operational independence and interagency parity, not policy autonomy; borrowing remains politically accountable.
- Framework: an explicit, distinct low-income debt-management framework, the global architecture’s missing chapter, that joins sustainability analysis, strategy, and diagnostics through a single perimeter, a single data layer, and a single escalation protocol. It would build on, not replace, real Fund, Bank, and partner capacity work through the Debt Management Performance Assessment, the Medium-Term Debt Strategy and the Low-Income Countries’ Debt Sustainability Framework, etc. The gap is not debt-issuance advice, recording systems or cost-risk practice; it is a country-owned operating map for the whole sovereign balance sheet: liabilities, guarantees, arrears, cash, bank exposure and warnings.
- Delivery: an architecture held to outcomes. The Fund, the Bank, UNCTAD, the Commonwealth, the Organisation for Economic Co-operation and Development, and regional agencies, as their own evaluators found, answering, with countries, to the scorecard below. To ‘we already do this’, the answer is one word: publish and explain. Training counts are inputs; a curve beyond 90 days is an outcome.
Figure 1. The five-year debt management integration scorecard
| Dimension | Core actions (sovereign-controlled, scored) |
| Balance sheet | Consolidated balance sheet published annually; guarantees, arrears, collateralised obligations, and central bank advances inside the perimeter; shadow deficit projected; borrowing, cash and issuance plans reconciled |
| Macro-financial | Sovereign-exposure dashboard published jointly by the debt office, the central bank and the regulator |
System signal |
Composite funding-stress indicator; statutory escalation ladder; record of warnings and plan adjustments |
| Meta | Budget, DSA and debt bulletin reconcile to one dataset |
| Accountability | Mandate in statute; regular reporting to parliament; committee review of guarantees and collateralised obligations. |
Source: Author’s framework
Note: Floor (assumed, not scored): the metrics prescribed by existing international guidelines. Scoring is binary – published or not, enacted or not. Tracked separately as market diagnostics: holdings concentration; auction cover, tails and under-subscription; the shadow deficit’s size.
The next round of distress is being assembled now, auction by auction, in guarantees that no central ledger carries. The debt office did not cause every shock; it is where shocks become rollover risk and where the warning appears first.
Two decades of relief and reform treated debt management as a capacity problem. In a low-income country, it is statecraft; the office must stand as an institution of the state, mandated, autonomous and accountable. Reforming it and holding the sovereign and parliament accountable for outcomes means the debt cycle finally has a circuit breaker.
Udaibir Das is Vice Chair of OMFIF, Member of the Bretton Woods Committee, Distinguished Fellow at the Observer Research Foundation-America, Visiting Professor at the National Council of Applied Economic Research, and Senior Adviser of the International Forum for Sovereign Wealth Funds.
Interested in this topic? Subscribe to OMFIF’s newsletter for more.