Off-Book Liabilities Across the Global South

Markets obsess over headline debt ratios. Governments obsess over published budget deficits. The real sovereign risk often sits elsewhere: in state-owned enterprises bleeding cash, guarantees quietly accumulating, subsidies becoming structural, power utilities collapsing under below-cost tariffs, and "temporary" emergency fixes that become permanent fiscal anchors. In 2026, the next wave of solvency shocks will not come from what states publicly admit they owe, but from what they have quietly, inevitably promised to absorb when the crisis arrives.

The modern sovereign balance sheet is elaborate theatre. The official script features familiar characters: "public debt" quoted as percentage of GDP, "fiscal deficit" debated in parliament, "primary balance" monitored by the IMF. The stage crew works in darkness: state-owned enterprises accumulating losses, loan guarantees signed quietly in ministerial offices, public-private partnerships with minimum-revenue clauses that kick in during downturns, central bank quasi-fiscal operations that look like monetary policy but function as hidden subsidies, and mounting arrears to suppliers that governments refuse to classify as debt. When the lights fail and crisis arrives, everyone discovers the crew has been running the show all along.

This is not a technical accounting footnote relegated to IMF staff reports. It is a primary explanation for why debt crises across the Global South so often appear sudden to markets, yet turn out to have been slow-building, visible to insiders, and politically managed through deliberate opacity. A country can maintain "stable" headline debt ratios of 50-60% of GDP while its national power utility accumulates circular debt equivalent to 4-6% of GDP, its development banks make politically directed loans that will never be repaid, its state-owned airlines and railways borrow in foreign currency against implicit sovereign guarantees, and its treasury quietly signs revenue guarantees for toll roads and power plants that embed long-dated fiscal commitments. The state does not disappear these risks through creative accounting. It only delays recognising them formally until crisis forces recognition.

By 2026, that recognition is arriving for multiple economies simultaneously. Not through sudden revelations, but through the collision of three forces: refinancing walls that demand full disclosure to creditors, foreign exchange stress that exposes dollar-denominated off-book obligations, and fiscal space so constrained that governments can no longer absorb hidden losses silently.

Quantifying the hidden balance sheet

The scale of off-book sovereign risk varies dramatically by country, but several patterns emerge consistently from World Bank and IMF data on fiscal transparency and contingent liabilities.

The pattern is consistent and alarming. Countries already facing debt distress (Pakistan, Egypt, Sri Lanka pre-default) carry additional hidden exposures of 20-30 percentage points of GDP that are economically real but fiscally unrecognised. When crisis forces recognition, published debt ratios can jump 15-25 percentage points overnight through SOE debt assumptions, guarantee calls, and arrears regularization, turning manageable situations into insolvency.

Even countries with apparently comfortable headline debt levels face material hidden exposure. South Africa's 71% official debt conceals estimated 20-25 points of additional state-owned enterprise obligations, primarily Eskom (electricity) and Transnet (ports/rail). Indonesia and Vietnam maintain better discipline but still carry 6-10 points of off-book risk through development bank lending and infrastructure guarantees.

The five channels where risk hides

Off-book liabilities accumulate through five primary institutional channels, each with distinct accumulation dynamics and crisis triggers. Understanding these mechanisms is essential for assessing which countries face imminent recognition crises.

State-owned enterprises: the sovereign risk factory

SOEs represent the single largest category of hidden sovereign exposure across emerging markets. The World Bank estimates SOEs account for 12-15% of GDP in annual economic activity across developing countries, employ 15-30m workers globally in major economies, and carry combined debt exceeding $8trn worldwide with approximately $2.5-3trn in emerging and frontier markets alone.

Energy utilities dominate SOE risk profiles. Across multiple countries, electricity is systematically priced below cost recovery to manage inflation and maintain political stability. The resulting losses are financed through accumulated arrears to fuel suppliers, emergency government transfers, short-term commercial borrowing at punitive rates, and deferred maintenance that progressively destroys generation and distribution capacity. This looks like consumer protection in the short term. It behaves like contingent sovereign debt in the medium term. It becomes an explicit fiscal crisis in the long term.

Pakistan's National Transmission and Distribution Company, along with distribution companies (DISCOs), exemplifies the terminal stage of utility-as-hidden-liability. Tariffs have been held 30-40% below cost recovery for decades. Theft and non-payment rates run 15-20%. Government ministries themselves are among the largest defaulters on electricity bills. The resulting circular debt—unpaid obligations cascading between generators, transmission, distribution and government—grew from $3bn in 2013 to $13bn by 2024. The state now pays $1.2bn annually merely to service the interest on circular debt, equivalent to 40% of the education budget, while load-shedding of 8-12 hours daily during summer costs the economy $2bn monthly according to Pakistan Business Council estimates. This represents 6% of GDP in annual economic losses from a "subsidy" meant to protect consumers.

South Africa's Eskom provides an even starker example of how hidden liabilities destroy both utilities and sovereigns. Eskom accumulated R423bn ($23bn) in debt by 2024, equivalent to roughly 8% of GDP, primarily from political instructions to maintain below-cost tariffs during the 2010s while coal costs rose, demand patterns changed, and systemic corruption under state capture extracted billions. The utility now requires government bailouts of R60-80bn annually ($3.3-4.4bn) while implementing Stage 6 load-shedding—meaning rotating blackouts exceeding 6 hours daily—that cost the South African economy an estimated R200-250bn annually ($11-14bn, roughly 3-4% of GDP) in lost production, damaged equipment, and foregone investment.

The fiscal impact extends beyond direct bailouts. Eskom's debt service absorbs R50-60bn annually. Its declining reliability forces businesses to install backup generators costing 15-25% of capital budgets and consuming diesel at 4-6x grid power costs. Manufacturing competitiveness collapses. Foreign investment avoids energy-intensive sectors. The "cheap" electricity delivered through below-cost tariffs costs South Africa 3-4% of GDP annually in lost growth, making it perhaps the most expensive "subsidy" anywhere in the developing world relative to benefits delivered.

Government guarantees: the contingent debt that becomes certain

Explicit government guarantees on SOE and private sector debt are systematically underestimated in sovereign risk assessment because they carry "contingent" classification until called. But contingent liabilities behave exactly like debt when macroeconomic stress arrives.

The pattern across high-risk guarantee portfolios is consistent: energy sector dominates (utilities, independent power producers, fuel importers), guarantees often denominated in foreign currency creating FX risk concentration, and beneficiaries already financially distressed meaning guarantees will be called in predictable sequence once external conditions tighten.

Sri Lanka's pre-default experience demonstrates how quickly contingent becomes certain. The government carried approximately $8-10bn in explicit guarantees (8-10% of GDP) on Ceylon Electricity Board debt, Ceylon Petroleum Corporation borrowing, and SriLankan Airlines obligations. When foreign reserves depleted in 2021-2022, these entities could no longer service dollar-denominated debt. Within months, guarantees were called, forcing the sovereign to assume liabilities at precisely the moment it lacked capacity to service its own obligations. This accelerated the path to default.

Turkey maintains one of the world's largest guarantee portfolios at $90-110bn (10-12% of GDP), concentrated in infrastructure PPPs with minimum-revenue clauses and currency guarantees. As the lira depreciated 80% between 2018-2024, the local-currency cost of servicing these guarantees exploded. The government now makes regular transfers to cover shortfalls in toll road revenues, hospital bed occupancy, and power plant capacity payments. What was "contingent" became a permanent budget item absorbing 1-2% of GDP annually.

Public-private partnerships: the long-dated bill nobody priced

PPPs are marketed as innovative financing that keeps debt off government books while delivering infrastructure quickly. In practice, PPPs often transfer commercial risk from private partners back to government through contract structures that embed implicit fiscal commitments.

The most common high-risk clauses include: minimum revenue guarantees that compensate concessionaires if traffic or demand falls below projections, exchange rate protections that reimburse currency losses on dollar-denominated financing, availability payments that reward asset delivery regardless of usage, and termination provisions that require government to compensate investors at multiples of equity if projects are cancelled.

Chile's toll road PPPs demonstrate how seemingly sophisticated contracts create hidden fiscal exposure. The government granted concessions for major highways with minimum revenue guarantees indexed to traffic forecasts that assumed 4-5% annual growth. When actual traffic grew only 1-2%, the treasury made regular compensation payments. By 2020, accumulated payments exceeded $3bn. These were not infrastructure investments. They were revenue transfers to investors who had been guaranteed returns regardless of project performance.

Colombia's hospital PPP programme similarly embedded minimum bed-occupancy guarantees. When actual utilization ran 30-40% below projections due to insurance coverage gaps and competing public facilities, the government paid compensation totalling $800m annually. The hospitals exist. Patients could use them if insurance covered private facilities. Instead, the treasury pays for empty beds while public hospitals remain overcrowded.

India's power sector PPPs include capacity payment mechanisms guaranteeing returns to independent power producers regardless of whether electricity is actually dispatched. Distribution companies facing overcapacity and weak demand still pay approximately $4-5bn annually in capacity charges for power they don't need. When state distribution companies cannot pay, state governments absorb losses. When state governments cannot pay, the central government intervenes. The "private" investment becomes public liability through contractual architecture.

Quasi-fiscal operations: monetary policy as hidden subsidy

Quasi-fiscal operations represent perhaps the least transparent category of hidden sovereign exposure. These are activities conducted by central banks, development banks, or other public financial institutions that have fiscal consequences but bypass normal budget processes and oversight.

Common mechanisms include: central bank directed lending to specific sectors at below-market rates, preferential foreign exchange allocation creating implicit subsidies through exchange rate differentials, recapitalization needs for state banks carrying directed loans that will never be repaid, and development bank lending for politically mandated projects without commercial discipline.

Lebanon's Banque du Liban provides the catastrophic endpoint of quasi-fiscal operations. Between 2016-2020, BDL implemented financial engineering schemes (primarily Circular 158 and subsequent variants) that offered commercial banks extraordinarily high returns on dollar deposits—15-20% when global rates were near zero—to attract capital inflows and defend the currency peg. These were not monetary operations. They were Ponzi schemes run through the central bank.

When capital inflows stopped in 2019, the scheme collapsed. BDL accumulated losses estimated at $60-70bn against reserves of only $30bn, creating negative net worth exceeding 200% of GDP. The Lebanese pound, pegged at 1,507 per dollar since 1997, collapsed to 89,000 on parallel markets by 2024, a depreciation of 5,800%. The banking sector became insolvent. Deposits were frozen. The economy contracted 40%. Lebanon experienced hyperinflation reaching 200%+ in 2022. This was a central bank balance sheet that destroyed the entire financial system.

Argentina's Central Bank (BCRA) demonstrates how chronic quasi-fiscal deficits prevent stabilization. The BCRA routinely finances government deficits through money creation when bond markets close, operates FX intervention schemes that guarantee losses when reserves are insufficient, and provides subsidized credit to strategic sectors at negative real rates. The resulting quasi-fiscal deficit runs 2-4% of GDP annually. Combined with fiscal deficits of 3-5% of GDP, Argentina faces total public sector deficits of 6-9% of GDP that must be monetized, producing structural inflation of 100%+ annually that no stabilization programme can overcome without addressing central bank losses.

Egypt's Central Bank accumulated significant losses through maintaining an overvalued exchange rate using borrowed reserves. When the pound was devalued from 8.8 to 30.9 per dollar between 2015-2024, the local-currency value of dollar liabilities exploded while dollar assets remained fixed. The resulting balance sheet stress required fiscal transfers and complicated IMF programme negotiations because programme success depended on exchange rate flexibility that would crystalize additional central bank losses.

Case study: when hidden becomes visible

Sri Lanka's 2022 debt crisis illustrates how off-book liabilities transform from accounting categories into economic reality during stress. The country's pre-default situation looked manageable based on headline figures: public debt of 104% of GDP, deficit of 11% of GDP, reserves of $2.3bn. Markets and multilaterals had seen worse and survived.

What headline numbers missed was the hidden balance sheet: Ceylon Electricity Board carrying losses of 1.5-2.0% of GDP annually from below-cost tariffs, Ceylon Petroleum Corporation accumulating arrears to international suppliers exceeding $1bn, SriLankan Airlines requiring annual bailouts of $200-300m, government guarantees on SOE debt totalling 8-10% of GDP, and accumulating arrears to suppliers, contractors and pensioners approaching 3-4% of GDP that were not classified as "debt" but represented forced borrowing from domestic creditors.

When foreign reserves depleted in 2021-2022, this hidden balance sheet crystallized simultaneously: SOEs could not service dollar debt, forcing guarantee calls. Import compression created fuel and medicine shortages. Arrears triggered supplier boycotts. The electricity utility implemented 10-13 hour daily blackouts. Public sector salaries were delayed. The accumulated pressure made default inevitable regardless of IMF support because the true fiscal deficit including hidden operations approached 18-20% of GDP, far beyond any plausible adjustment capacity.

The lesson: headline debt of 104% GDP was actually 120-125% including hidden exposures. The reported fiscal deficit of 11% GDP was actually 18-20% including quasi-fiscal operations and accumulating arrears. Markets discovered this reality only when reserves exhausted and all commitments came due simultaneously. By then, restructuring was the only option.

Diagnostic framework: detecting hidden balance sheet risk

To assess which countries face highest probability of hidden liability recognition crises in 2026, six variables provide reliable early warning. Countries failing multiple criteria simultaneously face elevated risk of sudden fiscal deterioration as off-book exposures migrate onto sovereign balance sheets.

Applying this framework to major emerging and frontier markets reveals clear clustering. High-risk countries include Pakistan (6/6 red), Egypt (5/6 red), Kenya (4/6 red), Nigeria (4/6 red), and Ghana (4/6 red). These are precisely the countries where recent debt crises have featured sudden recognition of previously hidden exposures. Moderate-risk countries include South Africa (3/6 red, primarily Eskom), Bangladesh (2/6 yellow), and Indonesia (1/6 red). Low-risk countries demonstrating good practice include Chile (0/6 red), Vietnam (1/6 yellow), and Thailand (1/6 yellow).

The pattern suggests transparency itself functions as risk management. Countries maintaining comprehensive fiscal risk statements, timely SOE reporting, and regular guarantee disclosures force political systems to acknowledge problems earlier when corrective action remains feasible. Countries maintaining opacity delay recognition until crisis forces revelation, by which point adjustment options have narrowed dramatically.

What triggers recognition

Hidden liabilities rarely surface randomly. They crystallize around predictable catalysts that force governments to acknowledge what markets already suspected.

Refinancing walls: When large sovereign maturities come due, governments negotiate with creditors using full disclosure of obligations. Off-book commitments emerge during due diligence, often discovering previously unknown guarantee stocks or SOE obligations that must be included in debt sustainability analyses.

Foreign exchange crises: Currency depreciation exposes dollar-denominated off-book liabilities that were tolerable in local terms but become unsustainable in hard currency. SOE FX borrowing, guaranteed project finance, and import arrears all explode in local-currency terms during devaluation, forcing recognition and assumption.

Utility service collapse: When power shortages, fuel queues, or water rationing become politically unbearable, governments inject emergency financing that brings accumulated losses onto budget. The bailout makes previously hidden deficits visible.

IMF programme negotiations: Fund programmes require comprehensive accounting of fiscal risks including SOE finances, guarantee stocks, arrears, and quasi-fiscal operations. Countries discover during negotiations that headline deficits understate true fiscal gaps by 5-10 percentage points.

Supplier boycotts: When accumulated arrears reach levels where fuel importers, medicine suppliers, or contractors refuse further delivery, governments must clear backlogs or face service collapse. Arrears that were "not debt" become emergency expenditure requiring immediate financing.

"The debt is not discovered. It is admitted. Usually under duress."

Reform pathways: how to make hidden liabilities visible safely

Several countries have successfully brought hidden exposures onto books without triggering crises through careful sequencing and transparency.

Chile's fiscal responsibility law (2006) requires comprehensive annual fiscal risk statements quantifying SOE obligations, guarantee stocks, PPP commitments, and contingent liabilities. This transparency allowed gradual absorption of previously hidden risks while maintaining market confidence through demonstrable awareness and planning.

Colombia's fiscal rule framework includes explicit caps on guarantee issuance as percentage of GDP and requires quarterly reporting on realization rates. This prevents accumulation of unsustainable contingent positions before they become crises.

South Africa's SOE reform proposals (partially implemented) envision corporatizing Eskom's transmission assets, writing down uneconomic debt, and implementing cost-reflective tariffs over 5 years. The approach acknowledges that hidden losses must be recognized and managed, not indefinitely deferred.

Indonesia's state enterprise revitalization programme systematically audited all SOEs in 2018-2020, reclassified non-viable entities, and either liquidated or merged 48 zombie companies carrying accumulated losses. This prevented future bailout demands by addressing insolvency early.

The common thread across successful approaches is political willingness to acknowledge problems before markets force recognition. Controlled disclosure on government's timeline provides time to build buffers, negotiate with creditors, and implement reforms. Crisis-forced disclosure leaves no options except emergency measures.

The balance sheet beyond the budget

Policy debates about sovereign sustainability obsess over budget deficits, primary balances, and debt-to-GDP ratios—all legitimate concerns, all insufficient. The real determinant of solvency in 2026 will be whether states can quantify, acknowledge, and manage the obligations that sit beyond the budget: in state firms, guarantees, subsidies, quasi-fiscal schemes, and accumulated arrears.

This is not primarily an accounting question, though better accounting helps. It is fundamentally a question of institutional honesty and state capacity. A government that cannot consolidate its own public sector accounts, that operates multiple parallel budgets through SOEs and central banks, and that systematically classifies liabilities as "contingent" until crisis makes them certain, cannot persuade creditors it understands its own risk profile. Markets price that opacity as risk itself, regardless of headline figures.

By 2026, the gap between headline debt and true sovereign exposure will determine which countries face manageable adjustment versus which face restructuring. Pakistan's 74% official debt conceals 15-18 points of additional SOE and guarantee exposure. Egypt's 97% debt likely exceeds 115-120% including off-book commitments. South Africa's 71% debt approaches 90-95% when Eskom and other SOE liabilities are included. Sri Lanka's 104% pre-default debt was actually 120-125%.

Countries maintaining transparency, implementing fiscal risk frameworks, and pricing hidden exposures into their debt sustainability analyses can manage these risks gradually. Countries maintaining opacity will discover true exposure only when crisis forces revelation simultaneously, at the moment when absorptive capacity is lowest and financing costs highest.

The hidden balance sheet is not hidden from markets or multilaterals. It is hidden from political systems unwilling to acknowledge uncomfortable realities. In 2026, those realities will demand acknowledgment regardless of political preference. The only question is whether recognition occurs on governments' terms through transparent disclosure and gradual adjustment, or on markets' terms through sudden repricing and forced restructuring.

Solvency ultimately depends not on what states formally classify as debt, but on what they will inevitably be forced to pay. The most expensive opacity is the kind that prevents governments from understanding their own balance sheets until the bill arrives.