Who Paid the Transition Bill? Bank Rescues, State Assumption of Debt, Guarantees, Subsidies, Bankruptcies and the Socialization of Private Losses from Slovenia to Kosovo

First: what does “socialization of private losses” mean?

The phrase is attractive, but easy to misuse. In its most precise form it means: a loss that would normally be borne privately is transferred, through the state budget, a public bank, a state guarantee, a tax write-off or another public mechanism, to the wider community. But not every public rescue measure is automatically that. If the state reimburses insured household deposits, finances severance payments, prevents an immediate collapse of the payment system and remediates environmental damage that would otherwise remain untreated, the intervention may primarily protect: third parties or the public system, rather than the owner.

This article therefore uses three distinct categories. A — protection of the owner's capital. The owner preserves ownership or a meaningful economic benefit while the public sector absorbs part of the losses. B — protection of the system, creditors or the population. Owners are wiped out or severely diluted while the state protects deposits, payment infrastructure, employees and essential public services. C — mixed case. The state absorbs part of the private losses, while meaningful losses are also borne by owners, subordinated creditors, banks, employees and suppliers. Most real-world cases belong to: C.

What exactly counts as the “bill”? This article distinguishes at least seven channels. Recapitalization. The state injects new capital into a bank or company. This is a direct public outlay or acquisition of a public financial asset. Transfer of bad assets. Non-performing loans are moved to a bad bank, a rehabilitation agency and a state-owned asset-management company. The bill is not automatically equal to the gross face value of the loans. The crucial figures are: transfer price and later recovery. State guarantee.

The state does not necessarily pay merely because a guarantee is issued. The fiscal cost materializes when: the guarantee is called. This article therefore distinguishes between guarantee issued, guarantee activated and actual payment from the budget. Tax and social-contribution arrears or write-offs. If a state-owned or private enterprise does not pay for years taxes, health contributions and pension contributions, it receives a form of: implicit or quasi-fiscal financing. Subsidy. The state may cover operating losses, price differentials, electricity costs, labour costs and short-term liquidity. A subsidy may preserve employment, a public service and/or an economically unsustainable business model.

Bankruptcy. Bankruptcy can distribute losses among owners, secured creditors, unsecured creditors, workers, the state and suppliers. Bankruptcy therefore does not mean: the debt disappears. It is: a legal ordering of who absorbs the loss first and by how much. Social cost. Not everything appears in the budget. The closure of a bank or factory may produce job losses, unpaid wages, missing pension contributions, local economic decline and migration. This article does not convert all of this into one invented monetary figure. It treats it as: a separate category of cost.

SLOVENIA — THE MOST MEASURABLE BANKING BILL

Slovenia's post-2008 banking crisis is one of the most transparently documented public-cost cases in the series. In 2013 and 2014 the Republic of Slovenia recapitalized six banks NLB, NKBM, Factor banka, Probanka, Abanka and Banka Celje. for a total of: EUR 3.647 billion.[1] At the same time, risky exposures with a gross value of roughly: EUR 4.9 billion were transferred from four banks to BAMC at a transfer value of about: EUR 1.6 billion.[1] Important: EUR 4.9 billion is not an additional EUR 4.9 billion of public loss.

This distinction is fundamental. EUR 4.9 billion represented: the gross exposure of transferred assets. Their transfer value was approximately: EUR 1.6 billion. BAMC subsequently managed the assets, sold them and collected claims. It would therefore be wrong to calculate: 3.647 + 4.9 = EUR 8.547 billion of pure taxpayer loss. That is not a valid accounting operation. Were private bank owners simply rescued? Not entirely. Slovenia's Ministry of Finance later explicitly stated that the burden of the 2013–2014 rescue was also shared by:

• former shareholders;
• holders of subordinated or otherwise qualifying bank liabilities.[2]

This matters. The public sector supplied the capital needed to stabilize the system, but old equity and part of the subordinated capital: were not simply preserved intact. So who paid? The state. Recapitalizations and the financial risk associated with BAMC. Shareholders and subordinated investors. Loss or cancellation of part of their rights. Banks and companies. Restructuring, asset sales and consolidation.

Employees. Workforce reductions and branch closures during subsequent restructuring. Was this a rescue of private losses? Partly. The banking crisis included bad private and corporate lending, high-risk credit expansion, leveraged ownership takeovers and falling collateral values. But the intervention was not only: a rescue of individual debtors. It was also intended to prevent systemic bank failure, protect deposits, restore capital adequacy and restore Slovenia's access to financial markets. The most precise formulation is therefore: socialization of a significant part of banking losses while part of the burden was also imposed on private capital.

CROATIA — BANKS AND SHIPYARDS: TWO DIFFERENT PUBLIC-COST MODELS

Croatia provides two very different channels:

• bank rehabilitation in the 1990s;
• decades of state support for shipbuilding.

Bank rehabilitation: the state takes over banks, but old shareholders can be wiped out. For the Croatian banking system in 1995–1998, the IMF described a mechanism involving write-off of bad loans, cancellation or temporary elimination of existing equity, government bonds to cover remaining losses, transfer of bad claims to a rehabilitation agency and later privatization of the banks[3]. At Riječka banka and Splitska banka: the entire existing equity was eliminated, and the state rehabilitation agency became the owner.[3] At Privredna banka: almost all old equity was written off, and after recapitalization the agency held about 90% of the bank.[3]

What was the directly documented public instrument? In 1996, about: HRK 6.1 billion of enterprise loans connected with Riječka, Splitska and Privredna banka were written off. Roughly: HRK 2.9 billion in government bonds were issued to support the banks, equal to about: 3.7% of Croatian GDP in 1996.[4] For Dubrovačka banka in 1998 the state additionally provided:

• about HRK 500 million of temporary liquidity;
• then about HRK 1 billion in rehabilitation bonds.[4]

Was this a rescue of old owners? For several major banks: no. Their equity was wiped out. The state primarily absorbed problematic parts of the balance sheet, protected the financial system and prepared the banks for later privatization. This is very different from: the state pays while the old shareholder keeps everything. Shipyards: long-term industrial policy becomes a large public bill.

In 2019 the Croatian government stated that between: 1992 and 2017 approximately: HRK 31.7 billion had been spent on rehabilitation or state subsidies for shipyards.[5] Of this, Uljanik and 3. Maj together accounted for about: HRK 13.3 billion.[5] This is a very different type of support from bank rehabilitation. Uljanik: private company, state guarantees.

In August 2018 the Croatian government itself stressed: Uljanik is a private company and that primary responsibility for its operations lies with management.[6] Yet the state had issued guarantees. From 2010 to October 2018 roughly: HRK 7.5 billion in state guarantees had been issued for the Uljanik group.[7] Later, because guarantees were called, the state actually paid several billion kuna. By March 2019 the finance minister cited roughly: HRK 3.1 billion in already paid guarantee calls.[8] This is a relatively clear case of risk transfer.

Ownership was private, while vessel financing was partly protected by: state guarantees. When the debtor could not meet its obligations, part of the financial burden moved: to the state budget. This is close to the classic meaning of: socialization of private credit risk. But the state did not pay only “the owners”. Guarantee payments primarily protected or repaid banks, financiers and contractual obligations of the shipyard. At the same time bankruptcy imposed losses on owners, workers and the local economy. Uljanik is therefore not a story in which: the private owner gained everything and the state lost everything. It is a story of: a private business model with a large public guarantee tail.

SERBIA — CLOSING BANKS INSTEAD OF RESCUING THEIR OWNERS

After 2000 Serbia inherited a deeply insolvent banking system. Four large banks Beobanka, Beogradska banka, Investbanka and Jugobanka Beograd, were placed into bankruptcy in January 2002.[9] Together they represented approximately: half of the banking system's assets.[10] Why did the state not simply rescue them?

The estimated rehabilitation cost was approximately: USD 3.8 billion in addition to already state-guaranteed obligations connected with the Paris Club, the London Club and frozen foreign-currency household deposits[10]. The chosen strategy was therefore: liquidation. This is an important contrast with Slovenia. But liquidation is not free. The IMF estimated the direct additional budgetary cost of banking restructuring at approximately: DEM 324 million in present value, or roughly: 1.2% of GDP in 2002.[9] This included approximately:

• DEM 271 million for liquidation costs, deposit payouts and severance;
• DEM 53 million for rehabilitation of three smaller banks.[9]

Who was protected? Individual depositors. They received payments or bonds according to the programme rules.[9] Employees. Around 8,000–9,000 workers in the large banks were covered by severance and social-support programmes.[10] Companies and other creditors. They were not all repaid in full. The IMF explicitly stated that companies would generally receive: only part of their deposits, while some categories of financial creditors would effectively receive no special compensation.[9] So who was not rescued?

The banks themselves: ceased to exist. This is not a case in which: the state recapitalizes a private bank and preserves the shareholder. It is: the state finances the orderly dismantling of an insolvent system and protects selected social groups. But there was an even larger legacy debt. The IMF also noted that, in addition to the new liquidation costs, the banking system carried older state-guaranteed liabilities, particularly those connected with the London Club, the Paris Club and frozen foreign-currency deposits. which were enormous relative to the economy at the time.[11] It would be misleading to attribute these liabilities solely to post-2001 privatization. A large part was: a legacy of the previous system, sanctions and the 1990s.

BOSNIA AND HERZEGOVINA — THE BILL OFTEN HIDES IN UNPAID TAXES AND PUBLIC-ENTERPRISE RECEIVABLES

Bosnia's problem is less concentrated in a single large: “bank rescue.” A large share of fiscal risk is spread across state-owned enterprises, entities, cantons, municipalities and pension and health funds. Aggregate: public enterprises with debt of about one quarter of GDP. An IMF analysis of state-owned enterprises estimated that in 2017: total public-enterprise debt was around EUR 4 billion or approximately: 26% of GDP.[12] Almost half of public enterprises were illiquid.[12] About: EUR 0.6 billion consisted of arrears in taxes and social contributions. equivalent to around: 4% of GDP.[12] Why are unpaid contributions a public cost?

Because the enterprise receives: liquidity today, while the pension fund, the health fund and the budget. do not receive the revenue they are owed. That is: quasi-fiscal financing. It may not appear as a conventional subsidy, but it has a real public effect. Direct support. For 2017 the IMF estimated that governments in BiH provided public enterprises with approximately: EUR 109 million or: 0.9% of GDP through transfers, subsidies and grants[12]. This is in addition to implicit support through tax and contribution arrears. Aluminij Mostar: electricity as a hidden financing channel. By the end of 2017 Aluminij Mostar had accumulated:

• more than EUR 120 million in cumulative losses;
• about EUR 188 million in liabilities.[13]

Roughly three quarters of the debt was owed to: Elektroprivreda HZHB, a publicly owned power company.[13] The IMF explicitly described this as: implicit government support.[13] Who bore the bill here? Not only the budget. Part of the burden was shifted to: another public enterprise. This is an important mechanism. A state can conceal a loss when one public enterprise does not pay and another public enterprise does not collect. Formally it is not always: a budget subsidy. Economically it is: a transfer of loss within the public sector.

Is this a direct privatization bill? Not necessarily. This is a key safeguard. Much of Bosnia's problem reflects the legacy of war, incomplete restructuring, continued public ownership and politically fragmented governance. This article therefore does not classify all public-enterprise debt as: “the cost of privatization.” More precisely: it is a cost of the broader post-Yugoslav transition and unfinished restructuring.

MONTENEGRO — KAP: A TEXTBOOK EXAMPLE OF A CALLED STATE GUARANTEE

KAP already showed in Sold Too Cheaply? Anatomy of the Most Controversial Privatizations of Former Yugoslavia: Price, Value, Debt, Land and Final Owners how public risk can reappear after privatization. In 2005 approximately: 65.44% of KAP was sold for: EUR 48.5 million. The state later guaranteed a large share of the company's new obligations. In 2012 the IMF reported:

• approximately EUR 350 million of total debt;
• around EUR 132 million of debt covered by state guarantees.[14]

The guarantee becomes a real budget bill. In 2013 Montenegro paid approximately: EUR 102.5 million because KAP guarantees were called.[15] The Montenegrin government itself described this as: an unforeseen expenditure.[15] That is more than twice: the original purchase price for the majority stake in KAP. Is this the cleanest this article case? It is one of the clearest. The sequence is straightforward:

1. the company becomes majority privately owned;
2. the state later guarantees debt;
3. the company cannot service the obligations;
4. the guarantee is called;
5. the budget pays.

This is a classic example of: private credit risk being transferred to the state. But precision is still required. KAP was not merely a private financial speculation. It was one of the country's largest exporters, a major employer and strategically important to the economy. The state therefore issued guarantees partly because of employment, exports and industrial policy. That helps explain the decision. It does not change the fact that: when the risk materialized, the budget paid a large part of it.

NORTH MACEDONIA — STOPANSKA BANKA: PUBLIC BALANCE-SHEET CLEANUP FIRST, PRIVATE SALE SECOND

Stopanska Banka is a textbook case of: rehabilitate → privatize. Before the sale, the state issued euro-denominated bonds in exchange for bad claims on the bank's four largest debtors. Nominal amount: DEM 235 million.[16] The bad assets were transferred to the: Bank Rehabilitation Agency.[16] The equity was then sold. For approximately: 85% of the ordinary share capital the agreed price was: DEM 94 million.[16] The government estimated that it would receive about: DEM 60 million from its own stake and premiums.[16] Those proceeds were intended to help service the new government obligations.[16] Who paid? The state.

It assumed problem claims and exchanged them for government bonds. Old owners. Their economic position was diluted or impaired through rehabilitation and restructuring. New investors. They paid for the cleaned-up bank and committed additional capital. Is this “private profit, public loss”? Partly, but the slogan is too crude. The bank before rehabilitation: was not the same economic asset as the bank after rehabilitation. The public sector absorbed historic losses, while the new owner paid for: a cleaned-up, better-capitalized institution. This is precisely the kind of transaction in which we must analyze separately the cost of the old crisis and the value of the new equity.

KOSOVO — A DIFFERENT MODEL: PRIVATIZATION FUNDS, THE WORKERS' 20% SHARE AND PUBLIC ENERGY RISK

Kosovo did not undergo a large post-socialist banking rescue comparable to Slovenia or Croatia. Its transition was more closely tied to liquidation of socially owned enterprises, sale of assets, distribution of sale proceeds to workers and creditors and later fiscal risks from public enterprises. Privatization proceeds do not simply go into the state budget. As of September 2026, the Privatization Agency of Kosovo reported approximately:

• 2,323 assets sold;
• around EUR 822.4 million in cumulative sales value;
• about EUR 171.9 million distributed under the workers' 20% share.[17]

PAK was still in 2025:

• confirming final worker lists;
• distributing workers' 20% shares;
• paying creditors in liquidation proceedings.[18][19]

This is an important contrast. Part of the privatization proceeds was not: unrestricted fiscal revenue for the government. It went to eligible workers, creditors and liquidation funds. The sale of socially owned property therefore also created: an obligation to distribute assets to former stakeholders. Kosovo later acquires a different type of public bill: energy. In 2021–2022, because of high electricity-import prices, the government allocated substantial transfers to cushion the energy shock. The World Bank cites:

• about EUR 20 million in support for electricity imports in 2021;
• about EUR 94 million in transfers or subsidies in 2022;

for roughly: EUR 114 million across the two years.[20] Is this a cost of transition privatization? No. And that is exactly why the example matters. It is: a fiscal risk from the public energy system and an external price shock. This article includes it to mark an analytical boundary: not every public bill in a post-Yugoslav economy is caused by privatization.

Kosovo now treats fiscal risk systematically. In 2025 the IMF noted that Kosovo's fiscal-risk framework explicitly monitors state guarantees, public enterprises, PPPs, litigation and the financial sector[21]. That is effectively an institutional response to the central question of this article: where can corporate or private risk turn into a future public liability?

WHO ACTUALLY PAID? A COMPARATIVE MATRIX

This table is: not a ranking of fiscal irresponsibility. The amounts are not directly comparable because they involve different years, different currencies, different sizes of economy and different definitions. Its purpose is to show: different channels through which losses were transferred. THE BIGGEST MISCONCEPTION: EVERY RECAPITALIZATION IS A GIFT TO OLD OWNERS. No. The Croatian case shows that in some rescued banks: existing ownership was wiped out.[3] In Slovenia, the burden was also shared by:

• shareholders;
• subordinated investors.[2]

In Serbia, the four largest insolvent banks were: closed.[9] Public money can therefore rescue depositors, the system and a new capital base. without necessarily rescuing: the old owners. SECOND MISCONCEPTION: A STATE GUARANTEE IS NOT A COST UNTIL IT MUST BE PAID. That is only partly true. Before activation, a guarantee is: a contingent liability. But even before payment it affects fiscal risk, the cost of borrowing, bank behaviour and company behaviour. Once activated it becomes: a real budgetary outlay. KAP and Uljanik are clear examples.[7][15] THIRD MISCONCEPTION: IF A COMPANY RECEIVED NO CASH SUBSIDY, THE STATE DID NOT SUPPORT IT.

Bosnia shows otherwise. A company can finance operations by not paying electricity bills, taxes and social contributions. This means it is effectively being financed by the public electricity company, the health fund, the pension fund and the budget. That is: an implicit subsidy. FOURTH MISCONCEPTION: BANKRUPTCY SHIFTS THE ENTIRE COST TO THE OWNER. No. In bankruptcy the owner normally loses equity first, but losses may also fall on employees, suppliers, unsecured creditors, local banks and the state through unpaid taxes. Bankruptcy is: a distribution of loss, not its disappearance. FIFTH MISCONCEPTION: PRESERVING JOBS IS FREE.

When the state subsidizes a company in order to preserve jobs, it is purchasing: time and social stability. That may make economic sense. But every year of support has: an opportunity cost. The money cannot simultaneously be used for schools, hospitals, infrastructure, tax reductions and another company. The question is always: does the public benefit justify the public cost? this article does not settle that politically. It shows: the bill. Taxpayers. Through recapitalizations, guarantees, subsidies and government bonds. Employees. Through layoffs, unpaid wages, reduced severance and lost social contributions.

Small shareholders. Through dilution, cancellation and bankruptcy. Subordinated and unsecured creditors. Through write-downs, partial repayment and long bankruptcy proceedings. Pension and health systems. Through: unpaid social contributions. This is particularly important in BiH. Public enterprises. Through: uncollected bills owed by other companies. Aluminij Mostar is a clear example.

WHO WAS MOST PROTECTED?

There is no single answer. It depends on the case. Banks. Often small depositors, the payment system and the new capital base. Industry. Often employment, exports and the local economy. Guarantees. Often the lender. because when the borrower defaults the state performs the obligation. That is why, when a guarantee is called, one must ask: who actually receives the payment? Frequently: a bank or another financier.

DID TRANSITION “PRIVATIZE PROFITS AND SOCIALIZE LOSSES”?

As a universal formula it does not fit every case. But in some documented cases the direction is clear: control and potential upside become concentrated privately, while part of the credit, guarantee or social downside is transferred, after failure, to banks, the state, employees and the local community. That has to be measured case by case, rather than dismissed in advance merely because shareholders and creditors also bore losses. The bill is not only a budget line.

When roughly 9,000 industrial jobs had disappeared in Gorenjska by 1994,[22] part of the transition bill was visible in households, emptier industrial complexes and lost local purchasing power. Even when bank stabilization was macroeconomically necessary, that does not make the earlier path to crisis less important to a worker who had already lost the company and the job. THE MOST IMPORTANT QUESTION: WHAT WAS THE ALTERNATIVE TO A RESCUE?

When we see a bill of three billion, the natural question is: why did the state pay at all? But the alternative scenario is not necessarily: cost = zero. Without intervention there may be bank failures, loss of deposits, a credit crunch, additional corporate bankruptcies, unemployment, higher social transfers and lower tax revenue. This does not mean: every rescue was justified. It means: the cost of rescue must also be compared with the probable cost of non-intervention. That counterfactual is often: the least documented part of public debate.

WHAT CAN WE STATE WITH HIGH CONFIDENCE FROM THE EVIDENCE? Slovenia recapitalized six banks with about EUR 3.647 billion in 2013–2014. EUR 4.9 billion of gross claims transferred to BAMC does not equal an additional EUR 4.9 billion of public loss. Shareholders and subordinated investors also bore part of the burden in Slovenia. Croatia wiped out old shareholder capital in the rehabilitation of some banks. Croatia spent approximately HRK 31.7 billion on shipyard support or rehabilitation during 1992–2017.

Uljanik was a private company for which the state assumed major guarantee exposure. Serbia closed four large insolvent banks in 2002 instead of fully rehabilitating them. Direct additional costs of Serbian banking restructuring were estimated at roughly DEM 324 million, or 1.2% of GDP. Not all creditors in Serbia were repaid in full. BiH public enterprises had debt of around 26% of GDP, large tax and contribution arrears, and direct subsidies.

Aluminij Mostar partly financed its business by not paying electricity bills to a public enterprise. Montenegro paid approximately EUR 102.5 million in 2013 because KAP guarantees were called. Stopanska Banka was cleaned up with about DEM 235 million in government bonds before private sale. Kosovo systematically distributed part of privatization proceeds to eligible workers as a 20% share. Kosovo spent approximately EUR 114 million in 2021–2022 to cushion the energy shock. Kosovo's energy subsidies are not proof of privatization costs.

Public expenditure is not automatically a rescue of the old shareholder. When private financial leverage becomes a public banking problem. In Slovenia, one important transmission channel can be seen quite clearly. The Bank of Slovenia later wrote that leveraged-buyout acquirers using small own-capital contributions caused large losses in the banking system.[23] The IMF estimated that at the end of 2011 financial holding companies used for highly leveraged privatization takeovers, together with construction, accounted for about 21% of all bank loans.[24]

Those figures do not allow a fair calculation of how much of the 2013 bank rescue belonged to any one takeover, to Merkur, Istrabenz, Pivovarna Laško or another group. That would require loan-by-loan and loss-by-loss tracing. What can be documented is the mechanism: privately organized ownership concentration financed with high bank leverage generated losses that later appeared on bank balance sheets; once the banking problem became systemic, the state entered the stabilization process.[1][23][24]

For a worker in a town where the factory had meanwhile reduced production or closed, this time lag matters. The job loss occurs locally and immediately; the banking loss becomes visible later, and the public rescue later still, spreading across the wider state. That is why the perception that control or upside could be concentrated while the cost of failure became dispersed is something that can be investigated through balance sheets and credit flows — not only through nostalgia or anger.

The most precise answer to the question “who paid the transition bill?”

Everyone — but not equally and not at the same time. The bill was paid by taxpayers through rescues and guarantees, shareholders through dilution and cancellation, creditors through write-offs, employees through layoffs and lost income, pension and health systems through unpaid contributions, public enterprises through unpaid receivables and future budgets through bonds and debt. The central finding is therefore not: “the state always rescued private owners.” Nor: “the market cleared all losses by itself.” The real picture is: a mixed system of loss redistribution in which the boundary between private and public risk repeatedly shifted — sometimes because of systemic necessity, sometimes because of industrial policy, sometimes because of political choice, and in some cases because of poorly designed guarantees or delayed restructuring. Conclusion of the Who Got the Socially Owned Assets? Major Privatizations, New Owners and Lost Companies from Slovenia to Kosovo–this article economic deep-dive.

Who Got the Socially Owned Assets? Major Privatizations, New Owners and Lost Companies from Slovenia to Kosovo asked: who received the socially owned assets? Sold Too Cheaply? Anatomy of the Most Controversial Privatizations of Former Yugoslavia: Price, Value, Debt, Land and Final Owners: for how much were they sold, and were they sold too cheaply? New Owners and Old Connections: Management Buyouts, Banks, Business Networks and Judicially Proven Abuse in Post-Yugoslav Privatization: who controlled transactions through banks, management buyouts and political-business networks — and what was actually proven in court? this article adds the last component: who absorbed the loss when the bill finally came due? With that, the four-part sequence: OWNERSHIP → PRICE → NETWORKS → LOSSES is substantively complete.