Ireland, Iceland and Cyprus each demonstrated the risks that arise when a small economy supports a banking system that is very large relative to national income. Crisis coverage often assumes that readers already understand terms such as haircut, bailout and bail-in. This guide explains those concepts and shows how they shaped the Cyprus debate.
1. How does a bank become insolvent?
A bank owns assets and owes liabilities. Its assets usually include cash reserves, customer and business loans, and securities such as government bonds. Its liabilities include deposits and money borrowed from other creditors.
Assets
- Cash and central-bank reserves
- Loans to households and businesses
- Government and private securities
Liabilities and capital
- Customer deposits
- Secured and unsecured borrowing
- Shareholders’ capital
Banks commonly operate with much less capital relative to assets than non-financial businesses. This is leverage. Leverage can support lending and returns, but it also means that a comparatively small fall in asset values can eliminate shareholders’ capital.
Losses may arise when borrowers cannot repay loans or when securities decline in market value. During the Cyprus crisis, exposure to restructured Greek government debt contributed heavily to losses. Once the realistic value of a bank’s assets falls below its liabilities, the bank is insolvent.
Why balance sheets can conceal trouble
Insolvency is not always immediately visible. Some assets are carried at accounting values rather than prices available in a rapid sale. Regulatory calculations may also apply risk weights and definitions that differ from ordinary balance-sheet arithmetic. These differences can delay recognition of losses. The archive’s earlier discussion of risk weighting examines that problem in more detail.
2. Liquidity pressure and bank runs
Insolvency and illiquidity are related but distinct. An insolvent bank owes more than its assets are worth. An illiquid bank may own assets of adequate long-term value but lack the cash needed to meet withdrawals today.
A bank run occurs when depositors withdraw funds rapidly because they fear the bank may fail. Other creditors can create similar pressure by refusing to renew short-term lending. If cash reserves are exhausted, the bank may have to sell loans or securities quickly. Such assets can fetch fire-sale prices below their reported values, converting a liquidity problem into insolvency.
Liquidity versus solvency
Liquidity concerns the timing and availability of cash. Solvency concerns whether assets are ultimately worth enough to cover liabilities. Emergency lending can address the first problem, but cannot by itself erase losses behind the second.
3. Liquidation and creditor haircuts
Liquidation closes a failed bank, sells its remaining assets and distributes the proceeds according to legal priority. Shareholders ordinarily absorb losses first because their capital is the residual claim. Secured creditors may have claims over specified collateral, while depositors and unsecured creditors rank according to the applicable resolution rules.
A haircut is the portion of a claim that is not repaid. If an unsecured creditor receives 50 for a claim of 100, the creditor has taken a 50% haircut. The size of that loss depends on the realised value of assets and the creditor’s place in the hierarchy.
The first Cyprus proposal contemplated losses for both larger and smaller deposits. It generated strong political opposition and was rejected by the Cypriot parliament. The final approach protected insured deposits while imposing substantial losses elsewhere in the failed institutions’ capital structures. That later outcome should not be confused with the initial proposal being debated when this article was first written.
4. Tools used to restructure a bank
Authorities do not always liquidate an insolvent institution immediately, particularly when they fear that disorderly failure could disrupt payments, lending or other banks. Resolution can combine several tools:
- Emergency lending: temporary funding intended to meet cash needs. It raises assets and liabilities together, so it does not automatically restore capital.
- Asset transfer: impaired assets are separated from the continuing bank and managed or sold elsewhere.
- Capital injection: public or private funds are exchanged for shares, directly increasing bank capital.
- Creditor conversion: eligible debt is written down or converted into equity, reducing liabilities and rebuilding capital.
Loans are most useful when a fundamentally solvent institution faces temporary liquidity stress. A bank with losses greater than its capital needs recapitalisation, creditor loss absorption or a combination of both.
5. Bailout versus bail-in
Bailout
Outside resources—usually public funds or guarantees—protect some stakeholders from losses they would otherwise bear. The precise meaning varies, so readers should check whether shareholders, creditors or depositors receive protection.
Bail-in
Shareholders and eligible creditors contribute to recapitalisation through write-downs or conversion of claims into equity. Each unit of loss absorbed internally reduces the amount that must come from outside the bank.
A full nationalisation can eliminate existing shareholders before public capital is supplied. A partial recapitalisation can dilute them. Unsecured creditors may be protected, written down or converted into shareholders, depending on the legal framework and the depth of losses.
Small deposits are generally treated differently because deposit-guarantee systems protect balances up to a defined limit. Larger uninsured balances can be exposed in resolution, although the rules differ by jurisdiction and period. The Cyprus controversy was intensified by the initial attempt to include deposits that were expected to be protected.
Why unconditional rescue creates moral hazard
When investors expect public authorities to protect them from losses, they may accept risks without demanding an appropriate return. This is known as moral hazard. Repeated protection can also give very large institutions a funding advantage if creditors assume those institutions will not be allowed to fail.
6. Two principles that clarify every bank rescue
Banking losses represent real economic losses
Accounting can determine when a loss is recognised and resolution rules can determine who carries it, but neither can make a genuine loss disappear. Bad loans may reflect failed investments; impaired bonds may reflect a borrower’s inability to repay. Once value is gone, it must be allocated among stakeholders.
One insolvent sector cannot rescue another indefinitely
Shareholders can lose only the capital they invested. Creditors can absorb losses only up to the value of their claims. A government can provide support only within its fiscal and political capacity, or with external assistance. If bank losses exceed all available buffers, the burden moves through the hierarchy until it reaches parties able—or compelled—to absorb it.
Cyprus faced an especially difficult constraint because its banking losses were large relative to the resources of the state. External support therefore came with requirements about how losses would be divided. Similar constraints appeared in different forms during earlier crises in Iceland and Russia, although their institutions, legal settings and outcomes were not the same.
What the Cyprus episode teaches
The central question in bank resolution is not whether losses exist, but how they are measured, recognised and allocated. Prompt recognition can limit further deterioration. Clear priority rules can reduce uncertainty. Credible deposit protection can reduce the risk of runs, while loss absorption by shareholders and eligible creditors can limit the burden placed on taxpayers.
Readers revisiting the crisis should distinguish the proposals under debate on 22 March 2013 from the measures subsequently adopted. This article is most useful as a guide to the mechanics of bank failure and as a record of how the policy dilemma appeared during a fast-moving event.