by Lars Jonung
Sweden's fix of its banks in the early 1990s is considered a model for today’s policymakers. This column reviews the main features of the Swedish approach and discusses its applicability to today’s banking problems. Policy must be carried out swiftly and openly, aiming at saving banks, not their owners or managers.
The Swedish banking crisis was part of a major financial crisis that hit the Swedish economy in 1991-93. Its origin should be traced to financial liberalisation in the mid-1980s that triggered a rapid lending boom. The pegged exchange rate for the krona prevented monetary policy from mitigating the boom by means of interest rate increases. The boom turned into bust and crisis around 1990, threatening a meltdown of the banking sector. The response of policymakers developed into the Swedish model for bank resolution. It comprises the following seven key features.
The Swedish model for bank resolutionPolitical unity
A central feature was the political unity across party lines which underlay the bank resolution policy from the very start. The Centre-Right government and the political opposition - the Social Democrats - joined forces and avoided making the banking crisis into a partisan political issue. This unity, initially forged by the determination of the major political parties to defend the pegged exchange rate of the krona, lasted throughout the crisis. Political unity guaranteed the passage through the Swedish parliament, the Riksdag, of measures to support the financial system. It also created policy trust amongst voters.
Blanket guarantee of bank deposits and liabilities
The Swedish government, in cooperation with the opposition, announced in a press release on 24 September 1992 – a critical month when currency pegs in several European countries were successfully attacked – that depositors and other counterparties of Swedish commercial banks were to be fully protected from any future losses on their claims. The guarantee was successful in the sense that foreign confidence in the solvency of the Swedish commercial banks remained intact.
In addition, this measure proved highly beneficial, as it expanded the options for the Riksbank to support commercial banks regardless of their financial position. The government guarantee of bank liabilities gave the Riksbank the option to lend to any commercial bank operating in Sweden, even to those that were on the brink of insolvency.
Swift policy action
Once it was fully understood that a serious financial crisis was in the making, the government, the Riksdag, and the Riksbank responded by taking decisive steps to support the financial system, in particular to help banks in distress. In this way, the confidence of depositors and counterparties in the financial system was strengthened at an early stage of the financial crisis by swift and determined action. Throughout the resolution of the crisis, confidence could then be maintained at a relatively low political cost.
An adequate legal framework based on open-ended funding
In December 1992, the Riksdag, passed legislation by an overwhelming majority to establish a Bank Support Authority, Bankstödsnämnd, as envisaged in the press release of 24 September 1992. The parliament approved open-ended funding for the Bankstödsnämnd, rather than settling for a predetermined fixed budget. This was a deliberate choice in order to avoid the risk that the Bankstödsnämnd would be forced to go back to the Riksdag to ask for additional funding at a later stage. The open-ended funding underpinned the credibility of the bank resolution policy.
The Bankstödsnämnd was set up as an independent agency at arm's length distance from the government, the Riksbank, and the Finansinspektion (the financial supervisory authority) to underline its independence. This construction fostered credibility and trust in its operations. The opposition was given full insight into its activities. The agency was staffed by professionals, and it began operation in the spring of 1993, shortly after being established.
Full information disclosure
Banks that turned to the Bankstödsnämnd with requests for support were obliged to give disclosure of all their financial positions, opening their books completely to scrutiny. This requirement facilitated the resolution policy and made it acceptable in the eyes of the public.
Differentiated resolution policy to maintain the banking system and prevent moral hazard
Banks that turned to the Bankstödsnämnd were dealt with in a way that minimised moral hazard. The policy priority was to save the banks, not the owners of banks or the bank managers. Banks in trouble were first asked to obtain capital from their shareholders. If they were not able to do so, present owners would have to surrender control and ownership before public support was given. Faced with this threat, private banks in Sweden made great efforts to raise capital from their owners. One bank, the SEB, decided to withdraw its request for government support. Banks that were in temporary difficulties could ask for government guarantees.
Out of six major banks, two were not expected to be profitable in the long run. They were taken over by the government with the aim of being re-privatised. Their assets were split into a good bank and a bad bank, the “toxic” assets of the latter being dealt with by asset-management companies set up by the Bankstödsnämnd, which focused solely on the task of disposing of them. When transferring assets from the banks to the asset-management companies, the government applied cautious market values, thus putting a floor under the valuation of such assets, mostly real estate. This restored demand and liquidity, and thus put a break on falling asset prices. The "good" assets of the two failed banks were transferred to a new bank that eventually emerged as Nordea, now one of the major Nordic banks.
The role of macroeconomic policies in ending the crisis
The bank resolution was facilitated by monetary and fiscal policy. The fall of the pegged exchange rate of the krona on November 19, 1992, following heavy speculative attacks, was an important move towards recovery. Once the krona was floating, it depreciated sharply, encouraging a rapid growth in exports. The ensuing fall in interest rates eased the pressure on the banking system. In July 1996, the crisis legislation and the blanket guarantee were abolished. The government allowed huge budget deficits to build up during the crisis, mainly as a result of the workings of automatic stabilisers, peaking at around 12% of GDP in 1993.
A successful bank resolutionThe Swedish bank resolution was successful. Sweden's banking system remained intact. It continued to function with no bank runs and hardly any signs of a credit crunch. It remained largely privately owned and became profitable shortly after the crisis.
In the long run, the net fiscal cost - the 'cost to the taxpayer' - turned out to be very low, close to zero. Figure 1, displaying the net fiscal costs for 39 systemic banking crises in 1970-2007, demonstrates Sweden's favourable ranking with a net fiscal cost of almost zero. The gross fiscal cost for the bank support policy amounted to around 4% of GDP initially.
Figure 1. Net fiscal costs from systemic banking crises, 1970-2007, per cent of GDP.
Source: Jonung (2009). click to enlarge.
Answering this question requires comparing the Swedish crisis of the early 1990s with the present global crisis. On a very general level, the similarities are striking. The two crises are both financial crises driven by identical forces – a boom fuelled by lax monetary policy and negligent financial oversight, later turning into a bust.
On the other hand, there are considerable differences. The Swedish crisis of the early 1990s was primarily a local phenomenon, or – more accurately - a Nordic one, as Finland and Norway also went into crisis at roughly the same time as Sweden. Being a small open economy, Sweden was able to abandon its pegged rate and obtain a significant and lasting depreciation of its currency that contributed to strong recovery. This option is hardly open to an individual country today because the present crisis is global.
The small size of the Swedish financial system in the 1990s facilitated the bank resolution policy. Policy-makers dealt with a limited number of banks - only six major banks - in an overall environment of trust in the banking system. This is in sharp contrast to the current situation in the US, for example, with thousands of banks of different types and many non-bank financial actors, and where public trust in the financial system and its actors (“Wall Street”) is extremely low.
The Swedish bank resolution policy was faced with a financial system that was much less sophisticated and much less globalised than the financial systems of today. There were no structured products, no sophisticated derivatives, hardly any hedge funds, less securitisation, and so on. Indeed, the ongoing crisis has been difficult for the authorities to manage, in part, because some traditional central banking tools – especially in the UK and the US – are not well suited, either legally or architecturally, to provide liquidity for the institutions most in need, including investment banks and insurance companies.
In addition, Sweden has a tradition of substantial public confidence in its domestic institutions, political system, and elected representatives. Such social capital made it easy for the government and opposition to reach swift and stable agreements on policy actions.
In spite of the differences between Sweden in the early 1990s and the world today, the Swedish experience holds lessons. A policy to support the financial system benefits from political unity and from being carried out swiftly and openly. The aim should be to save banks in distress, not their owners or managers. This minimizes moral hazard. The resolution policy should be implemented within a consistent and all-encompassing strategy, having a legal framework in which the administration of the support is left to experts acting at arm's length from the government, the central bank, and the financial supervisory authority. The support benefits from being financially open-ended to ensure the solvency of the financial system. The support should also be designed in a way that the public perceives as fair and just.
The Swedish case illustrates that the task of government during a financial crisis, or - in popular terms - the task of the taxpayer, is to serve as the capitalist or investor of last resort by recapitalising the financial system, thus dampening the effects of the financial breakdown on the real economy.
The Swedish formula cannot be fully imported by other countries due to institutional differences. Still, its guiding principles are applicable outside Sweden today, most prominently in four areas. First, the Swedish experience demonstrates that the threat of public receivership or nationalisation should be a real one as it forces the private sector to find private solutions. Second, the Swedish record suggests that banks in distress, nationalised as well as in private hands, should be split into a good and a bad bank, in order to get the financial system swiftly working again – more precisely, bad assets should be taken off the balance sheets of banks to prevent them from becoming "zombie" banks. Third, the bank resolution policy credibility is significantly enhanced by an open-ended financial commitment by the government. Fourth and finally, policy action should be swift and decisive to arrest the negative feedback loops arising during a financial crisis.
To sum up, the Swedish model of bank resolution should be used as a general template for countries facing financial crisis, but these countries will need to adapt the details of the implementation to their own circumstances.
Jonung, L., (2009) "The Swedish model for resolving the banking crisis of 1991-93. Seven reasons why it was successful." European Economy, Economic papers 360, European Commission, February 2009, Brussels.
