Sweden as a Useful Model of Successful Financial Crisis Resolution

The way Sweden handled its 1990s banking crisis has been offered as a useful case study in resolving systemic banking crises. This column discusses the merits of the Swedish experience relative to ideal resolution strategies.

Governments of the world’s rich nations are clearly in need of positive examples of how to fix broken banking sectors. While we have many such examples from the last century (Caprio, Hunter, Kaufman, and Leipziger 1998), one of the most recent is the Swedish crisis.

In the early 1990s, Sweden’s economy was nearly toppled by a banking sector swollen with bad loans from the preceding decade’s credit bubble. The Swedish government seized ownership of the largest financial institutions and publicly-capitalised asset managers were put in charge of managing the poorly performing banking assets and returning what could be salvaged back into private hands. In this column, we compare the Swedish government’s intervention to four identified principles of effective crisis resolution:

  • Transparency of asset losses up-front, and honest communication about the extent of public intervention
  • Politically and financially independent receivership
  • Maintenance of market discipline
  • Restoration of credit flows

Four principles of effective crisis resolution

A paper published by the Federal Reserve Bank of Cleveland, based on the analysis of evidence from financial crises across the globe, identified four practices common to successful financial crisis resolutions.


The most important attribute of successful crisis resolutions has been the transparency of the entire process. Triage and full public disclosure of associated losses clear the uncertainty surrounding financial institutions and make it possible for viable institutions to raise new funds from private investors or from the government if private sources are not available. Failing to acknowledge the true value of assets or the condition of troubled banks early on makes it easy for them to live on as propped-up “zombies” (as happened in Japan during the 1990s or with savings and loans in the US in the 1980s) – healthy on paper but economically insolvent.

Politically and financially independent receivership

Crisis resolutions have been most successful when they were handled by a politically and financially independent agency. Granting independence to those responsible for containing the crisis and restructuring shields decision makers from political pressures, which mount as institutions are closed and assets are liquidated. The decision to close a financial institution or a business must be economic, not political. Financial independence is necessary to give credibility to political independence – if a government agency holds the purse strings, it can dictate policy. Incidentally, a transparent process is essential to the success of the independent agency. Without transparency, investors and taxpayers cannot verify independence.

Maintenance of market discipline

A successful resolution strategy must maintain market discipline, lest it simply set the stage for future crises. Investors who assumed greater risks must be credibly exposed to loss and suffer the consequences of having ignored or failed to detect signs of trouble. Blanket guarantees of uninsured depositors and investors are an example of a policy manoeuvre that might lessen the pain of a crisis in the short run but also distort market discipline going forward.

Restoration of credit flows

Finally, a full crisis resolution must begin to restore credit flows within the economy. For that to happen, the creditworthiness of borrowers must be restored throughout the economy – a difficult task, given that the economic fallout from a crisis (such as rising unemployment) actually erodes credit quality further.

How did these principles work in Sweden?

Sweden’s crisis

Sweden’s crisis followed a massive surge in speculative real estate and consumer debt. In 1991, two of Sweden’s largest banks, Föreningsbanken and Nordbanken, fell below their required capital levels amid rising loan defaults. Afraid of a meltdown, the government guaranteed all of Nordbanken’s liabilities and took ownership of the bank, while arranging a guarantee for Förenings. When a third large bank, Gota, was taken over shortly thereafter, policymakers acted quickly to separate the good from the bad.

Government-held assets that were deemed viable were merged under one name, Nordbanken, and permitted to continue operating. Bad assets were transferred to two asset management companies – Securum for Nordbanken’s assets and Retrieva for Gota’s.

The asset management companies were charged with managing and liquidating the bad assets of these banks and taking on the assets of non-bank companies that were in default. Swedish legislators made sure that the companies were adequately capitalised and granted exemptions from regulatory rules that would have rushed their actions or limited their effectiveness, including a rule that required seized collateral to be liquidated within three years.

Often, the asset management companies became managers of otherwise private, failed companies, performing such tasks as hiring and firing, managing property, and changing operational strategies until their assets could be favourably sold. Their flexibility and financial resources shortened their own existence from an expected duration of 15 years to a few years. Liquidations were completed in 1997, and the companies’ remaining funds (less than half of their original capitalisation, in real dollars) were returned to the Swedish treasury.

Lessons learned

Sweden emerged from its credit market turmoil without the zombie banks and dismal growth that characterised Japan’s “lost decade”. This achievement is at least partly due to Sweden’s crisis containment and resolution strategies.

First, Sweden’s approach was remarkably transparent. The magnitude of losses was established by a Bank Support Authority, which was independent of the Ministry of Finance and the central bank. Good assets were separated from bad assets, and the full extent of the government’s involvement was clearly outlined. Transparency about losses from the outset likely avoided the “zombie” effect and what Douglas Diamond has called “evergreening,” a process whereby undercapitalised banks choose not to address problem loans because doing so would force asset write-downs, possibly prompting technical insolvency.

Sweden also extended considerable political and financial independence to the asset management companies, which allowed them to carry out their task with adequate resources. Doing so served as a public signal that their operations would not be subject to changing political winds. Similarly, Swedish officials’ relaxation of collateral liquidation requirements implied that the dispensation of assets would take place over an extended period of time. Arguments can be made either way about the best time to sell assets – selling early returns assets to private use and avoids investor anxiety about debt overhang, while selling gradually sidesteps a distressed-pricing feedback loop. In any case, asset managers were given flexibility to make their own decisions about the trade-off.

To restore credit flows, Sweden moved quickly to provide incentives to bank owners to inject additional capital into their banks or to inject government capital into banks directly, when necessary. Asset management companies played a key role in restoring the financial health of the non-bank companies they were operating. Some viable corporations were allowed to survive through capital injections, though in return the government acquired a majority of their shares so that taxpayers could profit from any upside. Recapitalised institutions could return to ordinary operation, gradually rebuilding the creditworthiness of the overall economy.

Sweden’s success at maintaining market discipline was perhaps more limited. Ideally, discipline is sustained by not saving undisciplined investors through issuing blanket guarantees and unlimited liquidity. In Sweden’s case, policymakers avoided the liquidity pitfall but ended up guaranteeing bank liabilities before the banks themselves were taken over. Edward Kane and Daniela Klingebiel have suggested an alternative to such incentive-skewing guarantees. They have argued that the optimal response to a systemic banking crisis is to call a bank holiday long enough for examiners to determine which banks are viable, while still giving insured depositors access to their funds. Doing so would insure business as usual for insured depositors without permitting uninsured investors to cash out before they’ve taken their share of unrecoverable losses.

Most of the criticisms that can be levelled at the Swedish crisis resolution are easy to make in hindsight. Overall, Sweden’s financial crisis containment and resolution strategy largely avoided mistakes that would skew uninsured investors’ incentives going forward. Its policies were enacted transparently, insured political independence, and attempted to restore credit flows in the broader economy. The Swedish case illustrates the trade-offs and considerations of market discipline that crisis managers must address if they are to minimise taxpayer losses and speed the return to a rebalanced, growing economy.

Disclaimer: This page contains affiliate links. If you choose to make a purchase after clicking a link, we may receive a commission at no additional cost to you. Thank you for your support!

About Kent Cherny 1 Article

Affiliation: Federal Reserve Bank of Cleveland

Kent Cherny is a research assistant in the Research Department of the Federal Reserve Bank of Cleveland. He joined the Bank in May 2008, and his work focuses primarily on financial institutions, capital markets, and financial intermediation.

Mr. Cherny holds a bachelor of arts degree in English and economics from The Ohio State University.

Visit: Federal Reserve Bank of Cleveland

Be the first to comment

Leave a Reply

Your email address will not be published.


This site uses Akismet to reduce spam. Learn how your comment data is processed.