Lessons from Swedish Bank Crisis Management

Pär Magnusson, Danske Bank, Copenhagen.
This article appeared in the October 2008 issue of Current Economics with permission of the author.

Key Concepts: Swedish Bank Crisis | Credit Guarantee | Bank Rescue |
Key Economies: Sweden |

Chronology of the Crisis

The Swedish bank crisis had its origins in the commercial real estate market, not primarily the residential real estate market, as is the case in the US today.

The first visible signs of a looming crisis were seen in the autumn of 1990 when the first so-called finance companies — who lent money to real estate investors mostly funded by issuance of commercial paper — had problems finding buyers for their debt. Moreover, these finance companies were often owned by banks.

In 1991 Första Sparbanken (First Savings and Loans, seventh-largest bank in Sweden at the time) disclosed severe credit losses in real estate lending, threatening its survival as the capital ratio fell below 8%. The Treasury gave a credit guarantee to the bank in order to improve its ability to raise funds on the market. There was still no sense of a 'systemic crisis’ among authorities.

Later in 1991 Nordbanken (which later evolved into today’s Nordea) disclosed large credit losses that took its capital ratio below 8%. The government guaranteed an equity emission of which it also bought a substantial part, taking the government’s ownership from 70% to 77%.

In 1992 both Nordbanken and Första Sparbanken once again found themselves in trouble. The government gave a new credit guarantee to Första Sparbanken, but this time it took 70% of the equity as collateral. The government also contributed to the process whereby the First Savings and Loans merged with a large group of other savings and loans to form what today is known as Swedbank.

The government also decided to buy out the remaining Nordbanken stakeholders (23%) for SEK2bn, since it believed it would be easier to restructure the company with sole ownership.

When it became clear that the credit losses were bigger than expected at Nordbanken, the government decided to create a new company, Securum, whose sole purpose was to take ownership of distressed debt. Some 25% of Nordbanken’s outstanding stock of credit was transferred to Securum.

During the summer of 1992 real estate prices declined quickly and this showed up as large credit losses on the banks’ balance sheets. In addition, the Swedish economy had gone into recession, and the final nail in the coffin was the money market turbulence that rocked the world from August 1992 onwards (remember George Soros vs. the Bank of England).

Since interest rates on deposits adjust more quickly than interest rates on outstanding loans the rate margins became negative among the banks. Moreover, the funding cost for banks on the capital markets surged, which resulted in higher lending rates for borrowers. The higher lending rates, in turn, led to loan defaults and bankruptcies on a large scale. Now it was clear that Sweden was indeed subjected to a ‘systemic crisis’.

Foreign banks cut credit lines to Swedish banks and Swedish banks had to give notice on outstanding loans to clients, which led to further defaults and bankruptcies.

On September 9, 1992 Gotabanken went bankrupt. The government immediately issued a guarantee that no counterparties to the bank would suffer losses. This entailed that it guaranteed all forms of bank debts, not just deposits (Sweden had no formal deposit insurance at the time). The guarantee was later extended to cover all banks. It did, however, make sure that stock owners should bear the cost and not be included in any guarantee.

On September 24, 1992 the government expanded the guarantee to be a general guarantee for all Swedish banks. The big question was what the limits to the guarantee should be? The answer was an unlimited guarantee in order to create the best conditions possible to rebuild confidence in the financial system.

By the turn of the year the government decided to create a new agency — the Bank Support Agency (BSA). The Agency was subordinated the Treasury Department, but any support measures should be taken in conjunction with the Riksbank, the Debt Office and the Financial Supervisory Authority.

Guiding Principles for BSA

Important principles for the BSA were:

1. All banks were eligible for support.

2. Support in the form of equity was preferable to debt, as the government would benefit from improvement in the stock when the bank recovered.

3. The government had the expressed wish not to take over banks as it had no interest in socialising the industry. Apart from ideological principles, the government recognised that it is unfortunate to be both a legislator and the owner of banks subjected to legislation in a competitive environment. Furthermore, there is always the risk of political considerations in the management of a bank that could make its operation suboptimal (e.g. by lending money out to pet projects at non-market rates). Only as a last resort would banks be nationalised, and if so, the government stake would be a temporary solution.

4. Minimisation of moral hazard. Whenever the BSA supported equity capital, existing stockholders would see the value of their equity decline correspondingly.

5. All participating banks had to disclose all known and expected losses and collateral values

6. A tricky conundrum was how the BSA should value debt and collateral. If the value was set too low, the banks could go bust. But if the value was set too high, taxpayers would risk making a bad deal. The guiding principle was to make conservative assessments rather than the opposite. Not least because the BSA thought too optimistic a valuation of real estate collateral could breed a negative surprise to the market if actual price developments undershot the BSA valuation. The perception of continued value loss on collateral would be very detrimental to the whole project of building confidence. However, there was the option for a more flexible approach to valuation as well. The BSA could under certain circumstances accept a valuation of the collateral based on discounted cash flow from the property for the next few years. Since valuation of collateral is a subjective process, the BSA created a Valuation Board made up of experts on real estate assessments in order to double check the valuations made by the banks themselves.

7. The BSA also had to assess whether or not a bank was worth rescuing at all. In doing this it established a three-tier model where it assigned bank to an A, B or C category. The A-category bank was an institution that was in decent shape and would likely not come under the 8% capital ratio limit. The B-category bank is expected to temporarily come under the 8% limit, but recover after a few years. The C-bank is not deemed viable at all. The A-bank ought to be able to manage the crisis itself by attracting new capital from existing or new owners. The B-bank would have to accept help from the government in order to make private investors confident enough to supply capital to the bank. The government would issue a ‘capital ratio guarantee’, which made sure the government would keep the capital ratio at 9% or more. The C-bank would be liquidated at once and the bad credits would be transferred to a company whose sole purpose was distressed debt management. This company was called Retriva. Parts of the C-bank that were deemed healthy enough to survive would be auctioned out to prospective buyers.

8. In order to expedite the process and prevent lengthy litigation processes from disgruntled existing equity owners, who would accept losses in return for the government taking a stake in the bank (see #4) a new law was passed. The law gave a board of independent judges the inalienable right to decide what the fair value of the existing stock owners’ equity should be. The decision could not be appealed.

9. The two companies formed to take care of distressed debt, Securum and Retriva, hired their employees mostly on a consultant basis. The reason for this was that the companies only were supposed to be temporary entities that should work towards their own liquidation. The government thus wanted to avoid that these large entities should ‘take on a life of their own’ and expand of their own accord into a financial behemoth rather than the opposite (empire building).

The Final Outcome and Cost

Securum started its operations on January 1, 1993. Its assets were a portfolio of non-performing loans. Securum became the owner of some 2,500 commercial properties with an estimated market value of SEK15-20bn, corresponding to 1-2% of all commercial real estate. Since it was thought that Securum should not flood the market with foreclosed real estate assets that would only depress prices and create further losses, it was decided that Securum should unload the assets in three ways: IPOs on the Stockholm stock exchange, corporate transactions outside the stock exchange, and transactions involving individual properties. Most of the sales were made in 1995-1996 when the real estate market had stabilised.

By and large it must be said that the Swedish bank crisis management was very successful. The two companies, Securum and Retriva, were expected to last for 15 years after their creation. They were done liquidating their assets after four years and ceased to exist thereafter.

The total cost incurred by the state is estimated to be some SEK35bn (corresponding to around 2% of GDP in 1997)1. The total investment by the state was SEK71bn. The proceeds from the asset sales, the dividend income from purchased bank equity, proceeds from sold stocks (Nordea) and the mark­to-market value of the remaining Nordea stock in 1997 was estimated to be some SEK36bn. However, there is wide consensus that the opportunity cost by not having dealt firmly with the situation would have been significantly higher.


1 Englund (1999), The Swedish Banking Crisis: Roots and Consequences. This article is largely based on the above paper and Ingves & Lind (1998), Om att hantera en bankkris.

Home Download Sample Order Economic Forecasts Contact

Copyright © 2008-2013 Consensus Economics Inc Site Map