Why Sweden? Because Sweden had its own financial crisis in the early 1990s, and by many accounts did a reasonably good job of pulling out of it. A housing bubble, fueled by cheap credit, collapsed in 1990, with residential real estate prices falling by 25% in real terms by 1995 and nonperforming loans reaching 11% by 1993, while the Swedish krona fell in value by 30%, hurting a banking sector largely financed by foreign funds. As Urban Backstrom said in a 1997 paper, “[the] aggregate loan losses [of the seven largest banks] amounted to the equivalent of 12 percent of Sweden’s annual GDP. The stock of nonperforming loans was much larger than the banking sector’s total equity capital.” In other words, the banking sector as a whole was broke.
So what did Sweden do? If the options on the table in the U.S. right now are (a) additional recapitalization, (b) an aggregator bank to buy up bad assets, and (c) nationalization, the Swedish solution included all three. First, in late 1992, the government guaranteed all bank creditors (but not shareholders), with no upper limit. Because investors did not at the time question the solvency of the government, this meant that they would continue to lend money to the banks, and the central bank provided unlimited liquidity just in case. Although the U.S. has guaranteed new debt issued by banks, and there is virtually an implicit blanket guarantee for at least the largest banks, there is still uncertainty among bank creditors, as witnessed by credit default swap spreads.
However, even if an insolvent bank has access to credit, it is still an insolvent bank, hoping somehow to become solvent, so it’s unlikely to lend or, even worse, it may be tempted to make extremely risky loans as the only possible path to solvency. As a condition of government support, government auditors reviewed the balance sheets of the all the banks involved, with the goal of taking writedowns immediately and showing the true state of affairs. When it turned out that two major banks, Nordbanken and Gota, were insolvent, they were nationalized (Nordbanken was already largely state-owned), giving the state control of over 20% of the banking system (by assets). Gota was merged into Nordbanken, which only held onto “good” assets, and the “bad” assets were moved to two new entities, Securum and Retriva. These entities were capitalized by the government, and bought 21% of Nordbanken’s assets and 45% of Gota’s assets. This is an example of the good bank/bad bank plan that has gotten so much attention lately. Nordbanken itself (the good bank) was recapitalized by the government, to the tune of 3% of GDP, and become a healthy bank, while Securum and Retriva were told to get whatever value they could out of the bad assets.
Securum and Retriva were run like a cross between private equity firms and asset management companies, both managing and improving assets and also finding buyers for the assets. According to the Cleveland Fed, they managed to return $1.8 billion out of their $4.5 billion in initial capital to the government, for a net taxpayer loss of $2.7 billion. (I can’t figure out if the government also lost money on the loan guarantee, although the sources I read implied that it didn’t.) And Nordbanken, after being run by the government, was eventually privatized (the government’s ownership share is now 19.9%), and the taxpayer recovered the capital put into it in the rescue. As I said above, this is generally seen as a success story, although the Cleveland Fed does have a sobering conclusion:
the cost of the crisis to Sweden was not limited to the capital spent by the [asset management companies]. There have been significant income and output losses associated with the crisis. In the early 1970s, Sweden had one of the highest income levels in Europe; today, its lead has all but disappeared. Cerra and Saxena (2005) found that the crisis caused a permanent decline in output that can explain the entire fall in Sweden’s relative income. So, even well-managed financial crises don’t really have happy endings.
The Swedish story is usually used as an argument in favor of nationalization, and that’s not an implausible inference to draw. But another lesson you can draw is that it’s not the nationalization per se that matters, but the pricing of the bad assets. The key was that the banks were forced to write down their assets in one shot and then to sell them to the bad banks at realistic prices. That cleaned up their balance sheets and, once they were recapitalized, allowed them to operate as healthy banks. As we said a long time ago, TARP was a fine idea as long as it paid fair value for assets and was combined with recapitalization to fill the resulting hole in bank balance sheets. The same holds for an aggregator bank. The problem would be letting the banks decide which assets they want to sell, and then letting them unload them on the aggregator bank at inflated prices. That solves nothing.
Originally published at the Baseline Scenario and reproduced here with the author’s permission.