The global financial crisis is rightly prompting calls for a rethink of how we regulate financial institutions and markets. Most such calls are focused on what might be called "fail-safe" regulations, designed to reduce the risk that institutions will make reckless lending and investment decisions. Even libertarian-leaning policymakers and thinkers, such as Alan Greenspan, are now concerned with the capacity of our globally connected financial system to spread failures of risk management from one institution to another.
In this crisis, institutions that bought up buckets of complex mortgage-linked securities found themselves facing huge losses as house prices fell. Their counterparts and clients, fearing the worst, provoked the worst by ceasing to do business with them. Others who wrote insurance against their failures-to-pay (credit default swaps) then lost huge sums as well, fuelling the fires of system-wide panic and default. But better regulation of lending standards and risk management, the argument goes, will prevent such systemic problems in the future.
History does not provide much comfort here. In financial markets, there are always new risks to take and new ways for risk management models and procedures to break down. Fail-safe approaches can also go too far: witness Japan in the early 1990s, when heavy-handed government intervention effectively shut down financial innovation. Furthermore, government policy promoting imprudent risk-taking-witness long-standing US congressional support for failed mortgage giants Fannie Mae and Freddie Mac-can overwhelm regulations intended to control it.
The key is to supplement prudent fail-safe interventions with safe-fail ones: interventions that recognise that institutional failures will continue to occur and that focus on limiting the systemic damage after they do.
A case in point is the mammoth global derivatives markets. Despite wild swings in prices, derivatives exchanges have not contributed one iota to market instability. This is because exchange-traded contracts are centrally cleared and trader defaults-which are rare because of continuously adjusted margin requirements-are absorbed by well-capitalised clearing houses. Compare the 2006 collapse of hedge fund Amaranth, whose derivatives exposures were on-exchange, with the 2008 collapses of Lehman Brothers and AIG, both of which had large exposures in non-cleared, over-the-counter CDSs. Amaranth's derivative defaults had trivial systemic ripples, while those of Lehman and AIG created major shockwaves. AIG invisibly built up huge under-collateralised sell positions on the back of a faulty credit rating. Yet if those contracts had been transacted on a trading platform with central clearing, margin calls would have short-circuited the strategy well before the company's September collapse. US and European regulators (whose institutions comprise the vast bulk of OTC trading) should require central clearing once volume barriers in a contract are breached. This will not prevent an institution from losing large sums in derivatives trading, but will stop its default from spreading big losses to others that may be far removed from the original transactions.
There are safe-fail macroeconomic counterparts as well. In August 2007, former Salvadoran finance minister Manuel Hinds and I spoke out at a Reykjavik conference in favour of Iceland unilaterally "euroising". At the time, the country had more than enough foreign exchange reserves to redeem all the krona in the country for euros at the then-current exchange rate. This would not have stopped the three large Icelandic banks from overextending, but it would have prevented national financial catastrophe. Countries that have adopted one of the two main internationally accepted currencies-the dollar (Panama, El Salvador and Ecuador) or the euro (think in particular of Italy, Portugal and Greece)-have effectively eliminated the risk of currency crisis (that is, not being able to pay short-term foreign debts for lack of access to "hard currency").
If we are wise and fortunate, in the future we will have corporate governance, capital standards and monetary policy regimes that better constrain the dangerous build-up of excessive leverage among consumers, banks and governments. But that is not enough. We need new safe-fail policies to prevent inevitable institutional failures from snowballing into economic crises.
This article appears in full on CFR.org by permission of its original publisher. It was originally available here.