AS GOVERNMENTS struggle to prevent the global financial crisis turning into a deep worldwide recession, attention is also turning to the longer-term problem: how to avoid a similar crisis happening again. When politicians meet in Washington DC in December they are likely to agree that the “loose touch” approach to financial regulation of the past two decades will have to give way to tighter controls. But the global financial system now operates at a level of complexity no one has ever tried to tame. How do we re-engineer it so breakdowns don’t happen again?
One place to start is the science of complexity itself. We now know that large interconnected systems, such as the weather, can behave in unexpected ways: for example, small changes can trigger fundamental shifts. New understanding of the principles governing such complex systems offers hope that the global financial system can be got under control. The snag is that politicians will have to accept that costs are likely to be involved.
Existing economic policies are based on the theory that the economic world is made up of a series of simple, largely separate transaction-based markets. This misses the fact that all these transactions affect each other, complexity researchers say. Instead, they see the global financial system as a network of complex interrelationships, like an electrical power grid or an ecosystem such as a pond or swamp. In a swamp, certain chemicals that normally keep pond life ticking over can, under the wrong circumstances, trigger an explosion in the numbers of one species – an alga, say – which then goes on to strangle all other life in the swamp.
Advertisement
“Complexity theorists see the financial system as a network of relationships, like an ecosystem such as a pond or swamp”
Similarly, they say, apparently unimportant changes that have crept into the global financial system may have triggered the current crisis. “Slow changes have been accumulating for years, such as levels of indebtedness. None on their own seemed big enough to trigger a response,” says Johan Rockström of the Stockholm Environment Institute. “But then you get a trigger – one investment bank falls – and the whole system can then flip into an alternative stable state, with different rules, such as mistrust.” To prevent events like this, governments need somehow to restructure global finance to limit these kinds of instabilities.
So how exactly has the financial system come to be so vulnerable? One key factor is that money can now flow more easily from country to country. This has stimulated trade and prosperity throughout the world, but it also means that an upset in one place can have severe and unpredictable consequences elsewhere.
Days before winning the 2008 Nobel prize in economics last week, Paul Krugman of Princeton University which concluded that the rapid increase in cross-border investments since 1995 is what allowed a local shock – the collapse in inflated US real estate values – to propagate globally, especially through highly indebted investment firms that can respond to a loss of money in one place by pulling back credit anywhere in the world. Krugman noted that “these channels are not yet part of the standard analysis”. This is exactly the kind of linkage that the complexity theorists say economists have been missing. “The source of the current problems is ignoring interdependence,” says Yaneer Bar-Yam, head of the New England Complex Systems Institute in Cambridge, Massachusetts.
For example, he says, financial firms calculate the risk involved in taking on a debt for each transaction separately, and then simply add them up to arrive at the total risk. This made the whole system look more secure than it actually was, because failure in some transactions can in fact multiply the risks of others. “They totally failed to account for couplings between them, which can change things dramatically,” Bar-Yam told 91av.
Warnings go unheeded
The banking industry itself was not entirely oblivious to this. In 2007 the Federal Reserve Bank of New York published a study, based in part on testimony from ecologists and engineers specialising in complex systems, which concluded that while vast sums were spent assessing the risks of individual investments, almost nothing was being spent on systemic risk – which could be much more grave. “It is really frustrating to me and others that the warnings were not heeded,” says ecologist Simon Levin of Princeton University, who was one of those who testified – all the more, he points out, because increased connectivity doesn’t just propagate trouble, it makes the whole system less diverse and more vulnerable to dramatic shifts.
According to Rockström, one of the key ways in which diversity was lost arose from the uniformity of criteria that have been used to judge economic success. One example of this is value-at-risk (VaR), the measure used by banks to report their potential losses from trading financial instruments. Since the late 1990s, it has been standard practice for banks to publicly report VaR measurements. When use of this measure was proposed, critics argued that this would encourage herd behaviour, with banks rushing en masse to sell off assets that were depressing their VaR numbers, but their concerns were ignored.
To prevent this sort of thing, complexity experts say that “firebreaks” that cut back connectivity should be built into the financial system, and that it must become less uniform. “The heterogeneity of the system must be restored,” says Levin. In other words, don’t let everybody become part of the same pond, all following the same rules.
“Firebreaks should be built into the system, even if this makes it more expensive for banks to do deals with each other”
This is unlikely to be popular with the banking industry: it lost diversity and gained connectivity in the first place because this cut costs and boosted profits. But such diversity is what allows ecosystems to remain resilient as conditions change; the same principle should apply to financial systems.
Achieving this is likely to be difficult, not least because in a complex network like the financial system, no one is in charge. And if international coordination were to happen, it could end up imposing even greater rigidity and uniformity. “Governments will have to be very careful, and set rules and limits for the system without actually telling people what to do,” says Bar-Yam. It can be done, he says, citing Wikipedia as a good model for such coordination.
Bar-Yam says, however, that he sees little evidence of such thinking in government responses to the crisis so far. For example, early in the crisis, US regulators put limits on the “short sellers” who speculate that a company’s stock value will stop rising. Short sellers were supposed to weed out weak firms, but a loosening of rules last year meant that they also brought down healthy ones. Bar-Yam says the result was like a predator-prey system run amok. Yet when regulators acted, they only stopped them attacking certain types of company, leaving them free to pounce on others. “It shows they still aren’t thinking about this as a connected system,” he says.
Bar-Yam thinks models of complex relationships such as those between predators and prey can help prevent such systemic problems, and show regulators how they can modulate market behaviour in a more sophisticated way. “We haven’t had the scientific tools to do this for very long. But we can now model the global system and capture the key collective behaviour that causes collapse.”
“At its core the science of complex systems is about collective behaviour,” Bar-Yam points out. “The invisible hand of the market is collective behaviour.” The problem till now, he says, has been that economic policy has failed to take into account the complexity and consequent unpredictability of such behaviour. In the absence of testable models, people “try to believe what they know really isn’t true”, he says – for instance, that real estate values always increase.
The remedy Bar-Yam proposes is to subject economic policies to verification with the same sort of rigour that is normal in science. That has never been done. “But with recent scientific advances, I believe we can now truly inform policy.”
