Via Brad Setser, Steve Johnson describes the recent sell-off in world markets:

The selling was largely driven by a sharp rise in volatility, with the Vix index, often referred to as Wall Street's "fear gauge" hitting a two-year high.

This increased the "value at risk" of leveraged investors such as hedge funds, forcing them to cut long positions. Many of the assets that had made the strongest gains this year, such as emerging markets, fell most sharply as a result.

This issue of VAR-driven sales on increasing volatility bring up a very general point about risk mitigation in financial markets: Any single risk-mitigation scheme widely adopted to reduce the risk of individual participants increases systemic risks to the market. Runs on banks, programmed stop-loss sales, and now sales at VAR thresholds are some examples of this that we've seen. All are dangers of risk management monocultures.

Dangers of derivative-based risk control are more subtle, because derivatives allow the party that would otherwise bear risk to hold their "insured" positions rather than scrambling for the doors. But if the bearers of transferred risk manage their exposures very similarly, then their collective behavior creates systemic risks that come back to bite holders of the underlying. For example, if bearers of credit risk in swap arrangements mechanically short the credit market to hedge risk as credit spreads grow, bond prices will fall and credit spreads widen just as surely as if uninsured bondholders had sold-off themselves. Derivatives allow for greater diversification and dispersion of risk, ideally to those who best placed to bear it. This probably does push back the thresholds at which dangerous risk-mitigation herd behavior kicks in.

But if the buyers of risk all turn out to look alike and act alike, if their tolerance for bearing risk is limited, if they have underpriced risk in tranquil times, then everybody had better watch out. Even for those who have trimmed exposure rather than assuming risk, derivative-based risk management, like diversification, can only protect against security or sector-specific risks, not against systemic risk. If everyone bearing any important category of risk has similar ("industry best practice") ideas about how to cut incipient losses, that's a source of systemic risk. Beware risk management monocultures.

In a conversation about a recent post of Michael Shedlock's, Eugene Linden likened financial risk to environmental toxins. His very astute observation was that the use of derivatives to disperse risk is much like the strategy of managing toxic chemicals by flushing them to the sea. At first blush, this is eminently sensible, because poisons when sufficiently diluted are not poisonous at all, and may be harmlessly absorbed by the environment. But in fact, the strategy is fraught with peril. Some toxins don't stay dispersed, particular agents accumulate them, and become poisoned and poisonous. In a bay, shellfish might accumulate dangerous levels of heavy metal, and become unfit for human consumption. In the financial world, we know that lots of risk has been created and "dispersed" via derivatives, but we don't have good information on whether all this financial risk is harmlessly diversified, or has concentrated dangerously in the hands of some speculative agents. The accumulation of metals by seabed mollusks does harm not only to the mollusk. Similarly, ill-advised accumulation of risk by hedge funds or investment banks could lead to consequences extending far beyond the risk-accumulator.

I thought this was a particularly good analogy, and have been thinking of other pollution metaphors in finance. "Global imbalance" is a shorthand commonly given for the situation in which some countries — in particular the United States — are spending more than they collectively earn, while others (China, Saudi Arabia, Russia) are selling their goods and services for debt. These habits have become embedded structural facts of various economies, implying changing the pattern — the US builds up debt, China and the oil states — will require "economic adjustment". Economic adjustment is a euphemism for hard times, and this arrangement is not just a private matter between a few countries. In today's interlinked global economy, US consumers provide a great deal of world demand and the US dollar is the global "reserve currency". China's population represents nearly one quarter of the world population, and its political stability depends upon economic growth. If "adjustment" doesn't occur smoothly, there could be a global economic downturn, political instability, resource wars, you name it. These are all improbable events, but they are risks borne by the entire world, and risks that grow with continued unbalanced world growth.

So, in a sense, "global imbalance" is also a lot like pollution. The United States, China, the elites who run the petrostates, are all "getting something" out of the present arrangement. (The US gets to consume inexpensive goods paid for with low-interest IOUs that may never be fully paid; China gets export-led rapid-fire development; and the petrostates get high oil prices without crimping demand, since the US is buying with debt and China is taxing and subsidizing oil consumption.) But, at the same time, they are poluting their own economic and political environment, and that of the rest of the world, with risk. Like polluters generally, "global imbalancers" continue to do what they're doing because in the short-term it's profitable, long-term costs are uncertain and won't be borne solely by the polluter. In economist-speak, global imbalance, like pollution, is an "externality".

Of course, with respect to pollution, industrial societies have been dealing with these problems for a long time. There's no formula for preventing pollution, but one approach that's worked very well both environmentally and politically are so-called "cap and trade" programs. I got to thinking about how we might "cap and trade" global imbalances.

I'm not the brightest bulb on the tree, but Warren Buffett just may be. He proposed a scheme called "import certificates", way back in October 2003. Revisiting it, what does it amount to? It is precisely a cap and trade scheme for global imbalance. I read this way back when, and thought it was kind of loopy. Rereading it now, it seems kind of brilliant. This Buffett guy, he's not so shabby. Why did this proposal get so little notice? It's remarkable in that it permits markets to determine who can most efficiently bear the cost of global rebalancing, and permits gradual application, as the initial "cap" on the US trade imbalance could be set at something like the present level, with a commitment reduce US overspending only gradually. It's financial innovation at its very best. Why hasn't this been done? Why don't we do it now?