Too Big to Fail? How About Too Big to Exist?

In 1996 a single power-line failure in Oregon led to a massive cascade of power outages that spread across all the states west of the Rocky Mountains, leaving tens of millions of people without electricity. Over the past year we have experienced a different kind of cascade in the financial system, which has produced the equivalent of a global blackout. Having studied the dynamics of cascades in complex systems, I suspect that the most damaging ones are impossible to anticipate with any confidence. The solution may therefore be to make the system less complex to start with, in order to reduce the chance that any one part can trigger a catastrophic chain of events. In the financial system, this means limiting how big companies are allowed to become. As governments struggle to fix the crisis, plenty of experts have weighed in on its causes, from excess leverage to lax oversight to faulty compensation structures. These explanations can account for how individual banks, hedge funds, and so on got themselves into trouble, but they gloss over the larger question of how all these institutions, acting independently, managed collectively to put trillions of dollars at risk without being detected. This risk was invisible because it was systemic—it resulted from the unpredictable interplay of myriad parts in the system. Think about power grids again. Engineers can reliably assess the risk that any single power generator in the network will fail under some given set of conditions. But once a cascade starts, they can no longer know what those conditions will be for each generator—because conditions could change dramatically depending on what else happens in the network. The result is that systemic risk, which can cause the system as a whole to fail, is not related in any simple way to the risk profiles of the system’s parts. Financial systems are arguably far more complex than power grids, but the fundamental problem of systemic risk is the same: Risk managers are only able to assess their own institutions’ exposure on the assumption that conditions in the rest of the financial world remain predictable, but in a crisis these conditions change unpredictably. No one anticipated that an investment bank the size of Lehman Brothers could collapse as suddenly as it did, so no risk managers built that contingency into their models. How do we reduce the risk of cascades in the financial system? One approach builds on the way regulators currently make judgments about systemic risk, in particular when they decide that some institutions are too big to fail. There are lots of problems with these judgments, as the Lehman Brothers fiasco revealed, but the most serious is that they are made after a crisis emerges, at which point only drastic responses are available. A better approach, therefore, would be for regulators to routinely review firms and ask: “Is this company too big to fail?” If yes, the firm could be required to downsize or shed business lines until regulators were satisfied that its failure would no longer pose a risk to the whole system. Correspondingly, proposed mergers and acquisitions could be reviewed for their potential to create an entity that would be too big to fail. Governments telling firms what they can and can’t do sounds like dangerous meddling in free markets. But antitrust law already permits regulators to prevent firms from growing in ways that stifle competition, and somehow our free market has survived. The current crisis has demonstrated that markets do not automatically control systemic risk, any more than they automatically create competition. Pragmatically speaking, therefore, government intervention is required to prevent markets from destroying themselves, and the relevant question is what kind of intervention is effective. The answer will be complicated, but it should include this simple principle: Firms should not be allowed to grow too big to fail in the first place.

Too Big to Fail? How About Too Big to Exist?

Summary.

Partner Center