Any discussion about the financial crisis and risk management involves some theorizing about derivatives. Who better to cast light on the subject than one of the architects of modern risk management, Robert C. Merton, a Harvard Business School professor and a winner of the 1997 Nobel Prize in economics. Merton’s Nobel was awarded in recognition of his role in developing a new method for valuing derivatives. Since the method’s publication, in 1973, the markets in derivatives have exploded, and today the notional value of outstanding contracts is estimated to exceed $500 trillion. The following interview is an edited version of recent conversations between Professor Merton and HBR senior editor David Champion. Many people believe that financial innovation, particularly derivatives, caused the global financial crisis. Do you agree? No. Derivatives are ubiquitous in the financial system, and thus will be part of any crisis, but the instruments themselves cannot be its cause. They are simply tools that can be used either functionally, to reduce risk, or dysfunctionally, in ways that increase risk without offsetting benefits. That said, innovation is structurally linked to an increased likelihood of a crisis: Successful innovation by its very nature initially outstrips our ability to regulate it, since existing infrastructural supports will be insufficient to safely apply its full potential. Out of 100 innovations, maybe only one or two will work out. It is not practical to spend untold resources to build a full infrastructure for all 100 at the get-go, before you know which will succeed. Of course, even if you did, you still couldn’t predict what all the effects of the innovation would be. In finance, as elsewhere, you have to make a trade-off between the benefits and risks of innovation. The only way to make us absolutely safe from innovation is to abstain from it altogether. In any event, to argue that we should ban derivative contracts when no major financial institution, including central banks, can function without them is tantamount to saying that we should ban cars to reduce the risk of global warming. It’s neither a meaningful nor a desirable aspiration. The only way to make us absolutely safe from innovation is to abstain from it altogether. You’ve said that understanding derivatives provides insight into the riskiness of conventional banking assets. How? Let’s suppose that you are a bank contemplating a loan application from a company. Before approving the loan, you might look for a guarantor, someone whose credit you don’t doubt. Once you have this guarantee and you make the loan, you have a composite asset of two parts—loan and guarantee—that is collectively a risk-free asset. Fair enough, you will say, for guaranteed debt. But what about loans that aren’t guaranteed? Well, actually, these loans are still guaranteed, only the guarantor is no longer a third party—it’s you, the lender, writing a guarantee to yourself that the borrower will repay. The asset still has two parts: the loan and the guarantee. But in this case you retain the risk, because the guarantee is a liability for you, not a third party. How big is this liability, and what is its risk? It turns out that the characteristics of a guarantee are exactly the same as those of the put option, one of the most widely traded derivative securities in finance. That means we can bring to bear all that we’ve learned over nearly four decades about pricing and measuring the risk of put options to see what’s happening on the liabilities side of the balance sheet when a bank makes a loan. What would that derivatives experience tell us about what happened in the crisis? It would help explain why the risks that banks faced grew so big so quickly and continued to do so even after they stopped new lending. One of the most important risk properties of an option is a feature called “convexity.” What this means is that as the price of an underlying asset falls, both the price and risk of a put option or guarantee increase disproportionately. And the further the asset’s price falls, the more sensitive the option price becomes to the next fall. So with the repeated declines in housing prices, the implicit put options that banks had extended in buying up mortgage debt increased in value by ever greater amounts, creating a huge liability that grew out of proportion to the fall in house prices and got still larger even after the banks stopped buying mortgage loans. Unfortunately, many of those responsible for firms with corporate or mortgage loan portfolios are not familiar with the concepts behind derivatives, which helps explain why so many management teams, boards, and regulators in the banking industry were surprised by how inadequate their capital provisions turned out to be. What advice do you have for dealing with the large portfolio of troubled loans owned by the taxpayer? I do worry that the people responsible for managing these troubled loans, along with various other assets acquired by the government, will act as liquidators and sell off whatever they can as soon as they can in order to get those assets off the government’s books and generate cash as quickly as possible. If that happens, the taxpayer will most likely lose out. Given the magnitude and diversity of assets involved, I’d prefer to see the government set up a kind of sovereign wealth fund, run as an investment portfolio by independent managers versed in financial technology, who can ensure that the government gets the full value for assets when they’re sold. Obviously, investment managers will have to sell the assets, but they should not feel pressured to sell at any price just because they’ve found a bidder. What can we do to make the financial system safer? We clearly need regulatory changes, both in the rules and structure and in their execution. As we expand the regulatory scope, we place greater demands on the regulators to exercise judgment on technically complex issues; if they are to do so effectively, provisions must be made for properly trained staffs. There’s also a danger that we’ll regulate too quickly and what we do will have bad unintended consequences. We saw that with Sarbanes-Oxley. Nevertheless, there are some regulations—for over-the-counter, or OTC, derivatives, for instance—we could enact right away that would probably give more than 90% protection against the kind of systemic breakdown we’ve seen. First of all, financial institutions should be required to provide explicit margin collateral against their OTC derivative liabilities. This would not be an undue burden because financial companies can manage collateral at low cost. We should also require all financial companies to report positions on a fair-value rather than a book basis, so that their collateral covers the actual liabilities of their derivative exposures. I would not advocate collateral requirements for nonfinancial companies that make it expensive for them to use derivatives to reduce the risks they face in their normal business. Another useful move would be to create a central clearing system for OTC derivatives trading of contracts that can be adequately standardized. When many different parties are making bilateral trades, the result is a gross, rather than a net, accounting of contract exposures. An exchange that functioned as the central counterparty would facilitate the netting out of contract exposures, making it easier to contain the cascading effects of one firm’s default. For custom derivatives or for nonfinancial firms that did not want to manage collateral, regulations could establish position limits to prevent these noncentrally cleared contracts from creating a systemic risk and to encourage the use of central clearing. A third step would require that senior management teams, board members, and regulators of financial institutions demonstrate adequate understanding of the financial technology and risk management concepts necessary to perform their oversight roles effectively. Do you see any other means for making financial markets safer? As another check on the system, the government could set up a financial equivalent to the National Transportation Safety Board, an idea suggested by Andrew Lo at MIT. The NTSB is not a regulator, but its findings, which are publicly available, inform the decisions of regulators, lawmakers, the industry, and the public. I think this is very sensible. Creating a strong independent body staffed with forensic experts to investigate failures within the financial system on a continual basis would help ensure that regulations are formulated with the best and least-biased information.
