Apologies for the reliance on the Financial Times today, but it happened to have a lot of good material.
The CEO of Harvard Management, Mohamed El-Erian, writes the occasional opinion piece, usually for the Financial Times, and I’ve always featured them because they are consistently thoughtful and well-argued.
I’m highlighting his latest FT piece, “How to reduce risk in the financial system,” for a different reason. As a faithful reader of El-Erian, I noticed a marked shift in his tone. He’s worried. And as a sophisticated investor who has an open-ended mandate, meaning he surveys more markets than most professionals (and almost certainly gets the best intelligence), that is troubling.
He articulates the core problem more clearly than I have seen anyone to date: in our brave new world of finance, we have new, sophisticated techniques that allow for much finer risk parsing and large-scale risk transfer. So the banking system is more solid that ever before because banks are no longer the repository of risk.
But that begs the question: then who is the bagholder? El-Erian tells us that it is increasingly insurance companies, pension funds, and other not-so-savvy players who for the most part lack the internal capacity to do due diligence.
To give you an idea of how bad it is, in a related post today, we discuss an article that tells us that there are only 50 managers who have sufficient skills to price CDOs, and that even the dealers can’t price them (not that the methodology is beyond them, but it takes huge amounts of time, and they can’t staff those areas that heavily). That means pretty much everyone who owns CDOs is clueless.
Here are El-Erian’s closing remarks:
What about the future? If left unchecked, systemic risk in the international financial system will increase owing to the combination of insufficient internal due diligence, excessive dependence on rating agencies, uneven supervisory coverage and politically driven legislative reactions. In the process, much of the initial beneficial impact of credit risk transfer technology may be negated.
Three steps can mitigate this new component of systemic risk: first, stimulate greater co-operation and sharing of expertise among the spaghetti bowl of supervisory bodies; second, encourage rating agencies to improve their modelling of new and complex derivative products; and third, induce new investors to evaluate the ratings issued by the agencies against improved internal risk management capabilities. The longer action is delayed in these three areas, the higher the likelihood that costly clean-up operations will be needed.
If this is the best in the way of solutions that a very well connected and thoughtful participant can come up with, we are in trouble indeed. Consider El-Erian’s first recommendation, that regulators should cooperate more. My sense is that they know that already and are endeavoring to do so. The Fed and other central banks have been talking among themselves more than before; the Bank of International Settlements is playing a necessary and more active role; and to a degree, the US regulators are communicating a bit more. But the US authorities not only have a history of not playing well together, but more important, their mandates give them differing perspectives and priorities that make it hard for them to work jointly. And some of these new instruments don’t fall neatly in anyone’s purview.
To illustrate: the SEC has started an investigation of CDOs. We doubt it will be very successful (unless they throw considerable resources against it) because the SEC’s enforcement apparatus isn’t well suited for this sort of thing. The SEC has an equity, not a debt markets focus, its enforcement division consists almost entirely of attorneys who are used to going after bucket shops and insider trading. By contrast, the Fed and banking regulators, who have been trying to understand this market better, have been unable to get deal documents because they aren’t accredited investors and lack the regulatory reach.
El-Erian’s second recommendation, “encourage the rating agencies to improve their models” is frankly a bit daft. First, no matter what you think of them, the rating agencies are already doing the best they can. Second, for them to make major changes at this juncture (as opposed to incremental changes that they could position as routine) when they are already getting a barrage of criticism, might create liability. Third, as we have discussed, the agencies are unable to hold on to talent. They serve as a training ground for investment banks and fund managers who can pay more. So it’s structurally impossible for them to do a systematically better job unless they can pay people better. And we don’t mean a little better, we mean a lot better, probably 3 to 5 times as much for their analytical talent (which then creates management problems, since in most organizations managers can’t stand having their juniors make more than them). Their business model simply won’t support it.
To be honest, I’m not at all sure what El-Erian means by his third idea, ” …induce new investors to evaluate the ratings issued by the agencies against improved internal risk management capabilities.” I don’t know how you “induce” investors to do anything. I think he means “compel” but doesn’t want to admit to himself that new regulations might be in order.
Reading El-Erian’s list confirmed my concerns that the problems in the marketplace are deeply rooted. Perhaps we’ll get lucky and somehow things will unwind themselves without too much pain. But if the problems worsen, we run the risk, as he points out, of legislative intervention aimed at symptoms rather than root causes, which is likely to make matters worse.
If the problems are indeed fundamental, that suggests we need a new regulatory regime, a sort of debt markets version of what the 1933 and 1934 Acts did for the equity markets. Incremental fixes won’t create integrated regulatory oversight. It would take a 1929 scale crisis to provide the impetus for root and branch reform.
From the Financial Times:
Regulatory authorities face two challenges that need to be addressed forcefully if they are to contain a new source of systemic risk in international finance. First, the increasing migration of complex market activities to supervisory bodies that lack the necessary sophistication to oversee them; and, second, a growing threat of politically motivated changes to regulatory regimes.
There has been much talk recently about the extent to which the proliferation of derivative products has allowed banks to manage their balance sheets better. By enhancing the ability to hedge and shift various risks, advances in what is called “credit risk transfer” technology have lowered the vulnerability of the international financial system to any individual bank crisis.
There has been less discussion about where the transferred risk has ended up and why. Increasingly, it is being borne by a new set of investors who previously had limited access to complex derivative products. These include insurance companies and public and private pension funds. They see the products as a way to earn higher yield.
The growing purchase by such investors of “structured products” is, in itself, acting as a catalyst for the creation of these products by banks. Indeed, given the considerable fees involved, banks’ business models are being reorientated away from the traditional structuring and holding of individual loans. Instead, the emphasis is now on originating and quickly distributing structured products.
For the purposes of analysing the implications for systemic risk, the new investors bring two important characteristics into play. First, many rely on external risk assessments rather than in-house due diligence, with a particularly heavy dependence on rating agencies; and second, they are supervised by bodies lacking the financial sophistication inherent in structured products. Both point to an increase in risk for the international financial system.
Recent analytical work raises concerns as to whether rating agencies’ (and others’) modelling of structured products has kept up with the massive growth in the volume and complexity of these products. The recent experience with US subprime products adds to such concerns. Worries centre on the “correlation modelling” that underpins rating designation. Given the leverage in many of the products, even a small change in correlation specifications can have a large impact on ratings.
Meanwhile, the responsibility for supervising the transfer in balance sheet risk increasingly falls outside the purview of those best equipped to handle such a complex task. Especially when compared with bank regulators and boards, bodies overseeing insurance companies and pension funds have had limited exposure to the structured products that increasingly populate the balance sheets they supervise. These concerns come at a time when politicians are looking more actively at the investment vehicles that, directly or indirectly, facilitate risk transfer to insurance companies and pension funds. Political activity will increase further should some of the new investors find themselves in the midst of large derivative-related losses.
In a recent FT View from the Top interview, Lloyd Blankfein, chief executive of Goldman Sachs, sounded a cautionary note based on something that he picked up at Harvard Law School. He remarked that politically inspired changes triggered by a reaction to a specific situation or an individual firm can have unintended negative consequences for the system as a whole.
What about the future? If left unchecked, systemic risk in the international financial system will increase owing to the combination of insufficient internal due diligence, excessive dependence on rating agencies, uneven supervisory coverage and politically driven legislative reactions. In the process, much of the initial beneficial impact of credit risk transfer technology may be negated.
Three steps can mitigate this new component of systemic risk: first, stimulate greater co-operation and sharing of expertise among the spaghetti bowl of supervisory bodies; second, encourage rating agencies to improve their modelling of new and complex derivative products; and third, induce new investors to evaluate the ratings issued by the agencies against improved internal risk management capabilities. The longer action is delayed in these three areas, the higher the likelihood that costly clean-up operations will be needed.
