How did securitization allow trillions of dollars of risky assets to be transformed into securities that were widely considered to be safe? Structured finance pools economic assets such as loans, bonds, and mortgages and then issues a prioritized capital structure of claims or “tranches” against these collateral pools. Because of the way claims are prioritized, manufactured tranches can be far safer than the average asset in the underlying pool. The repackaging of risks and creation of “safe” assets by structured finance spurred a remarkable expansion in the issuance of structured securities. Investors snapped them up, believing them to be almost risk-free—a belief reinforced by the rating agencies. The recent financial market crisis has demonstrated that these securities are actually far riskier than originally thought.
In this paper, which is to be published in the Journal of Economic Perspectives, Coval, Stafford, and Jurek highlight two features of the structured finance machinery that fueled its growth. They also explain why ratings have failed to do their intended job and make some recommendations. The authors argue that ratings agencies were extraordinarily overconfident in their ability to estimate the underlying securities’ default risks, and how likely defaults were to be correlated. They present a modeling exercise using the same tools as the rating agencies to show that “even modest imprecision in estimating underlying risks is magnified disproportionately when securities are pooled and tranched.” This fact opened the path to economic disaster when combined with the growth in subprime mortgages from $96.8 billion in 1996 to approximately $600 billion in 2006.
The other critical and neglected feature of the securitization process is the way in which it substitutes risks that are largely diversifiable for risks that are highly systematic and linked to economic events. Securities produced by structured finance are therefore much more vulnerable to severe economic downturns than traditional corporate securities of equal rating. The rating agencies used standard models and there is no evidence that they deliberately issued inflated ratings. The problem is that the models that generate ratings make no distinction between the different sources of default risk and so are “particularly useless for determining prices and fair rates of compensation for these risks.”
The authors conclude that, henceforth, we would be wise to “eliminate any sanction of ratings as a guide to investment policy and capital requirements.” Rather than looking to ratings, we should focus on measuring and judging the system’s aggregate amount of leverage and to understand the exposures that financial institutions actually have.