(Corrects paragraph two "writing" from "buying.")
By Jane Baird
LONDON, Oct 4 (Reuters) - Buyers of the top AAA tranches of collateralized debt obligations (CDOs) were not paid enough in the past for the true risk they took: that of an economic catastrophe bond, according to a Harvard Business School study.
Insurance companies, banks, pension funds and other big investors in these senior slices of CDOs could have got a lot more for their money by writing out-of-the-money puts on the U.S. S&P 500 index, another way to bet against economic catastrophe, Harvard professors Joshua Coval and Erik Stafford and doctoral student Jakub Jurek found.
What's more, they would not have been exposed to the liquidity crunch that has hit all CDO products since mid-July.
These investors now may demand higher spreads, which calls into question whether the economic model for cash CDOs works.
A CDO is a portfolio of credit risks that is divided into tranches. At the bottom, the riskiest tranche is exposed to the first few percent of losses from any credit in the portfolio. Default risk diminishes up the ladder to the top tranches, which lose after the others have gone under and so get AAA ratings.
Banks used CDOs to help sell billions of dollars of exposure to U.S. subprime mortgages. When subprime defaults rose, investors lost confidence in all CDOs as an asset class.
"Investors, perhaps for the wrong reasons, are now less interested in these products," Stafford said. "They think the ratings agencies got the probabilities wrong. We think that even if they got the probabilities (of default) right, they (CDOs) were totally mispriced."
Investors who bought the senior tranches relied on credit ratings to determine prices, he said. During the boom in the credit markets, banks packaged CDOs such that AAA-rated tranches would pay a few basis points more than AAA-rated bonds.
"A naive application of the law of one price says that a triple-A security should have the same yield (as any other triple-A security)," Stafford said.
"They (CDO arrangers) added value by giving you this yield advantage relative to the wrong benchmark," he said. "With the right benchmark, you are leaving a lot on the table for the risk you are bearing."
An out-of-the-money put, however, may not be an alternative for investors such as insurers and pension funds, whose mandates do not allow them to buy such instruments.
Vincent Matsui, an analyst with Fitch Ratings, argued investors in the top CDO tranches do have the technology and understanding of the different types of risk.
"All those institutions do apply very sophisticated analysis," he said. "At the end of the day, they, just like everyone else, are subject to the forces of supply and demand."
Before the crisis, the top tranches of investment-grade CDOs paid 12 to 15 basis points over risk-free rates. The AAA-rated tranches of collateralized loan obligations (CLOs), made up of junk-rated loans, paid about 25 basis points.
By contrast, the bottom equity tranches paid annual returns of 20 percent or more to the CDO arrangers that kept them or the hedge funds that bought them.
People in the market often call equity tranches "toxic waste", but Stafford said: "From an asset-pricing perspective, that's actually the best stuff to have."
At the bottom, the chance of a loss is high, but most of the time it is unrelated to the overall economy and risk is more diversified. "One or two companies defaulting is pretty random."
The risk of a loss at the top tranches, however, is purely a function of the economy and is, therefore, less diversified.
"As you move up (the CDO tranche structure), the fraction of the total risk that becomes economic becomes greater," he said. "By the time you're at the top (of an investment-grade corporate CDO), we are in states of the world that you haven't seen before, worse than the Depression."
Events over the summer showed that the market might indeed have been mispricing catastrophe risk, or systemic risk, versus the risk of individual defaults, or idiosyncratic risk.
Worries the credit crisis would hurt the global economy pushed up spreads, while the number of actual defaults remained at record lows. Within CDOs, pricing shifted as prices of the top tranches dropped, and spreads jumped to 40 basis points on investment-grade CDOs and as much as 100 basis points on CLOs.
The Harvard group has not yet updated its model with the latest pricing data but Stafford said current spreads for the senior tranches may be more in line with what they should be.
CDO arrangers, analysts and some investors say the top tranches are now the cheapest and are starting to attract buyers. Over time, they say, spreads on these tranches are likely to tighten again relative to lower tranches.
But Stafford says investors may not be willing to accept such low spreads again. And wider spreads for the senior tranches reduce the share of the pie available for the lower tranches in cash CDOs.
Stafford said this could limit the comeback of the product.
"If they (investors in senior tranches) step back and say that's not nearly enough and actually demand what I'm arguing they're owed, then it's not clear that the equity guys are going to be as interested in doing this," Stafford said.