Par-based investors in the triple-A CLO market may soon have their sense of security tested. Many still apparently cling to the assumption that their bonds will ultimately be repaid at par, based on the senior position of these bonds in the transaction structure and the corresponding protection from potential losses.
Our controversial view is that the secondary triple-A CLO market has been priced to an 'alternative' reality for quite some time. And we believe this pricing does not accurately reflect the chance that cumulative trading losses and defaults, coupled with significantly discounted net asset values, may challenge a par repayment of the triple-A tranche despite all the subordination.
Why is the market apparently ignoring this inherent risk? First and foremost, many of the original par-based investors cannot or will not quantitatively assess their triple-A CLOs based on any scenario other than repayment at par. Despite their exhaustive due diligence, these investors purchased the majority of the 2007 triple-A CLO paper at discount margins between 22 and 50 basis points over Libor, assuming that given 30% subordination (on average), they could accept this asymmetric payoff. A condition of their acceptance was the expectation that these triple-A CLO assets would be repaid at par and would consequently exhibit minimal price volatility. In fact, the primary risk these investors expected to contend with was not credit or market risk, but rather how quickly or slowly their cash flow would be returned based on the risk of optional redemption.
But what if triple-A CLO investors are repaid less than par? Does the current discount margin of 225 basis points as of August 2008 really offer value and compensation for these risks, even with it standing ten times greater than it did a year ago? In short, we believe the answer is no. The current triple-A CLO discount margin is an irrelevant measure of value, if par is not the final value to be received. A more realistic approach is to model the risk of receiving less than par repayment based on the actual portfolio liquidation net asset value at the point of greatest distress, simultaneously incorporating the expected losses from trading and default. In making the decision to buy, hold or sell, the investor must be willing to accept the possibility of less than par repayment.
CLO transactions are governed by very specific covenants and rules that are incorporated into the assigned ratings at issuance. However, these covenants, rules and corresponding ratings are based upon certain assumptions related to defaults, trading losses and, of course, the ultimate net asset value at the time of amortization or optional redemption of the CLO.
By way of example, let us look at a hypothetical CLO that is approaching the end of its reinvestment period, with loan losses from trading and defaults representing 15% of the transaction's original par value. If we assume the remaining assets are priced at some level materially below par (say 75% to 80% of par), we can quickly grasp the magnitude of the problem that confronts both the CLO collateral manager and the triple-A CLO investor. With 15% of the transaction's subordination gone, and the balance of the collateral valued at 20% to 25% below par, it is not difficult to see how the triple-A CLO tranche could ultimately be exposed to a less than par repayment.
It is estimated that more than $160 billion of CLOs were issued in 2007, and roughly 70% or $112 billion of that total was composed of triple-A tranches. However, if only a fraction of the 2007 triple-A issuance were to come to the market looking for bids, there would be few natural buyers capable of supporting the supply and preventing the price from declining to levels that would challenge the expected arbitrage relationships between the assets and liabilities of the CLO structure.
The catalyst to look out for is forced CLO unwinds because of managers' inability to limit the necessary selling that occurs when a CLO deal runs into trouble (i.e. when coverage and basket limitation tests are triggered). The quality of the loans sold and the timing of those sales will determine the amount of prepayments the triple-A investors will receive. And the quality of the underlying CLO portfolio will influence whether such an event will lead to less than par repayment at the triple-A CLO level, forcing existing par-based investors to accept the reality that losses are in fact likely.
However, investors who can endure significant mark-to-market volatility (expected from the exogenous risk that secondary supply might dwarf current demand) and are capable of quantitatively valuing the risk(s) of less than par repayment, may stand to benefit handsomely from deeply discounted purchases.
David Ardini is a portfolio manager for Franklin's Floating Rate Debt Group. He leads the management of the group's institutional CLO structured vehicles and institutional separate accounts, totaling approximately $1.8 billion in AUM.
Alex Yu is a research analyst/portfolio manager for Franklin's Floating Rate Debt Group.
Sam Peng is a quantitative research analyst for Franklin Advisers and its Floating Rate Debt Group.
The foregoing reflects the analysis and opinions of the authors as of September 2008. This material may help you understand the authors' investment management philosophy, but may not be relied upon as investment advice or an offer for a particular security. Statements of fact are from sources considered reliable, but no representation or warranty is made as to their completeness or accuracy.
(c) 2008 Asset Securitization Report and SourceMedia, Inc. All Rights Reserved.