The transactions described in Part I centered around issuers seeking to improve their capital ratios without exposing potential investors to the typical risks of an investment in capital securities. The first example discussed in Part II involves a transaction in which a senior unsecured creditor achieves a preferred position against other senior unsecured creditors, without taking collateral from the borrower.
Kangaroo Bond
The earliest incarnations of this idea involved a USD100 bond issued at par, but containing an unusual provision: if the issuer suffers a credit event, a second bond with an identical face amount must be issued to the investor free of charge. Once again, the apparent pointlessness is explained by noting that the transaction increases the investor's recovery against other creditors of equal seniority. If the recovery for general unsecured creditors is 50%, this particular investor will recover 100%. Since the investor will hold two senior claims aggregating USD200.
As described above, the issuance of the second bond would run afoul of essential bankruptcy principles in most jurisdictions, since the delivery of the second bond would represent a transfer of value from the issuer to a creditor for no consideration, in violation of typical "preference" and "fraudulent transfer" statutes such as Sections 547 and 548 of the U.S. bankruptcy code. Attempts to alleviate these concerns have centered on the issuance of the second bond on the date of purchase of the original bond--but with the second bond being held in escrow until a default occurs under the first. Under this variation, it is argued, the second bond is in effect issued at the same time as the borrower receives consideration from the lender, supposedly nullifying the objections raised under the preference and fraudulent transfer statutes.
Lawyers in most sophisticated jurisdictions have had difficulty signing off on any of these structures because a transfer of value from the borrower to a creditor is occurring after the date of bankruptcy. For these structures to work properly, it seems necessary, at a minimum, that all transfers of value from the borrower to the creditor should occur unambiguously on the same date as the original funding.
The structure that has perhaps come closest to overcoming these legal objections works as follows. Assuming an expected recovery value in default of 50% for senior unsecured claims, the borrower would issue a USD100 note to the investor at inception, in return for receipt of USD100 in cash. The borrower would simultaneously sell to the investor credit protection on itself, through a credit-default swap, in an aggregate face amount of USD200. The terms of the credit-default swap provide for cash settlement. These two transactions are depicted below.
Since both transactions are entered on the initial funding date, attempts to challenge the CDS transaction under the relevant bankruptcy statutes are more problematic. For the investor, this means that it holds under the CDS a claim of USD100 against the borrower upon a credit event; this claim, if recognized at the senior unsecured level, would be worth USD50. When added to the USD50 at which the note itself is trading, the investor's recovery reaches USD100. We note that the enforceability of the additional claim under the CDS has not been tested in court.
The lowering of risk for the investor means presumably that it should be willing to finance the borrower, under the note, at a favorable rate compared to other senior creditors. Assuming, for example that (i) the borrower's ordinary funding cost at the senior unsecured level is LIBOR plus 300 basis points, and (ii) the investor is prepared to provide financing at LIBOR plus 150bps on account of its reduced risk, the fee under the CDS would be set at 75bps annually, leaving the investor's net spread over LIBOR at (300 minus 2*75) 150bps).
It should be noted above that for senior unsecured recovery levels other than 50%, the transaction results in either a shortfall or windfall for the investor. At a recovery level of 40%, for example, the bond would recover USD40, while the swap would recover USD48 (= USD200 * 60% * 40%), leaving the investor with a shortfall of USD12. By the same token, if the recovery value is 80%, the investor receives USD80 under the bond, and USD32 under the swap (= USD200 * 20% * 80%), for an aggregate of USD112. Simple calculus reveals that the optimal scenario for the investor would be a recovery level of 75%, in which case her aggregate proceeds would reach USD112.50.
The structure above enhances the investor's position by increasing the size of the claim in bankruptcy from USD100 to USD200, assuming a 50% recovery. The question must be asked once again: should the company's financial statements disclose this contingent senior claim at inception, to alert readers that the senior claims against the company will increase by a sizeable amount if the borrower finds itself in a bankruptcy? A devious borrower could argue that this liability arises only if it enters bankruptcy and could conveniently adopt the position that its own bankruptcy is too remote to constitute a material liability that warrants disclosure.
Even in systems that require the marking-to-market of derivatives, and assuming the CDS is recognized as a derivative for this purpose, the MTM liability initially recognized on the balance sheet on account of the CDS would be a fraction of the face value. Assuming, for example that the company's credit is trading at 300bps in the regular CDS market, and that the CDS between the company and the investor has a maturity of three years, the MTM would equal at most the present value of the annual difference between the 75bps the company is earning under this CDS, versus the 300bps at which the credit trades in the regular market. This comes to around USD6 relative to a note face value of USD100. Of course, as the company's credit deteriorates and its price in the CDS market increases, the amount of this MTM increases.
Vanishing Debt
One final transaction works in a fundamentally different way from all those described previously. Rather than increasing the rights of the creditor in a bankruptcy, the transaction in question achieves the opposite: it removes that creditor altogether from the list of claimants once a company approaches bankruptcy. Consider for example a company rated AA, with senior unsecured debt trading at LIBOR plus 50bps. The company issues to the investor USD100 of five-year debt, with a coupon of LIBOR plus 200bps, but containing a clause which stipulates that if the company's credit rating falls to BB or lower, the investor's note is written down to zero.
For the investor, the note represents a simple trade-off calculation: the additional 150bps of yield versus the probability of losing everything if there is a major downgrade. The basic analysis presumably would center on the rating agencies' credit migration tables, which would enable the investor to gauge the probability of this downgrade. At a sufficient yield enhancement versus regular senior debt, the note will find buyers. From a valuation perspective, the note can be viewed as a regular senior note, plus a short position in a credit downgrade option, with the premium for the option reflected in the coupon enhancement.
For the issuer, the structure achieves a form of automatic deleveraging when its financial situation becomes precarious. When the credit rating is spiraling it is able to wipe out a chunk of its outstanding debt, slowing down or even reversing the decline in its rating. In many ways this represents a simpler form of tax-deductible equity than the complex schemes that have gained favor in the U.S. and Europe.
The most interesting questions arise in connection with the trading of the company's regular debt in the secondary market. Suppose the company's regular debt begins life also with a rating of AA. With each piece of bad news concerning the company's financial condition, the rating of the regular debt as well as that of the vanishing debt declines, resulting in a decline in the price of both instruments, but with the vanishing debt presumably declining more than the regular debt. Assuming, however, that the vanishing debt represents a sufficiently material portion of the company's total liabilities, there must surely come a point at which a further downgrade of the credit rating would be favorably viewed by the regular creditors: it will lead to the extinguishment of the vanishing debt and thus to a strengthening of the company's financial condition, possibly even leading to an upgrade of the regular debt. The question is whether traders of the vanishing debt would bid up the price of the regular debt as the company's rating approached the extinguishment level. Indeed, it is possible to imagine that in some circumstances, the holders of the regular debt would be bombarding the rating agency with calls arguing that the latest financial results justify an immediate downgrade.
This week's Learning Curve was written by Oussama Nasr, a senior consultant with DNA Training & Consulting in Beirut. He can be reached at oussamanasr@aol.com
