Bankers and investors putting on debt versus equity trades are drawing on the idea first expounded by Nobel economics laureate Robert Merton that equity can be thought of as a call option on the assets of a firm. If the share price dips below a certain level – implying a lower value on the firm’s assets and cashflows relative to its liabilities – default will follow. Bondholders, meanwhile, have sold an equity put option to the issuer and the spread on a corporate bond is the premium for taking that position.
KMV has popularized this theory over the past 10 years with its model providing – for a fee – an estimated default frequency (EDF) measure for corporates based on its proprietary database of historical default data. So the market was already attuned to this idea when CreditGrades – by Riskmetrics, with the backing of Goldman Sachs, JPMorgan and Deutsche Bank – was launched in June.
According to Jorge Mina, head of risk modelling at Riskmetrics, CreditGrades improve on KMV both in terms of being more transparent and more user-friendly. “The big difference is that certain parts of KMV, such as the historical database of default probabilities, are not market observable,” he says. “We’re not using historical default data but equity price volatility and balance sheet information.”
A flexible friend Users can log onto the site and use the CreditGrades model to calculate theoretical credit default swap spreads or the spread over Libor on a five-year bond for 10,000 companies publicly traded in Europe, the US and Asia. By plotting a company’s CreditGrade against its current actual credit default swap spread, investors can identify potential arbitrage situations. If the theoretical and actual default protection prices have been closely correlated for six months and then suddenly diverge, for example, alarm bells should ring.
Users can play around with different variables, depending on their view of what is going to happen to a company’s securities. Say that firm A’s stock price is at e10, its equity volatility is 40% and CreditGrades returns an indicative price of 150 basis points on its credit swap. Default swap protection for A is currently trading at 300bp, suggesting that the credit market and the equity markets have different views of A’s prospects.
There’s a potential trade to be done on the back of this disparity. An investor that spotted the disparity and decided to put on a long debt position by selling protection in the default swap market might then perform a sensitivity analysis to determine an equity delta.
The investor wouldn’t then just sit back and wait. If A’s equity price fell to e9 or e8 or e7, how would that affect the credit spread? Likewise, if the spread tightened by 10, 20 or 50bp, how would the equity market respond? “The model does not calculate the hedge ratio for you,” says John Tierney, head of credit derivatives research and strategy at Deutsche Bank in New York. “What it does is calculates what change in equity price would correlate to what change in bond spread.”
Earlier this year, for example, Spanish oil company Repsol’s bond spreads widened significantly because of fears about a potential default by its Argentine arm, YPF. But though the bond market was plunged into chaos by this potential credit event, with spreads widening by almost 90 basis points in a day, the equity market behaved as if the risk was immaterial and share prices stayed relatively stable.
There’s treasure everywhere Users of CreditGrades would have seen Repsol’s bond spreads suddenly gap out, while the theoretical default swap price remained the same, suggesting that expectations had got out of line. If an investor believed that this was only a temporary blip in Repsol’s credit prices and that default was unlikely, a possible trade might be considered. This would involve going long the bonds – where tightening is technically most likely to happen as this is where the widening took place – and short the equity. The investor could either sell shares short or buy equity put options. Sure enough, debt prices rallied. If default had occurred the short equity position would have insulated the investor from at least some of the losses.
The behaviour of Vivendi securities has also opened up arbitrage opportunities this year. From early January to early February, for example, Vivendi 1.25% 04 convertibles were trading 100bp below their theoretical spread. The same thing happened in the first three weeks of March. At the end of each of these periods there was an opportunity to make money as the theoretical and actual spreads converged. In this case investors would have gone long the credit and put on a short equity position – based on the delta implied by their own model.
But this isn’t an easy game for investors to play. The trouble with putting a model like CreditGrades on a publicly available website is that it creates the impression that capital structure arbitrage is just a question of feeding in a set of numbers and coming up with a failsafe trading idea. “A model is simply a tool. It tells you a company’s securities may be mispriced but you still have to do the fundamental analysis yourself,” says Doug Mallach, head of capital structure arbitrage marketing at Merrill Lynch in New York.
Even bankers closely involved in putting the model together say that, at best, it provides a rough and ready approximation of a debt-equity relationship. “There’s lots of basis risk involved in this,” says one. “It’s not like trading swaps versus treasuries. You’re dealing with two very different markets.”
