Trading

Go long junior mezz and short equity, says Morgan Stanley

Tuesday, August 18, 2009

In a research report entitled Positioning for a decline in default correlation, Morgan Stanley says that the rise in correlation during the credit crisis resulted in outperformance of equity tranches and outperformance of senior and super senior tranches relative to delta. Today, however, the best opportunities lie in the reverse of these trades, according to the report.

Although normalisation of market conditions should result in falls in equity – the same movement as in the first correlation market upheaval of May 2005 -  these tranches are a tough short as they are highly negatively convex for protection buyers, says the report.

Morgan Stanley suggests going long junior mezzanine tranches and short equity, especially in series nine of the CDX index, given its higher concentration of “tail” names.


<< END >>
Comment by: Daniel Schiemert. Posted 2 years ago

That is an interesting trade idea. But the spread dispersion could easily turn out to break the deal idea. The impact of correlation vs spread dispersion can be seen on the senior tranches, ( roughly i.e. increase the spread dispersion of the portfolio (keeping average spread constant) and the correlation goes up). And the equity tranche is similarly sensitive to small spread dispersion effects. Thus it might be interesting to consider to set a junior mezz tranche versus a more senior mezz tranche and try to catch the expected correlation move on them, rather than playing the equity versus a jun mezz. You will get the correlation effect more isolated. I see spread dispersion ( especially in copula models) one of the main drivers for problems in pricing when correlation is high.

Recent bond & loan issuance

>>More information from the Issuer Tracker

CFlux secondary 
CLO index levels:

Index
21 May
CFlux USD AAA  ↑ 96.2
CFlux USD AA  ↑

88.3

CFlux USD A  ↓ 84.1
CFlux USD BBB  ↓ 75.3
CFlux USD BB  ↓

74.1

CFlux USD EQ  ↑ 77.5

 

>> More information & historical data