|
PureBytes Links
Trading Reference Links
|
This is a long post....
Alot of big institutions place large bond spread trades between different
bonds of various credit rating. Alot of these traders look at historical
spreads on their Bloomberg terminals and when the spread blows out they go
long the spread. This decision is based on the view that the spread is at
historical highs and there is a low probability that it will keep blowing
out and that there is a high probability that the spread will contract back
to more normal levels.
These spread trades do not require alot of "rocket science". Quite the
contrary.
These are very basic spread trades in physical bonds and/or the swap curve.
In the bond world, these spread trades are "volatility hedged" so the
traders end up only making or losing money on the outright spread level.
With the global credit squeeze taking place in the last 1-2 months, the
spreads between federal Treasury government bond curves and other bond
issuers have blown out to new historical highs and the spreads are still
blowing out.
I used to trade these spreads in my old institutional days. The other day I
spoke to the guys
I used to work with and they actually have exactly the same kind of spread
trades on right now in the Australian bond markets. Alot of the
institutional traders have unloaded out of their portfolios their federal
government Treasury bonds and have "switched" or bought semi-government,
corporate or mortgage backed bonds instead. Alot of these positions were
put on 1 to 2 months ago. The spreads have blown out so the institutions or
"instos" have lost money on these spread trades. These portfolio managers
will just end up underperforming against their performance benchmarks coz
there was no leverage on these trades.
As an example of bond spreads blowing out, I bought a Korean Developement
Bank FRN (Bond) in August 96 and it was trading at the Australian
equivalent of LIBOR + 22pts. I later resigned from this institution in Feb
97. By Aug 97 it blew out and was trading at LIBOR + 80pts. By XMAS 97 it
was trading at LIBOR + 350pts and now it is currently trading at LIBOR +
950 pts. This institution is still holding this bond. Please note this is
an extreme example because I am talking about a Korean corporate bond
versus the Australian equivalent of US dollar LIBOR, but you can see the
concept here.
In the case of LTCM, they have just put on similar spread trades but with
many, many times the leverage. Spreads have blown out and LTCM have just
blown up!!
Regardless of the rescue of LTCM, they WILL HAVE TO UNWIND A PART OR ALL OF
THIS POSITION.
In Australia I am hearing 3 stories in regards to the spread trade that
LTCM put on. The first story was they went long Danish Bonds and short
German Bunds. The second story was they went long the British swap curve
and short Long gilts. The third story is they were long US mortgage backed
securities and short US Treasuries.
I think this is just a rumourmongering plot to ensure that nobody can
"guesstimate" which spread trade they actually have put on. All of these
spreads have blown out.
You never know, the above could be totally wrong and incorrectly reported
by the media. The spread trade could even be a basic yield curve flattening
or steepening trade in 1 or 2 different countries. You know, long T-Bonds,
short Eurodollars but volatility matched. No "rocket science" here.
Regardless of which spread position LTCM has, you, I and the bank dealers
out there will NEVER EVER know which spread position it is until it is
unwound. Could you imagine if George Soros ligitimately said, " Hey
everyone, I am about to unwind my $100 billion long US dollar short
Canadian Dollar position in the next few days!!". The entire world would
frontrun the position! It is in the interest of the Fed, LTCM and the banks
backing it that nobody knows the position until the exposure on this spread
trade has been unwound. The unwinding of a position like this will take a
couple of weeks to ensure that nobody can sniff out the real position. And
you can bet that the real outright position won't just be unwound "naked
with your pants down". They'll spread one side of the position with a
couple of different bond classes or swaps and then they'll unwind the other
side with other types of bonds. This will ensure that no-one can guess the
real underlying position. Every time LTCM will ask market makers for prices
in different bonds, those price makers will be analysing every single
market price that LTCM has asked for. The sharks will be circling.
|