Goldman, in a brilliant bit of legerdemain, invested (along with partners such as CV Starr and Perry Partners) $3 billion into its troubled quant fund Global Equity Opportunities. From its press release:
Many funds employing quantitative strategies are currently under pressure as recent conditions have resulted in significant market dislocation. Across most sectors, there has been an increase in overlapping trades, a surge in volatility and an increase in correlations. These factors have combined to challenge many of the trading algorithms used in quantitative strategies. We believe the current values that the market is assigning to the assets underlying various funds represent a discount that is not supported by the fundamentals.
Within its alternative asset platform, Goldman Sachs Asset Management manages Global Equity Opportunities (GEO), an equity long/short quantitative strategy fund. It had a net asset value of approximately $3.6 billion before the equity investment. Given the market dislocation, the performance of GEO has suffered significantly. Our response has been to reduce risk and leverage.
In addition, Goldman Sachs and various investors, including C.V. Starr & Co., Inc., Perry Capital LLC and Eli Broad, are making a $3 billion equity investment in GEO. We consider this an attractive investment opportunity. Existing investors in the fund will also have the opportunity to participate. The investment will also provide the fund with more flexibility to take advantage of the opportunities we believe exist in current market conditions.
Feliix Salmon argues that Goldman correctly sees a good investing opportunity. I’m not so sanguine (although some folks at Goldman may indeed see upside) and view this as yet another instance of Wall Street putting lipstick on a pig.
It’s easy to understand the tradeoff for Goldman. Its larger quant fund, Global Alpha, is down 26% for the year and the GEO fund was rumored to be in even worse shape, and it turns out it lost 28% this month. Bear Stearn’s hedge fund debacle is a stark warning of what happens when these funds go pear shaped: you sustain considerable damage to your reputation, stock price, and cost of funding (its CDS prices spiked up considerably) AND you still have to bail the hedge funds out.
By taking an anticipatory move, before investors have the opportunity to request to withdraw funds, Goldman signals confidence and hopefully stems a run on the fund. It’s almost a given that this was cheaper than a bailout even before you factor in the market disruption that the workout of this fund would have caused (what sexual and other favors were extended to induce to CV Starr and the other partners to go along is beyond me, although it’s possible that they had reason to want to avert further market turmoil).
FT’s Alphaville points out that this begs the question of what happens to Global Alpha:
$3bn? And this isn’t the fund that most of the concerned chatter has been concerning – that was the Global Alpha fund, reported to be 26 per cent down for the year. The North American Equity Opportunities fund has also been reported to be in trouble. The bank on Monday acknowledge that life at those two funds is also no picnic, but added: “at their current levels of equity capital, we believe the funds are positioned to actively pursue market opportunities.”
As to whether this is a good opportunity to invest in quant approaches, it doesn’t seem to be a sign of intelligence to double your bets in a failed strategy (this is a classic error of a panicked trader) particularly if you don’t have a clear grasp of why things went wrong and you therefore have not revised your methodology. Remember, Lehman’s Matthew Rothman declared what happened last week to be a “one in every 10,000 year event” and in the next breath said it happened three days running! By definition, facts have just proven out that what we have been through is patently not a one in 10,000 year event and perhaps the modelers need to go back to their computers and do some more study and revision before trading more on what are clearly erroneous assessments of probabilities.
The famed Greenspan put has conditioned investors to expect quick rebounds. Major Wall Street firms were buying subprime mortgage brokers as late as February of this year, believing the subprime stress to be short-lived. Yet it took the markets nearly a year to recover from the LTCM debacle. Bets that we will soon have markets as usual look premature.
You are correct that events have proved it is not a one in 10000 years event.
This people are mathematically savvy, but do not seem to understand that the difference between religion and science, is that in science the the truth and abilities of the rules (the models) is legislated by data.
This goes back to a point you made about fat tails.
Brad DeLong agrees with you
http://delong.typepad.com/sdj/2007/08/the-subprime-me.html
DeLong:
“Our strategy is fine. We were just hit by a sixteen-standard-deviation event.” “Then it didn’t happen: the universe isn’t old enough for even one sixteen-standard-deviation event to have ever happened.”
Tails are fat.
are these guys so incredibly smart that they’ve skipped Trading 101, i.e., ‘don’t throw good money after bad’
would suspect that the timing and the pseudo-confidence effort is a ploy to bolster confidence and stem redemptions, now that redemption season is upon us…interesting that money didn’t go into global alpha, their INTERNAL fund.
will pass on a quote from an old market hand (wish it were my own!):
“This is first crisis where the risks are so complicated by derivatives that if the truth be known …no-one knows what they have !!!….
What makes it more scary is all these derivative desks are staffed by PHD’s and who have never seen a market in panic mode before and were still smoking weed at Cambridge during 97-98…all flapping like fish out of water
However…for us old Jurrassic Park guys that have been in the mkt 25 years plus … we got the T shirt … been there…seen it….done it …A welcome return to way markets used to be ……a walk in the park
for us … !”
one in 10000 year event?! more like once every five. would you put money into a fund that didn’t have a risk model for that?