Analyzing Galleon’s returns

By Felix Salmon
October 22, 2009
The Pragmatic Capitalist gets his hands on Galleon's monthly returns, and finds them very suspicious:

" data-share-img="" data-share="twitter,facebook,linkedin,reddit,google" data-share-count="true">

The Pragmatic Capitalist gets his hands on Galleon’s monthly returns, and finds them very suspicious:

These guys just couldn’t lose. Whether the market was up or down they cranked out 25% returns like they were printing money. It makes you wonder just how long these guys were trading on insider information?

I have run the risk adjusted returns on hundreds if not thousands of portfolios throughout my career and I have never seen numbers like these. NEVER. There is virtually ZERO downside volatility in these figures… Gauging from the returns I would be willing to bet the insider trading was going on for most of Galleon’s existence and was likely much more rampant than currently reported.

I’m not completely convinced, for two reasons. Firstly, Galleon’s returns were pretty volatile: the fund was up 14.53% in May 1997, for instance, and then down 8.54% five months later, in October of the same year. That doesn’t seem abnormally consistent to me.

More generally, if you want returns which rarely turn negative, insider trading is not your strategy. Madoff-style outright fraud works, of course: you just report fictional returns instead of real ones. But Galleon isn’t being accused of making up its returns: it has the money it says it has. It just (allegedly) used illegal means to amass it.

A sophisticated insider trader isn’t trying very hard not to lose money. Quite the opposite, in fact: he’s putting a lot of money at risk, and knows there’s a significant downside. He just also knows that he’s got an informational edge which gives him an advantage over the rest of the market. Such advantages don’t always pay out, and as a result you’d expect an insider-trading strategy to show relatively frequent negative returns. Even if, over the long term, it was very successful.



We welcome comments that advance the story through relevant opinion, anecdotes, links and data. If you see a comment that you believe is irrelevant or inappropriate, you can flag it to our editors by using the report abuse links. Views expressed in the comments do not represent those of Reuters. For more information on our comment policy, see

I’ve been cranking through Galleon’s 13f’s since 2005. Just the long side and just the reported numbers at the end of each quarter. But I can say with certitude that Galleon was not getting alpha on the stocks it held over a few quarters. As best I can tell they just broke even on Google since 2005. Did OK on Apple. But most names they broke even or lost. The crushingly obvious explanation for Raj’s insider trading was that it was very very difficult to make positive returns the old fashioned way in the last few years.

Posted by Drew Robertson | Report as abusive

Excellent points Felix. What I find so hard to believe is that this fund, which had a relatively high standard deviation, only experienced 5 months of 4%+ negative vol in the 190 months I have as data.

That is just a mind blowing stat. Even the very best hedge funds (rentech, SAC) are much more volatile than this. It either means Raj was a genius or his big insider trading bets were substantially boosting his overall returns. This guy was shooting fish in a barrel based on his numbers….I just find it hard to believe that anyone is this consistently good.

I should add – it not his positive vol that you should focus on. That should be expected in a fund that is essentially cheating. But for him to have almost zero negative vol is truly remarkable. It raises some serious red flags.

*Such advantages don’t always pay out, and as a result you’d expect an insider-trading strategy to show relatively frequent negative returns.*

I’d like to understand this better (unless maybe we’ve just got wildly different ideas as to what counts as “relatively frequent”). The whole idea of inside information in past scandals has been that it usually pays out. Seems to me that assuming reliable inside informants, the main problem would be generating negative returns often enough so as not to draw undue attention to oneself.

Let’s say, for example, that Mr. Inside Trader is getting tipped off by someone (could be a lawyer, an admin, an IT guy, does not matter) at a law firm that their client is closing on a deal, soon to be announced, to acquire publicly traded ABC Co.

Mr. Inside Trader runs out and snaps up all the shares and all the calls that he reasonably can of ABC. Now here are all the likely outcomes I can come up with:

- ABC Co. acquired (whether in this deal or in some bidding war that may break out subsequently). Price of ABC goes up, Inside Trader wins.
- ABC Co. not acquired, nothing comes of the rumor. Possibly the information was bogus in first place, possibly the deal collapsed and information never became public. Price flat, basically a wash for Inside Trader (transaction costs).
- ABC Co. not acquired. Price falls, either because Mr. Inside Trader bought in after others had (based on hope, general attractiveness as target, same rumor or whatever) or for some other unrelated reason. Inside Trader loses.
- ABC Co. not acquired. Price goes up anyway for some random other reason. Inside Trader wins.

Posted by SelenesMom | Report as abusive

OK, so the difference between Galleon and Madoff is that Madoff just out-&-out lied on his statements & audits while Galleon used information garnered illegally.

It is clear that outsiders (including several commenters above) can spot a highly likelihood of illegal behavior–Ponzi or insider trading–by looking at readily available public statements.

Why can’t the SEC? And, to the extent commentators/analysts/bloggers find this informaiton out, why don’t they routinely (a) publish it and/or (b) report it to the SEC?

This info would certainly have been MUCH more helpful to investors before Galleon got the perp walk.

Posted by Lilguy | Report as abusive

maybe publish the monthly so we can see and judge. his tech fund was down 28 in 2008 and down 19 in 2002. down 9 in 2001. so the premise is wrong about how never loses etc. as can see from size of alleged insider trades, they were small % of what doing otherwise they would have lot more on the calls and lot more indictments. likely there are more, but not everything was inside otherwise wouldnt have lost big in some years. and the contention that no on has his numbers is just wrong. look at buffet/soros/sac/some tiger cubs/paulson/arbs…… seems that starting with a bias and then seeing numbers to back that up.

Posted by impartial | Report as abusive

It is been a pretty public secret that these guys are running on insider information for longer than this fund has been in existence. Trying to use numbers to prove it is questionable.

Posted by bob goodwin | Report as abusive

I haven’t looked at Galleon’s returns at all, but I did have dinner last night with a fairly well-connected tech investor and he said it’s been widely held for years that Galleon was trading on inside information.

Oh come on, if it was “widely held” for “years” then either all sorts of people would have piggybacked on their trades, or someone (such as disgruntled ex-employee, ex-girlfriend or boyfriend, etc.) would have snitched, or both. More likely everyone and his uncle now wants to look in the know but not so in the know as to be dragged into the slimy mess.

Posted by SelenesMom | Report as abusive

No, I can second thoughtbasket. The rumors about Galleon were persistent and loud. One of the problems with piggybacking insider info is asymmetric information. When insider info goes bad, it goes very bad and can turn a years worth of legit gains into negatives tout de suite. To be assured of making money, you would first have to certain that Galleon was putting on an “insider” trade and not a “normal” position; 2) you would need to see and trade the entire book of Galleon “insider” trades to make certain that the good “insider” trades beat the bad “insider” trades in aggregate. In practice, Galleon would be spreading the trade(s) around to mitigate piggybacks. I bet Galleon has at least 3 primes and a HUGE broker list.

Posted by WSJevons | Report as abusive