Just when investors jumped for joy when the news spread when George Soros (or, more exactly, his funds) was buying InterOil -- Soros alone took the share offering at $34 a couple of years ago -- some of them are now quite disappointed when the news came out that he has sold a significant part of his holdings.
We've never fallen for this particular investment trap. Not when he bought, even less now that he (partly) sold, and neither should you. Here is why.
The statistical nature of returns
Since InterOil's (NYSE:IOC) stock price actually went up considerably when Soros was selling, there are other funds that were buying. Now, the crux is this. Soros is undoubtedly one of the top three or top five investors of the last three decades. But how much of a guarantee is there that he is significantly more right about any single investment on any given point in time, compared to less stellar funds?
You have to realize that we're talking here about average returns comprising many hundreds, if not thousands of securities over many years. Soros, the seller, might, on average, be better than the next fund (the supposed buyer), but that says next to nothing about the chances of him being more right about any individual investment on any particular point in time. That is simply statistically insignificant.
Say Soros generated 22% return on average for the last 30 years (the actual figure is in that order, perhaps even a bit higher, a truly remarkable achievement). Say the buyer fund only generated a 15% average yearly return, significantly worse (although still quite good). Realize that although Soros' average yearly return is a lot better, this doesn't guarantee that he will be better each and every year. Even Soros might have a bad year, and the other fund a good one, outperforming Soros.
However, when we move to individual securities on particular points in time, the fact that Soros average performance is considerably better is no guarantee that he will be better on every security at every point in time. His average performance is statistically insignificant for any single security over a short period of time (or even a long period).
You have to realize that the returns on individual securities can vary wildly. Even the likes of Soros will have a considerable number of losses in individual securities and lesser funds will have numerous big winners. Only Soros will have slightly less disastrous investments and slightly more stellar ones.
When you realize that we're talking about hundreds, if not thousands of securities it will become clear that the differences between Soros predictive powers for any single security, compared those of lesser funds about the same security, are statistically insignificant.
Let us give an even more telling example. We riled against Whitney Tilson for his short case against InterOil. In our view, his case against InterOil is insufficiently substantiated (and we're really being polite here). Others point out that Tilson has a terrible track record with his shorts recently. Some even poke fun at this, and there is an anti-Tilson long fund that takes long positions in Tilson's shorts, and is doing very nicely.
The underlying argument here seems to be something like, "you see, Tilson has such a lousy record, he can't be right on InterOil." As much as we disagree with Tilson's arguments against InterOil (or rather, the lack of these), we're not falling for this line of attack. Yes, Tilson seems suffer particularly bad returns lately, especially with his short positions.
But perceived over longer-term, he doesn't have a bad track record, so he's going through something of a rough patch at the moment. Even that can reverse to the mean (stuff usually does). More significantly, his track record, let alone his short-term track record, doesn't hold much, if any, predictive value for any particular security at any time. Even if he had a really bad track record over 20 years or so, he will have had several big winners (although in this extreme case, it could start to matter statistically in some moderate way).
So unless one is consistently very bad, compared to a fund that performs consistently very good, there is little if any predictive value to be distilled from following the buying (and even less of selling) of one fund over the other on any single security. So we are ad odds with Tilson over his arguments about InterOil, his track record doesn't come into it, for us.
Selling is even less predictive than buying
While funds generally buy securities for one reason (the expectation of a rising share price, we'll leave dividends out of the equation here), selling might have multiple reasons, not all of these necessarily related to the prospects of the security in question. It's not uncommon even for insiders to sell when the company is on the cusp of achieving greatness.
It could be that the holder needs money, that his fund faced redemptions, that it follows a top-down investment approach looking at sectors and the macro investment climate (and Soros is certainly pessimistic about the latter).
In Soros case, he's closing his funds for the public, and we don't know what kind of effect this has on the size and composition of his portfolio. He could just be taking profits. It is also worthwhile, perhaps, to keep in mind that Soros is an old guy, with nothing to prove in the investment world. His reputation as an investor is well and truly cemented. He's also a multi-billionaire, at this age, money isn't likely to be his prime motivator anymore (and indeed, Soros has taken up a host of other, more public causes). The fund is mostly managed by his sons.
The truth is, when Soros was selling the stock price actually went up significantly and whether Soros sold because he is more likely to be right about InterOil than the buyers, or that he's likely to have more of an inside track compared to other big funds that didn't sell, we only know with hindsight. But we know that statistically, we can't tell. What's more, we can't even tell whether he sold for any reason pertaining to the prospects of the company, or for any other reason.
Should you sell?
Well, Soros shares found eager buyers, the shares are still up despite his selling. It's not hard to see why, within a couple of months, some if it perhaps quite a bit sooner, we will know:
- Formal re-approval of the new plans by the PNG government
- Who will be the operator of the planned LNG plants
- What part of Elk/Antelope, the resource, will be sold to whom and for what price (in all likelihood, quite a bit higher than today's valuation for the shares)
- Whether InterOil will partner to explore the vast exploration area (nearly 4 million acres with 54 leads and 12 possible reefs), with whom and at what price
- Most promising, the well they're drilling now at Bwata/Triceratops. This is a known gasfield (a well successfully demonstrated this in 1959) and recent seismics has both identified a reef and shown that the field is much larger than previously thought. Drilling could hit the top of the pay zone by the end of the month on most optimistic scenario.
We would stick around for the drilling at Triceratops/Bwata. The chance of something big is pretty good, there is gas, and there is likely to be a reef, the kind of reservoir that makes Antelope so productive.
Disclosure: I am long IOC.