In 2010 you first presented the idea of an overnight gold fund, citing your calculations that a $100 million hedge fund, starting in 2001 and going long gold on the PM to AM fix and short on the AM to PM fix, would be worth 2.6 billion dollars today. Talk to us about how you discovered this trade and what you have found out.
Sam Kirtley: The idea was spawned by a fair bit of noise about possible manipulation (and if not manipulation than at least discrepancies) in the way that gold trades in London and then in Asia. We sort of took these discrepancies to the next step; obviously as a trading opperation we're always looking to profit from things like this, so we looked at how we could design a trading strategy to profit on the back of it.
Given the pattern in which gold trades, large corrections tend to be concentrated within the London period between 10:30 and 3pm when those AM-PM fixes are made. During the Asia session and later in the US session, gold quite often tends to tick up. There were various hypotheses around that, but we were mainly concerned with how it is playing out in a practical sense.
You have a different perspective as you are in the Asia market region - the PM to AM fix, as opposed to your readers who are in the AM to PM fix.
Sam Kirtley: Absolutely. Most of our client base is in the US. I suppose it does give you a different perspective, being out here in New Zealand; the 24-hour trading environment is imposed on you because things are happening through the middle of the night in the US which you have to be aware of. And so you start to take note of how patterns change from various sectors when gold is traded in various areas of the world.
There's an organization called GATA [6] that has been putting forward for a long time that there is a bit of subtle manipulation going on. I'm not sure if we fully got along with it but there have been cases that have been settled - one by Moore Capital last year on "banging the close," which involves accumulating a position during the day and dumping it at the end of the day in order to manipulate a fix. Moore Capital had to settle that for $25 million, so that's evidence that there is some manipulation. [But] as I said, we are more concerned with the practical implications.
How has this trade changed as the gold trading environment has evolved over the last year?
Sam Kirtley: What really has startled us - and we were astounded by the returns in Augsust 2010 - is that not only has the discrepancy increased, but it's arguably increasing. As you mentioned a $100 million fund starting in 2001 doing the strategy would be worth 2.6 billion [2010]. Well if that fund had continued on for the last year and a half or so, it would now be worth $5.2 billion. So that's an increase of 143% in just over a year, which is quite startling really.
Initially, we thought if you are somebody trading gold at home and you don't want to hold a position overnight and you've just been long gold during the day in the UK, you would have lost money over 10 years. How could that be possible in an asset that has gone up 500%? That was when the alarm bells started ringing - the trigger for the investigation.
In terms of the rate of return, the annualized rate was about 37% a year from 2001. Now, the annualized rate is about 43%, which is quite an increase from just an extra year and a half of data. We would have thought if anything, [the rate] would dissipate as gold becomes more actively traded and the trading is increasing in the Asian regions with the Shanghai exhange, but apparently not - apparently the discrepancy continues to persist and persist more vigorously.
If you zoom in on the chart and look at that period over the summer where gold really rocketed upward, did that drive a lot of those increasing returns?
Sam Kirtley: If you look at the first six months of 2011, our short-intraday long-overnight index slightly outperformed gold. During the rally that we had from July through August it kept paced with gold - it matched it in terms of returns. But the stellar point of the strategy is when gold had that vicious correction, this is when the fund really started to take off. Indexing from 100 at the beginning of 2011, we were up to 140 when gold was picking up. But when gold dropped back to beingup about 20% on the year, the long overnight gold index was up something like 70% year-to-date. So when gold corrected, investors using this strategy would have sheltered and actually profited quite handsomely from it.
[This is] because you are short gold during a period - the middle of the London session and the first part of the New York trading session - when a lot of the damage was done. That, to us, is something that stands out as a really good part of the strategy: it not only works when gold is flying high but when it corrects severely as well.
You would think something like this would be mitigated through arbitrage quickly, but that doesn't seem to be the case here.
Sam Kirtley: I'm not sure why it hasn't happened. I think part of it is because gold is still a relatively tiny market in terms of the financial market as a whole, so maybe people aren't looking at it as much. And these returns are fantastic, but we most point out they are not including expenses - obviously you are going to have to cross a spread each day. But even crossing a spread each day, it's still an increase of 1700% since 2001. As I said, it's not something that works in theory but not in practice. Even if you put the practicalities in, it's still quite startling.
I wouldn't say it's a substitute for investors that might be long gold for various reasons. It's a compliment to a trading strategy. Even if one doesn't have any interest in gold as an investment, it could be an added traded strategy or something to chuck into a portfolio to add some alpha and a bit of diversity.
Are you thinking about setting up a fund to trade this strategy?
Sam Kirtley: Yes. We've had a great amount of interest in the fund. Initially, we were looking at going the hedge fund route, but there's been such a significant interest from the retail side that we are probably going to go with another vehicle.
How would you achieve exposure in the fund?
Sam Kirtley: Two ways we are looking at going about it. One is the simple, obvious route of switching the fund's exposure by futures twice a day, trying to target wherever gold happens to fix.
The other is setting up a tailored swap agreement with a major financial institution which would give us long exposure during the overnight session and short exposure during the day.With the swap agreement, you have the sort of counterparty risk that might deter people from investing in the fund, especially if a lot of the people interested in the gold space are perhaps concerned about the financial well being of the system and might not want to take that on.
Have you looked for this trade in other assets?
Sam Kirtley: We did a bit of an investigation into silver, which had reasonable results - but not as convincing as gold. I think the key when you are looking at these asset classes is that you need something that is the same wherever it is traded.
So, the commodity space is something that comes up as an obvious target. When you look at some other commodity products, like oil, one has to bear in mind that there is more than one type of oil, depending on where it is traded in the world.
Gold is exactly the same wherever it is traded. So, that is the standout candidate. We are looking at backtesting this strategy going back to the late 1970s to see how it performs in a bear market as well.
Do you have any thoughts as to why this opportunity might be present?
Sam Kirtley: I don't think there is anything as clear cut as what happened in April with Moore Capital. One bit of interest was late last year when the Swiss National Bank pegged the franc to the euro at a 1.20 managed float. We were quite startled because just before the announcement, there was a $50 smash in the gold markets.
Five minutes later, there was the announcement that the SNB was going to peg. We were a bit perplexed because you had a takedown of $50 in the gold market just 5 minutes prior to what we would consider a bullish announcement for gold.
If the Swiss franc, which had been a very safe haven, was now pegged to the euro at 1.20, you just eliminated one potential safe havens - the US dollar, yen, and gold - but we got the takedown.We're trying not to get hung up on the hypotheses that have been put forward and just focusing on refining this as a trading strategy. We could speculate all day on the reasons for it, but I don't think we are going to find a really definitive explanation.
