Tuesday, November 18, 2014
HOW HIGH WILL GOLD MINERS GO?: the answer will shock you!
EDIT: Ha ha, as of 11:30 AM Wednesday they all got they asses smacked down. Oh well - at this very moment GDXJ and GLD still haven't violated their EMA(10)s, so I guess we can just puzzle out whether this is a buy point or not.
So gold miners have popped really nicely. I feel kinda stupid at selling some stuff today after just a quick 10% gain, but still have a quarter of my money in B2Gold and intend to ride Clive into the sunset. I really wanted to have some cash back in the kitty, since I'm going to be busy tomorrow.
But how far back can the miners pop?
Once I take a position, I'd like to puzzle out how far it might go. I don't like to guess, and I don't want to sell early and leave a possible 50% win on the table - especially when I've been waiting all year for the chance to put this trade on.
So here are some slightly more involved charts than usual for you that might suggest a freaky answer.
First, here's $HUI:
Firstly, the chart will remind you that $HUI recently dropped from 240 all the way to 150. Yes, that was in the space of two months.
Well, now $HUI has taken back 30% of that fall and we shouldn't expect much more, right? In fact I can just imagine how many hobbyist TAs are logged into Blogger right now writing a post that calls this week's move a "pending 38% Fibonacci retracement". They're looking at that SMA-50 (blue line) as the sensible terminus for what would be a very short and vigourous bear rally.
I guess it makes sense for goldbug TAs to feel so hopeless. And so look at that SMA(50), and the ol' confirmation bias sets in.
But look at that other black line in the chart: that's the chart of $USD.
US dollar really started to pop at the beginning of September. When that happened, gold dropped. As Mickey Fulp recently and article-of-the-year-usefully pointed out, the gold move was -0.96 correlated with the USD move, which means absolutely nothing else was behind the drop.
But as I've kept screaming from the wasteland, gold never broke down ex-US.
But Americans saw gold break down in their currency.
And so you got this:
Man-oh-man that's a sexy looking chart, no?
$HUI perfectly tracked gold's US-denominated breakdown.
OK, not perfectly: the gold miners were actually resilient in July and August, before finally pooping a lung with an early-October divergence that ignored the tiny gold rebound, followed by a late-October utter collapse.
But gold never broke down ex-US.
What really happened, which all them white-ass fat honky Manhattan crackers like Ritholtz and Josh Brown never saw because they have their heads up the USA's ass and don't believe in the existence of any other country, was this:
Gold ex-US is back where it was just after the September breakdown. And as I said earlier today, USD might have moved just a bit too much and could maybe be due for a slight retracement.
If that happens, the GLD:UDN line could soon end up right back where it was in August.
And if it does, I won't be surprised to see $HUI get an awful lot closer to 240 than anyone's expecting. Because $HUI was at 240 for two and a half months with GLD:UDN where it was.
But why did $HUI drop so low to begin with? What caused sellers to swamp out buyers?
Betcha I know. Try this:
It dropped so low because all the white-ass Manhattan honkeys were looking for a pair trade to play the USD move with.
Well, as everyone with a Bloomberg byline has been parroting, gold "moves inversely to the US dollar". In reality it doesn't, it moves depending on a heck of a lot of different inputs - but over the past two months it has moved almost exactly inverse to the USD.
Spookily so, in fact.
I'd really like to see Mickey do more statistical work on gold and the US dollar, maybe broken down into quarters, to see how much the gold-USD correlation can historically vary from one quarter to the next. Because I really doubt that -0.96 is the norm.
It seems obvious to me that there is no reason whatsoever that any thing should correlate with another thing to the tune of -0.96 unless either it's mechanically connected, or it's a finely tuned derivative.
So what I think happened is this:
Some people probably decided to play the US dollar pop by going long-USD-short-gold. It makes extra sense to take that position, as the end of QE was also expected to cause a gold price drop.
But if a lot of people went long-USD-short-gold, then that pressure would swamp out all other inputs to the gold price-setting market mechanism, and the gold-USD correlation would approach -1 over that timeframe.
So I betcha the correlation only got that high because of the pair trade.
But then other people realized the gold miners suck more than gold does, so they went long-USD-short-GDX instead for the extra kick.
And probably some really insane wackjobs went one further and went long-USD-short-GDXJ.
And all that shorting of GDX and GDXJ would explain a lot of these companies really getting smashed down to zero. For those of you who read IKN's paid newsletter, I think it explains that big gold miner's collapse (the one he sold a week ago, ahem) and maybe even that big silver miner's collapse.
Problem is, GDX and GDXJ are really tiny little things compared to the potential population wanting to put on a long-USD pair trade. So their selling could have swamped everything else happening in the gold stocks.
And now, if USD drifts downward for a bit, they're going to have to close out those positions. Which means buying back all the gold miner stocks they just spent two months selling.
I'm not saying it's going to happen this way; but I am saying that $HUI dropped far more than it should have compared to gold ex-US, and I think the pair trade is a mechanism explaining why.
And I'm saying that I won't be surprised if $HUI makes it back to 220 at least.
Barring, of course, an OpEx smackdown: but the problem there is, last I've read, everybody has already gone short gold. Who's left to puke gold options into thin bids on the overnight Globex?
So, like I said: I'm not counting on it, but I'm going to sit and hold these positions til I can figure out some way that the above scenario might be wrong. Because this thing might just have legs.
Wait, what's that you just said?
Tax loss selling?
I betcha tax loss selling already happened when everyone got presented with a 40% paper loss in 2 months this October.