Why iShares Gold Trust ETF (IAU) Was On the Shelf by Dave Nadig, FactSet
Friday was one of those days where understanding ETF structure can be the difference between yawning over a press release or worrying that the sky is falling.
Here was the official notice from Blackrock about the iShares Gold Trust (IAU):
“iShares Delaware Trust Sponsor LLC, in its capacity as the sponsor of iShares Gold Trust (IAU), has temporarily suspended the creation of new shares of IAU until additional shares are registered with the Securities and Exchange Commission (SEC).”
What happened after that press release hit the newswire at 9:30 a.m. ET is pretty predictable. Reporters started calling around trying to understand if it was a big deal, and gold bugs used it as an opportunity to point out that, somehow, Gold ETFs were deceiving the public. This commenter on the Wall Street Journal website was typical:
“No one ‘forgot’ to register for new shares,(sic) there just simply is not the physical gold available to buy to match demand for those ETF shares. But, to admit this fact to the public would be to cause a run on physical gold …”
To explain how misguided this conjecture is, let’s dig into what actually happened (and, by the way, first thing Monday morning March 7, the fund was back open for creations.)
iShares Gold Trust – Shelf Filings
The first thing to understand is that the finance industry uses the initials “ETF” to describe a whole swath of different products structures. The vast majority of ETFs are actually regular old 1940 act registered mutual funds when you peel back the layers of regulation, and the 1940 act protects investors, as is well understood, and allows for the continuous issuance of new shares to anyone who wants them.
But there are lots of ETFs that aren’t built on that regulatory framework. The largest ETF in the world, the SPDR S&P 500, isn’t – it’s actually a Unit Investment Trust, which gives it a few unique quirks like not being able to do daily dividend reinvestment. Any ETF that’s tracking futures markets, like the huge U.S. Oil Fund (OIL), is actually a commodities pool, regulated more by the CFTC than the SEC. Giant volatility funds like the iPath S&P 500 VIX Short-Term Futures ETN are actually just bank debt that trades on an exchange.
And any ETF holding physical assets, like silver or gold, is organized as a “grantor trust.”
The grantor trust structure works beautifully for precious metals, as it lays out, very clearly, exactly how and what the fund can do on behalf of the trust holders (the shareholders of GLD). But one of the quirks of the structure is that, unlike a mutual fund, it has to register each share before it can be sold. When the iShares Gold Trust trust was initially formed, iShares registered 50 million shares, which, with the NAV struck at around $40, meant the firm expected about $2 billion of interest in the product back in 2004. Since then, iShares has occasionally registered additional shares, and in 2010, also split the shares 10 for 1 to drop the price from $120 down to $12.
Most recently, in March of 2015, iShares registered an additional 120 million shares. But when the first eight weeks of 2016 hit, and investors wanted $1.3 billion in new shares of IAU out of nowhere, iShares was left with not enough shares to issue.
But Why?
Why not register a billion shares and never have to worry about it? Because nothing is ever free. Registering shares costs real money. The filing last March for 120 million new shares cost Blackrock over $150,000 (figures are included with every filing). The filing made this morning for an additional 300 million shares cost over $300,000 just in SEC fees. The more shares they register, the more money gets paid to the SEC. It’s a balancing act, and one we’ve seen off and on in all of these types of products over the years.
Some folks saw this shortage hiccup and immediately flooded social media with hyperbolic headlines about how this was somehow about not being able to buy enough gold. That’s just not true either. See, iShares Gold Trust (or GLD for that matter) don’t actually buy gold at all – they receive gold from APs who create new shares. The reason they can’t create new shares is entirely about paperwork, not about whether anyone’s willing to sell APs London bars. And of course, people are always willing to sell at some price – that’s why we have a market to set the price of gold in the first place. So sure, maybe the demand for gold will continue to be crazy, and that will do what it always does when demand spikes – drive up prices. But the structure of the ETF won’t have anything to do with that.
And while it would be easy to point fingers, it’s worth noting that the impact on investors is actually nil right now. Theoretically, when it’s impossible for authorized participants to create new shares, and there’s investor demand, the price of an ETF can trade to a significant premium. We’ve seen it many times, such as when the Egyptian stock market closed and Van Eck was forced to stop creating new shares of the Market Vectors Egypt ETF (EGPT).
In that case, EGPT traded to a near 25% premium over the supposed fair value of the Egyptian stocks it held. But in a case like this, the market knows with complete certainty that new shares of IAU will be available within days, and so beyond a very minor reaction as the press release came out, iShares Gold Trust traded very close to fair value, and tracked the actual price of gold all day:
The above chart is all relative – so when gold was up 70bps on the day, there was a brief period where IAU was 20 basis points higher. Here it is presented as just the premium, like in the EGPT chart I led with.
Really, intraday tracking like that is about as good as you can expect on any ETF, much less one with a minor hiccup on the day. For comparison (and literally at random), here’s how the last ETF I wrote about, the First Trust Dorsey Wright Focus 4 ETF (FV), looked on Friday:
Nobody panicked because FV traded 20 bps high or low, and nobody should panic about iShares Gold Trust trading 20bps high or low. Would it have been less confusing if Blackrock had registered more shares weeks ago? Sure, I suppose. But is it a big deal? Hardly.
(It’s also worth pointing out that despite some media reports that the fund failed to track a 1% rise in gold yesterday, that’s not true either. At 4 p.m. ET, the difference between the price of NY spot gold and the price of IAU was less than the penny spread, in percentage terms. Gold can be a confusing market, with two London-based fixes a day, a widely quoted futures contract, and near continuous trading around the globe.)
Talk to Dave Nadig in our daily ETF briefing. Call: 1-800-721-6994, ID: 1383726 at Noon Eastern each trading day.
Receive stories like this to your inbox as they are published. Subscribe by e-mail and follow @FactSet on Twitter. If you are looking to source FactSet data or analytics in your publication, email [email protected].
