Nadig: Anatomy Of A Monster ETF Trade

How BZF traded 1,000 times its normal volume, and didn’t skip a beat.

I clearly need a new hobby, because I love looking at charts that show strange ETF-trading anomalies.

And there’s not much of a bigger anomaly than a seemingly illiquid ETF that suddenly trades 27 million shares without a hitch.

But more than just being interesting, digging into how these monster trades work can help us understand not only how to better trade ETFs, but can lead to interesting insights into what’s going on in the market.

First, let’s look at the ETF in question, the WisdomTree Brazilian Real ETF (BZF | B-91).

As you can see, this is an ETF that simply doesn’t trade much.

While it’s had its moments in the sun, it’s not uncommon for BZF to trade just a few thousand shares in a day. But look at the two spikes on the volume chart, one in October 2013, the other in January 2014. And you can see another one—albeit a smaller one—in 2011.

When I see something like this, it’s usually pretty obvious that this is a big player getting in or out of a position. If it was a mass frenzy, it’s unlikely we’d see such a clean jump, without any real ramp up or down in volume on similar days. So the first step in understanding the trade is to see who owns it.

In a larger ETF, that kind of forensic analysis is pretty much impossible.

So many people own shares of, say, the iShares MSCI Emerging Markets ETF (EEM | B-100), and it trades so much, that hundreds of millions of dollars can change hands and be lost in the mix.

Here, however, a quick look at 13F regulatory filings made with the Securities and Exchange Commission shows who our likely trader is:

13F filings are made by large investment managers to report their significant positions four times a year.

This list shows the top 10 big holders of BZF as of the end of 2013, and clearly Windhaven is the only holder large enough to have been making a trade this size. Depending on whom you use to track 13F filings—we’re using Bloomberg here—you can often get historical positions as well.

Again, Windhaven is obviously the big player both in the most recent trades and probably in the blip we saw in volume in 2011.

13F filings are problematic, however. Again, in a big ETF, so many shares are held by major custodians on behalf of their customers that movements of individual big players can be tough to discern.

In that case, really all you can try to gather is whether the big trade represents buying or selling. The tape doesn’t always tell us that, since, of course, every trade has both a buyer and a seller.

But we can look at the changes in the shares outstanding for ETFs to get a sense of buying pressure or selling pressure.

In this case, we just looked at the ETF.com Fund Flows tool and saw pretty clearly that the October trade was a buy, and the January trade was a sell.

With a few clicks, we can discern the exact investment play being made here: Windhaven made a big bet on Brazil’s currency—the real—in October when the fund was trading around $18.33. Whether unhappy with the trade, or because whatever they were hedging against no longer needed the hedge, they executed in January.

But how they make such an outside trade is the other half of the story. The first stop for me as I seek to unravel the mystery is to look at the intraday chart for the day of the big trade:

But wait! Nowhere in this intraday chart is there anything remotely like 27 million shares trading. The biggest block was 63,000 shares—a decent trade, but not a barnburner.

To really find this trade, you have to look deep into the nerdy weeds of trading: the complete time-and-sales for the day. This is available on most trading platforms, but here’s where you can find the trade on Bloomberg’s version:

There are a few things to note here.

The first is that the trade was printed after the market close. That’s why it didn’t show up on the intraday chart. The second thing to note is that it’s quoted to four decimal places of precision. That means this wasn’t a single trade, but has been derivatively priced.

That means it’s a combination of trading activity that resulted in a single aggregated price reported to the customer—and the market—as a block after the market closed. That’s what the “FT” means here as well, it’s a “Form-T” trade that’s done aftermarket through FINRA’s trade reporting facility.

The other thing of interest is the price—$17.0259 is substantially better than the actual price the ETF traded at most of the day. In fact, if you look back at the intraday chart, BZF only traded at more than $17 right at the beginning of the day.

It would be easy to think something weird is afoot here, but the story is probably much simpler.

I called Dave Abner, head of Capital Markets at WisdomTree, to talk about trades like this. He wouldn’t talk about this specific trade, for obvious reasons, but he was forthright about how the process works in a vacuum.

“Typically, big investors will let WisdomTree know that a trade like this is going to happen, so we can help give them guidance on the best way to execute the trade,” Abner explained.

And clearly that’s probably what happened here. And most likely, that call didn’t just happen with WisdomTree, it happened with the actual broker on the phone too. Let’s just guess it was Goldman Sachs, but it could have been any number of liquidity providers.

On that call, all the players developed a plan of attack.

“We consider ourselves to be mechanics experts,” says Abner. “We think we know more about how our funds work than most others in the industry. So every big trade, we generally have the conversation … not just on the way in, but on the way out, too.”

Again, on that call, Goldman (we’re guessing) and Windhaven developed a plan of attack for the trade.

It could have been as simple as: “Well, we know we’re doing a big redemption at the end of the day, so we’ll sell a pile of real nondeliverable forwards (NDFs) all day long knowing we’ll get them from WisdomTree when they trade in the shares.”

Or, they may be getting cash from WisdomTree, and simply need to lock in a price, which they do by hedging with the underlying stocks in the fund. It’s even possible they sold some shares in the open market throughout the day, and saved some for a redemption.

At the end of the day, the counterparty to Windhaven’s enormous trade (Goldman, or whoever it was), was acting as an authorized participant and presented the shares to WisdomTree and received either cash or NDFs in return.

Of course, it’s a crappy day to be at WisdomTree when this happens.

Nobody likes to see $500 million in assets walk out the door on an ETF that collects a 45-basis-point expense ratio totaling $2.25 million a year. But Abner was sanguine about that.

“If people work with WisdomTree’s Capital Markets team, they are going to have a good experience,” he boasts. “If they have a good experience, they’ll be back.”

This really highlights the beauty of the ETF structure the way few other trades I’ve chased down the rabbit hole can.

Again, BZF is an ETF that barely trades. But the underlying market—the currency market for the real—trades more than $60 billion a day, and that’s not including over-the-counter swaps. In that context, Windhaven’s $500 million bet doesn’t even move the needle.

And that’s why all the players in this story can shove an elephant through a knothole, and dominate BZF trading.

At the time this article was written, the author held no positions in the security mentioned. Contact Dave Nadig at [email protected].