As a kid, I loved the tales of Uncle Remus, where Br’er Rabbit always outsmarts Br’er Fox, famously convincing the Fox to throw him into the “briar patch”—his natural home.
Last week’s trading in the iShares Floating Rate Bond ETF (FLOT) reminded me a lot of that: a lot of brouhaha that ultimately just proved where FLOT really belongs.
What’s really going on? Let’s get into the briars:
First, let’s take a look at the chart that’s causing all the Twitter Chicken Little Sky Is Falling:
This plots the trading price of FLOT versus its end-of-day net asset value (NAV). A few things to note here. First, it historically trades at a slight premium. There are actually three independent reasons for this.
- FLOT has received pretty steady inflows all year long ($5.4 billion in new assets for the now $12 billion fund). Absent any other information, inflows mean you’d expect slight premiums in any ETF because it takes a little bit of buying pressure to push the price of an ETF over fair value. That’s what triggers the authorized participant community to do a creation, which we see as positive flows.
- As a bond fund, I suspect that the NAV is calculated on the bid. In other words, the price of each holding at the end of the day is determined by the fund accountant as “what can I sell this bond for?” Since the trading price of the ETF (especially if there’s positive flows) is based on actual activity, its closing price is much more likely to be at the midpoint of bids and asks, or at the “asks.” This created a “natural” perceived premium in almost all bond funds, all else being equal.
- FLOT owns illiquid securities, for which pricing is, at best, a guess.
Flow, Pricing & Underlying
The same three issues: Flows, pricing and underlying are potentially involved in understanding the trading of FLOT last week.
Let’s dive into this last point. By design, FLOT tracks an index made up of floating-rate bonds with maturities from one month to five years.
These aren’t the same as bank loans—another kind of floating-rate security generally issued by the junkiest of the junky issuers. These are primarily bonds rated “A” and above, issued by banks and some other giant conglomerates and international companies.
These floaters represent a very small portion of most companies’ debt, and can—optimistically—be said to trade by appointment. They are in demand by long-term holders (think big institutions) precisely because they offer some protection from rate increases.
But if you’re a big insurance company sitting on a position, you’re not likely to trade it out. You’re likely to just hold it until it matures in a few years.
Consider the trading activity in FLOT’s biggest holding, a Morgan Stanley floater that comes due in 2020:
No Big Deal
This is, as I said, the biggest—and we assume most liquid—position in the fund, and has changed hands just 655 times in 2018. So far in our very interesting December, it’s traded 30 times, for a total volume of 21,055 bonds traded at around par of $1,000, so around $21 million changing hands.
This is, to be blunt, not a lot in the grand scheme of things. Over the course of this year, about $1 billion has changed hands out of an issuance of some $3 billion for this particular note. It also often goes days with no trades at all. And this is the largest position in the fund. The least-liquid floaters in the portfolio haven’t even traded in the past week or so.
So it’s really no surprise when you see a disconnect between the “price” of these bonds and the price of the ETF that holds them. This potential disconnect is even bigger if there are any flows—which there have been. Here’s the premium/discount in concert with the actual flows into FLOT:
Here you can see the four days of negative flows that coincide with the perceived discount in the trading price of the ETF. Again, this is pretty normal for a fund holding less-liquid securities that has big outflows (in this case, a reaction to the action in the yield curve last week).
ETFs Working As Designed
What this isn’t, however, is a sign that somehow the ETF is broken. This is the ETF acting as price discovery for the market. The correct way to look at the chart above is that the action in the ETF—where there’s a lot of liquidity—acted to “set” the price of the underlying. And sure enough, the trading prices of the bonds came down in reaction.
But the other thing I feel obligated to point out is that this chart completely misrepresents the magnitude of what’s happening precisely because the value of FLOT versus its holdings has been so stable for so long.
At the worst of this, the trading in FLOT drove it to a whopping 0.60% below fair value.
Compare that to how other funds holding less liquid bonds fare on a regular basis:
This is the SPDR Bloomberg Barclays High Yield Bond ETF (JNK)—one of the bigger trading vehicles for junk bonds. Over the course of the last year, it’s swung from similar premiums and discounts in response to flows, and nobody loses their mind over it—investors just get that this is how this part of the market works.
Or consider the municipal bond ETF space and the VanEck Vectors High-Yield Municipal Index ETF (HYD):
Charts source: Bloomberg
Here a few days of redemptions drove the perceived discount to over 1%, and lo and behold, the ETF served—again—as price discovery, essentially “pulling” the price of the underlying bonds down to where the real trading was happening.
These moves in the less-liquid corners of the market are far from a crisis. Instead, these are precisely where these funds are designed to provide a real service to investors.
Or to put it in the sarcastic words of Br’er Rabbit: “Born and bred in the briar patch, Br’er Fox. Born and bred.”
Dave Nadig can be reached at [email protected]