You Can't Trust Bond ETFs for the Time Being

By Jim Wiandt

Add the bond discounts to the growing list of ETF debacles in the current market volatility.

An 8% daily discount to the NAV in the supposedly broadest bond index ETF (AGG) in the business? You've got to be kidding me. When it feels like you have no idea of what you are looking at in the markets, my inclination is to just stay away. Why would you use such a fund for safety when you've got no idea what's going on? Put it in an FDIC-insured savings account or a CD, maybe (zero-paying) Treasuries. But not in something that's trading all over the place like a lot of the bond ETFs are right now.

So the answer to Matt Hougan's "Can you trust bond ETFs?" question would appear to be "no" right now. 

In the current scenario, my money would be on the traditional fund structure that would actually have to liquidate the underlying to cover redemptions (though determining NAV for small bond holders would seem to be a similarly slippery proposition) than with the in-kind ETF structure ... but this would certainly be an interesting area to examine more closely. Neither scenario is pretty, is my guess, and that's worth looking at. How are the NAVs coming in on the large Vanguard funds, for example? And if they're giving the index NAV to redeemers, is that more accurate or is it in effect the existing shareholders subsidizing poor index pricing?

The most important point I want to make is that there is a slight mischaracterization of the bond ETF situation by Murray in his article, and Matt doesn't really address it directly in his blog. Murray mentions the possibility that maybe these funds should be traded like closed end funds, and bought at a discount. The closed-end analogy is an interesting one, but there's a very big difference.

In closed-end funds, there is a known underlying with known pricing, and if the fund is trading at a premium or discount reflects investor demand in the fund itself ... as redemption for the actual underlying is impossible. In the ETFs, the underlying index is not necessarily (I would say certainly is not, in these cases) reflecting the actual price of the underlying, and the ETF is effectively serving as the price discovery mechanism for essentially unknowable markets. My money is generally going to be with what the MARKET is saying on pricing.

In the case of these bond ETFs, my suspicion is that the ETF is actually closer to the actual NAV (whatever that is) than the underlying NAV price of the index is showing. It feels a bit like walking around in the dark ... but just like in the Malaysia country iShare back in the day, the ETF is really the only game in town in terms of price discovery mechanism. That doesn't say that it's still not a bit of a blind guess as to what the value of the bonds are. It's really hard to know that, and that makes me nervous.

So I think that these ETFs are interesting and valuable as a price discovery mechanism. But that doesn't mean I'm putting my money there. When I'm buying bonds, I'm buying a safe haven, and if I have little idea of what's going on, that to me is not a safe haven.

The other interesting aspect to this story is the SEC compliance issue with Barclays buying the Lehman indexes (see Eric Rosenbaum's outstanding article discussing the situation). My understanding has been that the SEC was unlikely to give BGI an exception allowing the same company to run the indexes that its ETFs are tracking because of potential conflict issues. I don't know if they'll be able to work out some scenario where there's a firewall/separate entity. Anyone who has an update on the status of that, I'd be interested in hearing... we'll certainly be looking into that.

Comments

[Indexor:]

Jim: iShares Muni Bond ETF (MUB) had the same problem. It appears the iShares fixed-income department is broken and may even be signaling that ETFs are not the right vehicle for fixed-income products. They have never proven liquid enough to trade more than a few hundred shares without moving the price, sometimes substantially. Of course recently, the spreads were not only a joke but the inability to sell positions even at the current bid was impossible. The discount to NAV problem was never addressed by iShares, it was simply abandoned. MUB lost 3% in the final two hours of trading one day last week while the NAV saw little change.

Sadly, after a negative experience with the iShares High Yield Bond ETF in July of '07, and now both AGG and MUB as well as BND, I'm done with fixed-income ETFs. I stopped using all Closed-End Funds several years ago for the exact same reasons. And If equity ETFs don't stop loaning out shares and start replicating their indices, there will no longer be any reason to use them going forward either.

What a sad way to destroy a fast growing industry which offered so much promise.

2008 Oct 15 12:06 PM

[oldman:]

What can you trust besides cash?

2008 Oct 15 02:14 PM

[Rhunzzz:]

Sooooo... back to open-end funds for fixed income? What a shame. Maybe that's why, according to the Richelsons, ETF managers aren't eating their own cooking.

2008 Oct 18 12:34 AM