Why Many Bond ETFs Now Trading At Discounts

Something strange is happening in bond ETFs.

Since the uptick in market volatility began, bond ETFs big and small have persistently traded at discounts to net asset value (NAV). (Read: "Understanding Premiums & Discounts.")

Some products have seen massive discounts; for example, the VanEck Vectors High-Yield Municipal Index ETF (HYD) traded a whopping 19.35% below NAV on March 12.

Other discounts have been more moderate, but no less striking. On Monday, the two largest bond ETFs on the market—the $72 billion iShares Core U.S. Aggregate Bond ETF (AGG) and the $51.5 billion Vanguard Total Bond Market ETF (BND)—traded at discounts of 4.43% and 6.17%, respectively. (Over the past year, AGG and BND usually trade at average premiums to NAV less than 0.10%.)

To understand better what's going on, we turned to legendary market maker Reggie Browne. With more than 30 years' experience in equity and derivatives trading, he has been making markets for ETFs since the vehicle's inception. As principal of one of the biggest market-making firms, GTS, nobody else has his fingers on the industry pulse quite like Browne. 

ETF.com: Why have we seen so many bond ETFs trading at discounts over the past two weeks?
Reggie Browne: It's a little complex, but I'll break it down for you.

First, let's start with liquidity in the marketplace. If you look at the top of the book of S&P 500 futures, and if you look at the Treasury futures, they're all abnormally wide and pretty thin. So I think there's less liquidity in the marketplace in general. That always happens in times of worldwide market stress. Liquidity just evaporates, because no one knows how best to price risk.

When fixed income ETFs trade at a 4% or 5% discount, it's because you're seeing stale prices in the underlying bonds, where the marketplace has fair valued them. The ETFs are giving real-time price action, where the bonds are actually trading. ETFs act as a forecast tool; they're based on some static benchmark, but trading at a discount in real time. That's also consistent for mutual funds—there's no free lunch.

ETF.com: So in times of stress, bond market pricing is more accurate for bond ETFs than for the bonds themselves?
Browne: Well, there isn't a lot of confidence from market makers right now that they can transact in the corporate bond space in size when they have to.

In 2018, the SEC did a study that showed only 20% of corporate bonds trade every day. But fair value agents, like IDC [Interactive Data Corporation], must come up with a fair value of the portfolio of corporate bonds. If those bonds don't trade every day, they still take a fair value, and a lot of times the prices are stale.

So current events are more a reflection on the fixed income market structure than it is on ETFs. If you look at what bond ETFs have done, they've brought more liquidity to the marketplace; they've driven spreads tighter to transact in large sizes. They've done their job. But the bond market structure isn't doing a good job of reflecting in real time where bonds are trading. ETFs are doing it for them.

ETF.com: Could you elaborate on that? What's wrong with the fixed income market structure as it is right now?
Browne: Let's compare [fixed income] to the equity market structure. If you look at how equities are traded, you have a national best bid/offer, you have top of the book, and you have affirmative obligation, where if you see the bid, you can hit it and transact.

In fixed income, you really don't have that. You have a bunch of dealers in a dealer network. There isn't a fixed income securities information processor (SIP), like you have in equity markets. You don't have anyone disseminating nationally and instantaneously the best bid/best offer and where the last trade occurred. You have issues around confidence about the ability to transact in corporate bonds.

There needs to be a discussion about the value of having a fixed income SIP, just as there is in equity markets. Some might say you have that in TRACE [FINRA's over-the-counter real-time price dissemination service, Trade Reporting and Compliance Engine], but even TRACE still has a lag in reporting corporate bonds.

What's needed is a national best bid and offer in corporate bonds, where the information is aggregated in one place and disseminated in real time. That would increase confidence.

ETF.com: How feasible would it be to apply what works in equities to the bond market, given that there are only 3,500 listed stocks in the U.S., but tens of thousands of individual corporate bonds?
Browne: There are 42,000 individual corporate bonds, and there are over 1 million different equity options and derivatives out there. If the option markets can find a way to disseminate prices efficiently, and the equities market can find a way to disseminate prices efficiently, then 42,000 CUSIPs shouldn't be a problem.

ETF.com: How would we go about doing this? What would the next steps look like?
Browne: There has to be an analysis done to see what's working well and what's not working well, and to root out which entrepreneurial themes people are trying to protect. Then I think they'll come to the conclusion that what the marketplace needs is more reliable, confident information in real time. It all comes down to raising confidence in the ability to transact.

There's already an SEC committee formed, FIMSAC [Fixed Income Market Structure Advisory Committee], that's looking into an analysis on market structure for fixed income. And there are some entrepreneurial enterprises, like MarketAxess and BondCliq that are trying to solve this problem, too. It's a problem that needs to be solved.

There's a lot of complexity to this. But the equity markets implemented positive changes successfully, through limit up-limit down. The next challenge is in terms of fixed income.

ETF.com: Any further thoughts on how bond ETFs have held up over the past two weeks?
Browne: On March 9, we celebrated the 30th anniversary of the global ETF industry, which started in Canada; ETFs were born out of the 1987 market crash. So you have 30 years of empirical evidence of how ETFs have operated across different types of market conditions. And by and large, ETFs have passed the test. Mission accomplished.

Contact Lara Crigger at [email protected]