U.S. Markets close in 5 hrs 27 mins

Bond Mutual Funds: Worse Than Bond ETFs?

Matt Hougan

There’s been a lot of talk recently about the fact that bond ETFs can trade to large discounts during periods of market distress. The reporting has been histrionic, arguing largely that ETFs aren’t a good vehicle for gaining bond exposure. The implication is that mutual funds are a better choice.

The irony of these arguments is that, by their own logic, they suggest the reverse; that bond mutual funds actually have bigger problems than bond ETFs.

The truth is, neither vehicle is perfect.

Bond markets are illiquid, and investing in illiquid markets has a cost—if you want to sell something illiquid during periods of market distress, you’re going to get less than you’d like. What’s important to realize is that the way you pay the piper for buying illiquid assets is different depending on whether you invest in a mutual fund or an ETF.

How Bond Funds (ETFs And Mutual Funds) Calculate NAV

The first step in understanding the issue here is to understand how net asset values are calculated for both ETFs and mutual funds.

The place to start is with equity investments. For equity mutual funds or equity ETFs, NAV calculations are simple. Take an ETF that holds all 500 stocks in the S'P 500 Index. To calculate the NAV, all you have to do, essentially, is look at the closing price for all the stocks in the S'P 500.

For bond ETFs—and particularly for bond ETFs tracking illiquid corners of the fixed-income market like municipal bonds—it’s trickier. There’s no official exchange for municipal bonds, and no consolidated tape, so there is no one agreed-upon price for what the value of any particular bond may be. Many muni bonds don’t even trade on a daily basis. They may go days, weeks or months between trades.

To calculate the NAV for a fund like the iShares National AMT-Free Muni Bond ETF (MUB), iShares relies primarily on bond pricing services. These prices are “provided directly from one or more broker-dealers, market makers, or independent third-party pricing services which may use matrix pricing and valuation models to derive values.”

In other words, it is the job of the bond pricing services to guess what a given bond is worth at the end of each day. They base those guesses on reported trades, surveys of bond trading desks, instant messenger chatter, or derivative pricing models that reference other areas of the market to estimate the value of individual bonds. They're smart, but they’re still just guesses.


How And Why Bond ETFs Trade To A Discount

The issue that’s come up repeatedly—and which the media has reported on repeatedly—is that, during periods of market stress, bond ETFs tend to trade at large discounts to their stated NAVs. On June 21, for instance, MUB closed at a 2.73 percent discount to its stated NAV.

Why does this happen? Let’s take a simplified example and see.

Imagine that you have a municipal bond ETF, and the entire portfolio is composed of one bond. On any given day, the NAV of the ETF is equal to the price of that bond.

Most of the time, the price of the ETF will be very close to the NAV. But let’s imagine that, during a period of market stress—like the one that happened on June 21, a disconnect occurs:The ETF closes the day at $95 but its reported NAV is $100.

The ETF industry argues that the NAV here is “wrong,” and that the bond pricing services are overestimating the true clearing price of the underlying bonds. Their argument is built on the principle of arbitrage.

The reason ETFs tend to trade close to their NAVs is that a group of institutional investors called “authorized participants” (APs) are able to create or destroy shares of an ETF at NAV on a daily basis.

If the price of an ETF dips below its NAV, APs can buy shares of the ETF on the open market, trade them in to the ETF issuer, and be paid the full value of the NAV. In this case, APs could go into the market, buy 100,000 shares of the ETF trading at $95/share, and trade them in to the ETF issuer for $100 each, minus fees.

The catch is that the ETF issuer will typically “redeem out” those shares “in-kind.” In other words, the APs won’t get $100 in cash for each share of the ETF they trade in, but rather, $100 in the securities that the ETFs hold (based on the bond pricing service’s valuation). To complete the transaction, the APs have to liquidate the bonds once they get them from the ETF issuer.

The reason the bond ETF trades down to $95 when its reported NAV is $100 is that APs don’t think they can actually liquidate the bonds for $100.

The truth is, during illiquid markets, they may not know what price they could really get for the bonds, so they err on the side of caution. Most likely, the true “value” of the bonds if they had to be liquidated that day lies between $95 and $100. APs are, after all, in the business of buying low and selling high, and they will keep the price of the ETF as low as they can—subject to competition from other APs—when there’s uncertainty.



What This Means For ETF Investors

What this means for ETF investors is pretty clear:During periods when investors want out of a particular illiquid ETF, they will likely receive less than the stated NAV. (Incidentally, it works in reverse:When everyone wanted in to MUB, it traded at a premium.)

It almost doesn’t matter if, as iShares and others in the ETF industry suggest—and I largely agree with this—that ETFs are the “true” market for bonds and the live prices generated by the ETF are closer to the true clearing price of the bonds than the matrix-calculated NAV.

During periods of market stress, spreads will widen on the underlying bonds, pricing will become uncertain and you will get less than NAV when you trade.  You should know that going in.

What It Means For Mutual Fund Investors

The presumption underlying most articles, however,  is that investors should ditch bond ETFs and use an alternative instead, like a bond mutual fund.  Here, they go astray.

With a bond mutual fund, you are guaranteed the ability to buy and sell at NAV on a daily basis.

Is this a good thing? If you want to sell during periods of market stress, absolutely. By guaranteeing at-NAV executions, bond mutual funds shield exiting shareholders from the true costs of liquidating a bond portfolio during periods of market stress.

But what about fund shareholders who don’t sell? It’s not so rosy.

Going back to our hypothetical scenario, let’s imagine you have two products:a bond ETF and a bond mutual fund. The NAV for each is $100. The market is stressed. The ETF is trading at $95, because APs don’t think they can actually liquidate the portfolio for $100. Mutual fund investors who redeem that day are thrilled:They get the full $100 NAV.

But what about shareholders who hold for the long term? To fulfill a redemption request, the bond mutual fund has to come up with $100 in cash. Under normal situations, it would sell “$100” worth of bonds. But we just agreed that the bond NAV isn’t real; the fund can’t actually sell those bonds for $100. To fulfill the redemption request, they will have to sell for more than the full $100.

This harms investors who don’t sell; they’re forced to subsidize the costs of other shareholders leaving the fund.

Worse, because mutual funds need to be able to process those redemptions overnight, they generally either keep cash on hand, creating cash drag, or have a credit facility that will show up as a fund expense. ETFs avoid both of these issues as well.

Obviously, this is all too black and white. ETFs can trade to “reasonable” and “unreasonable” premiums and discounts. Bond mutual funds can shield some, though not all, of this liquidity risk through that aforementioned cash cushion.

Neither product is perfect.

Ironically, one could argue that the traditional relationship between ETFs and mutual funds is upended in the bond space.

In this narrow case, investors who want to trade bond products during periods of market stress may be better off “trading” mutual funds at the end of the day; for investors who want to buy and hold for the long haul, the “tradable” ETFs are potentially the better option.

At the time this article was written, the author held no position in the security mentioned. Contact Matt Hougan at mhougan@indexuniverse.com .


Permalink | ' Copyright 2013 IndexUniverse LLC. All rights reserved