Liquidity Issues for Bond ETFs at Discount to NAV

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
Northern Flicker
Posts: 15363
Joined: Fri Apr 10, 2015 12:29 am

Liquidity Issues for Bond ETFs at Discount to NAV

Post by Northern Flicker »

There have been past threads regarding the cause of bond ETFs opening up persistent discounts to NAV. In the most general sense, this is caused by the underlying bonds not being liquid enough for authorized participants for the ETF to arbitrage away the gap by buying the ETF at a discount, redeeming the shares for the underlying bonds, and selling the bonds. Lather. Rinse. Repeat until the prices come together.

In extreme cases, particularly in thinner markets like those for a particular corporation’s bonds, there may not even be recent bond prices to compute NAV, leaving it up to accountants to make their best estimates to mark the bonds to market.

This has led some to believe that stale pricing is the only issue and therefore the ETF should reflect the accurate price for the bundle of bonds.

Activity in the TIPS market today (7/5/2022) seems to undermine that position. At the short end of the yield curve, VTAPX and VTIP are mutual fund and ETF share classes of the same fund. At the intermediate point on the curve, the ETF SCHP and the mutual fund FIPDX track the same index. Today’s movements:

VTIP -0.4%
VTAPX -0.08%
SCHP -0.3%
FIPDX +0.1%

I’m not aware that the TIPS market is suffering from stale pricing today.

My belief is that the correct position is that liquidity barriers to arbitrage are the general issue. When liquidity is in such short supply that a bond does not have current pricing, the problem becomes worse. But this does not mean that the ETF price reflects fair value. Not having a recent transaction is a stronger barrier to arbitrage, but it does not have to be that extreme to be a barrier. Clearly, mutual fund accountants take this into account when marking the bonds to market. If they believed that the ETF was fair value, they would value the underlying bonds so that the fund valuation matched or was close to the ETF value.

Neither price is trustworthy when a lack of liquidity undermines market efficiency. Arbitrage generally will bring the prices together somewhere in the middle, and fair value is likely to be somewhere in between the ETF trade price and NAV.
livesoft
Posts: 86075
Joined: Thu Mar 01, 2007 7:00 pm

Re: Liquidity Issues for Bond ETFs at Discount to NAV

Post by livesoft »

Discounts to NAV are great if you are buying bond ETFs.
Wiki This signature message sponsored by sscritic: Learn to fish.
User avatar
typical.investor
Posts: 5263
Joined: Mon Jun 11, 2018 3:17 am

Re: Liquidity Issues for Bond ETFs at Discount to NAV

Post by typical.investor »

Northern Flicker wrote: Tue Jul 05, 2022 5:19 pm Today’s movements:

VTIP -0.4%
VTAPX -0.08%
SCHP -0.3%
FIPDX +0.1%

Not disagreeing with your conclusion yet, but if NAV is accurate then why is VTIPX (investor share class) in at -0.12%?

Also, SCHP and FIPDX follow the same index, but FIPDX reports a much shorter duration. Maybe it's the same ... efforts to clarify apparently failed ... see viewtopic.php?p=6241373#p6241373

In any case, the 1, 3 and 5 year returns on VTIP are better than both VTAPX and VTIPX.

1-Year
3-Year
5-Year

VTIP (0.04% ER)
1.52
3.67
3.10

VTAPX (0.06% ER)
1.46
3.57
3.04

VTIPX (0.14 ER)
1.41
3.49
2.96
Northern Flicker wrote: Tue Jul 05, 2022 5:19 pm Clearly, mutual fund accountants take this into account when marking the bonds to market. If they believed that the ETF was fair value, they would value the underlying bonds so that the fund valuation matched or was close to the ETF value.
Perhaps Vanguard with it's different share classes might be able to do that, but there is no way Fidelity can look at SCHP and price FIPDX. They don't have the same bonds even following the same index.

Anyway, in a market with no buyers, why should the fair price be that of a market in which there are available buyers?

Let me ask this, in the 2020 liquidity crunch, what price would you have likely received if selling an individual treasury? Do you think you would have gotten closer to the NAV that mutual funds had it list at or closer the price it fetched as part of the ETF?

In any case, how do liquidity barriers to arbitrage explain the 1,3, and 5 year returns? Is it the suggestion that liquidity barriers as evidenced by today's returns will mean poorer returns for the ETF? If so, again what explains the 1,3 and 5 year returns?
Topic Author
Northern Flicker
Posts: 15363
Joined: Fri Apr 10, 2015 12:29 am

Re: Liquidity Issues for Bond ETFs at Discount to NAV

Post by Northern Flicker »

Barriers to arbitrage (low liquidity) generally have been short lived and should have minimal effect on long term return.

I was not trying to suggest that Fidelity uses SCHP market values to help value FIPDX. ETF providers compute the NAV for ETFs. There would be no notion of discount or premium to NAV without a computed NAV to which to compare. Moreover, it is bonds that may not have a bid for several days that I was referring to in that regard. TIPS have not exhibited that problem, at least not recently if ever.

There have been articles on line and threads on here suggesting that the ETF market price is the accurate pricing whenever there is a disparity. I am not in agreement with that position, and I believe that my original posting in the thread supports that position.
Chip
Posts: 3994
Joined: Wed Feb 21, 2007 3:57 am

Re: Liquidity Issues for Bond ETFs at Discount to NAV

Post by Chip »

Not sure if this is part of the explanation, but the record date for the July SCHP .374/share dividend was 7/5. Maybe an open end fund vs. ETF issue?
exodusNH
Posts: 10344
Joined: Wed Jan 06, 2021 7:21 pm

Re: Liquidity Issues for Bond ETFs at Discount to NAV

Post by exodusNH »

Northern Flicker wrote: Tue Jul 05, 2022 5:19 pm There have been past threads regarding the cause of bond ETFs opening up persistent discounts to NAV. In the most general sense, this is caused by the underlying bonds not being liquid enough for authorized participants for the ETF to arbitrage away the gap by buying the ETF at a discount, redeeming the shares for the underlying bonds, and selling the bonds. Lather. Rinse. Repeat until the prices come together.

In extreme cases, particularly in thinner markets like those for a particular corporation’s bonds, there may not even be recent bond prices to compute NAV, leaving it up to accountants to make their best estimates to mark the bonds to market.

This has led some to believe that stale pricing is the only issue and therefore the ETF should reflect the accurate price for the bundle of bonds.

Activity in the TIPS market today (7/5/2022) seems to undermine that position. At the short end of the yield curve, VTAPX and VTIP are mutual fund and ETF share classes of the same fund. At the intermediate point on the curve, the ETF SCHP and the mutual fund FIPDX track the same index. Today’s movements:

VTIP -0.4%
VTAPX -0.08%
SCHP -0.3%
FIPDX +0.1%

I’m not aware that the TIPS market is suffering from stale pricing today.

My belief is that the correct position is that liquidity barriers to arbitrage are the general issue. When liquidity is in such short supply that a bond does not have current pricing, the problem becomes worse. But this does not mean that the ETF price reflects fair value. Not having a recent transaction is a stronger barrier to arbitrage, but it does not have to be that extreme to be a barrier. Clearly, mutual fund accountants take this into account when marking the bonds to market. If they believed that the ETF was fair value, they would value the underlying bonds so that the fund valuation matched or was close to the ETF value.

Neither price is trustworthy when a lack of liquidity undermines market efficiency. Arbitrage generally will bring the prices together somewhere in the middle, and fair value is likely to be somewhere in between the ETF trade price and NAV.
Re VTIP/VTAPX, part of the discrepancy may be the different exdiv dates. VTIP went ex/payable/reinvest on 7/1, 7/5, and 7/7 respectively vs 6/29, 6/30, and 7/1. When you posted your original message, the 7/5 NAVs had not yet been calculated.

Alex_686 has laid out good reasons to trust the ETF value when there a meaningful discrepancy.

Otherwise I agree that the true value is somewhere between the market ETF price and the accountants' NAV. Each are valid in their own ways, though, in theory, the ETF price reflects the consensus of a much larger group of people / machine overlords. Wisdom of the crowds? Mob mentality? Who can say?
Topic Author
Northern Flicker
Posts: 15363
Joined: Fri Apr 10, 2015 12:29 am

Re: Liquidity Issues for Bond ETFs at Discount to NAV

Post by Northern Flicker »

The problem with trusting the ETF value is it is a valuation of a package not of individual securities. If some components are less liquid than others, it will bring down the value of the parcel.

Yesterday’s market action aside, TIPS ETFs have opened up discounts to NAV that persisted when other bond ETFs have despite there having been recent trades. The case laid out by alex_686 focused on the absence of recent trades to value the bonds.
Post Reply