Please read this before posting about ETF comparisons in March 2020

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
Topic Author
secondopinion
Posts: 712
Joined: Wed Dec 02, 2020 1:18 pm

Please read this before posting about ETF comparisons in March 2020

Post by secondopinion »

I have seen many posts comparing ETFs in March 2020 and coming to conclusions that they should change their allocations, especially concerning bonds. However, these posts ignore the fact that many ETFs were on steep discount. VTEB, for example, was over 9% on discount; yet, the holdings are good quality municipal bonds and did not drop like this in reality. Municipal bonds dropped a little then, but the discount exaggerated the situation.

ETFs carry the risk that their price does not reflect reality of their holdings.

Before coming to conclusions about one's allocation using March 2020 as a data point and using ETFs in general, please reflect on the fact that ETFs have had and may in the future have unusual behavior that the holdings themselves do not reflect.

As an investor, please pay note and research such anomalies carefully before coming to conclusions.
It is better to be half-wrong than have a 50% chance of being all-wrong. With the former, you will learn and have money to try again. Otherwise, you will never learn and will have nothing eventually.
000
Posts: 4857
Joined: Thu Jul 23, 2020 12:04 am

Re: Please read this before posting about ETF comparisons in March 2020

Post by 000 »

secondopinion wrote: Tue Apr 20, 2021 5:48 pm However, these posts ignore the fact that many ETFs were on steep discount. VTEB, for example, was over 9% on discount; yet, the holdings are good quality municipal bonds and did not drop like this in reality. Municipal bonds dropped a little then, but the discount exaggerated the situation.

ETFs carry the risk that their price does not reflect reality of their holdings.

Before coming to conclusions about one's allocation using March 2020 as a data point and using ETFs in general, please reflect on the fact that ETFs have had and may in the future have unusual behavior that the holdings themselves do not reflect.
I disagree. The ETF price was more likely to be accurate than the stale NAV of the underlying holdings. We are talking about an ETF that trades everyday with good liquidity versus a muni or corp bond that might not trade at all on a particular day and has large spreads when trading. How can a stale price be more accurate than the liquid ETF market price?

The real sin here is backtesting not how ETFs performed in March 2020. Imagine if you had a individual muni or corp bond you were trying to get out of then. You would have gotten slaughtered. It's not a problem of liquid ETFs but of the underlying holdings (i.e. how the bond market works).
burritoLover
Posts: 664
Joined: Sun Jul 05, 2020 12:13 pm

Re: Please read this before posting about ETF comparisons in March 2020

Post by burritoLover »

What do you think happens when you have to sell shares of an ETF when it’s trading at a discount? It’s totally relevant.
"Your money is like a bar of soap. The more you handle it, the less you’ll have." - Gene Fama
Topic Author
secondopinion
Posts: 712
Joined: Wed Dec 02, 2020 1:18 pm

Re: Please read this before posting about ETF comparisons in March 2020

Post by secondopinion »

burritoLover wrote: Tue Apr 20, 2021 6:22 pm What do you think happens when you have to sell shares of an ETF when it’s trading at a discount? It’s totally relevant.
What happens when you have to buy on discount? It works both ways. I did say that is a risk (and it is indeed relevant), but I am more worried about others making faulty conclusions about an asset class just because of an ETF's behavior.

But I want others to know mis-pricing happens.
It is better to be half-wrong than have a 50% chance of being all-wrong. With the former, you will learn and have money to try again. Otherwise, you will never learn and will have nothing eventually.
Marseille07
Posts: 3343
Joined: Fri Nov 06, 2020 1:41 pm

Re: Please read this before posting about ETF comparisons in March 2020

Post by Marseille07 »

I mean, the conclusion should simply be...don't touch illiquid ETFs, especially for Boglehead-style investing that might last for decades. The concern of 9% discount is valid, but it doesn't happen on popular ETFs like SPY & VOO.
Makefile
Posts: 717
Joined: Fri Apr 22, 2016 11:03 pm

Re: Please read this before posting about ETF comparisons in March 2020

Post by Makefile »

000 wrote: Tue Apr 20, 2021 6:00 pm
secondopinion wrote: Tue Apr 20, 2021 5:48 pm However, these posts ignore the fact that many ETFs were on steep discount. VTEB, for example, was over 9% on discount; yet, the holdings are good quality municipal bonds and did not drop like this in reality. Municipal bonds dropped a little then, but the discount exaggerated the situation.

ETFs carry the risk that their price does not reflect reality of their holdings.

Before coming to conclusions about one's allocation using March 2020 as a data point and using ETFs in general, please reflect on the fact that ETFs have had and may in the future have unusual behavior that the holdings themselves do not reflect.
I disagree. The ETF price was more likely to be accurate than the stale NAV of the underlying holdings. We are talking about an ETF that trades everyday with good liquidity versus a muni or corp bond that might not trade at all on a particular day and has large spreads when trading. How can a stale price be more accurate than the liquid ETF market price?

The real sin here is backtesting not how ETFs performed in March 2020. Imagine if you had a individual muni or corp bond you were trying to get out of then. You would have gotten slaughtered. It's not a problem of liquid ETFs but of the underlying holdings (i.e. how the bond market works).
It does seem that when an ETF sells a significant discount, either (1) the supposed NAV is too high and there actually is no discount, or (2) the "authorized participants" who are supposed to step in and buy ETF shares when there is a surplus on the market (just as they would create more shares to sell if the ETF is selling at a premium) weren't doing this, causing it to behave like a closed-end fund. How do we know it isn't the latter? Or is it guaranteed that the authorized participants did that sufficiently.
Topic Author
secondopinion
Posts: 712
Joined: Wed Dec 02, 2020 1:18 pm

Re: Please read this before posting about ETF comparisons in March 2020

Post by secondopinion »

000 wrote: Tue Apr 20, 2021 6:00 pm
secondopinion wrote: Tue Apr 20, 2021 5:48 pm However, these posts ignore the fact that many ETFs were on steep discount. VTEB, for example, was over 9% on discount; yet, the holdings are good quality municipal bonds and did not drop like this in reality. Municipal bonds dropped a little then, but the discount exaggerated the situation.

ETFs carry the risk that their price does not reflect reality of their holdings.

Before coming to conclusions about one's allocation using March 2020 as a data point and using ETFs in general, please reflect on the fact that ETFs have had and may in the future have unusual behavior that the holdings themselves do not reflect.
I disagree. The ETF price was more likely to be accurate than the stale NAV of the underlying holdings. We are talking about an ETF that trades everyday with good liquidity versus a muni or corp bond that might not trade at all on a particular day and has large spreads when trading. How can a stale price be more accurate than the liquid ETF market price?

The real sin here is backtesting not how ETFs performed in March 2020. Imagine if you had a individual muni or corp bond you were trying to get out of then. You would have gotten slaughtered. It's not a problem of liquid ETFs but of the underlying holdings (i.e. how the bond market works).
But mutual funds trade on NAV; so either the ETF is correct or the mutual fund is correct. The divergence is massive in the time of question. If the claim about ETFs is correct, then the mutual fund holders were paying a premium.
It is better to be half-wrong than have a 50% chance of being all-wrong. With the former, you will learn and have money to try again. Otherwise, you will never learn and will have nothing eventually.
typical.investor
Posts: 2611
Joined: Mon Jun 11, 2018 3:17 am

Re: Please read this before posting about ETF comparisons in March 2020

Post by typical.investor »

secondopinion wrote: Tue Apr 20, 2021 6:47 pm But mutual funds trade on NAV; so either the ETF is correct or the mutual fund is correct. The divergence is massive in the time of question. If the claim about ETFs is correct, then the mutual fund holders were paying a premium.
Yes, mutual fund holders were definitely paying a premium (relative to what bonds were actually trading at) to buy, but also selling at a premium if rebalancing out of bonds.

Perhaps one should hold some of each to act according to the situation, but occurrences when the treasury market become unhinged are pretty few.
Topic Author
secondopinion
Posts: 712
Joined: Wed Dec 02, 2020 1:18 pm

Re: Please read this before posting about ETF comparisons in March 2020

Post by secondopinion »

Marseille07 wrote: Tue Apr 20, 2021 6:41 pm I mean, the conclusion should simply be...don't touch illiquid ETFs, especially for Boglehead-style investing that might last for decades. The concern of 9% discount is valid, but it doesn't happen on popular ETFs like SPY & VOO.
VTEB is rather liquid, so is FLOT (a floating rate ETF); both with over a million shares average volume. The spreads are usually tight. VOO is more liquid; but not by much. BND has more volume than VOO, but it had over 6% discount in March 2020.
Last edited by secondopinion on Tue Apr 20, 2021 6:57 pm, edited 1 time in total.
It is better to be half-wrong than have a 50% chance of being all-wrong. With the former, you will learn and have money to try again. Otherwise, you will never learn and will have nothing eventually.
alex_686
Posts: 8301
Joined: Mon Feb 09, 2015 2:39 pm

Re: Please read this before posting about ETF comparisons in March 2020

Post by alex_686 »

secondopinion wrote: Tue Apr 20, 2021 6:47 pm But mutual funds trade on NAV; so either the ETF is correct or the mutual fund is correct. The divergence is massive in the time of question. If the claim about ETFs is correct, then the mutual fund holders were paying a premium.
Price can get very metaphysical in these cases. So maybe neither is are right.

However, as somebody who has actually struct a mutual fund's NAV, in my opinion the ETFs market price generally is more accurate than the accountant's estimate of NAV.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
alex_686
Posts: 8301
Joined: Mon Feb 09, 2015 2:39 pm

Re: Please read this before posting about ETF comparisons in March 2020

Post by alex_686 »

secondopinion wrote: Tue Apr 20, 2021 6:56 pm VTEB is rather liquid, so is FLOT (a floating rate ETF); both with over a million shares average volume. The spreads are usually tight. VOO is more liquid; but not by much. BND has more volume than VOO, but it had over 6% discount in March 2020.
In this case, the ETFs were often more liquid than the underlying securities. Often the ETF price was driving the underlying securities' price.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
Makefile
Posts: 717
Joined: Fri Apr 22, 2016 11:03 pm

Re: Please read this before posting about ETF comparisons in March 2020

Post by Makefile »

alex_686 wrote: Tue Apr 20, 2021 6:57 pm
secondopinion wrote: Tue Apr 20, 2021 6:47 pm But mutual funds trade on NAV; so either the ETF is correct or the mutual fund is correct. The divergence is massive in the time of question. If the claim about ETFs is correct, then the mutual fund holders were paying a premium.
Price can get very metaphysical in these cases. So maybe neither is are right.

However, as somebody who has actually struct a mutual fund's NAV, in my opinion the ETFs market price generally is more accurate than the accountant's estimate of NAV.
Yeah, but I'm wondering in a panic sell situation, is it possible for the panic selling to overwhelm the authorized participants' ability to correct the price?
alex_686
Posts: 8301
Joined: Mon Feb 09, 2015 2:39 pm

Re: Please read this before posting about ETF comparisons in March 2020

Post by alex_686 »

Makefile wrote: Tue Apr 20, 2021 6:59 pm
alex_686 wrote: Tue Apr 20, 2021 6:57 pm
secondopinion wrote: Tue Apr 20, 2021 6:47 pm But mutual funds trade on NAV; so either the ETF is correct or the mutual fund is correct. The divergence is massive in the time of question. If the claim about ETFs is correct, then the mutual fund holders were paying a premium.
Price can get very metaphysical in these cases. So maybe neither is are right.

However, as somebody who has actually struct a mutual fund's NAV, in my opinion the ETFs market price generally is more accurate than the accountant's estimate of NAV.
Yeah, but I'm wondering in a panic sell situation, is it possible for the panic selling to overwhelm the authorized participants' ability to correct the price?
I think - and I want to emphasis that is is a opinion - that you are asking the wrong question. It is a very much "chicken or the egg" question.

For 2020, 2008, and 200 the issue was that the market was frozen. The APs can' work if the market is not working. On the flip side, when the market is frozen the ETFs still trade, work, and help pricing.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
Topic Author
secondopinion
Posts: 712
Joined: Wed Dec 02, 2020 1:18 pm

Re: Please read this before posting about ETF comparisons in March 2020

Post by secondopinion »

alex_686 wrote: Tue Apr 20, 2021 7:09 pm
Makefile wrote: Tue Apr 20, 2021 6:59 pm
alex_686 wrote: Tue Apr 20, 2021 6:57 pm
secondopinion wrote: Tue Apr 20, 2021 6:47 pm But mutual funds trade on NAV; so either the ETF is correct or the mutual fund is correct. The divergence is massive in the time of question. If the claim about ETFs is correct, then the mutual fund holders were paying a premium.
Price can get very metaphysical in these cases. So maybe neither is are right.

However, as somebody who has actually struct a mutual fund's NAV, in my opinion the ETFs market price generally is more accurate than the accountant's estimate of NAV.
Yeah, but I'm wondering in a panic sell situation, is it possible for the panic selling to overwhelm the authorized participants' ability to correct the price?
I think - and I want to emphasis that is is a opinion - that you are asking the wrong question. It is a very much "chicken or the egg" question.

For 2020, 2008, and 200 the issue was that the market was frozen. The APs can' work if the market is not working. On the flip side, when the market is frozen the ETFs still trade, work, and help pricing.
When I think about it, it is like buying an international ETF; the stock market in many of the countries are closed when that ETF trades.

I see the point here; it is good to have the dialog to actually analyze it.
It is better to be half-wrong than have a 50% chance of being all-wrong. With the former, you will learn and have money to try again. Otherwise, you will never learn and will have nothing eventually.
000
Posts: 4857
Joined: Thu Jul 23, 2020 12:04 am

Re: Please read this before posting about ETF comparisons in March 2020

Post by 000 »

Makefile wrote: Tue Apr 20, 2021 6:44 pm It does seem that when an ETF sells a significant discount, either (1) the supposed NAV is too high and there actually is no discount, or (2) the "authorized participants" who are supposed to step in and buy ETF shares when there is a surplus on the market (just as they would create more shares to sell if the ETF is selling at a premium) weren't doing this, causing it to behave like a closed-end fund. How do we know it isn't the latter? Or is it guaranteed that the authorized participants did that sufficiently.
Certainly possible, but see what alex_686 wrote:
For 2020, 2008, and 200 the issue was that the market was frozen. The APs can' work if the market is not working. On the flip side, when the market is frozen the ETFs still trade, work, and help pricing.
typical.investor
Posts: 2611
Joined: Mon Jun 11, 2018 3:17 am

Re: Please read this before posting about ETF comparisons in March 2020

Post by typical.investor »

000 wrote: Tue Apr 20, 2021 6:00 pm The real sin here is backtesting not how ETFs performed in March 2020. Imagine if you had a individual muni or corp bond you were trying to get out of then. You would have gotten slaughtered. It's not a problem of liquid ETFs but of the underlying holdings (i.e. how the bond market works).
I see your point and agree but it's important to note though that those getting out of the market then (whether it's for ETFs, individual stocks or mutual funds) didn't necessarily get slaughtered. If you were moving to cash, yes maybe. If you were moving to buy stocks which were down extremely, then it worked out.

I did that several times with ETFs (muni and treasury) and it was to my advantage.

There several posts here by mutual fund holders complaining that by the time their order filled, that stocks had jumped. I noticed it on several occasions (each of the three times I traded in March 2020 actually) and so personally don't believe that holding bonds in mutual funds is actually necessary for rebalancing in a severe crisis.
000
Posts: 4857
Joined: Thu Jul 23, 2020 12:04 am

Re: Please read this before posting about ETF comparisons in March 2020

Post by 000 »

typical.investor wrote: Tue Apr 20, 2021 7:23 pm
000 wrote: Tue Apr 20, 2021 6:00 pm The real sin here is backtesting not how ETFs performed in March 2020. Imagine if you had a individual muni or corp bond you were trying to get out of then. You would have gotten slaughtered. It's not a problem of liquid ETFs but of the underlying holdings (i.e. how the bond market works).
I see your point and agree but it's important to note though that those getting out of the market then (whether it's for ETFs, individual stocks or mutual funds) didn't necessarily get slaughtered. If you were moving to cash, yes maybe. If you were moving to buy stocks which were down extremely, then it worked out.
I was referring only to the one-month price decline one would probably have to take to get out of an individual muni or corporate bond ("slaughter" is relative to the usual volatility and potential upside of bonds). Of course if one turned around and immediately bought stocks they obviously came out far ahead this time.
Marseille07
Posts: 3343
Joined: Fri Nov 06, 2020 1:41 pm

Re: Please read this before posting about ETF comparisons in March 2020

Post by Marseille07 »

secondopinion wrote: Tue Apr 20, 2021 6:56 pm
Marseille07 wrote: Tue Apr 20, 2021 6:41 pm I mean, the conclusion should simply be...don't touch illiquid ETFs, especially for Boglehead-style investing that might last for decades. The concern of 9% discount is valid, but it doesn't happen on popular ETFs like SPY & VOO.
VTEB is rather liquid, so is FLOT (a floating rate ETF); both with over a million shares average volume. The spreads are usually tight. VOO is more liquid; but not by much. BND has more volume than VOO, but it had over 6% discount in March 2020.
As alex_686 said, it's not just the ETF itself but the underlying has to be liquid. Both items have to be liquid, essentially.
alex_686
Posts: 8301
Joined: Mon Feb 09, 2015 2:39 pm

Re: Please read this before posting about ETF comparisons in March 2020

Post by alex_686 »

000 wrote: Tue Apr 20, 2021 7:27 pm I was referring only to the one-month price decline one would probably have to take to get out of an individual muni or corporate bond ("slaughter" is relative to the usual volatility and potential upside of bonds). Of course if one turned around and immediately bought stocks they obviously came out far ahead this time.
Let me extend on this a bit, and argue you are looking at this backwards.

Let us start off with the assumption that the ETF price is correct market price and the mutual fund price is wrong. This is the conventional wisdom. The accounting data that generates the NAV tends to be stale and lags the market.

The people who are selling the ETFs are getting the correct price. Yes, they are getting slaughtered but they are choosing to sell in the middle of of a panic.

The people who are selling mutual funds are selling at a stale price - a price that is set on prior days accounting data. As such they are getting a premium. That is, other shareholders are subsidizing the seller's premium. As a former mutual fund account I find this morally objectional - not that I have a better idea.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
000
Posts: 4857
Joined: Thu Jul 23, 2020 12:04 am

Re: Please read this before posting about ETF comparisons in March 2020

Post by 000 »

alex_686 wrote: Tue Apr 20, 2021 7:37 pm
000 wrote: Tue Apr 20, 2021 7:27 pm I was referring only to the one-month price decline one would probably have to take to get out of an individual muni or corporate bond ("slaughter" is relative to the usual volatility and potential upside of bonds). Of course if one turned around and immediately bought stocks they obviously came out far ahead this time.
Let me extend on this a bit, and argue you are looking at this backwards.
I don't follow. I was talking about selling an individual corporate bond versus corp bond ETF and suggesting to those unhappy with the ETF discount to try selling individual bonds during liquidity panics...
typical.investor
Posts: 2611
Joined: Mon Jun 11, 2018 3:17 am

Re: Please read this before posting about ETF comparisons in March 2020

Post by typical.investor »

Marseille07 wrote: Tue Apr 20, 2021 7:28 pm
secondopinion wrote: Tue Apr 20, 2021 6:56 pm
Marseille07 wrote: Tue Apr 20, 2021 6:41 pm I mean, the conclusion should simply be...don't touch illiquid ETFs, especially for Boglehead-style investing that might last for decades. The concern of 9% discount is valid, but it doesn't happen on popular ETFs like SPY & VOO.
VTEB is rather liquid, so is FLOT (a floating rate ETF); both with over a million shares average volume. The spreads are usually tight. VOO is more liquid; but not by much. BND has more volume than VOO, but it had over 6% discount in March 2020.
As alex_686 said, it's not just the ETF itself but the underlying has to be liquid. Both items have to be liquid, essentially.
I strongly disagree with your conclusion and believe it's not based on actually trading. Your conclusion makes no sense.

Even if the underlying isn't liquid (as in the case of municipal bonds and treasuries in an extreme crisis), the ETF will still reflect market prices. And as such, it's still possible to do well cashing out of (safer but depressed) assets such as bonds to rebalance back into even more depressed stocks to do Boglehead-style investing per your IPS (Investor Policy Statement).
Post Reply