Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Ferdinand2014
Posts: 2390
Joined: Mon Dec 17, 2018 5:49 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by Ferdinand2014 »

burritoLover wrote: Sun Dec 05, 2021 10:33 am I'm curious if there is any "500" indexes that are simply the 500 largest companies by market cap evaluated say, once a year. It isn't like we are talking about illiquid small stocks here. It seems you could have a simple rules-based strategy to handle this. Maybe that is what the CSRP is.

Yes:
https://fundresearch.fidelity.com/mutua ... /315911628
“You only find out who is swimming naked when the tide goes out.“ — Warren Buffett
Ferdinand2014
Posts: 2390
Joined: Mon Dec 17, 2018 5:49 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by Ferdinand2014 »

nisiprius wrote: Sun Dec 05, 2021 10:54 am
burritoLover wrote: Sun Dec 05, 2021 10:33 am I'm curious if there is any "500" indexes that are simply the 500 largest companies by market cap evaluated say, once a year. It isn't like we are talking about illiquid small stocks here. It seems you could have a simple rules-based strategy to handle this. Maybe that is what the CSRP is.
The CRSP is a firm and they provide many different indexes. The WSJ article is about one of them, the CRSP US Total Market Index.

CRSP also provides a CRSP U.S. Large Cap Index. It's tracked by the Vanguard Large-Cap Index Fund, VLCAX, and ETF, VV. Rather than including the 500 largest-cap stocks,
The CRSP US Large Cap Index includes U.S. companies that comprise the top 85% of investable market capitalization.
Maybe an arbitrary percentage-of-total-market-cap cutoff makes more sense than an arbitrary number-of-stocks cutoff. Anyway, it currently happens to contain 581 companies.

And for a long time, there has been the Russell 1000 index, which includes 1,000, not 500, but in a rules-based way based on capitalization, and a decent handful of funds and ETFs tracking it, include Vanguard's VONE.

The CRSP was founded because Merrill Lynch, at that time, ran occasional full-newspaper-page-sized ads--analogous to informercials--teaching investors about the stock market. They wanted to run one saying that common stocks were a prudent long-term investment for small investors. The SEC wouldn't let them publish it without evidence, so Merrill Lynch funded a project at the University of Chicago School of Business to research and produce that evidence.

Fidelity ZERO large cap index fund:
https://fundresearch.fidelity.com/mutua ... /315911628

“Normally investing at least 80% of assets in common stocks of large capitalization companies included in the Fidelity U.S. Large Cap Index, which is a float-adjusted market capitalization-weighted index designed to reflect the performance of U.S. large capitalization stocks. Large capitalization stocks are considered to be stocks of the largest 500 U.S. companies based on float-adjusted market capitalization. “


Comparison with VFIAX and FXAIX - 2 common leading S&P 500 index funds
https://www.portfoliovisualizer.com/bac ... ion3_3=100
Last edited by Ferdinand2014 on Sun Dec 05, 2021 10:06 pm, edited 1 time in total.
“You only find out who is swimming naked when the tide goes out.“ — Warren Buffett
mikejuss
Posts: 2749
Joined: Tue Jun 23, 2020 1:36 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by mikejuss »

Ferdinand2014 wrote: Sun Dec 05, 2021 9:59 pm
burritoLover wrote: Sun Dec 05, 2021 10:33 am I'm curious if there is any "500" indexes that are simply the 500 largest companies by market cap evaluated say, once a year. It isn't like we are talking about illiquid small stocks here. It seems you could have a simple rules-based strategy to handle this. Maybe that is what the CSRP is.

Yes:
https://fundresearch.fidelity.com/mutua ... /315911628
I apologize for my idiocy, but could you tell me how Fidelity makes money by selling an investment product with no expense ratio? I calculate that VTSAX brings in $520 million in revenue for Vanguard ($1.3 trillion x 0.04% = $520 million).
Last edited by mikejuss on Sun Dec 05, 2021 10:31 pm, edited 1 time in total.
50% VTSAX | 25% VTIAX | 25% VBTLX (retirement), 25% VTEAX (taxable)
Ferdinand2014
Posts: 2390
Joined: Mon Dec 17, 2018 5:49 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by Ferdinand2014 »

mikejuss wrote: Sun Dec 05, 2021 10:03 pm
Ferdinand2014 wrote: Sun Dec 05, 2021 9:59 pm
burritoLover wrote: Sun Dec 05, 2021 10:33 am I'm curious if there is any "500" indexes that are simply the 500 largest companies by market cap evaluated say, once a year. It isn't like we are talking about illiquid small stocks here. It seems you could have a simple rules-based strategy to handle this. Maybe that is what the CSRP is.

Yes:
https://fundresearch.fidelity.com/mutua ... /315911628
I apologize for my idiocy, but could you tell me how Fidelity makes money by selling an investment product with no expense ratio?
Share lending? Loss leader product? IDK.
“You only find out who is swimming naked when the tide goes out.“ — Warren Buffett
Northern Flicker
Posts: 15288
Joined: Fri Apr 10, 2015 12:29 am

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by Northern Flicker »

6NDone wrote: Sun Dec 05, 2021 6:45 am
Booogle wrote: Sun Dec 05, 2021 6:22 am Don't forget about VV.

VV would be the large and mid cap subsection of VTI.

(VO is actually 100% part of VV).

So it should have caught nearly all of Tesla's growth.
VV contains 600 stock, most of which is the S&P 500.
VO contains 380 stocks.

VV cannot contain 100% of VO. VV contains only about 17% mid caps .
The S&P500 is about 80% of the market by cap.

The CRSP Large Cap Index tracked by VV is about 87% of the market by cap.

The Russell 1000 is about 90% of the market by cap and includes the full large and midcap segments of the market.

Different indexing companies draw the lines differently, but VV has about 70% of the midcap segment of the market (87-80)/(90-80) = 0.7.
VTI
Posts: 282
Joined: Sat Mar 13, 2021 11:56 pm

Re: "The Mutual Fund That Ate Wall Street—Based on an Index Few People Know About": Debating VTSAX

Post by VTI »

nisiprius wrote: Sun Dec 05, 2021 6:27 pm [...]

When you are working down at the lower margin of adding and dropping stocks from a total market index, whatever horrors maybe happening, they are affecting such a microscopic percentage of the total portfolio value that I can take it philosophically. I lose no sleep over the possibility that, by shrewd human trading against a robotic index fund, someone has gotten the better of me to the tune of twenty cents.
According to my little brain, the issue is that VTSAX is paying a 3.6% fee for every new stock it incorporates, reducing its potential profit by 3.6% for each new stock, too.

(On top of that—and I don't want to drag this conversation off topic, it's my unsupported perception that many recent IPOs and SPACs aren't intended to raise capital. Instead, they're crash grabs from insiders over-eager to dump their bags on the unsuspecting public. As a VTSAX holder, I'm not happy that I'm contributing the enrichment of these insiders.)
Also available as an Admiral™ Shares mutual fund.
User avatar
burritoLover
Posts: 4097
Joined: Sun Jul 05, 2020 12:13 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by burritoLover »

Ferdinand2014 wrote: Sun Dec 05, 2021 10:00 pm Fidelity ZERO large cap index fund:
https://fundresearch.fidelity.com/mutua ... /315911628

“Normally investing at least 80% of assets in common stocks of large capitalization companies included in the Fidelity U.S. Large Cap Index, which is a float-adjusted market capitalization-weighted index designed to reflect the performance of U.S. large capitalization stocks. Large capitalization stocks are considered to be stocks of the largest 500 U.S. companies based on float-adjusted market capitalization. “
Yeah but what is this part getting at?
Using statistical sampling techniques based on such factors as capitalization, industry exposures, dividend yield, price/earnings (P/E) ratio, price/book (P/B) ratio, and earnings growth to attempt to replicate the returns of the Fidelity U.S. Large Cap Index using a smaller number of securities.
User avatar
nisiprius
Advisory Board
Posts: 52105
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by nisiprius »

burritoLover wrote: Mon Dec 06, 2021 7:03 am Yeah but what it this part getting at?
Using statistical sampling techniques based on such factors as capitalization, industry exposures, dividend yield, price/earnings (P/E) ratio, price/book (P/B) ratio, and earnings growth to attempt to replicate the returns of the Fidelity U.S. Large Cap Index using a smaller number of securities.
It's a well-known technique. It's gotten rarer since transactions costs got lower. It's a way of saving money by not literally buying all of the specific securities in the index. The Vanguard 500 Index fund launched with--can't find the exact number right now--something like 275 stocks and a sampling methodology. It's legitimate, but of course it's likely to lead to more tracking error than full replication. For a long time the Vanguard Total Stock Market Index Fund used sampling; not sure when it went to full replication.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.
User avatar
burritoLover
Posts: 4097
Joined: Sun Jul 05, 2020 12:13 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by burritoLover »

nisiprius wrote: Mon Dec 06, 2021 7:17 am
burritoLover wrote: Mon Dec 06, 2021 7:03 am Yeah but what it this part getting at?
Using statistical sampling techniques based on such factors as capitalization, industry exposures, dividend yield, price/earnings (P/E) ratio, price/book (P/B) ratio, and earnings growth to attempt to replicate the returns of the Fidelity U.S. Large Cap Index using a smaller number of securities.
It's a well-known technique. It's gotten rarer since transactions costs got lower. It's a way of saving money by not literally buying all of the specific securities in the index. The Vanguard 500 Index fund launched with--can't find the exact number right now--something like 275 stocks and a sampling methodology. It's legitimate, but of course it's likely to lead to more tracking error than full replication. For a long time the Vanguard Total Stock Market Index Fund used sampling; not sure when it went to full replication.
The Fidelty fund in question says it consists of 500 stocks so what are they meaning here unless they mean they may use this method at times and it may be 490 stocks or something?
sycamore
Posts: 6309
Joined: Tue May 08, 2018 12:06 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by sycamore »

Northern Flicker wrote: Sun Dec 05, 2021 11:09 pm
6NDone wrote: Sun Dec 05, 2021 6:45 am
Booogle wrote: Sun Dec 05, 2021 6:22 am Don't forget about VV.

VV would be the large and mid cap subsection of VTI.

(VO is actually 100% part of VV).

So it should have caught nearly all of Tesla's growth.
VV contains 600 stock, most of which is the S&P 500.
VO contains 380 stocks.

VV cannot contain 100% of VO. VV contains only about 17% mid caps .
The S&P500 is about 80% of the market by cap.

The CRSP Large Cap Index tracked by VV is about 87% of the market by cap.

The Russell 1000 is about 90% of the market by cap and includes the full large and midcap segments of the market.

Different indexing companies draw the lines differently, but VV has about 70% of the midcap segment of the market (87-80)/(90-80) = 0.7.
The CRSP Large Cap Index is described as a combination of Mega Cap Index + Mid Cap Index.

Per the 10/31/2021 portfolio pages of Vanguard's Mega Cap fund MGC and Mid Cap fund VO, there are 246 stocks in both the MGC and its index, but 380 in VO versus 372 in the Mid Cap Index. Not sure why VO has more stocks than its index, maybe different share classes?

Anyway, I would expect to see 246 + 372 = 618 stocks in the Large Cap index but Large Cap fund VV shows 594 in the index and 601 in the fund.

So even though Large is defined (described?) as Mega + Mid, in reality I'm guessing the Large index (and fund) is following its own construction / reconstitution rules, and that somehow causes the mismatch.

By and large, though, Large Cap VV includes most of Mid Cap VO but not quite 100%.
alex_686
Posts: 13286
Joined: Mon Feb 09, 2015 1:39 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by alex_686 »

Ferdinand2014 wrote: Sun Dec 05, 2021 9:59 pm
burritoLover wrote: Sun Dec 05, 2021 10:33 am I'm curious if there is any "500" indexes that are simply the 500 largest companies by market cap evaluated say, once a year. It isn't like we are talking about illiquid small stocks here. It seems you could have a simple rules-based strategy to handle this. Maybe that is what the CSRP is.

Yes:
https://fundresearch.fidelity.com/mutua ... /315911628
No.

Specifically, the fund you are citing uses a committee. In fact the index has been farmed out to S&P so I might guess it is the same committee running the more popular one.

Generally speaking, still no. All indexes uses committees because there are always subjective decisions to be made. Give me any set of rules based construction and I can talk about why it would fail. For example, how would you define "liquid" and "free float"? I can point to many examples where big liquid stocks causes problems.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
alex_686
Posts: 13286
Joined: Mon Feb 09, 2015 1:39 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by alex_686 »

burritoLover wrote: Sun Dec 05, 2021 10:18 am The S&P committee drug their feet on Tesla and did not add it to the 500 index even when it finally met their profitability metrics. That's the problem with the S&P 500 - you've got people making active decisions at times and not just due to making costs / liquidity reasonable for index followers. A total stock market index will have less of these active-managey type decisions.
Considering that all indexes have a committee why do you think S&P's committee is practically bad?

Yeah, they kind of drop the ball on Tesla. Kinda of. They overhauled their process when they let in lots of Tesla like companies during the dot.com boom. New startup companies that had a poor record of positive earnings. But they are being consistent which is kind of critical.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
User avatar
burritoLover
Posts: 4097
Joined: Sun Jul 05, 2020 12:13 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by burritoLover »

alex_686 wrote: Mon Dec 06, 2021 9:13 am
burritoLover wrote: Sun Dec 05, 2021 10:18 am The S&P committee drug their feet on Tesla and did not add it to the 500 index even when it finally met their profitability metrics. That's the problem with the S&P 500 - you've got people making active decisions at times and not just due to making costs / liquidity reasonable for index followers. A total stock market index will have less of these active-managey type decisions.
Considering that all indexes have a committee why do you think S&P's committee is practically bad?

Yeah, they kind of drop the ball on Tesla. Kinda of. They overhauled their process when they let in lots of Tesla like companies during the dot.com boom. New startup companies that had a poor record of positive earnings. But they are being consistent which is kind of critical.
I guess they have a reputation as being more active when it comes to subjective decisions? Sometimes that is proclaimed to be a benefit.

Maybe that is not a correct assumption. It is hard to get info on how each index makes these kind of decisions.
Northern Flicker
Posts: 15288
Joined: Fri Apr 10, 2015 12:29 am

Re: "The Mutual Fund That Ate Wall Street—Based on an Index Few People Know About": Debating VTSAX

Post by Northern Flicker »

VTI wrote: Mon Dec 06, 2021 12:32 am
nisiprius wrote: Sun Dec 05, 2021 6:27 pm [...]

When you are working down at the lower margin of adding and dropping stocks from a total market index, whatever horrors maybe happening, they are affecting such a microscopic percentage of the total portfolio value that I can take it philosophically. I lose no sleep over the possibility that, by shrewd human trading against a robotic index fund, someone has gotten the better of me to the tune of twenty cents.
According to my little brain, the issue is that VTSAX is paying a 3.6% fee for every new stock it incorporates, reducing its potential profit by 3.6% for each new stock, too.
I don't know where you got the purported 3.6% fee number from, but the index being tracked does not pay a fee to add a stock to the index, and VTI/VTSAX does a very good job of tracking its index. Costs are very low, and what costs there are typically are covered by securities lending revenue.
VTI
Posts: 282
Joined: Sat Mar 13, 2021 11:56 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by VTI »

alex_686 wrote: Mon Dec 06, 2021 9:09 am
Ferdinand2014 wrote: Sun Dec 05, 2021 9:59 pm
burritoLover wrote: Sun Dec 05, 2021 10:33 am I'm curious if there is any "500" indexes that are simply the 500 largest companies by market cap evaluated say, once a year. It isn't like we are talking about illiquid small stocks here. It seems you could have a simple rules-based strategy to handle this. Maybe that is what the CSRP is.

Yes:
https://fundresearch.fidelity.com/mutua ... /315911628
No.

Specifically, the fund you are citing uses a committee. In fact the index has been farmed out to S&P so I might guess it is the same committee running the more popular one.

Generally speaking, still no. All indexes uses committees because there are always subjective decisions to be made. Give me any set of rules based construction and I can talk about why it would fail. For example, how would you define "liquid" and "free float"? I can point to many examples where big liquid stocks causes problems.
I believe the original poster was correct. That fund uses the "Fidelity U.S. Large Cap Index", not the S&P 500 index.
Also available as an Admiral™ Shares mutual fund.
alex_686
Posts: 13286
Joined: Mon Feb 09, 2015 1:39 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by alex_686 »

VTI wrote: Mon Dec 06, 2021 9:29 am I believe the original poster was correct. That fund uses the "Fidelity U.S. Large Cap Index", not the S&P 500 index.
Never said it was.

The Fidelity U.S. Large Cap Index is created and maintained by Standards & Poor. Read the methodology.

Yes, it is a proprietary index that is sponsored by Fidelity but a fund sponsor is barred from actually running an index to avoid conflicts of interest.

My argument is that the S&P 500 has a investment committee run by S&P. And Fidelity U.S. Large Cap Index has a investment committee run by S&P. And that they are both large cap US domestic indexes with similar mandates. As such I strongly suspect that there is a fair amount of overlap between how the 2 committees are run.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
alex_686
Posts: 13286
Joined: Mon Feb 09, 2015 1:39 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by alex_686 »

burritoLover wrote: Mon Dec 06, 2021 9:21 am I guess they have a reputation as being more active when it comes to subjective decisions? Sometimes that is proclaimed to be a benefit.

Maybe that is not a correct assumption. It is hard to get info on how each index makes these kind of decisions.
I have never heard about this beyond the speculation of Bogleheads. My impression is that the S&P 500 is by far the most popular and well known index. As such it gets more scrutiny then the 3rd tier indexes. And by size comparison all of the other indexes are indeed 3rd tier.

And it is and is not hard to know how these decisions are made. There are lots of statements, articles, etc. written by current and former employees of the S&P 500. You can get a decent understand of their thinking. It is consistent with their actions.

I tend to use the analogy of umpires and referees. What size is a strike zone? What is pass interference? You need these subjective rules for the game to run. Some referees might be stricter, others looser. That is not a issue. The problem is if they favor team A over team B.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
User avatar
burritoLover
Posts: 4097
Joined: Sun Jul 05, 2020 12:13 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by burritoLover »

alex_686 wrote: Mon Dec 06, 2021 10:10 am
burritoLover wrote: Mon Dec 06, 2021 9:21 am I guess they have a reputation as being more active when it comes to subjective decisions? Sometimes that is proclaimed to be a benefit.

Maybe that is not a correct assumption. It is hard to get info on how each index makes these kind of decisions.
I have never heard about this beyond the speculation of Bogleheads. My impression is that the S&P 500 is by far the most popular and well known index. As such it gets more scrutiny then the 3rd tier indexes. And by size comparison all of the other indexes are indeed 3rd tier.

And it is and is not hard to know how these decisions are made. There are lots of statements, articles, etc. written by current and former employees of the S&P 500. You can get a decent understand of their thinking. It is consistent with their actions.

I tend to use the analogy of umpires and referees. What size is a strike zone? What is pass interference? You need these subjective rules for the game to run. Some referees might be stricter, others looser. That is not a issue. The problem is if they favor team A over team B.
So what was the reason they excluded Tesla when it met the profitability requirements?
alex_686
Posts: 13286
Joined: Mon Feb 09, 2015 1:39 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by alex_686 »

burritoLover wrote: Mon Dec 06, 2021 10:14 am So what was the reason they excluded Tesla when it met the profitability requirements?
They kind of did. First, what is your definition of "profitability"?

IIRC Tesla had non-sustainable positive earnings for 2 years due to accounting changes and sales of environmental credits. i.e., positive earnings were reached by one-off events that were not expected to repeat according to Tesla. It is debatable but S&D did communicate what was happening. For a bit of context here, S&P was badly stung by this type of stuff during the dot.com boom so had really tightened down in this area.

Then Tesla started making real positive earnings from core operations that are expected to be sustainable. And S&P found itself in a pickle. Tesla's size was going to make it hard to add to the index without causing front-running and other distortions. Maybe they fumbled the ball. They were slow. However, I am reluctant to Monday morning quarterback Sunday's game. In this case because I am not sure I could have come up with anything better.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
User avatar
burritoLover
Posts: 4097
Joined: Sun Jul 05, 2020 12:13 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by burritoLover »

alex_686 wrote: Mon Dec 06, 2021 10:40 am
burritoLover wrote: Mon Dec 06, 2021 10:14 am So what was the reason they excluded Tesla when it met the profitability requirements?
They kind of did. First, what is your definition of "profitability"?

IIRC Tesla had non-sustainable positive earnings for 2 years due to accounting changes and sales of environmental credits. i.e., positive earnings were reached by one-off events that were not expected to repeat according to Tesla. It is debatable but S&D did communicate what was happening. For a bit of context here, S&P was badly stung by this type of stuff during the dot.com boom so had really tightened down in this area.

Then Tesla started making real positive earnings from core operations that are expected to be sustainable. And S&P found itself in a pickle. Tesla's size was going to make it hard to add to the index without causing front-running and other distortions. Maybe they fumbled the ball. They were slow. However, I am reluctant to Monday morning quarterback Sunday's game. In this case because I am not sure I could have come up with anything better.
My understanding was that the S&P had a requirement of a certain number of consecutive profitable quarters (something like 4 or 5) before inclusion into the index. I think that is reasonable but when Tesla met this requirement, they decided against adding it that quarter. It sounded like the committee was making active decisions based on how overvalued they thought Tesla was? That isn't the sort of thing I want in what is touted as primarily a passive index.
Northern Flicker
Posts: 15288
Joined: Fri Apr 10, 2015 12:29 am

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by Northern Flicker »

It was not an active decision. They did not want to add it immediately after it qualified because that would have been too predictable, and played into the hands of index front-runners.
alex_686
Posts: 13286
Joined: Mon Feb 09, 2015 1:39 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by alex_686 »

burritoLover wrote: Mon Dec 06, 2021 10:48 am My understanding was that the S&P had a requirement of a certain number of consecutive profitable quarters (something like 4 or 5) before inclusion into the index. I think that is reasonable but when Tesla met this requirement, they decided against adding it that quarter.
So I asked you what "profitability" was. Is it fair to say that you are looking at earnings? That it would be o.k. to included companies that had no economic profits to speak of but had engaged in accounting manipulations (a.k.a. accounting shenanigans) to artificially manipulate earnings to generate positive quarterly earnings?

I suspect that is not what you meant but that is what Tesla kind of did. There is a fair amount of subjectivity when it comes to recognizing income. Because then you are saying we shouldn't accept S&P's subjective judgements but accept Tesla's subjective judgments. And why would Tesla's judgments be better and S&P's?

Once again I never heard about the overvalued bit. But maybe that is because I am in the wrong circles - I don't follow Tesla that closely. But I did follow it closely enough to know there were some real questions around its statements and profitability which were a bit shaky. Note - this is not a major criticism. It is hard to figure out profitability of a rapidly growing company like Tesla. or in the 1990's Amazon, Level 3, JDSU Uniphase, WebVan, etc.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
User avatar
burritoLover
Posts: 4097
Joined: Sun Jul 05, 2020 12:13 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by burritoLover »

alex_686 wrote: Mon Dec 06, 2021 11:06 am
burritoLover wrote: Mon Dec 06, 2021 10:48 am My understanding was that the S&P had a requirement of a certain number of consecutive profitable quarters (something like 4 or 5) before inclusion into the index. I think that is reasonable but when Tesla met this requirement, they decided against adding it that quarter.
So I asked you what "profitability" was. Is it fair to say that you are looking at earnings? That it would be o.k. to included companies that had no economic profits to speak of but had engaged in accounting manipulations (a.k.a. accounting shenanigans) to artificially manipulate earnings to generate positive quarterly earnings?

I suspect that is not what you meant but that is what Tesla kind of did. There is a fair amount of subjectivity when it comes to recognizing income. Because then you are saying we shouldn't accept S&P's subjective judgements but accept Tesla's subjective judgments. And why would Tesla's judgments be better and S&P's?

Once again I never heard about the overvalued bit. But maybe that is because I am in the wrong circles - I don't follow Tesla that closely. But I did follow it closely enough to know there were some real questions around its statements and profitability which were a bit shaky. Note - this is not a major criticism. It is hard to figure out profitability of a rapidly growing company like Tesla. or in the 1990's Amazon, Level 3, JDSU Uniphase, WebVan, etc.
Below is the criteria from S&P - to be clear, this is not what I'm wanting to see in particular, but what the S&P decided upon. Tesla sells carbon credits to other vehicle manufacturers - so at points in time before they were profitable from selling vehicles, a lot of those earnings were from these credits. I'm not sure what accounting shenanigans you are referring to? Or is just that they weren't profitable from their core business?
...a company should be a U.S. company, have a market capitalization of at least USD 11.8 billion, be highly liquid, have a public float of at least 10% of its shares outstanding, and its most recent quarter’s earnings and the sum of its trailing four consecutive quarters’ earnings must be positive.
alex_686
Posts: 13286
Joined: Mon Feb 09, 2015 1:39 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by alex_686 »

burritoLover wrote: Mon Dec 06, 2021 11:25 am Below is the criteria from S&P - to be clear, this is not what I'm wanting to see in particular, but what the S&P decided upon. Tesla sells carbon credits to other vehicle manufacturers - so at points in time before they were profitable from selling vehicles, a lot of those earnings were from these credits. I'm not sure what accounting shenanigans you are referring to? Or is just that they weren't profitable from their core business?
On the carbon credits - well, that is the reason why they had profitability for that one year. And Tesla is saying that they don't expect to be selling much in the way of carbon credits the next year. So how much weight to do you put on this transient event?

Primarily that they were not making money from their core business.

Step back and think why they have this rule in the first place. They want viable business that will be in the index for the long term. They want to minimize churn. They want to minimize companies trying to game the system.

Look at the flip side of the question. Is it better to be too strict or too lax? Ask how many shaky companies would be included on the index and then immediately kicked off if they used a looser standard? What would be the cost and churn in the index? If you are too strict you are going to make one set of errors, too loose a different set of errors.

This kind of ties back to the shenanigans from the dot.com boom. Lots of one-time events happening that really do not speak to the long-term viability of the company.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
User avatar
burritoLover
Posts: 4097
Joined: Sun Jul 05, 2020 12:13 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by burritoLover »

alex_686 wrote: Mon Dec 06, 2021 12:03 pm
burritoLover wrote: Mon Dec 06, 2021 11:25 am Below is the criteria from S&P - to be clear, this is not what I'm wanting to see in particular, but what the S&P decided upon. Tesla sells carbon credits to other vehicle manufacturers - so at points in time before they were profitable from selling vehicles, a lot of those earnings were from these credits. I'm not sure what accounting shenanigans you are referring to? Or is just that they weren't profitable from their core business?
On the carbon credits - well, that is the reason why they had profitability for that one year. And Tesla is saying that they don't expect to be selling much in the way of carbon credits the next year. So how much weight to do you put on this transient event?

Primarily that they were not making money from their core business.

Step back and think why they have this rule in the first place. They want viable business that will be in the index for the long term. They want to minimize churn. They want to minimize companies trying to game the system.

Look at the flip side of the question. Ask how many shaky companies would be included on the index and then immediately kicked off if they used a looser standard? What would be the cost and churn in the index? If you are too strict you are going to make one set of errors, too loose a different set of errors.

This kind of ties back to the shenanigans from the dot.com boom. Lots of one-time events happening that really do not speak to the long-term viability of the company.
Many of the questionable dot coms weren't profitable though. Even if Tesla said that the carbon credits were going to make up a smaller percentage of profits going forward, I'm not sure what relevance that has. That would mean that the S&P committee decided they could predict the future profitability of EV sales for Tesla or, at the very least, are making active decisions on which earnings count and which do not? Seems kind of ridiculous. I guess we won't know for sure as I don't think they've outlined the reasons behind their decisions. I'd rather have an index that sticks to rule-based metrics and only makes an active decision in the most extreme of circumstances (ex. not including a company that is accused of cooking their books by the SEC).
VTI
Posts: 282
Joined: Sat Mar 13, 2021 11:56 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by VTI »

alex_686 wrote: Mon Dec 06, 2021 10:03 am
VTI wrote: Mon Dec 06, 2021 9:29 am I believe the original poster was correct. That fund uses the "Fidelity U.S. Large Cap Index", not the S&P 500 index.
Never said it was.

The Fidelity U.S. Large Cap Index is created and maintained by Standards & Poor. Read the methodology.

Yes, it is a proprietary index that is sponsored by Fidelity but a fund sponsor is barred from actually running an index to avoid conflicts of interest.

My argument is that the S&P 500 has a investment committee run by S&P. And Fidelity U.S. Large Cap Index has a investment committee run by S&P. And that they are both large cap US domestic indexes with similar mandates. As such I strongly suspect that there is a fair amount of overlap between how the 2 committees are run.
I didn't realize that Fidelity contracts with S&P to maintain the Fidelity U.S. Large Cap Index. Thank you. Yes, there's no doubt that an S&P committee handles the Fidelity U.S. Large Cap Index, too.

You're more knowledgeable about this than I am, but in my mind, there's an important distinction to be made between the judgement-driven S&P 500 index and the more mechanical Fidelity U.S. Large Cap Index, which is described like this:
[...]the Fidelity U.S. Large Cap Index℠, which is a float-adjusted market capitalization-weighted index designed to reflect the performance of U.S. large capitalization stocks. Large capitalization stocks are considered to be stocks of the largest 500 U.S. companies based on float-adjusted market capitalization.
When I hear "committee-driven", I understand there to be an implication that human judgement is involved. I believe burritoLover has the same understanding.
Generally speaking, still no. All indexes uses committees because there are always subjective decisions to be made. Give me any set of rules based construction and I can talk about why it would fail. For example, how would you define "liquid" and "free float"? I can point to many examples where big liquid stocks causes problems.
Indeed, there are subjective decisions involved in defining the mechanical rules. That applies for both the S&P 500 index and the Fidelity U.S. Large Cap Index. However, in addition to those subjective definitions, the S&P 500 index is judgement-driven, whereas the Fidelity U.S. Large Cap Index is not. That's my understanding, at least.
Last edited by VTI on Mon Dec 06, 2021 2:41 pm, edited 1 time in total.
Also available as an Admiral™ Shares mutual fund.
Breaking Away
Posts: 9
Joined: Mon Dec 06, 2021 1:34 pm

The Mutual Fund That Ate Wall Street—Based on an Index Few People Know About - The Wall Street Journal.

Post by Breaking Away »

[Post merged into here --admin LadyGeek]

Interesting article in today’s WSJ

“The Mutual Fund That Ate Wall Street—Based on an Index Few People Know About”

(Not sure if I am formatting link correctly)

https://www.wsj.com/articles/mutual-fun ... 1638390382

[Link formatted by admin LadyGeek]

“That fund is the $1.3 tril­lion (yes, tril­lion, in­clud­ing all share classes) Van­guard To­tal Stock Mar­ket In­dex Fund (VT­SAX) and its ex­change-traded-fund shares. The fund, from Van­guard Group, now ac­counts for 10% of all as­sets in U.S. stock mu­tual funds and ETFs in the mar­ket, ac­cord­ing to Morn­ingstar Inc. No other mu­tual fund or ETF comes close to it in as­set size. The next largest is an $821 bil­lion Van­guard S&P 500 in­dex fund.”

“The para­dox is that this big­gest beast among funds is tied to the most unas­sum­ing of stock in­dexes—the CRSP U.S. To­tal Mar­ket In­dex, de­vel­oped at the Uni­ver­sity of Chica­go’s Booth School of Busi­ness.”
GP813
Posts: 1231
Joined: Wed Dec 11, 2019 9:11 am

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by GP813 »

Since Tesla became a publicly traded company S&P 500 and Total US are essentially tied. So choosing the year 2020 specifically and the performance of one quirky company in one very unusual year seems arbitrary.
sycamore
Posts: 6309
Joined: Tue May 08, 2018 12:06 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by sycamore »

VTI wrote: Mon Dec 06, 2021 1:34 pm
alex_686 wrote: Mon Dec 06, 2021 10:03 am
VTI wrote: Mon Dec 06, 2021 9:29 am I believe the original poster was correct. That fund uses the "Fidelity U.S. Large Cap Index", not the S&P 500 index.
Never said it was.

The Fidelity U.S. Large Cap Index is created and maintained by Standards & Poor. Read the methodology.

Yes, it is a proprietary index that is sponsored by Fidelity but a fund sponsor is barred from actually running an index to avoid conflicts of interest.

My argument is that the S&P 500 has a investment committee run by S&P. And Fidelity U.S. Large Cap Index has a investment committee run by S&P. And that they are both large cap US domestic indexes with similar mandates. As such I strongly suspect that there is a fair amount of overlap between how the 2 committees are run.
I didn't realize that Fidelity contracts with S&P to maintain the Fidelity U.S. Large Cap Index. Thank you. Yes, there's no doubt that an S&P committee handles the Fidelity U.S. Large Cap Index, too.

You're more knowledgeable about this than I am, but in my mind, there's an important distinction to be made between the judgement-driven S&P 500 index and the more mechanical Fidelity U.S. Large Cap Index, which is described like this:
[...]the Fidelity U.S. Large Cap Index℠, which is a float-adjusted market capitalization-weighted index designed to reflect the performance of U.S. large capitalization stocks. Large capitalization stocks are considered to be stocks of the largest 500 U.S. companies based on float-adjusted market capitalization.
FWIW, the index's actual methodology is more detailed than that. See https://www.fidelity.com/bin-public/060 ... dology.pdf and search for "Fidelity U.S. Large Cap Index" to see some interesting tidbits:

"The index sponsor is Fidelity Investments. Fidelity Investments has appointed S&P Dow Jones as Index Calculation Agent to calculate and publish the indexes in accordance with this methodology document. The index sponsor may appoint an alternative Index Calculation Agent at any time."

Investment Universe is "Largest 3,000 U.S. companies based on float-adjusted market cap*" and the footnote is "* Based on full list of stocks that meets liquidity and investability constraints; process detailed in Section 2."

Section 2 shows that some securities are excluded; there are availability, and liquidity / investability screens.

Section 3 shows how they construct the index given the universe & screens. Theoretically there could be less than 500 stocks in the Large Cap Index if enough stocks didn't pass the screen.

And other stuff...
VTI wrote: Mon Dec 06, 2021 1:34 pm When I hear "committee-driven", I understand there to be an implication that human judgement is involved. I believe burritoLover has the same understanding.
Generally speaking, still no. All indexes uses committees because there are always subjective decisions to be made. Give me any set of rules based construction and I can talk about why it would fail. For example, how would you define "liquid" and "free float"? I can point to many examples where big liquid stocks causes problems.
Yes, there are subjective decisions involved in defining the mechanical rules. That's applies for both the S&P 500 index and the Fidelity U.S. Large Cap Index. However, in addition to those subjective definitions, the S&P 500 index is judgement-driven, whereas the Fidelity U.S. Large Cap Index is not. That's my understanding, at least.
So there are some subjective decisions you're okay with, and others you aren't. Not that I'm disagreeing :)
VTI
Posts: 282
Joined: Sat Mar 13, 2021 11:56 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by VTI »

sycamore wrote: Mon Dec 06, 2021 2:22 pm [...]

So there are some subjective decisions you're okay with, and others you aren't. Not that I'm disagreeing :)
I'd say that statement is true for everyone!

However, in case it wasn't clear: For several reasons, I'm a big fan of the S&P 500 index. I've even been considering replacing VTSAX (VTI) with VFIAX (VOO, S&P 500).
Also available as an Admiral™ Shares mutual fund.
Ferdinand2014
Posts: 2390
Joined: Mon Dec 17, 2018 5:49 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by Ferdinand2014 »

alex_686 wrote: Mon Dec 06, 2021 9:09 am
Ferdinand2014 wrote: Sun Dec 05, 2021 9:59 pm
burritoLover wrote: Sun Dec 05, 2021 10:33 am I'm curious if there is any "500" indexes that are simply the 500 largest companies by market cap evaluated say, once a year. It isn't like we are talking about illiquid small stocks here. It seems you could have a simple rules-based strategy to handle this. Maybe that is what the CSRP is.

Yes:
https://fundresearch.fidelity.com/mutua ... /315911628
No.

Specifically, the fund you are citing uses a committee. In fact the index has been farmed out to S&P so I might guess it is the same committee running the more popular one.

Generally speaking, still no. All indexes uses committees because there are always subjective decisions to be made. Give me any set of rules based construction and I can talk about why it would fail. For example, how would you define "liquid" and "free float"? I can point to many examples where big liquid stocks causes problems.
No? The index is not based on the 500 largest cap weighted U.S. companies using the rules outlined by the index?
“You only find out who is swimming naked when the tide goes out.“ — Warren Buffett
Jack&Warren disciple
Posts: 120
Joined: Sat Sep 11, 2021 9:46 am

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by Jack&Warren disciple »

nisiprius wrote: Sun Dec 05, 2021 9:54 am Searching for the "best" is just going to keep you in a state of anxiety and constantly changing your portfolio when what you should be doing is leaving it alone. John C. Bogle wrote
Successful investing involves doing just a few things right and avoiding serious mistakes.
A total stock market index fund is an excellent way to capture U.S. equity returns. It's what I use myself (in my case, VTSAX which is the mutual fund version of VTI). It's what I recommend.

The article is really about the CRSP index specifically, and it's interesting. Several big firms offer total stock market index funds, tracking different providers' total stock market index funds, and I feel there's no compelling reason to choose one over another. CRSP, Dow Jones Total Stock Market Index (not the Dow Jones Industrial Average, which is completely different), Russell 3000... pretty much the same.

There must be well over a dozen offering S&P 500 index funds. There is no compelling reason for choosing total market funds over S&P 500 funds. There is no compelling reason for choosing S&P 500 funds over total market funds. Wherever there is a difference, people will have differing opinions on what is "best." The important things to keep in mind are:
  • A low-cost total market or S&P 500 index fund is an excellent way to capture U.S. equity returns.
  • The difference doesn't matter much.
  • Nobody can really say which is "best."
Here is a chart of Total Stock--VTSMX, the oldest version of the fund--and VFINX, the oldest S&P 500 fund--showing five year moving averages.

a) they've never been very far apart,
b) the small differences have leapfrogged each other.

Image

As you see, the biggest differences ever were:

4/1994 through 3/1999: Total market fund 23.31%, S&P 500 fund 26.16%
4/2001 through 3/2006: Total market fund 5.77%, S&P fund 3.84%

So average annual returns averaged over 60-month periods have never been more than about 2% different from each other.

But meanwhile, the difference between the worst and best five-year periods for the S&P 500 fund were:

3/2004 through 2/2009: -6.73%
1/1995 through 12/1999: 28.49%

Your choice of which index to use potentially could have made a difference of 2%.
The fluctuations of the stock market made a difference 35%.

We can see this even in the most recent data. The article mentions that the total stock market index beat the S&P 500 "partly because [of] Tesla." But over the past 60 months, the S&P 500 ETF, VOO, beat, or perhaps we should say tied, VTI:

Image

As for "CRSP versus rivals," here are four funds tracking three different total market indexes. Vanguard tracks the CRSP total market index; Fidelity and Schwab use the Dow Jones Total Stock Market Index (not the Dow Jones Industrial Average); iShares uses the Russell 3000.

Source

Image
Sounds like I might be splitting hairs here, but your most excellent chart I thought indicated CAGR of 3 basis points HIGHER for VTI vs. Fidelity. Now would that make a difference if you are talking a 7 or 8 digit amount to invest say over a 1-3 decade period?

I think that you, Jack, and Warren clearly hit the nail on the head. What's MOST IMPORTANT is investing long term in a market index low cost fund and it's a cherry on top if it's tax deferred or otherwise tax advantaged.
User avatar
LadyGeek
Site Admin
Posts: 95466
Joined: Sat Dec 20, 2008 4:34 pm
Location: Philadelphia
Contact:

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by LadyGeek »

Breaking Away, Welcome! I merged your post into the ongoing discussion.
Wiki To some, the glass is half full. To others, the glass is half empty. To an engineer, it's twice the size it needs to be.
alex_686
Posts: 13286
Joined: Mon Feb 09, 2015 1:39 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by alex_686 »

burritoLover wrote: Mon Dec 06, 2021 12:15 pm Even if Tesla said that the carbon credits were going to make up a smaller percentage of profits going forward, I'm not sure what relevance that has. That would mean that the S&P committee decided they could predict the future profitability of EV sales for Tesla or, at the very least, are making active decisions on which earnings count and which do not? Seems kind of ridiculous.
S&P was not speculating.

Tesla specifically said that the sale of credits to Ford was a one-time deal that that is should not be considered part of their on-going operations. From a accounting standpoint you have to disclose what revenue is coming from ongoing operations verse one off. Tesla declared it as a one-off, just like they would if they sold their headquarters or some other major asset.
burritoLover wrote: Mon Dec 06, 2021 12:15 pm Many of the questionable dot coms weren't profitable though.
Once again I am going to have to disagree with you. Lots of them were fantastically profitable - often with one-off transactions that were in the same class as the Tesla/Ford transaction. There were some pretty aggressive timings of income recognition to ensure that companies got into the S&P 500. To a certain extent you can manufacture earnings over the short run. There are so many great stories from the dot.com era. So after that debacle S&P rewrote the rules.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
Jack&Warren disciple
Posts: 120
Joined: Sat Sep 11, 2021 9:46 am

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by Jack&Warren disciple »

nisiprius wrote: Sun Dec 05, 2021 10:23 am A few words about the S&P 500's committee selection, because I feel this is often distorted a bit, particularly by people talking against index funds in general.

The S&P 500 is an index of five hundred "leading companies in leading industries." That means 505 stocks because five of them issue more than one share class.

The reason for it being "leading companies in leading industries" and not "the 500 companies with the largest market cap" is because in 1957, people didn't think of market cap as a natural or important criterion. In a vague way people assumed that the big stocks of good companies were the mainstream market, and that little companies were a kind of sideshow for speculators.

This changed in 1981 when an economist, Rolf Banz, published a paper that reported that the stocks of "small firms" had had higher risk-adjusted returns than the market as a whole. Problems in the data set exaggerated the size of the effect. Anyway, this touched off almost a fetish for small-cap stocks. People began to think of small-caps, mid-caps, and large-caps as obvious, natural categories.

S&P was not intending to produce a curated set of stocks, or, at least, not a cleverly selected one. They basically wanted a) an index that could be computed hourly, that b) covered as much of the market as possible. In 1957, 500 stocks was the limit. In fact,"The Datatron machine that computes the new index [is] in the Boston laboratories of Melpar, Inc," there apparently not being anything suitable in New York. But they had to choose the stocks somehow.

The goal was for the S&P 500 to include at least 90% of the market by capitalization, and initially it actually included more than 94%.
Your historical knowledge in this area is excellent and enlightening thank you :D
VTI
Posts: 282
Joined: Sat Mar 13, 2021 11:56 pm

Re: "The Mutual Fund That Ate Wall Street—Based on an Index Few People Know About": Debating VTSAX

Post by VTI »

Northern Flicker wrote: Mon Dec 06, 2021 9:27 am [...]

I don't know where you got the purported 3.6% fee number from, but the index being tracked does not pay a fee to add a stock to the index, and VTI/VTSAX does a very good job of tracking its index. Costs are very low, and what costs there are typically are covered by securities lending revenue.
I took an extreme liberty referring to the 3.6% front-running penalty suffered by VTI as a "fee".
Last edited by VTI on Tue Dec 07, 2021 1:10 am, edited 1 time in total.
Also available as an Admiral™ Shares mutual fund.
User avatar
burritoLover
Posts: 4097
Joined: Sun Jul 05, 2020 12:13 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by burritoLover »

alex_686 wrote: Mon Dec 06, 2021 6:04 pm
burritoLover wrote: Mon Dec 06, 2021 12:15 pm Even if Tesla said that the carbon credits were going to make up a smaller percentage of profits going forward, I'm not sure what relevance that has. That would mean that the S&P committee decided they could predict the future profitability of EV sales for Tesla or, at the very least, are making active decisions on which earnings count and which do not? Seems kind of ridiculous.
S&P was not speculating.

Tesla specifically said that the sale of credits to Ford was a one-time deal that that is should not be considered part of their on-going operations. From a accounting standpoint you have to disclose what revenue is coming from ongoing operations verse one off. Tesla declared it as a one-off, just like they would if they sold their headquarters or some other major asset.
burritoLover wrote: Mon Dec 06, 2021 12:15 pm Many of the questionable dot coms weren't profitable though.
Once again I am going to have to disagree with you. Lots of them were fantastically profitable - often with one-off transactions that were in the same class as the Tesla/Ford transaction. There were some pretty aggressive timings of income recognition to ensure that companies got into the S&P 500. To a certain extent you can manufacture earnings over the short run. There are so many great stories from the dot.com era. So after that debacle S&P rewrote the rules.
If the S&P committee was willing to throw out 4 quarters due to Tesla’s regulatory credits, it begs the question why did they add Tesla to the S&P 500 after the 5th quarter in which Tesla earned a slightly lower $397 million from the same credits and only $331 million in net income?
Northern Flicker
Posts: 15288
Joined: Fri Apr 10, 2015 12:29 am

Re: "The Mutual Fund That Ate Wall Street—Based on an Index Few People Know About": Debating VTSAX

Post by Northern Flicker »

VTI wrote: Mon Dec 06, 2021 9:31 pm
Northern Flicker wrote: Mon Dec 06, 2021 9:27 am [...]

I don't know where you got the purported 3.6% fee number from, but the index being tracked does not pay a fee to add a stock to the index, and VTI/VTSAX does a very good job of tracking its index. Costs are very low, and what costs there are typically are covered by securities lending revenue.
I took an extreme liberty referring to the 3.6% front-running penalty suffered by VTI as a "fee".
Was the 3.6% number reported by an active manager with active funds available to sell to you?

Total market indices have investability screens that only include stocks when they meet certain liquidity and other requirements. Without that, and index fund would be pushing the price up on itself trying to buy at higher volume than the market supported, and without enough liquidity for the market to establish a fair price efficiently.

And for everyone one of these that goes on to be a good performer, there are probably 8-10 or even 25 duds that go nowhere and are a drag on returns. Moreover, some will just become illiquid and the fund will just have to keep them in the portfolio as dead capital. Do you really want the index fund in which you invest to be a graveyard for illiquid, zombie securities? Bot only would they be dead capital, but they would accumulate over time and be a drag on admin cost.

I think the investability screens for a total market index increase returns, certainly on a risk-adjusted basis, but probably on an absolute basis as well. But the tradeoffs between liquidity risk and return of ultra-micro-caps is going on at the margin of a cap weighted portfolio. So not only is the magnitude of the overall effect unlikely to be as high as 3.6% (whether positive or negative) it is a miniscule fraction of the portfolio.
VTI
Posts: 282
Joined: Sat Mar 13, 2021 11:56 pm

Re: "The Mutual Fund That Ate Wall Street—Based on an Index Few People Know About": Debating VTSAX

Post by VTI »

Northern Flicker wrote: Tue Dec 07, 2021 3:11 pm
VTI wrote: Mon Dec 06, 2021 9:31 pm
Northern Flicker wrote: Mon Dec 06, 2021 9:27 am [...]

I don't know where you got the purported 3.6% fee number from, but the index being tracked does not pay a fee to add a stock to the index, and VTI/VTSAX does a very good job of tracking its index. Costs are very low, and what costs there are typically are covered by securities lending revenue.
I took an extreme liberty referring to the 3.6% front-running penalty suffered by VTI as a "fee".
Was the 3.6% number reported by an active manager with active funds available to sell to you?
The 3.6% was not reported by an active manager with active funds. It's from the linked WSJ article, and it was credited to FactSet, a financial data company.
Also available as an Admiral™ Shares mutual fund.
Northern Flicker
Posts: 15288
Joined: Fri Apr 10, 2015 12:29 am

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by Northern Flicker »

With the linked article behind a paywall, many of us could not read it. But FactSet's customers are in fact investment banks, active managers, etc. so still not an unbiased source.

Total market index funds have investability screens to improve returns net of cost. You can only get that 3.6% by taking one effect out of context and not considering the cost of trying to harvest that missed premium. And there is survival bias in that 3.6% number. That 3.6% is the missed premium on the microcaps that make it. How many that were filtered out of the index ultimately went bankrupt? That was the point of my posting. The 3.6% number does not incorporate losses avoided in the aggregation, only gains missed. It is a bogus number.
Northern Flicker
Posts: 15288
Joined: Fri Apr 10, 2015 12:29 am

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by Northern Flicker »

burritoLover wrote: If the S&P committee was willing to throw out 4 quarters due to Tesla’s regulatory credits, it begs the question why did they add Tesla to the S&P 500 after the 5th quarter in which Tesla earned a slightly lower $397 million from the same credits and only $331 million in net income?
You are blurring two decisions into one. The first decision was to add Tesla to the index. A second decision is the timing of the addition. The first decision considers whether they meet the requirements for membership. The second decision is to prevent front-running of the index change by adding some unpredictability to the timing.
User avatar
burritoLover
Posts: 4097
Joined: Sun Jul 05, 2020 12:13 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by burritoLover »

Northern Flicker wrote: Tue Dec 07, 2021 9:11 pm
burritoLover wrote: If the S&P committee was willing to throw out 4 quarters due to Tesla’s regulatory credits, it begs the question why did they add Tesla to the S&P 500 after the 5th quarter in which Tesla earned a slightly lower $397 million from the same credits and only $331 million in net income?
You are blurring two decisions into one. The first decision was to add Tesla to the index. A second decision is the timing of the addition. The first decision considers whether they meet the requirements for membership. The second decision is to prevent front-running of the index change by adding some unpredictability to the timing.
It’s announced ahead of time - like 5 weeks or something.
Northern Flicker
Posts: 15288
Joined: Fri Apr 10, 2015 12:29 am

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by Northern Flicker »

No it is not announced 5 weeks in advance. It is announced on a Monday or Tuesday with the change taking effect the following Thursday or Friday.

In the past, S&P announced changes on arbitrary Wednesdays after market close with the index change taking effect at the next market open. Portfolio managers for S&P500 funds had to work into the evening to get their trades submitted to brokers in the queue ahead of front-runners for execution at market open.
sycamore
Posts: 6309
Joined: Tue May 08, 2018 12:06 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by sycamore »

Northern Flicker wrote: Tue Dec 07, 2021 10:58 pm No it is not announced 5 weeks in advance. It is announced on a Monday or Tuesday with the change taking effect the following Thursday or Friday.

In the past, S&P announced changes on arbitrary Wednesdays after market close with the index change taking effect at the next market open. Portfolio managers for S&P500 funds had to work into the evening to get their trades submitted to brokers in the queue ahead of front-runners for execution at market open.
FWIW, here's the S&P methodology document.

It does mention a "five week" period for the S&P Total Market Indexing (which S&P 500 is based on)
The index is reconstituted annually, after the close of the third Friday in September. The index also rebalances quarterly on the third Friday of each calendar quarter as detailed in the index construction section. For both the annual reconstitution and quarterly rebalancing, the reference date to meet the eligibility criteria is five weeks prior to the effective date.
So it sounds like S&P is basing index decisions on data from five weeks prior to the effective date, but the announcement of those decision presumably is kept secret and not revealed until close to the effective date?

I see "Announcements are available to the public via our Web site, www.spdji.com, before or at the same time they are available to clients or the affected companies." but nothing specifically about days of the week. Maybe that's more of a policy thing than a "methodology" thing.
alex_686
Posts: 13286
Joined: Mon Feb 09, 2015 1:39 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by alex_686 »

Northern Flicker wrote: Tue Dec 07, 2021 10:58 pm In the past, S&P announced changes on arbitrary Wednesdays after market close with the index change taking effect at the next market open. Portfolio managers for S&P500 funds had to work into the evening to get their trades submitted to brokers in the queue ahead of front-runners for execution at market open.
That is not how you front run. If you put in your orders for the open you have already been front runned.

What you do is you buy (or short) the stock before S&P made their announcement. Which is not that hard to do. Ignoring the complex situation of Tesla the rules of S&P are pretty clear. In the good old days, the stocks would jump (or drop) by a good 10% from pre-announcement to post.

It was a easy way to make money. Not so much these days.
Former brokerage operations & mutual fund accountant. I hate risk, which is why I study and embrace it.
Northern Flicker
Posts: 15288
Joined: Fri Apr 10, 2015 12:29 am

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by Northern Flicker »

sycamore wrote: Wed Dec 08, 2021 8:37 am
Northern Flicker wrote: Tue Dec 07, 2021 10:58 pm No it is not announced 5 weeks in advance. It is announced on a Monday or Tuesday with the change taking effect the following Thursday or Friday.

In the past, S&P announced changes on arbitrary Wednesdays after market close with the index change taking effect at the next market open. Portfolio managers for S&P500 funds had to work into the evening to get their trades submitted to brokers in the queue ahead of front-runners for execution at market open.
FWIW, here's the S&P methodology document.

It does mention a "five week" period for the S&P Total Market Indexing (which S&P 500 is based on)
The index is reconstituted annually, after the close of the third Friday in September. The index also rebalances quarterly on the third Friday of each calendar quarter as detailed in the index construction section. For both the annual reconstitution and quarterly rebalancing, the reference date to meet the eligibility criteria is five weeks prior to the effective date.
So it sounds like S&P is basing index decisions on data from five weeks prior to the effective date, but the announcement of those decision presumably is kept secret and not revealed until close to the effective date?

I see "Announcements are available to the public via our Web site, www.spdji.com, before or at the same time they are available to clients or the affected companies." but nothing specifically about days of the week. Maybe that's more of a policy thing than a "methodology" thing.
The total market index and S&P500 are different indices with different index maintenance procedures.
User avatar
burritoLover
Posts: 4097
Joined: Sun Jul 05, 2020 12:13 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by burritoLover »

Northern Flicker wrote: Tue Dec 07, 2021 10:58 pm No it is not announced 5 weeks in advance. It is announced on a Monday or Tuesday with the change taking effect the following Thursday or Friday.

In the past, S&P announced changes on arbitrary Wednesdays after market close with the index change taking effect at the next market open. Portfolio managers for S&P500 funds had to work into the evening to get their trades submitted to brokers in the queue ahead of front-runners for execution at market open.
S&P announced that Tesla was going to be added on Nov 17, 2020. It was added to the index on Dec 21, 2020. Regardless, the market is efficient enough to prevent front running. If it wasn't, then anyone should be able to make money on listening to positive news on earnings reports.
Northern Flicker
Posts: 15288
Joined: Fri Apr 10, 2015 12:29 am

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by Northern Flicker »

alex_686 wrote: Wed Dec 08, 2021 9:32 am
Northern Flicker wrote: Tue Dec 07, 2021 10:58 pm In the past, S&P announced changes on arbitrary Wednesdays after market close with the index change taking effect at the next market open. Portfolio managers for S&P500 funds had to work into the evening to get their trades submitted to brokers in the queue ahead of front-runners for execution at market open.
That is not how you front run. If you put in your orders for the open you have already been front runned.

What you do is you buy (or short) the stock before S&P made their announcement. Which is not that hard to do. Ignoring the complex situation of Tesla the rules of S&P are pretty clear. In the good old days, the stocks would jump (or drop) by a good 10% from pre-announcement to post.

It was a easy way to make money. Not so much these days.
The trades made to try to front-run ahead of the announcement are based on predictions of stocks to be added or booted out. Most index changes are not like adding Google or Tesla, but are dropping a stock that no longer meets the membership requirement, and do not make hradlines in the financial press.

When index changes were announced at market close on a Weds, to take effect the following market open, if a fund manager did not focus on the timeliness of their trades, they would get front-runned on the trades for the actual changes once announced. Even among the S&P 500 funds there would be competition to trade first.
Northern Flicker
Posts: 15288
Joined: Fri Apr 10, 2015 12:29 am

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by Northern Flicker »

burritoLover wrote: Wed Dec 08, 2021 1:14 pm
Northern Flicker wrote: Tue Dec 07, 2021 10:58 pm No it is not announced 5 weeks in advance. It is announced on a Monday or Tuesday with the change taking effect the following Thursday or Friday.

In the past, S&P announced changes on arbitrary Wednesdays after market close with the index change taking effect at the next market open. Portfolio managers for S&P500 funds had to work into the evening to get their trades submitted to brokers in the queue ahead of front-runners for execution at market open.
S&P announced that Tesla was going to be added on Nov 17, 2020. It was added to the index on Dec 21, 2020. Regardless, the market is efficient enough to prevent front running. If it wasn't, then anyone should be able to make money on listening to positive news on earnings reports.
The date it was to be added was not announced Nov 17.

Correction, the date apparently was announced with more than the standard lead time for the Tesla addition, presumably due to the visibility of the change and volume of trades to take place.

It still was not based on S&P making an active management decision.
Last edited by Northern Flicker on Wed Dec 08, 2021 1:51 pm, edited 3 times in total.
User avatar
burritoLover
Posts: 4097
Joined: Sun Jul 05, 2020 12:13 pm

Re: Is VTI (based on CRSP U.S. Total Market Index) the best way to capture U.S. equity returns?

Post by burritoLover »

Northern Flicker wrote: Wed Dec 08, 2021 1:20 pm
burritoLover wrote: Wed Dec 08, 2021 1:14 pm
Northern Flicker wrote: Tue Dec 07, 2021 10:58 pm No it is not announced 5 weeks in advance. It is announced on a Monday or Tuesday with the change taking effect the following Thursday or Friday.

In the past, S&P announced changes on arbitrary Wednesdays after market close with the index change taking effect at the next market open. Portfolio managers for S&P500 funds had to work into the evening to get their trades submitted to brokers in the queue ahead of front-runners for execution at market open.
S&P announced that Tesla was going to be added on Nov 17, 2020. It was added to the index on Dec 21, 2020. Regardless, the market is efficient enough to prevent front running. If it wasn't, then anyone should be able to make money on listening to positive news on earnings reports.
The date it was to be added was not announced Nov 17.
Irrelevant.
Post Reply