"I see dead people" - and they are the best investors!

No, that doesn't make sense to me. One person is selling shares in the fund, one person is purchasing them. Doesn't matter if the latter were invested in the fund already beforehand, or not. But let's hear what others think. Wouldn't be the first time I was wrong.

I'd like to hear from others too. But I think the key is, selling some of a fund does not affect the fund's price - the fund is a reflection of the sum of the stocks holdings. You don't sell to other fund holders, you sell to the total market. Buyers buy from the total market.

Picture a neighborhood of 100 houses. If the housing market drops, and one person sells at a loss (to someone from out of town), that one sale doesn't change the value of the neighborhood. The value is whatever the value is that day, or the day before, or the next day. The one person realized a loss, but it had no effect on the value of the neighborhood.

Then another person sells 2 years later (to someone from out of town), when the housing market recovered. He realized a gain, and the earlier person realized a loss. Neither had any effect on the overall market. Each was just a transaction


-ERD50
 
Um, every time one person sells, another person buys, right? If the average (median?) investor trails the fund's performance, that can only mean many small fish sell to few large sharks in times of distress. That, or transaction costs.

According to Morningstar, the average fund investor receives less than the average fund return. They developed a measurement for that, called “investor returns gap”. Their fund statistics allow us to compare “fund returns” with “investor returns”. They also publish a yearly analysis of this, and rank funds and fund categories.

The reason the average investor receives less than the fund average is volatility and investor timing, which reduces return. This would lend credence to the OP thesis that average investor returns are higher when the investors don’t “buy and sell” vs “buy and hold”. It’s not a surprise that when M* ranks fund categories by the size of the investor gap, allocation funds have the smallest gap.

Here are a couple of links to M* articles on this https://www.morningstar.com/blog/2018/06/12/investor-returns.html and https://www.morningstar.com/blog/2019/08/30/investor-returns-fund.html
 

Don't have an M* account, so I cannot read your links unfortunately.

According to Morningstar, the average fund investor receives less than the average fund return.

Okay, so instead of buying and holding (and consequently enjoying fund-average returns), the average fund investor buys (somewhat) high and sells (somewhat) low. Who does he buy from, and who does he sell to? And what does this mean for his counterpart(s), who buy(s) low and sell(s) high?
 
Okay, so instead of buying and holding (and consequently enjoying fund-average returns), the average fund investor buys (somewhat) high and sells (somewhat) low. Who does he buy from, and who does he sell to? And what does this mean for his counterpart(s), who buy(s) low and sell(s) high?
Fund asset levels are not static, they can increase or decline. When a fund investor sells, the buyer is always outside the fund, and when a fund investor buys, it always represents an increase in the total fund assets.

The M* measure is based on fund asset levels. That is, they measure the return of a fund against changes in total fund assets. The measurement is pretty clean.
 
Picture a neighborhood of 100 houses. If the housing market drops, and one person sells at a loss (to someone from out of town), that one sale doesn't change the value of the neighborhood. The value is whatever the value is that day, or the day before, or the next day. The one person realized a loss, but it had no effect on the value of the neighborhood.

Then another person sells 2 years later (to someone from out of town), when the housing market recovered. He realized a gain, and the earlier person realized a loss. Neither had any effect on the overall market. Each was just a transaction

-ERD50

IMO it doesn't matter at all who you sell to, or whether that transaction influences the price of the commodity in question. I would always assume that none of my transactions move the market, but then I'm a small fish. :LOL:

Okay, so in your example, all owners who did not sell got 'fund-average' returns, meaning their homes appreciated at the 'market rate'. Seller #1 got less, because they sold low.
Your example ignores the following: Buyer #1 got excess returns by buying low! They may not have been home owners ('fund investors') before, but they became such with the purchase. And if they sell now or in the future, they will have gotten a better return than both the original buy-and-hold owners, and seller #1.

Makes sense?
 
When a fund investor sells, the buyer is always outside the fund, and when a fund investor buys, it always represents an increase in the total fund assets.

Those two statements seem to contradict each other. Are you trying to say that shares in the fund are not traded between individual investors, but first redeemed to the issuer by the seller, and then re-issued as new shares to a buyer?
 
As I understand it:

With a traditional mutual fund, all shares are bought and sold at close-of-business NAV with the fund manager acting as principal. Where there is an imbalance, the manager will buy or sell stocks as necessary.

With an ETF, most trades are between shareholders. Where there is an imbalance, a player called an "authorized participant" will create or destroy shares by buying or selling stocks. Personally I don't like this. The Authorized Participant has zero loyalty to the shareholders. only loose loyalty to the fund, and intense loyalty to its own P&L. In my experience whenever financial goals are not aligned, bad things can happen. I can see this potential when the market is moving quickly. I have no facts, only business experience and consequent gray hair, to support this worry.
 
Those two statements seem to contradict each other. Are you trying to say that shares in the fund are not traded between individual investors, but first redeemed to the issuer by the seller, and then re-issued as new shares to a buyer?
I am referring to open ended mutual fund. Yes, shares are not traded between individual investors. When you buy, you transfer money to the fund manager, who then goes to the market and buys assets. The net assets of the fund increase. When you redeem, you inform the manager, who then sell some fund assets and transfer the proceeds to you. The fund assets decline.
 
No, that doesn't make sense to me. One person is selling shares in the fund, one person is purchasing them. Doesn't matter if the latter were invested in the fund already beforehand, or not. But let's hear what others think. Wouldn't be the first time I was wrong.

The (open ended) mutual fund price is determined by underlying share prices at market close, not directly by the number of people buying and selling the fund. Far more investors are buying and selling the underlying shares in the open market, muting the impact of the fund operations.
 
Last edited:
I am referring to open ended mutual fund. Yes, shares are not traded between individual investors. When you buy, you transfer money to the fund manager, who then goes to the market and buys assets. The net assets of the fund increase. When you redeem, you inform the manager, who then sell some fund assets and transfer the proceeds to you. The fund assets decline.

Thanks for the explanation. I agree; under these circumstances the average investor can underperform.
 
As I understand it:

With a traditional mutual fund, all shares are bought and sold at close-of-business NAV with the fund manager acting as principal. Where there is an imbalance, the manager will buy or sell stocks as necessary.

With an ETF, most trades are between shareholders. Where there is an imbalance, a player called an "authorized participant" will create or destroy shares by buying or selling stocks. Personally I don't like this. The Authorized Participant has zero loyalty to the shareholders. only loose loyalty to the fund, and intense loyalty to its own P&L. In my experience whenever financial goals are not aligned, bad things can happen. I can see this potential when the market is moving quickly. I have no facts, only business experience and consequent gray hair, to support this worry.

When an ETF is popular and the demand for it is high, this may cause the ETF price to be higher than the sum of its constituent stocks. The ETF then would be traded at a premium, not unlike the situation with a favored closed-end fund. The way to keep the ETF price fair is to create new shares, by buying shares of the individual constituent stocks, then assemble them into the new ETF shares to meet this demand.

Conversely, if an ETF is traded below its fair price, meaning it is valued below the sum of its parts, then it makes sense to buy the ETF share to disassemble it and sell the parts.

The creation/redemption of ETF shares is to make sure that the price of the ETF truly represents the sum of its parts. There's nothing sinister about it. Without this mechanism, an ETF may trade below its fair value, or above it. Traditional closed-end funds have this problem of trading at a discount to fair value, or a premium to fair value. An ETF should trade at fair value, not differently than MFs at NAV.

This share creation/redemption mechanism has been done ever since the creation of the granddaddies of the current ETFs, which were called SPDR's (Standard & Poor's Depositary Receipts). I started investing in these back in the late 1990s.
 
Last edited:
.. There's nothing sinister about it. ...
I didn't say there was. My point is that the Authorized Participants' financial objective are not well aligned with those of the fund manager or of the fund shareholders. I learned long ago that situations like this can be problematical. "Good will" is not a strong glue for holding relationships together. Well-aligned financial objectives, OTOH, are strong glue.

This structure may never cause problems, but I think there is potential for problems in fast-moving markets. This function is really a sort arbitrage. Arbitrage can be hazardous when things are moving fast.

I would prefer to see the fund manager handling this function, just as the fund manager handles it in traditional mutual funds.
 
... This structure may never cause problems, but I think there is potential for problems in fast-moving markets. This function is really a sort arbitrage. Arbitrage can be hazardous when things are moving fast...

This mechanism is indeed an arbitrage, and it is intentional. The purpose is to keep the ETF price fixed to the price of its component stocks.

I will admit that I have not read prospectus of the recent ETFs, but the earlier SPDR's prospectus said that anyone could assemble a collection of individual stocks in the right mix, and deliver to the trustee in exchange for the ETF shares. And conversely, anyone could buy the ETF shares, turn them in to the trustee and demand a conversion into individual shares to be delivered back.

While I would not be able to keep track of the price of my ETFs in real time to see if it tracks the individual stocks, I would think many brokerages would use computers to detect any mispricing between the ETFs and individual stocks, and immediately do a simultaneous buy/sell as appropriately to arbitrage away any mispricing.

The above would ensure that I get fair value when I hold these ETFs. I would not get any premium, but not suffer from any discount either. And this is all good.

I would prefer to see the fund manager handling this function, just as the fund manager handles it in traditional mutual funds.

The ETFs can be traded in the open market, and not going through the MF manager. So, I do not know how an ETF and MF can be handled similarly.

My discussion so far has been about SPDRs or ETFs that are passive indexed funds. The constituents are defined and declared in advance for anyone to do fair pricing. When I held some of these early SPDRs, I even had the voting rights to the constituent shares.

Back then, some sector SPDRs had only 20 or so companies in them. A telecommunication sector ETF may define 100 shares of the ETF as consisting of 20 shares of ATT, 15 shares of Verizon, 10 shares of T-Mobile, etc... And because of the legal structure, you had to buy these ETFs in round lots of 100 only. Else, you would have fractional ownership of the shares of the constituent companies.

The above SPDRs have no manager, only a sponsor and a trustee.

Anyway, they have now actively managed ETFs, whose composition is subject to change by their manager. However, they are transparent, meaning that the composition is updated daily, in contrast with MFs that only disclose their holdings periodically.

Can shenanigans been done with these actively managed ETFs? I don't know.


PS. I just now recall that the old ETFs that could be traded only in round lots of 100 shares were called HLDR ("holder"). These HLDRs are grantor trusts, distribute dividends directly to shareholders, who also retain the voting right of the shares. These HLDRs may be too cumbersome to maintain, and they do not exist anymore.
 
Last edited:
"I see dead people" - and they are the best investors!

Um, every time one person sells, another person buys, right? .



That is true for individual stocks, not shares of an open end mutual fund. You are selling back to the fund so maybe your large shark analogy still applies.

Oops. I see this was covered very well in several posts above.
 
Last edited:
Back
Top Bottom