PF Investing Club: The Stock Market & Compounding Interest

[russ_watters]

Here's some quotes from "The Only Investment Guide You'll Ever Need" (2002)
Current ed, 2011:
https://www.amazon.com/dp/0547447256/?tag=pfamazon01-20

From the chapter entitled "Choosing (to Ignore) Your Broker"

There are no brokers who can beat the market consistently and by enough of a margin to more than make up for their brokerage fees...

[and even if there are and are willing to work for you, you small-timer]...there's no way for you to know who they are.

By and large, you should manage your own money (via no-load mutual funds).

[emphasis included]

The book includes lots of details and statistics about the [lack of] performance of managed funds to back up this advice. Including that - from actual tests - you are probably better off having a literal dartboard or random number generator manage your money than a broker/fund manager.

 

[Stephen Tashi]

I've heard so many landlord horror stories to keep me away from property.

I've never met anyone who was interested in making money.

By that I mean that I've never met anyone who was interested in making money and didn't have some other side conditions on how the money was to be made. I've known people who were interested in the stock market, but not real estate and people interested in real estate who didn't trust the stock market - and people interested in real estate who would only buy land and never buildings. Then there are people interested in buying gold and silver - and the relatively fewer interested in both buying and selling gold and silver. Even if we restrict the ways of money to legal ways of making it, I've never met anyone who could free his mind of other considerations and just concentrate on the best way of making money that was at hand. I can't do it and don't want to. I wouldn't want to be a landlord.

 

[Greg Bernhardt]

No, but if the next Great Depression comes, you can still live in your house. They don't even issue stock certificates anymore that you could wrap yourself into keep warm if you were a renter and got evicted because you can no longer pay your rent!

The question is, can I pay off the mortgage before the next great depression comes

 

[russ_watters]

The actual future prices aren't set by anything in the present.

I know. Let me rephrase for clarity:
On the large scale, the present prices are set by a consensus collection of metrics and logic today and the and future prices are set by a similar consensus collection of metrics and logic at that time.

That implies that you accept the current market price as the most accurate prediction.

No, as I've said before, the current market price is the current value of the company, period. It doesn't have an "accuracy", because it isn't a prediction: it just is.

It doesn't match up with high frequency trading. It matches up with human scale short term trading.

No, it doesn't match up with either. You're letting the fact that there are a lot of investors confuse you here. I buy into a mutual fund every payday when my 401K money gets taken out of my paycheck (26 times a year). I *never* sell. "Traders" (high frequency or human day-traders) sell about as often as they buy. The fact that the mutual fund as a whole will on average buy 125 times a year and sell 125 times a year has no bearing on me. It does not affect the value of my fund. And there is no way for the high-frequency or day-traders to affect the value of my fund by "trading" against it. That's what we're discussing.

[2. They only need to settle-up the difference between yesterday's and today's deposits and withdrawals, not the actual trades.]
Is daily trading not "short term" trading?

You're not responding to what I said. I'm feeling like there is a focus issue here, like you're trying to get a "win" on a technicality rather than come to an understanding on the issue we're discussing. You do understand that a fund might both take in and put out thousands of trades worth millions of dollars in one day and then aggregate them into one trade worth a tiny fraction, right? You recognize that the frequency and value of the churn is orders of magnitude different between this and "day-traders" and "high frequency traders", right? By and large (by orders of magnitude difference in number and value), these entities are not trading against each other so they cannot win or lose money from each other. That's what we have been discussing.

 

[russ_watters]

Sure, however no one speaks of short term trading in that functionality. Short term trading is taking a position for a "short" duration me it minutes or days or even weeks with the hopes that you beat a correction. ETF maintain their positions but have to rebalance their leverage ratio otherwise you'll see a divergence between the ETF value and the underlining assets. In fact, you can even say that the rebalance is really there in order to maintain their functionality not for gains.

Yes, I think this is a key point: managed funds are trying to beat the market (and usually fail), whereas index funds are trying to reflect the market.

 

[Question_]

Bonds tend to be associated with governments; but, companies also issue them at a much higher yield than governments do. Just something to keep in mind for the risk-averse.

 

[Vanadium 50]

Stephen, I am still having trouble understanding your point.

Let's imagine an "infinite Treasury bill" that costs $100 and pays $5 each year forever. Now imagine a stock that pays $10 in dividends per share every year. Guaranteed. You have this on stone tablets from the mountain. How much is a share worth? Can we all agree $200?

Now in real life Treasuries last for a fixed time, but accountants can correct for that. Also in real life, the earnings per share vary, but again, accountants know how to correct for that. The basic idea is still the same - if I know the exact returns every year from now until a long time from now, everyone will agree on the price of the stock: it's the price at which you get the same return from an equivalent safe investment, like a Treasury.

Of course we don't know what the exact future returns will be. People have different ideas of what this will be, but given their assumptions, the "fair price" should be calculable. In a very real way, the fair price is single number that characterizies a complicated future return profile. And because we may differ on what the future return profile is, we disagree on the fair price.

The market provides a consensus price. I take my idea of what the future returns are, and calculate that the fair price for XYZ is $110. If it's trading for $100, I will buy it. If I buy a lot of it, maybe the price moves to $101. This is how the market is factoring in my beliefs on how XYZ will perform in the future. The "market price" is a measure of the consensus (in a statistical sense) of what the future returns will be.

Do we agree on all of this? If not, let's get on the same page before moving on.

Now, how do computers factor into this? They can do so in two ways - they can make better decisions or they can make faster decisions.

On the better decisions front, we are already in a situation where the individual investor is handicapped. The Big Boys employ analysts who spend their whole days looking at data that might influence future returns. The individual investor simply doesn't have these resources. Trying to beat the market by out-analyzing the analyzers is, in my view, a fool's errand, with or without computers.

On the faster front, new information needs to be processed by a human brain. If the annual report says "Instead of the usual $1/share dividend, this year it will be only 75 cents because we are investing in new technology, and we expect future dividends to be $1.25/share." Does this mean the stock should go up, or should it go down? Clearly it depends on the credibility of the claim that this new technology will produce a permanent increase in profitability, and that means some human being needs to think about it. And that will limit how fast this new information can be priced in - it will happen at human-scale times, not computer-scale times.

 

[Vanadium 50]

btw I also bought a condo in 2007. In 2015 I sold it for 30% bath. Not always safe.

But not necessarily stupid. Let's say it was $200K in 2007 and $140K when you sold it, and you put 20% down. You'd have lost $60K in the price, and gained $11K in equity, and have spent about $800 a month for the mortgage. So it cost you $1300/month. How does that compare with renting? Maybe better, maybe not, but a house is more than an asset.

 

[Stephen Tashi]

By and large (by orders of magnitude difference in number and value), these entities are not trading against each other so they cannot win or lose money from each other. That's what we have been discussing.

1) The topic of day trading is not my original question.
2) Day traders are not necessarily trading "against each other". You choose to imagine a scenario where they are.
3) High frequency trading programs make profits so that short time-span version of trading is profitable to computer programs. Whether you want to call that "short term trading" or "day trading" can be a vocabulary debate.
4) The fact that human day traders supposedly don't make profits would support the idea that computers can out perform humans in short term trades.

You're not responding to what I said. I'm feeling like there is a focus issue here, like you're trying to get a "win" on a technicality rather than come to an understanding on the issue we're discussing.

5) It's not me that has a fixation on bad-mouthing active human-managed mutual funds on the basis of what I read in one book. I don't know why that crusade interests you. It's not the topic that I proposed discussing. You seem to think my question implies that I advocate investing in actively managed funds or advocate short term trading. I don't advocate either as a general policy. I don't do short term trading. I have invested in a few actively managed funds and am pleased with their performance. As far as I can tell, the statistical inferiority of the collection of human-actively-managed mutual funds versus index funds has no bearing on my question about how the increasing use of computer programs will affect long term investing - except that it might be an argument that computers will dominate long term investing. Index funds are essentially executing an algorithm that is defined by an index. There can be different indexes. Following a particular index might be a good idea and it might be a bad idea. Humans have invented different indexes. Computer programs might discover better indexes to follow.

 

[russ_watters]

Great post, @Vanadium 50,

I of course agree on all the background logic, but may have a slightly different take on the prospects for computers in trading:

On the "better" front, you're absolutely correct that we as individual investors have already lost (and will continue to lose) and have no hope of winning against professionals over the long-haul when it comes to deciphering and making decisions on the data. But Stephen's question is: can it get worse? My answer: certainly. Their ability to decipher data can only get better as they learn and computers help them do more and more complex analysis. But more importantly: It doesn't matter. If you follow the advice I've referenced and decline to play against the professionals, you can't lose to them.

On the "faster" front, I only half agree. Humans write the algorithms/programs, and as such I believe you can program a computer to replace a human, doing the same "better" job also faster, giving an additional edge over individual investors when it comes to making [snap] decisions. Ultimately, a piece of news does get processed by people - that's how the consensus impact of the news is eventually reached - but it takes time.

Say, for example, a drug in trials gets a surprise rejection by the FDA. The company stock plunges rapidly on the news, then people regain their senses over the next few days and the stock recovers most of its losses. It seems to me that it should be possible for a computer to recognize this, and if nothing else be the first to sell when the bad news hits the AP wire...and maybe even know roughly when to buy-back if the psychology takes tanking stocks on a consistent roller-coaster.

But again, this doesn't really matter to most of us. We can't lose this game if we don't play...or actually, we win this game by not playing, just like the guy who watched his two friends at the casino instead of playing with them went home a winner (second place and with an above-average return).

 

[Stephen Tashi]

Stephen, I am still having trouble understanding your point.

Let's imagine an "infinite Treasury bill" that costs $100 and pays $5 each year forever. Now imagine a stock that pays $10 in dividends per share every year. Guaranteed. You have this on stone tablets from the mountain. How much is a share worth? Can we all agree $200?

No, and that's why economic theory is so complicated. It has to deal with "utility" as a separate concept from "price" I wouldn't pay $200 for such a treasury bill today if there was another bill selling for $150 that paid $11 yearly dividends. I wouldn't pay $200 for such a treasury bill if preferred to spend my $200 to buy something right now. If I owned such a treasury bill, I might not wish to sell it for $200.

Modeling how the market price for such a treasury bill is reached is an interesting mathematical problem. You have a population of buyers and sellers with different "utilities" and different reserves of capital or bonds to use in pursuing their goals. If we assume a lot of buying and selling takes place we can formulate a model where some (dynamic) equilibrium price is reached.

The market provides a consensus price. I take my idea of what the future returns are, and calculate that the fair price for XYZ is $110. If it's trading for $100, I will buy it. If I buy a lot of it, maybe the price moves to $101. This is how the market is factoring in my beliefs on how XYZ will perform in the future. The "market price" is a measure of the consensus (in a statistical sense) of what the future returns will be.

The market price, no doubt, is affected by market participants' predictions on the future. Your personal beliefs affect he market price insofar as you have resources to buy or sell the stock in the market. However, there is no unique way to look at the current price of a stock and interpret it as a specific prediction about the future value of a stock. The market price is the result of complex interaction among participants with differing predictions and differing measures of utility and differing assets to deploy in the market.

It is true that a person can compare a stock selling for $200 per share with the idealized treasury bill in your example and consider whether his personal preference would be for the stock or for the treasury bill.

The market price set by a population of human participants is not necessarily the same price that would be set by a population of computer programs. The volatility of the market price is also not necessarily the same. ( Whether the financial power behind a population of computer programs would be similar to the financial power behind today's human investors is an interesting question. In the different case of short term trading, the HFT programs appear to have large resources behind them. )

One of my questions is whether it is safe to assume the historical percentage returns on stocks will be relevant to a future stock market where the market participants that might have the largest resources are computer programs. (I really do understand - at least as well as anyone else in the thread - how stock markets work. So far, I haven't seen any coherent argument that begins "This is the way stock markets work" and ends with "therefore the participation of computers will not affect the percentage returns on stocks". (And I also havent' seen an argument that concludes "therefore the the participation of computers will affect the percentage returns". ))

 

[NTL2009]

NTL2009 said: ↑
Did I say it was "sound"? No. I only said "it was". ...

...

(It interesting to me that many people who don't accept consensus opinions about politics, nuclear power, global warming etc, are willing to defer to the consensus opinion of "the market" about stock prices. I haven't followed your (NTL2009) posts on diverse issues, so I don't know if you are such a maverick. I myself am not such a maverick that I'm determined to ignore consensus opinions on all subjects, but when it comes to "the market" I'll be a little skeptical.

I think this has been answered by others, but since you framed it in a direct Q to me, I will respond to that as well.

Some of my earlier responses may not have been worded as well as I intended, but the above statement is reflective of how I feel/think. The stock prices are what they are. I don't concern myself with whether they "right" or "wrong" (and don't think that personal investors should either).

The difference, one that I think you have been missing, is we are saying the price is an accurate reflection of the consensus, not that the consensus itself is accurate.

To your point about consensus on other issues, I'll throw out a make-believe hypothetical to avoid hot-button issues. If a well done, rigorous and replicated study found that 20% of the people of voting age in the US thought that Perpetual Motion machines would be providing all our energy needs by 2030, then I'd have to say that the study was correct. But those 20% are wrong. See the difference? The stock market reflects the opinions, right or wrong.

How is this relevant (which is all that matters)? I don't think it has any relevance to the personal investor. If you are in the accumulation phase, you are probably (and should be) making purchases each paycheck (called "dollar-cost-averaging" in). So if the market consensus was a little 'wrong' one day to the positive, and a little 'wrong' the next day to the negative it all averages out for you. Same concept on the way out, as a retiree drawing down their stash. And if it was always wrong in the same direction, it would wash out. Relax.

Now there are some studies about how the average P/E of the market can be a good long term indicator, and some people use this to adjust their allocation between stocks/bonds (but generally using broad-based index funds for each). I personally don't subscribe to this, I think the market can be 'wrong' for too long a period of time, and too unpredictably to use this to advantage. And other, unforeseen events can intervene - being unforeseen, they could be positive or negative. So what to do? How about..."nothing"!

So I just go with the flow, and I've never been happier. It's wonderful and amazing that taking the easy, lazy, do-nothing way, actually ends up being the best way! How often does that happen? I feel like a real-life "Wally" from "Dilbert!

 

[NTL2009]

I'm starting to feel this should be broken off to a different thread, I'm not sure a hypothetical debate on the future of computer trading is useful (though maybe interesting) to a thread on personal finance.

...
One of my questions is whether it is safe to assume the historical percentage returns on stocks will be relevant to a future stock market where the market participants that might have the largest resources are computer programs. (I really do understand - at least as well as anyone else in the thread - how stock markets work. So far, I haven't seen any coherent argument that begins "This is the way stock markets work" and ends with "therefore the participation of computers will not affect the percentage returns on stocks". (And I also havent' seen an argument that concludes "therefore the the participation of computers will affect the percentage returns". ))

I think the answer actually is buried in some of the earlier replies, but let me give it a shot anyway:

So let's say XYZ corp trades today at $10/sh, and I have purchased some through a paycheck direct deposit, in the form of a broad based mutual fund that owns hundreds, maybe thousands of stocks. Shares of ABC Corp also are in that index, and also trade at $10/sh. I also buy these with every paycheck for many years.

30 years later, in our hypothetical world w/o significant computer trading, let's say those companies have done well, and are now trading at $300/sh. I start drawing the divs, and maybe a little principal from time to time, to fund my retirement.

OK, parallel world, with heavy computer trading:

Was $10 for XYZ and ABC corp "right"? I dunno, maybe I should have paid $9 for XYZ and $11 for ABC, or vice-versa, or $9 each or $11 each? And what about next paycheck? Same or different? And how about the $300 years later? High, low? Averaged over time - does it matter? I don't think so.

Maybe a bit more real-world - would computers have detected the tech bubble of the late 1990's into 2000? Let's say they did, I think that would mean tech would not have rose so high, and would not have crashed so hard. But that had zero effect on me, I kept my asset allocation throughout. Same with the 2008 scenario. Relax, don't worry, enjoy life.

"Smart" computer trading might lower volatility if we assume the consensus is wrong sometimes (too high or too low). That's a good thing for a retiree (drawing down), but actually, volatility is good for a dollar cost average investor - they get more shares on the dips! Overall, I wouldn't worry about it.

 

[Stephen Tashi]

The difference, one that I think you have been missing, is we are saying the price is an accurate reflection of the consensus, not that the consensus itself is accurate.

What I say is that the "consensus" price isn't a consensus about anything - i.e. it is not a consensus about any particular statement. So it isn't meaningful to say that the consensus price is accurate or inaccurate in a "global" sense. An individual can interpret the consensus price in terms of whether the price is a good deal or bad deal according his personal measure of utility.

The stock market reflects the opinions, right or wrong.

I agree that the stock market is affected by a variety of opinions about factual matters. But it is also affected by personal preferences and how much money different people have to pursue those preferences in the market. So the stock market reflects opinions, but not a specific opinion.

I'm starting to feel this should be broken off to a different thread, I'm not sure a hypothetical debate on the future of computer trading is useful (though maybe interesting) to a thread on personal finance.

I don't object to that. However, this portion of the thread had revealed some interesting variations in how participants think stock markets work and what market prices mean.

 

[NTL2009]

NTL2009 said: ↑
I'm starting to feel this should be broken off to a different thread, ...

I don't object to that. However, this portion of the thread had revealed some interesting variations in how participants think stock markets work and what market prices mean.

Don't bother on my account. Based on these responses, I won't be engaging further on this subject in this thread or any other. IMO, it has become pointless meandering, and I'm not learning anything. But to close that out from my side:

What I say is that the "consensus" price isn't a consensus about anything - i.e. it is not a consensus about any particular statement. ...

Sigh. Of course it is a consensus. It is the consensus on the price that buyers and sellers are willing to exchange their shares at, at that point in time. What you want to infer from that consensus is a different matter, but I really don't think it matters one wit to the personal investor with their money in broad based index funds.

So the stock market reflects opinions, but not a specific opinion.

Don't mean to be flippant, but really - "so what"? There's nothing actionable there for the personal investor, so they should "invest" their time learning about what can help them achieve their financial goals, rather than "angels dancing on the heads of a pin" discussion about what the price of a stock means to different people - which is what I'm going to do by dropping this conversation at this point.

Thanks.

 

[gleem]

but I really don't think it matters one wit to the personal investor with their money in broad based index funds.

Actually index funds being traded like stocks can be bid up/down if buyers/sellers feel the market will rise or fall say for example for a good or bad jobs report or interest rate increase.

There's nothing actionable there for the personal investor, so they should "invest" their time learning about what can help them achieve their financial goals, rather than "angels dancing on the heads of a pin" discussion about what the price of a stock means to different people -

A current price is very actionable. you can sell or short it if you think it is too high or buy more if you think it is a bargain.

 

[Question_]

The market provides a consensus price.

This would be a good assumption to take, maybe fifty plus years ago, when speculation wasn't as rife as it is nowadays (last I recall 80-90% of all trades on the market are/is speculation). I also want to mention that arbitrage, which is a measure to limit speculation is not as effective nowadays due to, well I'm not entirely sure what (would be a good thesis paper premise, probably due to most currencies being fiat based along with an overabundance of credit).

Basically, what I'm trying to say that the market isn't as rational as people think and more caution should be taken when making a long term investment.

 

[russ_watters]

[The market provides a consensus price.]
This would be a good assumption to take, maybe fifty plus years ago, when speculation wasn't as rife as it is nowadays (last I recall 80-90% of all trades on the market are/is speculation). I also want to mention that arbitrage, which is a measure to limit speculation is not as effective nowadays due to, well I'm not entirely sure what (would be a good thesis paper premise, probably due to most currencies being fiat based along with an overabundance of credit).

Basically, what I'm trying to say that the market isn't as rational as people think and more caution should be taken when making a long term investment.

I think you and Stephen are arguing against a definition/reality here. Nobody has to agree that the price of a stock they see in the newspaper is "good" or "correct", but it is, by definition, the consensus price.

If you offer to sell me something for $110 and I offer to buy it for $90, and we eventually make the trade for $100, $100 is the consensus price. It doesn't matter if you or I think we got ripped off, we made an *actual agreement*, an *actual transaction* at $100. That's what a "consensus" is.

 

[russ_watters]

Now there are some studies about how the average P/E of the market can be a good long term indicator, and some people use this to adjust their allocation between stocks/bonds (but generally using broad-based index funds for each). I personally don't subscribe to this, I think the market can be 'wrong' for too long a period of time, and too unpredictably to use this to advantage. And other, unforeseen events can intervene - being unforeseen, they could be positive or negative. So what to do? How about..."nothing"!

I posted a graph of it before and used it as a simplistic measure, but I do agree with you that it can vary over relatively long periods of time -- and that doesn't even necessarily indicate the market is under or over-valued during those times. More importantly, that means there is no way to know since as V50 pointed out, the P/E ratio marries today's P with last year's E, but next year's P will be based on next year's E. So what to do? How about..."nothing"! It will work itself out.

 

[russ_watters]

One of my questions is whether it is safe to assume the historical percentage returns on stocks will be relevant to a future stock market where the market participants that might have the largest resources are computer programs...So far, I haven't seen any coherent argument that begins "This is the way stock markets work" and ends with "therefore the participation of computers will not affect the percentage returns on stocks".

That is disappointed to hear, because I've given arguments in pretty much exactly that form. I feel like we're getting bogged down in irrelevant minutiae and as a result you aren't really responding to the answers to this discussion you started. I'll try to list them concisely so we can refer back to them by number:

1. Computer logic has to be the same as people logic because computers are programmed by people. Even if people learn to harness the computers and they spit out different answers that the humans decide they like/work better, causing the human logic to adjust, that's still humans adjusting human logic. So the starting premise of your question is a false dichotomy: adding [more] computers to the mix doesn't change anything in how the markets work. Actually, two:

2. People logic does change. So the idea that the logic is constant now and could change in the future (for better or worse) isn't valid because it already changes constantly (that's why the stock market goes up one day and down the next!). So adding [more] computers to the mix doesn't change the change.

3. HFT specific: short term logic and long term logic are fundamentally different and not competing with/dependent on each other. So computers used for HFT trading cannot change the long term returns of buy-and-holders.

4. Most directly: Given that in #1 and #2 I said that logic does change all the time, computers or not, doesn't that mean that if it changes a certain way that it can affect long term returns? No. Because one way or another, stock prices are based on the profits of the companies, whereas a logic change is a single event. A logic change this year can make a stock that rises at 8% a year rise more or less this year, but next year it will go back to that 8% growth rate if the earnings keep growing by 8% a year. This is a simple mathematical reality that I mentioned before, but if you're not seeing it I can show you in a table/spreadsheet how it works.

 

[Question_]

That is disappointed to hear, because I've given arguments in pretty much exactly that form. I feel like we're getting bogged down in irrelevant minutiae and as a result you aren't really responding to the answers to this discussion you started. I'll try to list them concisely so we can refer back to them by number:

1. Computer logic has to be the same as people logic because computers are programmed by people. Even if people learn to harness the computers and they spit out different answers that the humans decide they like/work better, causing the human logic to adjust, that's still humans adjusting human logic. So the starting premise of your question is a false dichotomy: adding [more] computers to the mix doesn't change anything in how the markets work. Actually, two:

2. People logic does change. So the idea that the logic is constant now and could change in the future (for better or worse) isn't valid because it already changes constantly (that's why the stock market goes up one day and down the next!). So adding [more] computers to the mix doesn't change the change.

3. HFT specific: short term logic and long term logic are fundamentally different and not competing with/dependent on each other. So computers used for HFT trading cannot change the long term returns.

4. Most directly: Given that in #1 and #2 I said that logic does change all the time, computers or not, doesn't that mean that if it changes a certain way that it can affect long term returns? No. Because one way or another, stock prices are based on the profits of the companies, whereas a logic change is a single event. A logic change this year can make a stock that rises at 8% a year rise more or less this year, but next year it will go back to that 8% growth rate if the earnings keep growing by 8% a year. This is a simple mathematical reality that I mentioned before, but if you're not seeing it I can show you in a table/spreadsheet how it works.

I don't think there have been any advancements in logic for a good while. I don't mean to be snarky but, what has affected the market equilibrium the most is the amount of cash supply and information, along with many many more participants in the market activity.

 

[russ_watters]

5) It's not me that has a fixation on bad-mouthing active human-managed mutual funds on the basis of what I read in one book. I don't know why that crusade interests you.

Because it is an important and common error and:

It's not the topic that I proposed discussing. You seem to think my question implies that I advocate investing in actively managed funds or advocate short term trading.

Since you stated directly that you don't have a mechanism to propose as to how computers might affect future returns (your real question), you leave us little choice but to speculate about your concern for you (which is one reason the discussion lacks focus). Since it is logically difficult to prove an open-ended negative, much of what we've done is propose individual hypotheses about how computers might affect future stock prices/investment returns and then try to follow the logic to see if the hypothesis pans out. Computers as active fund managers is one such hypothesis. Whether you agree with or care about any particular hypothesis someone else proposes to focus your question isn't in our ability to know in advance. The only way for this discussion to be better focused on your question is if you focus your question better. Because at this point there is quite literally nothing more to it than fear of the unknown.

 

[russ_watters]

I don't think there have been any advancements in logic for a good while. I don't mean to be snarky but, what has affected the market equilibrium the most is the amount of cash supply and information, along with many many more participants in the market activity.

Oh, I agree. I think it says something that over a very long period of time, through vastly different political, economic and technological climates, the market return and p/e ratio have returned to their historical averages.

That said, I *do* have some concerns about the long term viability of stocks as an investment vehicle - just not for the reason proposed.

 

[gleem]

For AI/computers in investing check this out.

http://www.cnbc.com/2017/06/16/ai-a...ibms-watson-is-getting-into-etf-business.html

So now we will have something to watch and evaluate.

 

[Vanadium 50]

Part Two (Part One will come later):

I am going to use the word rational to describe stock pricing based on the mechanism I describe above: i.e. if you had a choice of investing $X in the stock or $X in Treasuries, you would not pay a premium to select either side. Speculation can be either rational or irrational -if I believe that XYZ is going to turn around under new leadership, and I am more confident of this new leadership than the consensus, I should be willing to pay a higher price for it than the consensus. This is speculative, and it may be wrong, but it's not irrational.

In a sense 100% of long positions in any investment are speculative. Greg might not have bought his condo - or at least that particular condo - if he knew it would lose 30% of its value. I think when people complain of "speculation" they are really complaining about irrational speculation. Given that 70% of the market is owned by institutional investors, I believe that most decisions are rational. If it's true that 90% of the trades are pure speculation, it means that this 90% is going on between just that 30%. And so what if they are churning stocks among themselves?

There also seems to be the argument that the fact that stock prices are not (largely) static is a sign that their trading is irrational. Rational pricing is in relation to a safe investment. It would predict that P/E is inversely related to treasury yields. I was looking for a scatter plot, but all I could find is this:

As you can see, over ~50 years E/P and yields do track.

 

[gleem]

Treasureys are always considered a safe investment at least for the preservation of capital.. If the government fails everything fails.

 

[Stephen Tashi]

A further comment on whether the market price of a stock is a "consensus price":

Suppose company X has 99 shareholders that think 20$ a share is too low a price to sell their stock and one share holder who is pressed for cash and will sell his shares for $20 a share. Among the buyers of stocks suppose all but one of the potential buyers thinks $20 is too high a price to pay for a share of company X but there is one buyer who will pay that much. That one buyer buys shares from that one seller. The market price of the stock is $20 even though the majority of people who hold the stock or consider buying it disagree with that price.

In realistic market situation, things are not that extreme but the hypothetical example illustrates that market prices are not set by a broad consensus among shareholders and potential buyers. Market prices are set by the minority of buyers and sellers who agree to trade at the market price. (In fact if the index funds don't execute trades, they leave the market prices to be set by people like the much-despised day traders - people that actually execute transactions.)

 

[collinsmark]

The following will be elementary and obvious to most PF members reading this thread. But allow me give a brief, extremely basic and somewhat simplified tutorial on how stock market price is determined. I got the impression that there might still be some confusion as to how the market prices of stocks are determined.

Again, this is somewhat simplified, but the following image sums it all up:

Prospective buyers and sellers queue up in order based on their desired sell or buy price. Each person can instantly change their place in the queue (line) at any time merely by changing their desired sell or buy price. If those people at the front of each queue (line) agree on a price, the sale is made and that becomes the new market price. If those at the front of each queue do not reach a common price, a stalemate is reached; no transactions occur and the market price remains at the price it was when the last transaction occurred. Of course, this stalemate can alleviated at any time by someone jumping to the front of the line and making a transaction.

In reality, a slightly more detailed version than above, sales happen through brokers -- middlemen so to speak -- who handle the details of the transactions. A broker can be an automatic, online entity (e.g., Ameritrade, ETrade, etc.), it doesn't necessarily need to be a human.

• If you want to buy at a certain price, you can tell your broker to buy x number of shares if the market price ever falls to $y. If the market price ever actually does fall to $y that means that your order has reached the front of the queue where a transaction was made. It means that those that were in front of you in the queue either already bought or canceled their orders, and some seller was willing to sell at your buy price. Usually such orders are good for a certain time limit, typically 30 days.
  
• If you want to buy the shares immediately, you can tell your broker to buy them at market price. Your broker will put your order at the front of the queue and buy the stock at the front of the "sell" queue at whatever the lowest "put" price is.
• Similarly, if you have stock and want to sell x number of shares at a certain price, your broker can put your order in the queue at your specified "put" price. This order usually has a time limit, typically 30 days.
• And of course you can jump ahead in the queue and sell your stock at market price by telling your broker to sell at market. Your broker will then sell your specified number of shares to those at the front of the buyers' queue, at whatever price they have specified as their buy price.
• Whenever a transaction occurs, that determines the new, reported market price: it's whatever the price of the last transaction was.

So what happens if the price at the front of the sellers' queue is higher than the price at the front of buyers' queue (and nobody jumps to the front of either line)? Nothing. No transactions are made. The "market price" remains at what it was when then last transaction actually occurred. This isn't too uncommon in small-cap stocks, btw. It's not out of the ordinary for some very small-cap stocks to go for days or even weeks without a single transaction. And that means that the stock's reported market price remains constant throughout that time interval.

 

[russ_watters]

A further comment on whether the market price of a stock is a "consensus price":

Suppose company X has 99 shareholders that think 20$ a share is too low a price to sell their stock and one share holder who is pressed for cash and will sell his shares for $20 a share. Among the buyers of stocks suppose all but one of the potential buyers thinks $20 is too high a price to pay for a share of company X but there is one buyer who will pay that much. That one buyer buys shares from that one seller. The market price of the stock is $20 even though the majority of people who hold the stock or consider buying it disagree with that price.

I don't think that's physically possible because common trading procedures aren't set up to deal with such stupid requests, but in that case the consensus price is $20. Why are you bringing such a silly scenario into the discussion?

In realistic market situation, things are not that extreme but the hypothetical example illustrates that market prices are not set by a broad consensus among shareholders and potential buyers. Market prices are set by the minority of buyers and sellers who agree to trade at the market price. (In fact if the index funds don't execute trades, they leave the market prices to be set by people like the much-despised day traders - people that actually execute transactions.)

You're dancing around your own point: Yes, index fund owners don't execute many trades, so they don't set market prices, so they aren't "potential buyers", so they don't actively help form the consensus price*.

Again: The "consensus price" is the price you pay for a stock right now. And that doesn't just include the buyers and sellers, it also includes the holders*, who agree with the consensus by virtue of the fact that they have decided to hold at the current price.

And what bearing does this have on the question you raised about the potential for computer driven trading to impact market returns?

 

[Vanadium 50]

Why are you bringing such a silly scenario into the discussion?

This is my fault. I was trying to get everyone on the same page about the price of a security being a measure of its future income stream. I didn't anticipate we would get into a discussion about "what is price?"

 

[Stephen Tashi]

As you can see, over ~50 years E/P and yields do track.

I'm curious how an e/p ratio is computed for a stock index. For example, in the case of two companies with respective earings E1, E2 and respective prices (at some time? or averaged over some time?) P1, P2. Is the e/p ratio computed as (1/2) (E1/P1 + E2/P2) or as ( (E1 + E2)/2 ) / ( (P1 + P2)/2) = (E1 + E2)/ (P1 + P2) ?

Again: The "consensus price" is the price you pay for a stock right now.

it's as @collinsmark illustrates. The market price is the price at which some people buy and sell the stock.. The market price is not the "consensus" price that all buyers and sellers agree is appropriate.

And that doesn't just include the buyers and sellers, it also includes the holders*, who agree with the consensus by virtue of the fact that they have decided to hold at the current price.

People who don't trade disagree with the market price.

And what bearing does this have on the question you raised about the potential for computer driven trading to impact market returns?

I don't know. I myself haven't offered any argument that begins with the (false) assertion that the market price is a broad consensus of the value of a stock - and then proceeds to deduce that the advent of computer programs won't affect the long-term stock market.

 

[TeethWhitener]

I don't think that's physically possible because common trading procedures aren't set up to deal with such stupid request

This is absolutely possible, and is in fact exactly how a market order works. EDIT: Maybe it's the case that your broker won't let you set that kind of order, but if you're a market maker, you can offer any price you want. (Several trades along these lines popped up during the Flash Crash of 2010)

The idea of a "consensus" price as everyone's using it seems to be related to a concept called the efficient market hypothesis: the price of a security reflects all available information about that security. There are various schools of thought on how efficient the market actually is, and most of the evidence points toward "pretty efficient," which is borne out in the observation that most actively managed funds don't beat the indexes. But some very famous people (e.g., Warren Buffett and other value investors) have made a name for themselves working with the notion that individual securities in the market can be long-term mispriced, such that if you can identify a truly undervalued stock, you can beat the market as that stock's price reverts to a true reflection of the asset's underlying value.

I'm curious how an e/p ratio is computed for a stock index.

Different indexes compute them differently. Some weight the stocks, some don't include companies with negative earnings, some do the Shiller PE (averaged over 10 years to smooth out fluctuations from the business cycle). Generally people just go straight to the indexing company for their calculation of PE.

Changing topics: One thing that I've been thinking about lately that I simply don't know that much about is why the P/E ratio is what it is. Not in the sense that it's price divided by earnings, but specifically why is it that we consider a "normal" P/E to be ~15 or so. What's the underlying cause for the number, other than "that's what it's been historically?" The number seems to indicate that if you invest $1 in a random stock, you can expect to sell the stock for $15 at some point in the future. Assuming the stock market goes up ~10%/year, the time to double is ~7 years, so the time to get to 16x is ~28 years. So very roughly, 15x is the expected return on a random stock after a ~30 year investment window. This seems anthropically appropriate, if you consider investing as a 30-something year old and selling assets as a 60-something year old, but this gave me an interesting thought. The PE for the broad market is currently at ~25 and has been creeping up for a few decades. Under an efficient market, this would happen because we expect higher future earnings from the market. Higher future earnings imply (again assuming efficient market) either more productive companies or a longer time horizon. So is the market anticipating a future of glorious hyperproductivity, or just sensing that people in the future will retire much later than age 65? (Of course in all likelihood, the real answer is that the market isn't perfectly efficient and the price is just evidence of a bubbly market).

 

[NTL2009]

Actually index funds being traded like stocks can be bid up/down if buyers/sellers feel the market will rise or fall say for example for a good or bad jobs report or interest rate increase.
...
A current price is very actionable. you can sell or short it if you think it is too high or buy more if you think it is a bargain.

And for long term personal investors, it doesn't matter. Sure, the price will move around based on day-to-day noise, and none of us can know which price reflects the "true" long-term value. But a personal investor will make many buys over their accumulation phase (normally through a payroll deduction, or other regular purchase program), so all this averages out.

This is not directed at you gleem, but I thought this thread was about "PF Investing" - IOW, what the normal type of person frequenting this forum should pay attention to regarding personal finance decisions.

Instead, it seems to have turned into some esoteric discussion of the value of a stock or fund - something maybe better discussed in an advanced review of the science of Economics?

Heck, my Father was a salt-of-the-earth type, barely had a high school education, but did well for himself and his family. One of the things he told me "Something's only worth what someone else will pay you for it". Forget the rest of the discussion, that's where the rubber meets the road, and no PhD in Economics can really add anything meaningful to that for the average investor.

 

[NTL2009]

This is absolutely possible, and is in fact exactly how a market order works. EDIT: Maybe it's the case that your broker won't let you set that kind of order, but if you're a market maker, you can offer any price you want. (Several trades along these lines popped up during the Flash Crash of 2010) ...

No, not in the world of how the personal investor should be approaching things. The personal investor should be investing in highly liquid, broad-based, low cost, index funds. The reality is (theoreticals don't matter to us 'regular folk'), none of us is likely to place an order large enough to move the market of a large fund. The spread on a highly liquid ETF is tiny, and averages out over time. Just buy/sell as needed.

If you were to look at not just the bid/ask, but the depth and volumes of the bid/ask on an ETF like SPY, you'd find more than enough sellers to fill your order within a penny or two. Pennies are not make/break for a long term investor. There is no big delta to be concerned with. If some investor is holding out for 2x of the current price, it doesn't matter to me, because someone else will fill my order within a tenth of a percent of the current price.

The idea of a "consensus" price as everyone's using it seems to be related to a concept called the efficient market hypothesis:

No, not 'everyone'. Several of us are saying (over and over again), that it just "is", efficient or not, high, low, middlin' or not. It is what it is. I'll repeat what my father taught me "Something is only worth what someone will pay you for it". Period.

But some very famous people (e.g., Warren Buffett and other value investors) have made a name for themselves working with the notion that individual securities in the market can be long-term mispriced, such that if you can identify a truly undervalued stock, you can beat the market as that stock's price reverts to a true reflection of the asset's underlying value.

I am not Warren Buffet. I doubt any of us are. I believe (not sure, correct me if I'm wrong) that Warren Buffet also got involved in the management of those companies, not simply a passive investor. And here is what Buffet says :" Buffett is skeptical that active management and stock-picking can outperform the market in the long run, and has advised both individual and institutional investors to move their money to low-cost index funds that track broad, diversified stock market indices."

Changing topics: One thing that I've been thinking about lately that I simply don't know that much about is why the P/E ratio is what it is. Not in the sense that it's price divided by earnings, but specifically why is it that we consider a "normal" P/E to be ~15 or so.

Well 1/15 ~ 6.7%. I think it only means that on average, we expect annual earnings to be somewhere around 6.7% of the stock price. No guarantees of course, but I think historically, over the long run, entrepreneurs on average won't work for less (or won;t get funded for less). And the ones that don't earn near that eventually fail to compete with 'risk free' investments like Treasuries. And when they make far more than 6.7%, competitors come into that market for a piece of that sweet action, and it rarely lasts long. It's kind of like the equilibrium states we find in nature.

 

[TeethWhitener]

No, not in the world of how the personal investor should be approaching things. The personal investor should be investing in highly liquid, broad-based, low cost, index funds. The reality is (theoreticals don't matter to us 'regular folk'), none of us is likely to place an order large enough to move the market of a large fund. The spread on a highly liquid ETF is tiny, and averages out over time. Just buy/sell as needed.

If you were to look at not just the bid/ask, but the depth and volumes of the bid/ask on an ETF like SPY, you'd find more than enough sellers to fill your order within a penny or two. Pennies are not make/break for a long term investor. There is no big delta to be concerned with. If some investor is holding out for 2x of the current price, it doesn't matter to me, because someone else will fill my order within a tenth of a percent of the current price.

Maybe you meant to reply to someone else. This has nothing to do with what I posted. I was specifically replying to Russ Watters, clarifying the technical details of what exactly happens when a "regular folk" submits an order through their broker.

No, not 'everyone'. Several of us are saying (over and over again), that it just "is", efficient or not, high, low, middlin' or not. It is what it is. I'll repeat what my father taught me "Something is only worth what someone will pay you for it". Period.

"Consensus" is generally construed to mean "broad agreement." Your position implies that the word "consensus" is superfluous. Why say "consensus price" instead of just "price?"

And here is what Buffet says :" Buffett is skeptical that active management and stock-picking can outperform the market in the long run, and has advised both individual and institutional investors to move their money to low-cost index funds that track broad, diversified stock market indices."

In this case, Buffett was specifically referring to hedge funds which charge ridiculous fees and take a percentage of the gains. These fees can soar to effectively 15-20% of an entire portfolio. He made a bet that they couldn't beat the indexes, and he was right. Buffett himself has always been a staunch supporter of the Graham-Dodd school of value investing (Ben Graham was his advisor at Columbia), which explicitly rejects the efficient market hypothesis. Others have argued that his gains have largely been outsized beta with clever use of leverage, rather than pure alpha, (which may or may not be true) but his personal outlook--as far as I can tell--has always been as a value investor.

Well 1/15 ~ 6.7%. I think it only means that on average, we expect annual earnings to be somewhere around 6.7% of the stock price. No guarantees of course, but I think historically, over the long run, entrepreneurs on average won't work for less (or won;t get funded for less). And the ones that don't earn near that eventually fail to compete with 'risk free' investments like Treasuries. And when they make far more than 6.7%, competitors come into that market for a piece of that sweet action, and it rarely lasts long. It's kind of like the equilibrium states we find in nature.

I think you're probably right; this is a better way of thinking about it. In this case, the rising PE ratio that we observe simply reflects the fact that a rock-bottom treasury rate pushes more people into stocks as they go yield hunting, thus driving down returns.