trying to understand total market and value investing

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
jbolden1517
Posts: 868
Joined: Sun Jul 09, 2017 2:53 pm

Re: trying to understand total market and value investing

Post by jbolden1517 »

rkhusky wrote: What is the dollar value of Apple? Something like the book value? It's my understanding that most things don't have a fundamental value - things are worth what someone is willing to pay. A car or a house or even a painting. On that basis, if someone is willing to pay 10x the current price for Apple stock, then Apple stock is worth 10x the current price.
The dollar value of apple is the discounted value of all future dividends the stock will pay over its lifetime. Take the following sum:
2018 dividend / 1 year discount rate
2019 dividend / 2 year discount rate
2020 dividend / 3 year discount rate
.... (who knows the future say it gets shut down in 2087 and sold off)
2087 M&A cash payout / 70 year discount rate

That's what Apple is worth. http://www.investopedia.com/terms/d/ddm.asp

Now that's sum is hard to compute. Fundamental analysis is about how to estimate it. The reason it is possible to estimate is discount rate return grows faster than the USA economy by a lot So those later years don't matter as much.
User avatar
in_reality
Posts: 4529
Joined: Fri Jul 12, 2013 6:13 am

Re: trying to understand total market and value investing

Post by in_reality »

rkhusky wrote:
jbolden1517 wrote:
rkhusky wrote:Not sure I understand the aversion to large companies. If Apple split into separate companies focusing, e.g., on software, computer hardware, music hardware, music store, app store, phones, and entertainment devices, would that be more palatable?
Yes. They could be more easily evaluated and priced. Their earnings would be less opaque. Their capital structure less opaque. The assets less opaque. The margin risks (which is the reason for the quite attractive pricing) would be easier to disentangle Yes it would be substantially more palatable. Not to say Apple isn't a good stock at these prices. But the problem with a cap weighted index is not that it holds a lot of Apple, but rather if Apple's cap grew 10x and nothing else about the company changed the cap weighted index would be holding 10x as much Apple. A fundamentally weighted index would hold the dollar value of Apple constant as the stock price grew so it would be selling all the way up.
The economies of scale and Apple's model of tight integration of software and hardware have no bearing? So, you would increase your stake 7-fold if the technologies were spun off? Would that hold also for conglomerates like Philip Morris with 250+ brands? You would increase your stake in those brands 250-fold?

What is the dollar value of Apple? Something like the book value? It's my understanding that most things don't have a fundamental value - things are worth what someone is willing to pay. A car or a house or even a painting. On that basis, if someone is willing to pay 10x the current price for Apple stock, then Apple stock is worth 10x the current price.
It's not the size of the company. I don't mind large companies.

It's that Facebook, Apple, Amazon, Microsoft And Google have added a total of $600 billion of market capitalization this year — the equivalent of the GDP of Hong Kong and South Africa combined. As of 6.9.2017, Facebook, Amazon, Apple, Microsoft and Alphabet (Google) — had contributed about 42 percent of all stock market gains over the preceding year.

Recent analysis of FAAMG stocks and previous leaders from the tech crash found that these current tech stocks have advantages in cash flow, valuation and cash balances over the top five tech names in the first quarter of 2000 — just before the bubble burst. But the current group is behind in profitability, as measured by gross profits and total assets. The tech bubble names studied included Lucent, Cisco, Oracle and Intel. Microsoft was the only stock to make both lists.

Look, I have no ability to evaluate their worth. All I know is there exists a tendency for things to boom, and perhaps boom for a while, only to settle back to reality. Maybe this time will be different. I don't feel that comfortable paying so much for future growth and so have a value tilt that I am ok with despite value having done poorly after 2008. So has international. Either this FAAMG is truly the place to be, or perhaps people are all focused on the same thing.

The point is that any new contributions and reinvestment are going increasingly towards the future exceptions of this group. Which I guess isn't the worst thing. I am not telling anyone to pull out. That is not my message here.
jbolden1517
Posts: 868
Joined: Sun Jul 09, 2017 2:53 pm

Re: trying to understand total market and value investing

Post by jbolden1517 »

in_reality wrote: The point is that any new contributions and reinvestment are going increasingly towards the future exceptions of this group. Which I guess isn't the worst thing. I am not telling anyone to pull out. That is not my message here.
Its worse than that. Many of those companies are diluting which means it is not just their current share of new contributions and reinvestment but a much larger percentage than their current share as the index is going to read itself as underweight a diluting company. Anyway good for you in having a value and international position. Diversification helps a lot. There are times like 2008 where it doesn't matter much, but those are rare.
User avatar
packer16
Posts: 1488
Joined: Sat Jan 04, 2014 1:28 pm

Re: trying to understand total market and value investing

Post by packer16 »

There IMO is a further differentiation between value investing (buying something for less than it is worth) & academic SCV investing which is indexing into low growth highly leveraged firms (crappy firms) which at times may be value investments and at other times not. The later I think will not be a satisfactory long term strategy and will provide additional volatility for very little if any incremental return. However, this is where most of the money has & is flowing into recently. I have tracked this crappy investment strategy since the mid 1990s when the strategy was first widely implemented in SCV value funds & the returns have been unimpressive. Since 1997, the incremental returns have been on the order of 0.5% per year but the volatility has been high. If you timed this strategy when it was published by Larry Swedroe with his LP, you would be really disappointed. The underperformance is very high for a so-called persistent factor, which has disappeared over the last 10 years. IMO what you have is not classical value investing but a modification of trading strategy that is value in name only & should be called the crappy company strategy. In my experience investing in crappy companies is very difficult & add in the index buying for these firms and you have a recipe for underperformance.

Packer
Buy cheap and something good might happen
rkhusky
Posts: 17767
Joined: Thu Aug 18, 2011 8:09 pm

Re: trying to understand total market and value investing

Post by rkhusky »

jbolden1517 wrote:
The dollar value of apple is the discounted value of all future dividends the stock will pay over its lifetime.
...

Now that's sum is hard to compute. Fundamental analysis is about how to estimate it.
How does one compute that for companies that don't pay a dividend? Or do you not invest in those? It seems like a lot of work, and expensive too.

I like cap weighted indexes because I own the same percentage of each company in the index.
jbolden1517
Posts: 868
Joined: Sun Jul 09, 2017 2:53 pm

Re: trying to understand total market and value investing

Post by jbolden1517 »

rkhusky wrote:
jbolden1517 wrote:
The dollar value of apple is the discounted value of all future dividends the stock will pay over its lifetime.
...

Now that's sum is hard to compute. Fundamental analysis is about how to estimate it.
How does one compute that for companies that don't pay a dividend? Or do you not invest in those? It seems like a lot of work, and expensive too.
The the current year that's a 0.
2017 - $0 dividend / share
2018 - $0 dividend / share
but hopefully somewhere that becomes a non-zero figure
2028 - $3 dividend / share
2029 - $ dividend / share
2030 - $4 dividend / share

If that sum stays at $0 forever the company is worth $0. If not to be able to pay a dividend why bother having the underlying company exist at all? Baseball cards are much more fun to hold and trade than stocks.
rkhusky wrote: I like cap weighted indexes because I own the same percentage of each company in the index.
The question is about what is a company worth. Let's assume there were only 3 companies in the index. Each costs $1/share to own.

A will pay out $2/share when acquired by B next year. B will acquire A with all its cash and then go broke. C will struggle for 2 years and then go broke.

Do you want to buy $1 or each or rather buy $3 of A and double your money?
rkhusky
Posts: 17767
Joined: Thu Aug 18, 2011 8:09 pm

Re: trying to understand total market and value investing

Post by rkhusky »

jbolden1517 wrote: The question is about what is a company worth.
Unfortunately, it appears that that information is very difficult and expensive to determine, with no guarantee that it will outperform a simple cap weighted index. And it is equally difficult to identify someone in advance who will be able to consistently outperform passive cap weighted indexes.
avalpert
Posts: 6313
Joined: Sat Mar 22, 2008 4:58 pm

Re: trying to understand total market and value investing

Post by avalpert »

rkhusky wrote:
jbolden1517 wrote: The question is about what is a company worth.
Unfortunately, it appears that that information is very difficult and expensive to determine, with no guarantee that it will outperform a simple cap weighted index. And it is equally difficult to identify someone in advance who will be able to consistently outperform passive cap weighted indexes.
Not difficult or expensive at all - take a simplified, Finance 101 version of valuation models assume it actually represents reality and pretend that the numbers we plug in for future estimates are nowhere near as uncertain as they actually are. Boom, intrinsic value.
rkhusky
Posts: 17767
Joined: Thu Aug 18, 2011 8:09 pm

Re: trying to understand total market and value investing

Post by rkhusky »

avalpert wrote:
rkhusky wrote:
jbolden1517 wrote: The question is about what is a company worth.
Unfortunately, it appears that that information is very difficult and expensive to determine, with no guarantee that it will outperform a simple cap weighted index. And it is equally difficult to identify someone in advance who will be able to consistently outperform passive cap weighted indexes.
Not difficult or expensive at all - take a simplified, Finance 101 version of valuation models assume it actually represents reality and pretend that the numbers we plug in for future estimates are nowhere near as uncertain as they actually are. Boom, intrinsic value.
OK. Difficult and expensive to do it right with some reasonable expectation of being close to correct. Throwing darts is also not that difficult.
CantPassAgain
Posts: 577
Joined: Fri Mar 15, 2013 8:49 pm

Re: trying to understand total market and value investing

Post by CantPassAgain »

rkhusky wrote:Throwing darts is also not that difficult.
Exactly.
avalpert
Posts: 6313
Joined: Sat Mar 22, 2008 4:58 pm

Re: trying to understand total market and value investing

Post by avalpert »

rkhusky wrote:
avalpert wrote:
rkhusky wrote:
jbolden1517 wrote: The question is about what is a company worth.
Unfortunately, it appears that that information is very difficult and expensive to determine, with no guarantee that it will outperform a simple cap weighted index. And it is equally difficult to identify someone in advance who will be able to consistently outperform passive cap weighted indexes.
Not difficult or expensive at all - take a simplified, Finance 101 version of valuation models assume it actually represents reality and pretend that the numbers we plug in for future estimates are nowhere near as uncertain as they actually are. Boom, intrinsic value.
OK. Difficult and expensive to do it right with some reasonable expectation of being close to correct. Throwing darts is also not that difficult.
Well sure, if being correct is your goal - but if your goal is to sell your conclusions (even if only to yourself) then it works just fine.
jbolden1517
Posts: 868
Joined: Sun Jul 09, 2017 2:53 pm

Re: trying to understand total market and value investing

Post by jbolden1517 »

rkhusky wrote:
jbolden1517 wrote: The question is about what is a company worth.
Unfortunately, it appears that that information is very difficult and expensive to determine, with no guarantee that it will outperform a simple cap weighted index.
Over what timeframe? Take for example a Schwab fundamental index vs. a Schwab cap weighted index. On the fundamental you are weighting by the company's sales, cashflow and dividend. Real tangible facts about the company that over the long haul help to determine whether it can pay a dividend or not and how much that dividend will be. On the cap weighted index you are weighting by whatever stuff happened to sell for yesterday. Sometimes a reasonable number and sometimes a random number. If market's are efficiently pricing all assets then a cap weighted index beats the fundamental index by about .3% per year after expenses. That's your downside risk.

If there is ever a point though where the market is irrationally pricing stuff, where investors in the aggregate aren't doing a particularly good job in assessing the discounted value of future dividends then the Schwab Fundamentals fund will be heavily overweight the cheaper stuff and underweight the more expensive stuff.

So just imagine of 4 stocks A, B, C and D. Each stock is going to grow randomly at some number between 5-15 per year as a base but that growth varies by +/- 10 in any given year randomly.

Now imagine you have two investors. GI and VI. Both start with $3000. GI estimates any company to be worth 26/(26-last year's growth). GI will buy up to that price. VI will only buy the two cheapest companies in any given year. It will sell to GI for his price and buy from GI for his price. At the end of 10 years A, B, C and D liquidate for $1000 * product of the ten years of growth. And we create 4 more random companies and repeat.

I'll stop here before I get into indexes. Run a spread sheet. And look for yourself how much more profitable VI is than GI.

Now we have mutual funds.
FW holds 100% VI
CapW holds A,B, C and D equally. Though of course GI sets the price.

Your argument boils down to that CapW won't underperform FW by 30 basis points.

Let's stop here so you can disagree or agree.
rkhusky wrote: And it is equally difficult to identify someone in advance who will be able to consistently outperform passive cap weighted indexes.
The argument for active vs. passive and the argument for value weighting vs. cap weighting are distinct. You can do value weighting passively. You don't need active.
jbolden1517
Posts: 868
Joined: Sun Jul 09, 2017 2:53 pm

Re: trying to understand total market and value investing

Post by jbolden1517 »

avalpert wrote: Not difficult or expensive at all - take a simplified, Finance 101 version of valuation models assume it actually represents reality and pretend that the numbers we plug in for future estimates are nowhere near as uncertain as they actually are. Boom, intrinsic value.
Pity the evidence for value premiums prove that it isn't that uncertain. Simply doing the most simple valuations adjustments outperform by 200-300 basis points per year on average. At this point we have decades of evidence of even the most stupid metrics working well like buy companies trading at a low ratio to their asset's original acquisition cost

Image

Or of course your least favorite weight by dividend income:

Image

The market is so terrible at this, that you don't have to do it well to collect the premium. You just have to do it at all.

I do love how you claim all this financial expertise and are quite condescending about it. Yet your actual solution on how to allocate capital is to just buy random investments at random prices and hope for best.
avalpert
Posts: 6313
Joined: Sat Mar 22, 2008 4:58 pm

Re: trying to understand total market and value investing

Post by avalpert »

jbolden1517 wrote:
avalpert wrote: Not difficult or expensive at all - take a simplified, Finance 101 version of valuation models assume it actually represents reality and pretend that the numbers we plug in for future estimates are nowhere near as uncertain as they actually are. Boom, intrinsic value.
Pity the evidence for value premiums prove that it isn't that uncertain. Simply doing the most simple valuations adjustments outperform by 200-300 basis points per year on average. At this point we have decades of evidence of even the most stupid metrics working well like buy companies trading at a low ratio to their asset's original acquisition cost

Image

Or of course your least favorite weight by dividend income:

Image

The market is so terrible at this, that you don't have to do it well to collect the premium. You just have to do it at all.

I do love how you claim all this financial expertise and are quite condescending about it. Yet your actual solution on how to allocate capital is to just buy random investments at random prices and hope for best.
You do realize that none of those refer to intrinsic value or dividend discount models, right? You do realize that the value premium is widely believed to be a risk premium, right - and that they exist at the aggregate, not individual security level, right?

And no, I have never recommended buying random investments at random prices and hoping for the best - there is nothing random about investing at market weights.
jbolden1517
Posts: 868
Joined: Sun Jul 09, 2017 2:53 pm

Re: trying to understand total market and value investing

Post by jbolden1517 »

avalpert wrote:You do realize that none of those refer to intrinsic value or dividend discount models, right?
Your claim was that in the aggregate you couldn't apply valuation, this was a naive finance 101. You have been urging others to believe in a world of total nihilism where nothing can be known. There is no fact, no reality other social convention. In your world there is no truth.

What the evidence shows is that even applying the simplest valuations improves performance over the market. There are things that can be known understood and acted upon. And when one does that, it has measurable effects. Those effects are probabilistic and so much easier to see over long time periods and many stocks in the aggregate. But this is a mutual fund investor's board. They are talking about aggregates.
avalpert wrote: You do realize that the value premium is widely believed to be a risk premium, right - and that they exist at the aggregate, not individual security level, right?
It has to exist at the individual security level to exist in the aggregate. As for a risk premium that's a linguistic convention. When one assumes all markets are efficient then any means of getting return is called a risk premium. There certainly exist "risks" and risk dispersions that is situations where based on cashflow a particular type of investor is concerned about a particular type of risk and accepts below market returns in exchange for not carrying that risk.

An easy example is short term bonds vs cash. Over any meaningful time period of time short term bonds have essentially 0 risk of underperforming cash. Over short periods of time they can have minor fluctuations. Investors are heavily rewarded for taking on the risk premium for the first 2 years of duration risk, so much so and so predictably that we can talk about "cash drag".
avalpert wrote:And no, I have never recommended buying random investments at random prices and hoping for the best - there is nothing random about investing at market weights.
I think you would be hard pressed to prove. Define a reasonable testable distinction that exists between what you do and a policy of random investment. You have two investors A and B. A holds TSM. B buys 1000 stocks at whatever the market price is at weight .1% of his portfolio and let's it ride for a decade reinvesting dividends. At the end of the decade what is the distinction between them? Mind you I think B is likely to outperform A because equal weighting beats cap weighting and a decade isn't enough for B's portfolio to be as inefficient as A's. But you are arguing that A is in fact superior to B. So prove it. Prove that's not what you are doing more or less.
avalpert
Posts: 6313
Joined: Sat Mar 22, 2008 4:58 pm

Re: trying to understand total market and value investing

Post by avalpert »

jbolden1517 wrote:
avalpert wrote:You do realize that none of those refer to intrinsic value or dividend discount models, right?
Your claim was that in the aggregate you couldn't apply valuation, this was a naive finance 101.
I have never made such a claim. I have claimed that certain applications of the dividend-discount model as advocated by some here are based on a naive finance 101 mindset.
You have been urging others to believe in a world of total nihilism where nothing can be known. There is no fact, no reality other social convention. In your world there is no truth.
And again, I have not. In fact I am very specific at separating what can be known from what is just guesses - there is a very clear reality. As for 'truth', well that depends on truth of what - if you are looking for the true future worth of an investment today then you are right, the answer isn't out there.
It has to exist at the individual security level to exist in the aggregate.
Just the other day you were mentioning the effect when you have say a 5/6 chance of small outperformance and 1/6 chance of large underperformance. Think of how that might apply to the small and value premiums and why what exists in the aggregate need not at the individual level.
As for a risk premium that's a linguistic convention.
Not at all, it has specific meaning and implication in the academic theories.

avalpert wrote:And no, I have never recommended buying random investments at random prices and hoping for the best - there is nothing random about investing at market weights.
I think you would be hard pressed to prove. Define a reasonable testable distinction that exists between what you do and a policy of random investment. You have two investors A and B. A holds TSM. B buys 1000 stocks at whatever the market price is at weight .1% of his portfolio and let's it ride for a decade reinvesting dividends. At the end of the decade what is the distinction between them? [/quote]
Huh? The distinction should be obvious - though neither is an example of random investing.
But you are arguing that A is in fact superior to B. So prove it. Prove that's not what you are doing more or less.
Even in multi-factor models, A is on the efficient frontier. B is unlikely to be because it is unlikely to be an efficient means of investing in the SmB risk factor.
jbolden1517
Posts: 868
Joined: Sun Jul 09, 2017 2:53 pm

Re: trying to understand total market and value investing

Post by jbolden1517 »

avalpert wrote: I have never made such a claim. I have claimed that certain applications of the dividend-discount model as advocated by some here are based on a naive finance 101 mindset.
The claim that has been made here repeatedly is that investment return for an equity among all investors in the aggregate is determined by its eventual paid dividends discounted. That's the claim you have been arguing against. Absolutely that's finance 101 and algebra 1. You are very clear that anyone who would believe the most formulas that underly the entire theory of finance is stupid and naive. Yes you have made such a claim, repeatedly.
avalpert wrote:
You have been urging others to believe in a world of total nihilism where nothing can be known. There is no fact, no reality other social convention. In your world there is no truth.
And again, I have not. In fact I am very specific at separating what can be known from what is just guesses - there is a very clear reality. As for 'truth', well that depends on truth of what - if you are looking for the true future worth of an investment today then you are right, the answer isn't out there.
See your sentence above where you deny making this very claim.


avalpert wrote:
It has to exist at the individual security level to exist in the aggregate.
Just the other day you were mentioning the effect when you have say a 5/6 chance of small outperformance and 1/6 chance of large underperformance. Think of how that might apply to the small and value premiums and why what exists in the aggregate need not at the individual level.
I don't think that was me. I did say said something about large growth outperforming value in most 3 year periods. That doesn't mean that expected return for value isn't higher in all single year periods just that the distributions for value return and growth return aren't normal distributions.
avalpert wrote:
As for a risk premium that's a linguistic convention.
Not at all, it has specific meaning and implication in the academic theories.
Which is the origins of the linguistic convention, as I said.

avalpert wrote:
I think you would be hard pressed to prove. Define a reasonable testable distinction that exists between what you do and a policy of random investment. You have two investors A and B. A holds TSM. B buys 1000 stocks at whatever the market price is at weight .1% of his portfolio and let's it ride for a decade reinvesting dividends. At the end of the decade what is the distinction between them? But you are arguing that A is in fact superior to B. So prove it. Prove that's not what you are doing more or less.
Even in multi-factor models, A is on the efficient frontier. B is unlikely to be because it is unlikely to be an efficient means of investing in the SmB risk factor.
That's not proof that's just assuming the point in question. Obviously if a-priori A is on the efficient frontier then A is perfectly weighted. You are arguing in a circle: Stock prices are perfect because stock prices are perfect because stock prices are perfect. Responding with a proof of that type is intellectual nihilism. The reason you can't answer is there is no proof. It was a simplifying assumption for finance nothing more. The economics equivalent of starting with a perfectly spherical horse in a frictionless universe (https://en.wikipedia.org/wiki/Spherical_cow)

The biggest problem with believing that theory is it has been demonstratively disproven by the experimental evidence. Every testable hypothesis that CAPM asserts is contradicted by the evidence, including the evidence collected since CAPM was created which included instructions as to how to build a perfect portfolio. Those charts above being just one of many disproofs. They show quite clearly the market portfolio is not on the efficient frontier since for example a basket of low P/B stocks, has a higher risk adjusted return than the market portfolio. Moreover it has also been proven that the market portfolio when cut with non-correlating assets allows for the creation of portfolios which have a higher risk adjusted return than the market portfolio, which CAPM would assert are impossible.

You are arguing religion not finance.
rkhusky
Posts: 17767
Joined: Thu Aug 18, 2011 8:09 pm

Re: trying to understand total market and value investing

Post by rkhusky »

jbolden1517 wrote: So just imagine of 4 stocks A, B, C and D. Each stock is going to grow randomly at some number between 5-15 per year as a base but that growth varies by +/- 10 in any given year randomly.

Now imagine you have two investors. GI and VI. Both start with $3000. GI estimates any company to be worth 26/(26-last year's growth). GI will buy up to that price. VI will only buy the two cheapest companies in any given year. It will sell to GI for his price and buy from GI for his price. At the end of 10 years A, B, C and D liquidate for $1000 * product of the ten years of growth. And we create 4 more random companies and repeat.
This sounds like a Dogs of the Dow strategy, which didn't work out in the real world. That's the trouble with toy problems and simple algorithms. There are thousands of folks whose job it is to analyze and value companies, assisted by advanced computer algorithms and software. I doubt that any simple strategy that actually works is going to provide out-performance a few moments past publication.

I also don't think that simple metrics (or even complex metrics) can predict which companies are going to fail and which are going to succeed in the next 20 years, much less how much dividends they will be paying out.
jbolden1517
Posts: 868
Joined: Sun Jul 09, 2017 2:53 pm

Re: trying to understand total market and value investing

Post by jbolden1517 »

rkhusky wrote: This sounds like a Dogs of the Dow strategy, which didn't work out in the real world.
I agree that's exactly what it is like. I don't agree with you it didn't work when it became popular. So let's look at what happened when you selected the "dogs" from the broader market even after the theory became well know:

Image
rkhusky wrote: That's the trouble with toy problems and simple algorithms. There are thousands of folks whose job it is to analyze and value companies, assisted by advanced computer algorithms and software. I doubt that any simple strategy that actually works is going to provide out-performance a few moments past publication.
Value premiums have persisted for decades since they became well known to exist. You are absolutely right there are lots of people doing analysis. So if there are going to explanations for why these continue to exist they would need to be in terms of systematic bias. You would need to be looking for things like

a) an improper incentive structure where the interests of investors and the interests of stock analysts conflict.
b) An immutable advantage that corporations with low valuations have that allows them to deploy capital with a higher long term return because they have a low valuation.
c) A type of persistent systematic error that would cause random biases to accumulate in one direction and thus increase distortion.
d) Structural inefficiencies in markets that would cause them not to be able to respond to new information in an equal manner.
I also don't think that simple metrics (or even complex metrics) can predict which companies are going to fail and which are going to succeed in the next 20 years, much less how much dividends they will be paying out.
Then how do you explain the evidence that apparently they can? That happens in every country, it happens in all time periods long enough. Why?
rkhusky
Posts: 17767
Joined: Thu Aug 18, 2011 8:09 pm

Re: trying to understand total market and value investing

Post by rkhusky »

jbolden1517 wrote:
rkhusky wrote:I also don't think that simple metrics (or even complex metrics) can predict which companies are going to fail and which are going to succeed in the next 20 years, much less how much dividends they will be paying out.
Then how do you explain the evidence that apparently they can? That happens in every country, it happens in all time periods long enough. Why?
Is there a list of the dividend predictions of the stocks in the Total Stock Market from 20 years ago and how does that compare to the dividends currently being paid?
What are the predictions for dividends for the current Total Stock Market? (although I doubt I will be interested in this 20 years from now)
jbolden1517
Posts: 868
Joined: Sun Jul 09, 2017 2:53 pm

Re: trying to understand total market and value investing

Post by jbolden1517 »

rkhusky wrote:
jbolden1517 wrote:
rkhusky wrote:I also don't think that simple metrics (or even complex metrics) can predict which companies are going to fail and which are going to succeed in the next 20 years, much less how much dividends they will be paying out.
Then how do you explain the evidence that apparently they can? That happens in every country, it happens in all time periods long enough. Why?
Is there a list of the dividend predictions of the stocks in the Total Stock Market from 20 years ago and how does that compare to the dividends currently being paid?
For most stocks who are paying most dividends they are companies growing along with the economy and their dividend yield today is essentially their dividend yield then + inflation + economic growth -1%. That's the bulk of the dividends.

The next group are the dividend growers. Stocks who are increasing their earnings and often increasing their payouts. There are more sophisticated methods but again the easiest one is look at stocks that currently pay a dividend and whether they grew it last year or not:

Image

More sophisticated is look at companies with a long history of increasing dividends. For the SP500 there is an index of those with a 25 year history: https://us.spindices.com/indices/strate ... ristocrats ; there are also indexes for 10 year growers.

The next level down are those companies that don't pay a dividend yet or don't grow their dividend but could. High earners. You can estimate the dividend yield for a slowly growing company to be 1/2 their earnings yield. And as already mentioned. That outperforms.

For the high growers 20 years out the dividend yield is likely to still be near 0. But their earnings will likely have stabilized. Generally high P/E growers grow about 3% faster for a decade than slower growers. So you can estimate what their value would be 20 years from now using the formula but just assuming they grow faster.

As you get more complex you get into "fundamental analysis". The techniques have been around for a century:

https://www.amazon.com/Security-Analysi ... 0071592539

But it is done one stock at a time. For aggregates you don't need to be complex. The market ignores valuation in enough cases. Remember you don't have to find the perfect number. You just have to answer the question "is the current price way too high, way too low, or close enough to right"?
Topic Author
indexonlyplease
Posts: 1571
Joined: Thu Apr 30, 2015 12:30 pm
Location: Florida

Re: trying to understand total market and value investing

Post by indexonlyplease »

jbolden1517 wrote:Your thinking is correct. I'm a value investor. Here is basically the argument.

If you exclude control and arbitrage investors then TSM represents the collective judgement about the returns of each individual stock as determined by the entire aggregate of passive minority shareholders. By holding TSM you arbitrage away their research costs while achieving their precost return. In essence before cost you tie the average manager and after cost you win.

Now for the value counter argument.
1) Management of value firms have an opportunity to deploy their capital into a high return investment for shareholders (their stock price) that growth firms do not enjoy.

2) Investment managers mutual fund managers are compensated for assets under management not return. Pension fund managers are compensated usually on 3 year track records relative to a benchmark. Their payoff matrix is not the same as the investor's payoff matrix. The managers not shockingly follow strategies designed to grow their payoff even at the expense of return. So for example: growth stocks have many more highly positive 3 year periods than value stocks (which effects both mutual fund and pension fund managers, and in today's world likely hedge fund managers). Value stocks outperform during periods of time when new investment in mutual funds are low, growth stocks outperform at times when they are high....

3) Investors respond slowly to new information. There is a bias towards under reaction (contrary to the popular wisdom). I should mention widespread indexing is compounding this problem severely.

4) The winners curse. Assume there is an auction with S sellers and B buyers each of which is capable of evaluating an item and making a buy sell decision based on the price relative to their evaluation. Assume each of their individual evaluations has random error. if S > B the asset will tend to sell for below fair price while if B > S the assets will tend to sell above fair price. This becomes exploitable if one merely makes blind buy/sell decisions based on the ratio between S and B.

The result of the above is that their is a consistent bias of more positive returns by weighting stocks on value criteria. Just as indexing can outperform the average mutual fund manager by arbitraging the cost of research value can exploit by arbitraging away the bias in stock selection. Passive value achieves both objectives it pays neither for research and avoids the bias.

Given your user name I assume you like disagree with the above. So I'll stop here and let you object.
I will not object because I don't have enough knowledge. I am starting to understand the idea of tilting small cap value. But yes I did buy into the indexing a few years ago.
User avatar
privatefarmer
Posts: 779
Joined: Mon Sep 08, 2014 2:45 pm

Re: trying to understand total market and value investing

Post by privatefarmer »

The way I think of it is : I want to own companies for 30+ years. If the companies I own grow their earnings, the share price will increase at most likely the same rate to keep the P/E anchored at around the historical average (ie 15-20). It all comes down to : what companies will grow their profits the fastest over the next X number of years? Who the hell knows. Just bc a company is classified as "value" or "growth" doesn't tell you anything about the future widgets that they will sell or what their industries will look like over the next 30 years. So buy them all.

Now, some "growth" companies may have a P/E of like 80 which may seem crazy but the market has priced them that high and the market may be correct in predicting that their profits are going to shoot up so fast that it justifies their current valuation. Perhaps the market is using a totally different earnings estimate than you or I use, perhaps the P/E ratio the majority of investors in the actual market use is only 25 instead of 80? Or, if you look at price/book, that doesn't consider the "book value" of all the geniuses working at google and facebook, for example. How do you put a price on the intellectual property that they control?

All I know is that I know next to nothing. But I do believe that hedge funds, actively managed funds, and people like Warren Buffett will collectively price every stock better than I could so I might as well buy them all at current prices instead of trying to outsmart those who do this for a living. We simply do not know what stocks will grow their earnings the most over the coming decades and that really is the crux of the issue. If we invest for decade we hopefully will see our stocks more than double in price therefore whether you buy a stock at a P/E of 15 vs 25 really isn't as impactful if the stock is worth say 8x what you initially bought it at.
jbolden1517
Posts: 868
Joined: Sun Jul 09, 2017 2:53 pm

Re: trying to understand total market and value investing

Post by jbolden1517 »

privatefarmer wrote:The way I think of it is : I want to own companies for 30+ years.
This is a good outline. But there are a lot of minor points of disagreement here.

Let's start with that? Why shouldn't you want to own shares in companies that have the best returns and not own shares in companies that have substandard returns? What does 30 plus years get you? Now if you mean you want to hold a percentage of the USA market, sure that means you want to hold funds. But there is no particular reason you want those funds to want to hold any particular company.
privatefarmer wrote: If the companies I own grow their earnings, the share price will increase at most likely the same rate to keep the P/E anchored at around the historical average (ie 15-20).
That's not quite true. 30 years while not a huge amount of time in the life of a company is long enough to see a noticeable bias. Companies start in industries where the long term return on capital is on average well above the national average. As they invest capital the return on capital within those industries on average frequently declines, competition compresses margins. You should over say a 200 year period expect to see a u-shape where the P/E starts high (because a company is growing things like sales or I/P to use your example below). Then this number lowers as the returns on capital are realized and become unsustainably high. Then start to increase as the industry experiences declining margins and return on reinvested capital falls to well below the national average. A company with growing earnings on average is going to have a substantially higher P/E than one with falling earnings, and they should.
privatefarmer wrote: It all comes down to : what companies will grow their profits the fastest over the next X number of years? Who the hell knows.
Statistically companies in industries whose margins are well below average due to oversupply where demand is increasing. So companies with strong top line growth experiencing bottom line declines. Your typical growth stocks grow their earnings the fastest.
privatefarmer wrote: Just bc a company is classified as "value" or "growth" doesn't tell you anything about the future widgets that they will sell or what their industries will look like over the next 30 years. So buy them all.
Why is "buy them all the answer"? Why not buy who have structural advantages relative to market pricing?
privatefarmer wrote: Or, if you look at price/book, that doesn't consider the "book value" of all the geniuses working at google and facebook, for example. How do you put a price on the intellectual property that they control?
First off that number becomes relevant if you think the companies are likely to be bought. You would want to apply that to unsuccessful and advertising and social media companies. Successful ones you are valuing based on earnings not book. The book is a concern only in that you are going to be valuing their earnings so highly that you know if the company starts faltering the acquisition price will be considerably lower than what you pay today. So you should assign a higher risk premium because the floor is a long way down and move on from there.

The way you estimate the value of I/P for unsuccessful companies is a reasonably discounted fraction of what it costs to acquire the knowledge in the first place, assuming it is worthwhile. This is frequently how you evaluate bio-tech firms since most fail as businesses but are acquired for their research. For research that is promising something like 50% of the investment over the last decade. For research that is very promising you need expertise. For companies in the social media space (Facebook's space), their I/P seems to be mostly worthless, however their customer base has tremendous value. Google has too many sub businesses to do an estimate and they are too complex. Only in a situation where most of them are failing would that number be relevant.

Most companies are not that complex however. They are much much simpler.
privatefarmer wrote: All I know is that I know next to nothing. But I do believe that hedge funds, actively managed funds, and people like Warren Buffett will collectively price every stock better than I could so I might as well buy them all at current prices instead of trying to outsmart those who do this for a living.
The evidence shows that to be false, as I've shown above. They collectively price stocks worse than you could even using simple metrics. They do such a dreadful job on average you can still be only slightly less terrible and win.
privatefarmer wrote: We simply do not know what stocks will grow their earnings the most over the coming decades and that really is the crux of the issue.
First off their are many companies whose earnings are relatively stable or gently falling. A decade is a long time and their earnings may not be as stable as they were 10 years out. But they don't usually go from stable to highly unstable in that short a period of time. Not being able to value Google is not the same as not being able to value most stocks. Google is much more complex than the average company, an almost worst case.

Remember you aren't having to determine the perfect price. You only have to determine if on average the market price is likely too high or too low. You can be wrong 40% of the time and right 60% of the time and still show a huge profit margin for your trouble.
Even if there is no way to know the growth of earnings for most companies with about 7000 major USA stocks if you can know if for 2% that's 140. The next question is whether the market has persistent biases towards assuming that earnings growth is more stable than it is. The answer is yes. That's an exploitable inefficiency regardless of how good you are at solving the complex stock valuation problem.

For many companies it perfect price doesn't matter too much. Pick a company you know something about and try: https://www.trefis.com/companies
privatefarmer wrote: If we invest for decade we hopefully will see our stocks more than double in price therefore whether you buy a stock at a P/E of 15 vs 25 really isn't as impactful if the stock is worth say 8x what you initially bought it at.
Buying a company for 2x what it is worth vs one that is fairly priced costs you on average 7.1% of your return over a decade or 2.3% over 30 years. It makes a difference even for a long time buy and hold investor. This is not a small issue. Buying a company for 2x what it is worth vs. focusing your buying on ones selling for for 1/2 what it is worth, doubles those numbers. 4.6% is larger than the risk premium for equity over bonds. So if you are making this mistake consistently in your initial buys at the start of 30 years it would wipe out all your profits for buying equity over bonds even over a 30 year holding period. Now cap weighting doesn't force you to always make this mistake but it substantially biases you towards this mistake.

Valuation is not something to be cavalier about. You wouldn't be cavalier about what a 2.3% expense ratio would do to your returns on a mutual fund. Why be cavalier about an equally destructive habit?
User avatar
packer16
Posts: 1488
Joined: Sat Jan 04, 2014 1:28 pm

Re: trying to understand total market and value investing

Post by packer16 »

jbolden,

Although I agree that value investing works, I think the implementation of it matters quite abit. If the trends of the data you illustrate are true after they were published then you have alot of wealthy investors who invested in the value factor vs. not so wealthy non-value investors & the data does not support this. Also, it to a certain extent it too easy to implement for any excess returns to hang around for any period of time. I have been following this since the mid-1990s. As a simple example, I have developed a relative performance index of SCV vs. S&P 500. If simply investing in a value index fund were the answer for better returns (like investing in stocks versus bonds is a good strategy for this) you would see a persistent outperformence of SCV vs. the S&P 500. What we see instead is a small overall premium of about 0.5% per year & a huge amount of volatility. It looks as though the value premium is more cyclical than directional. Whenever this happens, the timing of your investment makes a huge difference in performance. This is a huge risk for an individual investor relying on growth to meet a goal over a specified period of time. I have shown this in the LP shortfall thread. There I started an LP after his NYT article about the LP was published & have shown if you implements a LP (higher weight to SCV & thus you can have more bonds & get about the same expected returns) you would have significantly underperformed a similar suggested S&P 500/Bond portfolio. This is nothing new. John Bogle describes this in Bogle on Mutual Funds published in 1991. The small amount of value premium versus the cost savings associated with lower fees and the consistency of the low-cost benefit is why Bogle focuses on costs versus value as way to get better returns.

I think a more plausible alternative is to identify a few securities that are fundamentally extremely undervalued on a number of dimensions (like from firm valuation & large holding company or voting discounts) using the standard valuation tools (DCF, relative multiple, etc.) this way you are relying on multiple ways for upside. Just my 2 cents.

Packer
Last edited by packer16 on Fri Aug 11, 2017 6:59 am, edited 3 times in total.
Buy cheap and something good might happen
User avatar
privatefarmer
Posts: 779
Joined: Mon Sep 08, 2014 2:45 pm

Re: trying to understand total market and value investing

Post by privatefarmer »

jbolden1517 wrote:
privatefarmer wrote:The way I think of it is : I want to own companies for 30+ years.
This is a good outline. But there are a lot of minor points of disagreement here.

Let's start with that? Why shouldn't you want to own shares in companies that have the best returns and not own shares in companies that have substandard returns? What does 30 plus years get you? Now if you mean you want to hold a percentage of the USA market, sure that means you want to hold funds. But there is no particular reason you want those funds to want to hold any particular company.
privatefarmer wrote: If the companies I own grow their earnings, the share price will increase at most likely the same rate to keep the P/E anchored at around the historical average (ie 15-20).
That's not quite true. 30 years while not a huge amount of time in the life of a company is long enough to see a noticeable bias. Companies start in industries where the long term return on capital is on average well above the national average. As they invest capital the return on capital within those industries on average frequently declines, competition compresses margins. You should over say a 200 year period expect to see a u-shape where the P/E starts high (because a company is growing things like sales or I/P to use your example below). Then this number lowers as the returns on capital are realized and become unsustainably high. Then start to increase as the industry experiences declining margins and return on reinvested capital falls to well below the national average. A company with growing earnings on average is going to have a substantially higher P/E than one with falling earnings, and they should.
privatefarmer wrote: It all comes down to : what companies will grow their profits the fastest over the next X number of years? Who the hell knows.
Statistically companies in industries whose margins are well below average due to oversupply where demand is increasing. So companies with strong top line growth experiencing bottom line declines. Your typical growth stocks grow their earnings the fastest.
privatefarmer wrote: Just bc a company is classified as "value" or "growth" doesn't tell you anything about the future widgets that they will sell or what their industries will look like over the next 30 years. So buy them all.
Why is "buy them all the answer"? Why not buy who have structural advantages relative to market pricing?
privatefarmer wrote: Or, if you look at price/book, that doesn't consider the "book value" of all the geniuses working at google and facebook, for example. How do you put a price on the intellectual property that they control?
First off that number becomes relevant if you think the companies are likely to be bought. You would want to apply that to unsuccessful and advertising and social media companies. Successful ones you are valuing based on earnings not book. The book is a concern only in that you are going to be valuing their earnings so highly that you know if the company starts faltering the acquisition price will be considerably lower than what you pay today. So you should assign a higher risk premium because the floor is a long way down and move on from there.

The way you estimate the value of I/P for unsuccessful companies is a reasonably discounted fraction of what it costs to acquire the knowledge in the first place, assuming it is worthwhile. This is frequently how you evaluate bio-tech firms since most fail as businesses but are acquired for their research. For research that is promising something like 50% of the investment over the last decade. For research that is very promising you need expertise. For companies in the social media space (Facebook's space), their I/P seems to be mostly worthless, however their customer base has tremendous value. Google has too many sub businesses to do an estimate and they are too complex. Only in a situation where most of them are failing would that number be relevant.

Most companies are not that complex however. They are much much simpler.
privatefarmer wrote: All I know is that I know next to nothing. But I do believe that hedge funds, actively managed funds, and people like Warren Buffett will collectively price every stock better than I could so I might as well buy them all at current prices instead of trying to outsmart those who do this for a living.
The evidence shows that to be false, as I've shown above. They collectively price stocks worse than you could even using simple metrics. They do such a dreadful job on average you can still be only slightly less terrible and win.
privatefarmer wrote: We simply do not know what stocks will grow their earnings the most over the coming decades and that really is the crux of the issue.
First off their are many companies whose earnings are relatively stable or gently falling. A decade is a long time and their earnings may not be as stable as they were 10 years out. But they don't usually go from stable to highly unstable in that short a period of time. Not being able to value Google is not the same as not being able to value most stocks. Google is much more complex than the average company, an almost worst case.

Remember you aren't having to determine the perfect price. You only have to determine if on average the market price is likely too high or too low. You can be wrong 40% of the time and right 60% of the time and still show a huge profit margin for your trouble.
Even if there is no way to know the growth of earnings for most companies with about 7000 major USA stocks if you can know if for 2% that's 140. The next question is whether the market has persistent biases towards assuming that earnings growth is more stable than it is. The answer is yes. That's an exploitable inefficiency regardless of how good you are at solving the complex stock valuation problem.

For many companies it perfect price doesn't matter too much. Pick a company you know something about and try: https://www.trefis.com/companies
privatefarmer wrote: If we invest for decade we hopefully will see our stocks more than double in price therefore whether you buy a stock at a P/E of 15 vs 25 really isn't as impactful if the stock is worth say 8x what you initially bought it at.
Buying a company for 2x what it is worth vs one that is fairly priced costs you on average 7.1% of your return over a decade or 2.3% over 30 years. It makes a difference even for a long time buy and hold investor. This is not a small issue. Buying a company for 2x what it is worth vs. focusing your buying on ones selling for for 1/2 what it is worth, doubles those numbers. 4.6% is larger than the risk premium for equity over bonds. So if you are making this mistake consistently in your initial buys at the start of 30 years it would wipe out all your profits for buying equity over bonds even over a 30 year holding period. Now cap weighting doesn't force you to always make this mistake but it substantially biases you towards this mistake.

Valuation is not something to be cavalier about. You wouldn't be cavalier about what a 2.3% expense ratio would do to your returns on a mutual fund. Why be cavalier about an equally destructive habit?
appreciate the reply. This actually has strengthened my beliefs towards active vs passive investing. Lets face it - you either truly are passive and just own everything at the cheapest cost (total market) or you have some sort of active strategy (even if it is "hybrid" active ie smart beta). Passive investors simply rely on all the smart active investors to set the market. This point has been said over and over again, but passive investors and active investors have the exact same return as a group before expenses. If you beat the market that has to mean another active investor lost to the market by the same amount. The only question you have to ask yourself is : are you more knowledgeable than the active investor on the other side of the trade? For me that is definitely a "no".

You commented "Most companies are not that complex however. They are much much simpler." -- This line of thinking is exactly how many (myself definitely included) can get into big trouble. Thinking that finding the true "intrinsic value" of a company and buying for less than that can be "simple" is playing with fire, IMO. The fact is, they aren't simple. They are multi-billion dollar companies that have very complex financials and industry outlooks. And, even if you understood everything about a company, nobody can know where an industry will go over any defined period of time. It's just not possible we have been proven time and time again that technology brings in new ideas/products that totally change the world. Apple was a dying company in the 90s now it's the largest in the world. Amazon has lost 80-90% of it's value at the snap of a finger and today it is taking over the world. Sears, Enron, IBM, Microsoft, etc etc all examples of huge companies that have not panned out, at least recently, to what most expected. I firmly do not believe that any companies are that "simple" to predict and I think having that amount of confidence can easily lead to poor returns.

"Why shouldn't you want to own shares in companies that have the best returns and not own shares in companies that have substandard returns?" - of course you do. and so does every other investor. again, for to you "win" and beat the market, someone else possibly equally or more intelligent than you must "lose". Or you can simply be guaranteed the average, before costs, and above average after costs via indexing.

"The evidence shows that to be false, as I've shown above. They collectively price stocks worse than you could even using simple metrics. They do such a dreadful job on average you can still be only slightly less terrible and win." -- now you may be correct, active managers as a group may do far worse than YOU could do. I don't know you, maybe you are a hedge fund manager who has had amazing success. But I do know me. And I am sorry but just as I wouldn't bet against a field of physicians when it comes to diagnosing a disease, I am not going to bet against a field of hedge fund managers or active managers when it comes to pricing securities.

It really comes down to how confident are you in your abilities to value companies. If you truly understand a company so well that you can come up with a more accurate intrinsice value vs what it's market value is, perhaps you should be a consultant for said company instead of just buying the stock? If you understand the industry and the companies that well, better than the collective wisdom of millions of investors. But I don't know how one could be such an expert in multiple industries across multiple companies. It seems that one would have to devote nearly all his/her time studying a single company/industry to become such an expert.

Anyhow, for me "value investing" is a pipe dream. Of course we would like to buy stocks on sale, everyone would. But to think that I could identify such a stock better than the collective wisdom of the market is unrealistic.
CantPassAgain
Posts: 577
Joined: Fri Mar 15, 2013 8:49 pm

Re: trying to understand total market and value investing

Post by CantPassAgain »

privatefarmer wrote:
jbolden1517 wrote:
privatefarmer wrote:The way I think of it is : I want to own companies for 30+ years.
This is a good outline. But there are a lot of minor points of disagreement here.

Let's start with that? Why shouldn't you want to own shares in companies that have the best returns and not own shares in companies that have substandard returns? What does 30 plus years get you? Now if you mean you want to hold a percentage of the USA market, sure that means you want to hold funds. But there is no particular reason you want those funds to want to hold any particular company.
privatefarmer wrote: If the companies I own grow their earnings, the share price will increase at most likely the same rate to keep the P/E anchored at around the historical average (ie 15-20).
That's not quite true. 30 years while not a huge amount of time in the life of a company is long enough to see a noticeable bias. Companies start in industries where the long term return on capital is on average well above the national average. As they invest capital the return on capital within those industries on average frequently declines, competition compresses margins. You should over say a 200 year period expect to see a u-shape where the P/E starts high (because a company is growing things like sales or I/P to use your example below). Then this number lowers as the returns on capital are realized and become unsustainably high. Then start to increase as the industry experiences declining margins and return on reinvested capital falls to well below the national average. A company with growing earnings on average is going to have a substantially higher P/E than one with falling earnings, and they should.
privatefarmer wrote: It all comes down to : what companies will grow their profits the fastest over the next X number of years? Who the hell knows.
Statistically companies in industries whose margins are well below average due to oversupply where demand is increasing. So companies with strong top line growth experiencing bottom line declines. Your typical growth stocks grow their earnings the fastest.
privatefarmer wrote: Just bc a company is classified as "value" or "growth" doesn't tell you anything about the future widgets that they will sell or what their industries will look like over the next 30 years. So buy them all.
Why is "buy them all the answer"? Why not buy who have structural advantages relative to market pricing?
privatefarmer wrote: Or, if you look at price/book, that doesn't consider the "book value" of all the geniuses working at google and facebook, for example. How do you put a price on the intellectual property that they control?
First off that number becomes relevant if you think the companies are likely to be bought. You would want to apply that to unsuccessful and advertising and social media companies. Successful ones you are valuing based on earnings not book. The book is a concern only in that you are going to be valuing their earnings so highly that you know if the company starts faltering the acquisition price will be considerably lower than what you pay today. So you should assign a higher risk premium because the floor is a long way down and move on from there.

The way you estimate the value of I/P for unsuccessful companies is a reasonably discounted fraction of what it costs to acquire the knowledge in the first place, assuming it is worthwhile. This is frequently how you evaluate bio-tech firms since most fail as businesses but are acquired for their research. For research that is promising something like 50% of the investment over the last decade. For research that is very promising you need expertise. For companies in the social media space (Facebook's space), their I/P seems to be mostly worthless, however their customer base has tremendous value. Google has too many sub businesses to do an estimate and they are too complex. Only in a situation where most of them are failing would that number be relevant.

Most companies are not that complex however. They are much much simpler.
privatefarmer wrote: All I know is that I know next to nothing. But I do believe that hedge funds, actively managed funds, and people like Warren Buffett will collectively price every stock better than I could so I might as well buy them all at current prices instead of trying to outsmart those who do this for a living.
The evidence shows that to be false, as I've shown above. They collectively price stocks worse than you could even using simple metrics. They do such a dreadful job on average you can still be only slightly less terrible and win.
privatefarmer wrote: We simply do not know what stocks will grow their earnings the most over the coming decades and that really is the crux of the issue.
First off their are many companies whose earnings are relatively stable or gently falling. A decade is a long time and their earnings may not be as stable as they were 10 years out. But they don't usually go from stable to highly unstable in that short a period of time. Not being able to value Google is not the same as not being able to value most stocks. Google is much more complex than the average company, an almost worst case.

Remember you aren't having to determine the perfect price. You only have to determine if on average the market price is likely too high or too low. You can be wrong 40% of the time and right 60% of the time and still show a huge profit margin for your trouble.
Even if there is no way to know the growth of earnings for most companies with about 7000 major USA stocks if you can know if for 2% that's 140. The next question is whether the market has persistent biases towards assuming that earnings growth is more stable than it is. The answer is yes. That's an exploitable inefficiency regardless of how good you are at solving the complex stock valuation problem.

For many companies it perfect price doesn't matter too much. Pick a company you know something about and try: https://www.trefis.com/companies
privatefarmer wrote: If we invest for decade we hopefully will see our stocks more than double in price therefore whether you buy a stock at a P/E of 15 vs 25 really isn't as impactful if the stock is worth say 8x what you initially bought it at.
Buying a company for 2x what it is worth vs one that is fairly priced costs you on average 7.1% of your return over a decade or 2.3% over 30 years. It makes a difference even for a long time buy and hold investor. This is not a small issue. Buying a company for 2x what it is worth vs. focusing your buying on ones selling for for 1/2 what it is worth, doubles those numbers. 4.6% is larger than the risk premium for equity over bonds. So if you are making this mistake consistently in your initial buys at the start of 30 years it would wipe out all your profits for buying equity over bonds even over a 30 year holding period. Now cap weighting doesn't force you to always make this mistake but it substantially biases you towards this mistake.

Valuation is not something to be cavalier about. You wouldn't be cavalier about what a 2.3% expense ratio would do to your returns on a mutual fund. Why be cavalier about an equally destructive habit?
appreciate the reply. This actually has strengthened my beliefs towards active vs passive investing. Lets face it - you either truly are passive and just own everything at the cheapest cost (total market) or you have some sort of active strategy (even if it is "hybrid" active ie smart beta). Passive investors simply rely on all the smart active investors to set the market. This point has been said over and over again, but passive investors and active investors have the exact same return as a group before expenses. If you beat the market that has to mean another active investor lost to the market by the same amount. The only question you have to ask yourself is : are you more knowledgeable than the active investor on the other side of the trade? For me that is definitely a "no".

You commented "Most companies are not that complex however. They are much much simpler." -- This line of thinking is exactly how many (myself definitely included) can get into big trouble. Thinking that finding the true "intrinsic value" of a company and buying for less than that can be "simple" is playing with fire, IMO. The fact is, they aren't simple. They are multi-billion dollar companies that have very complex financials and industry outlooks. And, even if you understood everything about a company, nobody can know where an industry will go over any defined period of time. It's just not possible we have been proven time and time again that technology brings in new ideas/products that totally change the world. Apple was a dying company in the 90s now it's the largest in the world. Amazon has lost 80-90% of it's value at the snap of a finger and today it is taking over the world. Sears, Enron, IBM, Microsoft, etc etc all examples of huge companies that have not panned out, at least recently, to what most expected. I firmly do not believe that any companies are that "simple" to predict and I think having that amount of confidence can easily lead to poor returns.

"Why shouldn't you want to own shares in companies that have the best returns and not own shares in companies that have substandard returns?" - of course you do. and so does every other investor. again, for to you "win" and beat the market, someone else possibly equally or more intelligent than you must "lose". Or you can simply be guaranteed the average, before costs, and above average after costs via indexing.

"The evidence shows that to be false, as I've shown above. They collectively price stocks worse than you could even using simple metrics. They do such a dreadful job on average you can still be only slightly less terrible and win." -- now you may be correct, active managers as a group may do far worse than YOU could do. I don't know you, maybe you are a hedge fund manager who has had amazing success. But I do know me. And I am sorry but just as I wouldn't bet against a field of physicians when it comes to diagnosing a disease, I am not going to bet against a field of hedge fund managers or active managers when it comes to pricing securities.

It really comes down to how confident are you in your abilities to value companies. If you truly understand a company so well that you can come up with a more accurate intrinsice value vs what it's market value is, perhaps you should be a consultant for said company instead of just buying the stock? If you understand the industry and the companies that well, better than the collective wisdom of millions of investors. But I don't know how one could be such an expert in multiple industries across multiple companies. It seems that one would have to devote nearly all his/her time studying a single company/industry to become such an expert.

Anyhow, for me "value investing" is a pipe dream. Of course we would like to buy stocks on sale, everyone would. But to think that I could identify such a stock better than the collective wisdom of the market is unrealistic.
Thank you for a much needed dose of sanity, Privatefarmer.
jbolden1517
Posts: 868
Joined: Sun Jul 09, 2017 2:53 pm

Re: trying to understand total market and value investing

Post by jbolden1517 »

privatefarmer wrote: appreciate the reply. This actually has strengthened my beliefs towards active vs passive investing. Lets face it - you either truly are passive and just own everything at the cheapest cost (total market) or you have some sort of active strategy (even if it is "hybrid" active ie smart beta).
I might even disagree with you there. I don't think there is a neutral strategy. Holding stocks based on cap weighting and being a float trader (what you are advocating described less mystically) is not IMHO fundamentally different from holding stocks based on sales weighting and being a sales volume trader which is not fundamentally different from holding stocks based on earnings weighting and being an earnings trader. There is a minor difference that your strategy is slightly cheaper to implement. So all strategies being equal you should come out slightly ahead. But you are definitely asserting that either all strategies are equal or for some reason float trading is the best strategy. That isn't a given.
privatefarmer wrote: Passive investors simply rely on all the smart active investors to set the market.
First off that's cap weighted indexers. The smart beta crowd are passive just as much as the cap weighted indexers. I would agree that cap weighted indexers ignore valuation and focus on float. So while they rely on the market to set price they very aggressively punish concentration (like stock buybacks or insider buying) and reward stock dilution. They don't let the market set prices there, they intervene.

privatefarmer wrote: This point has been said over and over again, but passive investors and active investors have the exact same return as a group before expenses.
I would agree it has been said over and over again. It is based on some oversimplifying assumptions about the market. In particular that all participants are passive minority shareholders who intend to profit from either capital gains or dividends on stocks. When indexing was one among many passive minority theories that simplifying assumption was probably fine. With cap weighted indexing (including quasi indexing among many "actively managed" funds) having become the dominant means of investing for passive minority shareholders it no longer is safe to oversimplify in that way. The same as the spherical horse. For rough purposes of estimation we can assume horses are spheres and the world is frictionless, once we need to get more precise the complexity overwhelms those simplifying assumptions.
privatefarmer wrote: If you beat the market that has to mean another active investor lost to the market by the same amount. The only question you have to ask yourself is : are you more knowledgeable than the active investor on the other side of the trade?
And that's a good example of the problem. Let's take the simplest counter example of mutual funds. A holds mutual fund B managed by C. D holds mutual fund E managed by F. A's excess return may have to come at the expense of D, but C's excess return doesn't have to come at the expense of F's. Both C and F in the aggregate draw returns from A and B investments in the aggregate. Most of the return for B and E goes to A and B. But C and F can engage in behaviors to boost their combined returns from B and E quite easily. C and F's incentives don't align with A and B's incentives.

The horse is not spherical.
privatefarmer wrote:
You commented "Most companies are not that complex however. They are much much simpler." -- This line of thinking is exactly how many (myself definitely included) can get into big trouble. Thinking that finding the true "intrinsic value" of a company and buying for less than that can be "simple" is playing with fire, IMO. The fact is, they aren't simple.
I just showed you multiple examples of really dumb strategies out performing the market. Ignore everything about a company and just focus on wheter it pays a dividend or not. Growers outperform non payers and cutters 10::1 over 30 years. Ignore everything about the business and just buy and sell based on the ratio to had assets. Outperforms by 20::1 over 70 years. That simple worked. Now obviously more complex works better. But it really is that dumb.
privatefarmer wrote: They are multi-billion dollar companies that have very complex financials and industry outlooks. And, even if you understood everything about a company, nobody can know where an industry will go over any defined period of time. It's just not possible we have been proven time and time again that technology brings in new ideas/products that totally change the world.
You don't need to know. You just need probabilistic edge. The more you know the more it helps.
privatefarmer wrote:
jbolden1517 wrote:
"The evidence shows that to be false, as I've shown above. They collectively price stocks worse than you could even using simple metrics. They do such a dreadful job on average you can still be only slightly less terrible and win."
-- now you may be correct, active managers as a group may do far worse than YOU could do. I don't know you, maybe you are a hedge fund manager who has had amazing success. But I do know me. And I am sorry but just as I wouldn't bet against a field of physicians when it comes to diagnosing a disease, I am not going to bet against a field of hedge fund managers or active managers when it comes to pricing securities.
I wouldn't either. If that was what they were doing they would crush me. If they were doing pricing as best they could they would excel. But they aren't pricing securities and trading for their own account. They are attracting investment dollars and either using them to support their much more profitable sell side business or extracting a lot of fees. For the sell side houses particularly they care about creating a market to support the equity and bond sales not returns. 6% of the sales volume far exceeds what they make from buy side clients. They often lose money on the buy side clients after the high expenses of customer acquisition and customer service. For buy side houses they have to attract investment. New investment correlates strongly with the economy. They need to outperform when people have excess money to invest even at the expense of long term total return.

You don't even have to be good at it, as those examples show. You just have to be doing it at all to win.
privatefarmer wrote: Anyhow, for me "value investing" is a pipe dream. Of course we would like to buy stocks on sale, everyone would. But to think that I could identify such a stock better than the collective wisdom of the market is unrealistic.
If it unrealistic then what explains the consistent outperformance of value investing even passive value investing? We have material examples of this happening for long periods of time with real money invested. It can't be unrealistic if it is observable reality.
Post Reply