The random stock portfolio bogey - beating the monkey

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
User avatar
fortyofforty
Posts: 1751
Joined: Wed Mar 31, 2010 12:33 pm

Re: The random stock portfolio bogey - beating the monkey

Post by fortyofforty » Mon Mar 30, 2015 6:38 pm

Back to beating the monkey, I understand that a monkey basket of stocks will, compared with a capitalization weighted index, tend to be overweight in mid and small cap stocks. Just based on the law of averages, that is inevitable. However, is there anything that requires the basket to be overweight in "value" stocks? Couldn't the number of stocks labeled or considered to be "value" be fewer than the number of growth stocks? Couldn't the monkey basket be overweighted towards small and growth, or just small and not tilt towards value?
Indexing works, not because of magic, but because of math. | Diligentia. Vis. Celeritas. - Jeff Cooper | Original Vanguard Diehard

rkhusky
Posts: 7909
Joined: Thu Aug 18, 2011 8:09 pm

Re: The random stock portfolio bogey - beating the monkey

Post by rkhusky » Mon Mar 30, 2015 6:57 pm

fortyofforty wrote:Back to beating the monkey, I understand that a monkey basket of stocks will, compared with a capitalization weighted index, tend to be overweight in mid and small cap stocks. Just based on the law of averages, that is inevitable. However, is there anything that requires the basket to be overweight in "value" stocks? Couldn't the number of stocks labeled or considered to be "value" be fewer than the number of growth stocks? Couldn't the monkey basket be overweighted towards small and growth, or just small and not tilt towards value?
I don't understand that either. Vanguard's Value, Mid Cap Value and Small Cap Value have about the same number of stocks as Growth, Mid Cap Growth and Small Cap Growth. So a randomly choosing monkey should select about the same number of each.

Another thought. If you are just randomly sampling the Total Stock Market with say, just 500 stocks. You may miss all the big players in a given industry. For example, you could easily miss Exxon Mobil, Chevron, Phillips, Marathon, BP and Shell.

User avatar
Uncle Pennybags
Posts: 1836
Joined: Tue Oct 28, 2014 2:05 am

Re: The random stock portfolio bogey - beating the monkey

Post by Uncle Pennybags » Mon Mar 30, 2015 8:42 pm

fortyofforty wrote:Back to beating the monkey,...
Beating sounds a bit harsh. May we spank the monkey?

Topic Author
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: The random stock portfolio bogey - beating the monkey

Post by Browser » Mon Mar 30, 2015 8:48 pm

rkhusky wrote:
fortyofforty wrote:Back to beating the monkey, I understand that a monkey basket of stocks will, compared with a capitalization weighted index, tend to be overweight in mid and small cap stocks. Just based on the law of averages, that is inevitable. However, is there anything that requires the basket to be overweight in "value" stocks? Couldn't the number of stocks labeled or considered to be "value" be fewer than the number of growth stocks? Couldn't the monkey basket be overweighted towards small and growth, or just small and not tilt towards value?
I don't understand that either. Vanguard's Value, Mid Cap Value and Small Cap Value have about the same number of stocks as Growth, Mid Cap Growth and Small Cap Growth. So a randomly choosing monkey should select about the same number of each.

Another thought. If you are just randomly sampling the Total Stock Market with say, just 500 stocks. You may miss all the big players in a given industry. For example, you could easily miss Exxon Mobil, Chevron, Phillips, Marathon, BP and Shell.
Here's Arnott's explanation of why value shows up in non cap-weighted indexes:
There's a value exposure to every one of these. Why is that? It's because if you break the link with price, price means that the higher-priced stocks get higher weight. Take price out of the equation, and the higher-priced stocks no longer are biased to have the higher weight. There is more of the lower-priced stocks, the lower valuation multiple stocks, than there are the high valuation multiple stocks--far more. And so if you strip that linkage away, you're going to wind up with a value tilt.
So what about Malkiel's monkeys? Here you find a value tilt. It's modest. Here you find a big size tilt, because darts are going to mostly land on small companies.
http://www.morningstar.com/cover/videoc ... ?id=613699
We don't know where we are, or where we're going -- but we're making good time.

User avatar
Maynard F. Speer
Posts: 2139
Joined: Wed Mar 18, 2015 10:31 am

Re: The random stock portfolio bogey - beating the monkey

Post by Maynard F. Speer » Mon Mar 30, 2015 9:25 pm

And Ibbotson's new paper (Risk and Return within the Stock Market) addresses this specifically:

It's not just risk you get paid a premium for, it's something about going against the grain too .. Simple explanation: the monkey exposes you to slightly more risk, and therefore slightly better returns .. More complicated explanation: the monkey is a contrarian to boot - so it may be spotting value in ways that don't easily show up through the simple methodologies we use to define value

http://www.zebracapm.com/research.php
"Economics is a method rather than a doctrine, an apparatus of the mind, a technique of thinking, which helps its possessor to draw correct conclusions." - John Maynard Keynes

User avatar
in_reality
Posts: 4529
Joined: Fri Jul 12, 2013 6:13 am

Re: The random stock portfolio bogey - beating the monkey

Post by in_reality » Mon Mar 30, 2015 9:33 pm

rkhusky wrote:
fortyofforty wrote:Back to beating the monkey, I understand that a monkey basket of stocks will, compared with a capitalization weighted index, tend to be overweight in mid and small cap stocks. Just based on the law of averages, that is inevitable. However, is there anything that requires the basket to be overweight in "value" stocks? Couldn't the number of stocks labeled or considered to be "value" be fewer than the number of growth stocks? Couldn't the monkey basket be overweighted towards small and growth, or just small and not tilt towards value?
I don't understand that either. Vanguard's Value, Mid Cap Value and Small Cap Value have about the same number of stocks as Growth, Mid Cap Growth and Small Cap Growth. So a randomly choosing monkey should select about the same number of each.
Even if the numbers are the same (value stocks and growth stocks), not all growth stocks are equal. For two companies in the same growth sector, they may have different valuations. Market cap weighting will give you larger exposure to the company with a higher valuation. Random choosing would give you an opportunity to select one of the more relatively valuey growth companies.

The common view is that if it's not on the "value" side then it's not value, but from a relative standpoint that is not necessarily true.

In any case, I don't think the random dart strategy works in the real world. They are reselecting the portfolio every year, aren't they? You clearly couldn't do it once and they hold it forever -- industries change too much. You'd have a pretty penny to pay in capital gains by doing it every year considering some would be realized as short term gains.

User avatar
fortyofforty
Posts: 1751
Joined: Wed Mar 31, 2010 12:33 pm

Re: The random stock portfolio bogey - beating the monkey

Post by fortyofforty » Mon Mar 30, 2015 9:48 pm

Browser wrote:Here's Arnott's explanation of why value shows up in non cap-weighted indexes:
There's a value exposure to every one of these. Why is that? It's because if you break the link with price, price means that the higher-priced stocks get higher weight. Take price out of the equation, and the higher-priced stocks no longer are biased to have the higher weight. There is more of the lower-priced stocks, the lower valuation multiple stocks, than there are the high valuation multiple stocks--far more. And so if you strip that linkage away, you're going to wind up with a value tilt.
That seems to presuppose that growth stocks are always more highly priced than value stocks. If value stocks become more popular, more in demand, then they will be higher priced than growth stocks. Price is not necessarily linked to capitalization weight, thereby weight in the index, except for that old horse the Dow Jones Industrial Average, and there is nothing to prevent value-labeled companies from rising to the top of the capitalization heap in the index.
Indexing works, not because of magic, but because of math. | Diligentia. Vis. Celeritas. - Jeff Cooper | Original Vanguard Diehard

Leeraar
Posts: 4109
Joined: Tue Dec 10, 2013 8:41 pm
Location: Nowhere

Re: The random stock portfolio bogey - beating the monkey

Post by Leeraar » Tue Mar 31, 2015 3:44 am

fortyofforty wrote:
Browser wrote:Here's Arnott's explanation of why value shows up in non cap-weighted indexes:
There's a value exposure to every one of these. Why is that? It's because if you break the link with price, price means that the higher-priced stocks get higher weight. Take price out of the equation, and the higher-priced stocks no longer are biased to have the higher weight. There is more of the lower-priced stocks, the lower valuation multiple stocks, than there are the high valuation multiple stocks--far more. And so if you strip that linkage away, you're going to wind up with a value tilt.
That seems to presuppose that growth stocks are always more highly priced than value stocks. If value stocks become more popular, more in demand, then they will be higher priced than growth stocks. Price is not necessarily linked to capitalization weight, thereby weight in the index, except for that old horse the Dow Jones Industrial Average, and there is nothing to prevent value-labeled companies from rising to the top of the capitalization heap in the index.
As a side comment, this takes us back to the "investing is not physics" issue. You can make statements about growth vs. value, but you cannot construct a controlled experiment to test your hypotheses. So, we are left with plausible explanations and interpretations of past data.

It is what it is. IMHO, it's sort of futile to look for greater underlying or governing truths.

L.
You can get what you want, or you can just get old. (Billy Joel, "Vienna")

User avatar
Uncle Pennybags
Posts: 1836
Joined: Tue Oct 28, 2014 2:05 am

Re: The random stock portfolio bogey - beating the monkey

Post by Uncle Pennybags » Tue Mar 31, 2015 10:18 am

Leeraar wrote:As a side comment, this takes us back to the "investing is not physics" issue. You can make statements about growth vs. value, but you cannot construct a controlled experiment to test your hypotheses. So, we are left with plausible explanations and interpretations of past data.

It is what it is. IMHO, it's sort of futile to look for greater underlying or governing truths.
How many monkeys get to pick stocks. Some will beat the S&P some won't. With an infinite number of monkeys with an infinite number of typewriters they will eventually produce the entire works of Shakespeare.

Image

User avatar
Maynard F. Speer
Posts: 2139
Joined: Wed Mar 18, 2015 10:31 am

Re: The random stock portfolio bogey - beating the monkey

Post by Maynard F. Speer » Tue Mar 31, 2015 11:46 am

fortyofforty wrote:That seems to presuppose that growth stocks are always more highly priced than value stocks. If value stocks become more popular, more in demand, then they will be higher priced than growth stocks. Price is not necessarily linked to capitalization weight, thereby weight in the index, except for that old horse the Dow Jones Industrial Average, and there is nothing to prevent value-labeled companies from rising to the top of the capitalization heap in the index.
Well at some point many of these 'value' stocks would become (by definition) growth stocks .. Because as their price is pushed up, it becomes higher relative to their fundamentals (failing the definition for a value stock), and if it didn't, these companies would certainly be experiencing above-average growth (the definition for a growth stock)

So value is always going to do a reasonable job of keeping you invested in the less popular ..
"Economics is a method rather than a doctrine, an apparatus of the mind, a technique of thinking, which helps its possessor to draw correct conclusions." - John Maynard Keynes

User avatar
fortyofforty
Posts: 1751
Joined: Wed Mar 31, 2010 12:33 pm

Re: The random stock portfolio bogey - beating the monkey

Post by fortyofforty » Tue Mar 31, 2015 2:00 pm

Maynard F. Speer wrote:
fortyofforty wrote:That seems to presuppose that growth stocks are always more highly priced than value stocks. If value stocks become more popular, more in demand, then they will be higher priced than growth stocks. Price is not necessarily linked to capitalization weight, thereby weight in the index, except for that old horse the Dow Jones Industrial Average, and there is nothing to prevent value-labeled companies from rising to the top of the capitalization heap in the index.
Well at some point many of these 'value' stocks would become (by definition) growth stocks .. Because as their price is pushed up, it becomes higher relative to their fundamentals (failing the definition for a value stock), and if it didn't, these companies would certainly be experiencing above-average growth (the definition for a growth stock)

So value is always going to do a reasonable job of keeping you invested in the less popular ..
I disagree. I don't think there is anything inherent in growth stocks that makes them necessarily higher in capitalization weight than value stocks, or higher in price. People tend to pay more for pieces of those companies, but that isn't necessarily the case. But that's not the real issue.

The only thing that makes a monkey basket of stocks smaller and more value-y than the overall market is numbers. There would have to be more smaller company stocks than larger (which is true) and there would have to be more value stocks than growth stocks (which is where the problem lies). I am not sure that a truly random selection of stocks would always tend to be more value-y than the capitalization weighted average.
Indexing works, not because of magic, but because of math. | Diligentia. Vis. Celeritas. - Jeff Cooper | Original Vanguard Diehard

rkhusky
Posts: 7909
Joined: Thu Aug 18, 2011 8:09 pm

Re: The random stock portfolio bogey - beating the monkey

Post by rkhusky » Tue Mar 31, 2015 3:09 pm

fortyofforty wrote: I am not sure that a truly random selection of stocks would always tend to be more value-y than the capitalization weighted average.
It depends on how you figure the line between growth and value. Suppose I use Price/Book to figure value/growth and further that stocks have a uniform distribution of P/B between 10 and 20. I could draw the line at 15 or at 12. Or I could draw two lines, one at 13 and one at 17, with 13<P/B<17 defined to be neither value or growth. If the distribution is not uniform, then I could draw the line at 15 and have an unequal amount of growth/value stocks.

User avatar
fortyofforty
Posts: 1751
Joined: Wed Mar 31, 2010 12:33 pm

Re: The random stock portfolio bogey - beating the monkey

Post by fortyofforty » Tue Mar 31, 2015 6:40 pm

rkhusky wrote:
fortyofforty wrote: I am not sure that a truly random selection of stocks would always tend to be more value-y than the capitalization weighted average.
It depends on how you figure the line between growth and value. Suppose I use Price/Book to figure value/growth and further that stocks have a uniform distribution of P/B between 10 and 20. I could draw the line at 15 or at 12. Or I could draw two lines, one at 13 and one at 17, with 13<P/B<17 defined to be neither value or growth. If the distribution is not uniform, then I could draw the line at 15 and have an unequal amount of growth/value stocks.
Of course. But my point remains. There is nothing that logically requires there to be more value stocks in a given sample than growth stocks, so nothing that requires a random sample of said stocks to contain more value stocks than growth. I'm still not 100% sure of that, but it seems logical to me.
Indexing works, not because of magic, but because of math. | Diligentia. Vis. Celeritas. - Jeff Cooper | Original Vanguard Diehard

User avatar
in_reality
Posts: 4529
Joined: Fri Jul 12, 2013 6:13 am

Re: The random stock portfolio bogey - beating the monkey

Post by in_reality » Tue Mar 31, 2015 6:58 pm

fortyofforty wrote:
rkhusky wrote:
fortyofforty wrote: I am not sure that a truly random selection of stocks would always tend to be more value-y than the capitalization weighted average.
It depends on how you figure the line between growth and value. Suppose I use Price/Book to figure value/growth and further that stocks have a uniform distribution of P/B between 10 and 20. I could draw the line at 15 or at 12. Or I could draw two lines, one at 13 and one at 17, with 13<P/B<17 defined to be neither value or growth. If the distribution is not uniform, then I could draw the line at 15 and have an unequal amount of growth/value stocks.
Of course. But my point remains. There is nothing that logically requires there to be more value stocks in a given sample than growth stocks, so nothing that requires a random sample of said stocks to contain more value stocks than growth. I'm still not 100% sure of that, but it seems logical to me.
Why does there need to be more value stocks in a random selection to see the value effect in the first place? This misses the point.

Try this, restrict your selectable universe to growth stocks only. Then do a random selection. Then compare the P/B of the market weighted portfolio to your selection. I bet you a cup of coffee that the random selection will be more valuey. That is what they are saying.

So yes by randomly selecting from the total market, likely you'd end up with 50% value stocks and 50% growth stocks (and in any case neither growth nor value stocks would be systematically more likely to be selected), but still your random selection would most likely have more value characteristics than a cap market portfolio. Of course, your random selection could be an outlier but on average it would be more valuey.

rkhusky
Posts: 7909
Joined: Thu Aug 18, 2011 8:09 pm

Re: The random stock portfolio bogey - beating the monkey

Post by rkhusky » Tue Mar 31, 2015 7:47 pm

in_reality wrote: Why does there need to be more value stocks in a random selection to see the value effect in the first place? This misses the point.

Try this, restrict your selectable universe to growth stocks only. Then do a random selection. Then compare the P/B of the market weighted portfolio to your selection. I bet you a cup of coffee that the random selection will be more valuey. That is what they are saying.

So yes by randomly selecting from the total market, likely you'd end up with 50% value stocks and 50% growth stocks (and in any case neither growth nor value stocks would be systematically more likely to be selected), but still your random selection would most likely have more value characteristics than a cap market portfolio. Of course, your random selection could be an outlier but on average it would be more valuey.
The only way that could be true is if the total market is not symmetric between value/growth and small/large. If this is indeed true, then there is likely more small value stocks than small growth and more large growth than large value, assuming the same value/growth dividing line for small and large stocks. Vanguard's index funds do show a slight asymmetry, but its not clear if they are using the same dividing lines for value/growth.

User avatar
fortyofforty
Posts: 1751
Joined: Wed Mar 31, 2010 12:33 pm

Re: The random stock portfolio bogey - beating the monkey

Post by fortyofforty » Tue Mar 31, 2015 9:19 pm

in_reality wrote: Why does there need to be more value stocks in a random selection to see the value effect in the first place? This misses the point.

Try this, restrict your selectable universe to growth stocks only. Then do a random selection. Then compare the P/B of the market weighted portfolio to your selection. I bet you a cup of coffee that the random selection will be more valuey. That is what they are saying.

So yes by randomly selecting from the total market, likely you'd end up with 50% value stocks and 50% growth stocks (and in any case neither growth nor value stocks would be systematically more likely to be selected), but still your random selection would most likely have more value characteristics than a cap market portfolio. Of course, your random selection could be an outlier but on average it would be more valuey.
Try this. Imagine a stock market consisting of 100 growth stocks and 100 value stocks. Then select at random 100 of those 200 stocks. Repeat a thousand times. I'll bet you a cup of coffee that the value and growth characteristics tend to balance each other out. Unless you're claiming that the value stocks are more value-y than the growth stocks are growth-y, then the weighting between growth and value characteristics should be the same. Maybe the 50 average value stocks in our monkey basket really would add more "value" characteristics than the 50 average growth stocks in our monkey basket add "growth" characteristics to that basket, but I don't think so.
Indexing works, not because of magic, but because of math. | Diligentia. Vis. Celeritas. - Jeff Cooper | Original Vanguard Diehard

User avatar
fortyofforty
Posts: 1751
Joined: Wed Mar 31, 2010 12:33 pm

Re: The random stock portfolio bogey - beating the monkey

Post by fortyofforty » Tue Mar 31, 2015 9:23 pm

rkhusky wrote:The only way that could be true is if the total market is not symmetric between value/growth and small/large. If this is indeed true, then there is likely more small value stocks than small growth and more large growth than large value, assuming the same value/growth dividing line for small and large stocks. Vanguard's index funds do show a slight asymmetry, but its not clear if they are using the same dividing lines for value/growth.
I agree with rkhusky. Time to deploy the monkeys.
Indexing works, not because of magic, but because of math. | Diligentia. Vis. Celeritas. - Jeff Cooper | Original Vanguard Diehard

User avatar
in_reality
Posts: 4529
Joined: Fri Jul 12, 2013 6:13 am

Re: The random stock portfolio bogey - beating the monkey

Post by in_reality » Tue Mar 31, 2015 9:49 pm

fortyofforty wrote: Unless you're claiming that the value stocks are more value-y than the growth stocks are growth-y
You still miss the point.

Whether value stocks are more value-y than the growth stocks are growth-y is irrelevant to arriving at a value tilt. Let's assume they are equal. Even with that assumption, it still holds that a random selection of stocks [even if the absolute number of "value" stocks and "growth" stocks is identical] will produce a value tilt relative to market weighting.

This is as there is variance among value stocks as to how value-y they are. There too is variance in growth stocks as to how growth-y they are.

Weighing by market capitalization will result in greater selection of the less value-y value stocks AND a greater selection of the more growth-y growth stocks. It's a fundamental characteristic of market capitalization. Simply put, the more value-y value stocks and the less growth-y growth stocks will have lower capitalization and will have a correspondingly lighter weight under market capitalization.

What your argument to the contrary amounts to basically says that if we have three growth stocks (say internet start ups) and our random selection has to choose one of them, and we have three value stocks (say financial companies) and our random method has to choose one of them [because on average random selection will choose equally between value and growth stocks], that it doesn't matter which growth stock and which value stock gets chosen. It does, though, matter which growth stocks gets chosen and which value stocks get chosen.

The fact of the matter is that there is a wide range of valuations in internet start ups, and a range of valuations in financial stocks, and that random selection on average will result in better odds that the most growth-y growth stock and least value-y value do not get chosen. Market capitalization will ensure the most growth-y growth stock and least value-y value stock get selected and in greater proportion relative to less growth-y growth stocks and more value-y value stocks.

Of course, the market has priced the growth-y growth stocks and the not so value-y value stocks and who are we to argue with market pricing? It's priced like that for a reason.

Holding more value-y stocks (from both the value and growth boxes) may out perform and it may underperform. There can be no guarantees about performance. Randomly selecting stocks, though, is guaranteed to on average produce a value tilt since market capitalization is systematically giving more weight to higher valuations [again valuations which the market it all it's combined wisdom has set], and a randomized weighting have the same systematic tilt towards growth. That, I think, is what confuses you. No, a random select does not tilt towards value. A market cap weighting tilts towards growth (not by selecting more growth stocks but by giving greater weight to less value-y value stocks and greater weight to more growth-y growth stocks) and so a random selection will have a value tilt relative to market capitalization.

TradingPlaces
Posts: 1245
Joined: Sun Nov 09, 2014 1:19 pm
Location: 30.286029, -97.530011

Re: The random stock portfolio bogey - beating the monkey

Post by TradingPlaces » Tue Mar 31, 2015 11:28 pm

There are a lot of sensible explanations:

1. Almost any sensible strategy you pick, that is not actively harvested, will show paper alpha. Try to implement it (trade into and out of it), and the alpha will disappear. This is almost tautalogical: those who trade, collect excess returns, in exchange for making the market efficient. At the same time, you would not expect anyone to trade for a losing strategy (unless, of course, this is the financial advisory industry, who trades OPM: other people's money),

2. I do agree that as a passive investing strategy, market-cap weighted strategy is pretty much the only one that would work. Otherwise, you are paying too much to market intermediaries (and remember, those guys get paid to make the market efficient). Even better: invest in passive benchmarks that reduce trading, e.g., VTI vs SPY. Avoid R2K like the plague.

User avatar
Epsilon Delta
Posts: 8090
Joined: Thu Apr 28, 2011 7:00 pm

Re: The random stock portfolio bogey - beating the monkey

Post by Epsilon Delta » Wed Apr 01, 2015 12:40 am

fortyofforty wrote:
rkhusky wrote:The only way that could be true is if the total market is not symmetric between value/growth and small/large. If this is indeed true, then there is likely more small value stocks than small growth and more large growth than large value, assuming the same value/growth dividing line for small and large stocks. Vanguard's index funds do show a slight asymmetry, but its not clear if they are using the same dividing lines for value/growth.
I agree with rkhusky. Time to deploy the monkeys.
First we don't really need the monkeys. We can look at the equally weighted portfolio. The mean properties of monkey portfolios will be the same as the means of the equally weighted portfolio.

If there is a negative correlation between market cap and value (however you want to measure value) then the equally weighted portolio will be more value-y, if there is a positive correlation the equally weighted portfolio will be more growth-y.
IMO the negative correlation seems more likely, but I can construct plausible examples either way. For somebody with the right data sets it should be easy to check actual data. It would be even better to check for multiple years under different market conditions.

User avatar
JoMoney
Posts: 8225
Joined: Tue Jul 23, 2013 5:31 am

Re: The random stock portfolio bogey - beating the monkey

Post by JoMoney » Wed Apr 01, 2015 2:14 am

Here's a bizarre one for you, even without a "monkey", the cap-weighted S&P500 exhibits alpha when compared to the "Total Market" index as the benchmark:
Image
http://www.etf.com/publications/journal ... nopaging=1
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

User avatar
fortyofforty
Posts: 1751
Joined: Wed Mar 31, 2010 12:33 pm

Re: The random stock portfolio bogey - beating the monkey

Post by fortyofforty » Wed Apr 01, 2015 6:08 am

in_reality wrote:
fortyofforty wrote: Unless you're claiming that the value stocks are more value-y than the growth stocks are growth-y
You still miss the point.

Whether value stocks are more value-y than the growth stocks are growth-y is irrelevant to arriving at a value tilt. Let's assume they are equal. Even with that assumption, it still holds that a random selection of stocks [even if the absolute number of "value" stocks and "growth" stocks is identical] will produce a value tilt relative to market weighting.

This is as there is variance among value stocks as to how value-y they are. There too is variance in growth stocks as to how growth-y they are.

Weighing by market capitalization will result in greater selection of the less value-y value stocks AND a greater selection of the more growth-y growth stocks. It's a fundamental characteristic of market capitalization. Simply put, the more value-y value stocks and the less growth-y growth stocks will have lower capitalization and will have a correspondingly lighter weight under market capitalization.

What your argument to the contrary amounts to basically says that if we have three growth stocks (say internet start ups) and our random selection has to choose one of them, and we have three value stocks (say financial companies) and our random method has to choose one of them [because on average random selection will choose equally between value and growth stocks], that it doesn't matter which growth stock and which value stock gets chosen. It does, though, matter which growth stocks gets chosen and which value stocks get chosen.

The fact of the matter is that there is a wide range of valuations in internet start ups, and a range of valuations in financial stocks, and that random selection on average will result in better odds that the most growth-y growth stock and least value-y value do not get chosen. Market capitalization will ensure the most growth-y growth stock and least value-y value stock get selected and in greater proportion relative to less growth-y growth stocks and more value-y value stocks.

Of course, the market has priced the growth-y growth stocks and the not so value-y value stocks and who are we to argue with market pricing? It's priced like that for a reason.

Holding more value-y stocks (from both the value and growth boxes) may out perform and it may underperform. There can be no guarantees about performance. Randomly selecting stocks, though, is guaranteed to on average produce a value tilt since market capitalization is systematically giving more weight to higher valuations [again valuations which the market it all it's combined wisdom has set], and a randomized weighting have the same systematic tilt towards growth. That, I think, is what confuses you. No, a random select does not tilt towards value. A market cap weighting tilts towards growth (not by selecting more growth stocks but by giving greater weight to less value-y value stocks and greater weight to more growth-y growth stocks) and so a random selection will have a value tilt relative to market capitalization.
You're still missing the point. There is nothing that requires growth stocks to have higher market capitalizations than value stocks, except it's been that way. If people start to pay more for value stocks (stocks graded as value by inherent characteristics independent of price) then value stocks would have a greater weight. It's fairly simple. A truly random selection for our monkey basket might produce an equal number of value and growth stocks, more growth stocks, or more value stocks. We are speaking, in this entire thread, about a random selection of monkey basket stocks. My point is that people might drive up valuations of value stocks if people become convinced that's the place to be.
Indexing works, not because of magic, but because of math. | Diligentia. Vis. Celeritas. - Jeff Cooper | Original Vanguard Diehard

User avatar
Maynard F. Speer
Posts: 2139
Joined: Wed Mar 18, 2015 10:31 am

Re: The random stock portfolio bogey - beating the monkey

Post by Maynard F. Speer » Wed Apr 01, 2015 7:29 am

JoMoney wrote:Here's a bizarre one for you, even without a "monkey", the cap-weighted S&P500 exhibits alpha when compared to the "Total Market" index as the benchmark:
Image
http://www.etf.com/publications/journal ... nopaging=1
That's because the S&P 500 is neither strictly passive, nor strictly cap-weighted

“It is ironic that the largest and most famous index, the S&P 500, is really an active fund in drag. It has momentum rules (market cap weighting), fundamental rules (four quarters of earnings, liquidity requirements) and a subjective overlay (committee input). Does that sound passive to you?” - Faber (Global Value)
"Economics is a method rather than a doctrine, an apparatus of the mind, a technique of thinking, which helps its possessor to draw correct conclusions." - John Maynard Keynes

Topic Author
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: The random stock portfolio bogey - beating the monkey

Post by Browser » Wed Apr 01, 2015 11:46 am

When you think about it, cap-weighting flies in the face of the re-balancing mantra. Everybody on the board seems to believe that it makes sense to rebalance between assets; i.e., sell the better performing assets to buy more of the worse performing assets. But the cap-weighting methodology doesn't do that; it is a trend-following strategy that allows your equity stake to tilt progressively toward the best performing stocks as their cap-weight grows. An equal-weight, monkey, or other non cap-weight strategy actually entails selling the best performing stocks to buy more of the worse performing stocks. Why is this not correct if you believe in rebalancing?
We don't know where we are, or where we're going -- but we're making good time.

rkhusky
Posts: 7909
Joined: Thu Aug 18, 2011 8:09 pm

Re: The random stock portfolio bogey - beating the monkey

Post by rkhusky » Wed Apr 01, 2015 2:34 pm

Browser wrote:When you think about it, cap-weighting flies in the face of the re-balancing mantra. Everybody on the board seems to believe that it makes sense to rebalance between assets; i.e., sell the better performing assets to buy more of the worse performing assets. But the cap-weighting methodology doesn't do that; it is a trend-following strategy that allows your equity stake to tilt progressively toward the best performing stocks as their cap-weight grows. An equal-weight, monkey, or other non cap-weight strategy actually entails selling the best performing stocks to buy more of the worse performing stocks. Why is this not correct if you believe in rebalancing?
I think this is a major negative of the equal weight portfolio. Rebalancing between large segments of the market, like stocks vs. bonds, US vs. Int'l, small-cap vs. large-cap, is not comparable to rebalancing between stocks. Stocks can go to zero. It is unlikely that the small-cap segment of the US market will go to zero. How would you like to have been rebalancing between Ford and GM during the recession? You would want to put some limits on the drops that you are willing to rebalance into, and then of course you would lose the upside if the stock does rebound. And you wouldn't be equal weight anymore either.

Plus, all that buying/selling has got to be expensive.

User avatar
JoMoney
Posts: 8225
Joined: Tue Jul 23, 2013 5:31 am

Re: The random stock portfolio bogey - beating the monkey

Post by JoMoney » Wed Apr 01, 2015 3:15 pm

Browser wrote:...Everybody on the board seems to believe that it makes sense to rebalance between assets; i.e., sell the better performing assets to buy more of the worse performing assets...
Not "everybody" follows a constant-mix rebalancing strategy. Not that it doesn't make sense, or that it's any better or worse than some other strategy. Some of us argue against the purported "rebalancing bonus". You can find lots of articles about Jack Bogle's thoughts on "rebalancing", and lots of other threads on the board regarding it. It's not a bad idea, but it's not necessarily something that is totally uncontested on this board either.

“I don’t rebalance… I leave it alone. I have not touched my asset allocation since March of 2000.” - Jack Bogle
http://www.more.com/reinvention-money/c ... -portfolio

http://www.morningstar.com/cover/videoc ... ?id=615379
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

User avatar
ogd
Posts: 4875
Joined: Thu Jun 14, 2012 11:43 pm

Re: The random stock portfolio bogey - beating the monkey

Post by ogd » Wed Apr 01, 2015 3:17 pm

Browser wrote:When you think about it, cap-weighting flies in the face of the re-balancing mantra. Everybody on the board seems to believe that it makes sense to rebalance between assets; i.e., sell the better performing assets to buy more of the worse performing assets. But the cap-weighting methodology doesn't do that; it is a trend-following strategy that allows your equity stake to tilt progressively toward the best performing stocks as their cap-weight grows. An equal-weight, monkey, or other non cap-weight strategy actually entails selling the best performing stocks to buy more of the worse performing stocks. Why is this not correct if you believe in rebalancing?
Cap-weighting is neutral vs momentum -- it does not trade either way in response to changes in momentum. Equal weighting fights momentum every step of the way. Momentum strategies amplify it . Cap-weighting does neither.

I don't think rebalancing is a mantra. The case for it across different risk classes is clear -- keeping a constant risk profile -- and it does not apply to rebalancing between stocks. The case for rebalancing between domestic and international is (in my book) similarly related to currency risk. Both of these are a way of imposing my individual demands (risk profile, US currency spender) on my portfolio since the market can't possibly do it right for investors with conflicting demands, all at the same time in a single set of prices.

The case for rebalancing between small, value, etc is IMHO quite a bit weaker, but you can still regard is as a way of (weakly) asserting your views about the market. The case for rebalancing between individual stocks is IMHO none at all; I might as well be rebalancing between stocks starting with A and stocks starting with X to counteract the arcane popularity of letters in the English language. In fact, I'm gonna be a little harsh and say that in my book equal weighting based on the rebalancing argument is akin to cargo cult investing. In general, arguments of the sort "if X is so good, then why not do X everywhere" don't convince me. "Mantras" have to live on their own merits when changing contexts.

In response to your earlier question about the benchmark, cap-weighted benchmark special because matching / beating the benchmark means you've matched / beat at least half of the invested dollars [*]. Whereas an equal weight BM can leave you beating the benchmark, but undeperforming almost everyone.

[*] The dollars invested in the asset class, that is. It's true that if you add bonds in the picture or even the global portfolio you lose some of this guarantee, but at least "US stocks" is a big pool. The logistics and ever-changing risk profile of all-asset cap-weighted portfolios make them unfeasible IMHO.

Topic Author
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: The random stock portfolio bogey - beating the monkey

Post by Browser » Wed Apr 01, 2015 6:06 pm

I'm wondering about midcap funds vs. largecap. With midcap, if a stock's capweighted value gets large enough it gets kicked out of the index, which is automatically contrarian since that stock is being removed from your holdings because it has grown too large. At the other end, if a smallcap stock has grown too large for it's index it gets promoted to the midcap index. So, you're adding that stock to your holdings because it is a successful smallcap stock. It would seem this natural process of removing and adding stocks to the midcap index is working favorably. On the other hand, a stock is likely to remain in the largecap index no matter how bloated and overvalued it gets. So a largecap index like the S&P500 is a corral for these kinds of stocks.
We don't know where we are, or where we're going -- but we're making good time.

User avatar
fortyofforty
Posts: 1751
Joined: Wed Mar 31, 2010 12:33 pm

Re: The random stock portfolio bogey - beating the monkey

Post by fortyofforty » Wed Apr 01, 2015 6:22 pm

I'll say it again. These types of civil, intellectual discussions are what keep me coming back to this forum. Thanks to all the participants, and to the moderators for not shutting it down.
Indexing works, not because of magic, but because of math. | Diligentia. Vis. Celeritas. - Jeff Cooper | Original Vanguard Diehard

User avatar
Maynard F. Speer
Posts: 2139
Joined: Wed Mar 18, 2015 10:31 am

Re: The random stock portfolio bogey - beating the monkey

Post by Maynard F. Speer » Wed Apr 01, 2015 7:04 pm

Browser wrote:I'm wondering about midcap funds vs. largecap. With midcap, if a stock's capweighted value gets large enough it gets kicked out of the index, which is automatically contrarian since that stock is being removed from your holdings because it has grown too large. At the other end, if a smallcap stock has grown too large for it's index it gets promoted to the midcap index. So, you're adding that stock to your holdings because it is a successful smallcap stock. It would seem this natural process of removing and adding stocks to the midcap index is working favorably. On the other hand, a stock is likely to remain in the largecap index no matter how bloated and overvalued it gets. So a largecap index like the S&P500 is a corral for these kinds of stocks.
Over the decade our FTSE 250 (UK mid-caps) has returned 5.6% more annually than the large-cap (FTSE 100) index .. Any way to lop off the larger companies to improve returns ..

On the other hand, the big blue chips at the top of the index tend to be the most established, resilient, global corporations (such as pharmaceuticals and electronics) - plus they often tend to be consistent, long-term dividend payers - so they can usually weather market storms better than mid-caps, which are reliant on strong domestic growth

So when you take into account the capital preservation, and the fact that over very long periods, about 75% of market returns have come from dividends, there's good reason to have exposure to those big companies ... I'd agree, for most people, just owning the broad cap-index probably isn't the best option
"Economics is a method rather than a doctrine, an apparatus of the mind, a technique of thinking, which helps its possessor to draw correct conclusions." - John Maynard Keynes

Topic Author
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: The random stock portfolio bogey - beating the monkey

Post by Browser » Thu Apr 02, 2015 12:31 am

Maynard F. Speer wrote:
Browser wrote:I'm wondering about midcap funds vs. largecap. With midcap, if a stock's capweighted value gets large enough it gets kicked out of the index, which is automatically contrarian since that stock is being removed from your holdings because it has grown too large. At the other end, if a smallcap stock has grown too large for it's index it gets promoted to the midcap index. So, you're adding that stock to your holdings because it is a successful smallcap stock. It would seem this natural process of removing and adding stocks to the midcap index is working favorably. On the other hand, a stock is likely to remain in the largecap index no matter how bloated and overvalued it gets. So a largecap index like the S&P500 is a corral for these kinds of stocks.
Over the decade our FTSE 250 (UK mid-caps) has returned 5.6% more annually than the large-cap (FTSE 100) index .. Any way to lop off the larger companies to improve returns ..

On the other hand, the big blue chips at the top of the index tend to be the most established, resilient, global corporations (such as pharmaceuticals and electronics) - plus they often tend to be consistent, long-term dividend payers - so they can usually weather market storms better than mid-caps, which are reliant on strong domestic growth

So when you take into account the capital preservation, and the fact that over very long periods, about 75% of market returns have come from dividends, there's good reason to have exposure to those big companies ... I'd agree, for most people, just owning the broad cap-index probably isn't the best option
It's true that large caps have been less volatile and suffered smaller drawdowns on average compared to midcaps. Of course, we know that midcaps are generally more valuey and are smaller than largecaps, so we'd expect this. The question is: do midcaps return more than largecaps, and if so is it just because they are riskier? As a first look, I compared the returns and volatility of largecap blend index fund to midcap blend using the data in the Simba spreadsheet from 1972-2014. What we find is that midcap blend returned an annual average of 14.4% compared to 11.8% for largecap blend, which is 21.6% higher. Midcap blend was also more volatile with an annualized SD of 19.7% vs. 17.8%, which was 10.8% greater. In 2008, micap blend dropped by 41.8% vs. 37.0%, which was a 13% greater loss than for largecap blend. But the extra return from midcaps isn't fully explained by their greater risk, since the return premium was about twice the size of their incremental volatility. This is seen when we compare the risk-adjusted returns (0.48 for midcaps vs. 0.38 for largecaps). So midcaps have provided some extra juice in the form of risk-adjusted returns. You got paid for the extra risk at about a 2:1 ratio. Of course, the monkey knows this -- he would have selected more midcap stocks.
We don't know where we are, or where we're going -- but we're making good time.

User avatar
fortyofforty
Posts: 1751
Joined: Wed Mar 31, 2010 12:33 pm

Re: The random stock portfolio bogey - beating the monkey

Post by fortyofforty » Thu Apr 02, 2015 6:28 am

I am in complete agreement that the monkey basket will contain more midcap and small cap stocks than the capitalization weight indices. That's math. But how can it be true that the largest capitalization companies tend to be big, old, slow, unexciting dividend paying large cap corporations, but also skew said index towards growth? I'm still not convinced that a monkey basket will contain more value-y stocks, by necessity.

S&P 500 Value (363 holdings):
1 Exxon Mobil Corp.
2 Berkshire Hathaway Inc.
3 General Electric Co.
4 JPMorgan Chase & Co.
5 Pfizer Inc.
6 Verizon Communications Inc.
7 Chevron Corp.
8 AT&T Inc.
9 Bank of America Corp.
10 Citigroup Inc.

S&P 500 Growth (322 holdings):
1 Apple Inc.
2 Microsoft Corp.
3 Google Inc.
4 Johnson & Johnson
5 Facebook Inc.
6 Walt Disney Co.
7 Intel Corp.
8 Gilead Sciences Inc.
9 Home Depot Inc.
10 Amazon.com Inc.

S&P 500 Blend:
1 G Apple Inc.
2 V Exxon Mobil Corp.
3 G Microsoft Corp.
4 G Google Inc.
5 G Johnson & Johnson
6 V Berkshire Hathaway Inc.
7 V General Electric Co.
8 V Wells Fargo & Co.
9 B Procter & Gamble Co.
10 V JPMorgan Chase & Co.

So, out of the top ten S&P 500 companies, four are growth, five are value, and one is in both indices. The total number of holdings in the underlying indices is also skewed towards value. Admittedly, this is from the S&P 500, so it's not the ideal measure, but it's my seat of the pants calculation this morning, but a monkey basket of stocks selected from this subset of the market will contain more value than growth stocks.
Indexing works, not because of magic, but because of math. | Diligentia. Vis. Celeritas. - Jeff Cooper | Original Vanguard Diehard

Topic Author
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: The random stock portfolio bogey - beating the monkey

Post by Browser » Thu Apr 02, 2015 9:43 am

Here are the results of equally weighted 10-stock random portfolio selected from the 500 stocks in the S&P 500 by a five-year old named Jimmy for each of the years 2012-2014.

Image

As you can see, beating the S&P 500 is child's play (even a monkey can do it). Here are Jimmy's selections for 2015:

Image

Let's check back at the end of the year to see how Jimmy did.

http://seekingalpha.com/article/2918596 ... hilds-play
We don't know where we are, or where we're going -- but we're making good time.

Topic Author
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: The random stock portfolio bogey - beating the monkey

Post by Browser » Thu Apr 02, 2015 10:29 am

An older post on the board discusses the "Roach Motel" characteristic of the S&P 500:

"In the old Black Flag commercials for their “Roach Motel” product, their slogan was “the bugs check in, but they don’t check out.” In some indexes, such as the S&P 500, stocks are selected by a committee based on having grown to become dominant in their industry (as measured by cap-weighted value) and most stay in the index until their price stagnates or falls and they are eliminated like pests by the committee. There is some evidence that the index acts like a "roach motel" in this article published in 2005: http://www.hussmanfunds.com/rsi/misfitstocks.htm"
The median annualized returns of removed stocks outperformed the shares of companies that were added to the index in every year going back to 1998. The median returns were positive in each year, even during the three-year bear market. The average annualized returns of the deleted companies were nearly as impressive. In five of the seven years, deleted stocks outperformed those added to the index.
http://www.bogleheads.org/forum/viewtop ... 10&t=52851

A random selection of stocks from the S&P 500 is likely to give more weight to the "pests" that have fallen from favor, and consequently their price has fallen and they have less weight in the S&P 500 index. These would be the more "valuey" stocks in the index. Apparently, it would be better to invest in the worst of these. We need an index fund based on the stocks that have just been kicked out of the S&P 500; we can call it the "anti-S&P 500 index", or the S&P 500 flunkies index".
Last edited by Browser on Thu Apr 02, 2015 11:48 am, edited 1 time in total.
We don't know where we are, or where we're going -- but we're making good time.

User avatar
Maynard F. Speer
Posts: 2139
Joined: Wed Mar 18, 2015 10:31 am

Re: The random stock portfolio bogey - beating the monkey

Post by Maynard F. Speer » Thu Apr 02, 2015 10:57 am

Browser wrote:Here are the results of equally weighted 10-stock random portfolio selected from the 500 stocks in the S&P 500 by a five-year old named Jimmy for each of the years 2012-2014.

Image

As you can see, beating the S&P 500 is child's play (even a monkey can do it). Here are Jimmy's selections for 2015:

Image

Let's check back at the end of the year to see how Jimmy did.

http://seekingalpha.com/article/2918596 ... hilds-play

Well I think the conclusion of that article is rather reductive ..

The *real* risk of investing in the stock market is that the market tanks .. If you look at how long it's taking Japan (once the world's 2nd largest economy) to recover from its crash in the 80s, this would've hit fund managers, indexes, children and monkeys alike .. a 20-30 year bear market is not a particularly rare occurrence

The reason a hedge fund usually lags is because a child's stock selection can easily all go to zero; while a good hedge fund is designed to stand up when every other asset class is tanking ... To say that's a damning critique of active management is the kind of thinking that will lead you to the poor-house
"Economics is a method rather than a doctrine, an apparatus of the mind, a technique of thinking, which helps its possessor to draw correct conclusions." - John Maynard Keynes

Twins Fan
Posts: 2775
Joined: Fri Mar 08, 2013 1:02 pm

Re: The random stock portfolio bogey - beating the monkey

Post by Twins Fan » Thu Apr 02, 2015 11:01 am

Uncle Pennybags wrote:
fortyofforty wrote:Back to beating the monkey,...
Beating sounds a bit harsh. May we spank the monkey?
I admit I have not read this thread... just tuned in to the last page to check it out. Too much on the last page for me to read through also, so I'm glad this was towards the top.

Every time I saw this thread title at the top of this section, I couldn't help but allow my juvenile mind to take over for a second. Glad to know I'm not the only one! :D :beer

Topic Author
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: The random stock portfolio bogey - beating the monkey

Post by Browser » Thu Apr 02, 2015 1:26 pm

Of course, we could backtest the results of Little Jimmy's random 10-stock portfolio to see how it would have fared over more than just the last three years. But fortunately, Arnott et al. have already done something similar:
It would be time-consuming and costly to arrange for a monkey to throw darts at the Wall Street
Journal’’s stock pages, not to mention tracking down 50 years of archived copies of their stock lists. So,
we simulate a dart-throwing monkey, picking a random 30-stock portfolio out of the top 1,000 largest
stocks by market capitalization once a year. We then equal-weight the random stock selections to form
the portfolio. We repeat the exercise 100 times, and examine both the individual year trials and the
average of the trials.
Of course, we know that Arnott's digital monkey also beat the cap-weighted index 99 out of 100 trials. So, Little Jimmy's success isn't just a fluke. You should have such a batting average, Joe Dimaggio.

http://www.q-group.org/wp-content/uploa ... ll_Hsu.pdf
We don't know where we are, or where we're going -- but we're making good time.

User avatar
Maynard F. Speer
Posts: 2139
Joined: Wed Mar 18, 2015 10:31 am

Re: The random stock portfolio bogey - beating the monkey

Post by Maynard F. Speer » Thu Apr 02, 2015 2:05 pm

Browser wrote:Of course, we could backtest the results of Little Jimmy's random 10-stock portfolio to see how it would have fared over more than just the last three years. But fortunately, Arnott et al. have already done something similar:
It would be time-consuming and costly to arrange for a monkey to throw darts at the Wall Street
Journal’’s stock pages, not to mention tracking down 50 years of archived copies of their stock lists. So,
we simulate a dart-throwing monkey, picking a random 30-stock portfolio out of the top 1,000 largest
stocks by market capitalization once a year. We then equal-weight the random stock selections to form
the portfolio. We repeat the exercise 100 times, and examine both the individual year trials and the
average of the trials.
Of course, we know that Arnott's digital monkey also beat the cap-weighted index 99 out of 100 trials. So, Little Jimmy's success isn't just a fluke. You should have such a batting average, Joe Dimaggio.

http://www.q-group.org/wp-content/uploa ... ll_Hsu.pdf
But of course the fundamental-weighted and inverse strategies are producing much better results than the monkey (and Jimmy)
"Economics is a method rather than a doctrine, an apparatus of the mind, a technique of thinking, which helps its possessor to draw correct conclusions." - John Maynard Keynes

rkhusky
Posts: 7909
Joined: Thu Aug 18, 2011 8:09 pm

Re: The random stock portfolio bogey - beating the monkey

Post by rkhusky » Thu Apr 02, 2015 2:34 pm

Browser wrote: Of course, we know that Arnott's digital monkey also beat the cap-weighted index 99 out of 100 trials. So, Little Jimmy's success isn't just a fluke. You should have such a batting average, Joe Dimaggio.

http://www.q-group.org/wp-content/uploa ... ll_Hsu.pdf
Gotta luv the final paragraph of the paper:
We omit from this study the discussion of transaction costs and investment capacity. This is done for
simplicity, given our purpose. At the same time, the costs and capacity differences between strategies
can make a significant difference for investors who are interested in assessing the true investment
benefits of these strategies. Given that both sensible and senseless strategies outperform for the same
reasons (value and small-cap tilts), potential investors would do well to base much of their decision on
the comparison of implementation costs associated with turnover and market price impact.
Wonder what their tables would look like with costs taken into account? Total return includes costs.

User avatar
Runalong
Posts: 138
Joined: Mon Jun 13, 2011 5:22 pm

Re: The random stock portfolio bogey - beating the monkey

Post by Runalong » Thu Apr 02, 2015 3:05 pm

"Let's check back at the end of the year to see how Jimmy did."

I non-randomly predict that TRV & TSO will outperform VTR & SJM the next 3 quarters.

It's too small of a sample to mean anything but let's check back at the end of the year and see if my prediction would have been worth more or less than 1¢ to Jimmy.

Topic Author
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: The random stock portfolio bogey - beating the monkey

Post by Browser » Thu Apr 02, 2015 3:17 pm

Wonder what their tables would look like with costs taken into account? Total return includes costs.
Well, Arnott reports that the monkey added 160 bps annually to the returns of the index. So I guess that wouldn't be too hard to figure out. If the monkey costs you more than an extra percent or so in costs per year, probably wouldn't be worth messing with. But, as long as you were trading reasonably liquid stocks annually the spreads should be low and round trip brokerage fees would be $14 at Vanguard. It is reported that spreads on the NYSE are less than 5 cents per share and a penny for widely traded stocks. To pick some figures out of the air, let's say your 30-stock portfolio consisted of 10,000 total shares of stock with a value of $250,000 (average $25 per share). The round trip spread for turning over all 30 annually would be an average of 2 - 10 cents per share or $200 - $1000; let's split the difference at $600. Add in the round trip brokerage fees of $14 * 30 = $420. Your total annual costs to turn over all 30 holdings per year would be about $1,000 or 0.4%. You can hold a large cap index fund at Vanguard for as little as 0.05% these days, so your monkey portfolio net return advantage would be cut from 1.6% before expenses to about 1.2% after expenses. Obviously your actual expenses would depend on the share price and number of shares you can purchase, but could still be worth doing. The monkey works pretty cheap these days.
We don't know where we are, or where we're going -- but we're making good time.

rkhusky
Posts: 7909
Joined: Thu Aug 18, 2011 8:09 pm

Re: The random stock portfolio bogey - beating the monkey

Post by rkhusky » Thu Apr 02, 2015 4:19 pm

Browser wrote:
Wonder what their tables would look like with costs taken into account? Total return includes costs.
Well, Arnott reports that the monkey added 160 bps annually to the returns of the index. So I guess that wouldn't be too hard to figure out. If the monkey costs you more than an extra percent or so in costs per year, probably wouldn't be worth messing with. But, as long as you were trading reasonably liquid stocks annually the spreads should be low and round trip brokerage fees would be $14 at Vanguard. It is reported that spreads on the NYSE are less than 5 cents per share and a penny for widely traded stocks. To pick some figures out of the air, let's say your 30-stock portfolio consisted of 10,000 total shares of stock with a value of $250,000 (average $25 per share). The round trip spread for turning over all 30 annually would be an average of 2 - 10 cents per share or $200 - $1000; let's split the difference at $600. Add in the round trip brokerage fees of $14 * 30 = $420. Your total annual costs to turn over all 30 holdings per year would be about $1,000 or 0.4%. You can hold a large cap index fund at Vanguard for as little as 0.05% these days, so your monkey portfolio net return advantage would be cut from 1.6% before expenses to about 1.2% after expenses. Obviously your actual expenses would depend on the share price and number of shares you can purchase, but could still be worth doing. The monkey works pretty cheap these days.
So, when are you going move your stocks to the monkey portfolio? :-)

And, when you do, what will your bond allocation be? And how would that compare to before the move?

User avatar
fortyofforty
Posts: 1751
Joined: Wed Mar 31, 2010 12:33 pm

Re: The random stock portfolio bogey - beating the monkey

Post by fortyofforty » Thu Apr 02, 2015 5:53 pm

OK, so it's very easy to beat the S&P 500 Index and the Total Stock Market Index. Very easy. So easy a child can do it. So, why isn't every little Jimmy doing it, then? It's not hard. Select a random basket of thirty stocks. Replace annually. Comb your hair for your appearance on the cover of Forbes. Close your eyes while your makeup is applied before your television appearance on CNBC. Practice your lines so you can give your speeches without hesitation. Retire to a villa in Italy, only coming out of retirement to become a guest professor at the Wharton School.

If it really were so easy, I can't see why money managers aren't all doing it. Do they not want to make money? Do they not see the accolades that would be heaped upon them for beating those bugaboo indices, year after year? Do they hate fame? Do they detest fortune?
Indexing works, not because of magic, but because of math. | Diligentia. Vis. Celeritas. - Jeff Cooper | Original Vanguard Diehard

User avatar
Maynard F. Speer
Posts: 2139
Joined: Wed Mar 18, 2015 10:31 am

Re: The random stock portfolio bogey - beating the monkey

Post by Maynard F. Speer » Thu Apr 02, 2015 6:14 pm

Well smart-beta strategies are still beating the child and the monkey in the studies .. and still, the time you enter the market, and how long you've been investing, will exert a much stronger basic effect on returns

Costs are the main reason the child and monkey fail in practice ... At what point a portfolio becomes cheaper, buying/rebalancing 100+ stocks vs paying virtually nothing for a Vanguard tracker

Also the problem of running 100 monkey portfolios is you effective smooth out risk - you've really just got an equal weight portfolio ... So forget Jimmy and the Monkey, you can buy an equal weight ETF right now .. Whether it'll continue to beat the market is anyone's guess ... But personally I've done better avoiding cap-weighted indexes (but to some extent the difference is a drop in the ocean over a large, diversified portfolio)
"Economics is a method rather than a doctrine, an apparatus of the mind, a technique of thinking, which helps its possessor to draw correct conclusions." - John Maynard Keynes

rkhusky
Posts: 7909
Joined: Thu Aug 18, 2011 8:09 pm

Re: The random stock portfolio bogey - beating the monkey

Post by rkhusky » Thu Apr 02, 2015 7:41 pm

Maynard F. Speer wrote: But personally I've done better avoiding cap-weighted indexes (but to some extent the difference is a drop in the ocean over a large, diversified portfolio)
Are you avoiding indexes or cap-weighting or both? If you don't have cap-weighting, what is your average ER?

User avatar
Maynard F. Speer
Posts: 2139
Joined: Wed Mar 18, 2015 10:31 am

Re: The random stock portfolio bogey - beating the monkey

Post by Maynard F. Speer » Thu Apr 02, 2015 8:30 pm

rkhusky wrote:Are you avoiding indexes or cap-weighting or both? If you don't have cap-weighting, what is your average ER?
Well I'm not religiously avoiding cap-weighted indexes - it's more how it's worked out ..

I've been favouring dividend indexes for a while (like Vanguard UK Equity Income) - which have done well as QE's had everyone seeking income .. And I've been researching low-vol and minimum variance indexes a lot this year, and my plan with Emerging Markets and Europe is to simplify my current holdings to roughly 50:50 low-cost value active, and low volatility ETF

I think my average ER is about 0.6%, but I may be able to get it down 0.5 .. So could be cheaper
"Economics is a method rather than a doctrine, an apparatus of the mind, a technique of thinking, which helps its possessor to draw correct conclusions." - John Maynard Keynes

Topic Author
Browser
Posts: 4857
Joined: Wed Sep 05, 2012 4:54 pm

Re: The random stock portfolio bogey - beating the monkey

Post by Browser » Thu Apr 02, 2015 8:44 pm

So, when are you going move your stocks to the monkey portfolio? :-)

And, when you do, what will your bond allocation be? And how would that compare to before the move?
rkhusky
Well smart-beta strategies are still beating the child and the monkey in the studies .. and still, the time you enter the market, and how long you've been investing, will exert a much stronger basic effect on returns
Well, like every unconventional strategy there's the matter of tracking error and behavioral risk. Maynard is absolutely correct. When you enter the market and how long you stay the course will trump your strategy. Unless you are able to stick with your unconventional strategy when it's taking on water it will end up not working. Is anyone going to be able to stick with the monkey strategy through thick and thin? Probably not. So most of us are probably better off picking a strategy with enough peer consensus that they can stick with it, even if it will probably turn out to be an inferior one. I'll probably continue to run with the Bogledom herd and maybe tilt a little here and there.

Here's a suggestion though. All the smart-beta people lay claim to being able to beat plain vanilla cap-weighted index funds, subject to tracking error risk. What say these claimants compare their results to the monkey? Apples compared to apples. If they can't beat the monkey they should be run out of the jungle.
We don't know where we are, or where we're going -- but we're making good time.

rkhusky
Posts: 7909
Joined: Thu Aug 18, 2011 8:09 pm

Re: The random stock portfolio bogey - beating the monkey

Post by rkhusky » Fri Apr 03, 2015 9:35 am

Maynard F. Speer wrote: I think my average ER is about 0.6%, but I may be able to get it down 0.5 .. So could be cheaper
Since you can probably get a cap weighted 3-fund portfolio for 0.1%, are your returns providing an extra 0.5% every year on average?

LeeMKE
Posts: 1888
Joined: Mon Oct 14, 2013 9:40 pm

Re: The random stock portfolio bogey - beating the monkey

Post by LeeMKE » Fri Apr 03, 2015 9:47 am

Unread postby ShiftF5 » Sat Mar 21, 2015 8:13 pm

I say we run with this -- "The Monkey Index"

We could do commercials with the monkey (in a suit) throwing the darts, driving his Mercedes, hanging out at his beach house, etc.

Then we IPO this thing.

This is Gold Jerry, Gold I tell Ya.

Who's IN?
I'm with ShiftF5. And I can't believe the amount of discussion this is getting. Hope Spring comes soon so folks can get outside and work off some of this energy.
The mightiest Oak is just a nut who stayed the course.

User avatar
Maynard F. Speer
Posts: 2139
Joined: Wed Mar 18, 2015 10:31 am

Re: The random stock portfolio bogey - beating the monkey

Post by Maynard F. Speer » Fri Apr 03, 2015 11:31 am

rkhusky wrote:
Maynard F. Speer wrote: I think my average ER is about 0.6%, but I may be able to get it down 0.5 .. So could be cheaper
Since you can probably get a cap weighted 3-fund portfolio for 0.1%, are your returns providing an extra 0.5% every year on average?
It's very difficult for me to construct a fair benchmark - and I'm targeting consistent returns rather than market returns ..

But certainly my decision to invest in the UK (where I'm principally invested) through dividend-weighted and active income targeting funds has resulted in consistent mid-single-figure out-performance .. This 12 months particularly has been very weak for the All Share index, where my UK portfolio's up about 12% over the same period

Of course, now I may well be up against some long-term mean reversion .. But I think a 0.5% premium (which is what cheap index trackers cost a few years ago) gives me a portfolio I'm much less worried about
"Economics is a method rather than a doctrine, an apparatus of the mind, a technique of thinking, which helps its possessor to draw correct conclusions." - John Maynard Keynes

Post Reply