Diversification beyond 80 Stocks is not necessary?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Topic Author
shellycoat
Posts: 8
Joined: Tue Jan 14, 2020 6:54 pm

Diversification beyond 80 Stocks is not necessary?

Post by shellycoat » Tue Jan 14, 2020 7:25 pm

Long-time reader (~3 Years)...First-time poster here!

I will jump right into my question. I know! The subject line sounds like "clickbait".

Is the following true?
A financial advisor told me that you do not need more than 80 individual stock for a diverse (equity) portfolio.

Notes:
I was very surprised when hearing that statement. Over the past 5 years, I became very interested in ETFs and really enjoy the low cost and broad diversification!

I was always under the assumption: the more stocks in an ETF the better the diversification. My current portfolio:

70/30 Equities/Bonds
Equity Mix:
70% Vanguard Total Stock Market (IE: VTI)
30% VANGUARD TOTAL INTL STOCK (IE: VXUS)
Bonds:
80% Vanguard Total Bond Market (IE: VBTLX)
20% Vanguard Total International Bond Index (IE: VTABX)

I appreciate the help!!!

snailderby
Posts: 693
Joined: Thu Jul 26, 2018 11:30 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by snailderby » Tue Jan 14, 2020 8:22 pm

Welcome to the forum! This article by Larry Swedroe reviews several studies that have been done on this topic: https://mutualfunds.com/expert-analysis ... re-enough/.

retired@50
Posts: 1053
Joined: Tue Oct 01, 2019 2:36 pm

Re: Diversification beyond 80 Stocks is not necessary?

Post by retired@50 » Tue Jan 14, 2020 8:39 pm

shellycoat wrote:
Tue Jan 14, 2020 7:25 pm
Is the following true?
A financial advisor told me that you do not need more than 80 individual stock for a diverse (equity) portfolio.
Based on reading I've done, I'd say it's likely to be true. However, who needs the hassle of selecting all those companies, ensuring that they are in diverse industries, and diverse sizes, etc. So, say you do all this work, and parse out your money among the companies, your results aren't likely to be any better than an index fund, which you can get for nearly nothing in expense ratio. Worse yet, who would pay an adviser to do all this work. Silly.

I value my time too much, and it appears you do too. Your portfolio looks spot on to me...

Regards,
Boggle - a game from Parker Brothers. Bogle - investor, founder of Vanguard.

User avatar
Wiggums
Posts: 2420
Joined: Thu Jan 31, 2019 8:02 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by Wiggums » Tue Jan 14, 2020 8:46 pm

shellycoat wrote:
Tue Jan 14, 2020 7:25 pm
Long-time reader (~3 Years)...First-time poster here!

I will jump right into my question. I know! The subject line sounds like "clickbait".

Is the following true?
A financial advisor told me that you do not need more than 80 individual stock for a diverse (equity) portfolio.

Notes:
I was very surprised when hearing that statement. Over the past 5 years, I became very interested in ETFs and really enjoy the low cost and broad diversification!

I was always under the assumption: the more stocks in an ETF the better the diversification. My current portfolio:

70/30 Equities/Bonds
Equity Mix:
70% Vanguard Total Stock Market (IE: VTI)
30% VANGUARD TOTAL INTL STOCK (IE: VXUS)
Bonds:
80% Vanguard Total Bond Market (IE: VBTLX)
20% Vanguard Total International Bond Index (IE: VTABX)

I appreciate the help!!!
This allocation looks good to me.

I would never want to select 80 diverse funds and maintain them.

stan1
Posts: 8038
Joined: Mon Oct 08, 2007 4:35 pm

Re: Diversification beyond 80 Stocks is not necessary?

Post by stan1 » Tue Jan 14, 2020 8:51 pm

While I do agree in concept that selecting 80 out of around 500 stocks in the S&P 500 can probably come close in terms of performance the advisor will charge you 1-2% per year to select those 80 stocks and change them every so often to show they are earning their keep.

Total Stock Market (VTI) is 0.03% per year.

It's next to impossible to compete on a level playing field with that cost difference, so the advisor has to put a lot more effort into "selling" you.

Your allocation in your current portfolio fine.

User avatar
nisiprius
Advisory Board
Posts: 39777
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Diversification beyond 80 Stocks is not necessary?

Post by nisiprius » Tue Jan 14, 2020 9:03 pm

I am put off by the ridiculous obviously false precision of the number 80.

Why is the financial advisor discussing this? By any chance is your financial advisor proposing to assemble a portfolio of 80 stocks?

I'd like to see the following article updated... and Bernstein focusses on showing how inadequate 15 or 30 stocks are, and doesn't really explore how many are "enough."

The Fifteen-Stock Diversification Myth
To be blunt, if you think that you can do an adequate job of minimizing portfolio risk with 15 or 30 stocks, then you are imperiling your financial future and the future of those who depend on you. The reason is simple: There are critically important dimensions of portfolio risk beyond standard deviation. The most important is so-called Terminal Wealth Dispersion (TWD). In other words, it is quite possible (in fact, as we shall soon see, quite easy) to put together a 15-stock or 30-stock portfolio with a very low SD, but whose lousy returns will put you in the poorhouse....

The reason is simple: a grossly disproportionate fraction of the total return [of the stock market] came from a very few "superstocks" like Dell Computer, which increased in value over 550 times. If you didn’t have one of the half-dozen or so of these in your portfolio, then you badly lagged the market. (The odds of owing one of the 10 superstocks are approximately one in six.) Of course, by owning only 15 stocks you also increase your chances of becoming fabulously rich. But unfortunately, in investing, it is all too often true that the same things that maximize your chances of getting rich also maximize your chances of getting poor.

If the O’Neal data are generalizable to stocks, and I believe that they are, then even 100 stocks are not nearly enough to eliminate this very important source of financial risk.

So, yes, Virginia, you can eliminate nonsytematic portfolio risk, as defined by Modern Portfolio Theory, with a relatively few stocks. It’s just that nonsystematic risk is only a small part of the puzzle. Fifteen stocks is not enough. Thirty is not enough. Even 200 is not enough. The only way to truly minimize the risks of stock ownership is by owning the whole market.
The "a very few 'superstocks'" is supported by other research; for example, But it is extremely interesting, and the idea that "the total return of the stock market" depends on a tiny number of superstocks is supported elsewhere. For example, there's a 2017 paper by Hendrik Bessembinder, Do Stocks Outperform Treasury Bills?
When stated in terms of lifetime dollar wealth creation, the best-performing four percent of listed companies explain the net gain for the entire U.S. stock market since 1926, as other stocks collectively matched Treasury bills.
Last edited by nisiprius on Wed Jan 15, 2020 7:50 am, edited 1 time in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
CyclingDuo
Posts: 2833
Joined: Fri Jan 06, 2017 9:07 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by CyclingDuo » Tue Jan 14, 2020 9:19 pm

shellycoat wrote:
Tue Jan 14, 2020 7:25 pm
Long-time reader (~3 Years)...First-time poster here!

I will jump right into my question. I know! The subject line sounds like "clickbait".

Is the following true?
A financial advisor told me that you do not need more than 80 individual stock for a diverse (equity) portfolio.

Notes:
I was very surprised when hearing that statement. Over the past 5 years, I became very interested in ETFs and really enjoy the low cost and broad diversification!

I was always under the assumption: the more stocks in an ETF the better the diversification. My current portfolio:

70/30 Equities/Bonds
Equity Mix:
70% Vanguard Total Stock Market (IE: VTI)
30% VANGUARD TOTAL INTL STOCK (IE: VXUS)
Bonds:
80% Vanguard Total Bond Market (IE: VBTLX)
20% Vanguard Total International Bond Index (IE: VTABX)

I appreciate the help!!!
Read the Wiki on Passively Managing Individual Stocks and the studies that are cited at this link:

https://www.bogleheads.org/wiki/Passive ... ual_stocks

In Common Sense on Mutual Funds, Jack Bogle suggests that a reasonable alternative to an index fund for some investors would be to hold a well-diversified portfolio of individual stocks, as long as they are held long-term, with a minimum of trading costs incurred. This article outlines some suggestions for how to build a portfolio of individual stocks to cover at least part of one's overall stock allocation. It will also attempt to summarize the advantages and possible pitfalls of doing so.

Note that the discussion here assumes that one is not trying to beat the market, but rather, by passively managing individual stocks create a "DIY index fund."


Key in on the number of stocks each study mentions (skew, costs, tax loss harvesting, etc...) and how they arrived at their data to see how close the 80 stock recommendation from the financial advisor was or not regarding matching the performance of the index. :beer

CyclingDuo
"Everywhere is within walking distance if you have the time." ~ Steven Wright

User avatar
patrick013
Posts: 2799
Joined: Mon Jul 13, 2015 7:49 pm

Re: Diversification beyond 80 Stocks is not necessary?

Post by patrick013 » Tue Jan 14, 2020 9:27 pm

shellycoat wrote:
Tue Jan 14, 2020 7:25 pm

Is the following true?
A financial advisor told me that you do not need more than 80 individual stock for a diverse (equity) portfolio.

The traditional theory of investing says that it is better to have more stocks than it is to have less. The studies usually find that 50 stocks will reduce standard deviation very well and after 100 stocks will reduce standard deviation very little additionally.

So if I had an index of stocks with above average growth historically and expected and below average LT debt ratio I would probably need 100 stocks in the index. At 50 stocks I might pass up stocks that would add profits and at say 500 stocks the index could be watered down with less quality companies that would reduce profits and price growth. So in concept SD is reduced and similar quality companies are included in the index up to a point as the index selection is to include the better companies then.

Whether this works in practice would need to be backtested to see if it is true. If the S&P 500 was the 400 it could include better stocks and be more profitable. SD should be about the same. If it only had 50 stocks additional stocks added could boost returns over the many cycles.
age in bonds, buy-and-hold, 10 year business cycle

User avatar
JoMoney
Posts: 8194
Joined: Tue Jul 23, 2013 5:31 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by JoMoney » Tue Jan 14, 2020 9:43 pm

If you were picking small cap stocks, you'd likely need a lot more.
If you stick predominately with large cap stocks, you'd likely be ok with even less.

The top 100 stocks by weight make up more than half the U.S. market.
Large cap stocks tend to be less volatile than small cap stocks. You can reach the point where adding more stocks to a large cap portfolio increases volatility pretty quickly as you have to start reaching lower into the riskier more volatile stocks to add more. The big question though, is whether or not the riskier more volatile stocks will add enough return to compensate to a higher "risk adjusted return"... which is not as easy to answer as the "efficient market" theorists will suggest with their models.
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

shess
Posts: 368
Joined: Wed May 17, 2017 12:02 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by shess » Tue Jan 14, 2020 10:21 pm

shellycoat wrote:
Tue Jan 14, 2020 7:25 pm
Is the following true?
A financial advisor told me that you do not need more than 80 individual stock for a diverse (equity) portfolio.

Notes:
I was very surprised when hearing that statement. Over the past 5 years, I became very interested in ETFs and really enjoy the low cost and broad diversification!

I was always under the assumption: the more stocks in an ETF the better the diversification. My current portfolio:
I think there's a bit of an "uncanny valley" issue, here. If I could replicate VTI (Vanguard Total Stock Market ETF) buy purchasing a single stock, then that seems like a great deal. It's exactly like the ETF, except I save .03%/year!

OK, that's crazy, but maybe I can replicate the market by investing in 10 stocks? Again, I'm saving the .03%/year, but now I'm starting to wonder how much overhead I'm willing to put up with to save that .03%? I'll have dividends coming in randomly, maybe stock splits, perhaps someone gets acquired, etc. None of that is incredibly complicated, but it's not like this is my job or something.

Now, let's ramp that up to 80. By that point, things are really all over the map, you're likely getting multiple dividend payments per week, it's almost certain that in any given year someone is acquiring someone or splitting their stock, perhaps multiple someones. So it's becoming increasingly unclear if that .03% cost is really all that bad for what you get!

So, put another dimension on it - I'd probably be willing to put up with that on a $100M portfolio. .03% of $100M is $30k/year to handle these issue. Seems like a fair reward, though if I had $100M maybe that $30k isn't so interesting. But am I willing to do it for the $30/year savings on a $100k portfolio? No way!

What is the financial advisor willing to do it for? I'd be surprised if they're going to match VTI's .03%. They'll probably make arguments about how individual stocks are better, tax-loss harvesting, etc, but IMHO those gains are arguable. The bulk of your TLH opportunities come in the first few years of ownership, but after a decade or so your positions have likely appreciated enough that TLH is unlikely.

To my mind, the main advantage of this kind of index replication for financial advisors is that it makes it much more challenging to move to another provider. You can transfer your assets "in kind", but nobody else is going to be interested in managing your existing account.

User avatar
Cubicle
Posts: 316
Joined: Sun Sep 22, 2019 1:43 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by Cubicle » Wed Jan 15, 2020 1:06 am

Assert allocation looks good.

"80" is a silly number with no meaning. Potentially 1 really good stock would be all you "needed".

User avatar
1789
Posts: 634
Joined: Fri Aug 16, 2019 3:31 pm
Location: OR

Re: Diversification beyond 80 Stocks is not necessary?

Post by 1789 » Wed Jan 15, 2020 2:44 am

Your portfolio looks great. Is financial advisor have an ETF to offer you with 80 stocks? Maybe he is trying to sell you that one. But yeah 80 sounds stupid, i would think a “carefully selected portfolio with 88” stocks perform better than a portfolio with 80 stocks.
"My conscience wants vegetarianism to win over the world. And my subconscious is yearning for a piece of juicy meat. But what do i want?" (Andrei Tarkovsky)

User avatar
danielc
Posts: 944
Joined: Sun Dec 10, 2017 4:48 am
Location: Iowa, USA
Contact:

Re: Diversification beyond 80 Stocks is not necessary?

Post by danielc » Wed Jan 15, 2020 3:12 am

shellycoat wrote:
Tue Jan 14, 2020 7:25 pm
Is the following true?
A financial advisor told me that you do not need more than 80 individual stock for a diverse (equity) portfolio.
Your financial advisor is an idiot and you should fire him for saying that.

If you were to model all stocks as following roughly the same random walk, then yes, you would find that 80 stocks gives you about as much diversification as 1,000. But this is a toy model that does not reflect how stocks work. First of all, 80 stocks is not even remotely enough to diversify sector risks. Secondly, stocks do not all follow the same random walk and a surprisingly small number of stocks are responsible for a vastly disproportionate portion of the stock market return. In the last decade or so, a handful of companies (I think mostly the FAANGs --- Facebook, Amazon, Apple, Google) were responsible for the majority of the stock market return and without them the overall market return would be pretty crappy. If you just have 80 stocks, you are basically guaranteed to miss out on the few lucky mega-winners of the next decade.

Seasonal
Posts: 632
Joined: Sun May 21, 2017 1:49 pm

Re: Diversification beyond 80 Stocks is not necessary?

Post by Seasonal » Wed Jan 15, 2020 3:24 am

Even if the statement were true, is the advisor arguing 80 is better than a broad index?

Blue456
Posts: 268
Joined: Tue Jun 04, 2019 5:46 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by Blue456 » Wed Jan 15, 2020 7:10 am

danielc wrote:
Wed Jan 15, 2020 3:12 am
shellycoat wrote:
Tue Jan 14, 2020 7:25 pm
Is the following true?
A financial advisor told me that you do not need more than 80 individual stock for a diverse (equity) portfolio.
Your financial advisor is an idiot and you should fire him for saying that.

If you were to model all stocks as following roughly the same random walk, then yes, you would find that 80 stocks gives you about as much diversification as 1,000. But this is a toy model that does not reflect how stocks work. First of all, 80 stocks is not even remotely enough to diversify sector risks. Secondly, stocks do not all follow the same random walk and a surprisingly small number of stocks are responsible for a vastly disproportionate portion of the stock market return. In the last decade or so, a handful of companies (I think mostly the FAANGs --- Facebook, Amazon, Apple, Google) were responsible for the majority of the stock market return and without them the overall market return would be pretty crappy. If you just have 80 stocks, you are basically guaranteed to miss out on the few lucky mega-winners of the next decade.
Isn't Wellesley also limited to few stocks, they seemed to have done well for the past several decades.

User avatar
nisiprius
Advisory Board
Posts: 39777
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Diversification beyond 80 Stocks is not necessary?

Post by nisiprius » Wed Jan 15, 2020 7:47 am

Let's consider the S&P 500, and let's consider broadly the idea of choosing a sample of "80 stocks" = 16% of them. Now add the idea, from a 2017 paper by Bessembinder, that all of the return (above Treasury bills) comes from just 4% of the stocks in it. I'll call these the "superstocks."

It would be great if we could pick 80 stocks and get all 20 of the superstocks, but we are indexers and we don't think we can do that; we don't even want to. Our goal is to have an 80-stock sample of the S&P 500 that matches the market, and, accordingly, contains 4% of 80 = 3.2 superstocks.

It's not worth bothering to be precise so we'll just assume that as we pick 80 stocks at random, each pick has a probability of 4% of being one of the superstocks. This just becomes an example of binomial probabilities. So finding an online calculator I find that we want 3.2 superstocks, and:

The chances of getting two or less is 20.9%.
The chances of only getting one or less is 12.7%.
The chances of missing all twenty superstocks, getting none of them, is 3.8%.

I find that last one daunting. 3.8% is more than the probability of rolling a total of two on a pair of dice. Happens to me in Monopoly all the time. On our premises, picking 80 stocks from the S&P 500 has a far-from-black-swan chance of missing all of the good stocks.

That's enough for me to conclude that there is a meaningful difference between 80 stocks and "all of them."

If it were 1970 and there were no cheap, easy way to get "all of them" then, of course, you'd settle for doing the best you could.

It may be possible to overcome the statistics by picking stocks in some way--such as by intentionally matching sector composition or other characteristics between our sample and the total market, or by outright stock-picking. But on the raw statistics, if we accept the "just a few superstocks" idea, eighty isn't enough to do it automatically.
Last edited by nisiprius on Wed Jan 15, 2020 9:23 am, edited 4 times in total.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

bondsr4me
Posts: 1298
Joined: Fri Oct 18, 2013 7:08 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by bondsr4me » Wed Jan 15, 2020 7:56 am

retired@50 wrote:
Tue Jan 14, 2020 8:39 pm
shellycoat wrote:
Tue Jan 14, 2020 7:25 pm
Is the following true?
A financial advisor told me that you do not need more than 80 individual stock for a diverse (equity) portfolio.
Based on reading I've done, I'd say it's likely to be true. However, who needs the hassle of selecting all those companies, ensuring that they are in diverse industries, and diverse sizes, etc. So, say you do all this work, and parse out your money among the companies, your results aren't likely to be any better than an index fund, which you can get for nearly nothing in expense ratio. Worse yet, who would pay an adviser to do all this work. Silly.

I value my time too much, and it appears you do too. Your portfolio looks spot on to me...

Regards,
+1....especially on "I value my time too much"
I like and own "some" individual stocks...but 80!...no way...not me.

aristotelian
Posts: 6689
Joined: Wed Jan 11, 2017 8:05 pm

Re: Diversification beyond 80 Stocks is not necessary?

Post by aristotelian » Wed Jan 15, 2020 9:08 am

Do you need more than 80 stocks? Probably not. But diversification is the only free lunch as they say. At .05% the cost is negligible. To go through the trouble of managing 80 stocks you should have a good justification and I see no reason *not* to diversify.

User avatar
nisiprius
Advisory Board
Posts: 39777
Joined: Thu Jul 26, 2007 9:33 am
Location: The terrestrial, globular, planetary hunk of matter, flattened at the poles, is my abode.--O. Henry

Re: Diversification beyond 80 Stocks is not necessary?

Post by nisiprius » Wed Jan 15, 2020 9:29 am

I think one of the unintended (????) consequences of zero commissions is that it is luring people out of mutual funds and back into portfolios of individual stocks.

I expect very soon brokerages will begin offering tools for managing 80-stock portfolios that let you use your own personalized ideas instead of some boring cookie-cutter index. Want the S&P 500 but without the FAANGs? Want to exclude tobacco but include marijuana? Want emerging markets but not Brazil but with the Índice Bursátil Caracas? You've got 'em... and the fee for the tool will be just pennies per day.

And it will be a great opportunity for advisors to offer customized, individualized portfolios with huge numbers of stocks.

We've already seen cries of pain in this forum from Wealthfront clients who went in for "direct indexing" and have discovered that it has made leaving Wealthfront hard to do.
Annual income twenty pounds, annual expenditure nineteen nineteen and six, result happiness; Annual income twenty pounds, annual expenditure twenty pounds ought and six, result misery.

User avatar
JoMoney
Posts: 8194
Joined: Tue Jul 23, 2013 5:31 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by JoMoney » Wed Jan 15, 2020 9:36 am

^ I can't help but think that "free trades" has, or will have, echoes of the advent of online discount brokers in the late 1990's. I'm not hearing lots of 'day trading' chatter yet... but we'll see.
"To achieve satisfactory investment results is easier than most people realize; to achieve superior results is harder than it looks." - Benjamin Graham

pkcrafter
Posts: 13748
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: Diversification beyond 80 Stocks is not necessary?

Post by pkcrafter » Wed Jan 15, 2020 11:41 am

Shellycoat:
Is the following true?
A financial advisor told me that you do not need more than 80 individual stock for a diverse (equity) portfolio.
Notes:
I was very surprised when hearing that statement. Over the past 5 years, I became very interested in ETFs and really enjoy the low cost and broad diversification!
Obviously, the advisor wants to set you up with 80 individual stocks. And why not, it makes him a generous commission.

https://www.nerdwallet.com/blog/investi ... ions-fees/


Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.

User avatar
CyclingDuo
Posts: 2833
Joined: Fri Jan 06, 2017 9:07 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by CyclingDuo » Wed Jan 15, 2020 12:08 pm

pkcrafter wrote:
Wed Jan 15, 2020 11:41 am
Obviously, the advisor wants to set you up with 80 individual stocks. And why not, it makes him a generous commission.

https://www.nerdwallet.com/blog/investi ... ions-fees/

Paul
Paul,

In terms of that list of fees from Nerd Wallet...

Stock trading commissions are gone. Dinosaurs. Extinct.
There are no ER fees for individual stocks.

Key being, a DIY Index fund that is passively managed by the investor themself vs. paying an AUM fee to a financial advisor of say .3% to 1% would be saved. There would be no trading cost fees and no ER fees. Even if one tracked the index fund with their DIY Index fund made up of individual stocks, the fees would be lower. That may or may not help/hurt performance, but just pointing it out.

Most would benefit more from the index funds as they don't have enough capital to deploy in their early investing years to diversify enough with individual stocks (if we are talking 50 to 70, or 80 - 100 stocks) that are both domestic and foreign. They also don't have to select the individual stocks.

The exception being something like a $0 cost account at M1 Finance that allows fractional share purchases and "pies" that can hold say 50 stocks with 2% invested in each of them, or 100 stocks with 1% invested in each of them (or whatever allocation one prefers to use). The built in rebalancing with each subsequent purchase makes it an interesting platform to create and easily maintain your own equal weighting (or whatever weighting you want) DIY Index fund.

Of course now the race to the bottom in fees regarding purchasing fractional shares (outside of dividend reinvesting) that were first available by the micro-investing/robo-investing platforms has come to the major brokerages when Schwab became the first to offer this last fall.

I don't know if there will be an increase in DIY Index fund creation by investors as a result of the low costs and availability, but it at least opens up a strategy that may have been cost prohibitive before for many investors to invest in such a way now and avoid ER fees of mutual funds/EFTs while matching the return of the market with a diverse basket of individual stocks.

https://www.bogleheads.org/wiki/Passive ... ual_stocks
"Everywhere is within walking distance if you have the time." ~ Steven Wright

randomguy
Posts: 8534
Joined: Wed Sep 17, 2014 9:00 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by randomguy » Wed Jan 15, 2020 12:22 pm

nisiprius wrote:
Wed Jan 15, 2020 7:47 am
Let's consider the S&P 500, and let's consider broadly the idea of choosing a sample of "80 stocks" = 16% of them. Now add the idea, from a 2017 paper by Bessembinder, that all of the return (above Treasury bills) comes from just 4% of the stocks in it. I'll call these the "superstocks."

It would be great if we could pick 80 stocks and get all 20 of the superstocks, but we are indexers and we don't think we can do that; we don't even want to. Our goal is to have an 80-stock sample of the S&P 500 that matches the market, and, accordingly, contains 4% of 80 = 3.2 superstocks.

It's not worth bothering to be precise so we'll just assume that as we pick 80 stocks at random, each pick has a probability of 4% of being one of the superstocks. This just becomes an example of binomial probabilities. So finding an online calculator I find that we want 3.2 superstocks, and:

The chances of getting two or less is 20.9%.
The chances of only getting one or less is 12.7%.
The chances of missing all twenty superstocks, getting none of them, is 3.8%.

I find that last one daunting. 3.8% is more than the probability of rolling a total of two on a pair of dice. Happens to me in Monopoly all the time. On our premises, picking 80 stocks from the S&P 500 has a far-from-black-swan chance of missing all of the good stocks.

That's enough for me to conclude that there is a meaningful difference between 80 stocks and "all of them."

If it were 1970 and there were no cheap, easy way to get "all of them" then, of course, you'd settle for doing the best you could.

It may be possible to overcome the statistics by picking stocks in some way--such as by intentionally matching sector composition or other characteristics between our sample and the total market, or by outright stock-picking. But on the raw statistics, if we accept the "just a few superstocks" idea, eighty isn't enough to do it automatically.
This logic is flawed. You are assuming the performance of a subset (S&P 500) is the same as the whole (the 20k+ stocks that showed up in the CRSP list). The is unlikely. The S&P 500 unlikely to have the big winners (i.e. riding a 62 million market cap AOL to 200 billion) but there aren't as many losers (i.e. yes large companies go to 0. But for every Enron that goes to 0 there are hundreds of pets.com that live and die in a couple of years). If you look at the historical returns of things like the DJIA or S&P 100, they do fine. They might have done better with different choices (i.e. adding apple in 2004 instead of 2015 would have really boosted the DJIA).

I would be happy with 80 stocks. It is just hard to come up with a good reason to go with 80 instead of buying an ETF for most situations.

shess
Posts: 368
Joined: Wed May 17, 2017 12:02 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by shess » Wed Jan 15, 2020 12:24 pm

nisiprius wrote:
Wed Jan 15, 2020 7:47 am
Let's consider the S&P 500, and let's consider broadly the idea of choosing a sample of "80 stocks" = 16% of them. Now add the idea, from a 2017 paper by Bessembinder, that all of the return (above Treasury bills) comes from just 4% of the stocks in it. I'll call these the "superstocks."

It would be great if we could pick 80 stocks and get all 20 of the superstocks, but we are indexers and we don't think we can do that; we don't even want to. Our goal is to have an 80-stock sample of the S&P 500 that matches the market, and, accordingly, contains 4% of 80 = 3.2 superstocks.

It's not worth bothering to be precise so we'll just assume that as we pick 80 stocks at random, each pick has a probability of 4% of being one of the superstocks. This just becomes an example of binomial probabilities. So finding an online calculator I find that we want 3.2 superstocks, and:

The chances of getting two or less is 20.9%.
The chances of only getting one or less is 12.7%.
The chances of missing all twenty superstocks, getting none of them, is 3.8%.

I find that last one daunting. 3.8% is more than the probability of rolling a total of two on a pair of dice. Happens to me in Monopoly all the time. On our premises, picking 80 stocks from the S&P 500 has a far-from-black-swan chance of missing all of the good stocks.

That's enough for me to conclude that there is a meaningful difference between 80 stocks and "all of them."
I don't have the background to adjust for it, but this feels like it overstates the possibility. If you had made your investment earlier than 2004, you'd have zero chance of picking Google or Facebook. Before 1997 you'd have zero chance of picking Amazon. That's three of the top ten components by market cap.

Sure, you could have a system which adapts over time to account for new entries, but in that case, you aren't dealing with "only 80 stocks" in any meaningful fashion.

wolf359
Posts: 1973
Joined: Sun Mar 15, 2015 8:47 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by wolf359 » Wed Jan 15, 2020 12:41 pm

shellycoat wrote:
Tue Jan 14, 2020 7:25 pm

Is the following true?
A financial advisor told me that you do not need more than 80 individual stock for a diverse (equity) portfolio.
That might well be true. However, it is easier to buy and manage 1-4 ETFs (that collectively hold hundreds or thousands of stocks) than it is to buy and manage 80 individual stocks.

The fact that those 80 stocks might give me diversification is irrelevant. I'd rather manage 1-4 individual assets than 80 individual assets.

pkcrafter
Posts: 13748
Joined: Sun Mar 04, 2007 12:19 pm
Location: CA
Contact:

Re: Diversification beyond 80 Stocks is not necessary?

Post by pkcrafter » Wed Jan 15, 2020 1:16 pm

CyclingDuo wrote:
Wed Jan 15, 2020 12:08 pm
pkcrafter wrote:
Wed Jan 15, 2020 11:41 am
Obviously, the advisor wants to set you up with 80 individual stocks. And why not, it makes him a generous commission.

https://www.nerdwallet.com/blog/investi ... ions-fees/

Paul
Paul,

In terms of that list of fees from Nerd Wallet...

Stock trading commissions are gone. Dinosaurs. Extinct.
There are no ER fees for individual stocks.
I'm not so sure. I think there are still some fee-based advisors around.

On another note - why did the advisor advocate 80 stocks? If not individual stocks with fees, he's going to suggest a specific fund.


Paul


Paul
When times are good, investors tend to forget about risk and focus on opportunity. When times are bad, investors tend to forget about opportunity and focus on risk.

User avatar
CyclingDuo
Posts: 2833
Joined: Fri Jan 06, 2017 9:07 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by CyclingDuo » Wed Jan 15, 2020 2:12 pm

pkcrafter wrote:
Wed Jan 15, 2020 1:16 pm
CyclingDuo wrote:
Wed Jan 15, 2020 12:08 pm
pkcrafter wrote:
Wed Jan 15, 2020 11:41 am
Obviously, the advisor wants to set you up with 80 individual stocks. And why not, it makes him a generous commission.

https://www.nerdwallet.com/blog/investi ... ions-fees/

Paul
Paul,

In terms of that list of fees from Nerd Wallet...

Stock trading commissions are gone. Dinosaurs. Extinct.
There are no ER fees for individual stocks.
I'm not so sure. I think there are still some fee-based advisors around.

On another note - why did the advisor advocate 80 stocks? If not individual stocks with fees, he's going to suggest a specific fund.
Yes, there are plenty of fee-based advisors around and probably always will be. I was simply pointing out two of the cost items in the list at Nerdwallet that you linked (trading commissions and expense ratio fees).

I guess we should clarify our terminology as the trading commission paid to a brokerage house was, in the past, paid whether you used an advisor or did it yourself for each stock purchase and sale. Many accounts that bought/sold enough trades per year qualified to not have to pay this trading commission charge, but it is now available to all. To clarify, that is a different charge than that of an advisor fee, or management fee which is where our miscommunication developed on the subject.

Anyway, back to the two of the fees that the Nerdwallet list included...

The first being stock trading commission charges for stocks/ETFs. Almost all brokerages now have recently gone to commission free trades for stocks and ETFs. So instead of paying $4.99 or $7.99 or $9.99 or whatever charge one paid when you buy or sell a stock, you now pay $0 commission for the trade. Not a big savings when dealing with large lots (as a buy and a sell would still total up to be less than an expense ratio fee for a mutual fund/ETF), but if only buying one, two, three, four, five, etc... shares - this cost savings is now a rather important savings for somebody accumulating shares and putting capital to work each and every paycheck by purchasing shares. Or tax loss harvesting. Or rebalancing.

The second being the expense ratio fee. Most mutual funds and ETFs, outside of the Fidelity Zero funds, have an expense ratio that is a percentage of the underlying fund charged on an annual basis (most often paid on a quarterly basis).

Removing the trading cost fees and the expense ratio fees improves the chance of matching the market with a diverse basket of stocks.

Why did the advisor recommend the number 80 for individual stocks? Read the studies cited in the Wiki for Passively Managing Individual Stocks. The number 80 is right in the "zone".
"Everywhere is within walking distance if you have the time." ~ Steven Wright

afan
Posts: 4475
Joined: Sun Jul 25, 2010 4:01 pm

Re: Diversification beyond 80 Stocks is not necessary?

Post by afan » Wed Jan 15, 2020 2:27 pm

To follow on from nisi's point.

There have been studies showing that you can get the volatility of a portfolio down to levels similar to the market using a relatively small number of stocks. That does not mean that you address at all the extreme asymmetry of returns. As he points out, you will likely miss most of the stocks that have great returns. You will have variance similar to the market, so you will be worse off than in an index fund.

The bigger question is why this even came up. It is an interesting exercise in sampling to see how many you need to mimic the overall market in aggregate but you would never want to actually do it. You can get the total stock market for free, so why bother picking individual stocks? If it is a marketing pitch for someone to create a portfolio of individual stocks then they have told you all you need to know. You don't want their services.
We don't know how to beat the market on a risk-adjusted basis, and we don't know anyone that does know either | --Swedroe | We assume that markets are efficient, that prices are right | --Fama

limeyx
Posts: 280
Joined: Wed Sep 07, 2016 5:34 pm

Re: Diversification beyond 80 Stocks is not necessary?

Post by limeyx » Wed Jan 15, 2020 2:48 pm

shellycoat wrote:
Tue Jan 14, 2020 7:25 pm
Long-time reader (~3 Years)...First-time poster here!

I will jump right into my question. I know! The subject line sounds like "clickbait".

Is the following true?
A financial advisor told me that you do not need more than 80 individual stock for a diverse (equity) portfolio.

Notes:
I was very surprised when hearing that statement. Over the past 5 years, I became very interested in ETFs and really enjoy the low cost and broad diversification!

I was always under the assumption: the more stocks in an ETF the better the diversification. My current portfolio:

70/30 Equities/Bonds
Equity Mix:
70% Vanguard Total Stock Market (IE: VTI)
30% VANGUARD TOTAL INTL STOCK (IE: VXUS)
Bonds:
80% Vanguard Total Bond Market (IE: VBTLX)
20% Vanguard Total International Bond Index (IE: VTABX)

I appreciate the help!!!
Isn't the problem though in picking the "right 80" and making sure that as soon as the "right 80" changes that you know immediately and can adjust to the new "right 80" ?

Seems like that's a tough problem to solve

beehivehave
Posts: 225
Joined: Thu Aug 08, 2019 1:21 pm

Re: Diversification beyond 80 Stocks is not necessary?

Post by beehivehave » Wed Jan 15, 2020 2:58 pm

Nonsense. Vanguard Total Stock has 3550 stock holdings. Why would they choose to complicate it if 80 was enough?
But who cares if it's 80 or 8000? Do you really need the hassle and cost of constantly tracking and rebalancing that many stocks?

randomguy
Posts: 8534
Joined: Wed Sep 17, 2014 9:00 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by randomguy » Wed Jan 15, 2020 4:37 pm

shess wrote:
Wed Jan 15, 2020 12:24 pm
I don't have the background to adjust for it, but this feels like it overstates the possibility. If you had made your investment earlier than 2004, you'd have zero chance of picking Google or Facebook. Before 1997 you'd have zero chance of picking Amazon. That's three of the top ten components by market cap.

Sure, you could have a system which adapts over time to account for new entries, but in that case, you aren't dealing with "only 80 stocks" in any meaningful fashion.
I don't think there is a constraint that you need to hold the same 80 stocks for the whole time period. You would expect to lose stocks occasionally to mergers. And you would probably do what the S&P 500/DJIA do and replace stocks occasionally. That is why you are paying the advisor the big bucks:) It also makes it hard to back test ideas like this. It would be doable to see what the return of holding say the 80 biggest S&P 500 stocks would have been historically (in real life you probably would want some type of filter to prevent churning of the bottom 20 stocks. Maybe something like needing a stock to be in the top 80 for 2 or 3 years to make a swap). Or if you are doing some type of sector balancing (maybe you hold the 7 biggest stocks in each of the 11 sectors and the next 3 biggest companies). I expect you would find that you would be getting a pretty good correlation with the S&P 500.

From 1980 to today, the DJIA (i.e. 30 stocks weight by stock price) has returned 9% real. The S&P 500 did 8%. Now the DJIA is far from the best index:), but I have a feeling if you created one that held say the 5 largest stocks in each sector with market cap weighting, you would have done fine over just about all the 10+ year periods. You will have some deviance from the returns of a total market fund but they aren't going to be huge.

Scooter57
Posts: 1169
Joined: Thu Jan 24, 2013 9:20 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by Scooter57 » Wed Jan 15, 2020 9:01 pm

beehivehave wrote:
Wed Jan 15, 2020 2:58 pm
Nonsense. Vanguard Total Stock has 3550 stock holdings. Why would they choose to complicate it if 80 was enough?
But who cares if it's 80 or 8000? Do you really need the hassle and cost of constantly tracking and rebalancing that many stocks?
The top 50 stocks in the total stock market index absorb most of your invested dollar. By the time you get to the 1000th stock in the index, listed by market cap size you are investing a fraction of a mill from each dollar in that stock. By the 2000th it is far less. That stock could be a ten bagger and you would only gain a mill on your invested dollar.

The concept behind Total Stock Market is a marketing gimmick that takes in a lot of investors.

langlands
Posts: 100
Joined: Wed Apr 03, 2019 10:05 pm

Re: Diversification beyond 80 Stocks is not necessary?

Post by langlands » Wed Jan 15, 2020 9:15 pm

I'm with randomguy on this one. The "superstock" argument is flawed. I think finding 80 stocks that roughly match the S&P is doable and this kind of exercise is par for the course for many quants at banks attempting to do portfolio replication. As randomguy mentioned, the Dow Jones in fact with only 30 stocks does a decent job. It's certainly nowhere close to ideal though and of course has all sorts of factor risk (size being the most obvious) that one would remove.

I personally read about the "superstock" idea about 3 months ago and remember being a bit befuddled initially. If it's really true that missing out on all of these "superstocks" is devastating, then individual stocks do seem really risky. Here's the problem I see with it. While it may be true that the top 4% of stocks is responsible for all the gains of the stock market, that really isn't as insane as it first sounds. As a simple thought experiment, imagine that 45% of stocks return -10%, 5% of stocks return 0%, and 50% of stocks return 10%. Notice that 5% of stocks here are responsible for the entire return of the stock market. And yet half the stocks got the max return. As long as roughly half your portfolio was in the top 50% of stocks (which is pretty likely with 80 stocks), you'd have matched the market.

Put another way, the 96-100 percentile of stocks account for all the return in the stock market. This does not contradict the statement that for instance the 86-96 percentile of stocks also account for all the return.

shess
Posts: 368
Joined: Wed May 17, 2017 12:02 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by shess » Wed Jan 15, 2020 9:20 pm

OK ... I think I'm leaning towards the notion that the cap weighting probably does mean that it doesn't matter all that much that you missed out on the best performers, simply because you wouldn't have that much allocated towards those performers. A rising tide raises all boats, and all.

BUT, it also feels like this is a question someone could just answer with data and modest analysis. Over a specific timeframe, you should be able to take your initial investment and calculate a ranked list of components which contributed to your returns.

langlands
Posts: 100
Joined: Wed Apr 03, 2019 10:05 pm

Re: Diversification beyond 80 Stocks is not necessary?

Post by langlands » Wed Jan 15, 2020 9:30 pm

Scooter57 wrote:
Wed Jan 15, 2020 9:01 pm
beehivehave wrote:
Wed Jan 15, 2020 2:58 pm
Nonsense. Vanguard Total Stock has 3550 stock holdings. Why would they choose to complicate it if 80 was enough?
But who cares if it's 80 or 8000? Do you really need the hassle and cost of constantly tracking and rebalancing that many stocks?
The top 50 stocks in the total stock market index absorb most of your invested dollar. By the time you get to the 1000th stock in the index, listed by market cap size you are investing a fraction of a mill from each dollar in that stock. By the 2000th it is far less. That stock could be a ten bagger and you would only gain a mill on your invested dollar.

The concept behind Total Stock Market is a marketing gimmick that takes in a lot of investors.
Completely agree. Actually, this is the easiest argument showing something is wrong with the idea that owning a small number of "lottery" stocks are absolutely essential for your portfolio to have good performance. If 80 stocks out of 3550 can't diversify, why can 500 stocks diversify? And yet VOO and VTI have negligible difference in performance and no one has issues owning VOO.

columbia
Posts: 2160
Joined: Tue Aug 27, 2013 5:30 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by columbia » Wed Jan 15, 2020 9:42 pm

Vanguard says “Hold my beer”:

38 stocks in Global ESG Select Stock Fund
https://investor.vanguard.com/mutual-fu ... file/VEIGX

columbia
Posts: 2160
Joined: Tue Aug 27, 2013 5:30 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by columbia » Wed Jan 15, 2020 9:51 pm

langlands wrote:
Wed Jan 15, 2020 9:30 pm
Scooter57 wrote:
Wed Jan 15, 2020 9:01 pm
beehivehave wrote:
Wed Jan 15, 2020 2:58 pm
Nonsense. Vanguard Total Stock has 3550 stock holdings. Why would they choose to complicate it if 80 was enough?
But who cares if it's 80 or 8000? Do you really need the hassle and cost of constantly tracking and rebalancing that many stocks?
The top 50 stocks in the total stock market index absorb most of your invested dollar. By the time you get to the 1000th stock in the index, listed by market cap size you are investing a fraction of a mill from each dollar in that stock. By the 2000th it is far less. That stock could be a ten bagger and you would only gain a mill on your invested dollar.

The concept behind Total Stock Market is a marketing gimmick that takes in a lot of investors.
Completely agree. Actually, this is the easiest argument showing something is wrong with the idea that owning a small number of "lottery" stocks are absolutely essential for your portfolio to have good performance. If 80 stocks out of 3550 can't diversify, why can 500 stocks diversify? And yet VOO and VTI have negligible difference in performance and no one has issues owning VOO.
The idea that someone *needs* 8200+ stocks (see VT) seems pretty ridiculous, but, hey, it’s not my money. It’s cheaper than all (?) of the direct alternatives, so I certainly get that.

usagi
Posts: 249
Joined: Wed Jun 05, 2019 1:08 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by usagi » Thu Jan 16, 2020 1:00 am

In general, I think what is going on is firms like Fido have seen the impact of low cost/zero cost funds and etfs and are aiming to give, at least, the appearance of one upping them. With zero fee trading costs, they can slip in a management fee and run a custom index, collect a fee and try and show a profit to the investor. How? The how is in the caveats. one it is only for taxable accounts and two it is via the magic of tax loss harvesting and tax management. I think Fido targets 1000 stocks (and allows you to exclude stocks you do not want (i.e. tobacco stocks, McDonalds, whatever).

The essence of it is they are not putting you into a index others are in, but in your own custom portfolio so the tax liability and ta advantages accrue to you. In Fido's case the fees are .20-.65% minimum of 200K.

I did not post this as a recommendation, but by way of explanation.

Cheers...

User avatar
CyclingDuo
Posts: 2833
Joined: Fri Jan 06, 2017 9:07 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by CyclingDuo » Thu Jan 16, 2020 8:50 am

beehivehave wrote:
Wed Jan 15, 2020 2:58 pm
Nonsense. Vanguard Total Stock has 3550 stock holdings. Why would they choose to complicate it if 80 was enough?
But who cares if it's 80 or 8000? Do you really need the hassle and cost of constantly tracking and rebalancing that many stocks?
This thread has actually turned into a rather good discussion.

These days, the commission free, no cost robo-advisors such as M1 Finance (to give one example) do all of that for you automatically. They reinvest the dividends and additional investments you make into the stocks that have dipped below their weighting automatically and do all of the paperwork for you. They also have a rebalance button you can click and voila - it's all rebalanced immediately back to your original weighting. I haven't had to use that rebalance button yet as the robo-advisor with additional contributions and reinvestment of dividends does such a good job of keeping the weighting balanced automatically.

It's a very easy way for an investor to passively manage a DIY Index made up of your own diverse basket of stocks. Or ETF's. I hold pies at M1 Finance. One pie is the Three Fund Portfolio with Vanguards VTI, VXUS, BND (70/30 allocation). One pie is Paul Merriman's Ultimate Buy & Hold Portfolio that uses 15 ETFs of which 10 are Vanguard and the other 5 are Wisdom Tree or SPDR ETFs (70/30 allocation). One pie is a diverse basket of stocks (started with 54, but Red Hat was purchased by IBM, so am down to 53 stocks and M1 took the Red Hat proceeds and distributed it among the other investments automatically). Dividends are all automatically reinvested. Once a week, the same amount is automatically invested into each pie from my bank. After the initial set up of it all, it has been completely hands off on my part. These are not my main investing accounts, but is a nice and easy way to invest money I make from a side gig on a continual basis and compare the performance of the three strategies (Three Fund Portfolio; Ultimate B&H; DIY Index Fund).

It's only been a short term duration of 18 months, but to compare to the benchmark of Total Stock Market and the S&P 500 thus far since the date I opened the account in June 2018...

DIY Index Fund: 36.35%
Three Fund Portfolio: 16.85%
Ultimate Buy & Hold Portfolio: 10.84%

VTI: 36.31%
S&P 500: 35.47%

Obviously, the 70/30 allocations should not be compared to the S&P 500 or Total Stock Market, but the DIY Index Fund I created is tracking right along with VTI and the S&P 500 to date. It is an interesting passively managed experiment for me, but correlates with the studies cited in the Wiki...

https://www.bogleheads.org/wiki/Passive ... ual_stocks
"Everywhere is within walking distance if you have the time." ~ Steven Wright

hoops777
Posts: 2765
Joined: Sun Apr 10, 2011 12:23 pm

Re: Diversification beyond 80 Stocks is not necessary?

Post by hoops777 » Thu Jan 16, 2020 4:28 pm

nisiprius wrote:
Wed Jan 15, 2020 7:47 am
Let's consider the S&P 500, and let's consider broadly the idea of choosing a sample of "80 stocks" = 16% of them. Now add the idea, from a 2017 paper by Bessembinder, that all of the return (above Treasury bills) comes from just 4% of the stocks in it. I'll call these the "superstocks."

It would be great if we could pick 80 stocks and get all 20 of the superstocks, but we are indexers and we don't think we can do that; we don't even want to. Our goal is to have an 80-stock sample of the S&P 500 that matches the market, and, accordingly, contains 4% of 80 = 3.2 superstocks.

It's not worth bothering to be precise so we'll just assume that as we pick 80 stocks at random, each pick has a probability of 4% of being one of the superstocks. This just becomes an example of binomial probabilities. So finding an online calculator I find that we want 3.2 superstocks, and:

The chances of getting two or less is 20.9%.
The chances of only getting one or less is 12.7%.
The chances of missing all twenty superstocks, getting none of them, is 3.8%.

I find that last one daunting. 3.8% is more than the probability of rolling a total of two on a pair of dice. Happens to me in Monopoly all the time. On our premises, picking 80 stocks from the S&P 500 has a far-from-black-swan chance of missing all of the good stocks.

That's enough for me to conclude that there is a meaningful difference between 80 stocks and "all of them."

If it were 1970 and there were no cheap, easy way to get "all of them" then, of course, you'd settle for doing the best you could.

It may be possible to overcome the statistics by picking stocks in some way--such as by intentionally matching sector composition or other characteristics between our sample and the total market, or by outright stock-picking. But on the raw statistics, if we accept the "just a few superstocks" idea, eighty isn't enough to do it automatically.
I keep seeing the argument about picking the super stocks but little mention of getting all of the dogs when you take the SP 500. I do not care either way but maybe we should also mention the worst performing stocks. I would imagine the SP 500 minus the 50 worst performing would be quite different.
K.I.S.S........so easy to say so difficult to do.

DB2
Posts: 532
Joined: Thu Jan 17, 2019 10:07 pm

Re: Diversification beyond 80 Stocks is not necessary?

Post by DB2 » Thu Jan 16, 2020 4:48 pm

Dow vs VTI from 2002 - 2019. 30 stocks vs 3550. 8.5 vs 8.51% CAGR.

https://www.portfoliovisualizer.com/bac ... ion2_2=100

User avatar
danielc
Posts: 944
Joined: Sun Dec 10, 2017 4:48 am
Location: Iowa, USA
Contact:

Re: Diversification beyond 80 Stocks is not necessary?

Post by danielc » Thu Jan 16, 2020 8:02 pm

langlands wrote:
Wed Jan 15, 2020 9:15 pm
I personally read about the "superstock" idea about 3 months ago and remember being a bit befuddled initially. If it's really true that missing out on all of these "superstocks" is devastating, then individual stocks do seem really risky. Here's the problem I see with it. While it may be true that the top 4% of stocks is responsible for all the gains of the stock market, that really isn't as insane as it first sounds. As a simple thought experiment, imagine that 45% of stocks return -10%, 5% of stocks return 0%, and 50% of stocks return 10%. Notice that 5% of stocks here are responsible for the entire return of the stock market. And yet half the stocks got the max return. As long as roughly half your portfolio was in the top 50% of stocks (which is pretty likely with 80 stocks), you'd have matched the market.
Stock returns roughly follow a log-normal distribution. Your example doesn't even remotely match that. The distribution of stock returns have a have positive skew; the point about 4% of stocks being responsible for the entire excess return of the market is just an illustration but it doesn't fully describe the distribution. The paper I linked to also says that 0.3% of stocks are responsible for 50% of the return of the market and 0.02% of stocks are responsible for 10% of the return of the market. With a random set of 80 stocks there is a high chance that you'll miss the top 0.3% of stocks.

langlands
Posts: 100
Joined: Wed Apr 03, 2019 10:05 pm

Re: Diversification beyond 80 Stocks is not necessary?

Post by langlands » Thu Jan 16, 2020 10:36 pm

danielc wrote:
Thu Jan 16, 2020 8:02 pm
langlands wrote:
Wed Jan 15, 2020 9:15 pm
I personally read about the "superstock" idea about 3 months ago and remember being a bit befuddled initially. If it's really true that missing out on all of these "superstocks" is devastating, then individual stocks do seem really risky. Here's the problem I see with it. While it may be true that the top 4% of stocks is responsible for all the gains of the stock market, that really isn't as insane as it first sounds. As a simple thought experiment, imagine that 45% of stocks return -10%, 5% of stocks return 0%, and 50% of stocks return 10%. Notice that 5% of stocks here are responsible for the entire return of the stock market. And yet half the stocks got the max return. As long as roughly half your portfolio was in the top 50% of stocks (which is pretty likely with 80 stocks), you'd have matched the market.
Stock returns roughly follow a log-normal distribution. Your example doesn't even remotely match that. The distribution of stock returns have a have positive skew; the point about 4% of stocks being responsible for the entire excess return of the market is just an illustration but it doesn't fully describe the distribution. The paper I linked to also says that 0.3% of stocks are responsible for 50% of the return of the market and 0.02% of stocks are responsible for 10% of the return of the market. With a random set of 80 stocks there is a high chance that you'll miss the top 0.3% of stocks.
Exactly, 4% of stocks being responsible for the entire return of the market doesn't fully describe the distribution. My toy example was meant to give explicitly a distribution in which not owning the top 4% of stocks isn't disastrous, even if they account for all the return. I agree it's not close to actual stock market returns, but I think it illustrates the basic point that when almost half the stocks have negative returns, you have to think carefully about what it means for 4% of stocks to explain the entire market performance.

Frankly, I really dislike the phrasing that "4% of stocks account for all the wealth creation in the market." I find it sensationalistic and very misleading, not befitting a serious academic paper. Superficially, this makes it sound like all the other stocks are worthless, but this is wrong. If you think of 100 people doing a task and you hear that 4 of them did all the work, you automatically assume the other 96 did nothing. This logic fails with stocks because a large fraction of them can have negative return and can even go bankrupt.

Again, the 96-100 percentile of stocks account for all the return in the stock market. This does not contradict the statement that for instance the 86-96 percentile of stocks also account for all the return. Ultimately, this is an empirical question about the precise distribution of stock market returns, but there is ample evidence (Dow Jones for instance) that with 30 stocks, you're getting fairly close to total stock market, and with 80 you'll probably do just fine in replicating. Put simply, even though stocks have positive skew, they aren't skewed massively enough so that 80 stocks can't replicate it.

YearTrader
Posts: 8
Joined: Thu Jan 02, 2020 6:46 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by YearTrader » Tue Jan 21, 2020 9:22 pm

Here's a simple simulation. Let's assume in a world that all the assets have covariance matrix like this:

Code: Select all

1.0 0.5 0.5
0.5 1.0 0.5
0.5 0.5 1.0
With equal amount of holdings, you can see that the total volatility of the portfolio decreases as the number of assets grow, due to diversification. But the marginal gain from each new asset diminishes eventually.
Image

Obviously the real world is much more complicated and I wouldn't simply rely on the magical number 80.

User avatar
tc101
Posts: 3257
Joined: Tue Feb 20, 2007 3:18 pm
Location: Atlanta - Retired in 2004 at age 54

Re: Diversification beyond 80 Stocks is not necessary?

Post by tc101 » Tue Jan 21, 2020 9:29 pm

From what I remember from my statistics and probability class many years ago, it is true that a random selection of 80 from any group of numbers has a high probability of matching the average of that group, so yes, 80 stocks from the S&P 500 has a high probability of matching the S&P 500. But why go to the trouble? Why not just buy the S&P 500 fund.
. | The most important thing you should know about me is that I am not an expert.

User avatar
willthrill81
Posts: 15190
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Diversification beyond 80 Stocks is not necessary?

Post by willthrill81 » Tue Jan 21, 2020 9:44 pm

tc101 wrote:
Tue Jan 21, 2020 9:29 pm
From what I remember from my statistics and probability class many years ago, it is true that a random selection of 80 from any group of numbers has a high probability of matching the average of that group, so yes, 80 stocks from the S&P 500 has a high probability of matching the S&P 500. But why go to the trouble? Why not just buy the S&P 500 fund.
That's my take as well. In the end, 80 stocks will probably come really close to matching the performance of the S&P 500 or TSM, but the effort isn't worth the handful of basis points (none at all if you buy FZROX) paid.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

YearTrader
Posts: 8
Joined: Thu Jan 02, 2020 6:46 am

Re: Diversification beyond 80 Stocks is not necessary?

Post by YearTrader » Tue Jan 21, 2020 10:02 pm

[/quote]
willthrill81 wrote:
Tue Jan 21, 2020 9:44 pm
tc101 wrote:
Tue Jan 21, 2020 9:29 pm
From what I remember from my statistics and probability class many years ago, it is true that a random selection of 80 from any group of numbers has a high probability of matching the average of that group, so yes, 80 stocks from the S&P 500 has a high probability of matching the S&P 500. But why go to the trouble? Why not just buy the S&P 500 fund.
That's my take as well. In the end, 80 stocks will probably come really close to matching the performance of the S&P 500 or TSM, but the effort isn't worth the handful of basis points (none at all if you buy FZROX) paid.
It also needs to be cap-weighted. Otherwise it'll bias towards small cap index I guess.

JBTX
Posts: 5768
Joined: Wed Jul 26, 2017 12:46 pm

Re: Diversification beyond 80 Stocks is not necessary?

Post by JBTX » Tue Jan 21, 2020 10:03 pm

I suspect that on the face of it, the 80 stock argument is true. The S&P 500 behaves almost identically to the TSM. It probably isn't a big leap that the top 80 drives most of the return and volatility of the 500.

Now if you are picking 80, will it be a static portfolio or a dynamic one? You would pick the top 80 of the S&p and you'd have some changes from year to year. If you pick a dynamic one, then you aren't likely to miss the "top performers", because the top performers will tend to start at the bottom of the 500 and make their way to the top. You will have the top performers in your "slice" on the way up.

Now if you pick 80 stocks, and held them come hell or high water, and never ever change then your outcomes could vary a lot more. Imagine picking 80 stocks in 1960 and holding them to now. How well that portfolio did would vary dramatically on which ones you picked.

From a practical perspective if an advisor is saying let him pick 80 stocks for you, run, away, as fast as you can. That is a completely pointless and potentially expensive (fees) strategy.

User avatar
willthrill81
Posts: 15190
Joined: Thu Jan 26, 2017 3:17 pm
Location: USA

Re: Diversification beyond 80 Stocks is not necessary?

Post by willthrill81 » Tue Jan 21, 2020 10:04 pm

YearTrader wrote:
Tue Jan 21, 2020 10:02 pm
willthrill81 wrote:
Tue Jan 21, 2020 9:44 pm
tc101 wrote:
Tue Jan 21, 2020 9:29 pm
From what I remember from my statistics and probability class many years ago, it is true that a random selection of 80 from any group of numbers has a high probability of matching the average of that group, so yes, 80 stocks from the S&P 500 has a high probability of matching the S&P 500. But why go to the trouble? Why not just buy the S&P 500 fund.
That's my take as well. In the end, 80 stocks will probably come really close to matching the performance of the S&P 500 or TSM, but the effort isn't worth the handful of basis points (none at all if you buy FZROX) paid.
It also needs to be cap-weighted. Otherwise it'll bias towards small cap index I guess.
I would call that more of a feature than a bug, but you're right that unless you cap-weight it, you'll get 'tracking error' vis a vis the TSM.
“It's a dangerous business, Frodo, going out your door. You step onto the road, and if you don't keep your feet, there's no knowing where you might be swept off to.” J.R.R. Tolkien,The Lord of the Rings

GrowthSeeker
Posts: 808
Joined: Tue May 15, 2018 10:14 pm

Re: Diversification beyond 80 Stocks is not necessary?

Post by GrowthSeeker » Tue Jan 21, 2020 10:34 pm

Step 1: present a fact
Step 2: misrepresent that fact as if it proves the mark’s current vendor a “false god”
Step 3: convince mark to buy my product

Fallacy: because there is some finite number (say 80, for example) beyond which further diversification provides a benefit which is smaller than some threshold, it then follows that there is no reason to own an ETF or mutual fund.

Reasons it’s a fallacy:
- other benefits besides diversity, eg low cost, greater simplicity
- there is more to diversity than just a calculated standard deviation
Just because you're paranoid doesn't mean they're NOT out to get you.

Post Reply