Best Number of Stocks for Optimum Diversification?

[Voodooking]

I have held varying amounts of stocks in my portfolio at various times, depending on how much spare money I had and how many good ideas I was able to generate. More recently I've been trying to identify the optimum number of stocks to hold, with the aim to sticking to this roughly depending on opportunities available at the time.

 

The information I have found seems to average out at about 30-ish stocks in a portfolio. I just wondered if anyone had any strong views on this in either direction.

 

Joel Greenblatt (in his excellent book, You Can Be A Stock Market Genius) explains that Statistics say that owning just two stocks eliminates 46 per cent of the nonmarket risk of owning just one stock. This type of risk is supposedly reduced by 72 per cent with a four stock portfolio, by 81 per cent with eight stocks, 93 per cent with 16 stocks, 96 per cent with 32 stocks, and 99 per cent with 500 stocks. Therefore, it seems to me that diversifying by buying up to 32 stocks gives the benefits and safety of diversification, but there is absolutely no point in buying more than 32...

 

Buffett and Munger; Ackman; Burry; Pabrai etc. seem to take this concept even further, saying that if you know what you're doing, you only really need about 3, 5, 10, 12 or 20 stocks, depending on which view you follow. However, I am not as skilled or experienced as them, so slightly more diversification seems to make sense. Also, I prefer equally weighted positions for this reason. I believe I can spot good opportunities, but I'm less confident in grading these as to their individual prospects.

 

Looking at the early Buffett Partnership letters, it seems like for the first few years at least, he invested in 15 to 20 "Generals" (Undervalued) and 10 to 15 "Workouts" (Event based) and an undefined number of "Controls" (Activist) positions. It seems reasonable to assume that WEB therefore had exposure to at least 25, and at most 40(?) securities at the one time? This would mean an average of 32.5 individual stocks were held by the Buffett Partnership in its early (most successful years).

 

I am tempted to buy 'up to' 32 stocks, depending on how many opportunities are available at the time, but stick to that as the upper limit. I'd be interested to hear what other people's views on this are and any justification for running smaller or larger portfolios.

[Hielko]

Why would you want to limit how many stocks you own? More stocks is better diversification, so if you find enough good idea's you should own as much as you can.

[Voodooking]

Because some people (WEB) hold the view that you are better putting more money into your number 1 idea rather than your 33rd idea...

[writser]

There is no single right answer to this question. First of all, your risk appetite depends on your age, expected future cashflows, your personality, your household, etc. Second of all, optimum allocation depends on the perceived upside and downside of your individual picks and how confident you are that your judgements are correct (and how confident you are that you are not overconfident). Example: imagine a cash box with some optionality that has already announced a distribution. A scenario with very high certainty, almost no downside. A 50% allocation might be justified here. However, if you allocate 50% of your portfolio to Tesla (high uncertainty, downside ~100% in a liquidation) I'd say you are suicidal.

 

I'd say in general it is better to err on the side of caution. Most of us are not as sharp as Warren and the market might be more efficient now than it was 50 years ago. Also, most of us tend to think we're smarter than we actually are (see the Valeant thread on this board amongst many others) leading to disaster if you concentrate. And finally, the risk of diworsifying is that you end up with market returns while the risk of concentrating is that you go broke. If you have ~20+ positions you allow for a few spectacular failures.

[DTEJD1997]

Hey all:

 

My Dad (with my help), has EASILY exceeded market returns for a long, long time now...he has a rather large portfolio, perhaps coming close to 100 different positions.

 

One of the things he has done is just buy & hold for a long time, sometimes for 20+ years.  He will sell a position, but it is relatively rare.  He usually just buys & holds on & on & on.

 

Sometimes it does not end well...but sometimes it ends up rather incredible...10x returns on a stock?  hahahahahaha, try 100x or more returns.  He has a few 100x returns, and several 40X-50X returns.  Of course, this has taken 10, 15, 20 years to achieve.

 

He bought Apple back in 1997.  Steve Jobs had just come back to the company....

 

He has held Seaboard since it was in the very low 100's....

 

He has NewMarket.....Southern Company, Checkpoint Software, National Beverage, and others that were bought 15-20 years ago.

 

The other interesting thing is the dividends he gets.  He has some positions where he gets back his initial investment with every year in dividends....a couple/few positions with every quarterly payment of dividends!  Talk about taking risk off the table!

 

He also has some incredibly obscure//hard to get positions.

 

So for him, a huge, incredibly diversified portfolio seems to work out very, very well.

[clutch]

And finally, the risk of diworsifying is that you end up with market returns while the risk of concentrating is that you go broke.

 

+1. It should be obvious to everyone the latter risk has much worse consequence than the former.

 

People should understand that Buffet is justifying his own decision to concentrate; To others he recommends ultimate diversification -- indexing.

[Guest longinvestor]

Because some people (WEB) hold the view that you are better putting more money into your number 1 idea rather than your 33rd idea...

+1. My reason learned from WEB. I cannot convince myself that this works any better than looking back at how I did make my gains; An overwhelming % of my gains have from just a handful of ideas. Like < 5. It has been hard enough to keep up with 5 names let alone 33.

 

Another thing that WEB/CM repeat like a parrot, most rich people in business got there by putting all their eggs into their one basket; Their apartment block of the corner store or the farm. Of course, the fact that they run it by themselves matters but if you are not the owner operator, that's the real reason why diversification has been a fool's game for me.

 

Find 33 people I can trust with my money? In hindsight, of the few I did make money with, there is only 1 that checks all of my boxes of trust. IMO, they don't exist in large numbers.

 

[oddballstocks]

If you take the concentration thought to it's final conclusion you should only own your best idea.  Why dilute your best idea with a second best idea?

 

I own a number of stocks, I've beat the market, other people have as well. 

 

I agree with the strawman about 1st vs 10th best idea IF you are sitting with a pile of cash and someone said "allocate this into an ideal portfolio right now."

 

The problem is that scenario never exists.  A portfolio is fluid.  Look at it this way.  Say you own 50 stocks and you have a 20% turnover, that's less than a stock sold and purchased per month.  Can you find one single good idea per month?  If you own 30 stocks and have the same turnover it's even less, it's one stock every two months.  Doesn't that encapsulate what WEB was saying?  In those two months you're putting all of your money to work in your best new idea.

 

Also look at what WEB does vs says.  His portfolio has a few dozen companies, he owns dozens more.  People argue it's because he has 'control' so that's alright.  I disagree, he knows the risk is blow up risk.

 

Do some modeling of returns.  You win by not losing.  If you merely match the market in up years, but avoid significant losses you'll outperform on a significant scale.  You don't need to hit home runs constantly.

 

There are people who come on here who talk about their 60% annualized returns with three stocks.  Search their posting history, they either have a mea culpa moment, or just silently disappear.  Moonshot returns are unsustainable.

 

[giofranchi]

I have not the time to study and evaluate complicated (but potentially very lucrative) investment ideas (I am not a full time money manager). Therefore, my process is straightforward: I invest on a constant basis (each month, a sort of dollar cost averaging process) in cos which have outperformed during the last 5 and 10 years (are probably doing something right), and which I believe could go on performing very well during the next 5 years. And my approach to diversification is the following: the more cos I find that qualify, the better.

 

Cheers,

 

Gio

[frommi]

Since nobody knows the exact future business development of a position it is very hard to know a priori which one of your ideas will really be your best idea.

Given that you have 100 roughly equal good ideas should you only invest in 5 or 10? Mathematically over the very long term the returns will be the same when you invest only in a subset of the 100 stocks, but the yearly volatility of returns will be a lot higher if you only invest in a subset instead of all 100.

 

The only reason to invest only in a subset of all ideas is if you are paid for more variability of the returns, like people managing OPM.

For example under 2+20:

Return series:                                        -5%, +50%, -5% = 35%, fees = 3*2%+10%=16%.

The same return with lower stddev: 10.5%,10.5%,10.5% = 35%, fees = 3*2%+3*2.1%=12.3%.

 

So higher std deviation leads to more fees.

 

For this group of people concentration makes sense, for everybody else more diversification is probably better. (At least if you belong to the unlucky part of the population)

[Jurgis]

There are people who come on here who talk about their 60% annualized returns with three stocks.  Search their posting history, they either have a mea culpa moment, or just silently disappear.  Moonshot returns are unsustainable.

 

To be fair, some of these people disappear upwards (i.e. open hedge fund, whatever and don't need CoBF or can't talk here anymore).

 

But, yeah, if you're one of the people who can do 30%+ for 5-10+ years on handful of ideas, you already know it. You won't be asking the question OP asks. If you are asking the question, you likely should diversify and/or own something like BRK which is like a diversified fund anyway.

 

 

I'm one of the people who cannot tell which one is the #1 idea, which one is #33. Sometimes #33 performs way better than #1 for me.

 

Not much else to add, I think people made some great points above.

 

 

[KCLarkin]

Why would you want to limit how many stocks you own? More stocks is better diversification, so if you find enough good idea's you should own as much as you can.

 

The amount of research time spent is fairly linear (unless you are running a quant or basket strategy). Your return on invested time drops with each stock added to your portfolio. Actually, for most investors, an optimal portfolio might be 80% ETFs and 20% in 2 or 3 stocks.

 

[Guest jeffswaldron]

I don't understand diversification. Why not just buy an index if you're worried about it?  I think of investing as buying a company that I wish I could own outright for a fair price. 

 

Do small private business owners get together and discuss how they can profit share so they are well diversified?

 

And before you pass judgement, I already know I am an ignorant fool.

[oddballstocks]

 

Do small private business owners get together and discuss how they can profit share so they are well diversified?

 

 

Yes, its called selling different products or services.  Business owners are always trying to diversify their product offering.

 

When was the last time you went to a hamburger place that only served burgers, just a single menu item, a burger on a bun?  They have fries, chicken sandwiches, onion rings, all sorts of options.  They're trying to capture as much of the market as possible with slight varieties.

 

I'd say it's exceedingly rare to meet a small business that just does one single thing.  Maybe the 15-yr old cutting grass.  But my guess is he'll trim your bushes too if you ask, or spread mulch. 

[Guest jeffswaldron]

 

Do small private business owners get together and discuss how they can profit share so they are well diversified?

 

 

Yes, its called selling different products or services.  Business owners are always trying to diversify their product offering.

 

 

I apologize if I had a negative tone.  I am seeking wisdom and correction of my ideas and in no way think I am correct.  Your response shuts me up for the most part.  My follow up is if a business is well diversified, and a stock is part ownership of a business, what sense does it make to calculate the optimal number of businesses to own. 

 

[Hielko]

Hey all:

 

My Dad (with my help), has EASILY exceeded market returns for a long, long time now...he has a rather large portfolio, perhaps coming close to 100 different positions.

 

One of the things he has done is just buy & hold for a long time, sometimes for 20+ years.  He will sell a position, but it is relatively rare.  He usually just buys & holds on & on & on.

 

Sometimes it does not end well...but sometimes it ends up rather incredible...10x returns on a stock?  hahahahahaha, try 100x or more returns.  He has a few 100x returns, and several 40X-50X returns.  Of course, this has taken 10, 15, 20 years to achieve.

 

He bought Apple back in 1997.  Steve Jobs had just come back to the company....

 

He has held Seaboard since it was in the very low 100's....

 

He has NewMarket.....Southern Company, Checkpoint Software, National Beverage, and others that were bought 15-20 years ago.

 

The other interesting thing is the dividends he gets.  He has some positions where he gets back his initial investment with every year in dividends....a couple/few positions with every quarterly payment of dividends!  Talk about taking risk off the table!

 

He also has some incredibly obscure//hard to get positions.

 

So for him, a huge, incredibly diversified portfolio seems to work out very, very well.

If he almost never sells and some positions are up 100x I can't see how he can have an incredibly diversified portfolio. Those few positions with insane returns will start to dominate the portfolio.

[DTEJD1997]

Hey all:

 

My Dad (with my help), has EASILY exceeded market returns for a long, long time now...he has a rather large portfolio, perhaps coming close to 100 different positions.

 

One of the things he has done is just buy & hold for a long time, sometimes for 20+ years.  He will sell a position, but it is relatively rare.  He usually just buys & holds on & on & on.

 

Sometimes it does not end well...but sometimes it ends up rather incredible...10x returns on a stock?  hahahahahaha, try 100x or more returns.  He has a few 100x returns, and several 40X-50X returns.  Of course, this has taken 10, 15, 20 years to achieve.

 

He bought Apple back in 1997.  Steve Jobs had just come back to the company....

 

He has held Seaboard since it was in the very low 100's....

 

He has NewMarket.....Southern Company, Checkpoint Software, National Beverage, and others that were bought 15-20 years ago.

 

The other interesting thing is the dividends he gets.  He has some positions where he gets back his initial investment with every year in dividends....a couple/few positions with every quarterly payment of dividends!  Talk about taking risk off the table!

 

He also has some incredibly obscure//hard to get positions.

 

So for him, a huge, incredibly diversified portfolio seems to work out very, very well.

If he almost never sells and some positions are up 100x I can't see how he can have an incredibly diversified portfolio. Those few positions with insane returns will start to dominate the portfolio.

Yes, some of the very large positions are starting to dominate the portfolio, such as Apple is probably over 10% now.  HOWEVER, he still has what I would call a very diversified portfolio.  His top 10 positions are maybe 1/3 of the portfolio?  He is also sitting on a rather large amount of cash, both absolute and percentage.  As the years have gone by he has built up cash and slowed down in terms of taking new positions.  I suspect that in the next big downturn he will start buying stuff again.

 

Some of the big positions (and huge gainers) are: AAPL, ADC, ANDE, ELXS, NEU, FIZZ, SEB, SCCO, MHGU, MO, CLWY, IBA, GSH. I am sure I'm overlooking a few...but he has done well on those.

[DooDiligence]

If you take the concentration thought to it's final conclusion you should only own your best idea.  Why dilute your best idea with a second best idea?

 

I own a number of stocks, I've beat the market, other people have as well. 

 

I agree with the strawman about 1st vs 10th best idea IF you are sitting with a pile of cash and someone said "allocate this into an ideal portfolio right now."

 

The problem is that scenario never exists.  A portfolio is fluid.  Look at it this way.  Say you own 50 stocks and you have a 20% turnover, that's less than a stock sold and purchased per month.  Can you find one single good idea per month?  If you own 30 stocks and have the same turnover it's even less, it's one stock every two months.  Doesn't that encapsulate what WEB was saying?  In those two months you're putting all of your money to work in your best new idea.

 

Also look at what WEB does vs says.  His portfolio has a few dozen companies, he owns dozens more.  People argue it's because he has 'control' so that's alright.  I disagree, he knows the risk is blow up risk.

 

Do some modeling of returns.  You win by not losing.  If you merely match the market in up years, but avoid significant losses you'll outperform on a significant scale.  You don't need to hit home runs constantly.

 

There are people who come on here who talk about their 60% annualized returns with three stocks.  Search their posting history, they either have a mea culpa moment, or just silently disappear.  Moonshot returns are unsustainable.

 

+1 "the Most Important Thing" (which I appear to have ignored in a few cases...)

[porcupine]

Can you truly understand and keep up with the 30 companies you own? Do you have the time to do this? These are important questions to ask yourself.

[Jurgis]

Can you truly understand and keep up with the 30 companies you own?

 

Can you truly understand 3 companies you own?

[porcupine]

Can you truly understand and keep up with the 30 companies you own?

 

Can you truly understand 3 companies you own?

 

I'd certainly hope so. 

[Jurgis]

Can you truly understand and keep up with the 30 companies you own?

 

Can you truly understand 3 companies you own?

 

I'd certainly hope so.

 

I know nothing about you, so don't take this personally. Maybe you do understand them.

 

Most people who think they understand the 3 companies they own, understand them at best very superficially.

 

I'd say that some CEOs that run a single company don't understand it truly.

For outside investor to truly understand multiple companies without being inside of them is IMO rather exceptional.

 

FWIW.

 

Good luck

 

8)

[porcupine]

Can you truly understand and keep up with the 30 companies you own?

 

Can you truly understand 3 companies you own?

 

I'd certainly hope so.

 

I know nothing about you, so don't take this personally. Maybe you do understand them.

 

Most people who think they understand the 3 companies they own, understand them at best very superficially.

 

I'd say that some CEOs that run a single company don't understand it truly.

For outside investor to truly understand multiple companies without being inside of them is IMO rather exceptional.

 

FWIW.

 

Good luck

 

8)

 

No offense taken.  I actually agree with you.  What I meant was that if you have the time to keep up with every company you own, then go ahead and own as many as you can keep up with... as long as you have enough information and the time to process and analyze that information to make a prudent decision... go for it. 

[Guest longinvestor]

Can you truly understand and keep up with the 30 companies you own?

 

Can you truly understand 3 companies you own?

 

I'd certainly hope so.

I'm with you. Only would replace certainly with an added adjective such as fairly or almost.

 

I believe that extraordinary trust is needed with both extremes. An investment in an index fund implies trust in the country's economic environment while the ultimate concentration in just one stock is trust in the steward.

 

More than your ability to understand, the willingness of the steward to want you to understand is key.

 

Look about you, you'll spot Waldo.

 

If the accounting is hard to understand, they don't want you to. Buffett.

[racemize]

Just to throw out my own data points on this, and it should be noted that I'm talking about 7 years of investing experience and not the 10+ of many here:

 

Generally, my highest conviction ideas do generally work out better than the lower ones, which is usually a function of price.  I'll also say that that does not mean that every high conviction idea works or that the #1 high conviction idea is better than 2-5.  However, there is a noticeable drop-off between high conviction and not high conviction.

 

Since 2012 I have used my IRA as the high-concentration/high-conviction portion of my investments.  And I am more aggressive on using options in the IRA (usually after my common position in my 'normal' investing vehicle have dropped significantly), but it doesn't always have leverage. 

 

The results are so far:

full-portfolio: ~20% a year

IRA: ~30% a year (miraculously never negative)

(Caveat, given the starting time period of 2010, these are unsustainable numbers in my view, but they are what they are at this moment)

 

So, either this is working out, or I'm going to blow up that IRA.  We'll see.

 

Anyway, I'm happy with 8-15 positions (probably holding cash if I'm down to 8 ) in my main portfolio.  The IRA has 3-6 usually.  To each their own.