Oildog, How I so disagree with your comment..  from another post, copied here. I think your portfolio has far too many funds.  When people talk about "not putting your eggs in one basket," it is in reference to securities, not mutual funds.  Generally, mutual funds are already widely diversified across a large number of securities, so there is no need to hold a large number of funds (unless you are allocating to specialty funds like FSDAX).  If anything, holding a large number of active funds makes it more likely that your portfolio will perform like a market index fund, but with higher fees.  I would look at the size of the portfolio, then choose the number of funds.  when you have critical mass, then the fund count should grow. I don't hold any more than 10 % in a single fund. I make sure the overlap, via xray stocks is minimal.   Owning multiple funds in the same classification, with no overlap is better than owning 30% in a given fund.  Ask Magellon owners.  As the portfolio grows, the number of holdings I own grows. If you are in the better funds, top 20%, then you'll do good.  Make sure you are in funds that don't correlate, you'll come out ahead with less risk, another fido stat.  So an approach with limited $ is to own diversified funds, then as it grows use re-balancing to obtain access to targeted funds.   my $0.02 + interest

Hi kerryvan, you might want to edit the title to something more germane like "number of funds in portfolio" rather than "RE: oildog."

I would look at the size of the portfolio, then choose the number of funds.  when you have critical mass, then the fund count should grow.

It depends.  I would agree with you in some limited circumstances, e.g. if you're only investing in focused funds and you need a minimum asset level to buy more than a few funds.   

In general, however, the key is to assess diversification at the level of your securities holdings, not your fund holdings.  For example, somebody might allocate 100% of their assets to a Target Retirement fund, e.g. TRRBX.  That's fine - you're diversified across thousands of stocks and bonds in all kinds of sectors, countries, and capitalizations.  You might want to add specific categories like small-caps or international exposure depending on your tastes, but your overall portfolio is well-diversified with a single fund.

On the other hand, I would not recommend a 100% allocation to a focused fund like CGMFX, OAKLX, CFIMX, etc.  An investment in any of these funds would leave your portfolio exposed to only about 20-30 stocks.  It probably makes sense to own at least several funds if that's your strategy.  That said, even if you own 10 focused funds, you aren't going to have a portfolio as diversified as a target retirement fund.  So comparatively, you're still pursuing a high-risk, potentially high-return strategy. 

In both cases, the important point is what your underlying securities profile looks like, not how many funds you have in your portfolio. 

Best,
Oildog 

 

We agree in part.

What I don't like to see as the portfolio grows, is being tied to a single, or few funds for performance.  Also, being in a 'well rounded' fund may limit the small caps, intl or other sectors that often have better performance.

Also, as the portfolio size grows to a critical mass,  swings can be tolerated without panic or jumping ship. This is assuming the fund is a sound investment.  For example, china last march. 

Or assuming that you have one targeted fund, therefore you should do well.  I have three china funds that have very little overlap in stocks, they don't have the same peaks/ drops.  I think this is better than putting all the money in a single china fund.  Hence understanding that investing in china is a risk, I'll put the eggs in three baskets.

 Bottom line is for all investors not to be a buy and forget!

I think I might have to disagree in part with Oil on this one because a target date fund, as a result of the mix of different styles, asset classes, etc. IS diversified, as you said.  A fund can have 150-200 holding (like a Magellan mentioned above) but because the stocks that are being selected fall into a similar criteria it can do quite poorly for a very long period of time.  In fact, if it is a flawed strategy it may never produce good returns.  So, even with hundreds of stocks in a particular portfolio, if the stocks are similar based on a discipline the manager implements, it might be considered diversified by "number" of stock holdings but does not produce the kind of diversification that most investors look for.  And what I mean by that, is owning securities that act differently in various market cycles.  Usually, owning several different funds (of course with different disciplines, asset classes, etc.) can create that kind of diversification.  I do agree however that many investors own way too many funds.  I will state for the record that I do have a huge bias in this case.  I dislike indexing and I have found it VERY difficult to find fairly young proven managers or teams that run funds that can meaningfully out perform over a LONG period of time.  With many proven managers being older it is tought to find someone you can stick with for a LONG period of time.  Many others don't pay attention to taxes and finally some of the best performance comes from funds that are not going to repeat performance because of asset bloat.  With that said, I have almost 80% of my invested assets with BRAGX and BRAIX.  51 yr old manager.  I would rather rely on a computer that a person (although the person must maintain and tweak the model to keep it a sound strategy) and he pays a lot of attention to taxes.  Yes, that is concentration but based on all of the issues I mentioned above, I find it very hard to build a portfolio spread out among several different funds that can outperform over a long period of time with tax efficiency (or at least tax aware).

80 % in two funds that have a 5 yr track record of 16-19 %..  not bad,

I'd rather have 10% in each of these and put some in funds like eurox, letrx, prlax that have demonstrated 40% over the same 5 yrs.

you'd have doubled your portfolio, rule 72.. in approximatley 4 yrs.

I would have doubled mine in 2 yrs, then again in 2 yrs, so at the end of 5 yrs I'd have more than tripled your's   with less sensitivity to the US market.  for example 2006 you'd be hurting. with a minor set back,  mine would have almost doubled in one yr for one fund.

I'll stick with a winner, yours gives me concern.

I'm looking at three simple portfolios published in Kiplinger's in May 2006. The long-term portfolio (5 funds, 10 years to retirement) allocates 25% to Bridgeway Aggressive Investors 2 (BRAIX); the mid-term (5 funds, 5 to 10 years) allocates 15%; and the short-term (8 funds, retired) only 10%. As retirement approaches, the allocation to BRAIX decreases as diversification increases. Asset classes perform differently at different points in the business cycle: Small-cap, mid-cap, large-cap, international stocks, even bonds all have their moments (sometimes years) in the sun. Too few funds can limit your total return, just as too many funds can flatten returns. As for the "ideal" number of stocks a mutual fund should own, I get a kick out this line from an interview with Acorn Fund's Ralph Wanger: "At any given time, I typically owned 300 stocks. Ten might soar, and that made all the difference."

BRAGX  nearly 20% since 1994   Outperformed in up market...outperformed in down market....did it tax efficiently and the guy is still 51.....Don't ever talk4 yr time period with me at least....for that matter talking less than 10 with me is a losing battle.

I disagree with your investment philosophy, but this is one where you really have to figure out what you're comfortable with.  There's not necessarily a correct answer.

My approach is similar to the one advocated by Warren Buffett.  To loosely summarize, Buffett's take on diversification is this: if you're clueless, pursue diversification; if you know what you're doing, pursue concentration. 

What I try to do is to thoroughly understand what I own.  Most investors just look at past performance and try to ride winners.  I read through all available reports, I dig through media archives, I study the strategy extensively, I look up historical material from various sources, etc. etc.  I also follow the academic literature on fund performance and I think I have a reasonable familiarity with how the fund industry works and what kind of incentives managers and fund management firms face. 

So I will choose a fund only if I have a very high degree of confidence in the choice after all factors are considered.  I invest with the intention to stay put indefinitely unless something materially changes such as the management, expenses, strategy, asset base, etc.  With this strategy, it doesn't make sense to own a large number of funds - I'm already investing in the fund(s) with the highest expected long-term returns, so I'm only diluting my expected returns if I add new funds.

There are several reasons why I don't recommend this strategy to others: 1. Most investors aren't willing to do this kind of "homework"; 2. Most investors are obsessively concerned with past performance; 3. Most investors do not have a basic understanding of probability and statistics; 4. Most investors let emotions dictate their investment decisions.  For these reasons, I don't think this would work for most investors. 

However, my point about pursuing diversification at the securities level rather than the fund level remains valid and applies to all investors. 

Best,
Oildog 

I agree with Oildog totally here... if you want to try active management your only chance for success is to study up, spread your money across a couple to several of the best managers you can find and ride it out... if you have a portfolio of 20 plus actively managed funds, if you think you are beating the indexes after taxes and transaction fees over a long period of time I think you are probably mistaken... after all, they call it an average because... its average!  The losers cancel out the winners leaving you with only the entertainment value of watching your portfolio over the years and a big tax bill... which might be worth it as long as you understand thats what you are doing... the only way you could possibly win that game is to select every single outperformer in advance... and the chances of that go down with each fund you add...

You dont really need more than one fund per asset class as the majority of funds are diversified enough to avoid stock specific risk on their own... you might have more than one if your funds are extremely focused or volatile or you might only have one fund if you can get into SGIIX and feel comfortable enough with the succession question... but certainly not more than two funds per asset class or you are giving up before you start...

When in doubt, do a M* Xray of your portfolio and look at the contribution of each stock from a percentage standpoint... if you have an endless list of stocks that contribute .005% to your returns, imagine how silly you are to pay that manager for his time picking that stock?

Ajw 

 

I have a question though as it pertains to this statement:

"so I'm only diluting my expected returns if I add new funds"

You are assuming that all other funds are likely to underperform the ones that you own, right?  If an investor owns 10 funds and there is an 11th fund out there that reasonably has the ability to perform better than the first 10 over some period of time, than by adding an 11th fund you are not diluting performance.  That would hold the same for individual stocks.  The question the investor needs to ask is: "at what point" does that additional fund no longer has the ability to equal or add to the performance and if it is not likely to equal or add to performance is it (because it better) going to lower the risk profile of the overall portfolio because of correlation.

[quote user="ajwells"]

I agree with Oildog totally here... if you want to try active management your only chance for success is to study up, spread your money across a couple to several of the best managers you can find and ride it out... if you have a portfolio of 20 plus actively managed funds, if you think you are beating the indexes after taxes and transaction fees over a long period of time I think you are probably mistaken... after all, they call it an average because... its average!  The losers cancel out the winners leaving you with only the entertainment value of watching your portfolio over the years and a big tax bill... which might be worth it as long as you understand thats what you are doing... the only way you could possibly win that game is to select every single outperformer in advance... and the chances of that go down with each fund you add...

You dont really need more than one fund per asset class as the majority of funds are diversified enough to avoid stock specific risk on their own... you might have more than one if your funds are extremely focused or volatile or you might only have one fund if you can get into SGIIX and feel comfortable enough with the succession question... but certainly not more than two funds per asset class or you are giving up before you start...

When in doubt, do a M* Xray of your portfolio and look at the contribution of each stock from a percentage standpoint... if you have an endless list of stocks that contribute .005% to your returns, imagine how silly you are to pay that manager for his time picking that stock?

Ajw 

[/quote]

This only is true if you are primarily investing in diversified funds.

Modern Portfolio Theory would argue that adding funds or stocks does not dilute the ability to outperform (I understand that not everyone agrees with MPT: in this case I do).  Just look at the S&P 500.  A diversified market cap weighted AVERAGE.   If you were to own the S&P 500 Index (market cap weighted or equal weighted) and then add to it the Russell Mid Cap Index, and MSCI Small Cap Index, you would have added 2 more funds/indexes and over 1000 stocks to the portfolio.  Did you increase or decrease your chance of outperforming?  Answer: Depends on if small and mid cap stocks outperform doing forward but by adding funds/indexes you did not necessarily decrease your ability to outperform.  In this case, over the long term you are actually more likely to outperform if history is a guide (as it relates to small/mid over large cap).  The same holds for managers.  The question simply becomes: at what point to you run out of a selection of managers that are able to outperform?  Are there only 5 managers out there?  10? 20? 30?  Who knows, but if you have 10 funds in your portfolio and there are 20 out there, then by adding an 11th you are are not necessarily diluting your returns.

 

 

Modern Portfolio Theory would tell you to avoid actively managed funds and hold a passive selection of non-correlated assets... I was of course talking about assembling a portfolio of actively managed funds in an attempt to exceed the returns of the overall market... in that case each fund and manager you add decreases your chances of deriving the most contribution from the managers who succeed... and can increase the volatility of your portfolio if all of your managers overweight a particular sector or industry and are wrong over an extended period

And in response to the other poster, my comment on the .005% contribution of a particular stock is emphasized by the holding of many diversified funds, but it can also exist in the portfolio of an investor who holds a big collection of less-diversified funds 

Ajw 

Warren Buffett's oft-quoted line to the effect that "if you're clueless, pursue diversification; if you know what you're doing, pursue concentration" stands in notable contrast with the large collection of diverse companies (everything from carpets and candy to jewelry and insurance) he has assembled under the umbrella of Berkshire Hathaway. Likewise, his stock selections run from financials (American Express and Wells Fargo) to food and beverage (Coca-Cola and Kraft) to railroads (Burlington), industrials (Ingersoll-Rand), and utilities. The common denominator is that he buys quality assets he understands at a discount, then patiently waits for their value to grow. Typically, he holds them indefinitely, but sometimes he takes profits (PetroChina) or quietly liquidates his position (Ameriprise).

the way buffett and soros invest is by buying the market space, betting against something and in general doing this that someone without their $$ can not play.  The can make move and a number of sheep will follow..  I'm sure it is nice, but not very many people can play that game.  Not a good analogy.

Two points:

1) I do not agree with all of MPT.  This just happens to be one of the points I agree with.

2) "I was of course talking about assembling a portfolio of actively managed funds in an attempt to exceed the returns of the overall market... in that case each fund and manager you add decreases your chances of deriving the most contribution from the managers who succeed..."

Sorry, but that is not true....the assumption is that you picked the "best performing" managers in existence.  That is a huge assumption.  It is highly unlikely that you happen to have picked the 10 (let's just use that number as an example) "best" funds out there and that the 11th will be a worse performer than the first 10.....There is a very good argument that would suggest that there is an 11th fund out there that would improve the performance of the first 10 because it is highly unlikely that you own the BEST 10 in the first place.  This is one point where I happen to agree with MPT.  I happen to only own 3 funds because there aren't many out there that fit my critieria.  So, with that said I am not suggesting (I don't want anyone to misunderstand my point) that investors own lots and lots of funds.  Investing is more than MPT.  A lot of it comes down to making educated guesses.  I made mine about certain managers and I suggest everyone make their own educated guesses as well.

Okay, so I think this discussion is getting a little off track.  In any given asset class I'd own 1-3 funds. LC, MC, SC.  (total 6-9 funds depending on asset base)

Then I would go after intl, broad based diversification LC, MC/SC,  then targeted fund for western europe, asia, latin am, china, india, brazil, then eastern europe, mid east... russia, ( additional 9-15 funds)  There is no way a fund manager that manages diversified intl will pick the best opportunities in a country market, they will swing to the hot companies.

then lastly, I'd go after sector plays,  definitely not buy and hold.  energy, basic materials..

For those of you that are so confident in US market, great!  Being overseas for the past two yrs, my protfolio has grown nicely. Funds gaining 20++ % and currency dropping 10-20%, very nice. 

Being confident that the manager of the 3-5 funds you own can make the best global choices on stock buy, wonderful... I'm not there..

78% markets outside the US,  if you want to use 60% of your money to chase 22% of the market,  go ahead.    while doing some research on funds this morning, I saw an interesting stat from a major US fund manager.  His ave was in the 20% range over time, he holds about 300 stocks at a time, it is 7-15 stocks that make the pop for the yr.  Interesting,  his average 'good' pick is 1-2%..

So by going with other funds in the same style/ market space, if the overlap is low, then the maybe you have a better chance of not having the dogs bring you down.

Do I feel safer when all the major markets are down and I still have funds that made money that day, you betcha!

 

With that said, I have almost 80% of my invested assets with BRAGX and BRAIX.  51 yr old manager.  I would rather rely on a computer that a person (although the person must maintain and tweak the model to keep it a sound strategy) and he pays a lot of attention to taxes. 

Hi valu,

May I ask what you own for international equity exposure?  I am trying to simplify my portfolio and for my taxable account am wondering if I should just pair BRAIX with VEU (and that's all!).  I am satisfied with Bridgeway solely covering me domestically, but have yet to find an equivalent for the international side.  Like I said, I may just settle for VEU.

At 80%, you don't believe the world economy will decouple from the US's?
     

 

You are assuming that all other funds are likely to underperform the ones that you own, right?  If an investor owns 10 funds and there is an 11th fund out there that reasonably has the ability to perform better than the first 10 over some period of time, than by adding an 11th fund you are not diluting performance.

Yes, but several things: 1. your initial search should have uncovered this 11th fund if it was adequately thorough; 2.  if something has changed that makes this 11th fund superior to your existing 10 funds, it might be worth considering whether one of the 10 funds should be dropped in favor of the 11th fund. 

The question the investor needs to ask is: "at what point" does that additional fund no longer has the ability to equal or add to the performance and if it is not likely to equal or add to performance is it (because it better) going to lower the risk profile of the overall portfolio because of correlation.

I don't have a strong preference regarding portfolio volatility as the probability that I will need to withdraw significant money from my investments in the short-term is effectively zero.  This is completely dependent on your individual circumstances, however, so I can sympathize with the point. 

 

Best,
Oildog

I do not own an international manager.  I am not recommending that others follow my lead on this point but I will explain why I have decided to invest in such a way (for myself).

With that said if you were set on owning international I would look at funds like JETIX/JETAX and ANJIX.  As you will see below, there are reasons I will not own either.....owned by big public banks/insurance companies, etc but they are the "type" of international fund I would own if I was going to.  They are unique, proven disciplines, either team managed or by a young manager and they are truly different than the index. 

1)  I have not been able to find a manager outside of the US that fits the criteria that I look for as BRAGX/BRAIX does on the domestic side.

        - young manager or team approach, private firm, proven discipline, no asset bloat, not index hugger/closet indexer, tax aware


2) I do not like dealing with Foreign Taxes as it further hurts long term after tax returns and is an April headache

3) International Indexes are flawed because of country weightings that make for easy active management outperformance.  EAFE is one of the easiest benchmarks to beat.  That is statistically proven.  FTSE All World, etc are all very similar...

4)  Most international managers have heavy weightings to Western Europe and Japan.  That is not good diversification from the US.  If you are going to invest overseas find a fund that is truly different.  (EG Allianz NFJ International Value: I will not buy for reasons listed above but that is a good example of a team approach and different kind of an international fund that can provide diversification and outperformance.)

5) Lastly, I believe that even if the US is to underperform that rest of the world a fund like BRAGX/BRAIX can still give great returns regardless.  They hold 40-70 stocks that look nothing like any index and therefore I believe over the long term, even in a difficult US market, it can provide wonderful returns.  I understand it is backward looking but the last 10 years are good example of that).  In essence, I can get the returns I am looking for over the long term (not 1,2,5 or 7 yrs but 10+yrs/15+yrs) with such a stock picking strategy that BRAGX/BRAIX implements.  That is the educated guess I am willing to live with understanding that my portfolio can potentially have less volatility and do better shorter term if other asset classes were part of the overall portfolio.

 Not sure this helps but it the approach I have decided to take until something comes along that fits all of my above listed criteria.  I just don't see that happening. 
 

Totally agree with your second point.  Unless you are in your distribution years, volatility does not play a factor except to help with the "sleep at night" factor many investors need.  That is why I own so much BRAGX/BRAIX.  I am not planning on withdrawing for 17 or so years (if ever) so I could care less about the volatility.  I am looking for what I believe is the best strategy to deliver the best long term returns.

 
To your first point, I would agree with it assuming that the individual is actually able to pick the best 10 funds for the next 10 years.  We are dealing in theory only here and since noone can accurately predict which funds will perform best over the next 10 yrs (arbitrary number) there is no way to know if adding the additional fund will help or hurt.  Therefore, there is no way to know if you are going to dilute or enhance your returns.  It is simply an educated guess.  That I think we can agree on?

 

Warren Buffett's oft-quoted line to the effect that "if you're clueless, pursue diversification; if you know what you're doing, pursue concentration" stands in notable contrast with the large collection of diverse companies (everything from carpets and candy to jewelry and insurance) he has assembled under the umbrella of Berkshire Hathaway.

the way buffett and soros invest is by buying the market space, betting against something and in general doing this that someone without their $$ can not play.  The can make move and a number of sheep will follow..  I'm sure it is nice, but not very many people can play that game.  Not a good analogy.

Both of these points are off the mark and reinforce my point that most investors are not willing to do their homework. 

1. Berkshire's current wide diversification is a function of a huge asset base that Buffett has accumulated over many years.  He has said many times that this is a liability and will reduce his performance.  Buffett's historical outsized returns were established through very concentrated bets in securities such as AXP, Geico, WPO, and KO, e.g. as high as 60% of the portfolio in AXP alone in the 1950s. 

2.  Buffett's strategy is completely different from Soros and does not depend on shorting.  He has taken advantage of Berkshire's heft to attract private sellers in recent years, but the returns from the initial 20-30 years were primarily based on long-only bets in individual stocks based on information gained from publicly available material. 

Best,
Oildog

Sorry for all the posts....I forgot to answer the last question of your post though.  I do think that "certain" parts of the world can and already have decoupled and that a manager that thinks outside the box (as mentioned in my prior post) can enhance returns and create the kind of diversification you are looking for.  The problem is that most managers do not think outside of the box and even for those that do, most do not fit my search criteria and I believe that my current managers can provide the returns I am looking for without going overseas (again, over long periods of time).

Thanks for the reply, valu. 

I too avoid bank-owned funds, so I would not invest in the funds you suggested.  One fund I would pair with BRAIX in a heartbeat if it were more tax-efficient is TAVIX.  Otherwise, there is no international fund I have found to be satisfactory and which I could trust that I could keep for the long haul. 

I believe BRAIX can own up to 10% in foreign equities.  Right now M* lists it at around 15% though.  At the top of my wishlist is an Aggressive 3 global fund.  I have tried to email Mr. Montgomery about this but received no reply.

 

the way buffett and soros invest is by buying the market space,

Both of these points are off the mark and reinforce my point that most investors are not willing to do their homework. 

okay, so what is this:

Over the past three years, Berkshire has spent $27.3 billion to buy seven companies in industries as disparate as aviation, fast food, and home furnishings. The $22 billion purchase of reinsurer General Re Corp., which closed late last year, was Buffett's largest ever.

He doesn't invest in stocks, He buys companies...  Most investors/ investment companies don't go around an buy the company..  ie,  not everybody can invest like this,  he changed the rules, and it worked for him..  Great job,  Soros did likewise, they played a different game an came out winners..

Buffett 'quotes/ rules' won't steer my way of investing because I can't play his game..

 

To your first point, I would agree with it assuming that the individual is actually able to pick the best 10 funds for the next 10 years.  We are dealing in theory only here and since noone can accurately predict which funds will perform best over the next 10 yrs (arbitrary number) there is no way to know if adding the additional fund will help or hurt.  Therefore, there is no way to know if you are going to dilute or enhance your returns.  It is simply an educated guess.  That I think we can agree on?

Two different issues: 1. You are uncertain about which fund has the highest expected returns.  2. You are uncertain about which fund will have the highest realized returns in the future.  Uncertainty about #2 is a given.  It's possible that A-Rod will hit fewer home runs than an average player in any given baseball season.  However, you will maximize your expected home runs by choosing A-Rod over the average player. 

Best,
Oildog

 

 

okay, so what is this:

Over the past three years, Berkshire has spent $27.3 billion to buy seven companies in industries as disparate as aviation, fast food, and home furnishings. The $22 billion purchase of reinsurer General Re Corp., which closed late last year, was Buffett's largest ever.

I don't have anything to add - your question indicates you aren't reading what I've already stated, and I'm not inclined to waste my time repeating myself to educate somebody who has no willingness to listen.  If you're curious, I suggest reading through all of the Buffet Partnership letters and Berkshire Annual Reports.  You will find your answers there. 

Best,
Oildog

[quote user="kerryvan"]

Oildog, How I so disagree with your comment..  from another post, copied here. I think your portfolio has far too many funds.  When people talk about "not putting your eggs in one basket," it is in reference to securities, not mutual funds.  Generally, mutual funds are already widely diversified across a large number of securities, so there is no need to hold a large number of funds (unless you are allocating to specialty funds like FSDAX).  If anything, holding a large number of active funds makes it more likely that your portfolio will perform like a market index fund, but with higher fees.  I would look at the size of the portfolio, then choose the number of funds.  when you have critical mass, then the fund count should grow. I don't hold any more than 10 % in a single fund. I make sure the overlap, via xray stocks is minimal.   Owning multiple funds in the same classification, with no overlap is better than owning 30% in a given fund.  Ask Magellon owners.  As the portfolio grows, the number of holdings I own grows. If you are in the better funds, top 20%, then you'll do good.  Make sure you are in funds that don't correlate, you'll come out ahead with less risk, another fido stat.  So an approach with limited $ is to own diversified funds, then as it grows use re-balancing to obtain access to targeted funds.   my $0.02 + interest

[/quote]

I'm late to this post, and I see there are 26 replies already.   I'm too lazy to read them, so with apologies I will make only a couple of comments:

1.  Oildog is right on target, 100%.   Fund count should absolutely not grow with portfolio size.   Where will it end?   When you get to $10,000,000, will you have 200 funds owning tens of thousands of stocks?   "Owning multiple funds in the same classification" is eventually just owning the index.

2.  Harris Associates manages private portfolios with a minimum size of $5 million.   Can you imagine getting to $4.9 million by owning 20 funds and several thousand stocks, and then cashing it in and giving your $5 million to Mr. McGregor to invest in about 50 stocks?

3.  My best recollection is that Buffett has $60 billion invested in about 35-40 companies, with 75% of that in the top 10.   Now I know, obviously, Berkshire's goals and objectives are different than most private investors', but the point is simply that you don't need to own 20 or more funds (or funds of funds!) and thousands of stocks.   If you want megadiversification, just buy Vanguard TSM. 

Concentration In Stock Picking

I just finished reading "The Intelligent Investor" by Bejamin Graham a few days ago.  It was a good read, although I can't say I learned much I didn't already know.  I'd already come around to the understanding that a declining market actually reduces risk, since you are buying companies more cheaply, while an increase in the market actually makes companies more pricey.  A bit counter intuitive, but quite true when you get right down to it.  I also already knew that I was not willing to invest the time and effort required to pick individual stocks and not do a shoddy job of it, hence why I invest in mutual funds. 

Anyway, I digress.  One thing that struck me that applies to this conversation is a section at the end of the book that was written by Warren Buffet (it was from some talk he had given).  In it he discusses how the disciples of Benjamin Graham's value investing philosophy did after Graham had retired. 

The interesting thing was that each of these disciples (Warren Buffet, Walter Schloss, and the guys who founded Tweedy Browne) had quite different ways of investing.  Some followed a concentrated approach (Buffet), while others held around one hundred or more stocks (Schloss, Tweedy Browne), yet all did quite well.  Some other examples were discussed as well.

Putting it briefly, I think it is wrong to assume that concentration in stock investment necessarily means better performance.  I believe it depends on the specific fund manager and their inclination.  

Another point is I do not feel that Graham was an advocate of extreme concentration.  His viewpoint seems to be that on average value investing techniques work, but there is always some degree of uncertainty, or chance involved.  By diversifying you are limiting the impact of chance on your results, and thus leveraging the advantage that value investing techniques give you. 

I believe this is a strong argument against extreme concentration, because just because an individual stock pick is your "best idea", doesn't mean it cannot prove to be wrong, possibly for no fault of your own.  For example, was Bill Nygren wrong about his investment in Washington Mutual, or just unlucky?  Chance can cut both ways. 

Diversification provides protection against random acts of chance, or if you prefer, unforeseeable conditions, that can arise and crush even your "best ideas". 

(Please note that this does not rule out concentrated funds, see the fund section below, but it does argue against holding only a single concentrated fund.)

Concentration In Regards To Mutual Funds

That said, there is a major difference between individual stocks and mutual funds.  That difference is the fact that a mutual fund, even a concentrated one, is already fairly diversified.  There is no need to spread your money around between a dozen managers for diversification purposes, because you'll already have that by owning as few as one or two funds (assuming that each fund is not super concentrated).  Even if your funds are all highly concentrated (20 holdings or so), you'll achieve adequate diversification with four or five funds. 

When it comes to choosing funds I feel there is only a select group of managers who are really worth investing in.  Which of them will perform the best going forward?  I don't know for sure, but I don't think it hurts to try to pare the list down to the point where you end up with a few of the most favorable conditions. 

Which mutual fund performs the absolute best is probably determined by chance as much as skill, but chance is unpredictable so it can't be used as a criteria.  We're left to focus on those traits which are readily identifiable which can indicate a fund has a chance of performing better.  Since a "great fund" by this sort of measurement has just as much chance of also benefiting from chance as a merely "good fund", I see no reason not to stick with the "best". 

 Fund count should absolutely not grow with portfolio size.   Where will it end?   When you get to $10,000,000, will you have 200 funds owning tens of thousands of stocks?   "Owning multiple funds in the same classification" is eventually just owning the index.

This going to an extreme, I know just to prove a point.  I get the impression people only hold 2-3 funds.  A couple US and maybe 1 foreign.   For investments less than 300K this may be okay.  I like my 10% rule for any single fund.

I would not own more than 2-3 funds in any single style, and then only for a good reason.  For intl exposure, a broad based fund is good for starters,  then add country/ region specific funds depending on the market.  Again, 2-3 funds max for a region.

So when the portfolio is 800K, I could easily see 12-16 funds.  At 1.4M, 18-22 funds, then at 10 M maybe 25 funds.

Harris Associates manages private portfolios with a minimum size of $5 million.   Can you imagine getting to $4.9 million by owning 20 funds and several thousand stocks, and then cashing it in and giving your $5 million to Mr. McGregor to invest in about 50 stocks?

No, I could never do that, I think it would be way too risky.  Also, I'm not sure with 50 stocks you will get good exposure to the world markets.  A successful fund manager may hold 200-300 stocks, only 7-10 are the winners for the performance.  I'd rather have better odds.  Hence, I don't go to Vegas...

I agree the fund of funds is a lousy way to invest.  The funds get rich, an index would be a wiser choice.

the overall point of the discussion was to hold more than a few funds for good world market exposure.

I just want to add that my outlook on investing globally is quite different from valunvstr's.  For the most part, I don't really care much about MPT, instead I concentrate on getting the best value.  If my fund managers can buy assets for 60 cents on the dollar, then I'll do fine, regardless of whether or not my portfolio is perfectly tailored to reduce asset correlation.  

Anyway, my viewpoint on investing globally is that it gives your managers the chance to find and invest in the best values, without being arbitrarily restricted by considerations such as country of origin.  Just pulling an example out of thin air, I couldn't care less whether Europe and Japan are correlated with the U.S. while Brazil isn't, that is no reason to invest in Brazil.  But, if a cheap company can be found in Brazil that offers a better value than a company in the U.S. or Europe, then that is a different story. 

Just a different way of looking at things, not necessarily that either of us is right or wrong. 

Kerryvan, one thing I think you may be underestimating is how your thinking changes (and needs to change) as your portfolio gets larger.  For example, when your portfolio is $800,000, a 1% shift in value becomes $8000.  If your portfolio gains or loses 10% you're looking at a gain or loss of around $80,000.  Putting $80,000 (or more) in a fund will not seem as aggressive to such an individual as it might be to someone whose entire portfolio totals $80,000 (just an example).  

Just as right now you might feel someone with only $5000 to invest is silly in feeling uncomfortable putting it all in one fund, you'll also feel differently when your portfolio is larger. 

That isn't to say I would invest 5 million in 50 stocks, if I had it.  I probably would not feel comfortable doing that.  But, I could see investing 5 million in around 100 stocks (200 max), and also a good amount of bonds. 

Kerryvan, one thing I think you may be underestimating is how your thinking changes (and needs to change) as your portfolio gets larger. 

I do understand this very well. I noticed I'm holding more funds that I would not have held many yrs ago.  In the early 80's I was in gold, and I got out 90% of it due to the swings.  Today, I'm in uupix  which I would recommend for the ave investor.  And the +- 10 % swings don't give me an ulcer. 

Why, because all the other funds provide the balance.  If I only held 10 funds then the daily changes of 10% would be a concern.  BTW only hold this fund in a tax free account.  the drop to $29 a share and the ride to $44  in the past 2 months did not cause a sleepless night.

Nagorak,

It's interesting that you bring up the difference between the Schloss approach of a more diversified value investing, versus Buffett's concentrated style.  And, I think it highlights Buffett's view (mentioned above by oildog) that the amount of diversification one has should roughly track the degree of lack of knowledge one has about one's investments.  That is, all other things being equal, the less you know what you're getting into, the more you should gravitate toward broad diversification.

From what I've read, Schloss was really just implementing a quantitative value approach, not unlike what many value index funds are based on today.  Schloss owned shares in many companies because he knew next to nothing about any of them.  If you were building a portfolio based only on the fact that companies in your portfolio were trading for less than two-thirds of book value, how many would you need to hold to feel comfortable?  Also, all other things being equal, if you found 1000 companies all trading at roughly half of book value, why would you include some to the exclusion of others if all you really know about any of them is this one metric?

On the other hand, Buffett's preferred style of investing is to invest in a very small number of companies where he thinks he understands all the important quantitative and qualitative factors that will affect the long term economic condition and position of those companies.  Given the increased thought about the economic prospects of a company, it doesn't seem unreasonable to put a large percentage of one's wealth into such an investment.  Indeed, this seems like what most small business owners should be doing when they "invest" in such enterprises.

"Putting it briefly, I think it is wrong to assume that concentration in stock investment necessarily means better performance."

What you wrote above is correct.  However, I think the "value" in seeking out concentrated bets is not that it ensures better performance, but that it offers increased odds for outsized gains.  This has nothing to do with personal tastes or predilections; it is simply a matter of probabilities, borne out in history by statistics.

Yes, value investors can do well following a more diversified strategy, but the outliers in performance, both positive and negative, will be those with more concentrated investment portfolios.

'tato
 

 

kerryvan wrote:
Okay, so I think this discussion is getting a little off track.  In any given asset class I'd own 1-3 funds. LC, MC, SC.  (total 6-9 funds depending on asset base)

This investing method is not consistent with what you said about overlap. If you hold three funds in value and three in growth you have lots of overlap. In addition, holding another three in blend overlaps everything. There is no need to hold blend if you hold value and growth.  Furthermore. Holding three funds in one class means that two are going to underperform the best one. If the goal of using active funds in to outperform, then just pick the best one. Holding nine funds in large cap almost insures you will have a closet index fund.

 

 

You miss understood what I wrote,  I'll try to make it more clear...  I would own up to 3 large caps,  up to 3 mid caps,  and up to 3 small caps...  I would look for minimal overlap, so one may be Large cap growth, midcap value, small cap growth..  My total number of funds for US is 6-9 total..

the rest of the discussion was stating the foreign breakdown.

[quote user="small potato"]

"Putting it briefly, I think it is wrong to assume that concentration in stock investment necessarily means better performance."

What you wrote above is correct.  However, I think the "value" in seeking out concentrated bets is not that it ensures better performance, but that it offers increased odds for outsized gains.  This has nothing to do with personal tastes or predilections; it is simply a matter of probabilities, borne out in history by statistics.

Yes, value investors can do well following a more diversified strategy, but the outliers in performance, both positive and negative, will be those with more concentrated investment portfolios.

'tato 

[/quote]

That is a fair point, and I agree.  Personally I prefer a bit more concentrated approach, but not super concentration.  My ideal mutual fund has about 40 holdings, as opposed to 100 or to 20.  Anyway, you're right the more concentration the more chance to deviate both positively and negatively, so it cuts both ways.