How concentrated should you be?

I have been planning an article on portfolio composition and concentration for some time. My portfolio composition to date has been quite simple. It consists of 25 to 30 positions, all more or less equally sized accept for cases where I bought more at a lower price. In spite of studies showing that 15 positions eliminate 80% of company specific risk in a portfolio I arbitrarily chose 25 to 30 positions as I knew my selections were not perfect and I wanted a bit more diversification.I hardly ever have more than 30 positions in my portfolio, as that is the maximum number of companies I can comfortably keep track of.


When doing research for this article I remembered a very good presentation Zeke Ashton of Centaur Capital Partners gave on portfolio composition and concentration at the Value Investing Congress in May 2007.


Zeke graciously agreed to let me use his presentation for this article.


The table below shows the composition of the portfolios of noted value mutual fund managers:

Fund	No of Ideas	% of Portfolio in Top 10 Ideas	% of Portfolio in Best Idea
Sequoia	29	61.5%	22.5%
Tilson Focus	61	55.7%	8.9%
Clipper	23	69.9%	12.6%
Fairholme	21	61.0%	16.2%
Oakmark Select I	21	56.3%	9.9%
Legg Mason Growth Trust A	34	39.0%	4.6%
Longleaf Partners	21	71.2%	13.8%
Weitz Value	35	51.2%	13.3%
Legg Mason Value A	45	45.8%	6.3%
Tweedy Brown Value	46	36.1%	5.5%
Third Avenue Value	43	64.7%	12.8%
Tilson Dividend Fund	28	50.5%	6.9%

Source: http://www.morningstar.com as at 10 Aug 09


As can be seen the portfolio concentration differs but they are all relatively concentrated with high percentage bets, up to 22.5%, in one company.


This differs substantially from the 100 plus securities in most mutual funds.


Common Value Investor Portfolio Models


The following are several common “portfolio models” used by successful value investors:


Ultra-Concentrated Portfolio Model

Definition:

Fewer than 10 stocks with large position sizes routinely comprising 20-25% of portfolio assets and larger.Â

Practitioners:

Chieftain, Eddie Lampert, Tom Brown,

The 10 Stock Model

Definition:

Standard position size of around 10%, though there may be one or two larger positions, and a handful of smaller positions for a total of 12-20 ideas. Practitioners:

Clipper, Mohnish Pabrai until the end of 2008, he now uses the 20 stock model below

Standard 20-Stock Model

Definition:

Standard position size for a good idea is about 5%, though best 2-3 ideas may be modestly larger and many ideas are somewhat smaller. Total portfolio of 25-40 ideas. Practitioners:

Many of the value mutual fund managers in the above table: Robert Hagstrom, Bill Miller, Wally Weitz, Longleaf Partner, Tweedy Brown Value, etc.

20-Stock Model (Super-sized)

Definition:

Essentially the same as standard 20-stock model, but two or three best ideas are “super-sized” to 10-15% of the portfolio, and there are fewer sub-5% positions. Total of 20-30 ideas.Practitioners:

Tilson Focus, Fairholme, Sequoia, Oakmark Select

The Centaur Capital Partners Portfolio Model


LONG POSITIONS

6.0% - 7.5%	Outstanding Idea, 1-2 Best per year, combines compelling valuation with significant margin of safety.
4.0% – 6.0%	Standard Great Idea, usually will be one of their top eight to ten ideas.
2.5% – 4.0%	Solid or even excellent idea with one or more minor risk factors, which might relate to business or industry quality, valuation, liquidity, political risk, or level of their conviction and ability to completely understand all aspects of the business.
0% - 2.5%	Interesting and sometimes compelling idea that may be very illiquid, may be a probability bet with a favourable asymmetrical reward to risk ratio, or may simply be a low quality business that is very cheap relative on a net-net working capital or price / tangible book value basis.

SHORT POSITIONS

>4%	Generally reserved for shorts or hedges involving the use of broad market or sector specific indices.
3 - 4%	Most compelling individual short idea where we believe risk is very low.
2 - 3%	Standard Excellent Idea, usually will have no more than two or three shorts of this size.
1 - 2%	Solid or even excellent short idea with one or more but certain risk factors to the short thesis might be present, such as high short interest, low float, low market cap, etc.
0% -1%	Generally reserved for short ideas utilizing a put option instead of common, where the probability of a good outcome might be low but the magnitude of a positive outcome might be significant to their performance.

The Reliability Pay-off

Portfolio structure is not about reducing volatility, it’s about increasing reliability.You want enough ideas to ensure that your sample size is big enough to reward skill and absorb the occasional bad outcomes, bad decisions, bad timing, or bad luck.Adding in a mix of five to ten short ideas further improves reliability.

Maxims Regarding Position Size

The goal for all investors should be to get the most value out of your best ideas without risking significant capital loss if you are wrong. Concentration isn’t a constant - it is idea and environment dependent. Your philosophy on selling will determine to some extent how many positions you hold at any one time. The more concentrated you are, the more rigorous you have to be and the more good ideas you have to reject.The more ideas you have, the harder you have to work.Diversification in terms of the valuation factors (Price/earnings or Price/book) also known as Factor diversification can be a good thing as it further diversifies your risk.Sample sizes matter. A certain minimum number of ideas is required to protect skilful investors from the capriciousness of luck and unexpected bad outcomes. If you can’t sleep well at night, either you don’t own the right stocks or you are running too concentrated a portfolio.

Practical Questions to Ask Yourself

What type of portfolio is consistent with your portfolio and personal risk / stress tolerance? Can you sleep at night if you have -10% months? or 25-30% peak-to-trough declines? What are your goals? Maximum returns (with the assumption of significant volatility) or satisfactory returns within a certain risk profile?How many ideas can you process and maintain?

What kind of ideas do you prefer? How stable is your capital base? Are your investors prepared for significant volatility? Do you have safeguards in place to protect your capital from fleeing at the worst possible time?Do you work best alone or do you prefer a team environment? A team is going to generate more ideas than a solo practitioner.

Final Thoughts

“Confronted with the challenge to distil the secret of sound investment into three words, we venture the motto, MARGIN OF SAFETY.”

Benjamin Graham, Intelligent Investor


Why is having a margin of safety important?

Valuation is an imprecise artThe future is inherently unpredictableHaving a margin of safety provides protection against bad luck, bad timing, or error in judgement. We believe that the principle of “margin of safety”is just as applicable to portfolio construction as it is to individual investment selection.

The positions in my portfolio have all, more or less, had the same weighting. Zeke's article has however changed my view.


Thinking back there was definitely a few companies that deserved a larger position due to their undervaluation and business quality.


I am however still comfortable with my 25 position portfolio as, in spite of by best efforts and thorough analysis, I was still wrong on a few companies. This was in spite of what I thought was a high probability winner.


As Zeke mentions above, valuation is an imprecise art and the future is unpredictable.


Tim du Toit

http://www.eurosharelab.com


Tim du Toit is the editor and owner of Eurosharelab. He has more than 20 year of institutional and personal investing experience. Tim is based in Hamburg, Germany. More of his articles can be found at http://www.eurosharelab.com