Citywire for Financial Professionals
Stay connected:

Citywire printed articles sponsored by:

View the article online at

If Warren Buffett doesn't diversify, why should you?

Citywire's Smart Investor says you only need shares and bonds for a well-balanced portfolio.


by Smart Investor on Dec 10, 2012 at 08:01

Followers of the most successful investor of all-time, Warren Buffett, may already know his thoughts on diversification. For those of you who do not, his views are quite simple: he doesn’t believe in it.

Of course, all of the textbooks and all of the advisers will tell you that he is wrong. Indeed, his mentor, Benjamin Graham, will tell you the same thing. However, maybe Buffett has a point; maybe diversification is at least overrated?

Diversification 101

Diversification can take two main forms for the private investor who runs their own portfolio: diversifying among asset classes and diversifying within each asset class. For example, you may hold a mixture of shares, bonds and commodities in your portfolio (which is diversification among asset classes) and then hold a number of different companies within stocks and bonds (which is diversification within asset classes).

The main reason used to justify diversification is that it spreads risk. In other words, if your  portfolio contains only shares and the stock market falls, then you will sustain large losses. Whereas if you had some government bonds and other assets in your portfolio (in addition to the shares) your losses may have been smaller.

The idea of spreading risk extends to diversification within asset classes, with a portfolio of 20 stocks being less risky than a portfolio of 2 stocks since if one company goes bust it will halve the latter and make a smaller dent in the former.

Risks allayed, what about rewards?

However, if the world’s richest investor does not diversify, why should you?

Sure, it may reduce risk but risk is only one side of the equation – reward is the other and diversification, whilst it undoubtedly reduces risk and volatility within a portfolio, can also reduce your reward.

This is especially obvious when investing in shares or funds. The long term performance of shares is generally accepted to beat other asset classes. This is open to some debate, but is generally accepted as fact in the investment community. Why, then, would you invest in assets which are not generally viewed as delivering the same or better level of return? Surely that is counter-intuitive?

To an extent, it is. This is because many major asset classes are closely correlated, since they rely upon the performance of the economy. Shares, higher yield corporate bonds and property are examples of assets which tend to perform better during periods of economic growth and worse during economic downturns. Why, then, would you invest in corporate bonds or property if they suffer when shares suffer and offer a lower long term return?

One up, one down

Surely you should be seeking out assets which are negatively correlated; i.e. when one goes down, the other goes up and vice versa. That way, you can continually buy low and sell high.

Going back to Benjamin Graham, this is exactly what he recommends: a portfolio made up of shares and bonds only, with a minimum of 25% in both assets and a maximum of 75% in one asset, gradually moving between the two as a see-saw does depending on where value is being offered by the market.

So, when shares are cheap you should move to 75% shares and 25% government bonds (gilts). When shares are expensive and government bonds offer high single digit gross redemption yields, move the other way. Of course, this will not be a sudden movement but a gradual one, taking advantage of cost averaging and low commission costs via aggregated orders.

Sign in / register to view full article on one page

25 comments so far. Why not have your say?

Income Investor

Dec 10, 2012 at 17:30

This is an important issue - if do too much diversification, you run the risk of buying the market and then you might as well buy an index ETF. But I think thoughful investors can do better.

For an income investor, gilts do not at the moment offer an attractive yield. My own portfolio (which I describe elsewhere on here) is 50% high-yield shares and 50% high-yield fixed income. It is not clear to me how the correlation between these stands - and let's not forget that correlation evolved over time. However, it seems to work for me at the moment: and as the shares component has risen to slightly over 50%, my next purchase will be fixed income..

report this

Bernard Bedford

Dec 10, 2012 at 17:39

Which annual Berkshire Hathaway letter to shareholders were you reading when you decided that Buffet didn't believe in diversification? He doesn't like gold and he doesn't take much interest in foreign companies, Tesco excepted, but if you look at the range of companies Berkshire run and the shares that they hold it's hardly without some diversification.

report this

an elder one

Dec 10, 2012 at 19:04

There's a lot of waffle indulged on diversification; my philosophy is buy what you like the look of with due diligence, stick with it and if things look to be going really pear-shaped then sell. The only time I failed in that regard was during the dot com bubble when I didn't sell soon enough because of the threat of capital gains; I've learnt that lesson.

I can't be bothered with bonds at all of any description and stick with equities (in essential resources and high yielders of good repute), some etfs and one or two funds in foreigns, but no deliberate attempt to diversify.

I've survived reasonably well over 20 years; not great; say average 5-6% year on year; some years bettered that; which satisfies me.

Income Investor's first para has it to rights.

report this

an elder one

Dec 10, 2012 at 19:10

I would add January 2011 - 2012 to date has been ghastly, though I've broke even nonetheless.

report this

Jeremy Bosk

Dec 10, 2012 at 20:04

I second An Elder One: except that I have gained about 20 per cent over that period - after spending £300 a month from capital and dividends. How much is skill and how much luck I really don't know. Sadly, I started from a really small amount and have not been able to add new savings since early 2008 when I had to accept early retirement.

The one thing I do that many people seem psychologically unable to do is cut my losses before they get too big (immediately when really bad news comes) and take profits when prices get silly.

My current portfolio - there have been purchases and sales since January 2011 - is:

PEQ, invests in US tech oriented private equity funds

CSN, runs (mostly closed) life funds

VEC, biotech company with inhalers for chronic obstructive pulmonary disease.



RUS, REIT (logistics in Russia)

GLIF, buys mainly US corporate loans and recently European CMBS

NBL, sells ancient coins, stamps and other collectables on line and at auctions

RECI, invests in commercial real estate loans and similar

As you will see, most are dividend stocks but all have shown some share price gain since I bought them, six per cent being the least.

They all offer some diversification outside the UK through either overseas assets and operations or overseas customers. There is a concentration on property and financial related companies. I regard the current aversion to these sectors as overdone. Europe is my next area of interest ;-)

report this


Dec 10, 2012 at 20:35

Finally, someone is waking up to the bullsh8t modern portfolio theory about correlation. after about 15 stocks you are at maximum reduction in specific risk, but not market risk.

if you were a businessman, would you buy 50 - 70 different companies just to ensure that if one failed you would not loose to much money? No, you would focus on one and know it really well (inside out).

as ben graham said, investing is most successful when it is most business like.

for as long as science and maths tries to dominate investment and investment decision making we will have theories that blow up - to the professors- Markowitz, Sharpe, Farma.... blah blah blah - none of them wealthy through investing in their own right, but writing texbooks on the subject.

one couldn't write book!

report this

an elder one

Dec 10, 2012 at 21:18

Yes jeremy, my stupid mistake was during the dot com bubble when prices got really silly and it was obvious from a personal literacy in computer IT that the vast majority of dot coms would fail because there was not sufficient room in the market place; it was the thought of paying all that capital gains to HMRC that delayed me and prices fell very quickly; I learnt the lesson then to ignore tax implications.

Actually working out profit over a 20 years period with all its vicissitudes is a challenge; I've used a piece of software called Quicken to work mine out; don't know how accurate it is; my more recent essay into the FT portfolio facility (mainly to get moving prices) suggests 200% and a bit totally (which is nearer 10% year on year I think; I'm no accountant) over a 10 year period; don't know quite what to make of it; I reinvest all dividends, but with a pretty meagre pension, including ex-employer, what's left from an Equitable Life AVC and the State we have to take out capital occasionally for home maintenance things.

Your's are some interesting ideas.

report this


Dec 10, 2012 at 22:02

I think I am a fan of Buffett. We try to buy good quality long term investments in areas we think we understand.

We diversify by dividing our funds between house property and equities. We don't hold any shares in companies in housing related businesses like housebuilders or estate agents.

We focus on getting good quality property which should rent well and be easy to manage and buying shares in what seem to be good quality businesses. These will be companies which look to be good for a long term hold, and make a decent compromise between a reasonable yield and growth of the business, (obviously most companies being more one or the other).

It seems to be a sound philosophy and we try to make it work.

report this

douglas gordon

Dec 11, 2012 at 05:59

Ah yes, Elder One - the CGT dilema > balancing the certain cost of CGT against possible losses on a falling stock that has made us money! I normally use my CGT allowance each year, not to realise capital, but to reduce the possible future gains by selling and often buying back the same or similar shares. Just a pity we can't carry forward unused allowances. It's all a gamble, eh?

report this

an elder one

Dec 11, 2012 at 13:50

yes Douglas! but doing that you've broker's charges to take into account; another complication.

report this

Jeremy Bosk

Dec 11, 2012 at 18:05

I use a flat rate, execution only broker which uses a polling system to get the best price from the market. This minimises the transaction costs. I also try to avoid very small transactions where the flat rate (£12.50 from Selftrade) would be too high a proportion of the cost. Of course I am a small investor so none of my deals would be more than small change on the institutional scale.

There are cheaper brokers out there, especially for frequent traders but I stick with what I know to be competent and reasonably priced. They also charge the same flat rate for deals in other countries represented in the Boursorama Group including France, The Netherlands and Germany. I am thinking of doing more in Europe because it is oversold. The internet and a subscription to the FT make research much easier than it used to be.

report this

douglas gordon

Dec 11, 2012 at 19:33

Yes Elder One - as Hamlet said so long ago ".... ay, there's the rub."

report this

Anonymous 1 needed this 'off the record'

Dec 12, 2012 at 09:21

God there's a lot of rubbish being talked here.

"someone is waking up to the bullsh8t modern portfolio theory"

The problem is not necessarily with MPT but actually more how it is implemented. Correlation is not constant, as we saw in 2008. MPT is not perfect, it makes some assumptions which are not always correct (markets are rational and efficient, have a normal distribution etc) but to denigrate one of the greatest mathematical discoveries of the century, with regards to investment management, is frankly just ignorant as you clearly don't appreciate the genius of Markowitz's work.

There is a real misconception about diversification reducing returns which is a complete fallacy. The individual performance of an asset is less important than the correlation it has to the rest of the portfolio. People always think that if one asset is negatively correlated to another then the returns will cancel each other out over time. Correlation is talked about so loosely, especially in some of the posts above.

Equities have a long term return of maybe 7-8%. It would be much higher, were it not for the periodic drawdowns they suffer. So, if you can include something in your portfolio that performs well when equities don't then it's quite simple to improve your overall returns and reduce your risk.

If you want the smoothest equity curve, you want to combine assets that have a time series correlation as close to 1 as possible and a return series correlation as close to -1 as possible.

I will also add that a portfolio that is 75% equities means that your beta risk is most likely to be over 90% so you are by no means diversifying your risk.

Metacognition......look it up in wikipedia

report this


Dec 12, 2012 at 09:47

Blah Blah Blah.... while you do your theories and believe what ever is printed in text books written by salaried professors in the 50's who have never made any real wealth in the open stock market - I will continue to read books written by the mega wealthy value investors who laugh at these theories.

Anyway professor, if you are a member of the CISI - obtain a report called MPT Reviewed by K Robertson. Might make you think about what you believe in.

Common sense investing.... .... try it in tthe real world.

report this

Anonymous 1 needed this 'off the record'

Dec 12, 2012 at 10:00

Spoken to like a true ignoramus.

I despise people like you

report this

an elder one

Dec 12, 2012 at 10:30

Oh dear, not another thoughtless dogmatist. 'Despise' is a nasty word to use, esp. against anyone one has probably never met; and cowardly written in anonymity.

report this


Dec 12, 2012 at 10:47


a mere difference of opinion shouldn't make you 'despise' people like me.

It is an exact replication of Dr G Farma who brough us the ultra bulls8t theory of Efficient Market Hypothesis ver1 ver2 ver3 - he and Buffet have had 40yr running arguement about his theory. Buffet is worth $50billiln or more and Farma?

Farma once shouted at the top of voice 'God knows the market is efficient!' - almost fanatical in his beliefs rather than learning from a real time investor.

report this

Anonymous 1 needed this 'off the record'

Dec 12, 2012 at 12:31

Let me get a few things straight

1) I had no wish for this thread to descend into a slagging match. I gave my carefully thought (and I would add, perfectly logical) opinion and it was him who responded with "blah blah blah" hence he I can only conclude he is ignorant....and yes I despise ignorant people, particularly those who are unaware of how ignorant they are. How predictable that I referred to this before he even posted his reply.

2) My opinion is based on many years of experience of managing both my own and more importantly, other people's money. I have worked for large organisations who have generally made lots of basic mistakes in the management of clients' investments and have been fortunate enough to look at the investment records of 1000s of clients....some of whom have made a lot of money and even more who have lost a lot. Now that I am working for a business that manages client money very successfully over a long period, and I mean in terms of returns, I think I have a pretty good idea of what works and what doesn't. So, I think my opinion is based on a lot of personal and professional experience rather than (in Value Investor's case) just on my own relatively meaningless personal portfolio. I could quite easily take a £50k portfolio and turn it into £100k in no time with a bit of luck. Does that I mean I am qualified to give people advice and think that I am some sort of investment guru?

3) I do not believe in EMH either

4) If you think Buffett simply just buys companies he believes are "cheap" from a valuation perspective then I can tell you now that you will fail if you think you are copying him

5) I find it depressing that someone can support another person who makes a statement that mathematics has no place in investing

That's all I'm going to say on the matter

report this


Dec 12, 2012 at 14:07

Anonymous 1 - sorry that this has lead to a slagging match. not what I wanted either.

A few things to put the record straight :

I guess your professional experience includes large companies like UBS, Credit Suisse, Legal & General or some other 'grand' investment managment firm.

for your information, I to have 'been there done that' - I have worked for large institutions too

my livelyhood is managing client money both retail and institutional

i advise clients on investments and have two CF30 functions - advisory and discretionary permissions from the FSA

i am a member of the CFA Society, CISI, CII

I have travelled to N america to study value investing as Buffet did based on ben graham's work

So yes, I think I am qualified to give people advice. I never suggested I was a guru.

I am pleased you dont believe in EMH

buying undervalued companies is fundamental to value investing, and infact any transaction that intends to make a profit. buy low, sell high?

I am pleased to say that it does work, contrary to your belief. why dont you take your argument to Woodford at Invesco or the value team at Schroders?

if you want further imperical evidence that there is life beyond theory, then follow James Montiers (CFA) work on MPT

thats all i have to say on the matter

report this

Anonymous 1 needed this 'off the record'

Dec 12, 2012 at 14:35

Whilst it appears we are similarly qualified and experienced, I actually think qualifications stand for very little in the management of client investments (after a certain point). The best money manager I know is a former engineer for example and it's logic and common sense above all else that matter.

You keep giving me examples of people who refute MPT and I don't know why. I rate Montier's work and already acknowledged that MPT has multiple failings in my first post....BUT you still cannot deny some of it can be applied with great success. For example, Markowitz basically proved that you can add a volatile product to a portfolio in order to make the portfolio less volatile, so long as it has the right correlation characteristics and you can slag that off as technical mumbo jumbo as much as you want but it's a fact.

Ray Dalio is one example (oh and he's quite rich by the way, seeing as this is how you seem to judge success) who applies the same principles that I do so I frankly couldn't care less what Neil Woodford or any other long-only manager does because they are just guys focusing on one specific investment strategy which, to give them their dues, they do rather better than most of their charlatan counterparts.

It's healthy to have different approaches and I wasn't even denigrating value investing. I do however find it an unhealthy obsession most people have with Buffett and was merely saying that that you cannot possibly copy his approach unless you have substantial personal wealth. Buffett buys controlling stakes in large companies which enable him to have a fair say in how the company is run...something someone with 100 shares on BP won't be able to replicate.

Ray Dalio has some excellent views on diversification which give it far greater meaning than the typical "don't put all your eggs in one basket" analogy that most people dumb it down to.

report this


Dec 12, 2012 at 14:53

Ok, all forgotten. lets agree to disagree. I do agree that one should diversify by sector etc

I dont agree that Buffet buys controlling stakes. he is not an activist investor. quite to the contrary, he buys companies with excellent managers and lets them get on with managing the companies (one of his criteria). if the manager/company is not producing his preferred return on equity or retaining cash which cannot be reinvested at a better rate than he can then he will take action.

his view, as any long only manager is the same - whether you own one share or 100% of the issued share capital. you are a business owner.

report this

Jeremy Bosk

Dec 12, 2012 at 16:55

I tried to read a book on MPT (several at one time or another). They made me dizzy. :-(

Looking in Sharescope, I notice that high volatility shares tend to have no or minimal profits, suggesting that the price is very dependent on future possibilities which vary a lot on slight changes in circumstance and opinion. Many of these are oil and mining explorers. High volatility shares also tend to be low beta - move less than the market. Looking at my own portfolio, the volatility ranges from 5.38 to 2.13 with most in the 2.x range. My beta ranges from 0.2 to 0.69. Looking at all the shares listed, the median beta is 0.2 and the median volatility is 2.93. So, I suppose that if any share differs greatly from those numbers you want to know why. Or you could chart a share against its sector and the FTSE100 and again look for what makes it different and why. With a spreadsheet or data mining facility, you might refine the numbers by adjusting for market cap.

Rambling done for now, have fun.

report this


Dec 12, 2012 at 18:04

You can invest how you want as long as YOU understand the risk.

I advice clients and diversify their portfolios to reduce risk. Mainly because they dont understand risk.

personally I never invest in anything but equities. Thats because I understand the risk and reward.

report this

Donald Chan

Dec 16, 2012 at 09:35

I've got a headache after all that.

report this

Anonymous 2 needed this 'off the record'

Dec 16, 2012 at 10:16

nice to have an intelligent board,most are rubbishI have never resolved the investor conundrum, when young no money,when old no time!

report this

leave a comment

Please sign in here or register here to comment. It is free to register and only takes a minute or two.

Sorry, this link is not
quite ready yet