"Robert, we just focus on a few outstanding companies. We're focus investors." - Warren Buffett

That quote appears on page one of Robert G. Hagstrom's excellent book The Warren Buffett Portfolio. Buffett was explaining to Hagstrom his position on diversification.

For most investors, who do not have the time or the inclination to research their portfolio properly, diversification is an important tool for ensuring that a disaster that destroys a certain sector won't wipe out your entire investment. If you have all your money invested in one or two stocks, and you don't keep up with events on those stocks, you are exposing yourself to a large risk of getting the "nothing" side of "all or nothing".

Diversification means not putting all your eggs in one basket. Diversification means putting various eggs into baskets strewn all over the place so that when predators come into the area they will only get a few of your eggs. Focus means carefully watching over your basket and keeping it from harm. The way to watch a basket in investment is to do your homework first, and keep up to date on events affecting your portfolio.

Just talking with other aus.invest readers I often hear from those with large portfolios complaints that with so many stocks to track, it is hard enough just keeping tally of the dividends and tracking the purchase dates of different parcels for CGT purposes, let alone thoroughly researching each and every company and keeping an eye on technical factors, if that appeals to the investor.

Warren Buffett does not believe in throwing money at everything, he says to pretend you have a punch card, you only get ten trades in your life, so if you blow them all you have to stop investing. In my first week as a novice trader I blew that punch card numerous times, selling a stock, buying it back immediately as it went up again, selling and watching it fall then buying back when the fall wasn't quite complete. I made money on every single trade (in fact, due to the bull market, and amazing luck, it took me many months before I made a single losing trade). Despite the hit ratio, my returns were slight, I made a few thousand dollars in trading, but I later calculated that if I had held onto my first stock for a month or two and then sold in the middle of a big fall, then bought my second stock and held onto it for a while, I would have made about $13000. I had carefully researched to find the right stocks, and managed somehow to snatch mediocrity from the jaws of great success.

Never take a trade lightly, unless you truly want to be a short term speculator (which I did, at the time). Buffett researches stocks for many years identifying what he sees as the intrinsic value of the top stocks in the world, he waits for many years until he finally gets a major buying opportunity (like the crash of 87, following which he made many of his most famous purchases, including Coca-Cola) - he buys when the value is right.

Buffett spurns diversification for the intelligent investor, he believes in careful analysis and knowing every single thing going on in a company and the economy. For the rest he endorses index funds.

The alternative is known in investment circles as portfolio theory. This is a range of techniques used to hedge against a variety of risks. For example, to overcome currency risk (the risk that your returns on a bullish stock will be killed by a falling currency), portfolio theory exponents advocate investing in several countries so as to balance out this risk. Investment in several sectors that have historically peaked at different times is another aspect of portfolio theory.

Hagstrom's book The Warren Buffett Portfolio is an excellent reference book for focus investors, whereas one book I read recently that was quite good with portfolio theory was Fortune Strategy by An Higgins and Arun Abey. There is no doubt considerable virtue in both methods, either concentrating on being invested in the best shares or a strategic game plan to take a measure of risk and optimise a portfolio for less active management. Both techniques are a far cry from "dumb" portfolio management, as practiced by the majority of stock market players, which consists of subscribing to a newsletter for hot tips and buying whatever looks good.

Neither approach could be described as a strict long term or short term strategy, Buffett is well known for holding onto his favourite stocks for many years but still has a turnover of more than 10% every year, as he sells his less successful stocks and gets into better investments, and similarly portfolio strategists allocate resources to compensate for various economic factors and may sell individual investments to get better value.

An interesting experiment was performed by Hagstrom in writing his second book, in order to assess the effects of diversification on ultimate potential performance, they ran a computer simulation of a number of random portfolios on the Standard and Poors Compustat database of common stock returns. They programmed the computer to randomly select, from 1,200 companies 12,000 portfolios of various sizes:

• 3,000 portfolios of 250 stocks

• 3,000 portfolios of 100 stocks

• 3,000 portfolios of 50 stocks

• 3,000 portfolios of 15 stocks

The average return of all the different portfolios was about the same for each portfolio size, though, as one might expect, the smaller portfolios showed the widest range between the highest and lowest returns. In fact it became a near statistical impossibility to beat the market as the number of stocks in the portfolios grew, though at the same time the chances of attaining a result far below market returns also diminished in the larger portfolios.

As most people are terribly bad stock-pickers, including fund managers, it makes a lot of sense to invest in larger portfolios. For those with talent, or in fact anyone who bothers to do any worthwhile research at all, large portfolios make no sense as it becomes correspondingly more difficult to achieve a very good result.