How does diversification affect investment returns?

[i4value]

Imagine that you have $100 invested equally in 2 stocks ie $ 50 each.

• A had 10 % return
  
• B had a loss of 5 %
  

Your total gain is the sum of $ 5 gain from A and a loss of $ 2.50 from B = $ 2.50 overall gain.

Now if you had invested the $ 100 in A alone, your gain would be $ 10. But if you had invested all in B alone, you would have lost $ 5.

This is the impact of diversification. It reduces the overall gain but it also reduces the overall loss

The more stocks you have given the same $ 100, the less potential gain. But then the risk of a loss is reduced.

Of course, you don't know how individual stocks will turn out so you try to have as many as possible so that if a few does badly the others will do well enough to more than offset the bad ones.

So, what is the optimum number of stocks. Studies have shown that the benefits of diversification become marginal after 30 uncorrelated stocks.

It has to do with the fact that the risk of a loss can be separated into systemic and non-systemic risks.

Non systemic risks are those specific to a company eg its product risks. Systemic risks are those the affect the whole economy eg depression.

If you have 30 uncorrelated stocks the theory is that you diversify away all the non-systemic risks. But you are still left with the systemic ones.

There are ways to reduce the systemic risks but it is not from having more diversified stocks.

Real life is not so simple.

• If you are an idiot and invest blindly, no amount of diversification will help you.
  
• It is very hard to find uncorrelated stocks. There is always some correlation. Remember how all fell in Mac last year due to Covid-19?
  

[ghchua]

Hi i4value,

One should not only look at diversification vs investment returns only. In fact, diversification helps in many other ways. For example, liquidity risk. A lot of stocks in the mid/small cap space on SGX are very thinly traded, with low bid/ask volume and wide spread. If you only hold a few of these stocks, chances are very likely that you will not be able to exit a position without selling down the price significantly.

Furthermore, with diversification, it allows one to receive dividends from his portfolio consistently from every stock for every month. This cannot be done with only a few stocks.

[i4value]

Given the risk and reward tradeoff, the challenge with diversification is how many stocks do you hold. Warren Buffet and Mohnish Pabrai holds a concentrate portfolio. I have about 30 to 40 stocks as I read somewhere that the benefits of diversification becomes marginal after this numbers.

[specuvestor]

There is such thing as diworsification when one focus on the number of stocks to fit the theory rather than understanding the stock

It's like a banker going into the streets to give loans betting on law of large numbers than try to understand the credit

[i4value]

You miss the point. I am a stock-picking value investor. So all my stocks in the portfolio were selected based on value investing principles. The question then is how many stocks should I have? This is where the portfolio performance studies come in. I target 30 to 40 selected stocks. I do not simply target 30 or 40 stocks. If I cannot find enough stocks, I stop. So there were times when I had only 20 stocks in my portfolio. The idea of diworsification is because the stocks in the portfolio are selected randomly.

[specuvestor]

If doing bottom up stock picking then diversification for uncorrelated return is probably not an important consideration. Generally it would be prudent for bottom up not to focus everything on say shipping or tech etc to reduce sector risk, rather than statistical correlation. So Buffett's portfolio though very tilted towards consumer related stocks cause he understands that business, or simply coincidental with US growth driver; it is also into financials, transport, energy and recently tech though most are still consumer driven ideas on bottom up basis.

[i4value]

I hope you realize that by not "... not to focus everything on say shipping or tech etc..." you are actually diversifying

[weijian]

If doing bottom up stock picking then diversification for uncorrelated return is probably not an important consideration. Generally it would be prudent for bottom up not to focus everything on say shipping or tech etc to reduce sector risk, rather than statistical correlation. So Buffett's portfolio though very tilted towards consumer related stocks cause he understands that business, or simply coincidental with US growth driver; it is also into financials, transport, energy and recently tech though most are still consumer driven ideas on bottom up basis.

Indeed, diversification looks to be the end result, rather than means to an end. As one gets more into investing....

(1) The more he understands, the more ignorant he/she realizes he/she is. And when he/she learns more as a result of his ignorance, about different business models, entrepreneur way of thinking etc....he/she gravitates towards having more diversification of stocks

(2) The more he/she observes, the more he/she realizes that every dog has its day. That is, markets work in cycles. So it might be better to "get lucky" than been skillful about been there (or not there) at the right time and right position sizing.

[EnSabahNur]

If doing bottom up stock picking then diversification for uncorrelated return is probably not an important consideration. Generally it would be prudent for bottom up not to focus everything on say shipping or tech etc to reduce sector risk, rather than statistical correlation. So Buffett's portfolio though very tilted towards consumer related stocks cause he understands that business, or simply coincidental with US growth driver; it is also into financials, transport, energy and recently tech though most are still consumer driven ideas on bottom up basis.

Indeed, diversification looks to be the end result, rather than means to an end. As one gets more into investing....

(1) The more he understands, the more ignorant he/she realizes he/she is. And when he/she learns more as a result of his ignorance, about different business models, entrepreneur way of thinking etc....he/she gravitates towards having more diversification of stocks

(2) The more he/she observes, the more he/she realizes that every dog has its day. That is, markets work in cycles. So it might be better to "get lucky" than been skillful about been there (or not there) at the right time and right position sizing.

Isn't diversification the means to an end result, which is risk control?
I don't consciously build my portfolio towards diversification, but I diversify because I am scared a wrong pick wipes out a chunk of my wealth.

[specuvestor]

Yes it's not for statistical correlation or law of large numbers or to hit 30 stocks portfolio but not be put all eggs in one basket. They are not the same. That's what I meant that Buffett's portfolio though very tilted to consumer /consumption it is still "diversified" in a common sensical way rather than diworsify to make the numbers or add negative correlation.

Last year and in many crisis shows that bonds and stocks correlation increased unexpectedly. LTCM and many times hedge funds or hedging strategies thought they are hedged but correlation broke down.

I hope you realize that by not "... not to focus everything on say shipping or tech etc..." you are actually diversifying