Do Not Diversify For The Sake Of Diversification - Corrected
In my previous article about diversification, JC definitely put me in my spot. I was wrong when I compared the S & P 500 to the Russell 2000 in an attempt to prove my "theory" that overdiversification results in poorer results.
As JC rightly pointed out, the S & P 500 is NOT diversified. To be exact, the S & P 500 is not adequately diversified. The S & P 500 is diversified in terms of industry but not in terms of market capitalization, which I completely neglected in my previous article. This short article explains it really well.
The whole aim of diversification is to spread your investments into different vehicles that, ideally, have negative correlations. In other words, the stocks should move in opposite directions of each other.
Of course, finding a pair of stocks with negative correlation is a formidable task. Suppose you found a pair of stocks — symbols ABC and XYZ — that have ideal negative correlation and one of them, ABC, is currently trading at a bargain. If they were to move in opposite directions, wouldn't it follow that the counterpart, XYZ, would be trading at a sky-high price? If this is the case, the only time XYZ is worth buying is when ABC is trading at a sky-high price. The problem now becomes until that happens you will be exposed to the unsystematic risk of owning only ABC. My point is it is impossible to find both stocks on opposite ends of the spectrum selling at bargain prices at the same time.
A problem I find with owning stocks with negative correlations is to take a stand. Let me explain. Assume you own two stocks ABC and XYZ. For simplicity, ABC sells only Windows whereas XYZ sells only Linux. Some of you may think this is not diversified. If this is diversified, XYZ will sell rocks or something other than computer operating systems. True, but if XYZ sells rocks, they won't be moving in opposite directions now will they? Meaning if ABC tanks and RST, who sells Linux, skyrockets, you will be missing out on RST although you may still have your rocks. So, for the sake of making an argument I want to win, we'll assume XYZ sells Linux. Now, you are faced with a problem: Do you believe that Windows will prevail over Linux? If you bought both ABC and XYZ, you're essentially saying "Neither." If so, why buy any of them in the first place?
Another mistake I made in my previous article was assuming diversification improves returns. Diversification is not meant to improve returns although many mutual funds have used the term to explain improved returns. Diversification is meant to reduce risk, not necessarily improve returns. The goal is to maintain the same return for reduced risks.
Nonetheless, most of the time, taking on reduced risks means accepting less rewards. It is possible to take on more risk than the expected return and vice versa. The former, if you will agree, is a move we will want to steer clear as much as possible.
Diversification is good because it dilutes the effect when one or two stocks in your portfolio takes a nosedive. Because your stakes in those stocks are small, they don't affect your entire portfolio much. But what most people fail to realize is diversification hinders your portfolio performance exactly the same way when one or two stocks in your portfolio skyrockets. So when you diversify, you are not only limiting your losses but also limiting your winnings.
David, another reader, partially agrees with me. David recommends holding 6 stocks to hit the sweet spot in diversification. Perhaps this formula worked well for him. I don't have enough experience to know how many stocks to hold. I'll leave it up to you to find out. Holding too many stocks for an individual investor may eliminate the benefit of diversity. When you own more stocks than you can follow, you are taking on risks unnecessarily even if your portfolio is diversified.
Obviously, diversification is not restricted to just stocks. Diversification always involves other investment vehicles including bonds, real estate, mutual funds and so on.
As I have said earlier, diversification is a necessary evil / good (whichever you fancy) to a certain extent. No one knows how much diversification one needs in order to achieve that perfect balance of risk and reward, except yourself.
I'm new to investing and I may be acting idiotically. At this stage of my life, I don't pay attention to diversity. I worry if I focused too much on diversification, I might miss out on some multi-baggers. If I happen to find a stock on sale, I buy it. I do not worry if my entire portfolio consists of technology companies. I would rather take my risks with companies I like than companies I hate simply for the sake of diversification. Buffett recently purchased both Home Depot and Lowes. Was he thinking of diversification when he bought them? I doubt it.
• VirtualBank gives you up to 5.13% APY for your savings!
• Blue Cash from American Express: 5% Back on Gas, Up to 5% cash back on eligible purchases, No Annual Fee, 0% Introductory APR for 6 months, Pay over time
• Try Starwood Preferred Guest® Credit Card from American Express®. Great rewards and first year fee free!
• Bad Credit Home Loans
• Life Insurance in 10 minutes if qualified. No medical exam.
After reading 1stmill's post about why he doesn't buy or sell individual stocks, I have newfound respect for him. He humbly admitted his shortcomings in picking stocks on his own. Despite playing down his knowledge and experience in investing, I'd say he's more successful than ... Read
I have recently started to explore the intimidating prospects of saving for college. I want to be prepared to send my kid(s) to a good college. For now, let's assume I intend to have only one kid. I will not have more unless I can ... Read
Many new investors, which I consider myself to be, have been poisoned by the diversify-or-die mentality. I'm not sure who started this crap, but my incriminatory finger is itching to point towards the mutual fund managers. Here's my theory about mutual fund manages... Mutual fund ... Read
I have a confession to make: I am a stalker. Yep, there you have it. I stalk people, in particular investing gurus. Well, to be fair, I don't stalk the gurus in person. I just follow their investing closely, very closely. One of my favorite ... Read
After I posted my scathing comments on the previous article, I realized that some of those criticisms might be unfair. I didn't realize I was dealing a stockpicker. I thought we were talking about risk-adjusted returns on an overall portfolio. While the whole idea of merely six stocks being adequate diversification is silly to academic finance and professional portfolio managers, it may make perfect sense to a stockpicker who ignores risk in search of "multi-baggers". In the end, it's your money. Have fun. Pick away. But realize you're playing what Charles Ellis called "The Loser's Game" in his seminal work by the same title.
Cheers,
JC
I would disagree with the argumetn that you need 30+ stocks to be diversified, considering the transaction costs can often be substantial. The following article on diversification may give you more insight about the issue http://www.rmi.gsu.edu/FSR/abstracts/Vol2_2/v2-2a1.pdf
To the contrary David, you're thinking like a stockpicker again. Good luck with that. More insight? Consider the alternative:
http://business.scu.edu/faculty/research/working_papers/pdf/statman_wp06_paper.pdf
Here's an excerpt: "Lack of diversification is costly. Investors who hold only 4 stocks in their portfolios forego the equivalent of a 3.3% annual return relative to investors who hold the 3,444 stocks of the Vanguard Total Index Stock Market Index fund. Why do investors forego the benefits of diversification?"
Another good paper: http://kuznets.fas.harvard.edu/~campbell/papers/clmx.pdf
This paper shows that while several decades earlier, it may have been true that adequate diversification could be had with 15-20 stocks, it is no longer true. The paper finds that, while volatility of the stock market as a whole has remained relatively steady, the volatility of individual stocks has increased. This implies that correlations between stocks have decreased, which is confirmed by the paper's empirical results. Thus, the benefits of diversification have increased, along with the need to hold more stocks to achieve those benefits. This confirms the prudence of holding highly diversified mutual funds instead of individual stocks.
More? http://www.ppca-inc.com/Articles/DiversByNumbers.pdf
This paper challenges the conventional wisdom that a randomly chosen portfolio of 15-20 stocks gives nearly all the diversification benefit of the market. "Fifteen-stock portfolios, on average, achieve only 75%-80% of available diversification, not the 90%-plus typically believed. Even 60-stock portfolios achieve less than 90% of full diversification." This confirms the prudence of avoiding individual stocks in favor of highly diversified mutual funds.
Another piece of information to throw out here is that diversification is actually more effective in reducing portfolio VOLATILITY than reducing overall portfolio risk. Risk and volatility obviously have a close relationship, but the point I am trying to make here is that in the origins of Modern Portfolio Theory, the concept of "Alpha" was originally developed as a measure of RISK taken on by a given portfolio. Through subsequent studies by academia (Fama & French I belive in this case) it has been determined (albeit not PROVEN) that "Alpha" is actually a better measure of portfolio VOLATILITY than it is of RISK. However, the consensus remains that Alpha can be reduced by portfolio diversification, and if volatility can be reduced for a given level of risk, this is beneficial to said portfolio, thereby validating Alpha as a relevant statistical measure for Modern Portfolio Theory.
Finaly, to speak directly to Fatboy's post, for an individual investor who is in the process of building a portfolio of individual stocks over time, it is best to think of diversification in terms of industry classes. For example, if you already 100 shares of Hewlett Packard, your next individual stock purchase should NOT be Dell! Buy a company that you think is the best pick in manufacturing or consumer cyclicals or whatever, just try to make your next pick from a sector that is far away from your current holdings so you can hope to increase your level of diversity over time.
The bottom line is that diversification has been PROVEN to be a good thing for a given portfolio. But Fatboy is making a correct point here in that too much of any good thing is bad. This is more easily quantified in the example of the popular mutual fund that diversifies to the point of being a glorified S&P 500 index fund due to large inflows of capital, at which point the fund will begin to lag the index by the amount of the expense ratio.
