The Role of Diversification in an Individual Stock Portfolio
by AAII Staff
It is frequently said that diversification is the only free lunch an investor will ever get, as this risk-reduction strategy doesn’t need to lead to subsequent reduction in return. Or does it?
Warren Buffett disagrees: "Diversification is a protection against ignorance. It makes little sense for those who know what they are doing."
For individual investors managing their own portfolio of individual stocks, both statements are correct.
In one extreme, stock investors often fail to diversify, holding just a handful of companies and subjecting themselves to unnecessary risk.
However, what a finance book will not tell you is that a portfolio consisting of just a handful of stocks also enormously impairs your ability to make rational decisions at the time when that ability is needed the most—under pressure. Managing this emotional reality is one of the more subjective aspects of risk management through diversification.
This article takes a short look at diversification and its role when managing a portfolio of individual stocks.
Don’t Bet the Farm!
The following happened to a good friend of mine—let’s call him Jack. He and his wife both worked for the largest insurance broker in the world, Marsh & McLennan. Over the years, Jack and his wife accumulated a large position of Marsh’s stock, which they were reluctant to sell.
Sometime in 2000 he asked me what I thought of their financial situation, having all this wealth in Marsh’s stock. I commented that although I didn’t see Marsh going out of business anytime soon, I would not recommend having all their net worth in one company. Although the probability of Marsh disappearing was very, very small, this couple’s lack of diversification was just not worth the risk, especially considering that both their personal income streams (paychecks) also came from Marsh.
Jack listened to my advice and agreed with it, but did not feel the urgency to do anything about it.
However, several years later, a major lawsuit was filed against Marsh by Eliot Spitzer (the then-state attorney general of New York) accusing the company of bid rigging and other malfeasances. Marsh’s stock was almost halved on the news, and talks about bankruptcy were in the air.
To my surprise Jack was very calm (considering that Marsh stock was his entire net worth at the time) when he called me to ask my thoughts on what he and his wife should do about their Marsh stock.
In this type of situation, you need to weigh the probabilities of possible outcomes. Bankruptcy, which was an improbable outcome for Marsh a day before the lawsuit was filed, suddenly became a lot more probable—or, at least the odds went from one in a gazillion to a remote but imaginable outcome.
If Marsh was just another stock (one of 15 or 20) in a diversified portfolio, the remote risk of its bankruptcy—the worst-case scenario—would be considered as one of the risks with appropriate attribution of probabilities to each outcome coming to fruition. But this is what theory doesn’t tell you: In the situation in which one cannot afford a low-probability outcome (and Jack could not afford it), one starts treating that outcome as having a much increased probability.
Jack was not diversified, and he did not have the luxury of looking at the worst-case Marsh scenario as just one of the low-probability outcomes, as it was a possible outcome whose consequences he could not afford.
After our conversation, Jack sold a good portion of his Marsh stock at a significant loss. I bet he’ll never look at diversification with the same indifference again.
Too Many Eggs or Too Many Baskets?
At the other extreme, investors who hold a large number of individual stocks incur another cost—ignorance. The dictionary defines ignorance as the condition of being uneducated, unaware, or uninformed—the cost Buffett is referring to. A very large number of companies in stock investors’ portfolios makes it impossible for them to know these companies well. Lack of knowledge leads to an inability to make rational decisions, which then causes investors to behave irrationally and hurt portfolio returns.
Another side effect of overdiversification is indifference to individual investment decisions. In a portfolio of a hundred stocks, an individual position might represent 1% of the total portfolio. The cost of being wrong is very small (if, for instance, the stock goes up or goes down 20%, the overall impact on the portfolio is only 0.20% on either side), but so is the benefit of being right. This breeds a semi-indifference to incremental decisions common among overdiversified buy-and-hold investors.
You need to strike an appropriate balance, weighing the consequences of either extreme. Academics disagree on the exact number of uncorrelated stocks needed in a portfolio to eradicate individual stock risk, but the number is usually given as somewhere between 16 and 25 stocks. This is another case where being vaguely right is better than being precisely wrong. I found that a portfolio of around 20 stocks is manageable and provides an adequate level of diversification; at this level, the price of being wrong is not too high, but every decision matters.
Taking diversification a step further, stress testing a portfolio (playing out different what-if scenarios) for probable risks coming to fruition is critical, as it exposes the weaknesses of the portfolio in the event possibility turns into reality.
Mental Accounting and Randomness
Behavioral finance has identified a mistake that is one of the most frequently visited potholes in investing: the mental accounting trap. That is, investors tend to segment wealth into mental accounts: stocks, real estate, bonds, and business. They then evaluate the short-term performance of each asset class in isolation and fail to notice their interaction in the context of the total portfolio. So, even though an investor may be diversified among loosely correlated asset classes, their entrapment in a mental accounting error precludes reaping the fruits that diversification, provided properly, has given them.
The mental accounting trap can also occur at the individual stock level. It is important to be diligent in your analysis, but it is more important to accept that you just will not be right all the time on every stock. Even if your analysis and investment process was right on the money, unpredictable (random) events may turn what was supposed to be a good investment into a loser. Over time investors make mistakes—that is a reality of investing. Even Warren Buffett, whom many consider the god of investing, occasionally loses money on individual stocks. Ironically, the original Berkshire Hathaway, the textile company bought in 1962 that Buffett later turned into a vehicle for his investments, was an investment that went bad. Buffett bought Berkshire Hathaway on the cheap, but the fundamentals of the textile business were deteriorating at a faster rate than he estimated.
There is a reason for a diversified portfolio of stocks—to allow room for losers (though not too many). When we make mistakes, we should try to learn as much as possible from them (assuming there is something to learn) and move on. If we let a mistake drive us crazy, the rest of the portfolio will suffer, as it will obscure our judgment. In my personal experience, it is an active process to force myself to concentrate on the overall portfolio—not something that comes naturally for most of us, myself included, since we analyze each stock individually before they make it into the portfolio.
I encourage you to pay close attention to fundamentals and not be overly sensitive to individual short-term stock price action, as more often than not it is a manifestation of random noise.
Nassim Taleb in the book “Fooled by Randomness” provides a great example. Using a Monte Carlo simulation engine, he created a hypothetical portfolio that goes up 15% a year, with a volatility (variation of returns from the average) of 10%—a risk-return combination most investors would kill for. Here’s his probability projection for making money over different time horizons, based on statistical probabilities:1 year 93%; 1 quarter 77%; 1 month 67%; 1 day 54%; 1 hour 51.3%.
If it were not for volatility, the investor in this portfolio would observe no “noise” (short-term return fluctuations), as the portfolio would go up by 0.0383% a day (or 1.17% a month, 3.55% a quarter, or 15% a year). However, the 10% volatility that was assumed sprays some real-world uncertainty into the portfolio—introducing “noise” to the mix.
The amount of noise observed by the investor increases with the shortening of the observed time period. So, even though an investor holding the portfolio described has a 93% chance of making money in any given full year, looking at the portfolio on a more frequent basis—let’s say every hour—the investor would observe the portfolio making money only 51.3% of the time—a tiny 1.3 percent advantage (the difference between 51.3% and 50%) over breaking even or losing money.
In other words, even though over the course of an entire year this investor has only a 7% chance of seeing an overall loss in the portfolio, the same portfolio observed on an hourly basis would disappoint the investor with losses 48.7% of the time. In fact, that small 1.3% edge for making money on an hourly basis will probably not be noticed by our data-overloaded investor, as the pain of losses has a greater negative emotional utility than the joy of gains.
While this example is right on the money, investors building their own portfolios of individual stocks will actually observe more noise than in this example, because individual stocks are more volatile than a portfolio of stocks. If you focus on the performance of individual stocks on a short-term basis, rather than on the overall portfolio, you will observe even more short-term volatility.
The lesson: Don’t act on noise! Focus on the stock in the context of the total portfolio, and focus more on whether the company is still on track, rather than on daily or weekly share prices.
A properly diversified equity portfolio should consist of stocks from different industries, of various sizes (from large capitalization to small capitalization), growth rates, valuations, and countries. At times the market will fancy a certain characteristic over another, which is what markets do and is at least in part why investors diversify: to lower overall volatility. I say ‘‘in part’’ because there are two more reasons: first, and the most important, to limit exposure to the true risk—a permanent loss of capital; second, to be able to maintain a rational state of mind. As mentioned earlier, a portfolio should have a small enough number of stocks that every decision matters, but not so few that the cost of being wrong in a stock is unbearable.
All that having been said, don’t make marginal (buy) decisions for the sake of diversification.
Randomness Can Be Your Friend
Randomness can work to your advantage, as it may at times drive stocks you own above their intrinsic values, providing you with an opportunity to sell earlier than you originally expected at the time of purchase.
It may also drive totally fine stocks below their intrinsic values, providing an opportunity to buy more.
I sometimes find myself guilty of missing this unique opportunity. When I look for new buying opportunities, my natural tendency is to start looking outside of my portfolio first. But this makes me neglect to formally consider buying more of stocks that I have already researched, even though I still believe the fundamentals are intact but for some random reason the stock has declined.
Perhaps a good check and balance on this common tendency to ignore holdings when looking for buy candidates would be to pretend I was 100% in cash and had the entire universe of eligible attractive stocks from which to choose.