Over the next few weeks, we’ll provide some answers and insights to this question, as well as show why holding such positions is almost always imprudent speculation, unless you have a very high marginal utility of wealth, and are fully prepared to accept the possibility, if not likelihood, of a highly negative outcome.

While we’ll focus on the issue of concentrated positions in individual stocks, the very same issues apply to mutual funds. For example, while investors can own a total stock market fund, such as Vanguard’s Total Stock Market Fund (VTSMX), many will invest in funds that own as few as 20 or 30 stocks, such as the Ariel Focus Fund (ARFFX). The issue of failing to diversify extends to mutual fund investors who fail to diversify geographically and limit their holdings to domestic mutual funds, or maintain only a small allocation to international stocks. And the same issue of failing to efficiently diversify also extends to those who invest in sector mutual funds.

We’ll begin by discussing the concept of compensated vs. uncompensated risks.

Equity investors face several types of risk, which is true of any risky asset, be it a stock or bond. First, there is the idiosyncratic risk of investing in stocks. This risk cannot be diversified away, no matter how many stocks, sector funds, or different asset classes you own. That’s why the market provides an equity risk premium.

Second, various asset classes carry different risk levels. Large-cap stocks are less risky than small-cap stocks and glamour (growth) stocks are less risky than distressed (value) stocks, at least in terms of classical economic theory. These two risks, size and value, also cannot be diversified away. Thus, investors must be compensated for taking them. This is the reason that the small stock and value stock premiums exist.

Another type of equity risk is the risk associated with an individual company. The risks of individual stock ownership can easily be diversified away by owning a passive asset class or index fund that basically contains all the stocks in an entire asset class or index. Asset class risk can be addressed by the building of a globally diversified portfolio, allocating funds across various asset classes (domestic and international, large and small, value and growth, and even real estate and emerging markets).

Because the risk of single-stock ownership can be diversified away, the market doesn’t compensate investors for assuming this type of (unsystematic) risk. And because the risk can be diversified away without lowering expected returns, why so many investors hold concentrated portfolios remains a puzzle.

• Confusing the familiar with the safe. Familiarity breeds overconfidence, leading to an illusion of safety. In contrast, the lack of familiarity breeds the perception of high risk. Overconfidence also leads to underestimating downside risks.
• Employees are often overconfident regarding the outlook for their own firm. Their familiarity leads to over-investment. In the same way familiarity leads to the concentration of assets in an investor’s home country (referred to as the home country bias, a global phenomenon), it also leads to over-investment in the stock of his or her employer.
• Investors with large gains, which can create a concentrated position, make the mistake of believing that they are playing with the house’s money. Here’s my favorite example of this phenomenon: a good friend of mine was either lucky or smart enough to buy Cisco at $5 per share. At the time, the stock represented a relatively small portion of his net worth. When the stock reached $80, his position at Cisco had become a substantial portion of his portfolio. I asked if he would buy any Cisco stock at the current price, and he said he wouldn’t. I then pointed out that if he wouldn’t buy any, he must believe that it was either too highly valued or he was currently holding too much of the stock and it was too risky to have that many of his eggs in one basket. Despite the logic of the argument, my friend, one of the smartest people I have known, steadfastly refused to sell some of his shares for the following reason: His cost was only $5, and the stock would have to drop about 95 percent before he would have a loss. I pointed out that this was the same mistake gamblers make when they’re ahead at the casinos and keep playing because their gains are the “house’s money.” Of course it’s not the house’s money. Having won it, it’s their money. And it’s a behavioral error, a result of a “framing problem,” to believe otherwise. A few months later, Cisco’s stock had hit $13, and my friend was still holding it.
• Treating the likely as certain and the unlikely as impossible. Despite even relatively recent examples (such as Enron, WorldCom, Lehman Brothers and Bear Stearns) investors have a tendency to think that disasters can happen to other people and other companies, but not to them.

Another cause for the failure to diversify is one we can call “rearview mirror investing.” The study Excessive Extrapolation and the Allocation of 401(k) Accounts to Company Stock found that strong past performance of an employer’s stock leads to overconfidence with respect to its future performance. Past performance was simply extrapolated into the future. However, great past performance usually results in high valuations. Thus, not surprisingly, participants who over-weighted their employer’s stock based on past performance earned below average returns.

Next week, we’ll continue our discussion about the perils of concentrated positions with a look at yet another cause for failing to diversify. It relates to what is often called the “endowment effect,” where an individual values something they already own more than something they don’t own yet.