John Maynard Keynes was probably the most influential economist of his time. He also managed the portfolio of an insurance company and the endowment of Kings College. In the 1940s, a debate raged between the ideas of broad portfolio diversification and portfolio concentration. Keynes was an advocate for concentration. In January of 1942, he purchased a large position in Elder Dempster for The Provincial Insurance Company. F. C. Scott of Provincial questioned why such a large position was purchased, Keynes replied,

"...that I preferred one investment about which I had sufficient information to form a judgment to ten securities about which I know little or nothing."

Sounds reasonable? Keynes' response illustrates the cognitive resource limitations of learning about the many firms needed to be diversified and possibly hints at a common heuristic adaptation: familiarity. In short, people prefer those things that are familiar to them.

Later, Harry Markowitz would present the mathematical case for diversification, win the Nobel Prize for Economics for his Modern Portfolio Theory, and effectively end the debate for scholars and investment professionals. Diversification allows investors to earn the highest expected return for the level of risk desired.

However, the behavior of the individual investor tells a different story. The portfolios formed by individual investors suggest that they hold surprisingly under-diversified stock portfolios and invest too large a portion of their retirement savings in their employer's stock. Even though both scholars and the media have advocated the benefits of diversification for decades, it seems absurd that actual portfolios have three (median) stocks or four (average) stocks.

Why do individual investors fail to diversify?

See Polkovnichenko, Valery, 2005, Household portfolio diversification: A case for rank-dependent preferences, Review of Financial Studies 18, 1467-1502.