At the end of 2013, Grinnell College had an endowment of $1.5 billion. Not bad for a student body of 1,700 situated on a 120-acre campus in Iowa. The endowment has reached such a blessed size because the Trustees of the school had done two things intelligently in the 1960s: they bought shares of Berkshire Hathaway, and they made a very large investment in (what was then) the very small tech startup Intel [Grinnell College was the principal capital source for Intel in its formative years].
What I find interesting about the story is this: Even though the Grinnell Trustees were around the most successful compounders of the 20th century, they couldn’t resist the urge to sell once the investment had grown to be a very large component of the portfolio. If you visit a library and dig up old Grinnell University magazines on the microfilm, you will see them talk about how much they were thankful for Buffett’s leadership (he became a trustee of the school in 1968) for recommending Multimedia stock to the Board and purchasing AVCO’s television station (WDTN) in Dayton for $13 and then helping the school sell the station five years later for $50 million. Buffett’s association with the school led to a quadrupling of the endowment over the course of eight years, if you tally up the effects of the Berkshire stock and Buffett’s recommendations. The Board was also thankful for Bob Noyce, the son of an Iowa minister that founded Intel with Gordon Moore. Grinnell’s magazines and public releases about both Buffett and Noyce spoke glowingly about the two men.
And yet, despite this esteem, the Grinnell Trustees couldn’t resist relinquishing some of the stock in Berkshire and Intel that had appreciated significantly. A third of the Berkshire stock got sold, and the entirety of the Intel stock got sold during the 1970s. To their credit, though, they did put a good chunk of their endowment into the Sequoia Fund, which grew at 14% annually (I hope all of you reading this end up with “mistakes” that compound at 14% per year). Had they held on to their Intel stock alone, they would have had an additional $18 billion in their endowment. Quite a difference between that and the $1.5 billion that they have now. Figuring out the worth of the Berkshire stock is above my pay grade, as they bought and sold Berkshire throughout the years and the Sequoia Fund also had about 30% of its assets in Berkshire over the years, so it would be difficult to measure the accurate effect of Berkshire Hathaway upon Grinnell other than to say that it has been substantial, and would have been even more substantial had Berkshire been purchased and held rather than traded throughout the years.
I bring this up for more than mere storytelling—instead, I mention this to highlight the fact that creating wealth looks very different than preserving wealth. When you preserve wealth, you can funnel assets equally into 100 different sources comprising 1% of your portfolio because slogging forward at a 4-7% rate is perfectly fine because your goals with the money are to: (1) maintain a lifestyle, and (2) modestly increase it at a rate above inflation. When preservation is your goal, you purchase every capital allocation decision in the spirit of “How much can this investment possibly hurt me?” rather than “How much money can this possibly make me?”
When I study people who have made very significant wealth, there is a recurring theme that does show up but gets rarely discussed in the financial media: Concentrated bets are involved at some point, either because someone deliberately made a large bet or because some particular investment kept growing, growing, and growing. My guess as to why this rarely gets discussed is due to: (1) ignorance of the importance of that principal cash-generating machine in one’s wealth story or (2) well-intentioned paternalism that seeks to prevent people from putting years of their labor into one stock that they do not thoroughly understand and then losing it all. Look at people who invested significant sums of their money into GTAT Technologies, only to encounter an eventual bankruptcy: the misery is real and substantial.
I write this because even if you diversify across 20 or 25 stocks initially, there is a good chance that over a stretch of 15+ years something like Nike, Disney, or Visa could come to represent a disproportionate amount of your net worth. If you do decide to sell, it’s important for you to run the numbers ahead of time and figure out what you’re giving up if they continue to grow at 12-15% per year. The Pareto Principle talks about how 80% of the success comes from 20% of the participants in a group. If you have a twenty-stock portfolio that is held for a couple decades, it is likely that around four of them will be responsible for an outsized influence on your personal compounding.
That’s why I often prefer diversification through cash rather than selling—if Becton Dickinson becomes 20% of your portfolio, you can divert your BDX dividends elsewhere and pool your other dividends + surplus cash generated from your labor to water down the effects of Becton Dickinson by investing new money elsewhere. Although there does come a point at which you shift from wealth creation to wealth preservation, the conventional wisdom about diversification, diversification, diversification should incorporate some kind of cautionary word that advises against discarding those excellent compounders.