In a perfect world — assuming everything goes as planned — the best time to sell an investment is at its peak price. In reality, that’s called luck.
That’s why selling is an imperfect art. The overwhelming outcome for most sell decisions is too soon or too late.
Warren Buffett actually recognized it early on to never expect perfection:
Never count on making a good sale. Have the purchase price be so attractive that even a mediocre sale gives good results. The better sales will be the frosting on the cake.
It’s a byproduct of Ben Graham’s concept of a margin of safety. A wide margin of safety (between price and value) leaves room for error but you still make a profit.
Buffett put so much focus on buying right, that selling took care of itself. And if he got lucky selling near the top, so be it.
Of course, his problem back then was similar to a problem most investors have at some point. There’s never enough cash when you find a new idea.
Buffett explained how he dealt with those situations during an interview with Adam Smith (aka George Goodman) in 1998:
ADAM SMITH: What makes you sell a stock?
WARREN BUFFETT: It doesn’t happen too often, but if we think something like zeros last summer were more attractive than equities, we’re still very reluctant to sell anything but we may sell– trim a few holdings, maybe sell a few of the smaller holdings. We don’t like to disturb holdings in great businesses.
But when I started in this business I had way more ideas than money, so I had to sell things. Every time I’d get a new idea I’d get all excited and then I’d check my bank account and I didn’t have any money. So I had to sell something. And the best reason to sell a stock is because you find something that you like much better. But it’s not our natural style to sell.
ADAM SMITH: I understand that, but when you do sell something, what are the characteristics of the sale?
WARREN BUFFETT: Well, we’ll pick the thing– if I need money, we’ll pick whatever we like the least at the present price. And that’s not necessarily a negative comment about the business. Everything I’ve ever sold has gone up subsequently, and they should because they’re good businesses. But I’ll pick maybe the one I feel the least sure of where it’s going to be 10 years from now, not necessarily the one that has the least potential. I like certainties, so if I feel 100 percent sure about one company being worth far more money 10 years from now and 95 percent sure about the second, I’ll sell the second.
Buffett gives more weight to holdings that are less likely to fail. In other words, avoiding losses takes precedent over potential upside.
He had a similar exercise for position sizing during his partnership days. The company he’s most certain about gets the higher weighting in the portfolio while the least certain gets the lowest weighting and is likely the first to be sold.
Adam Smith’s Money Game 1998