Buying is the decision everyone studies. The two that quietly decide how you actually do are how much you put into one company, and when you sell. Get sizing wrong and a single mistake can sink you. Sell for the wrong reasons and you hand back the gains the analysis earned.
How much to own
Source: Graham; Buffett; Lynch
The first rule of sizing is to survive. Never put so much into one company that being wrong about it would change your life. Businesses you were sure about still fail, frauds still surface, industries still get upended. Size every position so that a total loss there is a painful lesson, not a catastrophe.
Size to survive first, and to compound second. A position large enough to ruin you is too large, however good the idea.
Beyond that, how concentrated to be is a real choice, and the great investors land in different places. Buffett holds a handful of companies he understands deeply, and argues that wide diversification is protection against ignorance: if you know what you are doing, owning your best few ideas beats diluting them with your fortieth-best. Lynch, running a huge fund, owned hundreds. The honest answer for most people is in between. Own few enough that you can actually follow each business and have real conviction, and enough that one blow-up cannot wreck you. Owning forty names you cannot explain is not diversification, it is confusion.
Two hard limits sit underneath all of this. Money you will need within a few years does not belong in stocks at all, because you cannot control when the market will be down. And borrowing to invest turns a normal, survivable drop into a permanent loss, because the loan does not wait for the recovery.
When to sell
Source: Buffett; Lynch
A good sale is driven by the business or the price relative to value, never by the price alone. There are really only a few good reasons to sell:
- The thesis is broken. The reason you bought it has stopped being true: the moat is eroding, debt is piling up, the numbers are deteriorating year after year.
- You were wrong. Honest re-reading shows the business is not what you thought. Admit it and move on.
- You have found something clearly better. Capital is limited, so a much stronger opportunity can justify selling a weaker holding.
- It has become wildly overvalued. Not just a little rich, but priced so far above any sensible estimate of worth that the future return is poor no matter how good the business.
- It has grown too large. A winner can become such a big share of your portfolio that prudence alone says trim it back.
Let your winners run
Source: Lynch; Buffett
The most expensive habit in investing is selling the great compounders too early. Lynch called it pulling the flowers and watering the weeds: investors snip their winners to bank a quick gain and cling to their losers hoping to break even, which is exactly backwards. A wonderful business that keeps getting more valuable is the thing you were trying to find. Trading it away for the small satisfaction of a realised gain, and a tax bill, usually means swapping a proven compounder for a worse idea or for cash.
The bar for selling a great business should be high: the thesis genuinely broke, or the price became absurd. Short of that, time in a compounder is the whole point. The next chapter is about putting all of this into practice: how to actually start.