When to Sell a Stock: The Four Rules Institutions Use

Warren Buffett is the global symbol of buying and holding forever.

Here is what he actually did in 2024, from Berkshire Hathaway’s own annual report:

Berkshire Hathaway, 2024
Equity securities purchased	$9.2 billion
Equity securities sold	$143.4 billion

One year. Fifteen dollars sold for every dollar bought.

This is not hypocrisy. It is the point.

Buffett has exit rules. You have hope.

His record is a list of exits

IBM. He bought in 2011 — $10.7 billion, 64 million shares at around $170. He held it for years. Over the period he held it, IBM fell about 18% while the S&P 500 rose about 116%.

Then he sold it. All of it.

Airlines. He bought in 2016 and 2017. He sold every share in 2020, and said it in plain English:

“That was my mistake.”

Five words that most retail investors cannot say about a single position they own.

Why “hold forever” sounds so convincing

Because you only ever hear from the survivors.

Apple. Amazon. Coca-Cola. These are the stories people tell at dinner parties, and they are true.

Nokia. Kodak. Enron. Wirecard. Lehman Brothers. General Electric.

Nobody tells those stories about themselves. There is no YouTube video called “How I held Wirecard to zero.”

The people who held forever and were right are visible. The people who held forever and were destroyed are silent — and there are considerably more of them.

Survivorship bias is not a footnote here. It is the entire reason the strategy sounds safe.

And there is a version of failure that never announces itself

It is called thesis creep, and it is how intelligent people end up holding a corpse.

Watch how it works. It does not feel like a mistake at any point.

Step 1. You buy because revenue is growing 20% a year. That is your thesis. Good.

Step 2. Growth falls to 5%. You do not sell. You say: “Well, at least the dividend is solid.”

Step 3. The dividend is cut. You do not sell. You say: “The new CEO has a turnaround plan.”

Step 4. The board fires the new CEO. You do not sell. You say: “It’s a great company. It’ll come back.”

At no point did you decide to hold a declining business.

You just replaced the reason. Four times. Without noticing.

Every new reason to hold is a red flag you have talked yourself past.

The institutional rule is unsentimental: when the original thesis breaks, the default action is to sell. You do not invent a new story to justify an old decision.

The autopsy: General Electric

The perfect case, because nobody saw it coming and everybody could have.

GE was in the Dow Jones for over a century. It was sold to retirees as the safest thing a person could own — a stock for widows and orphans.

Then:

13 November 2017 — GE halves its dividend, from 24 cents to 12. It is only the second cut since the Great Depression.

That is thesis break number one. It is loud, it is dated, and it is public.

June 2018 — GE is removed from the Dow Jones Industrial Average.

October 2018 — the dividend is cut again. To one cent.

A simple rule — a dividend cut without a credible plan is a thesis breach — would have got you out in November 2017, before the worst of it.

Instead, most holders waited. Because it was GE.

”But it recovered”

It did. And this is the trap inside the trap.

Hold through all of it, and after eight years you are up meaningfully.

Over the same period, the S&P 500 returned roughly twice as much.

You endured eight years of genuine distress — dividend cuts, ejection from the Dow, the endless question of whether to sell — to underperform an index fund you could have bought and then ignored.

And the company you bought does not exist. GE had to split itself into three separate businesses to unlock any value at all. You are not holding General Electric. You are holding one division of a dismantled conglomerate.

That is not buy and hold. That is buy and hope.

Why your brain will not let you sell

This is not a character flaw. It is a documented, measured, replicated bias — and knowing its name is the first step to overriding it.

Terrance Odean analysed 10,000 real brokerage accounts — actual people, actual money — and published the result in the Journal of Finance.

Investors were 1.5 to 2 times more likely to sell a winner than a loser.

And here is the part that closes off every comfortable excuse: this was not explained by tax planning, not by rebalancing, and not by trading costs. Odean checked. It is not rational. It is just what people do.

It has a name — the disposition effect — and it is exactly backwards from what works:

You cut your winners short and let your losers run.

The reason is loss aversion

Kahneman and Tversky showed that the pain of a loss is roughly twice as intense as the pleasure of an equivalent gain.

So you sell Apple at +20% to “lock in the gain.” It feels like a decision. It feels responsible.

And you hold GE at −50%, because selling would make the loss real — and while you still hold it, it is somehow still just a number on a screen.

Professionals do not solve this by being braver. They solve it by writing the rule down before the emotion exists.

The four triggers

Every serious fund runs an Investment Policy Statement — the written constitution of the portfolio. It does not only say what may be bought. It says when a position must be sold.

And for a fiduciary this is not a preference. Holding a broken position on hope, with somebody else’s money, is a failure of duty.

Here are the four conditions that force an exit.

1. Thesis breach

The reason you bought is no longer true. Exit. Immediately.

A dividend cut with no credible plan. A regulatory change that removes the moat. The founder leaving. Whatever your thesis rested on, gone.

You do not renegotiate with yourself. You leave.

2. Valuation extreme

The price has run far beyond what the business is worth. Take the profit.

This one is harder, because it feels like abandoning a winner. It is not. It is recognising that you are now being paid for a future that has to be perfect.

3. Fundamental deterioration

The numbers are getting worse, not better.

Margins compressing. Return on capital falling. Two quarters of free cash flow below plan. Net debt to EBITDA climbing through your threshold.

The business is telling you something. Listen to it before the price does.

4. Opportunity cost

The capital would work harder somewhere else.

Capital is finite. Every day you hold a dead position, you are choosing it over every other opportunity in the entire market.

That is a decision. It just does not feel like one, because you are not doing anything.

And look at what is missing

• “It will come back.”
• “It’s already down so much.”
• “I can’t sell at a loss.”
• “I’ve held it this long.”

None of these are on the list. They are not investment reasons. They are feelings about your entry price — a number that exists only in your head and has no bearing whatsoever on what the asset is worth today.

To be completely fair: buy and hold does work

For a broad index.

On a total-return basis, the S&P 500 has not lost money over any twenty-year holding period in its history.

But understand why, because the reason is the whole argument.

An index is not passive about its holdings. It continuously ejects the companies that fail and admits the ones that succeed. Enron left. Kodak left. GE left.

The index sells your losers for you. Automatically. Without asking. Without your permission and without your feelings getting involved.

A single stock has nobody doing that.

If you hold a single stock forever, you have appointed yourself the person who never sells — including on the day you should.

The one question

If you own something today and are not sure whether to sell it, there is a single question that cuts through all of it. It is uncomfortable, which is how you know it is the right one.

Knowing everything you know now — would you buy this, today, at this price?

If the answer is no, then you are not an investor in that company.

You are a hostage to your entry price.

And the market does not know what you paid. It has never known. It will never care.

The practical structure

Core: a broad index fund. It does not depend on you being right about anything, and it quietly does the selling for you.

Satellite: individual positions, if you genuinely want them — with a fixed size and a written exit plan, decided before you buy, while you can still think clearly.

One page. Why I bought this. What would have to be true for me to sell it.

Write it when it is boring. You will need it when it is not.

Loyalty to a stock is expensive. Loyalty to a process pays you.

Educational content only — not investment advice, and not a personal recommendation. Companies named here are used as historical case studies, not as views on their present value. Speak to a qualified, licensed professional before acting.