When to sell a dividend stock, and why the price is a bad adviser

By Ryan Wimalasena · 21 September 2026 · 6 min read

The question usually comes up at one of two moments: when a holding is down 30%, or when it's up 80%. They feel like opposite problems. They're really the same one, because in both cases you're asking the share price to tell you something it doesn't know.

There are four decent reasons to sell a dividend stock. The reason you bought it has stopped being true. It has grown into too big a part of your portfolio. You need the money. Or you've found something that does the same job better, and you understand it at least as well. The price going up or down isn't on that list, although now and then it's a clue about the first one.

Why the price is a bad adviser

You bought for a reason. Maybe it was the income, or a business you understood, or a hole in your income calendar that this company happened to fill. The price moves for plenty of reasons that have nothing to do with yours: interest rates, a big fund rebalancing, a sector going out of fashion, a rough night on Wall Street. Sell on the price alone and you've handed the decision to a crowd of strangers acting on reasons of their own.

The research on this is fairly brutal. Brad Barber and Terrance Odean studied more than sixty thousand household brokerage accounts in the US during the 1990s, and the fifth of households that traded most trailed the market by about six and a half percentage points a year. Their stock picks were about as good as anyone's. What sank them was the cost of making so many of them.

In a separate study, Odean found that the shares people sold went on to do better over the next year than the shares they bought to replace them. That one stings, because every swap feels like an upgrade when you make it.

The most common version of the mistake even has a name, the disposition effect. People sell their winners too early and hang on to their losers too long, because selling a winner feels like banking a gain and selling a loser feels like admitting you were wrong. Let that run for a few years and a portfolio slowly fills up with its owner's mistakes. These are old American studies, but the same habit has turned up in other markets since, and I wouldn't assume Australians are immune.

1. The reason you bought it has stopped being true

This is the big one, and it's the only one the price can sometimes hint at. If you bought for the income and the board has cut the dividend because the business has changed for good, the case you bought on is gone, whatever the price does next.

A cut that pays for something sensible, like an acquisition the company can actually explain, is a different animal. The announcement usually says which kind it is, and the piece on dividend cuts goes through how to tell them apart.

2. It has grown too big

A nice problem to have, but still a problem. If one holding has grown into a much bigger slice of your portfolio than you'd ever have chosen to buy, trimming it is housekeeping. You're deciding how much of your money should ride on one company, and saying nothing at all about the company itself. It usually feels wrong while you're doing it. Position sizing has its own piece.

3. You need the money

The most sensible reason there is, and the one investing articles tend to skip because it isn't clever. A house deposit, retirement, a life that needs paying for. Selling for this is exactly what the money was for, and nobody should make you feel sheepish about it.

4. You've found something that does the job better

And you understand it at least as well as what you already own. That second part is the real test. Odean's result says the replacement usually does worse, so the new idea should have to earn its place before anything gets sold to make room for it.

Write down why you bought

Memory is a terrible record keeper. A holding that fell turns into something you always thought was risky, and one that rose becomes something you knew about all along. Your original reason quietly rewrites itself to match the price, so the price ends up making the decision anyway.

The fix is two sentences, dated, written when you buy. First, what has to be true for this to work. Second, what would tell you it hasn't. That second sentence is your sell condition, written by a calmer version of you who hadn't seen the price move yet.

Dividend Guardian has a journal for exactly this. You can add a dated private note in your own words and tie it to the holding it's about, and it sits in the same record as every dividend change and declared payment Guardian picks up. When something changes, your reason for buying is right there next to it.

In Australia, the timing can matter as much as the decision

If you've held an asset for at least twelve months, you can generally halve the capital gain before it's taxed. So selling at eleven months instead of thirteen can cost you more than a year of dividends. And if you reinvest, each new parcel has its own purchase date, which means the twelve month mark falls at different times within the same holding.

Every May and June there's also a wave of people selling to lock in losses before the financial year ends. They're selling for the calendar, and the company usually has nothing to do with it. The ATO has warned for years that selling and buying straight back, mainly to create a loss, can be treated as a wash sale and the loss disallowed.

So tax is a good reason to think about when you sell, and rarely a good reason to sell at all.

Watch the payment instead of the price

If you bought for income, your reason leaves a public trail, and that trail is the dividend. The board sets it knowing people are watching, it's announced formally, and it often moves before the market accepts that anything is wrong, sometimes in a sentence nobody bothers to quote. Watching the payment means watching your own reason. Watching the price means watching everyone else's.

That's the idea Dividend Guardian is built around. It keeps an eye on the dividends and declared payments for the companies you own and the ones you're watching, and when a company confirms a change, it tells you what that does to your income for the year. It won't ping you when the share price moves. That's deliberate, because a price alert mostly invites you to react to the least useful number on the screen. The monthly summary only goes out when something actually happened, so a quiet month means no email.

Most weeks, the right move is to do nothing at all. That's a lot easier to stick to when you know you'll hear about it if the dividend changes.

Dividend Guardian provides information and estimates from public data, not personal financial advice. Nothing here is a recommendation to buy, hold or sell anything, including anything you own. It's about how to make the decision, not what the decision should be. The studies mentioned are American and historical. Capital gains tax, the twelve month discount and wash sales depend on your own circumstances, so check them with the ATO or a registered tax agent. Guardian reports confirmed changes; it does not predict them.