When Do I Decide to Sell a Stock?

On the whole, my years as a deep value investor were an unmitigated disaster. But at least I made a few smart moves.

Like the time I bought Intertape Polymer shares during the depths of the financial crisis. A big chunk of the company’s revenues came from selling plastic wrap put around new houses, a part of the business that collapsed in 2007. I purchased shares in 2008 for $2 and change.

The company survived the Great Recession and started to recover. It took a few years but eventually I was sitting on a substantial profit, so I sold half of my position at $7.12 in 2012. The company continued to recover so I sold the rest of my position for $12.23 per share in May, 2013.

Proud of my astute investing skillz, I promptly moved onto the next stock, pretty much forgetting about Intertape forever. It continued to execute behind the scenes, buoyed by a rapidly recovering North American housing market.

The company was then acquired in 2022 for $40.50 per share. Oops.

This creates a fascinating discussion. Was my Intertape Polymer investment successful?

The easy argument is yes, of course it was. I had something like a 400% return on my stock, plus I got a couple of dividends when it resumed the quarterly payout in 2012. That’s a terrific return over just a few years. And, as the old saying goes, “nobody ever went broke taking a profit.”

But the more experienced I get at this whole investing thing, the more I think that it was a failure. Now don’t get me wrong — if all my failures were that good I’d be writing this from my private island — but it was a failure knowing that it’s unlikely whatever stock I bought with the proceeds went up another 400%.

I sold a perfectly good stock (potentially creating a taxable event) to move my cash into a worse investment. It might not be a failure, but I certainly don’t classify it as a smart decision.

It might help to look at it from a pure dollars and cents perspective. Say I bought 2,000 Intertape shares. We’ll call it a $4,100 investment. I sold 1,000 shares for $7.12 and 1,000 shares for $12.23. My total profit was $15,250.

If I would’ve held on until the bitter end, I would’ve earned a profit of $76,900. That translates into an opportunity cost of $61,650.

I firmly believe deciding when to sell is the toughest decision in investing. It’s infinitely harder than buying.

The average investor hasn’t thought about selling much either. I meet a lot of regular investors — it’s one of the perks of having this newsletter — and I’m usually pretty impressed with their buy processes. They have certain investment characteristics they look for and are generally pretty disciplined. Sure, they might waver a little (like the hardcore dividend investor I recently talked to with 10% of his portfolio in Bitcoin), but for the most part they’re making smart decisions on the buy side.

When it comes to selling, these same investors have no plan. It’s all driven by intuition and feelings. They typically sell for one of five reasons:

• The stock has gone up

• The stock was down and has gotten back to break-even

• They feel the company is overvalued

• They’re nervous about the underlying market

• They need to free up cash for another investment

On the surface these feel like good reasons to sell, but they’re actually not.

• A stock’s previous price has little to do whether it’ll be a good investment, since the market looks forward

• A forward-looking market especially doesn’t care what you paid for a stock

• An overvalued company can easily trade at a ridiculous valuation

• Trying to predict short-term market moves is a sucker’s game

• I’m most sympathetic to this reason — I also have a lot of ideas and only a certain amount of capital — but with no selling plan in place an investor is likely to sell the wrong stock at the wrong time

The reality is while these might feel like good reasons to sell, logically they’re quite poor. Doing so might make you feel good at the time, but they expose you to the same issue as my Intertape Polymer investment. You run into the potential of missing a massive winner because you sell too early.

The good news is I’m increasingly running into investors who have embraced one of Buffett’s many excellent philosophies. They pledge to never sell, knowing that the winners will more than make up for the losers.

Overall, this philosophy is excellent, and I won’t poke too many holes in it. An investor who buys without a plan to sell is much more likely to approach investing from its most businesslike angle. It also minimizes decision risk, which is a really important part of investing.

(Forget about what Google says, here’s how I define decision risk. The more decisions you force yourself to make in an exercise like investing, the more chances you have of making a bad decision. If we minimize the number of decisions we make, then we can get to the point where we virtually eliminate decision risk)

But we must also remember that even though Buffett advocates a never sell philosophy (“our favourite holding period is forever”), the fact is the Oracle of Omaha sells stocks all the time. Each quarter Berkshire publishes its 13F filing with the SEC which details its buy and sell decisions. And just about every quarter Berkshire is selling something. 

Viewing this on the website? Get this email forwarded to you? Subscribe to our free newsletter for great stuff like this straight to your inbox.

Did I mention it’s free?

My selling rules

I’m an early retiree with lots of spare time on my hands. I love it because it gives me the time to think deeply about the kinds of things other investors don’t have time for — like when to sell a stock.

After all that thinking, I condensed the reasons to sell a stock into a small list. I kept it small primarily to keep it simple, but mostly I realized there are only a few good reasons to sell a stock. I wanted create a list that would avoid the bad reasons for selling (like I outlined above) and really force me to think about the good reasons for selling.

So without further adieu, here are the reasons why I’ll sell a stock.

Let’s start off with the most common reason I’ll sell — thesis creep. One of the advantages I have as a writer is I get to share with y’all when I buy a stock. I get tell you what I’m thinking. This allows me to go back and see what I was thinking months later.

If I bought a stock based on one thesis and that thesis has either changed or it isn’t working out, I’ll sell. However, this rule comes with a twist, since it’s really easy to take one of the emotional reasons to sell I outlined above, twist it a bit, and call it thesis creep. So there’s a wrinkle to this rule.

But first, a real-life example.

I purchased a small position in Artis REIT (TSX:AX.un) in January, 2023, enticed by the company’s huge discount to net asset value. It had a plan to bridge that gap by selling assets at net asset value and use that cash to repurchase undervalued shares. The stock fell some 30% from my purchase price because the market was skeptical and because the whole REIT sector in Canada sold off.

I recently took a look at Artis’s 2023 annual report and found that even though its revenue was down some 25% on a year-over-year basis, it had only repurchased something like 7% of its shares. Proceeds of the asset sales were primarily being used to pay off the debt attached to those assets, not to repurchase shares. It wasn’t creating as much value as I first thought. I also started to get really concerned Artis was selling all its good assets (retail and industrial) and keeping the junk (suburban office). And finally, Artis’s earnings on a per share basis collapsed — getting to the point where it barely covered its dividend.

So I sold Artis.

But here’s where the wrinkle comes in — I gave myself a three month cooling off period before I hit the sell button. I was thinking about selling for months before I did. Why? Because a thesis like the one I believed about Artis takes a little while to work out, and it’s too easy to throw up your hands after a lackluster quarter and hit sell. I want the analytical part of me to make the decision, not the emotional part.

It’s also too easy in a situation like that to look at the share price and declare the strategy isn’t working. I look deeper, at the underlying results. Often the underlying results will be getting better and the company gets no credit for it. If the thesis is working behind the scenes but the stock isn’t getting any credit, I’ll hang on. In fact, I’m usually thinking about buying more in such a situation.

Want to know when I sell a stock?

Like Buffett, I like to buy a stock and then ideally hold it forever. Unfortunately, not every stock deserves such a lofty status. I help Premium subscribers separate the wheat from the chaff with access to all my buy and sell decisions, with an explanation of what I’m doing and why I’m doing it.

I help DIY investors choose great dividend stocks — which ultimately helps regular investors to make smarter choices, sleep better at night, and, most importantly, retire richer with more passive income.

It’s the very same strategy I used to retire before my 40th birthday.

Only $150 per year. Upgrade your subscription today!

The second reason I’ll sell is if the company has a years-long streak of no earnings growth. I put in the years-long qualifier because, again, I want to be patient. I want the stocks I invest in to grow their earnings over time, with those increased earnings hopefully translating into higher dividends.

I sold Extendicare (TSX:EXE) in 2023 despite holding the stock for a decade. I enjoyed the dividend and I still think long-term care is going to experience a nice demographic tailwind. But ultimately the company just couldn’t grow its earnings on a per share basis, and that’s what caused me to sell.

I also used that rule to sell Canadian Utilities (TSX:CU) earlier this year. Yes, that’s right, the king of dividend investing in Canada (in my own mind, anyway), sold Canada’s dividend growth king. In hindsight it was a simple decision — CU hadn’t grown earnings per share in years, and I doubt it’ll do so in next few years either.

The first two reasons to sell are more qualitative reasons. The remaining reasons are more quantitative in nature. They’re more yes/no decisions and are much easier to make than the first two.

The next sell rule is quite simple. If a company cuts their dividend because of poor business performance, I hit the sell button. I’m the first to admit dividend cutters can recover, and often do. In fact, the dividend cut often coincides with the bottom. Deep value Nelson loved sniffing around dividend cuts. But I’ve evolved a lot since then, and these days I really don’t want to be involved in any serious turnarounds. I’d much rather wait a little while and buy in when the company has already recovered.

The next rule mostly has to do with acquisitions and is closely related to the thesis creep rule. If I’m holding on to a stock because I’m speculating on a takeover, or a bidding war, it’s time to sell. This means that typically when a stock gets a takeover offer, I’m out.

This might seem ironic to some of you who have read my massive guide to merger arbitrage, odd-lot tenders, and other special situations. I’m happy to play in that sandbox and have consistently made a few thousand dollars a year doing so.

But this is where the thesis creep rule comes into play. Typically I buy a stock because it’s a good long-term business that’ll help me compound my capital at 10% per year, which is my hurdle rate. If the thesis changes from “good long-term business” to “speculation on a higher bid coming,” then that’s a thesis change, and I’m out. I’m happy to come back in and play the merger arbitrage play, but only after a sale and a cooling off period.

I did exactly that in 2020, selling Northview Apartment REIT in February when it announced it would be acquired. A few months later during the COVID-fueled market chaos, the deal looked to be in jeopardy. This was despite the buyers practically shouting at the top of their lungs they were still going to do the deal. I bought back in and my shares were taken private a couple of months later for a nice short-term return.

Finally, I’ll sell when a company hits a ridiculous valuation. This one is tricky and I’ve only ever exercised it once. One of the most important lessons I’ve learned about investing is this — the market doesn’t care what you think a reasonable valuation is. If folks are sufficiently excited about a name, it can easily go from a high to a ridiculous valuation.

One example is Costco, which trades at 44x earnings. Much of my career was spent in retail, so I understand what a wonderful business Costco is. But this is a company that is projected to grow the top line by 6% per year over the next few years. There’s no way (in my opinion, anyway), it’s worth more than 25-30x earnings. But the market doesn’t care what I think. Costco could trade at 60x… 75x… or even 100x earnings and it wouldn’t matter. Once it hits a ridiculous valuation then all bets are off.

I bring up Costco because up until a few weeks ago, I owned it in an account I manage for a family member. I sold at just over 50x earnings and cycled their capital into two much cheaper stocks.

Ultimately, I don’t believe a mature, blue-chip stock like Costco trading at 50x earnings is extremely risky. So I’ll take my money off the table.

My selling rules — condensed

To summarize, here are my sell rules in a nutshell:

• Sell when the thesis is destroyed

  • Giving myself a minimum 3 month cooling off period to ensure the thesis really is destroyed

• Sell when earnings growth collapses

  • A years-long streak of no earnings growth only

• Sell when a company cuts its dividend

  • No exceptions, no matter how much value Nelson protests

• Sell when an acquisition offer comes in

  • Even if you think the chances of a competing bid is high

• Sell when the stock trades at a ridiculous valuation

  • I use this one sparingly. A stock has to be really high before I’ll partake