Valuation Swap: A New Way To Approach Selling

I firmly believe that the decision to sell a stock is one of the toughest an investor can make.

The buying decision is much easier in comparison. Most investors are optimistic by nature — if you weren’t, you probably wouldn’t invest — so we can twist ourselves into believing even the most outlandish of bull cases. Greed wins out over fear, at least until the next bear market.

I’ve struggled with selling over the years. I’ve watched long-term winners explode higher after holding the stock for years as it did nothing. I’ve sold losers only to see them rally from that moment on, solely just to spite me. And I’ve sold long-term winners that were performing well way too early. One example of that last one — the stock was Intertape Polymer — locked in a nice five-figure gain for me, but holding on for the long-term would’ve turned it into a six-figure gain.

Once I realized I wasn’t a great seller, I started acting like a turtle who was happy to retreat into my shell. If there’s a chance I’ll make the wrong decision, then I won’t make any decisions, dammit. I’ll embrace a coffee can buy-and-hold forever philosophy.

Like I said in that post, I do plan on embracing a mostly coffee can philosophy. I the only problem is I over-diversified — I owned ~100 stocks at one point, and am now down the the low-80s, which is still too many. I have too many positions that don’t move the needle at all, and I’m slowly correcting that.

When I’m done with this exercise — it’ll likely take years — I plan to have core positions in 40-50 different businesses. Some might argue I’m overdiversified at that point too, but I think I’ll be fairly comfortable there.

Anyway, as part of my thinking about selling, I’ve come up with an alternate theory on when to sell for Canadian dividend investors. Let’s take a closer look.

Let’s begin by talking about how I buy.

I’ve talked about this approximately a million times at this point, but in case you’re new around here, the qualities I’m looking for include:

From there I look at valuation, hunting for bargains. I primarily use three factors to determine if a stock is a good deal or not.

I’m looking to buy attractive income streams as cheaply as possible, which means buying at a low valuation. I’m happy to patiently wait until the market cooperates with me, knowing that there’s always some corner of the market that is beaten up.

For instance, McDonald’s (NYSE:MCD) might not seem super cheap today. The stock offers a 2.8% dividend yield and a forward P/E ratio of about 20x. But both those numbers are quite attractive compared to the company’s recent history. 2.8% is about a 20% higher dividend yield than normal, and 20x fwd. earnings is close to the stock’s cheapest valuation this decade. Therefore, it’s attractive to me.

Anyway, this all begs the question — if looking for those qualities identifies good buys, then shouldn’t the inverse of my buy criteria produce good sell candidates?

Specifically, you’d be looking for:

One thing I’ve learned is even boring stocks can periodically become ridiculously overvalued. Case in point, Walmart (NASDAQ:WMT). I owned it for a long time, buying my position more than a decade ago for under 15x earnings. I patiently held on and the stock did incredibly well.

Walmart did well because it successfully pivoted from a physical retailer to a hybrid with a colossal online presence. This drove both top and bottom line gains. But the bigger story was valuation expansion. Walmart shares traded at 15x earnings a decade ago; these days they trade hands at about 40x earnings.

My view is 40x earnings is too damned expensive for a retailer, and so it was one of the stocks I sold earlier this year.

This plan has two massive advantages compared to a few disadvantages. One nice part of this plan is it creates a simple, rules-based sell process. It takes all the thought and guesswork out of when the best time is to sell a winner. The rules say to sell — just like the rules say when to buy — and you jettison the stock out the door. Thanks for the contribution, pal.

It won’t generate perfect results, but what method does?

The second advantage is the big one. By selling companies with paltry dividend yields and swapping them for companies with more robust yields, you’re giving yourself a raise with very little work.

Walmart offers a 0.9% dividend yield today. If someone sold Walmart and put the proceeds into H&R Block (NYSE:HRB) and its 4.1% dividend yield, that one move could represent a nice little raise.

Say Walmart was 4% of someone’s portfolio and just 1% of their income. The H&R Block swap would give this imaginary investor a ~4% raise immediately.

Now onto the disadvantages.

There are a couple reasons why this plan might be a bad idea.

The first is valuation is hardly an exact science. The Walmart example is a perfect one to illustrate my point. Let’s pull that graph back up.

In 2024, Walmart saw its valuation expand from just over 20x to almost 40x forward earnings in the span of a year. The highest valuation in the last decade was about 27x forward earnings. If the rule dictates we sell at 10-year high valuation, then shares would’ve been punted at about 30x earnings, or about halfway through 2024.

That would’ve been a bad idea. I would’ve left a lot of upside on the table.

One of the things I’ve learned about valuation is once you get a bunch of investors into a name that don’t care about what the stock is valued, then the sky is the limit. A value investor just isn’t going to pay 35x or 40x earnings for Walmart. But a growth investor sure would, especially if the story makes sense.

Get enough growth investors in, and all thought of valuation goes out the window. You get a stock that moves up bigly based primarily on multiple expansion alone.

Another issue with a rules-based sell approach is it will inevitably trigger more taxable events than a buy-and-hold coffee can portfolio will.

I was lucky with my Walmart sale. It happened in my wife’s RRSP, so we didn’t trigger any capital gains taxes when she punted the position to the curb. But much of our assets aren’t in tax-protected accounts. Anything that gets sold from those accounts triggers capital gains.

There are various tactics to reduce or defer capital gains — like selling in January, knowing the tax bill won’t be due until April of next year, or selling something at a loss to offset the gain — but they’re a little bit like putting a bandage on a gaping head wound. The damage is done at that point.

(At least I assume that’s how head wounds work. I am clearly not a doctor.)

Say I’m forced to put 20% of the proceeds from my Walmart sale aside for taxes. Suddenly the raise I’ve given myself has taken a haircut as part of my capital gets shipped to Ottawa. It takes away part of the very advantage we’re looking to get, which is a little bit counterintuitive.

(This is why I like yield on cost. Detractors will say it’s dumb, or it’s a vanity metric, or whatever, because you can easily sell something, pocket the gains, and switch to something else. This argument almost always ignores the tax man)

Finally, the sell when a stock is overvalued plan still comes with the potential for regret. It’s still introducing a sell decision and the potential to make a mistake into the investment process. In the wrong person’s hands it’s an invitation for disaster. Soon they’re back to using vibes or gut feels or break-even prices to determine when to sell.

Selling is something ideally done with a light hand. We don’t really want to encourage it.

Let me be clear: I don’t plan to implement a rules-based selling formula for my portfolio.

If I did, terrific companies like Canada’s banks, pipelines, and utilities — which are all trading for multi-year valuation highs — would be candidates to sell. Doing so would leave me with a large amount of cash and much lower dividend income. It’s not an ideal outcome.

But I think there is a certain amount of logic to it. There should definitely be a valuation where you’re willing to hit the sell button. Such a move doesn’t just potentially increase your income; it protects your gains as well. It may also help you sleep at night.

It has been a valuable exercise for me as I continue to get rid of small positions in my portfolio. As I look to sell these positions, I’m doing the analysis, looking at valuations, and considering dividend yields as I weigh which stocks to sell. I’ve found myself selling low-growth options that are trading for much higher valuations and that are offering much lower dividend yields than normal. It’s been a useful exercise.

Perhaps it’s better used as a “don’t buy” signal. The valuation is too high, so you pivot to buying something else. It keeps going up, gets ridiculously high compared to its history, and then the sell decision is made.