When to Sell a Stock

For most of us, deciding when (and what) to sell is a lot harder than deciding when to buy. It's not a subject that gets much attention, largely because it can depend on individual circumstances--when you bought the stock, how big a part of your portfolio it is, etc. Still, there are some solid rules for when selling is a good idea, and for when it's not. 

First, let's look at the times when breaking up really is the right thing to do. Whenever you think about selling a stock, run through these four questions.

Did You Make a Mistake?
In other words, did you miss something when you first evaluated the company? Perhaps you thought management would be able to pull off a turnaround, but the task turned out to be bigger than you thought. Or maybe you underestimated the strength of a company's competition, or overestimated its ability to find new growth opportunities. No matter what the flub, it's rarely worth holding on to a stock that you bought for a reason that's no longer valid. If your initial analysis was wrong, cut your losses and move on.

I did this several years ago with a manufacturer of commercial movie projectors. My growth expectations turned out to be way too high, because the multiplex-building boom of the 1990s was waning. Theater owners started to get into financial trouble, and they were a lot more worried about paying their bills than building new theatres. I was down quite a bit on the investment by the time I figured this out, but I sold anyway. Good thing, too, because the shares subsequently plunged to penny-stock territory.

Have the Fundamentals Deteriorated?
After several years of success, that raging growth company you bought has started to slow down. Cash is piling up as the company has a tougher time finding profitable, new investment opportunities, and competition is eating away at the company's margins. Sounds like it's time to reassess the company's future prospects. If they're substantially more grim than they used to be, it's time to sell, no matter what the stock price has done since you bought it. This applies especially to turnarounds, which can deteriorate in a hurry. As Warren Buffett has said, "The trouble with turnarounds is that they often don't."

Has the Stock Surpassed Its Intrinsic Value?
Let's face it: Mr. Market is a capricious individual, and sometimes he wakes up in an awfully generous mood and offers to pay you a price far in excess of what your investment is really worth. There's no reason not to take advantage of his good mood. What you need to ask yourself is how likely it is that your estimate of what the company is worth could go up over time. If your estimate of intrinsic value is likely to rise, than it’s worth waiting out periods of mild overvaluation.

Generally, it doesn't take much in the way of a valuation premium to convince me to sell stocks with minimal economic moats--when they get close to their fair values, I'm out. However, I won't sell stocks with wide moats--ones with fair values that are likely to rise over time--on valuation concerns unless they’re egregiously pricey, in which case I'll happily take Mr. Market's generous offer and move on.

Has the Stock Become Too Large a Part of Your Portfolio?
Any time one stock grows to become more than 10% of your portfolio, you should start thinking very carefully about how much risk you're taking on--even if you still think a company has great prospects, it's simply not prudent to allow it to take up too large a percentage of your portfolio. I learned this lesson the hard way a few years ago. I'd bought a few hundred shares of a small company called  EMC  back in 1995, and five years later it comprised more than a third of my portfolio. I sold some shares to lessen my exposure, but not nearly enough. As a result of my greed, I suffered more than my fair share of the tech downdraft. Sometimes we need to learn the hard way.

Of course, spreading your risks around applies to larger categories--such as sectors, market caps, and styles--in addition to individual stocks. Any time you find that too much of your portfolio is in one area of the market, then you've probably got yourself some good sell candidates. Morningstar.com's Instant X-Ray tool is great for this.

Now, let's look at some reasons why you shouldn't sell a stock.

The Stock Has Dropped
By themselves, share-price movements convey no useful information, especially since prices can move in all sorts of directions in the short term for completely unfathomable reasons. Moreover, the future performance of stocks is largely based on the expected future cash flows of the companies attached to them--it has very little to do with what the stock did over the past week or month. So when you're making a sell decision, look to the future, rather than the past.

The Stock Has Skyrocketed
It is so, so tempting to use the past performance of stocks in your portfolio to decide which to sell and which to keep. But remember: It matters little how those stocks have done in the past--what's important is how you expect the companies to do in the future. There's no a priori reason why stocks that are up a lot should drop, just as there's no reason why stocks that have tanked "have to come back eventually."

The trick is to stay focused on the future. Let's say I buy shares of a retailer that was having some major liquidity troubles but that was so cheap that its real estate alone was worth more than the enterprise value of the company. I also buy a former tech darling that's fallen on hard times but that I think can tough things out until technology spending improves. Six months go by, and the retailer has doubled, while the tech firm has fallen 30%. I should lock in my gains on the retailer, and hang on to the tech company until it turns around, right?

Wrong. The stock price movements say nothing about the health of the two businesses. Over the past six months, my hypothetical retailer has sold off some stores and made big improvements to the way it manages inventory, so the liquidity crisis is past. Moreover, same-store sales numbers are faring well, profitability is on the mend, and the stock trades at a reasonable price. Sounds like this one's worth hanging on to. Meanwhile, the tech company has struggled as a large competitor has initiated a brutal pricing war, a rebound in corporate IT spending looks even farther away, and a couple of senior managers have bailed out. Hmm--looks like it's time to cut my losses on this one because things look a lot worse than when I initially bought the shares.

The Stock Has Rebounded to Your Purchase Price
One of the silliest things I hear people say is, "I'll sell that dog when it gets back to break-even." The only thing that matters is your estimation of a stock's performance versus alternative investments. If you own a stock that's down 20% with very poor future prospects, and there are better investment opportunities out there, you’re better off booking the loss, getting the tax break, and putting the proceeds in the other investment than stubbornly hanging on. After all, cumulative investment returns of 0%--minus transaction costs--aren’t terribly attractive.

Shameless Plug
Morningstar's first full-length book on stock investing is out--look for it in your favorite bookstore, on BN.com, and on Amazon.com. It's titled The Five Rules for Successful Stock Investing, and it was written by yours truly in conjunction with Morningstar's stock analyst staff. The first half of the book covers the basics of fundamental analysis: a discussion about developing an investment philosophy, a mercifully short introduction to reading financial statements, a primer on valuation, and a guide to evaluating company management and assessing economic moats. The second half is a series of chapters on each sector of the market--from hardware to health care to banking--that explains industry jargon and helps you analyze complicated companies.

A version of this article originally ran 9on September 4, 2002.