Smart Reasons to Sell

Many regular readers of this portfolio have probably been wondering if our discussion to potentially sell Gap (NYSE: GPS) goes against the basic Rule Maker philosophy. After all, we've never done it before; for more than a year, we didn't have any selling criteria at all.

We make a big deal of our 10-year time horizon, and how holding through the bad times helps you profit from the good times. We didn't sell Intel (Nasdaq: INTC) when it was down (and still aren't thinking of doing so, despite our recent criticism). Coca-Cola (NYSE: KO) is down from where we bought it, and we're not looking to sell. Microsoft (Nasdaq: MSFT) fell 50% from its highs earlier this year and I'm still as outvoted on whether it belongs in our portfolio as the day it was purchased. So, what's the deal with Gap?

Phil Weiss and I have been around since this portfolio was launched, and Tom Gardner is still hanging around Fool HQ and can be trapped into having a conversation with us when you put glue on his chair. We (along with Al Levit) came up with these criteria in the first place, and the three of us (plus the new guys, Matt, Zeke, and Richard) are all in agreement that, if we're going to sell anything, Gap is a prime candidate. But, that's still an "if."

If you look at why we're so against selling, the constant trading that goes on in most portfolios is incredibly detrimental to returns. The taxes, spreads, and commissions steadily eat into the annual profits. And, statistically speaking, if the market's going up 10% a year (and it's done way better than that for the past couple of decades), having your money out of the market is betting against the flow of history. It's possible to win, but the odds are against you.

Also, we're not superhuman. It's easy to pick the lowest point and the highest point on a stock graph in retrospect, but almost impossible to do it while the line's still wiggling. Even if you're only off by a few days each way (Houdini-like accuracy), that can make a huge difference in your returns. We're not day traders; we prefer to have a life outside of investing. It's also easy to pick winning lottery numbers in retrospect. Try doing it without looking at the end results before making your decision.

Put those two points together, and think what happens when the bottom in the stock you want to buy comes a month after the peak in the stock you want to sell. They're different stocks, there's no reason for them to move in unison. So, your money's out of the market for a month; do that on a regular basis and the market's average is going to have an easy time pulling ahead of your portfolio.

On the other hand, we haven't been against buying. Since this portfolio's inception, we've added Cisco Systems (Nasdaq: CSCO), Nokia (NYSE: NOK), and JDS Uniphase (Nasdaq: JDSU). (Schering-Plough (NYSE: SGP) doesn't count. We meant to buy a second pharmaceutical from day one.)

Statistically speaking, the more you diversify, the more likely you are to get market-average returns. To beat the market, you've got to focus on a group of companies that do better than average. Preferably a small group -- the more you pick, the less likely they are to be uniformly better. We're toward the high end of the range we consider comfortably diversified. We don't need as many companies as we've got. Still, that's not an inherent reason to sell.

If we're holding for 10 years, does it mean we're going to sell all of our stocks in a rush 10 years after we buy them? No. That wouldn't make sense. Some of them will still be good investments, and taking a tax hit when you don't need to is dumb. Once we're over the 10-year hump, our sell decisions should be spread out a bit, based on when we decide we no longer want to keep each stock. Based on the stock, not just how long we've held it.

So, our 10-year time horizon should behave more like a half-life. There's a 50% chance within this time period that we'll sell the stock. (You may often have seen us state our horizon as five years. That's a better number to use as a half-life if you expect an average 10-year holding period. And, for you physics majors in the audience, yes I know it's not exact. I can look up the formula with the logarithm in it if I want to, but these are just rules of thumb.) And, with a half-life, some events occur sooner than average and some occur later. The decision is made on a case-by-case basis, looking at the costs and the potential benefits of each decision.

In the case of Gap, it's trading about where it was when we bought it, so there's not much of a tax hit to worry about. And, since we moved our portfolio to American Express (NYSE: AXP), we don't pay commissions on buys (and only pay for sells until we can get our portfolio balance above $100,000). So, redistributing the money into a new place doesn't cost us a commission. That's part of the reason we invest $500/month without worrying about commissions. We're not paying them when we do that. There are still spreads, but that's both unavoidable and fairly small. (If you'd like to compare American Express to some other discount brokers, see our Discount Brokerage Center.)

And, our decision to sell has nothing to do with the short-term movement of stock. We're not trying to sell at a peak or move the money into a new stock on a dip. We've been discussing this internally for a week already (based on an ongoing discussion on the discussion boards), and now we've taken that discussion to the daily reports. We're not exactly worrying about timing here. It's not that we're uncomfortable having our money in Gap at $23/share, it's that we're uncomfortable owning a chunk of Gap when debt and growth and such aren't satisfactory and don't look to improve.

This potential sell is not a decision we're taking lightly. But, if we've truly lost faith in the company, and we don't think hanging on to it has much chance of improving things, and the costs of switching are a price we don't mind paying to get our money working for us again... maybe it is time to sell.

-- Oak

Related Links: