Agree or not, good reading.
From MSN Investor
Print-friendly version
Posted 7/6/99
Jubak's Picks
--------------------------------------------------------------------------------
Check out Jim's top stocks for the next 12 months.
50 Best
--------------------------------------------------------------------------------
See Jim's list of the 50 best stocks in the world for the long term.
Jubak's Journal
Don't forget about those stocks you've sold
A stock you've unloaded doesn't have to crash and burn to make the sale a success. In fact, those shares might be among the best candidates to buy back again. Yahoo! is a good example.
By Jim Jubak
I hadn't realized exactly how odd my last column ("Up 29% for the quarter and still not happy") was going to be until I'd finished it. Most report cards on the performance of a portfolio concentrate on how the buys did for the period in question. Stock A was a stinker -- it went down and not up after the date of purchase; Stock B has been a disappointment in the short term but the future still looks bright . . . you know the drill.
But the performance review I wrote on Jubak's Picks focused on the sells of the last two quarters. Frankly, I wasn't out to make any big point -- my sells just struck me as more critical to the performance of the portfolio than the buys. But I have to say that I can't remember seeing another performance review with this much emphasis on sells.
And that led me to this column's question: Why don't we investors pay more attention to the success or failures of our sell decisions?
Here's my take on an answer. We really don't have a very good idea of how to define a successful sell. Part of the problem is that, thanks to the upward bias of stock prices over time, all sells look bad eventually. But just as important, a lot of sell decisions are attempts to limit potential risk -- and investment science still hasn't come up with a precise way to measure the value of avoiding a potential loss.
It's not pretty, but it works
I think most sells fall into one of two categories. First, investors sell to avoid a loss. Second, they sell because they believe better opportunities are available elsewhere. In the real world, the two blend together and most sells are motivated by a combination.
If we recognize and build on that fact of selling life, I think we can put together a practical, ad hoc solution that will let us judge the success of a sell better. It's not pretty, but it works. (I think this solution ideally is suited to the 12-to-18-month holding period of Jubak's Picks. How to think about selling in a buy-and-hold portfolio like my "50 Best Stocks in the World" will be the subject of my next column.)
For example, on Jan. 8, I sold Lucent Technologies (LU) out of Jubak's Picks at $57.63 a share. How did that sell do when measured against these two goals?
The stock had been on a tear since the end of November 1998, climbing from $43.50 to $57.63, a 32% gain, in less than six weeks. That run put the stock above my target price of $56 by July 1999. After doing the numbers, I concluded that the stock had run ahead of its fundamentals, was probably overvalued, and no longer offered a sufficient potential return over the next year to justify the risk. So I sold to preserve my profit.
Details
--------------------------------------------------------------------------------
Company Report
1-yr Chart
Earnings Growth Rates
Was that a successful sell? Well, yes -- for the next few weeks. Lucent did drop to $48 by Feb. 17, a five-week retreat of 17%.
Does a stock have to tumble?
But then Lucent started to recover. By early April, it had climbed back to my selling price. Over this longer, 12-week time period, the stock broke even, so my sell didn't really protect me from a loss. If a stock has to tank before a sell can be considered a success, then my sell of Lucent didn't succeed in this longer time frame. Measured against the standard of my first goal -- avoiding a loss -- my sale of Lucent falls short.
Is that a reasonable standard? Measured this way, most sells are dismal failures. An investor has to be extraordinarily lucky -- or very good, indeed -- to sell a stock just before it tanks.
Any investor who decides to sell to avoid a loss, however, actually is trying to protect against the probability of a downturn, not the certainty of one. That's why sell decisions made to avoid a loss tend to involve stocks where the price-to-earnings ratio has spiked upward to a figure that seems too high, or where a recent run-up in price has been too rapid. Both of these situations, we believe, signal an increased potential for loss.
So, to take an immediately applicable example, I'm going to sell Yahoo! (YHOO) with today's column. By Thursday, July 1, Yahoo! had run from my June 15 purchase price of $125 a share in Jubak's Picks (see my column, "Buy Yahoo! -- It can't get any worse") to $177, well past my September 1999 target of $170. And I think the stock is likely to rally into its earnings announcement after the close July 7.
No certainties, just probabilities
But I know from history that Internet stocks tend to retreat (the Internet investor's euphemism for "plummet") after this kind of anticipatory earnings rally. And I also feel that the Federal Reserve's June 30 declaration of neutrality on interest rates has only taken that issue off the table temporarily. So I think selling Yahoo! here is a reasonable attempt to limit a potential downside that, in the past, typically has run to as much as a 40% loss.L I V E V O T E
Voting ends 7/14/99
Rate Yahoo!
Buy
Sell
Hold
View results
Am I certain that Yahoo! will go down? Of course not. Am I certain that Tuesday, the day my sell takes effect, is the absolute top for Yahoo!? Again, no. But given the probabilities as I see them, this is a good time to sell a volatile "trading" stock such as Yahoo!
Selling in situations like this is an attempt to manage risk, and I think it's wrong to say that the only successful attempts to manage risk are those where the asset sold actually crashes and burns. Risk management is about limiting the potential damage an investor faces to an acceptable level in comparison to the potential return. Not all potential threats will materialize, just as not all potential returns will turn into real profits. I believe it's reasonable to consider a sell a success, therefore, even if the threat of a loss doesn't lead to an actual loss.
Like a buy, which needs to be re-evaluated constantly to make sure that the numbers still add up, a sell decision is only good at the moment it's made. Especially if the sell works out and the stock declines, the sell candidate might become a buy candidate in a matter of weeks.
But we practicing investors do need some way to separate sells that are reasonable attempts to limit loss from those that are little more than a surrender to the fears of the moment. Almost any sell can be justified if all that's required is some worry about the future. If we set the bar that low, then we ought to admit that we're selling whenever the mood strikes us.
Seeking out relative success
This is where the second goal of selling -- the sell because an investor feels there's a better opportunity elsewhere -- comes into play. If instead of demanding absolute success (the sold stock has to crash), I look for relative success, then I think I've got a standard that will let me separate reasonable and unreasonable sells.
In practice, selling of this sort doesn't involve the direct substitution of one stock for another. For example, I sold Lucent Jan. 8 and then FDX Corp. (FDX) not too long after. I could say that I sold Lucent to buy EMC (EMC), Nokia (NOK) or Nextel Communications (NXTL) -- all January or February additions to Jubak's Picks. But I could also say that I sold FDX to buy one of these.
In fact, I'd be hard-pressed to say that I sold Lucent in order to buy any specific one of these stocks. And in this case, if I intend to measure the success of my sell by comparing Lucent's performance to the returns from its replacement, it does matter which specific stock I call the replacement. The potential choices have returned about 15%, 35%, and 50% since I added them to Jubak's Picks. With Lucent up about 18% since I sold it, that sale is a failure or success depending on which comparison is appropriate. This method for labeling a success or a failure, frankly, seems rather arbitrary.
Instead, I think it makes more sense to measure a sell against the same kind of generic hurdle rates of return that I use in making buys. I sold Lucent in January because I calculated at the time that the potential return for holding Lucent for another year was just 18%. And that wasn't enough to justify holding the stock when I was asking buy candidates with similar risk profiles to show a potential 25% return. By doing that, I was comparing Lucent with other stocks of a similar type to see which would do better over the next year. And by selling, I was saying that I believed I could find stocks that would do better than Lucent would. You need to identify a sell looking ahead instead of looking backward.
The analysis never ends
The tricky thing about this kind of sell analysis is that it's never over. Like a buy, which needs to be re-evaluated constantly to make sure that the numbers still add up, a sell decision is only good at the moment it's made. Especially if the sell works out and the stock declines, the sell candidate might become a buy candidate in a matter of weeks.
It's tempting to put a stock away and forget about it once you've made a sell, but that's often a mistake. Look at the Lucent example again. On Jan. 8, I calculated that the stock would climb to $67.56 in a year, a gain of 18%. With that potential return, the stock didn't make my cut.
But, within weeks, the stock was trading at $48. That meant that if I still believed in my Jan. 8 target of $67.56, the new potential gain on Lucent after its decline was 41%. The sell had become a buy candidate.
Jubak's Archives
--------------------------------------------------------------------------------
Recent Jubak articles:
• Up 29% for the quarter and still not happy, 7/2/99
• How to trade stocks for the rest of the year, 6/25/99
• Averaging down? Check your emotions at the door, 6/22/99
More…
Of course, not every stock that declines becomes a buy. When I looked at Lucent at $48, I concluded that we were headed into a riskier period for technology stocks and that with interest rates a good bet to go higher, technology stocks would deserve lower P/E multiples. When I re-did my calculations, I lowered my target price on Lucent. The new potential return over a year was higher at 23%, but Lucent still didn't make the cut as a buy.
Constantly reviewing sells to see if they've turned into buys may seem like hard, time-consuming work, especially if the review doesn't result in an immediate buy. But this kind of review is actually one of the most productive ways to invest your limited research time and energy. After all, when I decided to study Lucent again, I already knew the company, had collected most of the numbers I needed and was familiar with the industries where the company did business. I'd done all that work when I initially purchased the stock.
No guarantees
It's only human nature not to want to revisit the stocks that you've sold, but ignoring past sells when you're looking for future buys is often a mistake. To avoid this "sell and forget" tendency, I put every sell I make into a single watch portfolio in the MSN MoneyCentral Portfolio Manager, along with the target price I calculated when I sold the stock. I try to run down that list once a week to see if a decline has made any of these past sells a candidate for research as a possible buy-back.
Of course, even taking this precaution won't guarantee that you won't make mistakes or allow your emotions to bias your analysis of a former sell. We do want sells to go down, and it's hard to admit that they might be headed back up again. Take it from an investor who knows -- this kind of emotional bias prevented me from seeing the potential in such previous sells from Jubak's Picks as TriQuint Semiconductor (TQNT) and Qualcomm (QCOM), even when I revisited them.
I still haven't found any foolproof way to banish all emotion from my investing decisions.
Details
--------------------------------------------------------------------------------
Research Wizard
Earnings Estimates
--------------------------------------------------------------------------------
Changes to Jubak's Picks
Sell Yahoo! (YHOO)
Yahoo! has moved solidly above my $170 target price in a rally typical of an Internet stock in the weeks before earnings are announced. I don't think the current Internet sector rally is going to stick -- I don't think most stocks will take out their 52-week highs -- because I'm expecting a resurgence of interest-rate worries as the summer progresses. I think this is a trading rally and I'm expecting a decline for Yahoo! after earnings are announced. I'm selling the stock with a 40% gain since I added it to Jubak's Picks at $125 on June 15. And I'll definitely be adding this one to my list of recent sells that are potential buys after a price decline.
|
