For those of you who were on vacation last week, let me bring you up to speed: The market was down. Nasdaq, Dow, S&P 500 -- down. Rule Breaker -- down. To save space, I'll list the stocks in this portfolio that were NOT down more than 4% last week:

Amgen (Nasdaq: AMGN). Saved by a timely upgrade.

What's wrong with Amazon?
Of all our down stocks, none was a bigger downer -- and, in the end, a bigger rebounder -- than (Nasdaq: AMZN), which dropped as much as 18% last Wednesday for no apparent reason. One may argue that there's always a good reason for Amazon to drop below $7 per share, but it doesn't happen very often. In fact, it hasn't happened since 1998. Amazon CEO Jeff Bezos and spokesman Bill Curry got in the public face and reassured it that there was nothing to report, other than that Amazon was still taking good care of its customers. That's a relief.

The conclusion, then, is that either some fund dumped the stock, or that several funds "rotated out," as W.R. Hambrecht analyst Kristine Koerber asserted. Rotate in, rotate out -- it's no different than late-summer field fallowing, just like our grandfathers used to do. No wonder America loves mutual funds.

I don't particularly care why it fell. It doesn't seem to have been anything fundamental to the business, so it doesn't affect how I feel about the stock. Being a fan of conspiracy theories and wild speculation, however, I've concocted a theory that lays the blame on a particular party: Janus Capital Management, a division of Stilwell Financial (NYSE: SV).

Janus is (or perhaps was) the third-largest shareholder in Amazon, after Legg Mason and Lincoln Capital Management. Last year, between its second- and third-quarter 13-F SEC filings (which disclose a fund company's holdings), Janus acquired some 8 million shares of Amazon, bringing its total holding to over 27 million shares. Since then, it has been selling off its stake. As of its last 13-F, Janus still had 14 million shares.

If Janus had been selling shares on a first in, first out basis, then it still had the 8 million shares it purchased last year in the third quarter -- maybe sometime after Sept. 5. By selling now, Janus could claim a short-term capital loss on its Amazon investment. It's my guess that they are doing just that.

Why would Janus care about taxes, you ask? After all, mutual fund returns as they have long been stated don't include the effect of taxes. That burden has always been foisted off on the shareholders in the form of capital gains distributions. These days, however, Janus has a new motivation to watch its tax bill: Starting in February 2002, it will have to report its funds' returns on an after-tax basis in its prospectus. If it can minimize short-term gains, it will save 38.6% of the Amazon loss -- which looks to be over $250 million, youch! -- in its tax bill. Even if it offsets long-term gains, it's still a 20% improvement. That is meaningful.

We're entering tax-loss-selling season, my friends, and it looks like it's going to be a bad one. (Writing that was fun. Isn't it every young person's dream to be a weatherman?)

What does that mean to us, the Rule Breaker Team?
This year, most of us have some down stocks, and the Rule Breaker is no exception. We're coming up on the one-year anniversary of our purchase of Human Genome Sciences (Nasdaq: HGSI), which is down about 50% since we bought it.

Now, I'm all for HGS. Of the stocks we considered for purchase last year, it was the best -- and it has performed the best, which is a whole 'nother story. Hey, HGS has beaten the Nasdaq Composite over the last year. That's something. Nevertheless, it would not be imprudent for us to consider selling HGS for tax reasons, now that we are reporting our returns after taxes. Here is the argument:

If we sold now, we would realize about $22,000 in losses. We've had two other sales this year, of Spiders (AMEX: SPY) and a portion of Amazon. The short-term loss in HGS, together with the loss from the Spiders, would have the tax effect of offsetting the $13,700 gain we realized in Amazon, negating the $2,700 tax we owe on it. The remaining $8,500 would get taken as a short-term capital loss, knocking another $3,300 off our tax bill. (We, like mutual funds, calculate our taxes without consideration of the $3,000 capital loss limit per year that the IRS imposes on you and me. In our cases, we would have to carry the remaining loss forward into next year, or perhaps think about selling only a portion of our HGS holding.)

Selling now, then, would effectively recover about $6,000 for the portfolio, which is 26% of the current value of our HGS holding. It's as though the stock bounced 26% in one day. Then, after the 30-day wash sale period passes, we can buy it back again. As long as it's not up more than 26%, we've benefited. At the moment, dazed and jaded as I am from the 18-month beating we've been taking, that seems like a reasonable bet.

One more minor point: A new super-long-term capital gains rate is now in effect. For stocks purchased after January 1, 2001, the capital gains rate is 18% instead of 20%, if you have held the stock for five years or more. While that 2% is not a big deal, and we would have to reset our five-year clock to start when we repurchase HGS, it offers a little icing for the tax-loss-selling cake.

Why should we let Janus have all the fun?

Brian Lund is a big fan of Janus -- the Roman god of thresholds, that is. You can't discuss liminality enough, in his opinion. Brian can't sell any of the stocks mentioned for a loss, since he doesn't own any of them. The Motley Fool is investors writing for investors.