Why We Sold

March 2, 2004

We don’t trade our model portfolio very often. Scan through the changes history for any model portfolio and see for yourself. Trading costs money. Frequent trading incurs excess costs by realizing taxable gains at inopportune times (and higher rates), generating commissions, and triggering short-term redemption fees. 

The main reason we don’t trade actively, though, is that our fund investing strategy is reasonably long-term – when we like an area of the market or category of fund we usually like it for reasons that are anything but trendy. 

Moreover, there are many categories of funds we will almost always have some exposure to - it’s just a question of how much: are we going to overweight or underweight high yield bonds at this time? 

What leads us to sell a holding is market movements or events that conflict with our original reason for purchasing. 

At the end of February we made some changes to all but one of our model portfolios. In our next mid-month portfolio commentary updates (March 15th) we will explain in more detail what was bought and sold in each portfolio and why. For now, we’ll discuss some broad explanations for the moves. 

Each model portfolio is designed around a certain target risk profile. Lower risk portfolios will generally fall less then higher risk portfolios in down markets, and usually climb less in up markets. This expectation has been our practical experience since we started the MAXadvisor newsletter. 

We do shift allocations of each portfolio based on our broad outlook for the market in general, and certain styles and categories of funds in particular. These adjustments do change each portfolio’s risk level, but not by much. In 2002 we sold our cash holdings in the Daredevil portfolio and increased our telecom and Japan exposure, a move that increased risk. At the end of February 2004, we sold emerging market bonds, a move that lowers risk. 

The risk profile of these investments change when they go up and down in price significantly – emerging market bonds are now more risky after they went up some 37%+ since we added them to the portfolios in April 2002. By easing up on hot areas of the market an investor is often just bringing their portfolio back to the pre run-up risk level.

By this analysis, the SP&500 becomes a less risky investment after it falls 25% then when it climbs 25% (in a relatively short time period). While every asset class has some intrinsic risk level (junk bonds are riskier then government bonds and stocks are riskier then both) we feel risk is also defined by the likelihood of losing money versus the likelihood of making money going forward. 

As prices rise the likelihood of making money tends to fall. There are times when an investor can rationalize price increases because the future outlook has improved too, but fundamentals can only explain so much.

There are several factors we examine when determining if an investment category is getting too expensive, notably the valuations in the category. We compare the expected returns to the current price, and contrast that return with the portfolio’s risk level. 

The trouble with such an analysis is that there are a lot of unknowns, in particular the estimate for what the future can bring. Often analysts have to lean on historic valuations and assume if they are at the high end future returns will be below average – a reasonable starting ground. For example, high yield or junk bonds pay a greater yield then safe government bonds do to reward investors because they are riskier investments. But there is a historic range on how much more the higher risk junk bond pays than the safer government bond. If these bonds are paying yields at the low end of that range – just a few percent more then a safe government bond, it’s a good time to cut back on these bonds.

One strategy that we have developed and found to work time and time again is to observe what other investors are doing with their money. For lack of a better term, call it “the last dollar left theorem”. A pessimist could call it the “greater fool hypothesis”.

This theory comes from years of observing mutual fund investor’s behavior firsthand. Fund investors have a tendency to get the most excited about an investment shortly before returns slip. On the flipside, they tend to get most pessimistic shortly before events turn around.

In addition to our own qualitative and quantitative analysis of this defeatist pattern, numerous studies have been compiled examining decades of data, which show a clear pattern of investors consistently under-performing the market largely from this buy-high, sell-low formula.

There are two explanations why returns suffer after fund investors hop into a particular category en masse: one is there is nobody else left to pay a higher price for the stocks and bonds they are indirectly buying, the other is that when this happens too much money chasing after limited dollars waters down returns for everyone.

Investing has finite returns, whether it is in bonds, stocks, or real estate. The more money that piles into an asset category, the lower the future returns are for everyone. Investors are only going to make so much money in say, biotech stocks. If millions of investors poor billions into biotech, those investors are going to have to share the rewards with more investors.

Investment categories go through cycles of popularity – from total aversion to complete captivation. Along the way from point A to point B most of the returns are made. Fund investors tend to be late to the party because they often buy whatever fund performed the best over the last few years. 

Often early investors put money in and earned great returns. Next more investors come in and raise prices from their buying. Eventually later stage investors see the action and hop on board, some for fundamental reasons (maybe the category is up for good reason, and early investors uncovered something) or simply because they are momentum investors who want to get in on all the action. 

The last stage of investor is often the fund investor, the late adopter. They read an article in the paper about the top funds in some category over the last few years, saw some impressive ratings or past returns, and hop on board. Their inflows of cash can raise prices even higher. 

Unfortunately at this point there is no one left to recruit – everyone who would want to invest is this area already has. There is almost nowhere for the area to go but down, as money eventually starts leaving, disappointed at the meager returns, the result of having to share profits with hoards of investors. With more speculative investments the disappointing returns and outflows can lead to major collapses in prices. 

After many months of carnage, fund investors bail. As they were the last ones in, they are often the last ones out. When they leave, a new cycle gets underway.

Of course, fund investors are not always dead right or wrong, and markets can keep going up or down even after performance chasers pile in. But perfect contrarian timing is not needed to make money from these late-to-the-party investors; over time buying into areas that fund investors are running from and selling out of areas they are packing into works.

We discussed other reasons back in 2002 for liking the equity market going forward, one of which was we could barely find a fund investor who wanted to take on any risk at all. At the time the most popular fund categories were government bonds and money markets. In fact, late 2002 clocked in some of the worst months for stock fund inflows – investors pulled billions out of stock funds. This turned out to be the bottom of the last bear market. The next year saw all the areas investors wanted nothing to do with in 2002 get red hot again. 

We could go on and on about how emerging market bond yields are not high enough to warrant the increased risk today, or that junk bonds still have sizable default risk issues that investors are simply not getting rewarded for at current prices (unlike last year’s prices), or that tech stocks have a collective P/E ratio that is so absurd revenue and income growth would have to be astounding over the next five years to rationalize the prices. But the main reason we cashed out of some of our higher risk bets at the end of February is that fund investors are clamoring back in to the market in record numbers.

January 2004 saw one of the biggest inflows into stock funds on record, some $47 billion, surpassed only by February 2000’s inflow of $48.3 billion (which was one month before the market started a multi-year collapse). Conversely, one of the worst months for stock fund outflows was recorded in late 2002, the bottom of the bear market. The pattern is pretty clear, and it goes back far longer then the last few years. 

Does this mean the market (and our funds) can’t go up more? Absolutely not. It just means the risk adjusted returns of the market going forward (particularly some of the hot areas we have owned) are going to be lower, and that means we ease up a bit. 

Increasing our exposure to higher risk funds in 2002 was only half of the formula for success – we have to ease up when the time is right. Bottom line: take risks when nobody wants to, play it safe when everyone throws caution to the wind.