37. The Dirty Little Secret of the Dow Jones Industrial Index

It's the second oldest index around: since its inception in 1896 it saw countless market cycles, from the great depression to the booms and busts cycles of recent years. It's often the first (or only) quoted index in news reports about the market. And it, supposedly, tracks the largest companies in the US - the bellwethers of the economy.

What can be wrong with that?

The dirty little secret is that the Dow Jones Industrial Average (DJIA) is a pretty crappy index, and a lousy way to measure market performance.

For a start, it tracks only 30 companies. Indeed, all are pretty big, from ExxonMobile to Bank of America, from Disney to Microsoft. But these are not even the largest companies in the US. In fact, the DJIA does not include Apple, which competes with ExxonMobile for the top spot. With so few companies in the index, can you imagine what including Apple would have done to the performance of the index in the last year?

Well, it's not difficult to figure this out, because the DJIA is a price-weighted index. In a nutshell, the value of the index is the sum of the prices of all the stocks in the index, multiplied by a factor, currently about 7.5. (The factor is adjusted when the companies are added or removed from the index.) If Apple were to be included in the index, the company meteoric rise from $200 to around $400 a share in recent years would have added 1500 points to the DJIA! Similarly, AIG stock's 90% collapse in 2008 contributed to a roughly 3,000 point drop in the index, only because the stock price was in the hundreds. Today, after a reverse-split, the stock trades around $22. A similar drop would contribute only to a 150 point drop in the index.

Even worse, the companies in the index have very different share prices. While ExxonMobile (XOM) is traded at $74 a share, Bank of America (BOFA) is traded around $6. This means that a 20% decline in BOFA accompanied by 2% appreciation of XOM will result in a modest gain to the index. If BOFA went bust tomorrow morning and its stock price went to 0, the direct impact on the DJIA would be a drop of 45 points.

Does it make sense to you? It doesn't make any sense to me. But yet, millions are watching this index and its movements religiously. Professionals, however, are likelier to use the S&P 500 as a benchmark for the market and the economy: it contains the largest 500 companies, and is cap-weighted, i.e., the contribution of different companies are weighted according to their total market value. Even fancy modifications of this index, like the equal-weight S&P I mentioned last week, are better than the ridiculous DJIA.

One reason people still use this aging monster is that, against all odds, it's highly correlated with the S&P 500 index. Check this chart, for example:

In fact, the picture from this graph shows that the DJIA is only 10% off the S&P over the last 5 years. Not bad for an index with such a questionable composition. Perhaps this shows how interrelated our economy has become, and how we all go up or go down together. Or perhaps I'm missing something, a magical touch that makes the DJIA the venerable index it is today. What do you think?

36. Index Games

I recently received a Consumer Reports publication including an article titled "New twists in index funds". The twist they present is not that new. What was new to me was the degree of deception in a publication I usually trust.

What is it all about?

Index funds, like their name, track an index of stocks. A reliable index for the US economy is the S&P 500, which tracks the performance of the biggest 500 companies in the US. It is market-cap weighted - in other words, it corresponds to the total value of the biggest 500 companies. Naturally, movement in bigger companies matter more than smaller companies: If ExxonMobile (the biggest component of the index) increases 10% in value, the index will move by a couple of percentage points. If the smallest company in the index moves by 10%, the index will badge very little. This makes sense to me - the economy of the US is more ExxonMobile, IBM and Apple than AK Steel Holding and Monster Worldwide (whose market cap is less than 1% of the the top components).

But, apparently, CR thinks otherwise:

Equal-weighted S&P 500 is a recent invention: take all 500 companies in the index, and weigh them equally. ExxonMobile is now as important as AK Steel Holding. Apple fortunes are as significant as Monster Worldwide.

Does it make any sense? Nope. But look at the graph! The blue line is higher than the red line! Must be a good thing, not so?

Kind of. As long as you bought before 2008 and sold now. I'm surprised that CR relies on such meager data to draw conclusions on the benefit of equal weighting. In fact, if you look at the graph carefully you'll see that between 2003 and 2008 the two indexes had identical performance, and in the last year the traditional cap-weighted S&P 500 is slightly better. The Equal-weighted version had the upper hand in two years - not a significant period of time for any long-term investor. And it can be easily explained by the observation that the modified index is heavily tilted towards small-cap companies. And yes, in some years small-cap will give better results than large-cap, in the same way that in some years large-cap is better than small-cap, banks are better/worse than technology, foreign markets are better/worse than domestic markets, etc. You can never predict these movements in advance, and a couple of good years for small-cap stocks don't say that they're superior. In fact, long-term research shows that although small-cap stocks, over the long run, have higher return than large-cap, they also carry a much higher risk. I would advice against anchoring anyone's portfolio on small-caps.

The solution - in my opinion - is to continue to diversify and invest in everything, from large cap to small cap, through more varied indexes such as the Wilshire 5000. Or invest in the S&P cap-weighted 500, if you prefer the increased stability of large-caps. But don't fall pray to short-term fads and tricks like the equal weight fashion. Besides paying higher fees for the novelty, you're not going to gain much.

Next week: yet another way to build an index, crazier than the two mentioned above, and yet used in the most popular index of all. Stay tuned!

35. Upside Down

Sometimes, it's worth looking at the world upside down.

When you look at falling stock prices, you should ask yourself: Am I a buyer or a seller?

If you're an investor, like me, you're a buyer. In this case, why should you be worried about falling prices? A market crash just means that stocks are cheaper, and you'll get more shares for you hard earned money. In fact, annual rebalancing and periodic, consistent investments force you, in a way, to buy low and sell high.

To illustrate this point, look at the stock market in the last 5 years. You can see the chart here:

The DJ opened at 12,090 at the beginning of this period, and closed at 11,433, or down 5.4%. Pretty dismal, all in all: a $100 invested in Sep 12, 2006 would give you $94.60 in Sep 12, 2011.

Now, suppose that instead of investing these $100 all at once, you invested $20 every year on Sep 12th. The DJ was reading the following values (approximately) on 2007 through 2010: 13,443, 11,388, 9,820, and 10,608. Even though the DJ ended up below its value at the of the period, your total investments will be worth $100.84, or 6.6% more than the alternative of buying and forgetting. Note that the time I selected was quite random, based on today. An investor who chose March 23rd as her annual day would have fared much better. An annual rebalancing against a conservative bonds funds would have had even more dramatic results: assuming a 60/40 balance and 0% growth for the bonds funds, your $100 would have been worth $120.80 at the end of the period: a gain of more than 20% in basically a flat market. Not bad at all! (I'm neglecting the value added from the bonds funds and also the commissions - these will not influence the result greatly.)

Why did this happen? Because in some years (such as the wonderful 2009), the market crash made stocks so cheap that your $20 got you much more than in so called "good" years. Furthermore, if you followed a rebalancing plan, you had to buy a lot of stocks this year. You've made market fluctuations and instability work for you.

In a sense, an investor who's not going to spend the money in the coming 5-10 years should hope for a market crash, in the same way that a renter who's looking to buy a house should be happy with the house bubble crash. As long as you believe that the long term prospects of the US economy are goods, you're going to win.

Let the sharks of Wall Street and hedge fund managers worry about falling prices. For us simple people, as Warren Buffet famously said, there's no reason to worry if hamburgers - or stocks - get cheaper.

34. What is your ROI?

ROI in the business world stands for Return on Investment. In simple terms, this is your gain as a percentage of your investment.

Mutual Funds often boast spectacular ROIs. Sites like morningstar.com use these ROIs to rank funds, giving 5 stars for top performers.

Readers of this column already know where I'm headed: of course, 5-star top funds are a trap. It's confusing and counter-intuitive: a tennis player who's won almost every match in the last 5 years is likelier to win the US Open than a player who lost half her games this year; a business that's been profitable every year for the last 10 years is liklier to be profitable this year as well; why isn't it the case with mutual funds?

There are many reasons for that. The main one is the famous "Past performance is no guarantee for future performance" warning, which appears in every fund prospectus (maybe they should get more graphic, like the new warnings on cigarette packs?) Statistically, there's no correlation between past success and future success. Investments that do very well one year can be a dog this year (Internet stocks anyone?)

Another reason is that funds may luck out: a fund manager with 10 funds or a 100 funds can choose the most successful one and promote it (while perhaps closing the least successful ones to improve his overall record.) A maximum of a hundred random results is a pretty high number, but it's still random and does not reflect any intrinsic special quality.

Perhaps the most mysterious aspect of ROI is that the ROI of a fund is not your ROI. Take for instance an Internet fund, established in 1995 and look at its record at 2005. You won't be surprised to hear the fund had meagre results: after all, the .com bust of the early years of the last decade erased 98% of the value of some companies (pets.com anyone?). Still, since the fund was incorporated in 1995, it managed to see the bubble inflate before bursting. The overall ROI over this 10 year period was positive, around 5% total. Pretty dismal, but not nearly as bad as the average ROI for an investor in the fund. As you can expect, as the .com bubble grew, more investors joined the party. While very few invested in .com stocks in 1995, by 1999 and 2000 billions of dollars poured into the market, and most investors bought these funds when prices were beyond laughable. These investors lost almost all their investments when the market crashed. The average return for an investor in this fund is about -95%.

The moral of the story? If a fund boasts ROI of 50% over the last 2 years, by all means, jump on the bandwagon and buy it. But only if you can buy it retroactively, starting 2 years ago. Buying it today will be as good (or as bad) as buying a fund that lost 50% over the last 2 years. And how confident will you feel doing that?

33. Reading the moostars

One of the weirdest sites out there is Decision Moose. This web moosite uses moosignals to time the market, and switch allocations between gold, bonds, US market stocks, etc. With a total of 9 mooselections, the head moose makes a weekly moosecall to divert all assets to one of these options, or stay put.

What can I say. The terminology is dumbfounding, and it contradicts everything I believe in. But I'll give this guy credit for two things: first, he managed to time the market pretty well over the last 10 years. Second, he's pretty honest about the entire venture. From the FAQ:

DOES MARKET TIMING WORK? 

Market timing is unproven. That said, every mutual fund salesman you'll meet-- except maybe the Vanguard 500 guy-- would have you believe that his fund manager is a better stock picker than anyone else in the world, and although few like to mention it, good timing is implicit in good picking. On the other side, academia continues to go to great lengths to disprove timing and promote diversified buy-and-hold investing. The controversy, then, is between a group with considerable practical experience, but a vested interest in timing's success (financial professionals who want to sell their expertise), and a group with no practical experience, but also no particular vested interest (academicians). Obviously, the creator of Decision Moose, a financial professional, thinks timing may indeed work, or he wouldn't be wasting his time on a site devoted to benchmarking a timing mechanism to prove its validity.

My main problem with this site is the old Anthropic Principle: our observations influence what we see via a selection bias; we wouldn't see the world as it is if we were not here, watching it. For example, this answers the question of why Earth is in such a perfect position relative to the sun: slightly closer, and temperatures will be too hot. Slightly further away, and it would be too cold. The anthropic principle simply says that if Earth was not in this exact location, we wouldn't be here to wonder about it - it's not a case of exceptional luck (or divine provenance, or intelligent design). Out of a billion planets, only the ones that develop life will wonder how come they happen to live in a planet with the perfect conditions for developing life.

How is this related to market timing?

Imagine a thousand brokers trying to time the market, each with his or her moose site. Every year, half of them will have above average results - by pure luck. The other half will have below average results, realizing they're no good at it and shut down their site.

After 10 years, you'll have on average about 1 moose site that managed to switch allocations consistently well, and beat the market every year. And it will be the only moose site around.

But this doesn't prove that next year it will have more than 50% chance of beating the market.

If you're not convinced, just look at analysts predictions: at any given day, about half will tell you to buy and half will tell you to sell. I don't feel I can trust either one. Half say we're headed to a terrible recession, and half say the worst is behind us. I just can't trust them, or even our dear, honest moose. But if you do, good luck and let's see in another 10 years how you've been doing!