Bill Rempel, a.k.a. NO DooDahs!

It wasn’t all that hard to do, either. Frankly, when multiple entries show up for something “sexy” like this “lost decade” thing, it almost demands a de-bunking. Here are some things to remember:

(1) the major indices are nothing more than semi-discretionary domestic U.S. large-cap mutual funds, run by committees. Unless you long-term trade the majority of your stake – ooops! “invest” the majority of your assets – in an index fund, or in a fund that benchmarks itself to one of these indices, the indices are nothing more than a potential relative benchmark for your performance. By the way, if the majority of your stake is in a U.S. stock market index fund, you’re an idiot! With foreign developed and emerging markets, commodities, REITs, and foreign and domestic bonds all easily available through mutual funds and ETFs, there is no excuse for passive players to not be diversified!

(2) the general adage that stocks outperform bonds, real estate, and commodities, is based on many different multi-year samples. The presence of one or several 10-year samples wherein that doesn’t hold, is not a disproof of the statistical fact that stocks, over any randomly-selected multi-year period, are among the best possible places for parking money, and for the passive player, they are an essential part of the picture.

Re: (1) and (2), keep in mind that, for active players, the indices’ performances are irrelevant.

(3) presenting the fact that the last so-many-days/months/years have underperformed, without presenting contextual information or actionable strategies, is the kind of “infornography” that gets read by the “Bloomers Babe” in a 15-second slot from the trading floor, but doesn’t do a damn thing to help anybody.

Without further ado, here’s some useful information about 10-year returns on the indices.

With monthly data on the Dow, there are 714 month-end data points where I can observe both a trailing 10-year annualized return, and a forward-looking 10-year annualized return. For the S&P 500, with it’s shorter history (that’s right, the S&P 500 didn’t exist before the 1950’s, so take those gurus with their S&P 500 charts going back to Stone-Age times with a grain of salt!), there are only 459 such data points. For example, today’s data doesn’t make the sample, because I don’t have the subsequent 10-year annualized return number yet, and the data point for Jan 1998 will have prior annualized returns from Jan 1988 and subsequent annualized returns to Jan 2008. No dividends were included.

There’s no linear statistical relationship between the prior 10-year annualized return and the subsequent 10-year annualized return. There is, however, for the S&P 500, and for the Dow (during the time that that S&P 500 also existed), a second-order relationship and a “weirdness” about the middle-distribution. If the charts are truncated in your browser, click them for a better view. I tested:

the entirety of the Dow history,

the S&P 500 history, and

the Dow for the period in which the S&P 500 existed.

The difference is one of timeframe. There are many economic and monetary differences between the various eras, including massive world wars and a fairly strict dollar-gold peg in the pre-1950 years, going through increasing globalization, the Bretton Woods era, and now the days of free-floating (some would say free-falling) fiat currencies, not to mention different approaches to monetary policy, in the U.S. and abroad. These many differences mean that people who draw tight comparisons between eras should be flogged – or at least, not listened to overmuch, but I’d prefer to see them flogged.

The pattern in the post-WWII is that a recent weak or strong decade tends to be followed by … a strong decade; that is, strong (but not insanely strong) decades are reinforcing, and weak decades are mean-reverting. Of the strong decades, only the exceptionally strong decades tend to have a lull after them; and the mid-range return decades tend to be followed by one extreme or the other.

I then looked at the 834 data points where I had 10-year trailing annualized returns for the Dow, and categorized them. They are rounded to the nearest percent for this histogram:

You can note that the distribution is bimodal, with 10% and 3% being points of interest. If you consider the last 10 years to be “lost,” then a +3.6% annualized is “lost” to you. Remember that this doesn’t include a dividend yield!

In terms of where the last 10 years fits on the distribution, STOP YER WHINING! Roughly 32% of the time, the Dow performs worse than it did over the last decade. Boo-hoo!

That chart also shows the reason why stocks are so powerful an investment tool in the long term; 86% of the time, if you have a 10-year window, the stock indices are practically as safe as bonds! That is, take any random starting point, and wait ten years, you get your principle back 86% of the time, while being paid dividends, not to mention having a pretty darn good chance of capital appreciation, and that’s just buying the index!

The S&P 500 distribution of 579 data points is similar, although

it appears to be trimodal, and truncated at the negative end. This is because there wasn’t an S&P 500 in the 1930s during the Great Depression! (Remember that the next time some bozo shows you a 150-year chart of the S&P, and ping them for the detail on their *estimation* of who the S&P members were, back in the Jurassic era) If there were, I bet the distribution would be almost exactly the same as the Dow’s. The trailing 10-year return of +2.1% annualized (no dividends!) looks worse in context, precisely because it covers the more recent history; about 12% of the possible 10-year periods were worse than the most recent 10-year period, which really makes the recent past unexceptional.

Final Thoughts on “Lost Decades”

RELAX! The worst that can be said (if you look at 70+ years of history, not advisable in my book) is that the last 10 years of returns on the index don’t mean jack squat for predicting the next 10 years of return on the index. The best that can be said is, in the modern post-WWII era, low-returning decades like the one just past are typically followed by average to above-average decades. What’s to worry?

The indices’ performances are irrelevant as benchmarks, anyway. Your system, your asset allocation, and your personal triumph over the fear and greed responses you have to the market, will be the REAL determinants of your success.

The last 10 years were neither all that bad, nor all that unusual. The average passive schlump’s best protection against year-to-year or decade-to-decade variations in index returns is to diversify, and keep in mind the time-frames for which their money is “locked up” in stocks. By “diversify” I mean more than just allocate to domestic U.S. bonds; I mean foreign stocks, emerging market stocks, foreign bonds, and commodities – the WORKS! It’s exceptionally easy to do in today’s world.

The above-average active schlump has so many tools available that the indices’ performances are completely irrelevant to them. Many of them are available here, on this blog. Enjoy!

If you liked this post, you might be interested in subscribing to my RSS feed. There’s a convenient “Add to Any” icon on the top of the right sidebar. If you prefer, you can get a nightly RSS email update sent on the days that I post! You might also be interested in The Rempel Report, where I detail the results of my mechanical systems trading!