Past performance is no guarantee of future performance... but is anyone listening?

Most mutual funds end their advertisements with this statement: "past performance is no guarantee of future results". I don't know why they bother because investors don't seem to act like they care. In fact, one phenomenon that we know characterizes investors is that many of them try to invest in whatever asset class, fund or stock has done well in the past.

I was reminded of this return chasing phenomenon by an article I read on Permanent Portfolio, a fund that has been around a while but has generally struggled to survive for most of its life, with about $ 50 million in money under management a few years ago.  Starting in 2007, the fund's assets have exploded and it is now up to $ 5 billion. Its growth coincides with its superb performance over the period.
http://online.wsj.com/article/SB10001424052748704858304575497662644720960.html?mod=WSJ_hps_MIDDLEThirdNews
So what? Only about a third of the fund is invested in stocks; 25% is invested in gold and silver and 35% in cash. Its performance over the last three years can be explained largely by the fact that it is under weighted in stocks and over weighted in precious metals. In other words, its success comes almost entirely from its asset allocation, rather than asset selection.

History provides a cautionary note on why return chasing often comes to a bad end. Momentum investing, which is what this comprises, requires two components to succeed:
1. Long runs in returns. You need asset prices to move in the same direction for long periods. If good years are followed by bad years and vice versa, you will end up with whip lash as a momentum investor. In statistical terms, you need strong positive correlation in returns across time. Whether there are long runs in returns is an empirical question and there is evidence that is hopeful. There have been long historical runs (positive and negative) in returns in stock and bond markets and even longer runs in currency and commodity markets. The evidence is less positive when it comes to individual stocks; while there is some short term momentum, it is much weaker than in the asset marekts. Thus suggests that momentum investing is more likely to pay off for market timers playing the asset allocation game than for individual stock pickers.

2. Detecting the end of the runs: All good things come to an end and long runs in prices (up or down) end at some point in time, with the size of the correction directly proportional to how long the run lasted in the first place. A momentum investor can see years of positive returns wiped out in the course of a few weeks or even days. The evidence is not as positive on this factor. Models and investors that claim to detect imminent market corrections don't do very well, at least in the long term.  However, there may still be hope. I am not a great believer in technical analysis, but this is is one place where price and volume charts may help, especially in assessing how close to the cliff you are. I have a chapter on momentum investing in my investment philosophies book that you may find interesting and you can download it by clicking below.
http://www.stern.nyu.edu/~adamodar/pdfiles/invphil/ch7.pdf

If you are a momentum investor, making money on your gold investment right now, I am happy for you. I just hope that you have a mechanism that  will tell you when it is time to get out. With momentum investing, knowing when to sell is even more important than knowing when to buy.