Blog #46 Fool’s Game

Most of us aren’t investment geniuses, so a passive “buy-and-hold” strategy will beat “active” investing hands down most of the time. But we are restless creatures; buy-and-hold sounds like a boring cop-out.

Active investment is about market timing and stock selection. It is the stuff that mutual fund and hedge fund managers do for the fat fees they charge. Legions of ordinary investors also try to pick stocks and predict markets on their own.  How these individual and professional investors actually perform has piqued the curiosity of academic researches for decades. Fortunately, we do have a trove of hard evidence on the usefulness or otherwise of active investing from studies of mutual funds, and to a lesser extent, of individual investors. What evidence there is shows one thing – it is hard to consistently beat a buy-and-hold strategy once you factor the costs. My focus today is on market timing.

Market timing is irresistible. Who can resist feeling bullish about the market after hearing the evening news that the STI has reached another record high? The term “trend chasers” describes investors who trade on momentum – buying after market rises and selling after market falls. Psychology research shows that our brains are ‘hardwired’ to see patterns even in randomness. Trend chasing is an example of behavior arising from our tendency to extrapolate patterns from the past.

The opposite of trend chasing is contrarian investing. Contrarians are more inclined to buy stocks after the market has fallen and sell them after the market has risen. Because contrarians go against the crowd, active contrarians feel often smug, believing their trading style be a cleverer strategy than following the trend. Of course, some investors are both momentum and contrarian traders, just at different times.

Regardless of whether you are momentum or contrarian, as long are trying to predict the market, sectors, or individual stocks to trade than invest for the long term, you are a market timer.  You are implicitly saying, “who gives a #$@% about the long term? I just want to make a quick buck.”

Is that a problem?  I would say YES if market timing forms the core of your investment strategy. The reason is simple. Markets aren’t that easy to conquer. Especially in the short run (days, weeks, even months), stock prices are pulled by a multitude of factors, including observable fundamentals (GDP, inflation, interest rates and the like) and market sentiments (much harder to measure). The interplay of these factors explain why stock prices zig and zig randomly from day to day, making consistent accurate predictions very hard. In that sense, market timing isn’t that different from betting in casinos where the odds are stacked in favor of the house. Similarly, market timing is a fool’s game because the odds of winning are stacked against you. While it is true that for every loser, there is a winner, randomness implies that nobody will be consistent winner. Randomness also means that occasional wins may be due to pure luck rather than prediction skills.

For more direct evidence on market timing, I will now share with you the findings of a detailed research study. The original study was done in the US some years back. The current extends this research using more recent data (to 2016).

The study looks for evidence on the effectiveness of market timing by comparing two concepts of average return over a sample period. The first average return is called buy-and-hold average return (BHR). The second average return is called dollar-weighted averaged return (DWR).

BHR is the return one earns on average by following a passive buy-and-hold strategy which does not attempt market timing. In the study, we assumed that investors buy a particular market index at the start of the sample period and hold this index until the end of the sample period. Details of sample periods are discussed below.

DWR gives more weight to periods when more money is held in the stock market and less weight in other periods. Like BHR, we compute DWR for the whole market comprising all types of investors. Some of these investors may be momentum investors, while others are contrarians. It doesn’t matter which group they belong to. DWR will incorporate the ‘net effect’ of these investment styles into the calculations. The important thing is that DWR will be different from BHR if investors as a whole deviate from a passive buy-and-hold strategy.

The key question we want to know is this: on the whole, is DWR higher or lower than BHR?  If DWR beats BHR, hats off to market timers!  However, if BHR beats DWR, then this result confirms that market timing is a bad idea. The phrase “on the whole” means we are talking about all investors in the aggregate and not singling out individual investors or even subsets of investors.

The data for the study consists of monthly values of market capitalisation and returns for the following markets:

  • NYSE-AMEX
  • NASDAQ
  • World
  • Europe
  • Asia
  • Japan
  • Singapore
  • Hong Kong

The sample period is 1926 to 2016 for NYSE-AMEX and 1974 to 2016 for NASDAQ. For the non-US markets, the data runs from 1970 to 2016. All data is monthly and are obtained from Thomson Reuters Datastream.

The table below shows the results where the average returns are presented as annual figures. The last column (p-values) are the results of a statistical test to check whether any difference between BHR and DWR are real or simply due to chance. A p-value of less than 0.05 indicates that the difference is real.

Dichev_1.jpg

The table is self explanatory. It shows that BHR is always greater than DWR (i.e. all markets and across different sample periods).  For the big NYSE-AMEX market, the difference between BHR and DWR is 1.24% a year. This may not seem much, but trading cost has not been factored into the calculations and such costs will significantly drag down DWR. Also, 1.24% compounded on a sum of $100,000 over 20 years gives a dollar return of almost $28,000, a non-trivial sum.

The table also shows that market timing is particularly damaging for certain markets like NASDAQ, HK, Japan and Europe where the annual difference between BHR and DWR is more than 2%. In all cases, p-values below 0.05, indicating that the return differences are not due to chance.

We can say more about investors from the data. The next figure shows the relationship between past returns and future money flowing into the US stock market. As shown by the straight line, this relationship is positive, meaning that on the whole, investors are trend chasers.

Dichev_2.jpg

So, if market timing is so bad, why do so many investors do it?

I believe a big part of the answer is that we are born ‘pattern seekers’. All of us are hardwired to predict. This instinct to seek patterns is something that nature has endowed us over the millennia for survival. Think of our cave men ancestors roaming the savannahs for food. For them, the ability to differentiate food from predators, to predict when and where each group gathers is clearly life critical. The mistake of modern humans is to assume that what is easy to predict in the natural world is also easy to do so in a world ruled by abstract concepts like stock markets, currencies, and economic fundamentals. In doing so, we underestimate the impact of uncertainty so central to financial markets, and confuse randomness for systematic patterns in the desperate search for superior returns. As Gandhi once said: “To those who are desperately hungry, God appears in the form of bread.” (paraphrased).

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

w

Connecting to %s