Wednesday, October 5, 2011

Long-Term Investing Works!

We hear the phrase all the time: "Invest over the long term! Don't punt in the stock market!" Is that really a wise investment?


We started with a familiar market again - the Straits Times Index. Our period: 25 years starting from the year 1976 and ending in the year 2000. This period included the 1979 Oil Shock, the 1986 Singapore recession, the 1997 Asian Currency crisis, and other world events which had an impact to a certain degree on Singapore, which is so exposed to external influences.

We first took a tally of all the positive growth periods in the Straits Times Index on a 1-year basis. There were 15 out of 25 years in which the Straits Times Index had positive growth. This means that if you had randomly picked the start of any year from 1976 to 2000 to buy into the Straits Times Index and you had a holding period of 1 year, you would have had a 60% chance of coming out ahead.

We then tried the same experiment on a holding period of 3 years. The result was that out of twenty-three 3-year periods, eighteen of these were gains. So, if you had invested in the Straits Times Index at the start of any year from 1976 to 2000 with a time horizon of 3 years, you would have had a 78.26% chance of coming out ahead (quite a large improvement over 60% isn't it?).
We also experimented with 5 and 10-year holding periods. Take a look at the results in the following chart.

Chart 1

Source: S&P Micropal

As you can see, when you extend the holding period to 10 years, there is a 100% chance that you would have made money. The question now is, does this apply only to the Singapore stock market? What if we were to try the same experiment on other stock markets? Well, we did, on six other stock markets:

USA (represented by S&P 500)
Technology (represented by NASDAQ)
Europe (represented by MSCI Europe)
Taiwan (represented by the Taiwan Weighted Index)
Hong Kong (represented by Hang Seng)
Japan (represented by the Nikkei 225 Index)
World (represented by the MSCI World USD)

Chart 2

Source: S&P Micropal

As can be seen, the increasing percentages of positive periods for all the stock markets would seem to suggest that, at least historically, the longer your holding period, the greater the possibility of making money from an investment in one of the above equity markets.

In fact, in six of the markets shown above, you would have made money 100% of the time if your holding period were 10 years.

At this point, some may ask: "I can come out just 5% ahead after holding a Singapore fund for 10 years. That is a far cry from the 21.89% I would gain even if I were to keep it in a fixed deposit giving me 2% per annum!" That could be the case. After all, investing in equity markets are always riskier than investing in fixed income instruments.

So, we now choose to be more strict in our analysis. Instead of just having a positive gain, we choose to take the periods where the annualised gain (or average gain per year) is more than 4%. This is pretty reasonable since even the CPF special account can only promise 4% per year.

If an investment grew by 4% per year, after 3 years, it would have grown 12.49%. After 5 years, it would have grown 21.67%, and after 10 years, it would have grown 48%. This is because of the miracle of compounding (we won't go into that today as that isn't the focus of our study).


We also analysed the Straits Times Index with a stricter criterion. We counted periods where the returns were over 4% per year, over 12.5% for 3 years and more, over 22% for 5 years and more, and over 48% for 5 years and above. These would then be counted as a percentage against the total possible 1-year, 3-year, 5-year and 10-year periods over the last 25 years. The results are shown in Chart 3.

Once again, you can see that the rising trend over longer holding periods does not change. Also, even taking into account that our criterion is now far stricter, when you extend the holding period to 10 years, the probability that you would have gained a compounded 4% per year was 93.75%. Thirteen out of fifteen 10-year periods saw a gain of more than 93.75%, where the Singapore Straits Times index was concerned.

Just as with our initial analysis, we extended this to cover the six stock markets and the world index mentioned in Chart 2. The results are shown in Chart 4.

Charts 3 and 4

Source: S&P Micropal

The same rising trend can be seen for every stock market that we measured. For six of these stock markets, there was a higher than 90% probability that if you had invested in the stock market index at the beginning of any year since 1976, with a 10-year horizon, you would have made better than 4% per year. If you wanted to time the markets (meaning that you jumped into a market with all your money, then jumped out again after 1 year to hop into another market), then from a statistical point of view, your chances of making better than 4% per year would be significantly less and your transaction costs would be higher as well. We did not try to see if a shorter holding period of 6 months or less would give better results - we suspect that they won't.

One more interesting conclusion you might arrive from looking at Charts 4 and 2 is that the US stock market represented by the S&P 500 Index is a less volatile stock market than others. It had a 100% probability of having a 5-year gain, and also has one of the highest probabilities of having an average compounded 4% gain over the various periods measured. We believe that this is due to its low reliance on external fund flows. Most of the capital that make up the stocks in the US market are from US investors. Even during downtrends, US investors are not outward looking preferring to hold cash, or shift to bonds. This results in a more stable market environment where there are fewer drops, or spikes. As we mentioned in our previous article "Do Top Performing Markets Always Shine," there are times when other stock markets may outperform the US stock market. However, these stock markets may also be more volatile.


These analyses are statistical in nature and have their underlying limitations. The two main limitations would be analysing past performance data (which are not a guarantee for future returns), and not taking into account transaction costs.

Nevertheless, this is balanced by two considerations. Regarding the limitation of looking at past data, the figures taken were from an extremely long period of 25 years, during which there were quite a few wars (especially in Middle East), there was a global oil shock, and recession. So, if you are a believer that history repeats itself, then you will also come to the same conclusion: Long-term investing works far better than short-term punting.

The limitation of transaction costs is balanced by the fact that we were measuring only the indices and not active funds that invested in these markets (most of the funds in Singapore do not have such long histories). If you choose a good fund manager, one that is able to outperform the market index he invests in over the long run, you will not only be able to cover any possible transaction costs, you may even have gains better than the market index.

In conclusion, we first state that our objective here was not to find the best stock market to invest in. We recognise that each stock market has its peculiarities, and some have historically been more volatile than others. However, based on our analyses, we can at least discover a common trend among stock markets regardless of their size, volatility and other attributes. This common trend is: Long-term investing has a higher statistical chance of giving returns than short term investing.

No comments:

Post a Comment

Related Posts Plugin for WordPress, Blogger...