Just how safe are you investing in the JSE? Your safest and lowest risk option (other than sticking your money under a mattress) is to stick your money in a fixed deposit or bank account. But doing this means you barely keep up with inflation, and depending on how much you are investing, you may even be eroding your money after inflation and income taxes take their toll.

Alternatively, you can invest in the JSE. The graph below shows the accumulated returns over the last 10 years of these two options if you had invested R1.00 since 1998. A daily FNB Call-rate was used to calculate the Bank option and we assumed the JSE investment was in the ALSH index.

The JSE beats the Bank even after the massive Crash of 2008, where the JSE fell more than 40%. In the medium to long run, it would seem you are far better off with the JSE. But higher returns must mean higher risk right? Well just how risky is the JSE? We set out to find the answer...

PROBABILITY OF WINNING OVER LAST 50 YEARS

We looked at the JSE ALSH (All-Share) index over the last fifty (50) years. This gave us the closing price for the index for 12,588 days. For

Since we are using such a large and statistically significant sample set (12,588 data points), this effectively gives us an indication of the

The below chart shows what percentage of various short-term holding strategies managed to produce in excess of the growth shown along the x-axis. For example, 60.7% of the 12,588 1-month portfolios (shown by the blue line) managed to show more than or equal to 0% growth. This means, that if you invested in the JSE on ANY day in the last 50 years you would have had a 60.7% chance of NOT LOSING MONEY after a month. Similarly 9,195 one-year investments out of the total of 12,588 showed >= 0% growth, which means 73% probability of not losing money.

We see from the short term chart above that it would appear that you have a better chance (probability, shown by the vertical axis) of exceeding a certain growth target (shown on the horizontal axis) the longer you hold your investment. We now look at longer term holding periods. Since we are looking at whole years, and to allow us to compare apples to apples, we will use the compound annual growth rate (annualised total growth) as our growth targets.

This extends upon the short term back-tests, showing us that the longer you hold your investment for, the greater your chances of meeting your growth targets. If you hold your investment for 1 year, you have a 73% probability of not losing money. But if you hold it for 3 years, that probability jumps up to 87.3%, and holding for 5 years, you have a 95.4% chance of not losing money.

But wait, something strange happens if you want to exceed 12% growth per annum. The theory flips around! If you want higher per annum growth rates, you will have to hold for shorter periods! If you seek 30% growth per annum your best chances are with a 1 year holding strategy which showed a 27.2% propensity to beat 30% over the last 50 years, as opposed to the 3 year holding period which only beat 30% growth per annum 13% of the time. The 1-year strategy has double the probability of exceeding 30% growth per annum than the 3 year strategy!

Remember that we have not included brokerage charges and dividends in the above examples. Brokerage will effect returns by a maximum of 2% (1% for the initial buying and 1% for the selling, provided your portfolios are in excess of R10,000) and is not that material. Dividends on the other hand, can make a very material but positive difference, especially for longer holding periods.

BEST RISK/REWARD HOLDING PERIODS

The next question we seek to answer is "what were the best risk adjusted holding periods over the last 50 years?" The below chart shows the average compounded growth of each holding period taken over 50 years, together with the standard deviation and the "risk/return ratio" (average growth divided by standard deviation).

We see that although the 1-year holding periods delivered on average the higher annual growth rate, it was accompanied by far higher standard deviation resulting in the highest risk ratio. Ideally an investor wants an R/R ratio of greater than 0.8 and this only happens after 2 years. The 3 year holding strategy delivered an average compound of 14.0% on a deviation of 13.2% which gave a risk ratio of 1.07. Given that after 3 years you only incur capital gains taxes of 10% versus income taxes of 40% on your profits, it would appear then that 3 years holding periods are your "sweet spot" for minimum holding period. The maximum "sweet spot" is 6 years after which the risk/reward ratio "tilts down" and delivers diminishing returns.

SOME MORE INTERESTING PROBABILITY CURVES

CONCLUSION

The JSE is far less risky than you think. You can use time in the market to significantly reduce your risk

Holding shares in the ALSH index for 6 years delivers the best risk/reward ratio. The six year strategies offer the best ratio of returns to deviation of returns, the lowest risk of losing your money and the highest probability of meeting your growth objectives.

The above tests over the last 50 years certainly prove that whilst "timing in the market" can lower your risks (ie only buy when things are cheap and sell when they are expensive), its "time in the market" that can make the greatest impact on lowering risk and maximising growth probabilities.

Making investments when shares are cheap (such as now) and coupling this with a Market Timing Strategy and "time in the market" (>2-3 years)

We have completed a similar exercise as the above, but for individual JSE shares, which you can view in our JSE RiskCurves pages.

Alternatively, you can invest in the JSE. The graph below shows the accumulated returns over the last 10 years of these two options if you had invested R1.00 since 1998. A daily FNB Call-rate was used to calculate the Bank option and we assumed the JSE investment was in the ALSH index.

The JSE beats the Bank even after the massive Crash of 2008, where the JSE fell more than 40%. In the medium to long run, it would seem you are far better off with the JSE. But higher returns must mean higher risk right? Well just how risky is the JSE? We set out to find the answer...

PROBABILITY OF WINNING OVER LAST 50 YEARS

We looked at the JSE ALSH (All-Share) index over the last fifty (50) years. This gave us the closing price for the index for 12,588 days. For

*each*of these days we started an investment in the ALSH and then looked at how each investment would have grown after 1,2,3,6 and 12 months, as well as 1,2,3,...10 years (the holding periods) For each holding period we then tallied up the percentage of investments that managed to beat various growth rates ranging from 0% (ie what % of investments managed to not make a loss) through to 30%.Since we are using such a large and statistically significant sample set (12,588 data points), this effectively gives us an indication of the

*probability*of an investment in the ALSH beating a particular growth rate. We then plotted*Probability Curves*on the charts below.The below chart shows what percentage of various short-term holding strategies managed to produce in excess of the growth shown along the x-axis. For example, 60.7% of the 12,588 1-month portfolios (shown by the blue line) managed to show more than or equal to 0% growth. This means, that if you invested in the JSE on ANY day in the last 50 years you would have had a 60.7% chance of NOT LOSING MONEY after a month. Similarly 9,195 one-year investments out of the total of 12,588 showed >= 0% growth, which means 73% probability of not losing money.

We see from the short term chart above that it would appear that you have a better chance (probability, shown by the vertical axis) of exceeding a certain growth target (shown on the horizontal axis) the longer you hold your investment. We now look at longer term holding periods. Since we are looking at whole years, and to allow us to compare apples to apples, we will use the compound annual growth rate (annualised total growth) as our growth targets.

This extends upon the short term back-tests, showing us that the longer you hold your investment for, the greater your chances of meeting your growth targets. If you hold your investment for 1 year, you have a 73% probability of not losing money. But if you hold it for 3 years, that probability jumps up to 87.3%, and holding for 5 years, you have a 95.4% chance of not losing money.

But wait, something strange happens if you want to exceed 12% growth per annum. The theory flips around! If you want higher per annum growth rates, you will have to hold for shorter periods! If you seek 30% growth per annum your best chances are with a 1 year holding strategy which showed a 27.2% propensity to beat 30% over the last 50 years, as opposed to the 3 year holding period which only beat 30% growth per annum 13% of the time. The 1-year strategy has double the probability of exceeding 30% growth per annum than the 3 year strategy!

Remember that we have not included brokerage charges and dividends in the above examples. Brokerage will effect returns by a maximum of 2% (1% for the initial buying and 1% for the selling, provided your portfolios are in excess of R10,000) and is not that material. Dividends on the other hand, can make a very material but positive difference, especially for longer holding periods.

BEST RISK/REWARD HOLDING PERIODS

The next question we seek to answer is "what were the best risk adjusted holding periods over the last 50 years?" The below chart shows the average compounded growth of each holding period taken over 50 years, together with the standard deviation and the "risk/return ratio" (average growth divided by standard deviation).

We see that although the 1-year holding periods delivered on average the higher annual growth rate, it was accompanied by far higher standard deviation resulting in the highest risk ratio. Ideally an investor wants an R/R ratio of greater than 0.8 and this only happens after 2 years. The 3 year holding strategy delivered an average compound of 14.0% on a deviation of 13.2% which gave a risk ratio of 1.07. Given that after 3 years you only incur capital gains taxes of 10% versus income taxes of 40% on your profits, it would appear then that 3 years holding periods are your "sweet spot" for minimum holding period. The maximum "sweet spot" is 6 years after which the risk/reward ratio "tilts down" and delivers diminishing returns.

SOME MORE INTERESTING PROBABILITY CURVES

CONCLUSION

The JSE is far less risky than you think. You can use time in the market to significantly reduce your risk

*and*boost your returns.Holding shares in the ALSH index for 6 years delivers the best risk/reward ratio. The six year strategies offer the best ratio of returns to deviation of returns, the lowest risk of losing your money and the highest probability of meeting your growth objectives.

The above tests over the last 50 years certainly prove that whilst "timing in the market" can lower your risks (ie only buy when things are cheap and sell when they are expensive), its "time in the market" that can make the greatest impact on lowering risk and maximising growth probabilities.

Making investments when shares are cheap (such as now) and coupling this with a Market Timing Strategy and "time in the market" (>2-3 years)

*virtually eliminates your risk of losing money*. Given the magnitude of the upside in historical JSE returns versus leaving money in the bank, the risks of playing on the JSE are certainly worth it.We have completed a similar exercise as the above, but for individual JSE shares, which you can view in our JSE RiskCurves pages.