Showing posts with label total risk. Show all posts
Showing posts with label total risk. Show all posts

Sunday 14 June 2009

Assess risk - and then double it

Risk assessment is not always an exact science.

Occasionally, you hear of investors who have been hit by losses so big that they find themselves "out of the game". The reason is the same in every case: they have not managed their risk.

The most important is to always have a rough idea of how much money you could lose if the markets move against you, and you should be able to withstand that loss if necessary.

Risk assessment is not always an exact science. Judging how much a position can move against you will be nothing more than a gut feeling. It is difficult to be more scientific about it because:
  • using history as a guide is not always effective, as the world is always changing.
  • even in normal conditions, there is a lot of 'volatility of volatility', as the market goes throught quiet and crazy periods.
  • the size of the theoretical maximum loss is all of the investment, because a price can go to zero. But this is hardly expected to happen.
  • sophisticated statistical analysis has often proved inadequate, which is why LTCM had come unstuck.

Here is how one investor estimate his maximum loss for each position based on what he feel could happen in a normal environment. A normal environment is one which applies four years out of every five. (For every 5 years of investing, you can expect to meet 1 bear year.)

Step 1: He assumes for risk purposes:

  • Blue chip stocks will not fall by more than 25% in the four years out of five. So for every $100 invested in the big names, he could expect to lose $25.
  • Smaller stocks are normally more risky. These will not fall by more than 50% in the four years out of five. Therefore, for every $100 invested in a small stock, he was risking $50.

With this estimate for each position, he can simply add them all to get an idea of his total risk ($R). This gives an estimate of how much he could lose in reasonable circumstances - four years out of five.

Step 2: For the one bad year in five, it could be worse than that. The loss will be worse. For this reason, he assumes that he could possibly lose double that amount ($2R).

Step 3: Making some deduction to my total risk. It would be fair to expect that not all of his rainy days will happen together. This is the benefit from diversification. Therefore, he can make some deduction to his total risk if he feel not everything can go wrong at once. (But be careful, some big name hedge funds have come unstuck by underestimating how their positions are correlated.)

To summarise:

This simple and logical technique of risk assessment involves:

  • for each position, assess how bad a loss could be in a normal environment;
  • double the amounts; and
  • add up the potential losses, and take some off the total if it is justified by diversification.

The idea is to be comfortable with the total risk level. It is vital that you could withstand that loss, because a disaster may happen. So simply choose the size of positions so that potential losses are manageable. No market is too risky if the position is not too big.

With the right approach, you should be able to "stay in the game". Do not take too much risk.