Effect of volatility on rebalancing corridors

Schweser SS 16, Reading 46, LOS 46 f, p 152 “Generally speaking, the higher the volatility of the rest of the portfolio (in relation to the asset class in question), the smaller the corridor.” Parenthesis added above our mine not schweser. I think I understand this, but want to here others thoughts.

all they are saying is that if the corridor is let’s say 5% an usual movement are 4% ( high volatility) you have a good chance from 4% overweight to jump to 8% overweight pretty quick ( i know it’s a stupid example)

i had probs with this concept also - i guess you are more concerned with the increased downside risk of having more vol, hence higher vol should have lower corridor. That being said - this must be weighed up with the cost of rebalancing this stock (ie if illiquid)

This is somewhat counter intuitive IMO. My original thinking when I read this was: If the volatility (of the asset or the rest of the portfolio) is high, you would want wider corridors. This would avoid frequent rebalancing (high trading costs). However, the concept purports differently: If volatility is higher you would want tighter corridors in order to avoid asset classes from deviating greatly from desired exposures. Coming to your specific point on the ‘rest of the portfolio’: Think of the portfolio as just two classes: The asset, and the rest of the portfolio (in other words think of the rest of the portfolio as one single class). High volatility in either of the two classes will warrant tighter corridors. Hope this helps.

Magix Wrote: ------------------------------------------------------- > This is somewhat counter intuitive IMO. > > My original thinking when I read this was: If the > volatility (of the asset or the rest of the > portfolio) is high, you would want wider > corridors. This would avoid frequent rebalancing > (high trading costs). > > However, the concept purports differently: If > volatility is higher you would want tighter > corridors in order to avoid asset classes from > deviating greatly from desired exposures. > > Coming to your specific point on the ‘rest of the > portfolio’: Think of the portfolio as just two > classes: The asset, and the rest of the portfolio > (in other words think of the rest of the portfolio > as one single class). High volatility in either of > the two classes will warrant tighter corridors. > > Hope this helps. Ok, Magix. So if I have a high volatility asset class, say A. I would want a narrow corridor. If the rest of my portfolio is low volatility I may be able to widen my corridor on A a little bit. But if I have a high volatility asset class, say A. I would want a narrow corridor. If the rest of my portfolio is high volatility I may be able to narrow my corridor on A even more. Is that how you understand it?

Well if one asset class (A or Rest of Portfolio) is volatile, it would create volatility in the other asset from a total portfolio perspective. In other words, if A’s weightage in the portfolio increases from 30% to 50% it will lead to a decrease in the weightage of the rest of the portfolio from 70% to 50%. So IMO all you need to know here is that if there is high volatility in either the asset or the rest of the portfolio, you would want tighter corridors in the portfolio. This is because volatility in either will create overall portfolio volatility.

I wouldn’t agree with that I think you should only care about volatility of class in relationship with the whole portfolio. In rebalancing we don’t really think about volatility as a mean variance method, but as volatility related to the portfolio If you think of it like that- that is separating into asset class volatility and portfolio volatility you might miss some stuff - let’s say portfolio goes down 10% and asset class goes down 10 %- in this case volatility is high for both but no need to rebalance or reconsider corridor width. In this case for our purpose we can say that the asset did not move ( remained at the target weight) That is how I see things - volatility makes sense only compared to volatility of portfolio

Magix you look at it as A. Asset class moves and portfolio stays the same B. Asset class stays the same & portfolio moves I think you should look at it as Asset Class changes compared to Portfolio changes

I agree with Magix here. At the time you are setting the corridor you will consider the volatilites (among other factors). If any of the asset classes has a high volatility you will have a narrower corridor, simply because it makes divergences from the target weights more likely. Of Course you need to consider all the other factors (costs, correlation etc) , the volatility is not the only factor you consider. In the example you mentioned if subsequently the Portfolio and the asset class both go down by 10%, then you will not rebalance because it still is within it’s corridor.

sv exactly so then why would you care about asset class volatility and portfolio volatility separate and not look at it as volatility of asset class in comparison to portfolio?

Considering your example- >If you think of it like that- that is separating into asset class volatility and portfolio >volatility you might miss some stuff - let’s say portfolio goes down 10% and asset class >goes down 10 .%- in this case volatility is high for both but no need to rebalance or reconsider corridor width. >In this case for our purpose we can say that the asset did not move ( remained at the >target weight) Volatilty does not imply they will move in the same direction. Consider the following cases: 1. Portfolio moved up 20% Asset class went down 20% . We need to rebalance. 2. Portfolio moved up 20% Asset class return was 0% We need to rebalance. 3. Portfolio return 0% Asset class moved up 20% within the portfolio we need to return There are obviously other cases here but the point I am trying to make is that in case 1 both the portfolio and the asset class are highly volatile but it does not mean that you need a narrower corridor. In general if the asset class is more volatile or the rest of the portfolio is more volatile (reflected in a higher portfolio volatility all else (i.e correlation, asset class weights) being the same) you need a narrower corridor. I think what you are probably thinking of is the correlation among the classes. A higher correlation would imply that returns move together more often and hence a wider corridor all else equal. So in the case you mentioned Portfolio up 10% asset class up 10% probably has higher asset class correlations than the portfolio I mentioned above (in case 1) implying a wider corridor.