Ask Colin

I have always thought that the longer the time frame the more reliable the data (I think I heard someone once say the less 'noise' in the data.) If this is not the case I understand why the 2% rule should be used irrespective of time frame, but if it is the case would you not have

different rules for different time frames?

I think the concept you are thinking of is relating to returns rather than volatility. It has been demonstrated by several people that the returns from stocks are quite volatile in the short term (say one to five years), however the returns over longer periods (say 20 to 30 years) are much les volatile. There is a good discussion of this in Jeremy Siegel Stocks for the Long Run. It is also documented in a booklet you can obtain free from Vanguard.

This does not really have any bearing on the applicability of the 2% rule. That rule relates to money management where we are determining position size in relation to risk and size of trading capital.