� By the way, do you know the time frame for Grantham's efficient frontier graphs, Exhibit 5 in the article you linked? The time frame must be short, since Grantham makes a point of how the curves have flattened recently. �
Sorry, I don't know the time period.
Grantham's forecasts used to cover 10 year periods, but he moved to 7 years in mid 2001. I don't have a reference for you but do remember reading something to the effect that after studying historical cycles (from overvaluation to undervaluation) for many different asset classes he settled on 7 years. He rationalized 7 years by saying that 10 years muted the damage inflicted by a contracting multiple on the real total return but that annualizing the damage over five years was just too scary for his clients. I hope I got that right.