I still favour direct use of waist measurement in combination with hip and height as being simple, reliable and something that correlates well with early mortality in middle ages and on.

While for the same person *at the same time*, the use of subcutaneous fat as a simulcrum of total body fat is reasonable. I'm not sure that over longer periods this would hold up well or be a basis for making a highly informed decision about the desired end state. In particular many elderly people seem to lose much of their subcutaneous fat (and muscle) while still having a large midriff. I'm fairly sure that they would have had different body composition in their middle age and youth in most cases, whether they were healthy or overweight.