The process is the issue, not the model

I have great respect for Mr Roberts's views but in this piece he seems to be suggesting that just because a lot of providers use optimisers to determine their composition, this makes model portfolios flawed. I would instead suggest that it is more the fact that optimisers are tools which are particularly prone to the 'garbage in, garbage out' problem which many of their users fail to recognise which makes the problem more about the optimiser (or rather those who follow them slavishly 'because it gives the right answer' than the model portfolio per se. There is, in my view, nothing wrong with using model portfolios, provided that they are based on a clear and robust investment philosophy which everyone in the firm accepts and understands. There is, on the other hand, everything wrong with using a third party's optimiser whose input assumptions, calculations and constraints ("which fund are we trying to shift this month?") you do not understand to produce a portfolio composition even for a single investor.

Posted by: Robert Lockie