Elroy Dimson

has been, I think, overwhelmed by momentum returns, but it's high cost.

And so whether you do that effectively, if you can control costs very well,

then size and value are pushed down as the relatively unpopular factors now.

But for a long time, and after Paul Marsh and I created the first small cap index in the UK, which mirrored what Rolf Banz had done as Dimensional was being set up, he had done that for the US, the astonishing performance of small caps, which continued over a very long period

up into about two thirds of the way into the 1980s.

That all evaporated.

That's a fact and I think we now recognise that there are factors and premia.

There are attributes which are associated with differential performance, which may be good or may be bad.

attributes which may be associated with a premium because they are giving you exposure to stock characteristics that people don't want to be exposed to and therefore makes those stocks more cheap.

I think they should be monitored.

If you were looking at institutional active portfolios, then you'll often find that there are inadvertent back to tilts.

For example, some charities are constrained to spending income.

What that means is that they run the danger of influencing their asset manager to buy high yielding stocks.

So if you're aware of those factor effects, you can discover that in my example, the fund manager buys too much of the high yielders because that's the only way that the charity can actually access the money that it's making.

There are other similar things.

People after small caps have done well,

want to buy small caps.

And if they do that through a pooled vehicle, mutual fund, or an ETF, small caps are expensive to trade.

Those strategies can be expensive.

And so you need to understand some of these subconscious influences on the way a portfolio gets constructed.