Comprehensive Retirement Income Products, page-34

Just quickly scrolling through the 1H18 presentation, though (slide 19), I noticed that there was actually a breakdown of the Life COE margin drop for the first half. Product margins caused a 15bp drop and lower asset yields a 19bp drop vs pcp in 1H (partially compensated by other factors, so that the overall drop was by 14bp); so that looks roughly in line with what I was guesstimating above.

Actually, I was reading that wrong, my apologies.

In the 1H18 presentation, it was the overall “product margin” (i.e. the Life COE margin before Normalised Capital Growth) that dropped by 15bp. So, that was not the impact from product mix changes.

Of those -15bp, -19bp was the contribution from lower asset yields, partially offset by a +4bp net positive contribution from lower interest costs and distribution expenses.

With regard to the product mix impact on the COE margin, all it says (slide 20) is that the higher percentage of Japanese/GIR products had a negative impact on the COE margin, but such impact is not explicitly quantified.

So, conceivably, the product mix impact was already embedded in the 19bp drop from lower asset yields in the previous slide. When it comes to isolating the impact from product mix changes alone (i.e. at constant asset yields), it still looks reasonable to me to assume that it represents no more than 30% of the total, as previously estimated.

Cheers