My original question was valid. The two numbers are calculated independently of each other, ignoring the other loss. The capital coverage ratio of 166% is sort of meaningless because if all the interest losses were paid then there would be nothing left to cover capital
I think this hits on exactly what I've been trying (and failing) to formulate into coherent words, for about the last hour now!
Let me try and put it in ways I understand, and see if we're talking about the same thing.
The vast majority of the £65m fund is a projection. The projection part (£59m) is made up of two very interrelated things - future investor interest (£36m) and PF inflows (£23m).
These items are counted separately as if they're two stores of value, but are of course in reality just (part of) the profit that comes back from borrower repayments.
So the £59m is illusory to some extent. It's not double counting exactly, but is far more sensitive to unexpected losses than it might appear.
edit: It would also presumably be dramatically susceptible to borrowers defaulting earlier (on average) than forecast. As if that happens, you end up with:
1) future investor interest going down
2) PF inflows going down
3) Expected losses going up
In this way, each unexpected or unexpectedly early loss has a sort of triple whammy effect to the ratios.