The proof is in the pudding with Affirm, a specialist in consumer credit and the famous "buy now pay later" scheme that's popping up all over the place on e-commerce sites.

Astonishingly, over the last four half-years, Affirm has set aside between a third and a half of its gross profit for doubtful loans. What's more, the company seems to be finding it difficult to resell its loan portfolios as it once did.

This is pushing the accounts further and further into the red, and calls into question the viability of the business model. Faced with a worrying equation - increasingly loss-making operations, soaring leverage - it will be important to quickly get our heads around this.

In theory, rising interest rates had squeezed net interest margins, before they were restored. According to Affirm, the average yield on loans granted is now close to 20% (sic), compared with 16% two years ago, while the cost of financing has risen from 4% to 7%.

Now that's a banker's dream! Doubts about borrower quality aside, of course...

Be that as it may, the "disruptive" fintech model bears a striking resemblance to that of predatory lending. The genre has the particularity of going from strength to strength in a short space of time, before the music suddenly stops when the first cascading defaults appear.

And yet: before the great financial crisis of 2008-2009, at least sub-prime specialists were making money... But even with its usurious rates, Affirm continues to lose money.

The group's communication on its financing capacity also lends itself to suspicion. During the last major stress on the credit markets, Affirm had to quickly raise $500 million in additional debt. Its CFO explained at the time that this should be seen as a sign of strength rather than weakness; many saw it as exactly the opposite.

Given these factors, and despite meteoric sales growth, it seems hard to justify a valuation of almost six times tangible equity. As with Block, the atmosphere in this room is very special and reserved for a very discerning audience.