An excellent post that covers CCP's edge in its Wallet Wizard product. That is, by identifying reformed non-payers, CCP can offer a lower interest rate, or as an alternative, make a greater profit. It takes a general demographic that is risky, and deals with a sub-component that is less risky, leaving other lenders to work with the dross.
There is a minor error in your thinking displayed in the sentence, "Obviously doubtful debt provisions are utilised in the last few years when a debt eventually becomes uncollectable." The accounting for PDLs does not mirror the accounting for trade and loan debt, where provisioning is usual. The reason being is that the debt of a PDL is already "provisioned" (in a sense) by the low cents in the dollar of face value that CCP pays for PDLs. What CCP does is set a profit target for a PDL, and it is prepared to pay what it expects to collect, less what it wants as a profit. If it pays 20c in the dollar for a PDL, it typically means it expects to collect 40c, so what it does is effectively amortise the cost of the PDL by allocating half what it collects to amortisation.
Historically, CCP's amortisation has been about 47% of collections, which suggests it collects a little more than twice what it pays for PDLs. Typically its estimates are accurate (in part because estimating errors offset each other), so "adjustments" for differences in reality compared to expectation are minor. I suspect that CCP sets that percentage of collections slightly on the high side, so that by year 5 it has probably fully amortised the PDL, and so what is collected in Year 6 carries no amortisation - it just falls into the "adjustment bucket". Because of the last point, the profitability of collecting on old debt has an element of bonus in it that offsets the high collections expense of old debt. This is one of CCP's many buffers to average the rough with the smooth to keep their collections staff busy through cycles when PDLs are easily bought at times, but not at other times..
Expand
(20min delay)