While I understand most of your points, there is one thing that I'm struggling to understand. If the company has consistently outperformed its underwriting money multiplier over the last few years, shouldn't we be seeing material impairment gain for these past few years?
This is a great question and is a key metric in monitoring the thesis moving forwards. I believe the answer is multi-faceted:
Firstly, it's important to recall that the absolute $ value of PDP investment was very soft in FY21 (where the actual money multiple today is much higher than that set at inception). It's also important to recall that the actual money multiple expected to be collected fluctuates as the business goes about collecting on these PDPs. Given the company only started disclosing this money multiple data in the last year, its hard to track exactly when Pioneer changed their forecast for their actual money multiple for the FY21 vintage. On top of all this, if there is an impairment writedown of the portfolio it needs to be recorded in full on the income statement and as such masks the effect of what would have been any impairment gain throughout that period absent any writedown (just the nature of the amortized cost method of accounting which lends itself towards conservatism). In FY22, Pioneer incurred a one off writedown of their portfolio which showed up as an impairment loss in the P&L. Meanwhile in FY23, the company did indeed record a relatively significant impairment gain of ~$3.8M (especially when you compare this to the prevailing size of PDP investment over the previous years).
And then once again, in FY24 the company took a precautionary impairment on their portfolio of $18M as described in the write-up which again would have masked any would be 'impairment gain' absent the writedown.
Despite all of this, 1H25 saw impairment gains tick up to $3.8M. It's a metric to continue to monitor as the company progresses and one I'll be keeping a close eye on.
Just following up here. Apologies I was mistaken with the last point I made in my interpretation of the accounting.
If a particular vintage or tranche underperforms, an impairment loss must be recognised in full. However, this does not prevent gains from being recognised on other outperforming vintages.
Gains and losses are assessed and recorded individually at the tranche or asset level, based on revised cash flow expectations and the unchanged EIR.
That said, from a portfolio-level perspective, the income statement reflects the net position of these gains and losses. As such, a material loss on one tranche can visually mask the gains recorded on other vintages, even though those gains are being correctly recognised in accordance with the standard.
What this means in reality is that FY22 and FY24 really didn't show for much in the way of impairment gains.
Given the lowly level of investment in FY21, there would be little to come through in impairment gains in an absolute $ sense on this years vintage.
However, the fact we didn't see much in the way of impairment gains to offset the precautionary writedown for FY24 in the accounts is something to note.
It means that we should see greater impairment gains in the outer years to reflect the outperformance of the Actual M.M versus the U/W M.M. As expected, we did indeed see a significant 1H25 uplift in impairment gains of $3.8M. I expect this pattern of elevated impairment gains to continue into the next few earnings prints and if they don't, I will be reconsidering this position.
Thank you for the update. That is my understanding as well. Could it be that Pioneer is using an EIR that is closer to 40% and is therefore not showing significant impairment gains?
I don't believe it could be a 40% EIR that is being booked at the overall portfolio level. This is because their cash flow collection characteristics combined with the disclosed U/W MMs don't lend for this when you model it out. See my table in the write-up at a 1.8x money multiple, with the standard cash flow characteristics of an average vintage which lends for a 29% EIR. If you were to do this same exercise at a 1.9x money multiple with the same cash flow collection patterns, then the EIR only increases to ~31%. In reality it’s hard to be precise here given we don't have the data of their underlying investments but directionally I believe there is a wide gap in the last few years between the average EIR booked at inception versus the actual average IRR performance at the portfolio level and this deduction is based off recently disclosed data of cash flow characteristics and U/W MMs versus Actual MMs. I hope this helps.
You say that PDPs are "acquired at cents in the dollar". Can you be more specific? In a liquidation scenario, you are estimating they are sold for 10 cents/dollar. How much of a haircut is that? If the market is practically a duopoly now, who would be the buyer? If it is only CCP, they would be in a position to demand a very good price. Does the same logic apply to PAs?
Part of your thesis is the banks only want to deal with PNC or CCP. I would expect contractual constraints on those being sold on, or the banks' desire/efforts for safe counter-parties would be gamed. Is there anything in the contracts that limit on-selling? Would that constrain the sale of those assets in a liquidation?
PDP pricing will generally vary amongst the different categories of PDPs: e.g utility versus personal loans versus credit cards versus mortgages etc. It will also differ depending on whether the PDPs have been sold from big 4 banks, whose debt packages are considered higher quality. Also, the stability of pricing changes. In general, PDPs sold from big 4 banks are markedly stable long-term with only 2 key factors - interest rates and unemployment rates - serving as potential drivers for short term price changes. On the other hand, pricing for non-bank lenders is much more prone to fluctuations.
That's why I focussed on the net IRR profile and collection curve dynamics as opposed to trying to game the specific pricing in any way.
In saying that, my research suggests the average PDP purchase price from big 4 banks is around the 18c mark on the $ for Pioneer. These are consistent, high-quality debt packages and that's why I baked in 10c in the hypothetical liquidation scenario - a price markedly lower than that of which PNC purchase from the big 4 players.
On the duopoly situation, that strictly refers to the Australian market. There is no reason a larger international player - i.e. a PRA Group who are very good operators in the U.S. yet haven't been able to meaningfully penetrate Australian markets - couldn't lob a bid. The reality is that for any scaled and well-funded competitor to meaningfully penetrate Aus markets within a time horizon of less than 5 years, they would need to acquire either PNC or CCP. So there are plenty of international PE firms that could also be interested.
The ACCC would almost certainly not allow CCP to acquire Pioneer on the basis of monopolizing such a sensitive industry.
Re contractual obligations: PNC do not on-sell these debt packages to any other players and they are conditional upon PNC operating within the framework of the banks' approved collection procedures. In the hypothetical liquidation scenario, I don't believe this would be affected and believe the contracts are more centred around collection behaviour.
I don't believe there would be constraints regarding a transaction in the hypothetical liquidation scenario. There has been meaningful industry consolidation in recent years which hasn't then gone on to hinder the acquirer's capacity to service the newly expanded book - i.e. see Credit Corp's 2022 acquisition of Collection House and Credit Clear's 2023 acquistion of ARMA Group.
Hi Jesse,
Excellent write up, really insightful!
While I understand most of your points, there is one thing that I'm struggling to understand. If the company has consistently outperformed its underwriting money multiplier over the last few years, shouldn't we be seeing material impairment gain for these past few years?
Hi Evan,
Thanks for the kind words!
This is a great question and is a key metric in monitoring the thesis moving forwards. I believe the answer is multi-faceted:
Firstly, it's important to recall that the absolute $ value of PDP investment was very soft in FY21 (where the actual money multiple today is much higher than that set at inception). It's also important to recall that the actual money multiple expected to be collected fluctuates as the business goes about collecting on these PDPs. Given the company only started disclosing this money multiple data in the last year, its hard to track exactly when Pioneer changed their forecast for their actual money multiple for the FY21 vintage. On top of all this, if there is an impairment writedown of the portfolio it needs to be recorded in full on the income statement and as such masks the effect of what would have been any impairment gain throughout that period absent any writedown (just the nature of the amortized cost method of accounting which lends itself towards conservatism). In FY22, Pioneer incurred a one off writedown of their portfolio which showed up as an impairment loss in the P&L. Meanwhile in FY23, the company did indeed record a relatively significant impairment gain of ~$3.8M (especially when you compare this to the prevailing size of PDP investment over the previous years).
And then once again, in FY24 the company took a precautionary impairment on their portfolio of $18M as described in the write-up which again would have masked any would be 'impairment gain' absent the writedown.
Despite all of this, 1H25 saw impairment gains tick up to $3.8M. It's a metric to continue to monitor as the company progresses and one I'll be keeping a close eye on.
I hope this helps to clear things up.
Hi Jesse,
Noted! Thank you for your quick response. Looking forward to reading more of your stuff!
Hi Evan,
Just following up here. Apologies I was mistaken with the last point I made in my interpretation of the accounting.
If a particular vintage or tranche underperforms, an impairment loss must be recognised in full. However, this does not prevent gains from being recognised on other outperforming vintages.
Gains and losses are assessed and recorded individually at the tranche or asset level, based on revised cash flow expectations and the unchanged EIR.
That said, from a portfolio-level perspective, the income statement reflects the net position of these gains and losses. As such, a material loss on one tranche can visually mask the gains recorded on other vintages, even though those gains are being correctly recognised in accordance with the standard.
What this means in reality is that FY22 and FY24 really didn't show for much in the way of impairment gains.
Given the lowly level of investment in FY21, there would be little to come through in impairment gains in an absolute $ sense on this years vintage.
However, the fact we didn't see much in the way of impairment gains to offset the precautionary writedown for FY24 in the accounts is something to note.
It means that we should see greater impairment gains in the outer years to reflect the outperformance of the Actual M.M versus the U/W M.M. As expected, we did indeed see a significant 1H25 uplift in impairment gains of $3.8M. I expect this pattern of elevated impairment gains to continue into the next few earnings prints and if they don't, I will be reconsidering this position.
Hi Jesse,
Thank you for the update. That is my understanding as well. Could it be that Pioneer is using an EIR that is closer to 40% and is therefore not showing significant impairment gains?
I don't believe it could be a 40% EIR that is being booked at the overall portfolio level. This is because their cash flow collection characteristics combined with the disclosed U/W MMs don't lend for this when you model it out. See my table in the write-up at a 1.8x money multiple, with the standard cash flow characteristics of an average vintage which lends for a 29% EIR. If you were to do this same exercise at a 1.9x money multiple with the same cash flow collection patterns, then the EIR only increases to ~31%. In reality it’s hard to be precise here given we don't have the data of their underlying investments but directionally I believe there is a wide gap in the last few years between the average EIR booked at inception versus the actual average IRR performance at the portfolio level and this deduction is based off recently disclosed data of cash flow characteristics and U/W MMs versus Actual MMs. I hope this helps.
Hi Jesse,
Yes, very helpful. Thanks again!
You say that PDPs are "acquired at cents in the dollar". Can you be more specific? In a liquidation scenario, you are estimating they are sold for 10 cents/dollar. How much of a haircut is that? If the market is practically a duopoly now, who would be the buyer? If it is only CCP, they would be in a position to demand a very good price. Does the same logic apply to PAs?
Part of your thesis is the banks only want to deal with PNC or CCP. I would expect contractual constraints on those being sold on, or the banks' desire/efforts for safe counter-parties would be gamed. Is there anything in the contracts that limit on-selling? Would that constrain the sale of those assets in a liquidation?
Hey Whirly, thanks for your questions.
PDP pricing will generally vary amongst the different categories of PDPs: e.g utility versus personal loans versus credit cards versus mortgages etc. It will also differ depending on whether the PDPs have been sold from big 4 banks, whose debt packages are considered higher quality. Also, the stability of pricing changes. In general, PDPs sold from big 4 banks are markedly stable long-term with only 2 key factors - interest rates and unemployment rates - serving as potential drivers for short term price changes. On the other hand, pricing for non-bank lenders is much more prone to fluctuations.
That's why I focussed on the net IRR profile and collection curve dynamics as opposed to trying to game the specific pricing in any way.
In saying that, my research suggests the average PDP purchase price from big 4 banks is around the 18c mark on the $ for Pioneer. These are consistent, high-quality debt packages and that's why I baked in 10c in the hypothetical liquidation scenario - a price markedly lower than that of which PNC purchase from the big 4 players.
On the duopoly situation, that strictly refers to the Australian market. There is no reason a larger international player - i.e. a PRA Group who are very good operators in the U.S. yet haven't been able to meaningfully penetrate Australian markets - couldn't lob a bid. The reality is that for any scaled and well-funded competitor to meaningfully penetrate Aus markets within a time horizon of less than 5 years, they would need to acquire either PNC or CCP. So there are plenty of international PE firms that could also be interested.
The ACCC would almost certainly not allow CCP to acquire Pioneer on the basis of monopolizing such a sensitive industry.
Re contractual obligations: PNC do not on-sell these debt packages to any other players and they are conditional upon PNC operating within the framework of the banks' approved collection procedures. In the hypothetical liquidation scenario, I don't believe this would be affected and believe the contracts are more centred around collection behaviour.
I don't believe there would be constraints regarding a transaction in the hypothetical liquidation scenario. There has been meaningful industry consolidation in recent years which hasn't then gone on to hinder the acquirer's capacity to service the newly expanded book - i.e. see Credit Corp's 2022 acquisition of Collection House and Credit Clear's 2023 acquistion of ARMA Group.
I hope this helps.
Well done mate! Thanks.
No worries - Appreciate the kind words!