oxdoc
Posts: 19
Likes: 20
|
Post by oxdoc on Dec 3, 2019 21:37:49 GMT
With more than 85% recoveries on the earliest defaults shown, there's a huge difference between "defaults" and "losses due to defaults" (i.e. the "bad debt" figures). The latter are much better defined - a default rate BEFORE recoveries is entirely dependent on your internal policy of exactly which stage any given loan is marked as "defaulted", whereas losses due to defaults are pretty unambiguous once enough time has passed for all the defaults and recovery actions to have come to a conclusion. I agree. But I would presume that AC would use a definition of defaults for their predictions that is at least reasonably consistent with how they actually decide when loans are in default - but maybe that presumption is wrong and the cause of the difference. Though, having an ambiguous definition of what counts as a defaulted loan is itself not a good thing in my view. I re-read my original post and see that I wrote that the "difference between the expected and actual losses" is large, when I think I should have written "default rates" rather than "losses" there, so sorry if that's the cause of the confusion. It seems to me most reasonable to interpret the default projections as the anticipated losses due to defaults. The statistics given are called "default rates", though - if they were meant to be losses, I'd expect they would be called that. If this is unclear (I haven't looked at it beyond reading this thread) it would seem reasonable to me to ask AC to clarify things. stuartassetzcapital , can you comment? That would be appreciated, thanks.
|
|
corto
Member of DD Central
one-syllabistic
Posts: 851
Likes: 356
|
Post by corto on Dec 3, 2019 21:55:16 GMT
The definition of what credit, default and bad debt events are is on the statistics page linked to in post 1. There is a little ambiguity but supposedly AC sticks to these definitions.
|
|
|
|
Post by stuartassetzcapital on Dec 4, 2019 6:44:32 GMT
Yes but that applies to both the "expected" and "actual" default rates, so it doesn't look to me like it should account for any of the large difference between the values. The default rate doesn't depend on recoveries anyway. Even if AC are good at predicting losses, being bad at predicting default rates would seem to say something about their ability to assess the risks of loans. The loss rate also includes the recoveries and the 3.85%/3.75% figures look to refer to that - the default rates and losses figures could be consistent if recoveries were somehow much better than predicted, although I doubt that's the case, so I still feel confused... With more than 85% recoveries on the earliest defaults shown, there's a huge difference between "defaults" and "losses due to defaults" (i.e. the "bad debt" figures).
The latter are much better defined - a default rate BEFORE recoveries is entirely dependent on your internal policy of exactly which stage any given loan is marked as "defaulted", whereas losses due to defaults are pretty unambiguous once enough time has passed for all the defaults and recovery actions to have come to a conclusion.
It seems to me most reasonable to interpret the default projections as the anticipated losses due to defaults.
It's possible of course that the original way those figures were presented was as anticipated losses BEFORE recovery action, and they're just trying to quietly redefine it retrospectively for the older tranches due to the number of loans they subsequently classed as defaults before recovering most or all of the money... but ultimately it's the end result we're interest in!
I can confirm a few things. Defaults are loans in difficulty and the default rate merely gives the potential for a loss and is not directly connected to any likely loss rate on those defaulted loans which is expected to be substantially lower in reasonable conditions due to the security we take. Remember slightly elevated probability of default rates (and therefore an element of risk) is what helps create our higher interest rates. The property security is what is intended to bring this risk back down to a much lower overall loss level. It is therefore a very incorrect assumption to say a year's forecast default rate is its expected loss rate - that is absolutely wrong as the two are separate and assuming that will likely result in very large (positive) deviations between your expected outcome and actual outcome once all loans are repaid or recovered. The FCA requires us however to display Default Rate and does not require us to display the Loss Rate and clearly people are confusing the former for the latter which we have always feared but have not option to not display it - hence why we also show losses as a non regulation-required extra piece of data to seek to make things clearer to lenders but clearly the display of a default rate as required is confusing some people and making them think the default rate is the loss rate, as it might be in a poorly run unsecured lender for example, but there is nothing we are allowed to do about that. The reason for the difference between actual losses and defaults in our case is the property security principally means a defaulted loan doesn't necessarily lead to a loss. The early years had simpler credit models and default rate forecasting and our default rates ended higher than predicted as shown on the tables. However also bear in mind the very low volume of loans that makes statistical forecasting in the early years very uncertain and so the results do not necessarily relate to probability of default assessments being out due to the small data samples in those early years. Later years do not show this deviation as noted above but as the tables show not all loans are repaid yet so there is room for movement however the data shows a substantial improvement over the years at present as noted by others above. The data is the data and there is no adjustment permitted under the regulations and to answer another question I believe we have always had one of the most consistent processes for recording a default or other credit event in our statistics meaning our data in those default and loss tables for all years is very accurate and they have not benefited from a step change in how we record things as suggested above. I hope this helps and in the end only what is on the defaults and losses page counts as that is the regulatory requirement to publish our data and in a specified way. It will be updating very shortly too under the new FCA rules this month.
|
|
jlend
Member of DD Central
Posts: 1,840
Likes: 1,465
|
Post by jlend on Dec 4, 2019 12:18:25 GMT
BRISMO publish some independent analysis of the AC loan book. Some 1 and 3 year return stats are on their website, updated regularly. brismo.com/market-data/For the more detailed info you need to purchase their reports. |
|
|
|
Post by davee39 on Dec 4, 2019 22:09:06 GMT
Defaults can stay defaults indefinitely, so losses are nil. It seems even when guarantors are declared bankrupt and the period of bankruptcy has expired there is no intent to declare a default a loss. The loan just sits forever with six monthly updates informing 'no change - next update in six months'.
Alternatively it turns out that the asset can not be called in because under an administration its value drops by two thirds, so let the borrower prevaricate for several years and convince lenders to go along with it.
No amount of spin can hide the fact that there are some large undeclared losses, or that the provision funds are unlikely to be much help in compensating losses in the now closed accounts. The Green Loan Fiasco seems to be stretching A/C's balance sheet to the limit and there is plenty more rotten fruit to be dealt with.
|
|
