|
|
Post by devonlad on Nov 1, 2015 18:00:52 GMT
I agree pretty much with mikes1531 and mrclondon. However I know for certain that when a loan goes really bad and the provider gets a whole load of negative publicity as a result they hurt like hell. I have had personal experience of this happening. I with many others had vented my feelings on this Forum and attempted to contact two of the principles by their preferred route. Neither were available or attempted to contact me at the time. Then I had the opportunity to fire a burst of red hot shrapnel at the them on the TV. Within an hour of the programme going out I had an email asking me when one of the principle's could contact me. Subsequently I did speak to the principle involved and he explained that the negative contributions on this site and my little contribution via TV had hurt their public image badly. Further he offered me a direct line for communication if I got "upset" with them again.
Lesson: P2P providers are seriously concerned about maintain a good public image. So it is worth using this this Forum when you think that you have a genuine grievance and failing that try going public. Sorry to be pedantic. What you are saying is fair enough, but it grates to read it. As a matter of principle, I feel I should point out that it was the PRINCIPALS whom you attempted to contact and not the "principles" or "principle's". Sorry. That apart your post is really rather interesting.
|
|
mikes1531
Member of DD Central
Posts: 6,453
Likes: 2,320
|
Post by mikes1531 on Nov 1, 2015 18:21:03 GMT
AC, for example, assumed portfolio default rates between 1-13% and recoveries of 60-90% over the cycle. This is consistent with the recovery range assumptions above and also with the type of default probabilities I can see data for (see below) I expect samford71 has access to additional info because of his underwriting relationship with AC, but what they say on their website is... This is based on a 20% rate of 'Loss Given Default". They explain what that means in an earlier paragraph, but I think they have a mistake in their explanation. What they say is... If a property is valued at £500k, and they've lent £400k (80%), and after a default they recover £300k (60%), then they may have a loss of 20% of the value of the property, but the loss is a quarter of what they've lent, so ISTM that the Loss Given Default actually is 25%. Am I misunderstanding something? AC's estimate of a 2.5% future default rate would be a considerable improvement over their past/current performance. Their website says they've lent a total of £82M. Looking at the list of current loans, I see £7M that have defaulted, and I see an additional £3M of defaulted (and, I should add, successfully recovered) loans on their list of repaid loans. To me, that looks like a historical default rate of 10/82, or 12%. And that number can only increase if any of the existing loans were to default at some point in the future. Hopefully AC have learned from their early experience and really can achieve a 2.5% default rate going forward. As an investor and future shareholder, I have my fingers crossed.
|
|
mikes1531
Member of DD Central
Posts: 6,453
Likes: 2,320
|
Post by mikes1531 on Nov 1, 2015 19:20:34 GMT
TLC 9 is a roll up of capital and interest and is about half way through its 5 year term. Max bid rate allowed is 4.25%, and currently no takers. TLC9 pays at 7% pa payable monthly and has been doing so. As at August it was 'worth' £ 128,142 against a capital value of £ 106,000.
Let me start by saying I'm totally unfamiliar with this loan and the platform it's from. If this loan is paying interest monthly at 7% p.a. and has about 2.5 years until maturity, how can it be 'worth' a 21% premium? Wouldn't anyone who bought it at that premium and held it to maturity receive less than their investment back from the combination 2.5 years of interest and returned capital? To me, that looks like a negative yield to maturity. What don't I understand?
|
|
mikes1531
Member of DD Central
Posts: 6,453
Likes: 2,320
|
Post by mikes1531 on Nov 1, 2015 19:33:55 GMT
In the majority of the business cycle, default rates will be below cycle averages, often just 1-3% (see figure in my prior post). So assuming a 2.5% default rate and PV(recovery rate) of 80% (LGD of 20%) sounds plausible. However, when the downturn inevitably arrives, defaults can rise 5x, so you can observe a short burst of 10-15% default rates, and recovery rates will fall somewhat (as most asset values fall). If his is what AC mean when they refer to a 2.5% default rate -- and I have no reason to believe otherwise -- then I'd be willing to bet that it isn't how the average reader interprets it. I'd think that what the average investor wants to know is the platform's best estimate of the difference between the nominal rates shown on the platform's loans and the pre-tax return they can expect to achieve after considering the likely level of defaults. For a platform to quote a bad debt expectation that only applies during the 'good times' and isn't applicable across a whole cycle strikes me as rather misleading.
|
|
|
|
Post by solicitorious on Nov 1, 2015 19:41:28 GMT
In the majority of the business cycle, default rates will be below cycle averages, often just 1-3% (see figure in my prior post). So assuming a 2.5% default rate and PV(recovery rate) of 80% (LGD of 20%) sounds plausible. However, when the downturn inevitably arrives, defaults can rise 5x, so you can observe a short burst of 10-15% default rates, and recovery rates will fall somewhat (as most asset values fall). If his is what AC mean when they refer to a 2.5% default rate -- and I have no reason to believe otherwise -- then I'd be willing to bet that it isn't how the average reader interprets it. I'd think that what the average investor wants to know is the platform's best estimate of the difference between the nominal rates shown on the platform's loans and the pre-tax return they can expect to achieve after considering the likely level of defaults. For a platform to quote a bad debt expectation that only applies during the 'good times' and isn't applicable across a whole cycle strikes me as rather misleading. And that is assuming there is something to "recover" from items which are supposed to be "security", which - as we have seen already - adds considerable flexibility to the ordinary meaning of these words.... 
|
|
|
|
Post by reeknralf on Nov 1, 2015 19:43:49 GMT
These are just the sort of figures I want, but don't know where to find. Where do these data come from? by far the most common decile recovery is 90-100% (100% is the most common percentile). This is not surprising given we're talking about sub 70% LTV structures; theoretical recovery is 100%. To get averages of 67% and even 85%, however, that means that a minority of cases are recovering far below that number. If most recoveries are 90% or more, to end up with an average of 67%, you need to be completely wiped out on every third default. Seems hard to credit, but I've no reason to doubt the data, so I guess this is what we have to look forward to. On the bright side a 6% default rate with average 67% recovery still yields 9%, which I can live with. 12% default/67% recovery yields 6%.
|
|
mikes1531
Member of DD Central
Posts: 6,453
Likes: 2,320
|
Post by mikes1531 on Nov 1, 2015 20:53:59 GMT
by far the most common decile recovery is 90-100% (100% is the most common percentile). This is not surprising given we're talking about sub 70% LTV structures; theoretical recovery is 100%. To get averages of 67% and even 85%, however, that means that a minority of cases are recovering far below that number. If most recoveries are 90% or more, to end up with an average of 67%, you need to be completely wiped out on every third default. Seems hard to credit, but I've no reason to doubt the data, so I guess this is what we have to look forward to. On the bright side a 6% default rate with average 67% recovery still yields 9%, which I can live with. 12% default/67% recovery yields 6%. : I'm having difficulty duplicating the results you have in your last paragraph. Can you please explain how you calculated them?
|
|
|
|
Post by tybalt on Nov 1, 2015 20:57:03 GMT
TLC9 pays at 7% pa payable monthly and has been doing so. As at August it was 'worth' £ 128,142 against a capital value of £ 106,000.
Let me start by saying I'm totally unfamiliar with this loan and the platform it's from. If this loan is paying interest monthly at 7% p.a. and has about 2.5 years until maturity, how can it be 'worth' a 21% premium? Wouldn't anyone who bought it at that premium and held it to maturity receive less than their investment back from the combination 2.5 years of interest and returned capital? To me, that looks like a negative yield to maturity. What don't I understand?
Because it has accumulated extra income at more like 10% -
1. ThinCats are about to change software which will render this discussion even less relevant than now. 2. Too much of the valuation depends on the purchasers tax rate and the defaults on individual loans within the portfolio
|
|
