alanh
Posts: 556
Likes: 560
|
Post by alanh on Apr 29, 2019 11:04:11 GMT
Its also important to note that just because a loan is suspended doesn't mean there is a problem with it. For example there are many multi tranche loans which get suspended while drawdowns are processed. Other loans may be suspended for lender votes. Just using "suspended" as a criteria does not paint an accurate picture.
|
|
sl75
Posts: 2,092 
Likes: 1,245
|
Post by sl75 on Apr 29, 2019 12:14:09 GMT
Unfortunately the data you are using here is not fit for this purpose. Which purpose?
The purpose I had was highlighting a very sharp distinction I'd noticed between the first approx 600 loan IDs where there are quite significant numbers of loans suspended, and the remaining 400+ loan IDs where almost no loans are suspended. The other point that I was intending to highlight was that over a sufficiently long time, I would expect a portion of the older loans to be removed from the list, as the underlying loans were either recovered (from borrower/underlying asset/guarantor/insurer as appropriate), or all recovery avenues get fully exhausted so that the loss is crystallised with the loan book value reduced accordingly, and provision funds paying out for investment accounts that use them. On the face of it this process still does not seem to have occurred to any significant degree even for the very oldest loans on the platform.
If you're wanting to use some of the specific details for some other purpose, you would of course need to cross-reference against other details that AC aren't providing in that download. e.g. using outstanding loan amounts if you want a statistic derived from that rather than from AC's "loan amount", and someone else wanted dates for these different cohorts of loans.
This was very much intended as a set of simple and easily reproduced stats that others could use as the starting point of their own investigations rather than a final word on the matter!
|
|
|
|
Post by df on Apr 29, 2019 12:36:18 GMT
I get the feeling AC have a somewhat better record than other platforms. Proplend, Blendnetwork & Capitalstackers appear excellent, but unlike AC, availability is a problem with those. I know AC had a bad year 2016 with energy & bridging loans, but they appear to be on top of things now. The information they provide, pre & post loan offerings, are excellent. Their overall default %, after recoveries, appear on face value, to be quite acceptable if you are outside their protected funds. What is the general view ? I would have been fine if I didn't invest in Great&Green accounts. My diversification in MLA is adequate enough to face defaults and access accounts deliver what was promised. I can't measure performance of my AC funds until the recoveries of loans stuck in GBBA/GEA materialise.
|
|
bg
Member of DD Central
Posts: 1,368
Likes: 1,929
|
Post by bg on Apr 29, 2019 13:42:26 GMT
Unfortunately the data you are using here is not fit for this purpose. Which purpose?
The purpose I had was highlighting a very sharp distinction I'd noticed between the first approx 600 loan IDs where there are quite significant numbers of loans suspended, and the remaining 400+ loan IDs where almost no loans are suspended. The other point that I was intending to highlight was that over a sufficiently long time, I would expect a portion of the older loans to be removed from the list, as the underlying loans were either recovered (from borrower/underlying asset/guarantor/insurer as appropriate), or all recovery avenues get fully exhausted so that the loss is crystallised with the loan book value reduced accordingly, and provision funds paying out for investment accounts that use them. On the face of it this process still does not seem to have occurred to any significant degree even for the very oldest loans on the platform.
If you're wanting to use some of the specific details for some other purpose, you would of course need to cross-reference against other details that AC aren't providing in that download. e.g. using outstanding loan amounts if you want a statistic derived from that rather than from AC's "loan amount", and someone else wanted dates for these different cohorts of loans.
This was very much intended as a set of simple and easily reproduced stats that others could use as the starting point of their own investigations rather than a final word on the matter!
Sorry I mean the quoting of percentages of loans by value that are suspended. If you are using loan amount then it is inaccurate as for many of the loans significant capital repayments have been made. I would absolutely expect there to be more suspensions in older loans. It stands to reason that the longer a loan runs the more chance there is of it running into problems. Loans with numbers over 600 all originated in either 2018 or 2019, I would hope not many of them were suspended yet. Loans before that drew in 2013-2018. Of course, I also believe AC are now originating higher quality, lower risk/rate loans but I do not have data to back that up.
|
|
sl75
Posts: 2,092
Likes: 1,245
|
Post by sl75 on Apr 29, 2019 14:18:29 GMT
Sorry I mean the quoting of percentages of loans by value that are suspended. If you are using loan amount then it is inaccurate as for many of the loans significant capital repayments have been made. I would absolutely expect there to be more suspensions in older loans. It stands to reason that the longer a loan runs the more chance there is of it running into problems. Loans with numbers over 600 all originated in either 2018 or 2019, I would hope not many of them were suspended yet. Loans before that drew in 2013-2018. Of course, I also believe AC are now originating higher quality, lower risk/rate loans but I do not have data to back that up. Weighting by "loan amount" seemed more correct than quoting the first line with 16 loans out of 20 as (80%). As you say, an even better weighting would be by the "capital remaining", but AC do not make that information is not available in the data. That's one reason I didn't even consider quoting decimals of % except for the one that would otherwise have rounded to 0%.
I suspect that adding more data "for context" muddled the main point... reading down the number of suspended loans per group of 200 IDs, we see 16, 21, 21, 2, 0. A general trend of "more suspensions in older loans" is indeed expected but here we seem to have more of a sudden change. A secondary point is that recovery processes are much slower than many investors seem to assume, and even the very oldest loans have not been resolved either as amounts that have been recovered or as unrecoverable losses (the latter would surely be removed from the active loan value rather than merely suspended).
|
|
iRobot
Member of DD Central
Posts: 1,680
Likes: 2,477
|
Post by iRobot on Apr 29, 2019 14:23:30 GMT
Tend to agree and am concerned that the recently increased restrictions on viewing details for problem loans portends a worsening loanbook state. The main reason more likely is that confidential information about the state of recovery might leak into the public and have a potentially negative effect on the recovery processes. There have been a few interviews in the press that highlighted the problem. I don't disagree with secrecy when circumstances dictate, I'm just slightly cynical as to 'blanket' withdrawal of information.
|
|
sl75
Posts: 2,092
Likes: 1,245
|
Post by sl75 on Apr 29, 2019 14:35:31 GMT
I don't disagree with secrecy when circumstances dictate, I'm just slightly cynical as to 'blanket' withdrawal of information. On most other platforms I can think of, you only ever got to see updates about loans you're directly involved in (either as a current investor or as a potential buyer if the loan was open for purchase).
The "withdrawal of information" is merely returning AC to the established norm that other similar platforms have been using for years previously.
One point I do note, although not sure whether it's deliberate or an accidental oversight, is that all the relevant controls are enabled on "repaid" loans, so even unaffected investors can view all updates on those (and indeed ask to buy or sell some units of loans classified as repaid!)
|
|
ilmoro
Member of DD Central
'Wondering which of the bu***rs to blame, and watching for pigs on the wing.' - Pink Floyd
Posts: 11,329
Likes: 11,549
|
Post by ilmoro on Apr 29, 2019 15:08:22 GMT
Well its now 17 out of 20, but soon to be 16 out of 19, on the first 200 (140)
Fairwell to the 12% GGs 
And 22 out of 70, soon to be 21 out of 69, on the next 200 (repaid efficiently without suspension)
|
|
bg
Member of DD Central
Posts: 1,368
Likes: 1,929
|
Post by bg on Apr 29, 2019 15:21:43 GMT
Sorry I mean the quoting of percentages of loans by value that are suspended. If you are using loan amount then it is inaccurate as for many of the loans significant capital repayments have been made. I would absolutely expect there to be more suspensions in older loans. It stands to reason that the longer a loan runs the more chance there is of it running into problems. Loans with numbers over 600 all originated in either 2018 or 2019, I would hope not many of them were suspended yet. Loans before that drew in 2013-2018. Of course, I also believe AC are now originating higher quality, lower risk/rate loans but I do not have data to back that up. Weighting by "loan amount" seemed more correct than quoting the first line with 16 loans out of 20 as (80%). As you say, an even better weighting would be by the "capital remaining", but AC do not make that information is not available in the data. That's one reason I didn't even consider quoting decimals of % except for the one that would otherwise have rounded to 0%.
I suspect that adding more data "for context" muddled the main point... reading down the number of suspended loans per group of 200 IDs, we see 16, 21, 21, 2, 0. A general trend of "more suspensions in older loans" is indeed expected but here we seem to have more of a sudden change. A secondary point is that recovery processes are much slower than many investors seem to assume, and even the very oldest loans have not been resolved either as amounts that have been recovered or as unrecoverable losses (the latter would surely be removed from the active loan value rather than merely suspended).
I think some of the reason for that pattern is the speed of origination which has increased rapidly. The first 600 loans were originated over a 5 year period, the last 400 in 15 months, so certainly a much faster rate. I would be interested to know as well how many loans are being taken directly by institutions as they may influence the data. I do think AC made some mistakes in the early days though. There was one broker in particular that brought quite a few loans that ran into trouble which they stopped using.
|
|
iRobot
Member of DD Central
Posts: 1,680
Likes: 2,477
|
Post by iRobot on Apr 29, 2019 15:24:18 GMT
I don't disagree with secrecy when circumstances dictate, I'm just slightly cynical as to 'blanket' withdrawal of information. On most other platforms I can think of, you only ever got to see updates about loans you're directly involved in (either as a current investor or as a potential buyer if the loan was open for purchase).
The "withdrawal of information" is merely returning AC to the established norm that other similar platforms have been using for years previously.
One point I do note, although not sure whether it's deliberate or an accidental oversight, is that all the relevant controls are enabled on "repaid" loans, so even unaffected investors can view all updates on those (and indeed ask to buy or sell some units of loans classified as repaid!)
But I am in many of the loans just not at a £ level deemed sufficient to know what's going on when suspended. Not that I'm concerned over those as my exposure is minimal. It's more about the platform decision to implement this change and what it is indicative of for the future.
|
|
criston
Member of DD Central
Posts: 1,204
Likes: 628
|
Post by criston on May 5, 2019 13:41:31 GMT
Does anyone have information on the following.
1) What is the longest time that a defaulted loan is still going through recovery?
2) What is the worst defaulted loan in terms of percentage recovered?
3) What is the average percentage recovery for defaulted loans.
3) When a default occurs in one of the funds with a provision fund, does the recovery process have to get to the point where it is deemed 'the end of the road' before the provision fund kicks in ?
4) If the answer to 3) is yes, is it much different to handling a default in the Manual Lending Account, apart not being paid off by the provision fund.
My thinking is, if the average loss of defaults after recovery is say, less than 2%, why not pick your own in the MLA, to get an average percentage around 8%, rather than the GBBA at 6.25%.
|
|
ilmoro
Member of DD Central
'Wondering which of the bu***rs to blame, and watching for pigs on the wing.' - Pink Floyd
Posts: 11,329
Likes: 11,549
|
Post by ilmoro on May 5, 2019 14:36:52 GMT
Does anyone have information on the following. 1) What is the longest time that a defaulted loan is still going through recovery? #35 probably2) What is the worst defaulted loan in terms of percentage recovered? #104-110, or the #437-40, as in none. 3) What is the average percentage recovery for defaulted loans. Hard to quantify as many are ongoing, would need manual calculation.3) When a default occurs in one of the funds with a provision fund, does the recovery process have to get to the point where it is deemed 'the end of the road' before the provision fund kicks in ? Not the end of the road, ie written off, but far enough that any further recovery looks unlikely4) If the answer to 3) is yes, is it much different to handling a default in the Manual Lending Account, apart not being paid off by the provision fund. No difference as there arent different loans in the auto accounts compared to MLIA.My thinking is, if the average loss of defaults after recovery is say, less than 2%, why not pick your own in the MLA, to get an average percentage around 8%, rather than the GBBA at 6.25%. Depends on how much effort you want to devote to it and whether you want protection against losses. Losses on MLIA could be anything from 0-100% depending on what loans you invest in, level of diversification etc.
|
|
criston
Member of DD Central
Posts: 1,204
Likes: 628
|
Post by criston on May 5, 2019 14:53:42 GMT
Thanks ilmoro;
I note most of 'Trading Suspended' loans have 'Activity' greyed out.
If holding these loans, are you informed by email, as to what is happening.
|
|
cb25
Posts: 3,528
Likes: 2,668
|
Post by cb25 on May 5, 2019 15:15:16 GMT
Thanks ilmoro; I note most of 'Trading Suspended' loans have 'Activity' greyed out. If holding these loans, are you informed by email, as to what is happening. They may well appear greyed out to you if you have no holding in them. For investors, updates will be in Activity section and/or emails.
See this thread. E.g. this entry by pikeman Mar 16, 2019 at 7:44am "I asked chat line yesterday and they explained investors would no longer have access to ACTIVITY, DOCUMENT and Q&A tabs on suspended loans in which they had no monies invested."
|
|
criston
Member of DD Central
Posts: 1,204
Likes: 628
|
Post by criston on May 5, 2019 15:57:16 GMT
Just found a sort of answer to one of my earlier questions written by Assetz.
'Note: of loans in Default originated in 2013-2015, 85% of the amount lent has been or is expected to be recovered.'
|
|
