### Author Topic: Loss assumptions  (Read 11002 times)

#### hoggy1

• Sr. Member
• Posts: 401
##### Re: Loss assumptions
« Reply #15 on: February 12, 2015, 04:54:49 AM »
Thanks everyone. Good discussion. The loss data in LCs 10K/10Q is more than I hoped for. Its been some time since I looked at them and I forgot that data was there. My thanks to Fred.

Steve

#### hoggy1

• Sr. Member
• Posts: 401
##### Re: Loss assumptions
« Reply #16 on: February 14, 2015, 04:38:51 PM »
LC says 23% of IGP notes, 58% of late-16 notes, and 75% of late-31 notes go on to be written off. Any one of you with large note holdings ever try to check these figures in your own account?

Those numbers are "conditional probabilities" -- given that a loan is in IGP .....

To get a better charge-off picture without the conditionals, I usually refer to the charts on their 10-K/10-Q filings.

Here is the latest:
(from http://www.sec.gov/Archives/edgar/data/1409970/000119312514397209/d794350d10q.htm)

Hey Fred,

I was starting to work with the data in the 10Q a portion of which you included above. But I noticed something odd. Look at the graph again (and the accompanying tables in the 10q). For every grade and term (36,60) the first 4 months show 0.0% charge-off. Well you say, that's because they don't "charge-off" the loans until at least 4 months without payment! I'll argue then LC has screwed the pooch here and their charts are all screwed 4 month right. What we are really interested in is for charged-off loans, how many payments were made? 0.0% implies less than 1 in 1000 notes defaulted in any of the 1st 4 months. But we have all seen notes that made 1 or 2 payments and then defaulted albeit they were not "charged-off" till month 6. I think  there is a bit of a fiddle going on when LC at its sole discretion decides when to charge-off a note when in fact the note actually defaulted 4 or more months earlier.
Steve

#### Fred93

• Hero Member
• Posts: 2252
##### Re: Loss assumptions
« Reply #17 on: February 14, 2015, 04:50:59 PM »
For every grade and term (36,60) the first 4 months show 0.0% charge-off. Well you say, that's because they don't "charge-off" the loans until at least 4 months without payment! I'll argue then LC has screwed the pooch here and their charts are all screwed 4 month right. What we are really interested in is for charged-off loans, how many payments were made?

I agree.  This is an example of where you really need to understand what the data is before you can evaluate it.  I would add that because there is a 4 month skew, the charts for 36 month loans should go past 40 months, but they seem to stop at 36.  A lot of people have misunderstood the data because they didn't realize that what actually matters to lenders is when the borrower stops paying.

#### AnilG

• Hero Member
• Posts: 1123
##### Re: Loss assumptions
« Reply #18 on: February 14, 2015, 05:23:49 PM »
Hey Fred,

I was starting to work with the data in the 10Q a portion of which you included above. But I noticed something odd. Look at the graph again (and the accompanying tables in the 10q). For every grade and term (36,60) the first 4 months show 0.0% charge-off. Well you say, that's because they don't "charge-off" the loans until at least 4 months without payment! I'll argue then LC has screwed the pooch here and their charts are all screwed 4 month right. What we are really interested in is for charged-off loans, how many payments were made? 0.0% implies less than 1 in 1000 notes defaulted in any of the 1st 4 months. But we have all seen notes that made 1 or 2 payments and then defaulted albeit they were not "charged-off" till month 6. I think  there is a bit of a fiddle going on when LC at its sole discretion decides when to charge-off a note when in fact the note actually defaulted 4 or more months earlier.

From http://www.federalreserve.gov/releases/chargeoff/, Charge-offs are the value of loans and leases removed from the books and charged against loss reserves.

The charge offs are when loan payment is late by 120+ days. Even if a loan doesn't make any payment, minimum time to charge off after issue is 4+ months. You will not see any charge offs for first 4 months unless the borrower died or filed for bankruptcy and the settlements happened before 4 months. As long as a loan exist on the books, it can't be considered charge off. Charge offs have nothing to do with how many payments were made.

LC is not doing anything different than what other lending institutions would do.
---
Anil Gupta
PeerCube Thoughts blog https://www.peercube.com/blog
PeerCube https://www.peercube.com

#### hoggy1

• Sr. Member
• Posts: 401
##### Re: Loss assumptions
« Reply #19 on: February 14, 2015, 05:44:34 PM »
Thanks, but can I then be certain the charge-offs in month 5 made 0 (zero) payments.
Steve

#### AnilG

• Hero Member
• Posts: 1123
##### Re: Loss assumptions
« Reply #20 on: February 14, 2015, 06:00:40 PM »
No.

Official definition of charge off is when the loan comes off the books of lender, in this case Lending Club. In case of charge off due to death/bankruptcy, the loan can be charged off any time the settlement happens. In other cases, the charge off happens when lender decides that it has made sufficient effort for collection and recovery and there is no chance of recovery, typically it will be 120+ days but not guaranteed. I have seen as long as 180 days.

For practical purposes, you can review the loan history to see when typically loans are charged off after the last payment and use that as guideline to estimate. You don't need to be 100% accurate, just approximately right.

Thanks, but can I then be certain the charge-offs in month 5 made 0 (zero) payments.
---
Anil Gupta
PeerCube Thoughts blog https://www.peercube.com/blog
PeerCube https://www.peercube.com

#### rawraw

• Hero Member
• Posts: 2795
##### Re: Loss assumptions
« Reply #21 on: February 14, 2015, 07:55:39 PM »
Just to clarify, the 120 days applies to consumer loans.  Business loans are a little different.  If you are interested in the standards most financial institutions adhere to, here they are: https://www.fdic.gov/regulations/laws/rules/5000-1000.html

#### hoggy1

• Sr. Member
• Posts: 401
##### Re: Loss assumptions
« Reply #22 on: February 15, 2015, 07:37:05 AM »
Yeah, I should have guessed there were reporting rules (FASB, SEC, etc) for all these forms. Temporary brain death. Just haven't spent a lot of time with 10Qs before.
Steve