-
30-05-2017, 08:55 AM
#2261
Member
I have made an analysis of my Auto-lend loans, the majority were taken between October and December last year.These loans are only 5-7 months old now and 12.63% of my A/Ls' are in arrears. (Grade range A5-D3) Yet all my other active loans, minus the A/Ls', taken over the last 2 years manually are only 3.32% of the total in arrears. Half of these are in the E and F grades.
This leads me to think that there is one Auto-lend filter that can never be made available- "gut-feeling"; when I view a loan manually, as well as a financial evaluation I also get an overall impression or "gut-feeling" as to whether I should take the loan or not. I reckon I have been quite successful with limited arrears and defaults in A_D grades.
-
30-05-2017, 09:46 AM
#2262
yeah, nah
Originally Posted by permutation
I have made an analysis of my Auto-lend loans, the majority were taken between October and December last year.
Have you considered the time that these were taken as a possible cause i.e. pre-Christmas loans. It's another one of those graphs that I would love to see from Harmoney - month of year loan started vs default and/or arrears.
-
30-05-2017, 09:48 AM
#2263
Member
So has anyone had any luck getting money recovered on defaulting loans?
Just want to check out that Harmoney's collection service isn't some sham marketing.
Ive had no successful collections albeit from a smaller sample of 5 loans on 2.2k of lending.
-
30-05-2017, 10:22 AM
#2264
Member
Originally Posted by Saamee
Any other ideas how this all could go wrong ( apart from the obvious of genuine write offs )
Here are some ideas that spring to mind:
- Lending money is easy; it's getting it back that is the tricky part - and that is only seriously tested during a downturn. It sounds like Harmoney is not recovering much of it now during a booming economy, it will probably be a lot worse during a slow economy.
- Harmoney's methodology to assess the borrowers and deciding whether or not these borrowers should be put through the platform for funding.
- Harmoney has not been through a full credit cycle so it's difficult to assess the experience of the management team.
- It's unclear what the full ramifications would be if Harmoney went out of business.
- Change their loans underwriting model in order to increase their volume of loans.
- Their continued ability to identify fraud by borrowers.
- Unreliable forecasts of ROI and default rates.
Last edited by icyfire; 30-05-2017 at 10:31 AM.
-
30-05-2017, 10:44 AM
#2265
Member
Originally Posted by myles
Have you considered the time that these were taken as a possible cause...
NO and there's no reason.
Since 17/12 to 01/03 I have taken 164 loans manually (A_D grades only) and to date have 2/164 arrears being 1.22%. So I stick by my previous post!
Last edited by permutation; 30-05-2017 at 11:01 AM.
Reason: addendum
-
30-05-2017, 10:56 AM
#2266
Hello
The credit matrix is too loose to be sustainable, think South Canterbury under Lachie McLeod. The default models are unproven, from what everyone is saying collections are next to nothing and essentially they seem to be trying to bit off everything the big banks are currently turning away with tighter credit policies.
If the economies at the bar at its starting to get late, then these guys are still pouring shots.
-
30-05-2017, 10:59 AM
#2267
yeah, nah
Originally Posted by permutation
NO and there's no reason.
Personally I think their is - rash decisions are made on loans in the lead up to Christmas, and the volume typically increases significantly...
-
30-05-2017, 01:28 PM
#2268
yeah, nah
Three Economists do some data mining looking for ways to predict the likelihood of whether a borrower would pay back a loan based on the language they use:
How to Predict If a Borrower Will Pay You Back
The short of it:
Good: debt-free, lower interest rate, after-tax, minimum payment, graduate
Bad: God, promise, will pay, thank you, hospital
-
30-05-2017, 04:34 PM
#2269
Member
Originally Posted by myles
Three Economists do some data mining looking for ways to predict the likelihood of whether a borrower would pay back a loan based on the language they use:
How to Predict If a Borrower Will Pay You Back
The short of it:
Good: debt-free, lower interest rate, after-tax, minimum payment, graduate
Bad: God, promise, will pay, thank you, hospital
A good read, goes into big data too. And how casinos use it lol.
No ones ever approached me at sky city and said "are you having a bad night sir?"
-
31-05-2017, 09:12 AM
#2270
Member
[QUOTE=Since 17/12 to 01/03 I have taken 164 loans manually (A_D grades only) and to date have 2/164 arrears being 1.22%. So I stick by my previous post![/QUOTE]
I'm with you on this permutation. I've had four autolend loans totalling $400 out of 245 loans totalling $172,000. Of the four, one was repaid inside 6 months and two are constantly in and out of arrears. All were taken in October and December last year. I've decided that manual investing using a long developed lending background is best for me. And this month I had my first write-off: $285 on a $325 E2 taken out January 2016 which after I did it I wondered about the head space I was in at the time.
Tags for this Thread
Posting Permissions
- You may not post new threads
- You may not post replies
- You may not post attachments
- You may not edit your posts
-
Forum Rules
|
|
Bookmarks