Author Topic: liquidation strategy on folio  (Read 10241 times)

mo

  • Full Member
  • ***
  • Posts: 158
    • View Profile
Re: liquidation strategy on folio
« Reply #15 on: May 27, 2016, 06:43:57 PM »
It used to be that there were a lot of states that could only buy notes on Folio in which case it made sense that notes sold for a mark-up since there was significant demand that couldn't be satisfied by LC directly.  Currently though there are only four states that can't invest in LC loans directly (PA, NC, NM, and AK).  There are really only two reasons I can think of that you would want to buy a note on Folio instead of directly from LC. One you are getting a nice discount or two you can't buy any other way.

The only time I buy from Folio is when I can get nice loans at a 3-5% discount.

http://blog.lendingclub.com/is-lending-club-available-in-my-state/
« Last Edit: May 29, 2016, 09:32:24 AM by mo »

Rob L

  • Hero Member
  • *****
  • Posts: 2065
    • View Profile
Re: liquidation strategy on folio
« Reply #16 on: May 27, 2016, 07:16:28 PM »
Human nature.  Sellers are greedy and want to at least try to get what they can.  I started at a 2.5% universal mark up on 1400 current notes.  I sold 4 in two days.  I lowered all to 1% this AM and I have sold 71 notes so far--so I assume the best of my notes are getting picked off.  Think I'll give it a couple more days and then go to par and let that sit for a week or so. 
I'm not employing any robots and not analyzing my portfolio at all.  Just gradually lowering my markup/discount --waiting--rinse and repeat.  This makes the most sense to me.
At some point--I hope I reduce my exposure to LC notes by 50% and at that point might just ride things out--depending on new about LC's funding efforts etc.

I know what you are doing is an excellent approach, but I simply can't bring myself to do it. An EMH'er without complete faith; a venal sin no doubt but hey this is Folio and known to be inefficient. I simply must know as much about the notes I put up for sale as I possibly can; and hopefully more than the average buyer. I know this is not possible for non-performing notes, but for just issued and current I should be able to do it. Most of my money is there anyway. Thus the last frenetic week writing a few thousand lines of code to see things more clearly, running linear regression models on the issued and current notes in the smallnotes file (the market is IMHO myopically priced big time), etc. If the market went crazy discount I swear I'd be a buyer because I couldn't help myself (I do think there's plenty of time before LC dies). It's nowhere near that so I'll continue my fool's errand.

sommers

  • Jr. Member
  • **
  • Posts: 77
    • View Profile
    • Email
Re: liquidation strategy on folio
« Reply #17 on: May 29, 2016, 07:16:55 AM »
Human nature.  Sellers are greedy and want to at least try to get what they can.  I started at a 2.5% universal mark up on 1400 current notes.  I sold 4 in two days.  I lowered all to 1% this AM and I have sold 71 notes so far--so I assume the best of my notes are getting picked off.  Think I'll give it a couple more days and then go to par and let that sit for a week or so. 
I'm not employing any robots and not analyzing my portfolio at all.  Just gradually lowering my markup/discount --waiting--rinse and repeat.  This makes the most sense to me.
At some point--I hope I reduce my exposure to LC notes by 50% and at that point might just ride things out--depending on new about LC's funding efforts etc.

I know what you are doing is an excellent approach, but I simply can't bring myself to do it. An EMH'er without complete faith; a venal sin no doubt but hey this is Folio and known to be inefficient. I simply must know as much about the notes I put up for sale as I possibly can; and hopefully more than the average buyer. I know this is not possible for non-performing notes, but for just issued and current I should be able to do it. Most of my money is there anyway. Thus the last frenetic week writing a few thousand lines of code to see things more clearly, running linear regression models on the issued and current notes in the smallnotes file (the market is IMHO myopically priced big time), etc. If the market went crazy discount I swear I'd be a buyer because I couldn't help myself (I do think there's plenty of time before LC dies). It's nowhere near that so I'll continue my fool's errand.
Not sure about excellent---so far I've sold about $2K of my $70K remaining notes.  This AM I moved everything to par.  It looks like if they don't move in a day---they don't move.  I guess its the bots.  My entire portfolio are $50 notes.  I can't see how anyone could examine each one of them and value individually.  No way do I have the time or inclination to do that.  I'll see what the next 24 hours brings and then I may go to 1% discount (so a 2pt hit after LC's fee--I think)  Then I"ll likely look into one of the bot programs --Peer Cube and Lending robot seem to be the most popular

Rob L

  • Hero Member
  • *****
  • Posts: 2065
    • View Profile
Re: liquidation strategy on folio
« Reply #18 on: May 29, 2016, 10:13:56 AM »
]
I can't see how anyone could examine each one of them and value individually.  No way do I have the time or inclination to do that.  I'll see what the next 24 hours brings and then I may go to 1% discount (so a 2pt hit after LC's fee--I think)  Then I"ll likely look into one of the bot programs --Peer Cube and Lending robot seem to be the most popular

A bot program would probably be very helpful. I haven't looked at Lending Robot's Folio sell automation (my bad) but have looked at Peer Cube. It's a nice tool and gives enough flexibility by filtering to let you price better notes (say FICO up) higher than inferior ones (FICO down). I would have used it myself but there were a few things I wanted to do that I couldn't and I decided to build my own. Not finished yet and it's been a lot of work.

AnilG

  • Hero Member
  • *****
  • Posts: 1095
    • View Profile
    • PeerCube
Re: liquidation strategy on folio
« Reply #19 on: May 29, 2016, 01:52:07 PM »
Just curious, what are the few things you wanted that were not available with PeerCube? We usually incorporate features requested by subscribers pretty quickly. We could have saved you lot of time and work in developing your own tool. We already have done lot of hard work that you didn't need to do.


A bot program would probably be very helpful. I haven't looked at Lending Robot's Folio sell automation (my bad) but have looked at Peer Cube. It's a nice tool and gives enough flexibility by filtering to let you price better notes (say FICO up) higher than inferior ones (FICO down). I would have used it myself but there were a few things I wanted to do that I couldn't and I decided to build my own. Not finished yet and it's been a lot of work.
---
Anil Gupta
PeerCube Thoughts blog https://www.peercube.com/blog
PeerCube https://www.peercube.com

rubicon

  • Full Member
  • ***
  • Posts: 103
    • View Profile
Re: liquidation strategy on folio
« Reply #20 on: May 29, 2016, 03:59:53 PM »
Human nature.  Sellers are greedy and want to at least try to get what they can.  I started at a 2.5% universal mark up on 1400 current notes.  I sold 4 in two days.  I lowered all to 1% this AM and I have sold 71 notes so far--so I assume the best of my notes are getting picked off.  Think I'll give it a couple more days and then go to par and let that sit for a week or so. 
I'm not employing any robots and not analyzing my portfolio at all.  Just gradually lowering my markup/discount --waiting--rinse and repeat.  This makes the most sense to me.
At some point--I hope I reduce my exposure to LC notes by 50% and at that point might just ride things out--depending on new about LC's funding efforts etc.

I know what you are doing is an excellent approach, but I simply can't bring myself to do it. An EMH'er without complete faith; a venal sin no doubt but hey this is Folio and known to be inefficient. I simply must know as much about the notes I put up for sale as I possibly can; and hopefully more than the average buyer. I know this is not possible for non-performing notes, but for just issued and current I should be able to do it. Most of my money is there anyway. Thus the last frenetic week writing a few thousand lines of code to see things more clearly, running linear regression models on the issued and current notes in the smallnotes file (the market is IMHO myopically priced big time), etc. If the market went crazy discount I swear I'd be a buyer because I couldn't help myself (I do think there's plenty of time before LC dies). It's nowhere near that so I'll continue my fool's errand.
Not sure about excellent---so far I've sold about $2K of my $70K remaining notes.  This AM I moved everything to par.  It looks like if they don't move in a day---they don't move.  I guess its the bots.  My entire portfolio are $50 notes.  I can't see how anyone could examine each one of them and value individually.  No way do I have the time or inclination to do that.  I'll see what the next 24 hours brings and then I may go to 1% discount (so a 2pt hit after LC's fee--I think)  Then I"ll likely look into one of the bot programs --Peer Cube and Lending robot seem to be the most popular

I'm guessing that your most of your notes are just shy of one year? The peak default rate is right at the one year mark so I would guess that if you can hold your notes past the one year mark you will get some demand for selling a seasoned portfolio.

BruiserB

  • Sr. Member
  • ****
  • Posts: 418
    • View Profile
liquidation strategy on folio
« Reply #21 on: May 29, 2016, 08:40:23 PM »
Just took a look at the Folio listings for issued and current notes. There are 530,714 total notes listed.

167,136 are listed at a discount, (31%).
    19,560 of them are at a -2% discount or more (3.7% of total)
     1,499 of them are at a -2% discount or more that are NeverLate FicoTrend UP (only 0.6 % of total)
         Of these 995 have outstanding principal of $50 or less (0.19% of total):
               436 are 36 month term and 559 are 60 month term
               117 have a YTM that is at least 1% greater than the original note interest rate (0.02% of total)

I don't see a lot of compelling buy's here.



101,909 are listed at par (19%)
261,669 are listed at a markup (50%)
I don't doubt it's a buyers market, but the sellers don't know it or don't care.
By definition everything listed now is overpriced or it would have sold already; right?

Human nature.  Sellers are greedy and want to at least try to get what they can.  I started at a 2.5% universal mark up on 1400 current notes.  I sold 4 in two days.  I lowered all to 1% this AM and I have sold 71 notes so far--so I assume the best of my notes are getting picked off.  Think I'll give it a couple more days and then go to par and let that sit for a week or so. 
I'm not employing any robots and not analyzing my portfolio at all.  Just gradually lowering my markup/discount --waiting--rinse and repeat.  This makes the most sense to me.
At some point--I hope I reduce my exposure to LC notes by 50% and at that point might just ride things out--depending on new about LC's funding efforts etc.


So are you taking all of your notes and progressively lowering the price until you have half left?  If you do that, you'll be left with the half that the market sees as least desirable.

Another strategy would be to take a random set of half of your notes and then progressively lower prices on that set until they all sell. You'd sell some at higher discounts but you'd still be left with a more normal distribution of notes remaining in your account.

Not sure which would be the better strategy over time but the two strategies would surely produce different results.


Sent from my iPhone using Tapatalk

rubicon

  • Full Member
  • ***
  • Posts: 103
    • View Profile
Re: liquidation strategy on folio
« Reply #22 on: May 29, 2016, 10:12:26 PM »
pretty sure these two strategies are logically equivalent.

Rob L

  • Hero Member
  • *****
  • Posts: 2065
    • View Profile
Re: liquidation strategy on folio
« Reply #23 on: May 29, 2016, 10:37:34 PM »
Just curious, what are the few things you wanted that were not available with PeerCube? We usually incorporate features requested by subscribers pretty quickly. We could have saved you lot of time and work in developing your own tool. We already have done lot of hard work that you didn't need to do.

Here's the list I had in mind (1 & 2 seemed nice to have, 3, 4 and 5 seemed important):

1) Filter by interest rate (inclusive range).
2) Filter by portfolio. Lets me select notes with various characteristics that I want to list and put them in a separate portfolios to keep track of them. Prevent listing notes in portfolios I want to protect.
3) Filter by FICO points change since loan origination, not just trend (up, flat or down).
4) Filter by NeverLate.
5) Set maxdiscount. If other note(s) are currently listed same loan and original issue do not attempt to underbid if the existing discount is more than my maxdiscount.

Some may be worth doing and some not. Since I have so little knowledge or experience with Folio I didn't think it appropriate to ask for changes. Besides, the best way to learn is by doing, and I need to do a lot of learning.

AnilG

  • Hero Member
  • *****
  • Posts: 1095
    • View Profile
    • PeerCube
Re: liquidation strategy on folio
« Reply #24 on: May 30, 2016, 12:05:30 AM »
Rob,

Thanks for sharing your good suggestions. See my responses inline below. A few are planned for next iteration. Some have just too many exceptions to implement reliably with our limited resources.

For next iteration: 1) and 5) are definite Yes, 4) is May be, and 2) and 3) are definite No.

You may want to contact me through email or contact form at the site next time. I can give you guidance and tips how to get somewhere with existing methods or changes that can be made in near future. 

Quote
1) Filter by interest rate (inclusive range).

Interest Rate and few other options will become available in the next iteration of Folio automated selling feature. We are currently working on first iteration of Folio automated buying feature. Once released our next project is to modify our first iteration of Folio selling and incorporate user feedback as well as our own observations.

Quote
2) Filter by portfolio. Lets me select notes with various characteristics that I want to list and put them in a separate portfolios to keep track of them. Prevent listing notes in portfolios I want to protect.

Managing portfolio and associated changes has been very difficult and source of most errors. There are just too many things users can do with portfolio on LC website that we don't have visibility to and cause the most issues. Next iteration of Folio selling is not going to address portfolio based listing management.

Anyway, I think you might be trying to introduce manual overhead by trying to protect some of your notes through portfolio. What happens if one of the protected notes become late? Are you going to manually monitor every day your portfolio folders on LC site and move the notes so that they can be listed for sale? There is a definite hierarchy of premium/discount based on loan status, FICO trend, and grade that you can use to protect your desired notes (assuming high quality notes through loan status and probably grade) by having a separate filter for listing for sale for such notes. There is no limit on how many filters you can have on PeerCube for listing for sale.

Quote
3) Filter by FICO points change since loan origination, not just trend (up, flat or down).

Latest FICO Score and Application FICO Score  were our favorite data points for detecting Notes at Risk. Unfortunately, Lending Club API no longer provides these data points through API. It is only available in notes_ext.csv file that requires manual download from LC website and upload to PeerCube. The required User intervention creates whole host of issues with regard to staleness of these data points at the time of order placement. Only FICO data point available through notes API is FICO trend (Up, Flat, Down) so that is what we use to avoid using stale and potentially erroneous latest FICO score data.

Quote
4) Filter by NeverLate.

There is high probability that Never Late flag will show in our next iteration. The main issue is that this flag has to be infer from variety of different data sources. It can't be determined for certain loans for example, new loans and loans never listed on Folio previously. Any inference of this flag is also not 100% deterministic.

Quote
5) Set maxdiscount. If other note(s) are currently listed same loan and original issue do not attempt to underbid if the existing discount is more than my maxdiscount.

Yep this will definitely be in next iteration. We plan to provide field for both min and max discount so users can specify a range within which to list notes for sale.


Here's the list I had in mind (1 & 2 seemed nice to have, 3, 4 and 5 seemed important):

1) Filter by interest rate (inclusive range).
2) Filter by portfolio. Lets me select notes with various characteristics that I want to list and put them in a separate portfolios to keep track of them. Prevent listing notes in portfolios I want to protect.
3) Filter by FICO points change since loan origination, not just trend (up, flat or down).
4) Filter by NeverLate.
5) Set maxdiscount. If other note(s) are currently listed same loan and original issue do not attempt to underbid if the existing discount is more than my maxdiscount.

Some may be worth doing and some not. Since I have so little knowledge or experience with Folio I didn't think it appropriate to ask for changes. Besides, the best way to learn is by doing, and I need to do a lot of learning.
---
Anil Gupta
PeerCube Thoughts blog https://www.peercube.com/blog
PeerCube https://www.peercube.com

sommers

  • Jr. Member
  • **
  • Posts: 77
    • View Profile
    • Email
Re: liquidation strategy on folio
« Reply #25 on: May 30, 2016, 07:00:44 AM »
Just took a look at the Folio listings for issued and current notes. There are 530,714 total notes listed.

167,136 are listed at a discount, (31%).
    19,560 of them are at a -2% discount or more (3.7% of total)
     1,499 of them are at a -2% discount or more that are NeverLate FicoTrend UP (only 0.6 % of total)
         Of these 995 have outstanding principal of $50 or less (0.19% of total):
               436 are 36 month term and 559 are 60 month term
               117 have a YTM that is at least 1% greater than the original note interest rate (0.02% of total)

I don't see a lot of compelling buy's here.



101,909 are listed at par (19%)
261,669 are listed at a markup (50%)
I don't doubt it's a buyers market, but the sellers don't know it or don't care.
By definition everything listed now is overpriced or it would have sold already; right?

Human nature.  Sellers are greedy and want to at least try to get what they can.  I started at a 2.5% universal mark up on 1400 current notes.  I sold 4 in two days.  I lowered all to 1% this AM and I have sold 71 notes so far--so I assume the best of my notes are getting picked off.  Think I'll give it a couple more days and then go to par and let that sit for a week or so. 
I'm not employing any robots and not analyzing my portfolio at all.  Just gradually lowering my markup/discount --waiting--rinse and repeat.  This makes the most sense to me.
At some point--I hope I reduce my exposure to LC notes by 50% and at that point might just ride things out--depending on new about LC's funding efforts etc.


So are you taking all of your notes and progressively lowering the price until you have half left?  If you do that, you'll be left with the half that the market sees as least desirable.

Another strategy would be to take a random set of half of your notes and then progressively lower prices on that set until they all sell. You'd sell some at higher discounts but you'd still be left with a more normal distribution of notes remaining in your account.

Not sure which would be the better strategy over time but the two strategies would surely produce different results.


Sent from my iPhone using Tapatalk

I have found folio to be cumbersome.  I don't have the time nor inclination to become an underwriter/credit analyst.  I just want to lighten up if not entirely liquidate. I still think that simply lowering the markup/discount over time should yield about the same end result as any other strategy.  Why wouldn't it?  I am now at 0--with the long weekend--I'll probably leave it there into mid week.  Then I'll go negative--1/2% at a time---giving each level some time to "bite" At some point--I'll get to a discount level that I'll have a decision and may end up just keeping some/most notes and hoping for the best.  I would think these bot buying programs have to be doing sweeps constantly---so maybe even giving each reduction 24 hours might be enough. 
Again, I was content to just let my notes roll off (they are all 3 year terms).  But the LC shennanigans have wised me up--this thing is weighted to the big boys and this little guy is getting out of the way

Rob L

  • Hero Member
  • *****
  • Posts: 2065
    • View Profile
Re: liquidation strategy on folio
« Reply #26 on: May 30, 2016, 09:47:01 AM »
Rob,

Thanks for sharing your good suggestions. See my responses inline below. A few are planned for next iteration. Some have just too many exceptions to implement reliably with our limited resources.

For next iteration: 1) and 5) are definite Yes, 4) is May be, and 2) and 3) are definite No.

You may want to contact me through email or contact form at the site next time. I can give you guidance and tips how to get somewhere with existing methods or changes that can be made in near future. 

Quote
1) Filter by interest rate (inclusive range).

Interest Rate and few other options will become available in the next iteration of Folio automated selling feature. We are currently working on first iteration of Folio automated buying feature. Once released our next project is to modify our first iteration of Folio selling and incorporate user feedback as well as our own observations.

Quote
2) Filter by portfolio. Lets me select notes with various characteristics that I want to list and put them in a separate portfolios to keep track of them. Prevent listing notes in portfolios I want to protect.

Managing portfolio and associated changes has been very difficult and source of most errors. There are just too many things users can do with portfolio on LC website that we don't have visibility to and cause the most issues. Next iteration of Folio selling is not going to address portfolio based listing management.

Anyway, I think you might be trying to introduce manual overhead by trying to protect some of your notes through portfolio. What happens if one of the protected notes become late? Are you going to manually monitor every day your portfolio folders on LC site and move the notes so that they can be listed for sale? There is a definite hierarchy of premium/discount based on loan status, FICO trend, and grade that you can use to protect your desired notes (assuming high quality notes through loan status and probably grade) by having a separate filter for listing for sale for such notes. There is no limit on how many filters you can have on PeerCube for listing for sale.

Quote
3) Filter by FICO points change since loan origination, not just trend (up, flat or down).

Latest FICO Score and Application FICO Score  were our favorite data points for detecting Notes at Risk. Unfortunately, Lending Club API no longer provides these data points through API. It is only available in notes_ext.csv file that requires manual download from LC website and upload to PeerCube. The required User intervention creates whole host of issues with regard to staleness of these data points at the time of order placement. Only FICO data point available through notes API is FICO trend (Up, Flat, Down) so that is what we use to avoid using stale and potentially erroneous latest FICO score data.

Quote
4) Filter by NeverLate.

There is high probability that Never Late flag will show in our next iteration. The main issue is that this flag has to be infer from variety of different data sources. It can't be determined for certain loans for example, new loans and loans never listed on Folio previously. Any inference of this flag is also not 100% deterministic.

Quote
5) Set maxdiscount. If other note(s) are currently listed same loan and original issue do not attempt to underbid if the existing discount is more than my maxdiscount.

Yep this will definitely be in next iteration. We plan to provide field for both min and max discount so users can specify a range within which to list notes for sale.


Here's the list I had in mind (1 & 2 seemed nice to have, 3, 4 and 5 seemed important):

1) Filter by interest rate (inclusive range).
2) Filter by portfolio. Lets me select notes with various characteristics that I want to list and put them in a separate portfolios to keep track of them. Prevent listing notes in portfolios I want to protect.
3) Filter by FICO points change since loan origination, not just trend (up, flat or down).
4) Filter by NeverLate.
5) Set maxdiscount. If other note(s) are currently listed same loan and original issue do not attempt to underbid if the existing discount is more than my maxdiscount.

Some may be worth doing and some not. Since I have so little knowledge or experience with Folio I didn't think it appropriate to ask for changes. Besides, the best way to learn is by doing, and I need to do a lot of learning.

Hi Anil,
Thanks for the response. Hope the suggestions were somewhat helpful,

The portfolio filtering (2) also struck me as a way to" dip my toe into Folio selling" so to speak. I'd put a bunch of notes into a sell portfolio as an experiment  and see how things work. It eliminates the possibility of my making a filter mistake that puts my entire account at risk. Had no idea it had subtle gotcha's, etc. Maybe there's another way to accomplish the same idea; not for active sell management purposes but simply as a firewall. As a user if you gave me a set of  rules and limitations regarding such portfolios that make the technical solution more tractable I'd follow them (or not use the portfolio filter). Anyway, like I said, just a nice to have.

Update; Another way to implement the "fire wall" would be for the user to control the maximum $ amount listed, or the maximum number of notes listed. Something that lets the faint of heart get started and build confidence before going all in.

I well understand the problems posed by FICO points and NeverLate. As an end user I have access to mynotes.csv and that makes things a bit easier than what you have to deal with.

Rob

« Last Edit: May 30, 2016, 01:09:22 PM by Rob L »

nmay2k

  • Newbie
  • *
  • Posts: 49
    • View Profile
Re: liquidation strategy on folio
« Reply #27 on: May 30, 2016, 10:12:29 AM »
Yes, it is true by gradually reducing the process of the entire portfolio you are selling your best notes first. Case in point, at a 2% mark-up, I was surprised to see a just issued note sell. So although in theory you might want to sell "everything", if you hang on to a note long enough to see it issued these days (a month or so), then they are probably one you don't want to sell (even at a 2% mark-up). I also have a "mature" portfolio to avoid discount issues when trying to sell notes within the last 5 payments if you will. When I reduce the price, I can see my higher interest loans go first (Cs in my case), leaving me with As & Bs. So I think it makes sense to try an parse them by grade as well. This will avoid having a portfolio of dogs in eye of the secondary market. Why not sell what you perceive as dogs?

Rob L

  • Hero Member
  • *****
  • Posts: 2065
    • View Profile
Re: liquidation strategy on folio
« Reply #28 on: May 30, 2016, 10:40:57 AM »
I also have a "mature" portfolio to avoid discount issues when trying to sell notes within the last 5 payments if you will.

Need a bit of education;  what are the issues with near end of life notes? (I'm always the last to know.)


SeattleSun

  • Full Member
  • ***
  • Posts: 217
    • View Profile
Re: liquidation strategy on folio
« Reply #29 on: June 01, 2016, 01:04:07 AM »
.
OK stand by for a dumb question.

I was looking at selling out my Prosper notes as I now have an opportunity that just beta tested at 1/2% per week.

I have 1291 notes and started in 2012 most are $100 notes.

Dec-13-2012   B   740-759      Current   60   19   $42.72   18.82%
Feb-04-2013   B   800-819      Current   60   21   $45.60   17.47%

1) haven't LC/prosper lending rates been falling since these notes were issued,
2) maybe these FICO have gone up since being issues, but 700 was my floor to lend to.

Q.  Should stuff like this be easy to sell?

TIA

Clueless in Seattle.