r00lish67
Member of DD Central
Posts: 2,692
Likes: 4,048
|
Post by r00lish67 on Aug 26, 2019 17:22:57 GMT
I was a little bored, and so decided to make a couple of 'alternative' Lendingworks graphs based on data that I find interesting. Graph 1: The aim of this graph is to show the interest LW receive from borrowers (blue line) in relation to what they have given out to lenders in the form of interest (green bar) and what they forecast to write off as bad debt after recoveries (red bar, stacked on top). Edit: Please note Matthew's (LW rep) feedback in his post below as it indicates an error in the graph ( excerpt: The lender yields you've included are annualised, as are the APRs. However, the loss rates used are lifetime rather than annualised. For longer loan terms the annualised loss rate is substantially lower e.g. 5% lifetime loss rate on a 48 month term equates very roughly to around 2.5% annualised. So your red bars are going to be approx. double what they should be) Two things that I find noteworthy: 1) As I see it, for the years 2016-2018 in simple terms Lendingworks have had to pay out more than they have coming in, even on a gross basis. What the chart doesn't include of course are LW's running costs, which would add substantially to the stacked columns. 2) Despite having a significantly higher borrower APR than any other year, the forecast for 2019 bad debt is about the same as it was for 2016. ...Ok that burned rather more time than i thought...really need to make dinner. More later/tomorrow perhaps. Edit: Just wanted to make clear that this can't be a totally exhaustive exercise, as we have no view of any arrangement fees or other income that LW make. Similarly we don't have a view of their other costs. It's just an indication.
|
|
zlb
Member of DD Central
Posts: 1,422
Likes: 333
|
Post by zlb on Aug 27, 2019 8:52:59 GMT
Fabulous. Think this says a lot more than 'what size of provision fund is'. OK, I'll ask the dull question. What's the data source and is it reliable? It looks reliable, but that latest estimate of defaults looks strange...
If the red 2016-18 is from the provision fund, or insurance (the 'shield'?) Then can that be counted as a company loss? Are you assuming that anything over the blue line is from platform monies?
|
|
r00lish67
Member of DD Central
Posts: 2,692
Likes: 4,048
|
Post by r00lish67 on Aug 27, 2019 10:33:58 GMT
Fabulous. Think this says a lot more than 'what size of provision fund is'. OK, I'll ask the dull question. What's the data source and is it reliable? It looks reliable, but that latest estimate of defaults looks strange... If the red 2016-18 is from the provision fund, or insurance (the 'shield'?) Then can that be counted as a company loss? Are you assuming that anything over the blue line is from platform monies? Thanks. The source is the Lendingworks stats page (risk return tab) The red 2016-18 bars are taken from the provision fund. Any projected recovery from insurance should be included in LendingWorks' forecast bad debt rate. In some cases, the forecast bad debt rate is actually lower than the actual bad debt rate (e.g. 2015), which might reflect LW waiting for an insurance claim to pay out (or not, guesswork in this case). As far as I'm aware, all borrower defaults are claimed against the PF (although I believe there are some non-PF protected loans, but not sure if these apply to us in any way). In any case, it doesn't technically come directly off their bottom line. It's just indicative that from 2016 onwards, it seems pretty clear that the loans LW have written have been a substantially loss-making business. The above said, there is plenty of room for error in my analysis. There may be high loan arrangement fees that add to LW's income for example. Equally though, there's the rather substantial matter of all of LW's running costs - Rent, staff, utilities, IT, marketing etc etc. I struggle to imagine any additional income not listed exceeding the unstated costs, but I'm sure LW's Matthew will correct me if I'm wrong.
|
|
|
Post by propman on Aug 27, 2019 11:08:38 GMT
I can't access the graph, anyone else having a problem?
|
|
|
Post by Matthew on Aug 27, 2019 15:06:34 GMT
Fabulous. Think this says a lot more than 'what size of provision fund is'. OK, I'll ask the dull question. What's the data source and is it reliable? It looks reliable, but that latest estimate of defaults looks strange... If the red 2016-18 is from the provision fund, or insurance (the 'shield'?) Then can that be counted as a company loss? Are you assuming that anything over the blue line is from platform monies? Thanks. The source is the Lendingworks stats page (risk return tab) The red 2016-18 bars are taken from the provision fund. Any projected recovery from insurance should be included in LendingWorks' forecast bad debt rate. In some cases, the forecast bad debt rate is actually lower than the actual bad debt rate (e.g. 2015), which might reflect LW waiting for an insurance claim to pay out (or not, guesswork in this case). As far as I'm aware, all borrower defaults are claimed against the PF (although I believe there are some non-PF protected loans, but not sure if these apply to us in any way). In any case, it doesn't technically come directly off their bottom line. It's just indicative that from 2016 onwards, it seems pretty clear that the loans LW have written have been a substantially loss-making business. The above said, there is plenty of room for error in my analysis. There may be high loan arrangement fees that add to LW's income for example. Equally though, there's the rather substantial matter of all of LW's running costs - Rent, staff, utilities, IT, marketing etc etc. I struggle to imagine any additional income not listed exceeding the unstated costs, but I'm sure LW's Matthew will correct me if I'm wrong. Thanks for your analysis r00lish67 - if you keep that up we'll have to offer you a job... One immediate thing that jumps out is that you're not comparing apples with apples. The lender yields you've included are annualised, as are the APRs. However, the loss rates used are lifetime rather than annualised. For longer loan terms the annualised loss rate is substantially lower e.g. 5% lifetime loss rate on a 48 month term equates very roughly to around 2.5% annualised. So your red bars are going to be approx. double what they should be. Any remaining gap between APR and the red bar then represents: 1. LW spread - both in the form of upfront fees and share of NIM; and 2. Any surplus or deficit funding in the Shield i.e. contributions not used/overused. It's actually a useful piece of analysis as the FCA's P2P PS requires more disclosure around platform revenues versus lender returns (to ensure investors are aware of cases where the platform is charging 50% but only paying lenders 5%). This sort of graphical representation could be an interesting way of doing this. Interested to hear your thoughts. Thanks
|
|
r00lish67
Member of DD Central
Posts: 2,692
Likes: 4,048
|
Post by r00lish67 on Aug 27, 2019 15:11:23 GMT
I can't access the graph, anyone else having a problem? No idea why it's not showing for you, but here's a link: i.imgur.com/x7kP8eT.gif
|
|
r00lish67
Member of DD Central
Posts: 2,692
Likes: 4,048
|
Post by r00lish67 on Aug 27, 2019 15:33:47 GMT
Thanks for your analysis r00lish67 - if you keep that up we'll have to offer you a job... One immediate thing that jumps out is that you're not comparing apples with apples. The lender yields you've included are annualised, as are the APRs. However, the loss rates used are lifetime rather than annualised. For longer loan terms the annualised loss rate is substantially lower e.g. 5% lifetime loss rate on a 48 month term equates very roughly to around 2.5% annualised. So your red bars are going to be approx. double what they should be. Any remaining gap between APR and the red bar then represents: 1. LW spread - both in the form of upfront fees and share of NIM; and 2. Any surplus or deficit funding in the Shield i.e. contributions not used/overused. It's actually a useful piece of analysis as the FCA's P2P PS requires more disclosure around platform revenues versus lender returns (to ensure investors are aware of cases where the platform is charging 50% but only paying lenders 5%). This sort of graphical representation could be an interesting way of doing this. Interested to hear your thoughts. Thanks Hi Matthew, good point on my error, I will amend my original post. Re: graphs, I think it'd be a great idea to use something like this for lenders, except more accurate and complete of course! Other things that might be useful that weren't in the previous set of graphs on LW's page (off the top of my head). 1) Net shield utilisation forecast e.g. something showing that LW are expecting PF cash to fall/rise to not unduly concern lenders who don't have that forward view. 2) The impact of the insurance aspect of the products. I think lenders should be given a fair view of the impact of insurance, as it seems a common misconception that LW's insurance will cover most/all PF losses. 3) Some breakdown of how contractual future income and forecast future losses are generated perhaps. This said, LW already seem to disclose far more than most lenders in terms of statistics, certainly compared to FC/RS. Hopefully that means the FCA's disclosure requirements will fall more lightly on you than other platforms. Oh, and a job would be lovely, providing I can work a 10 hour week entirely from my home in my dressing gown
|
|
|
Post by propman on Aug 27, 2019 16:04:08 GMT
I can't access the graph, anyone else having a problem? No idea why it's not showing for you, but here's a link: i.imgur.com/x7kP8eT.gifI think it may be a firewall issue, will try again from home.
|
|