|
|
Post by nardge on Feb 13, 2019 22:08:51 GMT
Quick Query -
When I first invested in P2P, large sums were invested, and autobid 'parcels' were equally large as a result...
Funding Circle 'Balanced' has defaulted a single loan part (£49.60), and Zopa 'Plus' several (£97.50 in total).
I believe that if I'd drip-fed the initial funds, the 'parcels' would have been smaller, and consequently,
the above default losses much smaller too? Is my thinking right?
Therefore, so as to create smaller 'parcels', it'd be wise to only invest by dripfeed in Zopa in future
(turning autobid on and off), thus creating tiny 'parcels', and only tiny losses as a result? Is this correct?
I know neither of the above P2P have a safeguard, though that hasn't been to my great detriment as yet.
With Kind regards
|
|
benaj
Member of DD Central
N/A
Posts: 5,597 
Likes: 1,736
|
Post by benaj on Feb 14, 2019 0:14:12 GMT
Zopa lends out new transfer in money in 1% chunk, minimum loan part is £10. For example, transferring 10k to Zopa plus, each loan part is £100. If money is being drip feed £1k ten times, the portfolio will have 1000 £10 chunks instead. In terms of default, it's all down to luck / probability. blog.zopa.com/2018/11/14/core-plus-hows-going/For some investors, they may have much higher / lower defaults in the short term, say just 100 loan chunks. When money is being drip feed over time, the portfolio will have more equal loan chunks and number of defaults will trend towards Zopa's projected default rate. It seems Zopa has improved the plus and Zopa plus in 2018 has lower defaults then loans originated in 2017. My experience with Zopa plus 2017 wasn't great. |
|
aju
Member of DD Central
Posts: 3,500
Likes: 924
|
Post by aju on Feb 14, 2019 0:31:55 GMT
Not sure what autobid is but my recent strategy with new money since SG cover stopped has been to feed in money at £10/loan using the following levels as a guide. 10 - 1999 = £10 loans 2000 - 2999 = £20 3000 - 3999 = £30 and so on. Don't get caught trying to lend too quickly either but once some of the 1999 has been lend you can drip feed in up to the relevant levels at any one time. Zopa has said many times they are working on allowing the lender to set the loan levels but so far that's never materialised probably to prevent people complaining about slow lending rates. One thing I will say is that once I have lent out the current sums I turn on re-invest. You would need some considerable sums of money to get >£10 loans on the reinvest side of things so relend will probably be ok there. Perhaps thats also why Zopa says that lending defaults will settle down after a period of relending has been churned for a while. I have also been lending across Plus/Other at the rates of 10%/90% and more recently with a couple of ISA transfers I have lent 20%/80%. I would point out that this theory is unproven but as benaj says it should tend towards Zopa's advertised rates. The thing I have experienced with very large amounts invested and using this £10 rule is that one gets a default at least once a month and sometimes can get several of them but month on month the worst I have got is 68% of month return has been crystalised into defaults. That said my rates for this last year have been in the region of 4-5%. That's fine for me others would not accept this. One other thing to bear in mind with my figures is that there is about 25% of my lend covered by SG still in Invest and slightly less in ISA side. Mrs Aju is similar. Also something to bear in mind is that headline rates drop off when reinvest is not turned or when running down an account. All that said I have been looking elsewhere recently and am winding down our Invest accounts and at present re-investing into RS at 6%+. I may decide later to move ISA's funds over into RS as well but for the moment I'm still getting to grips with that platform. RS has a Protection Fund but who knows what the next couple of years holds for investments aftrr 29th March, probably. HTH PS none of this £10 loans theory is proven long enough yet, i think it needs about another 2 years I'd think to be sure. |
|
paule
Posts: 44
Likes: 10
|
Post by paule on Feb 14, 2019 9:59:27 GMT
Zopa lends out new transfer in money in 1% chunk, minimum loan part is £10. For example, transferring 10k to Zopa plus, each loan part is £100. If money is being drip feed £1k ten times, the portfolio will have 1000 £10 chunks instead. In terms of default, it's all down to luck / probability. blog.zopa.com/2018/11/14/core-plus-hows-going/For some investors, they may have much higher / lower defaults in the short term, say just 100 loan chunks. When money is being drip feed over time, the portfolio will have more equal loan chunks and number of defaults will trend towards Zopa's projected default rate. It seems Zopa has improved the plus and Zopa plus in 2018 has lower defaults then loans originated in 2017. My experience with Zopa plus 2017 wasn't great. And as you will see it wasn't great for me either (a current return of 1.5% pa over two years). Zopa really took its eye of the ball here as far as i am aware. Mathematically speaking there should be no difference return wise on 100 X £10 or 10 X £100 loans but that's the theory, one bad large loan can kill your return where as if you split it into little chunks you're more likely to get closer to the average %age of defaults
|
|
jeremy12
Member of DD Central
Everything's frozen
Posts: 83
Likes: 38
|
Post by jeremy12 on Feb 15, 2019 18:11:18 GMT
Yes I picked up on the £1999 threshold and drip fed my ifisa earlier this year whilst still satisfying the criteria for the bonus payment - a tricky balancing act! Now i have 2000 x £10 loans in the isa so it should approximate the headline rate. I still have some older funds which are similarly widely spread and some still protected by the old zopa guarantee. So my overall rate is a complex mix which makes it hard to follow my actual rate of return. I use RS too but I think a steep depletion of its pf would make actual returns significantly below the posted rates.
|
|
|
|
Post by fuzzyiceberg on Feb 19, 2019 11:10:04 GMT
Now i have 2000 x £10 loans in the isa so it should approximate the headline rate. Actually what your ISA return will approximate is the actual achieved rate for the period you were investing. This may be more or less than the headline rate dependent on how good (or otherwise) Zopa's credit decisions were. From the fairly old and unacceptably aggregated (and so not that useful) data available on their website, Zopas decisions were particularly poor in 2017 and 2018 (defaults running well above what was expected at origination) and so it is likeley that the 'headline' rates for those periods will be missed by most investors. Which sadly includes me.
|
|
aju
Member of DD Central
Posts: 3,500
Likes: 924
|
Post by aju on Feb 19, 2019 19:04:30 GMT
Now i have 2000 x £10 loans in the isa so it should approximate the headline rate. Actually what your ISA return will approximate is the actual achieved rate for the period you were investing. This may be more or less than the headline rate dependent on how good (or otherwise) Zopa's credit decisions were. From the fairly old and unacceptably aggregated (and so not that useful) data available on their website, Zopas decisions were particularly poor in 2017 and 2018 (defaults running well above what was expected at origination) and so it is likeley that the 'headline' rates for those periods will be missed by most investors. Which sadly includes me. I couldn't agree more about zopa 2017 and 2018 decisions. We have started to get quite a bit of defaults, obviously with larger amounts in small £10 loans its bound to increase but I wonder how much of Zopa's bad decisions are around the increasing use of AI in their decisions. For me having been in the Software business for more years than I can remember both before I retired and increasingly so since I retired on a personal level its always amused me how this AI milarky is being lorded as some sort of savior. My personal opinion here is that from the state of Zopa's front end that we all see gives the impression that if their AI is modelled in a similar way then that could be a possible reason for the defaults. It's a simplistic assumption I know but Zopa QA does seem to have been assigned to the Zopa bin in my view.
|
|
