|
|
Post by josc66 on Nov 4, 2014 16:44:30 GMT
Hmm. The secondary market is my favourite place in Bondora. I mainly purchase overdue loans there that people want to sell me very urgently. Anyone else investing mainly in the SM? If so any medium/long term results?
|
|
|
|
Post by analitikas on Nov 5, 2014 9:37:44 GMT
Hi josc66,
Key basic intro facts: investing only into verified Estonian loans across ABC600-1000; no more additional deposits; keeping cash reserve of up to 1% of portfolio value; trying to sell current loans with 7-40% premium to keep the cash level in place.
I am currently testing a very similar strategy - reinvesting into Estonian loans via investment profiles and purchasing overdue Estonian loans from the Secondary Market. I am purchasing only loans at discount (up to 0% premium/discount) and the ones that have only few days overdue (up till 15 days).
The strategy is now in place for around 2-3 months, but I have done detailed monitoring for only two months. I have noticed that my position in internal Bondora rankings increased and that my Bondora ROI improved. However, my internal calculation, using XIRR method, show very little change during the last two months of the experiment. I calculate the total value of my portfolio as the sum of 1) Current loans x 100% 2) Overdue loans x 90% 3) 60+ Overdue loans x 25% 4) Cash x 100%. During this period, my 60+ days overdue increased by 20%, while current and overdue loans remained at the very similar levels.
I will continue with this strategy for a while and update you on my findings.
|
|
|
|
Post by spanner on Dec 2, 2014 13:27:09 GMT
Hi, have been following this post with interest, the ability to make money by trading late loans in the Bondora secondary market appeals. However when I looked at it several months ago, it seemed like the market was pricing the risk consistently too low. I have recently stumbled across this article on AltFi: www.altfi.com/data/analysis/545. The analysis is basic but it seems to confirm my thoughts. I'd be interested to know whether I'm thinking about this in the wrong way and there are probability adjusted positive returns to be made in buying late loans at (relatively modest) discounts? |
|
|
|
Post by wiseclerk on Dec 2, 2014 17:33:31 GMT
The Altfi post neglects that a default is not final, but rather followed by collections. Expectations on how much is recovered differ of course, but expectations of some investors for the Estonian loans are high. Look at these charts which Bondora published in Oct. 2013 www.p2p-kredite.com/diskussion/grafiken-ausfallraten-2009-2012-t2028.htmlIf a curve crosses the x-axis it means more has been recovered on average than was due (helped by the penalty interests). So if an investor would have bought overdue Estonian loans originated in 2009 or 2010 at a moderate discount that could have been a good deal (note that the secondary market was not launched until early 2013, so in practise that would not have worked for many loans from that origination time). And of course you have to factor in the time collection takes and the risk that it is unpredictable how recovery rates will develop in future (and for other national markets) Hope that helps |
|
james
Posts: 2,205
Likes: 955
|
Post by james on Dec 2, 2014 19:51:50 GMT
Wiseclerk has mentioned some of the problems with that basic analysis. Here's a summary of some more:
1. percentages going on to default are different for each country.
2. percentages going to default are different for each credit grade and for assorted borrower properties.
3. recoveries vary very greatly between countries, perhaps 70%+ of the defaulted value for Estonian A loans.
4. percentage of amount lent at default isn't the same thing as percentage of loans going to default.
5. time to default varies for different markets, hence value at default time varies as well.
6. there are some apparent automated systems operating that within a few minutes of a payment will buy the loan before the seller has time to change the offered price.
Beyond that it's a market and opinions about probability of default and recovery will vary between buyers. Income tax rates and personal preferences also mean that different people will place different values on loans.
Say a loan has a base interest rate of 28% and is A1000 income and expenses verified with no other interesting properties. It's late by two days after three on time payments. Current loans of that nature typically have interest rates of 15%. What is the correct projected return to offer a buyer? What is the correct projected rate to offer the buyer if the loan instead has a base interest rate of 15? How do the discounts or premiums and projected profits or losses for the seller differ in the two base interest rate cases? What effect does knowing that there are buyers who will buy within minutes of the loan receiving a payment have on the pricing decision and decision whether and when to try to sell?
Same loan, now 7 days from default. Should you sell? What premium or discount? How does knowing that a buyer will buy within minutes of a payment affect your pricing decision and willingness to sell?
My partial answer to the pricing decision is that knowing that there are buyers who will buy only if there is a payment means that it is not possible to offer an economically perfect fair price to the buyers who will take the risk of default because it is necessary somehow for the seller to allow for that risk of selling with a discount that is not justified by the new loan state after the payment is made.
|
|
JamesFrance
Member of DD Central
Port Grimaud 1974
Posts: 1,323 
Likes: 897
|
Post by JamesFrance on Dec 3, 2014 10:29:42 GMT
Presumably the sellers think they will be better off by selling loans at the discounts or premiums they set. The buyers must think they will gain by buying them.
They cannot both be right, so the secondary market at Bondora has become a gambling activity and nothing to do with investment. I think it disadvantages investors so that a few bot operators and Bondora make money at our expense.
I will avoid it unless it is restored to the more sensible previous premium and discount levels. It would be interesting to know who asked for it to be changed, as Bondora said it was by request. No doubt the individual concerned is making a lot of money at the expense of investors, which I don't think is desirable for a P2P Investment platform. I wonder what the regulator would think of this?
|
|
james
Posts: 2,205
Likes: 955
|
Post by james on Dec 3, 2014 12:34:29 GMT
I've sold six loan parts at 40% premium. The buyer projected returns were between 87% and 98%. Do you really think that's unreasonable for income and expenses verified Estonian A1000 loans where recovery is likely to exceed 70% on average? Rescheduled formerly defaulted loans with lots of accumulated penalties. I actually do think it's unreasonable, I think a 40% premium cap was too low for those loans and would be more low later if they had six months of on time rescheduled payments. I'm one of those who asked for higher premiums, partly to cover this case and partly to cover income tax differences.
Of course asking 40% for a loan at 15% and the first payment not due yet would not look like a good deal and a way to filter for sensible buyer returns would be useful. I don't think that this use of high premiums is very useful, though if the buyer rate is sensible it could still make sense for both buyer and seller.
It is possible for both buyer and seller to be right. Consider a person with 40% income tax rate selling to a person with 0% income tax rate. The income tax margin offers the opportunity for the seller to convert taxable income to untaxed capital gains and the buyer to still make good money.
Both can also be right if they each have different preferences for dealing with bad debt and one doesn't want to do that and is willing to pay a buyer to take on the potential bad debt. This isn't a purely monetary gain but it can still be good for both buyer and seller.
If you've suffered a loss, perhaps as a result of those who buy a deeply discounted loan just after a payment is made, you have the option of seeking redress from Bondora and taking them to the FOS if they refuse.
|
|
|
|
Post by spanner on Dec 3, 2014 16:22:48 GMT
Fascinating discussion! Thanks for your thoughts Wiseclerk and James(s). Buying a loan at a big premium carries its own risk, no (ignoring default risk)? What if the loan were to repay early the next day? You'd lose the premium overnight? What maths make a 40% premium justifiable?
|
|
|
|
Post by wiseclerk on Dec 3, 2014 16:31:33 GMT
I posted this recent blog post by Orchard analysing resale market performance in another thread already: www.orchardplatform.com/blog/secondary-transactions-and-pricing-on-bondora/Look at the charts, especially the second one and you will see that the vast majority of successful listings has been between at between 1 and 5% premium. On the other hand you can also see - quite understandably - that prices tend to be higher for those loan parts that had more than 6 or more than 12 repayments. If you want to get a statistical feeling for what happens on the resale market I recommend that article as introduction. |
|
james
Posts: 2,205
Likes: 955
|
Post by james on Dec 3, 2014 16:40:44 GMT
If the loan is repaid the next day the premium is lost. This is very unlikely for defaulted loans because the borrower would probably not have the money to repay easily.
I think that the first loan that defaulted for me fully repaid all capital, interest and penalties. I sold some parts of that loan and kept other parts. The buyer got a good deal on that loan because I could only sell at 5% premium. They probably made more than 90% annual return on their purchase. The more recent sales at 40% premium with 87-98% projected annual return could also be a very good deal. Or not, it depends on collection rate.
Today I offered to sell a loan with interest rate of 33% for 60 months at a 15% premium. It is rescheduled because the borrower is doing military service and will make no payments until that is finished. Buyer projected return is 35%. A 30% premium for this loan gives a buyer rate of 27%. 40% would probably be about 25% annual buyer return.
|
|
|
|
Post by spanner on Dec 4, 2014 16:59:06 GMT
Hi Wiseclerk, I read the Orchard blog - there's no mention of late vs current loans and difference in sale price. This would seem an obvious thing to investigate, no?
|
|
