09
Jun
Jun
Lending Club late loan is current again
Surprisingly, my Lending Club late loan is now current again. That puts me back to 100% current loans at an average interest rate of 11.23%. Thank you to the borrower for catching back up on your payments.
Wow, that is nice to see. That doesn’t usually happen with P2P “lates” does it? I imagine the % of loans that go late and come back to current is pretty low.
It was certainly a pleasant surprise and I doubt that it happens often. I saw a post on it sometime in relation to Prosper data and the recovery rate was very low — something like only 10% of loans ever recover after going 15 days late — but I can’t find that post right now. Although, Prosper has acknowledged that this is an issue and seems to be working on it.
I’m in on the same loan that you had go late. Let’s keep our fingers crossed that borrower will make the 6/12 payment!
Congratulations on the loan recovery. Now if only a few of my late loans would step into line
I think Prosper does a better job of explaining the overall risk associated with investing in P2P loans (or any loan for that matter).
[...] Personal Loan Portfolio Experience in P2P Lending on Prosper and Lending Club « Lending Club late loan is current again [...]
@ Evelyn: I will be keeping them crossed.
@ Daniel: Thanks. I was surprised by it.
Interestingly, I also thought Lending Club did a good job of listing the risks in P2P investing.
Lending Club spells it out in general terms, which a responsible
organization should do. However, when I went to setup a portfolio
plan with Prosper, they laid it out in numbers. See the second image
in this post http://www.step3prophet.com/2008/05/prosper-update-2.html
. It illirates nicely that even if your average loan interest rate
is 16%, your real return will be closer to 9%.
I see what you mean and I like that feature at Prosper too.
I don’t know the exact date that lender guidance was introduced at Prosper, where they subtract a risk number from your interest rate to give you an idea of your probable return. However, it was after about two years of data had been collected on loans. Some people still believe that it is not accurate since it is not based upon more than three years of data which is the length of the loans.
Lending Club does not have nearly enough history yet to list out risk on particular loan numerically. If they tried doing that right now I think it would be a mistake.
Actually, I think Lending Club could list out the risk of an aggregate
portfolio, just as Prosper does for a bidding plan. Prosper’s numbers
are based on Prosper historical data because the traditional risk
models don’t fit a Prosper loan portfolio. With Prosper, interest is
not directly related to risk. With Lending Club, the interest rate is
based on the borrowers risk, and historical default rates for the
banking industry in general should apply. From LC’s website:
http://www.lendingclub.com/info/historical-defaults.action . Also,
my own evaluation of LC portfolios: http://www.step3prophet.com/2008/04/running-some-nu.html
P.S. I like your site redesign.
Sounds right Daniel. Let’s see what LC has up their sleeves though
when they finally reopen. I have been in contact with Rob there
several times since “quiet” and he indicated that we should hear
something very soon.
I have mentioned it somewhere on this site before – I believe that for a given risk level (i.e. credit score, loan amount, historical late loans, etc) that the Prosper Marketplace default rate and the Lending Club default rate should be practically identical. Otherwise, it would imply that the same borrower would be more likely to default on Prosper than Lending Club due to a different lending model. I can believe that the same borrower might receive different interest rates by a few percentage points due to the different models. However, a few points difference in an interest rate should not impact the default rate significantly.
Only time will tell if there is a difference in default rates and then still the statistics will not be precise because they use different credit scoring agencies and divide their credit scores into different ranges.
Certainly, Lending Club’s overall default rate will be lower because of the number of loans that Lending Club rejects due to poor credit scores.
Prosper too believed that their default rates would follow industry norms and they did not. In fact, it was theorized that due to the person-to-person nature of the loans that the default rates would be lower than industry norms. That is laughable now.
If I ran Lending Club, I would not list a risk adjustment until I had more data.
But why are we even talking about Lending Club’s risk disclosures when we cannot lend there now? Do you think Lending Club will be back?
I do think Lending Club will be back, if they don’t burn through all their cash before the come out of there quiet period. Your post on the burn rate was quite informative BTW. But back to the discussion on rates of return. It’s not that an individuals interest rate will make them more or less likely to default. An individual’s circumstances are unknown and unrelated to interest rate paid on borrowed money. But for a group of borrowers with similar credit profiles, a statistical model can be developed. The more loans a lender has, the more closely their portfolio should resemble the historical model of defaults. What this really affects is the lender’s return. The interest rate an individual borrower pays is based on the historical default rate for their credit profile to compensate the lender for defaults from that group. Because lower credit scores tend to have higher default rates, they pay more interest as a group to compensate the lender for individual loan defaults. However, Prosper’s bidding model dissociates interest rates from risk, which means a lender may not be adequatly compensated for defaults. I don’t think most lenders researched that very heavily, so I like that Prosper spells it out clearly.
Also, Prosper’s default rates aren’t grossly out of line for a given credit score. Prosper’s defaults don’t conform to traditional historical models because Prosper will accept people outside those models, while Lending Club does not.
Overall, I agree with what you are saying, but I still think that we may be passing each other somewhat.
Daniel: “Prosper’s bidding model dissociates interest rates from risk, which means a lender may not be adequately compensated for defaults.”
I am not sure that it disassociates the two. Although, it can get out of whack sometimes like the crazy early Prosper Pennsylvania loans that were beyond reason.
Daniel: “Prosper’s default rates aren’t grossly out of line for a given credit score.”
I agree more with this second statement. They can often get a little out of line, but it is not often that they go ridiculously out of sanity.
[...] the site map. Thanks!I mentioned earlier that I had one Lending Club loan fall behind, but the loan recovered in less than 30 days. I now have one Lending Club loan more than 30 days [...]
It?s nice to now finally locate a web site where the blogger is knowledgable.
Lending Club now subtracts their expected default rate and their annual expense rate from the posted interest rate when you add a loan to your portfolio. This gives you more reasonable expectation. For example, a 13% loan would have about a 1.7% default rate, and then a .7% annual expense rate so the expected rate of return is closer to 10%. It’s a nice feature to see on the site.