Free Updates

Exclusive content to your inbox for FREE!

Lending Club Stats Now Include Recovery Rate

by Peter Renton on March 18, 2011

Lending Club Bad Debt Recovery Rates

Last week Lending Club quietly made the addition of recovery rates to their performance page. The graphic above (click on it for the full size version) shows recovering rates for all their p2p loans that were not current as of the end of August 2010.

What Does Recovery Rate Mean?

First let’s define what they mean by recovery rate. When a loan enters the “In Grace Period” it means the borrower is late on their loan payment. In Grace Period means the loan payment is between 1 and 15 days late. They become officially late if the payment has not been made within 15 days of the due date.

In this chart above, what Lending Club has done is look at all the loans that were not current at the end of August 2010. They took a six month period and worked out how many of these loans were recovered. You can see that 83% of loans that were officially In Grace Period  at the end of August had been fully or partially recovered by the end of February 2011.

Now, they do say fully or partially recovered and so you might think that there would be a big difference between the two, but this is not the case. Let’s provide an example to illustrate. Say a borrower has a $20,000 loan with a $500 monthly payment. In August 2010 he falls behind and and enters the Late (15-30 days) category. To be included in the recovery numbers the borrower must have made full payments on all his past due balances. I double checked this with Lending Club.

So to take our example further this borrower must have made at least $2,500 in loan payments (excluding late fees) to have been considered fully or partially recovered. To be included as partially recovered the final loan payment (to bring the account current) must have only had a portion paid. If it was not made at all then it would have been included again in one of the late categories.

Why is Recovery Rate Important?

According to the chart above there is an 83% chance that a loan will be recovered from In Grace Period, a 75% chance that a loan will be recovered from 15-30 days late and (most surprising to me) a 55% chance the loan will be recovered if 31-120 days late. This is good news for those people who have just noticed one of their loans slip into Late status.

Many people assume that if a loan in your portfolio moves into In Grace Period then it is most likely going to default. These stats from Lending Club point out that this is not really true. I know from personal experience that many of my loans have gone into In Grace Period or even 31-120 days late and then they return to current and the borrower continues making monthly payments on time.

Now, I know many investors sell their notes on the trading platform as soon as they enter the In-Grace Period status. This may well be prudent if you can get close to the face value of the note, but if you are taking a 25% hit on the value of these notes then maybe you should reconsider.

My Take on This Change

This is a step in the right direction for Lending Club in that it provides more data for investors. But more needs to be done. Ideally the statistics page at Lending Club should be more interactive and more like the loan filter at Lending Club has all the data, but they are not allowing investors to access it easily. In talks with Lending Club management they have indicated their intention to provide a much better statistical analysis page in the future which would be a welcome addition for many investors.

While both Lending Club and Prosper provide their entire loan database for download, neither company provides a robust front end to inquire into that database. While I applaud Lending Club for providing this additional information there is still no way to run filters on their entire database except through a third party site like

Lending Club has indicated that they will be updating their recovery rates monthly. It will be interesting to see the trends in these numbers. Right now, I think they are better than most people (I speak for myself at least) expected. But it would be great if they get even better.

Like what you read?
Then please join over 1,000 people who receive exclusive weekly p2p lending tips, and get a FREE COPY of my ebook, Understanding Peer to Peer Lending. Just enter your name and email below:
Dan B March 18, 2011 at 11:01 pm

Ok this whole “fully” or “partially” recovered line doesn’t work for me because it’s way too nebulous. So if a note goes late & the borrower makes 1 payment & then disappears then that note counts as partially recovered?? Well that’s comforting. I mean come on. Why the smoke & mirrors? What this category should show is what percentage of notes return from each of the categories to being “CURRENT” again. It’s as simple as that.

And incidentally………..why look at just the ones that went bad on Aug 2010 when you have over 3 1/2 years worth of historical data? Why not look at all the notes since the beginning that have entered into each of the “late” categories & what percentage came back to being current & remained current?? That’d be a true recovery rate that actually means something.

But the thing that bothers me the most is that I know that “they” are aware of all these arguments that I’ve laid out…………………& they still decided to come out with this crap!

Lou March 19, 2011 at 6:56 am

This data allows us to create a Probability tree.

If you have 100 loans go into the Grace Period, statistically speaking, approximately 2 (1.86) will end up in Default (according to the Data Lending Club posted).

Herndon, VA

Peter Renton March 19, 2011 at 9:10 am

@Dan, I am sure you can surmise an answer to the money questions you pose, but here is my take. Lending Club obviously wants to provide data that is correct but that also paints themselves in a positive light. The reason I believe they don’t provide 3 1/2 years of data is because their underwriting standards have changed dramatically in that time and no doubt their collection rates have improved. I agree with them on this point. Providing historical data might actually be misleading because there is a smaller chance of default now when investing in notes than there was three years ago. Having said that, it would be more useful to at least see a trend over a few months than just the one month snapshot.

@Lou, Thanks for the chart. I am no statistics expert but that doesn’t sound right to me. If there is an approximate 3% annual default rate for all notes, that would mean far more than 2% of notes default when entering In Grace Period. Don’t you think?

Dan B March 19, 2011 at 10:56 am

@Lou………that doesn’t sound right & just can’t be right. But, if you think it is correct then I encourage you to go to the secondary market & scoop up all the “in grace period” notes that are selling at 4, 5, 10+% discounts. If you’re correct 98 out of every 100 you buy will go back to being current.

@Peter……….. I don’t know what came over me when I wrote my last post because I can’t imagine why anyone would doubt the accuracy & pure intentions of anything that’s supplied by a p2p lending company. By providing these unbelievably useful pieces of information, LC has set a new standard of “transparency” that all of us can only hope to aspire to………….LOL

Lou March 19, 2011 at 11:09 am

I think the discrepancy has to do with what Dan B pointed out, the loan may get current for awhile and then slip back into Grace Period or Late. So, the Probability tree is just a snapshot of what could happen to a loan, whereas the 3% default rate is comparing total loans to defaults. I could be wrong and it could be a rounding error on LC’s part. I’m not a statistics expert either and my initial analysis was based on my memory, but I did crack open my old textbook to check it.


Mike March 19, 2011 at 5:16 pm

Wonder what the recovery rates are for Chapter 7 bankruptcies….had another one in the last few days. An A3 loan, no less! LC will figure out a way to make it seem like it’ll be greater than 50%, I’m sure.

Peter Renton March 19, 2011 at 6:51 pm

@Dan, Lending Club are obviously doing what they think is best to grow their business. I think they want to be transparent as long as that transparency helps business.

@Lou, Be interested to hear Lending Club’s feedback on this or from some other statistics expert. Because your logic made sense to me, but the final number didn’t seem right. I am still baffled by it.

@Mike, Boy, with Chapter 7 bankruptcies I can’t imagine the percentage is much above zero for recovery rates. That is the whole purpose of that kind of bankruptcy – wipe everything out and start over. Bummer.

Dan B March 20, 2011 at 3:07 am

@Peter……….Then that’s not transparency at all. Transparency requires the disclosure of both the good & the bad. Most businesses don’t disclose the bad………….but then they don’t claim to practice transparency either. Selectively disclosing only the favorable details that help grow your business while implying that you’re being transparent in your disclosures can even be interpreted by some to be outright deception.

Peter Renton March 21, 2011 at 10:25 am

@Dan, I am going to defend Lending Club on this one. Transparency is always a relative thing, because absolute transparency is impossible. For example, we can’t know (nor should we) the inner workings of their management team, their detailed financial projections, their relationships with VC’s or many other things that may have a direct impact on our investment.

I will say this, though. Having looked at many other p2p lenders both here and in other countries Lending Club is the most transparent of all of them. Doesn’t mean there isn’t room for improvement, and I will continue to push them on this, but I think they are doing a reasonable job of it. As for the data you can download their spreadsheet and double check most things for yourself.

Now, in the interests of my transparency, I obviously have a vested interest in this. This blog receives income from both Lending Club and Prosper for referrals. So, while I try to remain objective, I may spin things in a more positive light than others with little or no vested interest.

Dan B March 21, 2011 at 3:33 pm

@Peter………Well I’m not sure if anyone actually gives a crap about the “inner workings” of their management team………….. In addition, I for one have little faith in any of their “detailed financial projections” & their “relationships with VC”, at least their financial ones are filed with the SEC & therefore public information.

As much as I’d like to applaud you volunteering that you derive some income from these companies, I think everyone here with a quarter of a brain already knew that. You Peter are after all, more often than not, the lone voice that comes to the defense of LC when a sharp criticism is made. And that’s fine & understandable. The question that I’m sure you’ve already considered is at what point you & this blog becomes perceived as nothing more than a mouthpiece of the p2p companies…………….& whether you care or not if/when it does become that. I don’t think you’re there yet, but…………….

Peter Renton March 21, 2011 at 4:50 pm

@Dan, I am just going to keep writing articles and comments calling it as I see it. It is up to others to judge whether or not I am just a mouthpiece of the p2p companies or whether I add value to the conversation. I will always strive for the latter.

Aaron March 22, 2011 at 5:14 pm

I am not a statistics EXPERT, but I think I know why Lou’s numbers looks so strange to most of us. There is no need for Lou to do any math. Here is the text directly under the graphs:

How to read these graphs: of the loans in Grace Period in August 2010, 83% was partially or fully recovered by February 2011.

You see, 83% of every loan that went into grace period in the month of August was not recovered. That means a whopping 17% was doomed as soon as the grace period began.

The thing that gets me is that they include the first two graphs in 15 day increments. That could potentially mean that LC uses a single loan that goes into grace period on August 1st and uses it for both graphs. The same hold true for the 30-120 days late graph. These are all statistics of what EVENT occured in August and not the cumulative totals that we assume when looking at the graphs.

Peter Renton March 22, 2011 at 8:53 pm

@Aaron, You bring up an interesting point. The pie charts don’t say as of August 31st, they say for the month of August, so it is highly likely that a loan may make into into two or more charts. This may well explain Lou’s statistical anomaly, I was thinking that these charts were cumulative totals. Thanks.

Dan B March 23, 2011 at 6:07 am

This whole “late” thing is so much more complicated than it needs to be. In any case, in an effort to be more fair & more balanced than Fox News, I’ll throw this little twist into the equation & state that in some instances late notes aren’t necessarily that disastrous.

I have 2 late notes left in my portfolio right now. They have been in late status since August & September 2010 respectively. They’re both on payment plans & have made 7 payments (principal & interest) & 5 payments (interest only) respectively since going late. Though I completely understand that this is not a typical situation, it does point out that in these specific instances Lending Club paints a picture that’s a bit dimmer than it actually is by keeping these notes as “late” even though they’re making monthly payments………….but have not caught up on a payment they previously skipped.

Peter Renton March 23, 2011 at 6:39 am

@Dan, Thanks for brining up the payment plans, I hadn’t even thought about those. Most people who go on these kinds of plans are genuinely interested in paying back their loan and are a better risk than those who just stop paying.

Previous post:

Next post: