Another Analysis of Default Rates at Lending Club and Prosper

I have been following the batch of loans that were issued between July 2009 and June 2010 for quite some time now. You can read my first analysis of these notes here and the second analysis here. This latter analysis was done over a year ago.

Many of the notes issued during this time period have now reached maturity and those that haven’t will do so within the next nine months. Regardless, we can assume that the vast majority of principal has been paid back on these loans so we can get a very good idea now of their overall performance.

Below is the table showing the results from both Lending Club and Prosper.

LoansDefaultsDefault RateDefault LossLate LoansLate %Int. RateROI
Lending Club781283610.70%7.34%2082.66%12.47%5.51%

Here are a few points to make about this table:

  1. This is a comparison of 3-year loans only, the first two analyses inadvertently included a few five year loans at Lending Club.
  2. For the Lending Club data I used Lendstats and for the Prosper data (other than ROI) I used Prosper’s Performance page.
  3. The Default Rate is not an annual rate, it is the total percentage of notes that have defaulted.
  4. The Default Loss percentage is the total amount of principal that has been lost. This number is lower than the Default Rate because with most defaults some payments are made before the loan defaults.
  5. The ROI column is not exactly an apples to apples comparison – I use Lendstats for the Lending Club ROI number and Prosper Stats for the Prosper number. Lendstats no longer provides analysis of Prosper data.
  6. Prosper has a higher default rate but also a much higher average interest rate that more than compensates investors for these defaults.

Even though the ROI numbers were not calculated using the same formula it is pretty clear that Prosper has outperformed Lending Club over this time period. This is confirmed by the Lendstats numbers from their home page that show Prosper with a clear advantage over Lending Club for notes issued in 2009 and 2010.

Comparing Lending Club and Prosper Returns on Lendstats

More P2P Loans Have Reached Maturity

One of the great things about investing in p2p today is that there is a bigger database of completed loans than ever before. We finally have some Prosper 2.0 loans that have matured and it is pretty clear that they came out after their quiet period and provided excellent returns. Those investors who were brave enough to invest in Prosper 2.0 right after they opened back up have been rewarded very well for the risk they took.

Now, the past is no guarantee of future results and this article is not meant to say that Prosper is a better investment than Lending Club today. In fact, the Lendstats numbers seem to indicate that Lending Club is doing a little better than Prosper for notes issued in 2012.

What is great for investors, though, is that we will be able to see for ourselves. By the end of this year all 2009 loans will be in the books. There will be no more guessing on loss rates, the data will speak for itself.

I will do another update on this batch of loans next summer when all the notes will have matured. But I will be surprised if we see numbers that are much different from today.

I am interested in hearing from others. What do you think of this data?



  1. Grant says

    I think this data just shows that Prosper and Lendingclub are two different companies with two different risk pools and nothing more. For example, I suspect Lendingclub loans are weighted more heavily into the lower risk a/b/c loans vs Prosper and will therefore show lower rates of return. Only by comparing apples to apples can we figure out which platform offers truly better returns –I’d hypothesize this isn’t possible as the underwriting criteria of each is both distinct and propriatary.

    Thanks for the great work!

    • Roy S says

      I think Michael at nickelsteamroller attempted to make an apples to apples comparison, but I’m not positive. I don’t think we’ll ever be able to get a true comparison between the two companies, and I know Prosper has outright stated several changes they have made to their underwriting model on their blog. So it is difficult to compare newer Notes with older Notes on the same platform. This is just an article of interest. Probably the best takeaway is that Prosper and LC are two different companies with two different UW models (not to mention collections, fee structures, etc), and diversifying between them is probably a good idea.

  2. Blake says

    I think initially the ground game for both companies was to find money (investors) and fund borrowers. Their metrics seemed to revolve around who could become Coke instead of Pepsi the most quickly, i.e., who can fund the most loans.

    But now we are all analyzing how to increase returns. We want long-term stability. Its the same reason you invest in mutual funds with long-term track records. Doesn’t matter if they’re the biggest, just give me the solid return.

    So now the question is becoming: will meet the bottom-line issue that concerns new and old investors alike which is: COLLECT MY MONEY.

    With the market for P2P obviously very real, now it will be interesting to see who can really come through on collecting-operations and not just funding-operations.

    • says

      Blake, I have spoken with the head of collections at both companies recently and I wrote a report in each case:

      Having done this research, though, I am not convinced collections will make a big difference. At most you might see a 1% increase in return from a truly stellar collections program. The problem with most borrowers who default is that they are in severe financial trouble, often through no fault of their own, and no amount of skillful collection practices will be able to see a successful repayment of a loan.

      I think long-term stability is a focus for both companies now as it is pretty clear now that p2p lending is here to stay. And long term track records here will be key to attracting investors.

      • Roy S says

        “most borrowers who default is that they are in severe financial trouble, often through no fault of their own?” I disagree here. Yes, there are times where things are so out of control that you really can’t blame them. But I think for the majority, it is just a lack of financial knowledge and/or discipline. In a more hopeful sign, you do see a lot of debt consolidation loans. I like to think that people are starting to take responsibility for their behavior (even if it is only because they really don’t have any other choice). Here, though, the fault is their own. The problem is that, while they are attempting to readjust their behavior, they are either: 1. Already in over their head in debt; 2. Want to get out of debt but lack the knowledge/skills to effectively do so; or 3. Still just undisciplined enough to actually follow through. At least that’s the lens through which I view the defaults.

        • Reya says

          There aren’t many people for whom #2 is true. You have to be REALLY dumb for that to be you–willfully stupid for most people.

      • Blake says


        You bring up a good point, and I have seen that post–it’s great.

        But until the debtor declares bankruptcy, I would think some hard pursuit of the debt would be in order. Two of my three loans in default right now are making 6 figures.

        It would be interesting to find any statistics about the default rate on credit cards to see how P2P collection practices compare. I have a friend who owns a debt collection agency…I’ll try to find some info.

        Other items that would be interesting to see improved upon over time in the P2P space (I don’t know about plausibility):

        Ability to spot and deny fraudulent loans
        Possibility of Lending Club sending funds for debt consolidation loans directly to credit card companies
        And while this is probably extremely long-term, it would be interesting to see if P2P ever gets into secured debts (car, home, savings).

        • Dennis says

          Hey, I’m very curious about their default rate, are those statistics published all true, that they have single digit default rate?

          Blake- of what rating were your 3 loans?

          • Blake says

            Sorry Dennis, I don’t understand what you mean by “are those statistics published all true, that they have single digit default rate?”

            The three loans I have currently have in default are D3, D3, and C2. Most of my notes are D grade to E grade.

          • says


            You should check out this article from Prosper’s blog a couple of months back that had a graph of credit card delinquencies and charge-offs going back over 20 years:

            And Dennis, like Blake, I am not sure which numbers you are referring to when you ask: “are those statistics published all true?” The statistics I quote in the table above are from publicly available information.

      • Reya says

        ” often through no fault of their own,”

        Ummmmm…..NO. If people didn’t make stupid decisions with credit, places like Lending Club wouldn’t exist, and the personal loan market would be maybe 2-5% of what it is now. Choosing good loans is about choosing the people who are the least foolhardy with their money or who are ready and able to turn a corner.

        Or people who will at least ride the straight-and-narrow long enough to pay off the loan before falling into stupidity and cupidity again……..

  3. Walter says

    I think a guest post from a bankruptcy attorney would be a good idea, just to give us a clearer idea about the pertinent bankruptcy fraud laws for a p2p borrower. Perhaps this has been covered in another post somewhere?

  4. says

    Think Peter’s right, re: collections. Short term, intense focus on hard-ass collection efforts may make not-sufficiently-diversified lenders feel better, but, long term, sufficient diversity’s really the only way to go, so, that’s the way both platforms are going. When there’s money, non-marginal, to be made in chasing people down, that market / those services will come…

    Meantime, diversify to the point that you can sleep well at night.

    • Dan B says

      I agree with Peer-Lend & Peter on the whole “collections efforts” thing. No one likes to be stiffed & it seems that a lot of the understandably strong emotions comes within the context of having been stiffed, deceived etc……………….as opposed to a financial gain from potential collections efforts.

  5. Dan B says

    Peter……….In regards to the “late” numbers comparisons, how are you treating the “in grace period” & “payment plan” loan numbers at LC vis a vis Prosper, given that Prosper doesn’t have those 2 categories?

    • says

      Dan, Good point. Late loans at Prosper and Lending Club are treated differently. There are no payment plans loans at Prosper and more importantly the In Grace Period loans at Prosper do not show as late on their statistics page. This means that the late loans number for Lending Club will be somewhat inflated. If we exclude all Payment Plan and IGP loans then the Lending Club late loans number is only 1.33%.

  6. Blake says

    I agree as well that at the end of the day you have to evaluate if the money is financially worth pursuing, BUT…

    If the collection efforts aren’t strong now, then money won’t be paid back. Peter talks about how some borrowers are so broke that they can’t pay it back, but unless they are unemployed without income or have claimed bankruptcy, they are probably paying the creditors they have that bark the loudest. That’s how you get your money back from people who aren’t managing their money.

    If you have strong collections, you set precedent for the future. Also if you wait, you increase the likelihood of having to settle the debt in the future at pennies on the dollar.

    Wal-mart learned this lesson, albeit in a different context. They were receiving false injury suits from customers and would pay to settle rather than litigate. They decided to fight even the smallest suits and reversed a perception that they would be intimidated. I happen to think collections is of the same vein.

    The fact that in Peter’s interview with Lending Club the lady wouldn’t disclose how many in-house collections agents tells me it’s not something they are all that proud of, especially coming from a company that will give you historical data on every freaking note on their platform.

    • says

      When it makes sense for the p2p platforms to have attorneys on staff in every jurisdiction, I’m sure they’ll hop right on it. Until then – and it’s gonna be a while, so accept it – diversify.

        • says

          Not being snippy with you, just trying to point out that it’s not economically viable for the platforms to do what you suggest *at the moment*. In a perfect world, I agree – your suggestion makes total sense. In the current world, think it loses more than it makes, ergo, probably not happening anytime soon, so, I’m acting accordingly, etc.

  7. Blake says

    I think we agree much more than it has come across ha ha.

    I’m not asking for improved collections to all magically happen right this moment. While I can see how theoretically an emphasis on collections could “lose more than it makes” I guess I haven’t seen any solid data to back that up–once again, very well may be true. I don’t want them to pursue collections just to prove a point if I’m gonna lose more money in the LT.

    Another consideration: when you think about LC’s model, they are making their money much more on the front end. So they’ve got to just want to FUND and as long as the screening, and as you said, the diversity, is high enough to generate respectable returns, they’ll continue to attract lenders and make their bucks off the borrowers.on the front end. The 1% service fee they charge investors probably isn’t all that appealing when it’s on late accounts.

    So unless they see overall ROIs drop in relation to our expectation, you’re right I don’t think they see much value in being aggressive.

    I’d like to see them auction off our notes more, but maybe they’re not doing that because, as you’ve said, they know when they hit a certain growth benchmark, they can go back and really pursue the late ones at cost that makes sense.

    • says

      Blake/Peerlend, My observations of Lending Club is that this year they have made a more stringent effort with collections. But the proof will be in the pudding. I would like to see them share statistics with investors as to how they are improving in bringing late borrowers back to current. They have made a couple of half hearted efforts in this regard but nothing recently. I am going to continue to try and push for some more information here.

  8. Matt says


    This is one of the best and most informative posts that you have put together. IMO, the key to maximizing returns is to understand the default rates. I think it would be interesting to see a chart identifying the default rates by occupation during this same time period.

    As far as implementing a successful collection process, I would expect an increased return of 2-3%, on the low end. Never underestimate the power of garnishment and levying on bank accounts.

    • says

      Matt, I agree that trying to lower default rates is one of the most important ways to maximizing returns – that and making sure you have a fully diversified portfolio. There are plenty of criteria where default rates can be very different for different kinds of borrowers and I admit I have never looked closely at occupation.

      Regarding your comment on collection practices I think your estimate is high. Although Lending Club in particular believe it is worth considerable effort as they are putting more effort into getting judgements and garnishments. I would love to see some metrics here but they are reluctant to share anything with me.

      • Matt says


        I’m of the belief that occupation is one of the most useful criteria in evaluating a loan. Some of the best returns come from Doctors, Lawyers, Scientists, Pilots, RN’s, Computer Programmers, Clergy, etc. As far as I’m aware, Judge is the only occupation to not have a single default on Prosper. Yes, I know the sample size is small (approximately 40).

        Since Lendstats is down, it is more difficult to examine Prosper data. I think more analysis of occupations would be very insightful.

Leave a Reply

Your email address will not be published. Required fields are marked *

Notify me of followup comments via e-mail. You can also subscribe without commenting.