Thursday, June 21, 2012

Lending Club Loan Length and Default Rate

If you are new to peer to peer lending, check out the introductory article An Introduction to Peer-to-Peer Lending by Peter Renton.

Loan Length

Lending Club issues loans with two different maturity lengths - 36 months (3 years) and 60 months (5 years). The loans with 60 month term were first introduced to Lending Club platform in May 2010, a little over two years ago. In 2011, the loans with 60 months term became very popular as more than 50% of the loans listed on Lending Club had 60 month loan repayment term. The chart below shows the number of loans listed every quarter with 36 and 60 month terms. It appears fewer loans with 60 month term are being listed in 2012.

Default Rate

The chart below shows the percentages of loan with 36 and 60 month terms in either charged-off and default status or in grace period, late and performing payment plan status. As the chart shows, the charged-off and default rate for 60 month term loan is almost twice the default rate for 36 month term loan for the same quarter of loan listing date.

Originally, I assumed that fewer borrowers may default with longer maturity of loan as borrowers will have smaller monthly payment. The chart shows this not to be the case. In fact, the borrowers are twice as likely to default with loans of longer maturity.

This observation is inline with Michael's real-life experience of much higher default rate with 60 month notes as described in his blog post Balancing Your Portfolio by Loan Term - 36/60 Month. A quick back-of-the envelope calculation for his portfolio default and late payment rate is shown below. It will also help some readers who requested examples of using default rates.

Assuming Michael has portfolio of 100 notes with 76% of portfolio in 60 month notes and all notes were listed in second quarter of 2010.
Number of 60 month notes in portfolio = 76% * 100 = 76
Number of 36 month notes in portfolio = 100 - 76 = 24
From the above chart, the default rates for 36 month and 60 month notes are 6.24% and 11.45% respectively. Similarly, the late payment rates are 3.36% and 3.61% respectively.
Default and late payment rate for 36 mo. notes = 6.24% + 3.36% = 9.60%
Default and late payment rate for 60 mo. notes = 11.45% + 3.61% = 15.06%
Number of 36 mo. notes in default or late = 9.60% * 24 = 2.304
Number of 60 mo. notes in default or late = 15.06% * 76 = 11.4456
Percentage of notes in default or late that are 60 mo. notes = 11.4456 / (2.304 + 11.4456) =  83.2%
Michael reported 95% of his notes in default or late are 60 month notes. Considering our assumptions of all notes issued in same quarter, we came up pretty close with 83.2% of notes in default or late should be 60 month notes.

Days between Loan Listing and Issued Date

The chart below shows the number of loans issued as function of days between loan listing and issued date between 2010 and 2012 for both 36 month term and 60 month term. The dotted line represents decile and each decile is 10% of loans. For example, one decile line indicates 10% of 36 month term loans were issued within three days of listing while 10% of 60 month term loans were issued within four days of listing. Similarly, two decile line indicates 20% of 36 month term loans were issued within six day of listing while 20% of 60 month term loans were issued within seven days. The chart shows that loan length has negligible influence on days required to fund and issue the loans. Update July 6, 2012: Please ignore the decile lines on the chart and related findings. There is a mistake in my interpretation of how Tableau calculated the deciles.

The chart below shows the percentage of loans with status charged-off and default or in grace period, late, and performing payment plan for both 36 month term loans and 60 month term loans for 2010 and 2011. The default rate for 36 month term loan is much higher when such loans were issued about 3 weeks after being listed. The default rate for 36 month term loans rises when it takes longer to issue loans after being listed. The default rate for 60 month term loan shows very different pattern. The 60 month term loans issued within four days of listing have much higher default rate.

Key Takeaways

  • Reducing the monthly repayment amount by lengthening the loan maturity doesn't reduce the risk of borrower default. In fact, the borrowers are more likely to default with longer maturity debt.
  • The coat-tailing strategy, described in my previous post Days Between Listing and Issuance of Lending Club Loans - Rush at Listing Expiration, may be a good strategy to invest in 36 month notes.
  • Personally, I have stopped investing in 60 month notes until I further understand the mechanics of setting interest rates for such loans and able to observe the default rate of 60 month loans to maturity.

Interested in Data Visualization, check out The Visual Display of Quantitative Information and New Perspectives on Microsoft Excel 2010: Comprehensive.


  1. Great article Anil. This is really a must read for every Lending Club investor. I have recently reduced my percentage of 60-month loans because I was seeing this same trend. When these first batch of 60-month loans reach maturity I think we will find that they underperformed 36-month loans significantly.

    Maybe this is why in January LC made the conscious decision to pull back the number of 60-month loans on their platform. They said the reason was investor demand - which may have been caused by the underperformance that you demonstrate here.

    1. Peter, thanks for the encouragement. I am working on couple more posts on loan length that will shed much more on 60 month loans and associated issues.

      I am glad to read that LC tightened up on 60 month loans. I was really surprised to notice the large number of 60 month loans listed last year.

      Overall, the issues come down to lenders' lack of knowledge of consumer lending practices, LC's lack of focus on educating borrowers and lenders, and most probably regulations that force LC to follow same consumer lending practices as financial institution.

      Thanks for your comment. Keep up your good work in promoting P2P lending.


  2. I've been with Lending Club for 9 solid months now and next week I will enter my 10th (based on payments received). I currently have exactly 300 notes. Of those, 62% are 60-month loans and 38% are 36-month loans. I currently have 0 loans that are 16 - 31 days late, 0 default/charge-off loans, and 2 loans 31 - 120 days late. My NAR is 18.39%. One of my late loans is on a payment plan so I expect that one will be made current. The other just entered collections. Both of those notes are 36 month loans, I have no 60-month late loans (yet).

    When I choose loans I could care less whether they are 36 or 60 months. My criteria is based solely on ability to pay and the use for the money the borrower states (I don't fund business, auto, or medical). I consider myself a newbie at this and won't feel experienced until I have results from at least 2 or 3 years of investing, so I'm not offering anything yet other than questions. The question I have is do you really think well qualified 60-month borrowers will default at a higher rate than well qualified 36 month borrowers (comparing apples to apples)? Is it possible that the higher 60-month default rate in your study resulted from a higher concentration of high risk candadates? That is to say, is it possible that the 60-month loans are already riskier (because of their higher numbers in riskier asset classes) than the 36-month loans? I saw no mention of asset class (A vs. B vs. C, etc.) and 60-month default rates. From my limited expericience with Lending Club I've noticed that there are far more 60-month applicants In the higher risk classes than the lower.

    Obviously I've taken on more risk with the higher NAR I currently have, but the 60-month notes I have are little different from the criteria of the 36 month notes. Am I being foolish to expect the same results from my 60-month notes as my 36? I'm very early in this game and have a 5 year (investment) plan for Lending Club, so any useful advice is highly appreciated.

    1. Dennis,

      Thanks for your detailed comments.

      Either you are very lucky or you have cracked the code of high-risk lending considering your NAR and status of loans. "Ability to pay" is the key criteria but very difficult to determine. How do you determine ability to pay?

      Personally I feel, in aggregate, defaults for both 36 mos. and 60 mos. should be similar as both loan lengths have pros and cons, for example longer duration when unpredictability strikes vs. higher monthly repayment.

      My next two blog posts this coming will address the Loan Length w.r.t. credit grade and interest rate. Are your 36 months and 60 months loans have similar default risk? This is the key question. My guess from your statement "I could care less whether they are 36 or 60 months" is that you are picking 60 month loans that have much lower default risk in comparison to comparable 36 month loans.

      Stay tuned for my posts this week as they may shed better light on your questions.

      Thanks for reading and contributing to blog posts on P2P lending.


    2. "How do you determine ability to pay?"

      - DTI must be less than 20%
      - Monthly payment must be less than 10% of gross income
      - Must be employed (no self-employed)
      - Minimum 5 years at same job
      - Minimum 10 year credit history
      - 0 default history
      - Minimum yearly income must be $25,000

      From the above I at least know the borrower has the ability to pay as opposed to someone who has far too much debt, a shady employment record, and an irresponsible payment history. I also avoid certain professions such as those that involve some sort of sales (dependent on commissions). There are other things I look for but I'm not going into all that here.

      As you might imagine it takes a lot of time and patience to find the right borrower, but I do find them. I invest heavier in D and E notes, but lately have been buying more F as I gain more confidence. Obviously with an 18.39% NAR I've taken on more risk, but that is what I want as I have mostly conservative investments in my overall portfolio (not one penny in the stock market).

      As I'm only going into my 10th month with Lending Club, I know the results I'm getting now may not be indicative of what to expect going forward. I'm still just a newbie, my notes are mostly young, and I'm still learning. I expect I may get those defaults everyone talks about, but so far so good, I remain cautiously optimistic.

      "My guess from your statement .... you are picking 60 month loans that have much lower default risk in comparison to comparable 36 month loans."

      Actually just the opposite. Try finding 36 month notes in F and G class notes, it's almost impossible. It's even hard to find them in E class. Most of my 36 month notes are therefore in th A through D class with not many in D.

      Maybe I've just been lucky as you suggest, but until I start seeing poorer results I'm not changing anything. Like I've said, I'm still just a newbie at this so I wouldn't read too much into my results yet. Maybe in a few months I'll be crying rivers over bad loans.

      I look forward to your future posts and thanks for allowing me to comment here.

    3. Dennis,

      Your criteria for ability to pay is very interesting as these are the factors Lending Club doesn't consider in assigning Credit Grade. Thanks for sharing your criteria. Most lenders' criteria, I see on the net, tend to focus on factors that are already included by LC so those just result in restrictive loan selection. It will be interesting to see how this pans out as loans age. Keep us updated on your performance. It will also be interesting to compare when I reach to the factors associated with borrowers in my analysis.

      As you will see from my post tomorrow morning, the default risk for your 60 month loans is most probably similar or better than that for 36 month loans.



  3. I follow a VERY similar strategy to Dennis, my MOST important criteria being that the monthly loan payment must be less than 10% of the income of the person. I also use the no default, and 10 years of credit history criteria, minimum of 3 years at same job, DTI under 20%.

    I also have been investing in mostly D, E, F, and of over 1,000 notes purchased, I've had 1 charged off note, 1 in default, and I have sold off roughly a dozen "problem" notes over the last 8 months... but I would say that overall, the strategy of finding borrowers with the ability to pay has been working well.

    My NAR has been steadily increasing since switching strategies and is now over 12% (was under 10% when I was strictly investing in A/B/C notes for the first 6-8 months). I expect this gradual trend to increase and settle at around a 15 to 16% NAR.

    1. Rajuabju,

      Thanks for sharing and confirming Dennis strategy. It will be interesting for me to see if data supports this strategy, when I reach these factors in my analysis.

      I appreciate you taking the time to comment and reading my blog.



  4. Anil,

    I'm new here -- but not new to P2P lending. Thanks very much for the insightful post. I have been investing in P2P for over a year now, and have interestingly refrained almost entirely from participating in 60-month notes. This was due in large part to my early perception of the fragility of the asset class. While my concerns around risks of fragility are much less prominent than they once were -- I'm maintaining the vast majority of the portfolio in 36-month notes for many of the same reasons cited above.

    Have a few different portfolios running where I'm employing a few different strategies -- but all told, I think early on I picked the right strategy for the wrong's to hoping I'm picking the right strategy for the right reasons from here on out.

    Thanks much for your writing; appreciate it greatly and please keep it up. Best,


    1. Daniel,

      Thanks for the encouraging words and welcome to the blog. I am enjoying analyzing the historical loan data file from Lending Club. Also hoping these insights will help me and my readers in coming up strategies grounded in data.

      I am thinking of back-testing some of the mentioned strategies with historical loan data file. Let me know if you need any help in back-testing your strategies. :-)

      Thanks for your comment. Your feedback and suggestions are always welcome.


  5. 15-16% long term NAR isn't very reasonable. I have been on the site since early 2009 (Feb?). I have around 600 total notes between $25-$50/piece. My criteria is much like yours: DTI<15%, no defaults, at least 3 years employed, >$4K/month, <$25K loans, Grade B and higher (mostly C), etc. etc. I am around 9.5% NAR and should be dropping to around 8.5% by the time the next few loans default (a few bankruptcies). I don't invest in businesses, medical.

    The long term NAR is much lower than people who have been on the site for 1-2 years would tell you.

  6. Seems to me that once you get past the scammers and marginal loans, once you put some filters against the higher risk loans your likely defaults are going to be due to major life events like death, divorce, job loss, major injury/illness. Unpredictable, and the rates of these things happening vary greatly with factors such as local cost of living variance, divorce rate, death rates, etc. I think this explains some of the fatness around higher rates of default in some states, because those appear to have higher costs of living and divorce rates. Other states have low cost of living and divorce rates, but higher rates of death and illness or its difficult to find jobs there. Even a great borrower with great credit and the ability to repay the loan easily can turn into a deadbeat in 5 seconds with zero predictability due to injury or divorce. The only way you overcome this is diversification.

    With that in mind, with a 60 year loan you're adding two more years of the potential for one of those life events to occur. In that vein, its not fair to compare a 36 to a 60 straight up, but rather a 36 year and the last 2 years of another 36 to even things up, because lets face it are you going to take the money and spend it when the 36 month loan ends, or are you going to reinvest the $? Yep, you're reinvesting at least a fair bit of it! In fact, since a majority of default situations occur in the first year, you might be taking on more risk by buying 36's and then reinvesting in another 36 when the first one is paid off.

    I do see from some of the data that the 60's historically had higher first few month defaults. I suspect lending club had a learning curve early on about who to select and what to charge for rates. Otherwise I'd see no reason why with all things being equal, a borrower would stop paying in the first few months of a 60 month loan with a smaller payment than the same qualified borrower would with a larger payment on a 36.

    One thing I focus on, which sadly lending club won't let you filter on, is I rarely write a loan where the loan payment is more than 10% of the borrowers gross, I'm really careful of people asking for lots more money than their revolving debt shows and I avoid a lot of weddings and home improvements to borrowers with huge revolving debts who just decided it was a good idea to spend $30k on a wedding or have to do emergency repairs on their house right after they bought 50k worth of toys. I try to focus on debt/credit consolidation.

    I don't think people will pay 15-20% plus of their gross to a debt pile for 3-5 years reliably. Its too big of a nut for too long without much gratification coming back.

    You also have so much front loaded interest on the 5 years, that a default in the 5th year just doesn't hurt as much as a default in the 3rd year of a 36 month loan.

    So I think the chart you want to see is projected loss per month of engaged investment, rather than straight up performance of a 36 vs a 60. I'm betting that matching a 3 year plus the last 2 years of another 3 year since its already a mature loan.

    The other factor that has to be adjusted for is the interest rate. They pay a lot more on the 60's than the 36's, adjusted for the same credit score and other risks. So even in a straight up analysis with more defaults on the 60's, are you making enough extra in the front loads to accommodate that?

    After one year, I have about the same defaults on 36's as 60's.

    1. Tomh,Thanks for your detailed comments. There are quite a few sites including PeerCube that help you filter loans using lot of different available attributes. I will try to address some of the raised points in future blog posts.