Who to Lend to at the Lending Club

Posted on Oct 21, 2016

Technological disruption is affecting many industries, and dusty old consumer lending is no exception. Peer-to-peer lending -- private individuals lending to one another rather than from banks -- has been growing exponentially over the past 10 years, and Lending Club is a lead player. This post analyzes trends in Lending Club's loan portfolio and shares an interesting observation that may help lenders maximize their returns.

But first, how exactly does Lending Club work? Say you have some spare cash. Lending Club now gives you an alternative to lending it to your crazy uncle. Instead, you can lend it to a...stranger! That may sound scary, but Lending Club handles the movement of money back and forth between you and the borrower. Rather than using Lending Club to borrow for new investment or consumption, borrowers are typically trying to consolidate existing debt. The below word cloud illustrates the prevalence of "debt consolidation"  and "credit card" in the free text field "loan purpose". Borrowers want to consolidate their debt at Lending Club because they think they can get a better interest rate than they pay on their credit cards. 


Most used words when applying for Lending Club loan

This leads to the other important service Lending Club provides:  they estimates the credit risk -- the risk of not getting paid back -- of each loan. They give each loan a “grade” (just like at school), and based on this grade, assign an interest rate. The lender can choose from a spectrum of low risk, low return to high risk, high return loans. As you can see in the below chart, this interest rate has changed quite a bit over time.  The interest rates on the high grade loans have been stable over time, but the rates on lower grade loans have gone up considerably. You can also see that the difference in interest rates between loan grades is quite high. For example, you may be paying 10% more interest if you have a grade C loan as opposed to a grade A loan.  That’s $500-1000 on a $10k loan!

intratetrendSo how does Lending Club determine this all-important loan grade? The industry standard for estimating credit risk is an individual’s FICO score, which is used for everything from credit card limits to mortgages. You can see that back in 2007, Lending Club’s loan grade metric was highly correlated with FICO scores: high grades were associated with high FICO scores. However, over time, the range of FICO scores used to give both high grade loans and low grade loans increased. 

Particularly intriguing were the low FICO score people getting grade A loans and the high FICO score people getting lower grade loans. As context, you would have trouble getting a typical mortgage at all if your FICO score were much below 650. But if you had a FICO score of 750-800, you would have banks chasing you down the street with free toasters to get your business! I investigated this latter group further. How do these credit golden children end up with, say, a grade C loan, paying 10% more than one would think they should? I called this group “outliers”.


If these outliers were, indeed, unjustly given lower grade loans than they deserved, we should be able to see that in the default rates of the loans. However, the chart below illustrates that, although the outliers’ default rates are a bit lower than peers with the same loan grade but lower FICO scores, they are not generally low enough to justify being bumped up a grade. For example, the red C grade bar would have too high a default rate to justify being moved to a B or A grade loan, despite the fact that all its constituents have FICO scores above 750. In short, Lending Club assessed the outliers' credit risk accurately, despite their relatively lofty FICO scores. 


So how does Lending Club figure this out? How do they know that your crazy uncle with an 850 credit score really deserves a grade C loan? The short answer is we don’t know. However, there are characteristics that suggest some possibilities for further analysis. 


For example, in the above graph it's clear that the loan amount seems to be a little higher for outliers than for non-outliers. Also, it looks as though the outliers have more recent credit inquiries and are more likely to have recently opened a credit account than non-outliers. This is reflected in the shape of the violin plots for Mths_snc_rct_inq and Mths_snc_rcnt_acct. Therefore, it may be the case that a potential borrower applying for a larger loan than usual and opening up credit accounts recently may be interpreted by the Lending Club as warning signs that their credit risk is higher than that implied by their FICO score alone.

In summary, the Lending Club's credit rating practices have matured considerably over the past 10 years, particularly with respect to their divergence from the FICO score-based credit ratings. However, the FICO score is still a useful metric for a lender: although the Lending Club has a more accurate algorithm for measuring default risk than FICO, within a given loan grade borrowers with high FICO scores tend to default less frequently than borrowers with low FICO scores in the same loan grade. This is valuable info for a potential lender!

Sample Code:

About Author

Jason Sippie

Jason has an eclectic skill set including programming, data warehousing, business intelligence, and risk management, spanning Pharma and Finance domains. With one career in technology consulting and a second in financial services, he is excited about leveraging these...
View all posts by Jason Sippie >

Leave a Comment

Jeannetta Massy December 14, 2017
Isn't the world great, it has been good to me, thank you wix!!
Adalberto Fribley December 12, 2017
{Hey|Hey man|Hi|How's it going?|How are you doing?|what's up?|What's new?|What's going on?|How's everything?|How are things?|How's Life?|How's your day going?

View Posts by Categories

Our Recent Popular Posts

View Posts by Tags

#python #trainwithnycdsa 2019 2020 Revenue 3-points agriculture air quality airbnb airline alcohol Alex Baransky algorithm alumni Alumni Interview Alumni Reviews Alumni Spotlight alumni story Alumnus ames dataset ames housing dataset apartment rent API Application artist aws bank loans beautiful soup Best Bootcamp Best Data Science 2019 Best Data Science Bootcamp Best Data Science Bootcamp 2020 Best Ranked Big Data Book Launch Book-Signing bootcamp Bootcamp Alumni Bootcamp Prep boston safety Bundles cake recipe California Cancer Research capstone car price Career Career Day citibike classic cars classpass clustering Coding Course Demo Course Report covid 19 credit credit card crime frequency crops D3.js data data analysis Data Analyst data analytics data for tripadvisor reviews data science Data Science Academy Data Science Bootcamp Data science jobs Data Science Reviews Data Scientist Data Scientist Jobs data visualization database Deep Learning Demo Day Discount disney dplyr drug data e-commerce economy employee employee burnout employer networking environment feature engineering Finance Financial Data Science fitness studio Flask flight delay gbm Get Hired ggplot2 googleVis H20 Hadoop hallmark holiday movie happiness healthcare frauds higgs boson Hiring hiring partner events Hiring Partners hotels housing housing data housing predictions housing price hy-vee Income Industry Experts Injuries Instructor Blog Instructor Interview insurance italki Job Job Placement Jobs Jon Krohn JP Morgan Chase Kaggle Kickstarter las vegas airport lasso regression Lead Data Scienctist Lead Data Scientist leaflet league linear regression Logistic Regression machine learning Maps market matplotlib Medical Research Meet the team meetup methal health miami beach movie music Napoli NBA netflix Networking neural network Neural networks New Courses NHL nlp NYC NYC Data Science nyc data science academy NYC Open Data nyc property NYCDSA NYCDSA Alumni Online Online Bootcamp Online Training Open Data painter pandas Part-time performance phoenix pollutants Portfolio Development precision measurement prediction Prework Programming public safety PwC python Python Data Analysis python machine learning python scrapy python web scraping python webscraping Python Workshop R R Data Analysis R language R Programming R Shiny r studio R Visualization R Workshop R-bloggers random forest Ranking recommendation recommendation system regression Remote remote data science bootcamp Scrapy scrapy visualization seaborn seafood type Selenium sentiment analysis sentiment classification Shiny Shiny Dashboard Spark Special Special Summer Sports statistics streaming Student Interview Student Showcase SVM Switchup Tableau teachers team team performance TensorFlow Testimonial tf-idf Top Data Science Bootcamp Top manufacturing companies Transfers tweets twitter videos visualization wallstreet wallstreetbets web scraping Weekend Course What to expect whiskey whiskeyadvocate wildfire word cloud word2vec XGBoost yelp youtube trending ZORI