LendingClub Analysis: Insights from Issued Loans

Avatar
Posted on Feb 17, 2020

LendingClub Shiny App

Introduction

This dashboard was created by using a random sample of 500,000 records from the LendingClub data on issued loans. The original data set contained more than  2 million records. In order to create the Shiny app, the data needed to be sampled to meet the size requirements. Throughout this post, I will refer to the data as the sample set. The data consisted of loans issued from 2007 to 2018. Again, the data consisted of only loans that were issued to customers and not loans that were declined.  

Briefly, I will discuss total funding by State. Next, I will discuss why people utilize LendingClub’s loan services. After that, the grades of issued loans will be examined. More specifically, the types of loan grades by purpose and the frequency of loan grades over time will be discussed. To conclude, funded loans and payment totals will be explored. This is meant to serve as an analysis of why people utilize LendingClub’s loan service as well as an initial look at the outcomes of loans.

 

What is LendingClub?

LendingClub is a leading peer-to-peer lending service. The LendingClub (2020) company was founded in 2007. The company helps individuals acquire loans from other investors. LendingClub screens individuals for loan approval. Following screening, investors can analyze various metrics suited to their investment needs and risk tolerance. Once an individual is approved and investors choose to fund a loan, LendingClub aids the transaction process and services the loans. Investors and LendingClub earn money from the fees or interest. Meanwhile, borrowers can use the loans as they need.

 

Applicants & Loan Purpose

Loans from LendingClub were issued in nearly every State except for Iowa. California had the most frequent number of loans, followed by Texas and New York. Total funded amounts by state reflect these frequencies. California had the highest total amount of loans funded and the highest total payments received.

LendingClub acquires information from borrowers, such as to why they are utilizing their services. The most frequented purpose for a borrower to use LendingClub was for “debt consolidation”, “credit cards”, or “home improvement.” This trend was closely maintained over time as well. However, the “other” category closely followed home improvement loans.

Applicant income varied greatly, however, the median income was $67,000. The Consumer Financial Protection Bureau (2019) noted that the debt-to-income ratio is a measure of an individual’s monthly debts divided by their gross income. The debt-to-income ratio is one measurement lenders use to asses an individual’s capacity to pay the loan in full. The average debt-to-income ratio was 18.54%. Additionally, most individuals either had a mortgage (245,366) or rented a home (198,493).

 

Issued Loan Grades

Issued loan grades were determined by LendingClub. Grades ranged from “A” to “G”. “B” loans had the highest proportion of issued loans at 29.4%. “C” loans had the second highest at 28.7% and “A” loans were the third highest at 19.2%.  This trend was observed over time, however, starting in early 2017, “A” grade loans started increasing. “A” grade loans even surpassed all other issued loan grades October 2018.

Next, I examined the proportion of grades within each loan purpose. The highest proportion of “A” grade loans was observed in the “credit cards” (27.4%) purpose and for cars (25.4%). The highest proportion of “B” grade loans were observed in the same purposes. The proportion of “B” grade loans for “credit cards” was 34.1% and 30.8% for cars. The highest proportion of “C” grade loans were observed for “educational” (33.7%) or for “moving” (32.3%) purposes.

 

Loan Statuses

Approximately, 46% of loans were “fully paid” at the observation point. Additionally, 40% of individuals were “current” on their loan and 11.6% had their loans “charged off.” Kagan (2019) noted that a debt being charged off does not wipe the debt from the consumer and the “charge off” status can have serious implications for an individual’s credit. Typically, a charge-off occurs after 180 days without a payment. The lender determines the funds to be uncollectible, however, the individual is still responsible for the loan until it is paid or settled. Defaults and charge-offs are important for investors because they may be at risk of losing a portion of their investment with these loan statuses.

Below, the status of loans with respect to the issued grade and loan purpose are presented. We can see that “A” grade loans, compared to “B” or “C” loans, were less likely to be charged off. “B” grade loans had a higher likelihood of default, however, “B” loans were the most likely to be fully paid. “C” grade loans had the highest likelihood of being late in both categories. 

“Debt consolidation” had the highest likelihood of occurring in each loan status. Thus, while debt consolidation loans were more likely to be fully paid, they were also more likely to be “charged off” or in “default.” In the sample, only four purposes had defaults.

 

Loan Funding and Repayments

Overall, $7,524,284,350 in loans were requested and $7,521,835,375 were funded. At the time of observation $5,915,792,962 had been received in total payments (including interest and principal). Below is the amounts of loans requested per month compared to the total payments received per month.

It is important to provided context when discussing loan funding and repayments. As noted above, 46% of the sample had fully paid their loan, 40% of individuals were “current” on their loan, and approximately 12% had been “charged off.” To surpass the negative balance, there needs to be an approximate change in 21.35% of the total loans paid. Loan terms for LendingClub could be 36 months or 60 months. Due to the term lengths of loans and the observation point, 60 month loans from 2015 may not have been fully accounted for. Similarly, 36 month loans from 2017 may not have been fully accounted for. That being said, I estimated the potential expected profit for investors and LendingClub.

To calculate the estimated potential profit, I had to estimate the number of months into the term an individual was at. I did this by subtracting the last payment date from the date the loan was issued. I rounded to the nearest whole month. Then the months remaining could be derived by subtracting the number of payments from the term length. Utilizing the number of payments left in a loan multiplied by the monthly installment gave the estimated balance remaining. Finally, to estimate the total potential profit, I added the already paid amounts to the estimated remaining balance and subtracted the funded amount. The result was $3,907,621,984. This assumes that all remaining balances were paid without charge off.

However, this calculation could be more conservative. The percentage of loans that were not “fully paid” or “current” was about 13%. Again, to be conservative I rounded to 14%. I subtracted 14% from the estimated amount remaining and recalculated the estimated profit to be $3,135,708,931. That works out to be a 41% profit given these conditions. Furthermore, a 20% decrease in the remaining total would result in a 37% profit.

Conclusion

Overwhelmingly, the most frequent purpose for individuals to use LendingClub was for “debt consolidation.” The most common issued grades issued were “A,” “B,” or “C” grade loan. While the proportion of grades within each loan purpose varied “B” or “C” loans held the highest proportion of grades across most loan purposes. With respect to loan statuses, most loans were either paid off or “current.” The “debt consolidation” purpose held the highest proportion of purposes across all loan statuses, including “charged off,” “default,” “current,” and “fully paid.” Finally, I discussed the funding amounts and total payments and estimated various potential profits.

While the LendingClub was founded slightly before an economic recession, it has positioned itself to provide loans to millions of individuals across the country. With loans funded exceeding billions of dollars, consumers have a unique way to acquire personal loans while investors have found a unique way to diversify their portfolio.  

 

 

References

Consumer Financial Protection Bureau. (2020). What is a debt-to-income ratio? Why is the 43% debt-to-income ratio important? Retrieved from https://www.consumerfinance.gov/ask-cfpb/what-is-a-debt-to-income-ratio-why-is-the-43-debt-to-income-ratio-important-en-1791/

Kagan, J. (2019). Charge-off. Retrieved from https://www.investopedia.com/terms/c/chargeoff.asp

LendingClub. (2020). How it works. Retrieved from https://www.lendingclub.com/public/how-peer-lending-works.action 

Kaggle. (2020). Lending Club loan data. Retrieved from https://www.kaggle.com/wendykan/lending-club-loan-data

 

 

 

About Author

Avatar

Tyler Kotnour

Tyler Kotnour graduated with his Master of Public Administration degree in 2018. Upon graduation, Mr. Kotnour worked in consulting conducting research and program evaluation. His primary role involved analyzing public health data for governments and non-profits. A major...
View all posts by Tyler Kotnour >

Leave a Comment

No comments found.

View Posts by Categories


Our Recent Popular Posts


View Posts by Tags

#python #trainwithnycdsa 2019 airbnb Alex Baransky alumni Alumni Interview Alumni Reviews Alumni Spotlight alumni story Alumnus API Application artist aws 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 Bundles California Cancer Research capstone Career Career Day citibike clustering Coding Course Demo Course Report D3.js data Data Analyst data science Data Science Academy Data Science Bootcamp Data science jobs Data Science Reviews Data Scientist Data Scientist Jobs data visualization Deep Learning Demo Day Discount dplyr employer networking feature engineering Finance Financial Data Science Flask gbm Get Hired ggplot2 googleVis Hadoop higgs boson Hiring hiring partner events Hiring Partners Industry Experts Instructor Blog Instructor Interview Job Job Placement Jobs Jon Krohn JP Morgan Chase Kaggle Kickstarter lasso regression Lead Data Scienctist Lead Data Scientist leaflet linear regression Logistic Regression machine learning Maps matplotlib Medical Research Meet the team meetup Networking neural network Neural networks New Courses nlp NYC NYC Data Science nyc data science academy NYC Open Data NYCDSA NYCDSA Alumni Online Online Bootcamp Online Training Open Data painter pandas Part-time Portfolio Development prediction Prework Programming PwC python python machine learning python scrapy python web scraping python webscraping Python Workshop R 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 Selenium sentiment analysis Shiny Shiny Dashboard Spark Special Special Summer Sports statistics streaming Student Interview Student Showcase SVM Switchup Tableau team TensorFlow Testimonial tf-idf Top Data Science Bootcamp twitter visualization web scraping Weekend Course What to expect word cloud word2vec XGBoost yelp