The Power of Data & The Next Generation Credit Model
LendingClub is built on data. We use it in every aspect of our business – from operations, to credit, to the way we hire employees – and as the basis for continued innovation on behalf of investors and borrowers alike.
Starting September 8, 2017, we are rolling out the fifth-generation credit model, which further leverages machine learning and 10 years of LendingClub data to better assess and price credit risk. The model went live for certain borrowers on September 8 and was ramped up to the vast majority of borrowers a few days after. (A small borrower segment will continue to be evaluated under the prior credit model and we are working to roll it out for those borrowers next year.) LendingClub has helped 1.5+ million borrowers since 2007; each borrower provides unique insight that can be leveraged in future credit decisions.
The new credit model is the latest in LendingClub’s long history of innovation: it’s a more powerful, complex, multi-variable model that takes advantage of a wealth of proprietary data and best-in-class modeling techniques. It’s more predictive than a borrower’s FICO score alone, and is 24% better at differentiating the likelihood of a borrower charging off than the fourth-generation model.
Going Beyond FICO
The fifth-generation credit model continues a long tradition of innovating beyond traditional FICO-based underwriting.
It’s why we are able to improve financial inclusion and increase access to lower interest rates for borrowers. Through powerful models we are better able to assess risk: we can facilitate loans to people who traditional lending programs miscategorize as too risky but are actually good borrowers, as well as screen out borrowers who seem safe but we believe are risky based on our use of rich data. A recent study done by researchers from the Federal Reserve emphasized these results and found that LendingClub “has enhanced financial inclusion and allowed some borrowers to be assigned better loan ratings and receive lower priced credit.”
Highlights of the New Model
Since our founding in 2007, we have used data to our advantage. The latest credit model is built on the most extensive data set ever and validated on recent data, which allows the model to better reflect the current environment. It also leverages the longest track record in the industry – LendingClub has rich data based on the last 10 years of underwriting.
The fifth-generation model specifically distinguishes itself from earlier models using more unique data and machine learning techniques:
- The model incorporates more data points per borrower than ever before.
- It also uses dozens of new customized attributes that are predictive in assessing risk. For example, instead of using aggregates, the new model uses very granular views of credit data which discern individual borrower actions vs. a simple aggregate (e.g. a borrower’s credit card balance per credit card vs. his total credit card balance). As another example, the model makes more extensive use of trended data, which provides insight into a borrower’s credit behavior over time rather than a snapshot into a borrower’s credit behavior at a point in time. Dozens of new custom variables like these improve the model’s predictive power and are proprietary to LendingClub.
- The model further distinguishes itself by leveraging new machine learning techniques to analyze the massive data sets that come from looking at so many data points on each borrower.
Impact to Investor Returns
The new model is 24% better at differentiating the likelihood of a borrower charging off than the fourth-generation model. This means the model better assesses and prices an individual borrower’s risk in addition to the risk of the total population.
In total, we expect that investors will see more lower-risk borrowers on the platform. Some higher-risk borrowers who may have received an offer previously will be screened out, while other borrowers will qualify for lower prices. The end result is that loan volume will shift toward higher quality grades (grades A and B) overall. Projected investor returns are mostly unchanged by grade and term, ranging from 4-9%. Expected charge-off rates for the platform as a whole are going down slightly (from 6.2% to 6.0%) as we expect to see more lower-risk borrowers on the platform as a whole.1
See the table below for projected returns by grade and term.
We see the fifth-generation credit model as the latest innovation in our continuous enhancement cycle, and one that helps us continue to serve borrowers and provide investors with solid risk-adjusted returns.
Safe Harbor Statement
Some of the statements above are “forward-looking statements.” The words “anticipate,” “believe,” “estimate,” “expect,” “intend,” “may,” “outlook,” “plan,” “predict,” “project,” “will,” “would” and similar expressions may identify forward-looking statements, although not all forward-looking statements contain these identifying words. Actual performance of the overall platform has differed from projected performance in the past, and could differ in the future. Factors that could cause actual results to differ materially from those contemplated by these forward statements include: increases to unemployment rates, particularly if such increases are concentrated in populations with a greater propensity to take loans facilitated by our platform; changes to consumer credit behaviors; stagnation or reduction in the growth of the nation’s gross domestic product or uncertainties created by political changes associated with a change in presidential administrations, the Company’s ability to continue to attract and retain new and existing retail and institutional investors; competition; and demand for the types of loans facilitated by the Company and those factors set forth in the section titled “Risk Factors” in the Annual Report on Form 10-K, filed with the SEC. LendingClub may not actually achieve the plans, intentions or expectations disclosed in forward-looking statements, and you should not place undue reliance on forward-looking statements. Actual results or events could differ materially from the plans, intentions and expectations disclosed in forward-looking statements. LendingClub does not assume any obligation to update any forward-looking statements, whether as a result of new information, future events or otherwise, except as required by law.
 “Projected Annualized Net Credit Loss (w/ Prepayment)” also known as Expected Charge-Off Rate, is LendingClub’s projection of the aggregate dollar amount of loan principal charged-off, net of any amounts recovered and accounting for the impact of amounts prepaid, as an annualized percentage of the aggregate dollar amount of loan principal for all loans issued under the Prime Program after September 8, 2017. Projected Annualized Net Credit Loss (w/ Prepayment) is not a promise of future results and may not accurately reflect actual charge-off or prepayment rates. Actual charge-off and prepayment rates experienced by any individual portfolio may be impacted by, among other things, the size and diversity of the portfolio, the exposure to any single loan, borrower or group of loans or borrowers, as well as macroeconomic conditions.
“Average Interest Rate” is the hypothetical loan grade and term mix used to calculate weighted average interest rate was generated by applying the fifth generation credit model to certain applications received from February through May 2017. This hypothetical mix is for illustrative purposes only and is not a guarantee or indication of future inventory distribution. This analysis uses elements of backtesting. Backtesting is hypothetical and is provided for information purposes only. Backtested data has inherent limitations, including that historical borrower populations are not necessarily indicative of future borrower populations.
 “Projected Return” is a measure of the estimated annualized return rate on invested principal (meaning for all funds then invested in Notes or loans) using an internal rate of return (IRR) methodology using a monthly term. Monthly cash flow projections are calculated as follows: the scheduled principal and interest payments based on the Interest Rate, minus the amount of such principal and interest payments lost due to the Expected Charge-Off Rate, minus Expected Fees. Monthly IRR figures are annualized by multiplying the monthly IRR figure by 12. Projected Returns are calculated based on grade and maturity mix for issued loans for the period of February through May 2017. Projected Return is not a promise of future results and may not accurately reflect actual returns. Actual returns experienced by any individual portfolio may be impacted by, among other things, the size and diversity of the portfolio, the exposure to any single Note or loan, borrower or group of Notes, loans or borrowers, as well as macroeconomic conditions. Individual results may vary and projections are subject to change. The information presented is not intended to be investment advice, guidance, or a guarantee of the performance of any Note or loan. Notes are offered by prospectus filed with the SEC and investors should review the risks and uncertainties described in the prospectus prior to investing. Actual results may vary.
“Interest Rate” is equal to the weighted average stated borrower interest rate for the loan grade or mix of loan grades (whichever is applicable) of the grade and maturity mix described in the “Average Interest Rate” disclaimer above.
“Expected Charge-Off Rate” is defined above as “Projected Annualized Net Credit Loss (w/Prepayment).”
“Expected Fees” for loan purchasers means the aggregate estimated impact of LendingClub’s servicing fee (1%), collection fee (18%), recovery fee (18%), and an administrative fee (0.10%).
“Expected Fees” for Note investors means the estimated impact of all applicable fees as well as the impact of interest not earned during the administrative holding period in the first month (2 business days). Applicable fees are LendingClub’s service fee and collections fee (if applicable). LendingClub charges an investor service fee of 1% of the amount of payments received within 15 days of the payment due date. The service fee is not an annual fee and may therefore reduce annual investor returns by more or less than 1%. We estimate the collection fee based on expected charge-off rates and the expected number of late payments that will be collected on past due loans with a given grade and term. For more detail on LendingClub fees for Note investors, please click here. Individual results may vary and projections can change. Past performance is no guarantee of future results.