A credit score is a metric used by companies to assess your worthiness to borrow money or your credit risk. Every time you apply for a credit card, car loan, house, or other debt, instrument lenders will request your credit score. These scores are core to the P2P lending business and essentially share a direct correlation to the loan grade in many cases. As the borrower credit score decreases, so will the loan grade. This causes the interest rate to increase proportionally. Of course there are other factors that will influence the rate, but credit score is always the first that comes to mind. While no single credit score calculation exists, there are several that are considered to be industry standards. Lending Club and Prosper both use credit scores in their underwriting process, but they each use different score types. Lending Club uses FICO while Prosper utilizes Experian ScoreX. Information on ScoreX is not publicly available, so I am unable to give a details break down.
FICO was developed by Fair Isaac Corporation, named after its creators. FICO scores use a value weighted approach encompassing five metrics to determine credit risk:
- 35%: Payment history – Late payments on bills, such as a mortgage, credit card or automobile loans.
- 30%: Credit utilization – The ratio of current revolving debt to the total available revolving credit or credit limit.
- 15%: Length of credit history – As a credit history ages it can have a positive impact on its FICO score.
- 10%: Types of credit used (installment, revolving, consumer finance, mortgage)
- 10%: Recent searches for credit – Credit inquiries (our favorite P2P filter), which occur when consumers are seeking new credit. There are two types, soft and hard. Soft do no impact your score, and this is the type that Lending Club and Prosper use for borrowers.
I’d also like to point out that inquires will remain on your FICO for up to 2 years, but only 1 year will be used as a factor in your score.
A person’s FICO score will range from 300 to 850 (500 points total).
In 2006, in an attempt to compete with FICO, the three major credit-reporting agencies introduced VantageScore. VantageScores use a value weighted approach encompassing six metrics to determine credit risk:
- 32%: Payment History – How timely and consistent your payments are, same as with FICO.
- 23%: Credit Utilization – Debt-to-credit ratios and how much credit is available.
- 15%: Credit Balances – What your total debt is; most likely, delinquent debt is counted more harshly than current debt.
- 13%: Length of credit history.
- 10%: Recent searches for credit - How recent and many new hard inquiries and new accounts there are.
- 7%: Available Credit – How much credit can be accessed, total credit line.
A person’s VantageScore score will range from 501 to 990 (489)
Not much is known about this score, since access to it is restricted. Two metrics of interest they boast is the ability predict default and also which accounts will have a higher recovery.
A person’s ScoreX score will range from 300 to 900 (600)
It’s pretty easy to see there is a significant difference in scale and in how the scores are calculated. VantageScore assesses the actual credit limit and balance while FICO is more concerned with this expressed as a percentage. Is one better than the other? It’s debatable. Both view payment history and credit utilization as the most important factor.
If you look at the Lending Club vs Prosper page on Nickel Steamroller, you will see that LC currently has an average FICO of 714.93 while Prosper (2.0 only) has a current average ScoreX of 711.82. There is no direct conversion between the scores.
In a world dominated by FICO scores, it is important to realize what the scores really mean. Prosper is clearly using a different risk model and this must be taken into consideration when comparing.
Peter at www.sociallending.net pointed out that Prosper actually uses ScoreX. Originally I had thought they used Vantage. I confirmed this to be true in their 10K filing: Annual Report on Form 10-K for the fiscal year ended December 31, 2010. Thanks Peter.