@Anonymous wrote:
@Anonymous wrote:Look at it this way: Income Estimation is basically another "score" like FICO or FAKO that lenders will use as a PART of their decision. If there is not adequate information to make a decision on income, then they will ask for verification, just like they may ask for ID verification on some apps.
Sometimes CCCs want a utility statement and copy of DL for ID. Sometimes they want paystubs or W2 for income.
The "estimation" equation will try to eliminate the need for those docs part or much of the time.
Personally, I don't believe they could come up with a statistically validated method to impute income by using credit limits or even spending patterns, but who knows? I have seen various state governments trying to impute income in child support cases and they seldom hit the mark. I hope CRAs can do better.
As long as there is a mechanism whereby an applicant can supply validated proof of income to overcome errors made by the CRAs, I think it might be ok. Besides, I would think that to comply with the ECOA CCCs will not have a really high threshold for approval at lower credit limits. If they later decide to raise a CL would the law require them to verify an increased income?
Sure they can. It is called data interpolation.
When you apply for credit or loans, you often or usually must disclose an income and possibly debt disclosure. Most prime lenders or those within a given "risk" model will grant credit based upon ranges of acceptable DTI.
Based upon your approval of credit or loans, and the size of credit or loans.... a hypothesis or interpolated estimated may be derived. The more TL's you have (law of large numbers) the more accurate the estimation potential.
Furthermore, it is likely that the estimation equation will keep track of "known characteristics of lenders." Discover will have credit CL ranges different from other lenders.
When you take these data points, then compare with other data elements such as FICO. By comparing the CL ranges and the individual CL's to FICO ranges and individual ranges, and then interpolating with income tables (not your income but ranges of income to known elements of credit) a 'likely income estimation' can be derived.
The estimations will become more accurate over time as more data is gathered and the algorithms are refined based upon measurments of accuracy over time.
This will never be definitive, just as FICO is not definitive. It is just one more tool for lenders to use to both to comply with regs and also to statistically improve their risk lending policies.
