Credit Scores Do Not Predict Mortgage Default Risk – Orange County Register

By the 1990s, almost everyone, everywhere in mortgage country, was buying into the power of FICO scores. From policy makers to rating agencies, they have firmly believed in the accuracy of FICO in determining a borrower’s risk of default.

It didn’t matter that the FICO formula was a better kept secret than the recipe for Coca-Cola syrup.

How did that work?

An updated May 20 working paper from the Federal Housing Finance Agency, Fannie Mae and Freddie Mac’s regulator, provided some amazing conclusions.

In the lead up to the 2008 financial crisis, mortgage risk increased for all borrowers, not just those with low FICO credit scores.

Still, subprime borrowers have been burdened with miserable credit features such as prepayment penalties, rising interest rates, payment adjustments, and balloon payments over the past three decades. Meanwhile, mortgage lenders rewarded borrowers with a higher FICO score with preferential rates and fewer points because they believed they were safer bets.

The FHFA paper reviewed 200 million mortgages issued from 1990 to 2019 and estimated a “stressed default rate” for each. That estimate determined what the default risk of any loan would have been if it had been issued at the start of the 2008 financial crisis.

Subprime borrowers, defined as borrowers with a credit score below 660, had a “stressed default rate” that was broadly in line with borrowers with higher credit scores, the study found.

How many of those 200 million borrowers had higher monthly mortgage payments or were charged more points, processing and subscription fees?

Over the past 30 years, how many consumers have been denied home ownership because their scores were low? And let’s not forget those whose credit files are too thin to produce credit scores – especially relevant in underserved and low-income communities.

Before FICO scores – and coupled with Fannie’s automated underwriting engine called Desktop Underwriter and Freddie Mac’s Loan Prospector to determine creditworthiness – we used common sense in the mortgage business.

When it came to credit errors, applicants were asked to explain a paper trail and provide a supporting document. For example, a poor late payment over a tightly limited time frame due to a serious medical matter could be supported by an independent third party – a doctor’s letter.

Underwriters had a certain amount of autonomy to make judgments. Common sense and credible excuses from borrowers are null and void in the credit scoring world. A 1 point difference in your mean FICO score would negate you while all other matters are the same.

Take into account the accuracy of the underlying data used to formulate credit scores. On March 24, the Consumer Financial Protection Bureau presented its 2020 Annual Report to Congress. Credit and consumer complaints accounted for more than 58% of the 542,300 complaints received.

Of the approximately 314,500 credit and consumer complaints, 191,300 concerned fake credit bureaus.

Credit bureaus, on the other hand, stopped reporting tax liens and judgments a long time ago because in our privacy-laden era it is difficult to match the right credit stain to exactly the right person.

Unpaid tax liens and unpaid judgments are a better indicator of a borrower’s character. This is important when it comes to risk assessment.

However, these behaviors are no longer considered between the credit bureaus and the FICO folks.

Fast forward to the CARES Act Forbearance Credit Reporting Kerfuffle. According to a Federal Reserve blog post on the 19th Bank of New York.

The Fed warned that “the concept of creditworthiness, a device for distinguishing good borrowers from bad borrowers, may lose some of its power to signal creditworthiness to lenders, at least for some time.”

Given this potentially myth-shattering FHFA working paper on FICO scores, where do we go from here?

“This is more of a question for lenders than credit bureaus,” said Francis Creighton, president and CEO of the Consumer Data Industry Association. “These are questions about how lenders assess risk based on the data.”

Now that the FHFA knows what they know, they should at least require Fannie and Freddie to finish minimum-qualifying FICO scores and mortgage markups for those with low scores. Affordable home ownership could be America’s asset.

Freddie Mac rated news: The 30-year fixed rate averaged 2.95%, 5 basis points lower than last week. The 15-year fixed interest rate averaged 2.27%, 2 basis points lower than in the previous week.

The Mortgage Bankers Association reported a 4.2% decrease in mortgage application volume from the previous week.

Bottom line: For example, if a borrower receives the average 30-year fixed rate on a compliant loan of $ 548,250, the payment last year was $ 59 higher than this week’s $ 2,297.

What I see: Well-qualified local borrowers can get the following fixed-rate mortgages at 1-point costs: a 30-year FHA at 2.25%, a 15-year conventional at 1.99%, a 30-year conventional at 2.625%, a 15 -Year one-year conventional high balance ($ 548,251 to $ 822,375) at 2.125%, a 30-year conventional high balance at 2.75%, and a 30-year fixed jumbo at 3.25%.

Eyecatcher credit of the week: A 30-year fixed price of 2.875% with no costs.

Jeff Lazerson is a mortgage broker. He can be reached at 949-334-2424 or [email protected] His website is www.mortgagegrader.com.

Comments are closed.