Will dropping tax lien data from credit reporting lead to bad loans?
In just over two weeks, the three major credit bureaus will make a significant change, deleting the last remaining scraps of tax lien data that exists in consumer reports, an estimated 5.5 million records.
But a question hangs over the April 16 deadline — what will happen to credit scores as a result?
While a definitive answer isn't clear, some argue that the existence of a tax lien says a lot about a person's ability to repay debt. They worry that eliminating that information completely from credit reports could blind lenders to serious risk because it will raise credit scores.
“We feel the data, when accurately linked to the consumer, is important,” said Ankush Tewari, senior director of credit risk assessment at LexisNexis Risk Solutions, a provider of public record data. “It is important for lenders to know, for example, if a consumer has a $50,000 tax lien that they have an obligation to repay before issuing them new credit. So if you’re an auto lender and you’re considering giving this consumer a loan for a $35,000 automobile, it’s important for the lender to be aware that the consumer also has this other lien obligation that they’re also paying off.”
The change is part of a series of steps taken by the credit bureaus, which last July eliminated civil judgment records — notes that a consumer owes a debt to a court as a result of a lawsuit — from credit reports, as well as half the tax lien data they had.
That change and the next step of purging the remaining tax lien data are a result of settlement agreements between the bureaus and 31 state attorneys general, which said that as of July 1, 2017, public record data given to the credit bureaus had to contain name, address, and Social Security number and/or date of birth, and had to be refreshed at least every 90 days. At issue were concerns that the credit bureaus had trouble linking public record data to credit reports.
“A lot of judgments and liens were linked to the wrong people, so someone may share your first and last name, maybe live in a different part of the country, and they might have a lien or judgment that might get linked to your file," Tewari said.
Equifax, TransUnion and Experian said last summer that 96% of civil judgments and 50% of tax liens didn’t have enough identifying information, so they dropped them from credit reports.
In a statement, TransUnion noted that "since then, our ongoing review and monitoring of tax lien data transmitted by our third party vendor has led us to decide to cease reporting tax lien data and to remove all remaining tax liens from our consumer credit reporting database.”
Experian and Equifax declined to comment for this article.
After July 16, the only public records that will remain in consumers' files will be bankruptcies, which apparently do come with sufficient publicly indentifiable information.
It’s hard to say with certainty how deleting these records will affect consumers’ credit scores and their ability to get credit. Tewari, for one, is certain that once the public record data has been scrubbed, people who may not have otherwise been issued a loan because they had liens or judgments on their file have now been issued loans.
But there won't be a definitive answer for some time, as it typically takes 18-24 months to see any uptick in default rates.
Some outside sources predict little impact, however.The Consumer Financial Protection Bureau studied the matter by analyzing credit scores in June, just before the majority of public data was eliminated, and then in July, August and September; it released a report in February. The agency concluded that “the small number of consumers who had civil judgments or tax liens and experienced a score change large enough to improve their credit profile suggests that any effects on overall model predictiveness (either positive or negative) are likely minimal.”
The CFPB study found that before all this started, in June 2017, 6% of consumers had a judgment or tax lien on their credit report. The study also found that such people tend to be subprime. About 71% of consumers with a civil judgment or tax lien had credit scores below 620.
“Removing tax liens from credit files will have very little impact on credit scores for people that have good credit,” said Steve Ely, CEO of eCredable. “But if you are struggling with credit and have a week credit file, this will help you. The consumer who has a tax lien is now automatically having a negative entry removed from their credit file.”
To lenders that base lending decisions primarily on credit scores, consumers with tax liens will look more creditworthy than they are.
“This is not good if you’re a bank, a credit card issuer or an auto lender,” Ely said.
But because this affects subprime consumers far more than prime consumers, only subprime lenders will be hurt, he said.
“This is not going to hurt Bank of America or Citi or Chase,” Ely said.
Another factor is that consumers who had tax liens or civil judgments in their credit reports also tended to have other signals of credit riskiness: about 89% of consumers with judgments or liens had a delinquency of 90 days or longer on their credit record, compared with 34% of consumers without those black marks.
In its statement, TransUnion noted that some consumer credit scores may change after the removal of public data records from credit reports. “However," it said, "analysis by leading scoring model companies last year showed the change to be minimal.”
This was a reference to studies FICO and VantageScore conducted last year that indicated their credit score models would be little affected by the removal of public records. Both companies have an incentive to maintain the status quo.
The accuracy problem
In the bigger picture, accusations of inaccuracy have plagued the credit reporting industry for years. Horror stories abound. The consumersunion.org site shares a story of a consumer named James from Fort Walton Beach, Fla., who said his credit score dropped from 780 to 620 when a hospital turned over to a collection agency a medical bill that belonged to someone with the same last name and birthday month (but not the same day or year).
“No matter what I did, from contacting the credit bureau, contacting the collection agency, contacting the billing department of the hospital, nothing could get the obviously incorrect charge removed from my credit report,” he said. “The worse thing was that at the time, I was applying to buy a home. Needless to say, the lender did not want to lend money because of the drop in score.”
In 2012, the Federal Trade Commission conducted a study that found that 26% of consumers had at least one potentially material error in their credit reports, and that 13% experienced a change in their credit score as a result of modifications to their credit report after a dispute.
With an estimated 200 million Americans having credit reports, that would suggest around 52 million consumers have potentially material errors on their credit reports. However, the FTC study found that only 2.2% of the credit reports reviewed had errors that when corrected resulted in a material change in credit score.
Some fintechs cite credit report inaccuracy as a reason to create “FICO-Free Zones” and focus on other means of analyzing creditworthiness, such as cash flow and ability to pay bills.
However, at banks and traditional lending companies, credit reports and credit scores are deeply embedded in the machinery used to lend money and grant credit cards. The cost and effort of upending this would be enormous. And the existing framework is sanctioned by regulators, who have cast a suspicious eye on the use of so-called alternative data.
The CFPB began a study of the use of alternative data last year. It have given one online lender, Upstart, a “no-action letter” in which Upstart can continue to use alternative data such as education information in lending and sends the agency reports on the loans it makes and declines.
The credit reporting industry is not going away anytime soon. What banks have to hope is that it continues to work on accuracy, security, privacy and other issues.
Editor at Large Penny Crosman welcomes feedback at firstname.lastname@example.org.