Over-Inflated Credit Scores Leave Consumers / Investors At Risk In A Recession (#GotBitcoin?)
Consumer credit scores have been artificially inflated over the past decade and are masking the real danger the riskiest borrowers pose to hundreds of billions of dollars of debt. Over-Inflated Credit Scores Leave Consumers / Investors At Risk In A Recession (#GotBitcoin?)
That’s the alarm bell being rung by analysts and economists at both Goldman Sachs Group Inc. and Moody’s Analytics, and supported by Federal Reserve research, who say the steady rise of credit scores as the economy expanded over the past decade has led to “grade inflation.”
This means debtors are riskier than their scores indicate because the metrics don’t account for the robust economy, skewing perception of borrowers’ ability to pay bills on time. When a slowdown comes, there could be a much bigger fallout than expected for lenders and investors. There are around 15 million more consumers with credit scores above 740 today than there were in 2006, and about 15 million fewer consumers with scores below 660, according to Moody’s.
“Borrowers with low credit scores in 2019 pose a much higher relative risk,” said Cris deRitis, deputy chief economist at Moody’s Analytics. “Because loss rates today are low and competition for high-score borrowers is fierce, lenders may be tempted to lower their credit standards without appreciating that the 660 credit-score borrower today may be relatively worse than a 660-score borrower in 2009.”
The problem is most acute for smaller, less sophisticated firms that lend to people with poor credit histories, deRitis said. Many of these types of lenders rely mainly on the data supplied by Fair Isaac Corp., the so-called FICO score, and are unable or choose not to include other measures — such as debt-to-income level, economic data or loan terms — into their models for measuring risk, he said.
Car loans, retail credit cards and personal loans handed out online are the most exposed to the inflated scores, according to deRitis. This kind of debt totals around $400 billion, with nearly $100 billion bundled into securities that’s been sold to investors, data compiled by Bloomberg show.
What has analysts concerned is that cracks have already begun to show up in the form of a rising number of missed payments by borrowers with the highest risk, despite a decade of growth. And now with the economy showing signs of weakness, as seen with the recent inversion of the Treasury yield curve, those delinquencies could grow and lead to larger-than-expected losses for investors in riskier asset-backed securities.
“Every credit model that just relies on credit score now — and there’s a lot of them — is possibly understating the risk,” Goldman Sachs analyst Marty Young said in an interview. “There are a whole bunch of other variables, including the business cycle, that need to be taken into account.”
Fair Isaac Corp. created its FICO credit score product in 1989, and it’s still used by more than 90 percent of U.S. lenders to predict whether a would-be borrower is an acceptable risk. Most scores range from 300 to 850, with a higher score purporting to show that someone is more likely to pay back debts. A competitor, VantageScore, was created in 2006 by the three major credit raters Experian, TransUnion and Equifax.
The concern that’s come up, Goldman and Moody’s say, is that lenders haven’t adjusted their underwriting standards as average credit scores have risen during one of the longest economic recoveries on record. So as cracks start to appear in the economy, someone whose credit score rose to 650 from 550 since the Great Recession may pay their bills more like they did 10 years earlier.
“Borrowers’ scores may have migrated up, but inherently their individual risk, and their attitude towards credit and ability to pay their bills, has stayed the same.” deRitis said. “You might have thought 700 was a good score, but now it’s just average.”
Big banks and lenders have been savvy enough to recognize the problem and include many other factors besides credit scores in their underwriting. This is probably true for some of the smaller lenders too.
FICO acknowledges that the credit score alone may not be enough to make informed underwriting decisions, and other factors need to be considered.
“The relationship between FICO score and delinquency levels can and does shift over time,” said Ethan Dornhelm, vice president of scores and predictive analytics at FICO. “We recognize there’s a lot more context you can obtain beyond a consumer’s credit file. We do not think that score inflation is the issue, but the risk layering on underwriting factors outside of credit scores, such as DTI, loan terms, and even trends in macroeconomic cycles, for example.”
But according to Goldman’s Young, the change in scores helps explain why missed payments on auto loans have significantly risen in recent years despite low unemployment, increasing wages and a relatively strong economy.
In February, the Federal Reserve Bank of New York said the number of auto loans at least 90 days late exceeded 7 million at the end of last year, the highest total in the two decades that the data has been tracked. Meanwhile, the subprime segment of auto-loan asset-backed securities has seen 30-day delinquencies rise 81 percent since 2011, driven by looser underwriting due to rising competition between lenders, according to S&P Global Ratings.
Marketplace lending — loans handed out online — has been flashing signs of stress. Missed payments by consumers and writedowns for online loans bundled into bonds increased last year, according to PeerIQ, a New York-based provider of data and analytics for the consumer lending sector.
“We don’t see the purported improvement in underwriting just yet,”’ PeerIQ wrote in a recent report tracking marketplace lending.
Russell-Dowe also avoids the retail credit card sector. So-called private label credit cards — those issued by stores, rather than big banks — saw the highest number of missed payments in seven years in 2018, according to credit bureau Equifax.
She urges investors to do the difficult work necessary to figure out how each lender approaches underwriting and to determine whether they take other factors into consideration besides just scores.
“As an investor it’s incumbent on you to do that deep credit work, which means you have to know as much as possible about how things should pay off or default,” she said. “If you don’t think you’re being paid for the risk, you have no business investing in it.”
The Secret Trust Scores Companies Use to Judge Us All
In the world of online transactions, trust scores are the new credit scores—but good luck finding out yours.
When you’re logging in to a Starbucks account, booking an Airbnb or making a reservation on OpenTable, loads of information about you is crunched instantly into a single score, then evaluated along with other personal data to determine if you’re a malicious bot or potentially risky human.
Often, that’s done by a service called Sift, which is used by startups and established companies alike, including Instacart and LinkedIn, to help guard against credit-card and other forms of fraud. More than 16,000 signals inform the “Sift score,” a rating of 1 to 100, used to flag devices, credit cards and accounts owned by any entities—human or otherwise—that a company might want to block. This score is like a credit score, but for overall trustworthiness, says a company spokeswoman.
One Key Difference: There’s No Way To Find Out Your Sift Score.
Companies that use services like this often mention it in their privacy policies—see Airbnb’s here—but how many of us realize our account behaviors are being shared with companies we’ve never heard of, in the name of security? How much of the information one company shares with these fraud-detection services is used by other clients of that service? And why can’t we access any of this data ourselves, to update, correct or delete it?
According to Sift and competitors such as SecureAuth, which has a similar scoring system, this practice complies with regulations such as the European Union’s General Data Protection Regulation, which mandates that companies don’t store data that can be used to identify real human beings unless they give permission.
Unfortunately GDPR, which went into effect a year ago, has rules that are often vaguely worded, says Lisa Hawke, vice president of security and compliance at the legal tech startup Everlaw. All of this will have to get sorted out in court, she adds.
Another concern for companies using fraud-detection software is just how stringent to be about flagging suspicious behavior. When the algorithms are not zealous enough, they let fraudsters through. And if they’re overzealous, they lock out legitimate customers. Sift and its competitors market themselves as being better and smarter discriminators between “good” and “bad” customers.
Algorithms always have biases, and companies are often unaware of what those might be unless they’ve conducted an audit, something that’s not yet standard practice.
“Sift regularly evaluates the performance of our models and tries to minimize bias and variance in order to maximize accuracy,” says a Sift spokeswoman.
“While we don’t perform audits of our customers’ systems for bias, we enable the organizations that use our platform to have as much visibility as possible into the decision trees, models or data that were used to reach a decision,” says Stephen Cox, vice president and chief security architect at SecureAuth. “In some cases, we may not be fully aware of the means by which our services and products are being used within a customer’s environment,” he adds.
When an account is rejected on the grounds of its Sift score, Patreon sends an automated email directing the applicant to the company’s trust and safety team. “It’s an important way for us to find out if there are any false positives from the Sift score and reinstate the account if it shouldn’t have been flagged as high risk,” says Ms. Hart.
There are many potential tells that a transaction is fishy. “The amazing thing to me is when someone fails to log in effectively, you know it’s a real person,” says Ms. Hart. The bots log in perfectly every time. Email addresses with a lot of numbers at the end and brand new accounts are also more likely to be fraudulent, as are logins coming from anonymity networks such as Tor.
These services also learn from every transaction across their entire system, and compare data from multiple clients. For instance, if an account or mobile device has been associated with fraud at, say, Instacart, that could mark it as risky for another company, say Wayfair—even if the credit card being used seems legitimate, says a Sift spokeswoman.
The risk score for any given customer, bot or hacker is constantly changing based on that user’s behavior, going up and down depending on their actions and any new information Sift gathers about them, she adds.
For Our Protection?
These trustworthiness scores make us unwitting parties to the central tension between privacy and security at the heart of Big Tech.
Sift judges whether or not you can be trusted, yet there’s no file with your name that it can produce upon request. That’s because it doesn’t need your name to analyze your behavior.
“Our customers will send us events like ‘account created,’ ‘profile photo uploaded,’ ‘someone sent a message,’ ‘review written,’ ‘an item was added to shopping cart,’” says Sift chief executive Jason Tan.
It’s technically possible to make user data difficult or impossible to link to a real person. Apple and others say they take steps to prevent such “de-anonymizing.” Sift doesn’t use those techniques. And an individual’s name can be among the characteristics its customers share with it in order to determine the riskiness of a transaction.
In the gap between who is taking responsibility for user data—Sift or its clients—there appears to be ample room for the kind of slip-ups that could run afoul of privacy laws. Without an audit of such a system it’s impossible to know. Companies live under increasing threat of prosecution, but as just-released research on biases in Facebook ’s advertising algorithm suggest, even the most sophisticated operators don’t seem to be fully aware of how their systems are behaving.
That said, sharing data about potential bad actors is essential to many security systems. “I would argue that in our desire to protect privacy, we have to be careful, because are we going to make it impossible for the good guys to perform the necessary function of security?” says Anshu Sharma, co-founder of Clearedin, a startup that helps companies combat email phishing attacks.
The solution, he says, should be transparency. When a company rejects us as potential customers, it should explain why, even if it pulls back the curtain a little on how its security systems identified us as risky in the first place.
Mr. Cox says it’s up to SecureAuth’s clients, which include Starbucks and Xerox, to decide how to notify people who were flagged, and a spokeswoman said the same is true for Sift.
Companies use these scores to figure out who—people or potential bots—to subject to additional screening, such as a request to upload a form of ID.
Someone on a travel service buying tickets for other people might be a scammer, for instance. Or they might be a wealthy frequent flyer.
“Sometimes your best customers and your worst customers look the same,” says Jacqueline Hart, head of trust and safety at Patreon, a service for supporting artists and creators, which uses Sift to screen transactions on its site. “You can have someone come in and say I want to pledge $10,000 and they’re either a fraudster or an amazing patron of the arts,” she adds.
Rewrite of Lower-Income Lending Rules to Advance In December
Comptroller of the Currency Joseph Otting to move forward alone if necessary after OCC talks end with Fed.
A top bank regulator is poised to propose changes to bank lending requirements that could potentially transform the way lenders make billions of dollars in loans, investments and donations to customers in lower-income areas.
The proposal from the Office of the Comptroller of the Currency to regulations of the Community Reinvestment Act, which requires banks to serve borrowers of all income levels who reside near their branches, could also make it easier for banks to meet certain lending requirements, particularly in poorer neighborhoods. The agency is planning to release its overhaul sometime in December.
Banks would be encouraged to make loans to lower-income borrowers based on the geographic concentrations of their deposits, in addition to the locations of their physical branches, Comptroller of the Currency Joseph Otting said in an interview.
“We think our approach will generate sizable more dollars into communities across America,” Mr. Otting said.
Under discussion since the earliest days of the Trump administration, a formal proposal has been repeatedly pushed back as the OCC sought to coordinate with two other banking agencies on the changes.
It isn’t yet clear if one of the other agencies—the Federal Deposit Insurance Corp.—will jointly propose the changes with the OCC, though Mr. Otting said he is “very optimistic” it will. The third regulator, the Federal Reserve, isn’t expected to be a part of the OCC’s overhaul after the two regulators disagreed on how to refashion the rules, Mr. Otting said.
The Fed was working with the OCC and FDIC until last week, but it isn’t expected to back the OCC’s proposal.
That split could generate industry pushback if the rules aren’t aligned across the three agencies by the time they are completed.
“At this point in time, we’ve just got to move forward,” Mr. Otting said.
The OCC oversees approximately 70% of community lending activity and estimates banks made $482 billion in such loans in 2017, amounting to 4.1% of deposits. Its proposal would apply to some 1,200 banks, including some of the biggest firms, such as JPMorgan Chase & Co. and Wells Fargo & Co.
The breakdown with the Fed revolved around whether to emphasize the number of CRA loans—units, in bank-speak—or the dollar amount of CRA activity. Mr. Otting said his proposal would account for both. The Fed, he said, was more focused on units.
“At this point the Fed is out,” he said. “We just fundamentally can’t get to the same spot” for the proposal.
“We agree that it is appropriate to update and modernize the regulations implementing the CRA. We worked diligently for months with the FDIC and OCC to agree on a common approach,” the Fed said in a statement. “It is unfortunate that these efforts have so far not been successful and that the agencies were unable to reach agreement on metrics that would be tailored to bank size and business model and reflect the different credit needs of the local communities that are at the heart of the statute,” it added.
The Fed said it continues to believe that the best outcome would be a joint proposal.
An FDIC spokesman declined to comment.
First enacted in 1977, the CRA aimed to combat redlining—a practice of not lending in certain minority neighborhoods. At a time when many banking services are provided online, there is growing consensus among policy makers that the rules should be updated.
Mr. Otting, a former bank executive who has made overhauling the regulations a priority, once faced public pushback from community groups in the midst of a merger between his former firm OneWest Bank and CIT Group Inc.
In recent years, the law has become a source of conflict between community groups that want the rules to be enforced more strongly and bankers who argue the regulations are too bureaucratic and haven’t kept up with technological changes, among other criticisms. Banks are typically examined every three years on their CRA efforts. A bad grade effectively prohibits mergers.
The OCC proposal will be subject to a public comment before it can be completed.
Mr. Otting said his proposal will aim to set clear standards and an illustrative list of the types of activities that qualify for CRA consideration. “People don’t have a list of what qualifies” under current rules, he said.
The changes would redefine where CRA activity counts by updating bank assessment areas, which at present are mapped around the geographic proximity to their branches. Banks would still be required to lend in lower-income neighborhoods near their branches, but online lenders and other banks with deposits outside of its branch networks would be examined based on areas where they have significant deposits in addition to their branch neighborhoods.
Mr. Otting said his agency’s proposal would aim to close loopholes in the existing rules. For instance, banks would have to ensure loans aren’t only made in areas defined as poorer neighborhoods, but that they go to lower-income borrowers, as well.
Similarly, banks’ assessments would revolve around the amount of time their CRA loans remain on their books. A bank that buys a CRA loan to meet its obligations under the law and then quickly resells it to another bank wouldn’t get as much credit as a lender that originated a loan and kept it on its balance sheet.
Bank Regulators Propose Overhaul of Lending Rules For Poorer Communities.
Overhaul is priority for comptroller of the currency, but it isn’t backed by Federal Reserve.
Federal banking regulators on Thursday proposed an overhaul of rules governing how banks lend hundreds of billions of dollars in low-income neighborhoods, setting up a potential break with the Federal Reserve.
The overhaul is a priority for Comptroller of the Currency Joseph Otting, who says it will boost lending under the Community Reinvestment Act and make existing requirements more transparent and consistent. The act requires banks to serve borrowers of all income levels in their communities.
The plan is backed by a second regulator, the Federal Deposit Insurance Corp., but not by the Fed, which is considering a separate overhaul, according to people familiar with the central bank’s thinking. Community groups have also objected to the plan.
Officials at all three regulators say they hope they can ultimately agree on a plan. “We worked very hard to try to get aligned with the OCC on a proposal, and my hope is that we can still do that,” Fed Chairman Jerome Powell said Wednesday. “I don’t know whether that will be possible or not.”
Friction over the proposal between the Fed and OCC has been palpable behind the scenes. In November, Mr. Otting sent an email to Fed governor Lael Brainard, who is leading Fed efforts on the issue, saying the central bank was slow-walking the overhaul, according to people familiar with the message. Fed officials objected to that characterization and said the OCC was moving forward with a plan that lacks analytical rigor, the people said.
“Any modernization of the Community Reinvestment Act must further the goal at the heart of the statute—encouraging banks to meet the credit needs of local low- and moderate-income communities,” the Fed said in a statement on Thursday afternoon, after the overhaul was announced. “No decisions have been made about how the Federal Reserve will proceed.”
Advocacy groups and some Democrats in Congress say the plan could inadvertently restrict lending in low-income areas. And some bank lobbyists have privately expressed concern it will increase costs of compliance with the act.
Mr. Otting dismissed those concerns. “If you don’t like this, you are either economically advantaged by the current structure or you don’t understand it,” he said in an interview this week. In discussions with major banks, “nine out of 10 are supportive of the direction we’re heading.”
In 2014, community groups opposed the acquisition of OneWest Bank by CIT Group Inc. Mr. Otting was chief executive of OneWest at the time, and Steven Mnuchin, now Treasury secretary, was chairman. OneWest was accused of abusing homeowners during the foreclosure process, a charge the bank disputed. The merger went through, but Mr. Otting has since made an overhaul of the CRA a priority.
In an acknowledgment that the proposed framework could be costly to banks, the OCC and FDIC said lenders with less than $500 million in assets could opt out. That threshold would cover about three-quarters of banks overseen by the FDIC, though the firms generate a fraction of the roughly $480 billion in annual CRA lending and investment.
The CRA was passed in 1977 to combat redlining, a practice where banks wouldn’t lend in lower-income communities. Regulators and Congress have turned the act into an extensive public test evaluating how many loans, branches and investments a bank has to serve the poor.
Evaluations are based on a complex formula that includes loans to home buyers and small businesses and the number of branches in lower-income areas. Bad grades can result in restrictions on bank mergers and other activities.
The proposed overhaul would more precisely define the types of lending and other activities that qualify under the act and redefine the geographic areas where they occur, known as assessment areas. If successful, it would allow regulators to see more easily if lending matches up with deposits.
Banks would still be required to lend in lower-income neighborhoods near their branches, but under the new rules they would also be examined based on areas where they have significant deposits.
The OCC and FDIC proposal will be subject to public comment before it can be completed as early as next year. Banks overseen by the OCC conduct about 70% of activities under the low-income lending rules. Those overseen by the FDIC and the Fed each account for about half of the remaining business.
Credit Scores Rise Amid Recession
The average credit score hit a record in July after millions of Americans had lost their jobs, scrambling lenders’ underwriting models.
Millions of Americans lost their jobs and skipped debt payments this year. You wouldn’t know it looking at consumer credit scores.
While the coronavirus was pummeling the U.S. economy, Americans’ credit scores—a metric used in nearly every consumer-lending decision—were rising. The average FICO credit score stood at 711 in July, up from 708 in April and 706 a year earlier, according to Fair Isaac Corp. FICO 0.09% , the score’s creator. Early estimates suggest the average score has held steady through mid-October at the July level, which is the highest since FICO began keeping track in 2005.
The increase is largely thanks to the unprecedented financial assistance the government and lenders rolled out to consumers after the pandemic took hold in the U.S. Stimulus payments and expanded unemployment benefits helped many borrowers keep up with their bills and, in some cases, even pay down their debt. Widespread payment holidays on mortgages, auto loans and student loans freed up funds and kept credit reports clean.
American consumers’ ability to withstand such a severe economic shock is undoubtedly good news—an outcome that few would have predicted in the pandemic’s early days. But for lenders, the rise in credit scores is yet another confounding factor that is making it difficult to assess risk.
During the last downturn, loan delinquencies rose along with unemployment, and credit reports reflected missed payments in short order. That hasn’t happened this time, yet millions of Americans are still out of work and surviving on unemployment benefits. The disconnect has scrambled lenders’ underwriting models and sent them in search of new ways to evaluate applicants’ creditworthiness.
One big fear is that consumers’ credit quality could begin to sour if Congress doesn’t reach a deal on additional aid for the unemployed. “We’re afraid that in a couple months there could be real damage to credit reports,” said Francis Creighton, chief executive of the Consumer Data Industry Association, which represents credit-reporting firms.
FICO scores, which range from 300 to 850, are calculated using the information in consumers’ credit reports, including the ratio of credit-card debt owed to total spending limit, payment history and prior loan applications. They don’t take into account employment history or income.
Ethan Dornhelm, vice president of scores and predictive analytics at FICO, said the scores are typically lagging indicators. In the last downturn, credit scores bottomed out at 686 in October 2009—months after the recession officially ended.
“First the macro stress occurs, and then it takes a few months for the strain to show up in people’s credit reports,” he said. Deferment programs and government stimulus “are having a further effect of pushing out that stress for many people.”
‘We are convinced that new ways of assessing credit eligibility, like reviewing personal cash-flow data, offer a truer reflection of risk…’
— Steve Smith, Finicity CEO
What’s more, many borrowers’ credit profiles appear to have improved in recent months. A decrease in credit-card spending helped lower total outstanding card debt. Government stimulus and lender deferment programs helped borrowers stay current on their debts.
Dee Donnell’s credit score was in the low 500s—deep in subprime territory—in February, when she lost her job at a health-insurance company. The 45-year-old used part of her severance to pay off roughly $10,000 of credit-card debt, which she said was costing her $600 in minimum monthly payments. She was able to cover her other bills with her remaining severance, savings and her $1,200 stimulus payment.
In July, Ms. Donnell got a job working in compliance for a startup. Her credit score is now nearly 700.
“Covid forced me to really look at my finances,” she said.
Lenders have been tinkering with their underwriting models since the pandemic began, according to senior executives at large banks, in an effort to avoid approving loan applicants who are unemployed and on the verge of running out of government assistance. Some are also looking to identify existing customers at higher risk of default.
To spot these risks, big banks including JPMorgan Chase & Co., Bank of America Corp. , Wells Fargo & Co. and Citigroup Inc. are seeking to augment credit reports and scores with real-time income or cash-flow data, according to people familiar with the matter.
That involves reviewing their own bank-account data when evaluating certain applicants. Some are weighing using information from other financial institutions. (In most cases, they would need an applicant’s permission to use that bank-account data in the underwriting process.)
Lenders are interested in the data to evaluate new loan applicants, specifically, those seeking credit cards and other loans that don’t typically require income documentation. Some are also discussing using cash-flow analytics—for example, a lag in deposits indicating a recent layoff—to determine whether to cut existing borrowers’ credit lines.
Finicity, a financial-technology company, has been in talks with banks about providing some of this data. “We are convinced that new ways of assessing credit eligibility, like reviewing personal cash-flow data, offer a truer reflection of risk, and many banks are quickly coming around to that idea,” said Chief Executive Steve Smith.
Dormant credit-card accounts are seen as particularly risky these days. Lenders are worried customers will reach for them when unemployment runs out, so they are trimming credit lines and closing some accounts altogether. Even customers with credit scores in the mid-700s—generally seen as more creditworthy and eligible for lower interest rates—haven’t been spared.
Despite lenders’ concerns, there is evidence that Americans are giving priority to debt payments in the pandemic. In a June survey of about 1,300 households, the Federal Reserve Bank of New York found that those who got stimulus payments used 35% of the funds to pay down debt.
Ben Rohrs, 42, was laid off from his job as a technology product manager in March. He and his wife paused their roughly $5,000 monthly mortgage payment for six months, and used the freed-up funds—plus unemployment and his wife’s income—to keep up with their credit-card bills.
Mr. Rohrs started a new job last week. His credit score is close to 830, around where it was in March.
“I just feel really lucky that there was the increased unemployment insurance, and having a mortgage deferral was huge,” he said.
Your Questions And Comments Are Greatly Appreciated.
Monty H. & Carolyn A.Go back