Why Americans Are Falling Behind on Auto Loans At Their Highest Level Ever

Episode Summary

Executive Summary: Odd Lots examines whether the U.S. consumer is truly weakening or just becoming more uneven. Guest Ricard Bondebo of VantageScore argues the economy is increasingly K-shaped, driven less by income alone than by wealth, homeownership, and balance sheets. He says delinquencies are rising in autos, mortgages, and student loans, but the stress is nuanced and not yet a collapse.

Main Topics: Measuring consumer stress in a K-shaped economy (Priority: 5/5): The hosts open with uncertainty about the state of the consumer: spending remains strong in places, sentiment is weak, and aggregate numbers hide diverging realities across income and wealth groups. What VantageScore does and why scoring matters (Priority: 5/5): Bondebo explains VantageScore’s role as a competing credit-scoring model created by the three bureaus to improve competition, prediction, and credit access beyond the legacy FICO framework. How credit scores are built and why they are relative (Priority: 5/5): The discussion focuses on credit scoring as a probabilistic, relative measure that must reflect the period in which it is pulled and the consumer segment being evaluated. Wealth, not just income, drives credit outcomes (Priority: 5/5): Bondebo argues that homeownership, stock ownership, and other assets explain more of the current divergence than income alone, especially among higher-income households. Auto loans as a leading stress point (Priority: 4/5): Auto lending has become the riskiest consumer credit category due to higher car prices, higher interest rates, insurance, and repair costs, despite tighter underwriting. Student loans and insurance as additional pressure points (Priority: 4/5): The resumption of student loan payments and rising auto/home insurance costs are adding incremental strain and may push more households into delinquency next year. Mortgage model changes and expanding access (Priority: 4/5): A rule change now allows newer credit models like VantageScore 4 in mortgage underwriting, potentially broadening access, improving prediction, and reducing systemic risk.

Key Arguments: Consumer weakness is real but uneven; wealthy households can offset weaker spending among lower-income households, making the aggregate picture hard to read. Credit scores are relative measures of risk, not absolute ones; the same score means different things in different macro environments. Income and wealth are not the same; homeownership is a major cushion that helps explain why some higher-income households weather stress better than others. Pandemic-era stimulus and forbearance temporarily improved consumer credit profiles, but inflation and rising costs later reversed that improvement. Higher-income borrowers began showing rising delinquencies after lower-income borrowers because reserves delayed the pain, not because they were immune. Auto loans deteriorated because vehicle prices, loan balances, rates, insurance, and maintenance all rose together, making the total ownership cost much higher. Student loan delinquency rates surged when reporting resumed after forbearance, showing a delayed but significant credit reset. Insurance increases can be the final shock that turns manageable budgets into delinquency, especially for car- and home-heavy households. Current trends do not suggest an imminent credit collapse, but they do point to a more fragile consumer that could be vulnerable to a new shock.

Data Points: VantageScore reach: 33 million more people scored - Bondebo says newer models and trended data allow VantageScore to score tens of millions more consumers than legacy approaches. Mortgage model update: Version 4 in use; Version 5 releasing this year - He describes VantageScore as rewriting the model over time to reflect current data and behavior. Historic gap in scoring: No score if no credit activity in past 6 months - A limitation of older scoring approaches that excludes many consumers, including deployed military members. Credit-history requirement: At least 6 months of history - Older models often fail to score people new to credit or with recent activity gaps. Population excluded by legacy score: About 20% of the U.S. population - Bondebo says the classic FICO score still does not score roughly one-fifth of Americans. High-income threshold used in discussion: $150,000+ household income - VantageScore’s internal analysis of a higher-income cohort in delinquency trends. Mortgage score cutoff (historical): 620 - Previously, conforming mortgage access was tied to a minimum FICO cutoff. High-income delinquency trend: Highest year-over-year increases last year; down in 3 of 4 months since July - Bondebo says delinquency momentum for higher-income households has recently improved. Low-income delinquency trend: Around 8% year-over-year increases - He says lower-income households continue to show persistently high delinquency growth. Student-loan delinquency before COVID: Around 9% to 11% - Historical average student loan delinquency rates before the pandemic forbearance period. Student-loan delinquency after reporting resumed: Over 20% - When student loans returned to credit files in 2025, delinquency rates were more than double the historical norm. Student-loan delinquency later in year: About 17% to 17.5% - He says some borrowers have since cured, but many are still under pressure. Auto delinquencies: At or near record highs - Bondebo says auto delinquencies continue rising and are now the riskiest consumer credit category. Mortgage delinquency history: Still much lower than 2010 or 2008-09 levels - Mortgages are rising but remain comparatively well behaved. Consumer cash balances: Checking-account cash is coming down - He cites JPMorgan data suggesting households have less liquidity to absorb shocks.

Pivotal Quotes: "AI shouldn't eliminate them, it should elevate them." — Advertisement / Palantir: A pre-interview promotional spot framing AI as worker-enhancing rather than replacement-focused. "Credit score is not an absolute measure of risk. It's a relative measure of risk." — Ricard Bondebo: He explains why the same score means different things across different economic periods. "The consumer is much more fragile than they used to be." — Joe Weisenthal: Closing reflection on the cumulative stress from student loans, auto costs, insurance, and uneven wealth.

Implications: Consumers are not in collapse, but many are closer to the edge than aggregates suggest. Lenders should rely on richer, more current data; listeners should expect continued pressure from autos, student loans, and insurance even if a broad recession has not yet arrived.