Where Are Mortgage Delinquencies Rising the Most?

The New York Fed reports aggregate household debt rose $191 billion in Q4 2025 to $18.8 trillion (up $4.6 trillion since end-2019), with mortgage balances up $98 billion to $13.2 trillion and credit card debt up $44 billion to $1.28 trillion. Mortgage 90+ day delinquencies increased in Q4, driven disproportionately by borrowers in lowest-income zip codes (from ~0.5% in 2021 to nearly 3.0% by late 2025); nationally about 1.3% of mortgage balances became seriously delinquent in 2025. Worsening local labor markets (counties with largest unemployment increases saw ~0.6pp rise in delinquency flows vs ~0.2pp where unemployment was stable) and falling local home prices are correlated with higher mortgage delinquencies, signaling regional credit stress despite relatively strong credit scores on newly originated mortgages.

Analysis

Market structure: Rising delinquencies are concentrated in lowest-income zip codes (90+ days up from ~0.5% to ~3% since 2021) and regions with >1.0ppt unemployment moves, creating winners (credit-data vendors like EFX, diversified national banks, agency MBS) and losers (regional banks, non‑bank originators/servicers, private‑label RMBS, regional homebuilders). Expect pricing power to shift toward firms with scale, low credit exposure and servicing capabilities; originator volumes will compress and risk premia on non‑agency paper should widen by tens to low hundreds of basis points if trends persist.

Risk assessment: Tail risk is a localized labor‑market shock that cascades (e.g., an additional 1.0–1.6ppt county unemployment rise could lift mortgage delinquency flows by ~0.4–0.6ppt nationally and materially widen MBS spreads). Immediate (days) risk: equity repricing of regional banks; short term (3–6 months): ABS and non‑agency RMBS losses; long term (quarters) risk: weaker housing markets amplifying defaults. Hidden dependencies include servicer liquidity/operational strain and concentrated state exposures (Florida, parts of Midwest).

Trade implications: Tactical ideas—short concentrated regional bank exposure and non‑agency mortgage risk while owning agency MBS and data/servicing franchises. Use pair trades to isolate credit risk (long JPM/BLK vs short KRE), and optioned downside protection on regional bank ETFs and homebuilder names (3–9 month put spreads). Monitor triggers: bottom‑quartile 90+ delinq >2.0% or county unemployment +1.0ppt to add size.

Contrarian angles: The market may overreact—aggregate mortgage serious delinquency (~1.3% newly delinquent in 2025) remains near long‑run norms and median new‑mortgage FICO >750, so forced selling could create buyable dislocations in agency MBS and select bank credit. History (post‑2010 regional stress) shows contained spillovers without broad systemic stress; hedge for regulatory backstops and prepare for mean reversion if spreads overshoot.

Key Decisions for Investors

• Establish a 2–3% short position in regional bank ETF KRE via a 3–6 month put spread (strike band 10–15% below spot) to capture near‑term downside from concentrated mortgage/consumer stress; hedge with a 1–1 long in JPM or BAC to isolate regional credit risk.
• Allocate 3–5% overweight to short‑duration agency MBS (ETF: MBB) for defensive yield and capital‑preservation versus non‑agency RMBS; increase exposure if non‑agency RMBS spreads widen >50bps versus Agency Treasuries.
• Initiate a 1–2% long position in Equifax (EFX) equity to play increased demand for granular credit data and risk‑modeling services; add on pullbacks >8% as momentum to outsource credit analytics accelerates.
• Open a 2% notional short on homebuilders (PHM or DHI) via 4–9 month put spreads sized to risk tolerance; add if zip‑level home‑price indices in Gulf Coast counties fall a further 3–5% or if mortgage 90+ delinquencies in bottom income quartile exceed 2.5%.
• Set alerts and increase hedges if (a) county unemployment change >+1.0ppt for counties representing >5% population or (b) national newly‑serious mortgage delinquencies rise >0.5ppt quarter‑over‑quarter; on trigger, increase short/regional bank and non‑agency RMBS protection by another 1–2%.