The Shock: A deep recession hits. Unemployment doubles from 4.5% to 10%. House prices crash 29%. The stock market drops 54%. Commercial real estate falls 40%. The Fed slashes rates to near zero. BBB corporate spreads blow out to 5.7 percentage points. This is 2008 again — maybe worse.
The Method: 476 digital twins — synthetic consumer personas representing a cross-section of US borrowers — were asked 13 questions across four domains: payment priority, default tipping points, negative equity behaviour, and post-crisis recovery. Each twin responded based on their individual financial profile, attitudes, and circumstances.
The Purpose: To provide the behavioural layer underneath a bank’s DFAST loss model. Macro regressions tell you how much defaults rise. Twins tell you who defaults, when, and in what order.
BNPL is the canary in the coal mine for Gen Z. 71% of Gen Z twins cut BNPL first, vs. 41% of Boomers. For a bank monitoring early-warning signals of consumer stress, BNPL delinquency in the under-35 segment is the leading indicator — it will show up before credit cards, well before mortgages.
Twin data aligns with the TransUnion payment hierarchy. Cars are protected, unsecured debt gets cut. The twins add a layer TransUnion’s historical data cannot: BNPL as a new, lowest-priority debt class that will show stress signals first in the next downturn, particularly among younger borrowers. LendingTree reports 41% of BNPL users already made late payments in the past year — before any recession.
86% of twins would either exhaust all other options or sell before missing a mortgage payment. Only 12% crack within the first two months of income loss. This suggests mortgage default rates in the early months of a recession may be lower than macro models predict — but when they do come, they come in a concentrated wave from the most vulnerable segments.
| Generation | Keep Paying | Sell / Negotiate | Stop Paying |
|---|---|---|---|
| Gen Z | 29% | 60% | 11% |
| Gen X | 61% | 35% | 4% |
| Boomers | 79% | 20% | 1% |
Gen Z is 11× more likely to strategically default than Boomers. At 30% negative equity, 11% of Gen Z twins would stop paying outright and 60% would try to exit. A macro model that applies a single strategic default rate across the mortgage book is materially wrong. Loss-given-default models need a generational overlay.
Twin overall strategic default rate (3.6%) closely matches the Fed’s observed rate (~4%) from the 2008 crisis. The moral aversion to strategic default is real and shows in the data. But the generational breakdown is new — Gen Z at 11% is materially higher than any historically observed rate, and there is no 2008-era data on this cohort. This is a forward-looking risk that historical regressions cannot capture.
93% of twins say bank loyalty depends entirely on crisis-period treatment. This matches industry data: 62% of bank customers say they would switch if treated impersonally, and responsive support is a top-3 retention driver. The implication is clear — investment in forbearance programmes, responsive customer service, and proactive outreach during a downturn directly determines post-crisis churn. Banks that go dark during the recession will lose customers on the other side.
Consumers will exhaust every other option before considering bankruptcy or formal debt management. This means the payment hierarchy — BNPL → credit cards → personal loans → car → mortgage → utilities — will play out in full before formal insolvency enters the picture. Loss models should expect a long tail of delinquency before charge-offs in any segment.
BNPL gets cut first by 49% of twins, rising to 71% among Gen Z. This is a new debt class that did not exist in 2008, so historical regressions cannot model its behaviour. Real-world data already confirms the risk: LendingTree reports 41% of BNPL users made late payments in the past year, up from 34%. Gen Z holds 64% BNPL usage with 71% carrying multiple simultaneous loans. In the next downturn, BNPL delinquency in the under-35 segment is the earliest stress signal available.
Overall strategic default at 30% negative equity is 3.6%, closely matching the Fed’s observed ~4% rate from the 2008 crisis. But Gen Z is 11× more likely to walk away than Boomers (11% vs. 1%). A loss model applying a single strategic default rate across the mortgage book will understate risk in young-borrower segments and overstate it in older segments. This generational divergence has no historical precedent to calibrate against.
TransUnion’s established hierarchy from 2008 was: auto > credit cards > mortgages. Twin data confirms the structure but adds BNPL below credit cards. The full hierarchy in 2026 is: mortgage/rent > utilities > car > personal loan > credit card > BNPL. During the 2008 recession, consumers prioritised credit cards over mortgages when homes went underwater — that pattern persists, but BNPL now absorbs the first wave of stress.
86% of twins would either exhaust all other options or sell before missing a mortgage payment. Only 12% default within the first two months of income loss. This suggests the early-stage mortgage delinquency rate in a recession may be lower than a linear macro model assumes — with losses concentrated later and in specific vulnerable segments rather than spread evenly.
93% of twins say post-crisis bank loyalty depends on how they were treated during the downturn. Only 6% would stay unconditionally, and only 1% would definitely leave. Industry data corroborates this: 62% of customers will switch if treated impersonally. Banks that invest in forbearance and responsive service during the recession retain customers. Banks that don’t face a churn wave on recovery.
99% say bankruptcy or debt management is only considered as an absolute last resort. No segment shows proactive consideration of formal insolvency. The full payment hierarchy will play out — from BNPL to credit cards to personal loans to secured debt — before charge-offs or formal defaults materialise. This implies a long delinquency-to-default runway that can be used for early intervention.