The Cross-Border Default Premium: Why the Same Deal Is 2.5x Riskier Across a Border
Ask a credit committee what changes when a deal crosses a border and you will usually hear about currency, withholding tax, and enforceability. What you rarely hear is a number.
Our research on 13,317 private credit transactions worth $11.8 trillion between 2004 and 2025 puts one on it. Cross-border deals default at 6.46%. Domestic deals default at 2.56%. That is a 3.91 percentage point premium — cross-border transactions default at roughly two and a half times the domestic rate, and the gap persists across multiple empirical specifications.
It is not just more defaults — it is faster ones
A raw default rate hides the time dimension. A deal that defaults in year seven is a very different proposition from one that defaults in year two, even if both eventually appear in the same statistic.
Cox proportional hazards estimation gives the sharper answer: cross-border deals face a hazard ratio of 2.86. At any given point in a deal's life, a cross-border transaction is nearly three times more likely to default than an otherwise comparable domestic one. The premium is not a late-life artefact. It is present throughout.
Why does it persist?
This is the part that should trouble anyone comfortable with the idea that capital markets have gone global. The frictions that generate this premium are the ones that are hardest to arbitrage away:
- Information asymmetry. Monitoring a borrower is harder at distance. Management meetings are less frequent, local market intelligence is thinner, early warning signs surface later.
- Legal complexity. Enforcement runs through an unfamiliar system. Security may be perfected differently, or not at all. Restructuring outcomes are less predictable.
- Monitoring cost. Everything above is fixable — for a price. The price is the premium.
None of these are eliminated by faster data or better derivatives. They are structural.
What this means for a credit process
If your PD model has no cross-border variable, it is mispricing. A single global PD curve applied to a book containing both domestic and cross-border exposures will systematically underestimate risk on the cross-border portion and overstate it on the domestic portion. The errors do not cancel; they concentrate exactly where the exposure is largest.
Three practical implications:
- Price the premium explicitly. A 3.91pp differential is not noise. On a large exposure it is the difference between an adequate and an inadequate risk-adjusted return.
- Reflect it in capital, not just pricing. The hazard ratio of 2.86 means the timing of losses is different too, which matters for lifetime ECL under IFRS 9 and for economic capital horizons.
- Be honest about which side of the border your comparables sit on. Benchmarking a cross-border deal against a domestic comparable set imports the wrong base rate.
The uncomfortable question
If distance still costs this much in a market with no shipping, no tariffs, and instantaneous settlement — what exactly did integration buy?
That is the subject of our companion analysis on distance elasticity in private credit flows, where the gravity model puts the friction at -2.41, comparable to physical goods trade.
Findings from "Cross-Border Shock Transmission in Private Credit Markets: Evidence from Global Deal-Level Data" (2025), presented at the Bank of England Agenda for Research (BEAR) Conference 2026.
Need this applied to your portfolio?
We turn research into models and advice you can act on — economic capital, climate ICAAP, and model governance.