When a parent company guarantees a subsidiary's third-party borrowing, the subsidiary typically secures a lower interest rate than it would have obtained on a standalone basis, because the lender's credit risk is reduced. That reduction in borrowing cost is a real economic benefit, and the arm's length principle requires that the guarantor be compensated for providing it. The question that generates the most transfer pricing controversy in this space is not whether a guarantee fee is payable, but how much it should be, and several competing methodologies routinely produce very different answers for the same transaction.
The Yield Approach
The most commonly used and most intuitive method compares the interest rate the borrower would pay without the guarantee (based on its own standalone credit rating) against the rate it actually pays with the guarantee in place. The difference, the yield spread, represents the benefit conferred by the guarantee, and the guarantee fee is set as some portion of that spread, reflecting that the benefit should be shared between guarantor and borrower rather than captured entirely by one side.
The Cost Approach and Expected Loss Method
An alternative approach prices the guarantee based on the guarantor's expected cost of providing it, essentially, the expected loss the guarantor bears from the probability of default multiplied by the loss given default, similar to how a commercial guarantor or insurer would price risk. This method draws on option-pricing and credit-risk modelling techniques and is often used where reliable market-based comparables for the yield approach are unavailable.
The CDS Spread Approach
Where a comparable Credit Default Swap exists for the guarantor's or a similarly rated entity's debt, the CDS spread can be used as a market-observed proxy for the cost of credit protection, and by extension, for the arm's length guarantee fee. This approach has gained traction because CDS spreads are transparent and update in real time, but its reliability depends heavily on finding a genuinely comparable reference entity, tenor, and seniority, a challenge that mirrors the comparability demands of the CUP method more broadly.
Implicit Support and the Group Benefit Question
A persistent controversy is the extent to which a subsidiary already benefits from implicit group support, meaning the market's assumption that a parent would not allow a subsidiary to default, even absent a formal guarantee. Several jurisdictions, drawing on cases in this area, hold that only the incremental benefit above and beyond implicit support should be compensated, meaning the guarantee fee should be based on the credit rating uplift attributable specifically to the explicit guarantee, not the full gap between standalone and group ratings.
Guarantee Fee Documentation in Practice
A robust guarantee pricing file typically presents the standalone rating derivation for the borrower, a clearly reasoned notching adjustment to arrive at the implicitly supported rating, and a final incremental adjustment isolating the explicit guarantee's contribution before applying the yield spread, cost approach, or CDS benchmark. Where more than one methodology produces materially different results, the practitioner should be prepared to explain the divergence rather than silently selecting whichever number is most favourable, since tax authorities increasingly expect a reconciliation across methods as standard practice for material guarantee arrangements.
Conclusion
Guarantee fee benchmarking should present the yield-spread approach, the expected-loss or cost approach and, where a comparable CDS exists, the market-spread approach side by side, with the final fee isolating only the incremental rating uplift attributable to the explicit guarantee, net of implicit group support. A single-method guarantee fee, unsupported by a rating notching analysis, is one of the most frequently adjusted positions in financial transaction transfer pricing audits.
