How to Run an Intercompany Loan Benchmarking Analysis

An intercompany loan analysis is a sequence of dependent steps. Each one relies on the output of the step before it, and skipping a step leaves the analysis exposed at exactly the point a reviewer will test.

The practical takeaway is that pricing is the last step, not the first. A rate benchmarked without the delineation, capacity, and rating work behind it prices an instrument that may not be respected, in an amount that may not be arm’s length, against comparables of the wrong risk. Before running any search or pulling any market data, fix the sequence and work it in order.

The five steps below are worked in order, each on the output of the one before it, and the whole sequence is then run through a single worked example.

Establish what the transaction actually is, from the conduct of the parties and the terms of the arrangement, not from the label on the intercompany agreement. This step has two parts.

Confirm the characterization as debt. Weigh the arrangement against how independent parties would have structured it:

• Fixed maturity: is repayment expected on a date certain?
• Repayment obligation: is the borrower unconditionally bound to repay?
• Enforceability: can the lender compel payment on default?
• Source of repayment: can the borrower service the loan from its own cash flow?
• External borrowing ability: could it have raised comparable debt outside the group?
• Capitalization: is the borrower thinly capitalized after the advance?

Where the arrangement lacks the ordinary indicia of a loan, part or all of it may be delineated as equity, which removes interest deductibility rather than merely adjusting a rate.

Record the economically relevant characteristics. These are the loan’s comparability factors and they govern every later step: principal amount, currency, maturity and tenor, seniority and ranking, security and collateral, covenants, repayment profile, optionality, and purpose of the funds. Note in particular that currency and tenor will fix the risk-free base in Step 4, while the remaining characteristics shape the spread.

Test how much of the debt is arm’s length. This is the quantum question, and it comes before pricing because a rate benchmarked onto a principal that exceeds the borrower’s capacity is exposed regardless of how well the rate itself is built.

Assess realistic borrowing capacity from the borrower’s cash-generating ability, cash-flow stability, available assets, and the leverage its sector and market tolerate. In practice:

• Read coverage ratios (interest coverage, debt service coverage) alongside leverage ratios (debt to EBITDA, debt to equity).
• Set the threshold each ratio must clear from how comparable independent borrowers of a similar profile are actually financed.
• Identify the arm’s length principal: the amount the borrower could raise and service on its own account. Any excess above it may be recharacterized, with interest on the excess disallowed.

Check the interest limitation rule separately. Confirm whether the resulting interest survives any fixed-ratio interest limitation rule in the borrower’s jurisdiction. That test operates independently of the arm’s length capacity test, and clearing one does not clear the other.

The borrower’s credit rating drives the credit spread and is usually the most contested input in the analysis. Derive it in three moves:

• Establish the standalone rating from the borrower’s own financials, using a quantitative scorecard grounded in its coverage, leverage, profitability, size, and stability, refined by a documented qualitative overlay for factors the scorecard does not capture.
• Assess implicit support. A subsidiary benefits from group membership even without any formal guarantee, and lenders price that expected support. Judge how strong the group is and how important and integrated the borrower is within it.
• Notch upward from the standalone rating toward the group rating to reflect that support, without exceeding the group rating. A core, integrated subsidiary of a strong group attracts a larger uplift; a peripheral one may attract none.

The resulting rating is the one carried into pricing, and because it carries a large share of the arm’s length answer, its derivation must be fully documented.

With the principal fixed and the borrower rated, price the rate. The CUP method is generally the most appropriate, and an internal CUP, drawn from a party’s actual third-party borrowing or lending, is the most reliable benchmark where one exists. Where none exists, construct an external CUP as a risk-free base plus a credit spread:

• Set the risk-free base from the loan’s currency and tenor, reading a highly-rated government yield in the loan’s currency at the point on the curve matching its maturity.
• Build the credit spread from comparable bonds or loans matched to the borrower’s rating, the loan’s currency, and its tenor, then adjust for differences in seniority, security, covenants, and structure.
• Date the comparables to the pricing date, because interest rates move and stale market data is not comparable however well matched on credit.
• Aggregate to a range using the interquartile range and interpolation method applied in any benchmarking study, then add the range to the risk-free base to produce the all-in arm’s length range.

Test the actual rate against the constructed arm’s length range:

• Within the range: generally treated as arm’s length, no adjustment required.
• Outside the range: the median is the usual point of adjustment, and the direction of the miss shows which jurisdiction bears the exposure. A rate set too high strips deductible interest from the borrower’s country; a rate set too low leaves too much profit there and understates the lender’s return.

Document the full study so a reviewer following the steps arrives at the same conclusion. The audit trail records:

• The delineation and characterization analysis.
• The capacity assessment and the comparables behind its thresholds.
• The rating build, with scorecard inputs and implicit-support reasoning.
• The loan comparables and their screening.
• The yield construction and the final range and conclusion.

Lock the methodology (the pricing date, the comparable screens, the averaging and percentile conventions) before running the search, so the outcome is not reverse-engineered.

The Sequence in One Pass

To show how the steps connect, consider a single fictional facility carried through the workflow.

Each figure is illustrative, and each step depends on the one before it. Change the rating from BBB to BB and the rate moves materially; change the capacity multiple and part of the principal may fall outside arm’s length debt. That dependency is the reason the sequence matters.

A consolidated worklist for the full analysis. Each item should be completed and evidenced in the study file.