Underwriting
Risk-based loan pricing
Work out the rate a loan has to carry before it makes money. Enter the risk and the funding cost, and see the break-even rate, the rate that hits your target return on equity, and what happens when default rates come in worse than you priced.
Inputs
All-in annual interest rate on the loan.
One-off, spread across the term below.
What your debt funding costs you annually.
Annualised PD for this risk grade.
Share of exposure lost after recovery.
Origination and servicing cost to carry one file.
Equity share funding the loan. The rest is debt.
Result
Break-even rate
Covers funding, expected loss and opex. Zero profit.
Rate at 18% ROE
Break-even plus the return your equity has to earn.
Headroom vs break-even
RAROC at your rate
Expected loss / year
Annual profit waterfall, per loan
Stress: default rate doubles
Still clearsPD 3.00% → 6.00%
Break-even becomes
RAROC becomes
Estimate only. This is a single-period, pre-tax model: it holds exposure flat over the year rather than amortising it, treats capital as a fixed share of the loan, and spreads the origination fee straight-line across the term. Your own funds-transfer pricing, capital rules and recovery assumptions will move these numbers.
Method
How the break-even rate is built
A loan has to cover four things before it earns anything: the cost of the money you lent, the losses you expect across a book of loans like it, the cost of running the file, and the return your shareholders expect on the equity tied up against it.
The first three give you the break-even rate. Funding cost is your cost of funds applied only to the debt-funded portion, since the equity slice does not pay interest. Expected loss is probability of default multiplied by loss given default. Operating cost is the annual cost to carry one file, expressed against the loan amount. Origination fees are spread across the term and subtracted, because they offset what the rate has to earn.
Adding the return on allocated capital gives the target rate. RAROC then works backwards: net contribution divided by the equity allocated to the loan, which is the number to compare against your hurdle.
The part most pricing sheets skip is the stress row. Expected loss is the single input most likely to be wrong, because it is the only one that depends on a forecast rather than a contract. If doubling the default rate turns the loan loss-making, the price was carrying no risk buffer at all.
The arithmetic here is easy. The hard part is the PD you feed it. Truffles builds credit models that produce calibrated, explainable default probabilities from financial statements and alternative data, so the number going into this calculation reflects the borrower in front of you rather than a grade-level average.
See how the credit models workFAQ
Common questions
What is a risk-based break-even rate?
The interest rate at which a loan exactly covers its own costs: funding, expected credit loss, and operating expense, net of fees. Below it the loan destroys value even if the borrower never misses a payment, because the price does not cover the average loss across loans of that risk grade.
How is expected loss calculated?
Probability of default multiplied by loss given default, applied to the exposure. A 3% PD with 45% LGD gives a 1.35% annual expected loss rate. It is an average across a portfolio, not a prediction about one borrower, which is why it belongs in the price rather than in the credit decision.
Why does cost of funds apply only to part of the loan?
Because equity does not pay interest. If 15% of the loan is funded by allocated capital, only the remaining 85% carries your cost of funds. The equity portion is compensated through the target return on equity instead, which is added on top of break-even.
What is RAROC and what is a good number?
Risk-adjusted return on capital: net contribution after expected loss, divided by the equity allocated to the loan. It is the figure to compare against your cost of equity or hurdle rate. A loan can look profitable on margin and still fail on RAROC if it consumes too much capital for the return it produces.
Why stress the default rate rather than the interest rate?
Rate and funding cost are contractual and observable. The default rate is an estimate, and it is the input most likely to be wrong in the direction that hurts. Doubling it is a crude but honest test of whether the price carries a genuine risk buffer.
Price the risk you actually have
Bring us a portfolio and we will show you what calibrated default probabilities do to your pricing, your provisioning, and your approval rate.