SME and Corporate Lending Modelling
We build credit risk models for business lending, designed for thin data and heterogeneous borrowers.
Specialist
Lending
Credit risk where the borrower is a business, an asset or a project rather than a consumer, and the data is thinner for it.
Specialist Lending Deliverables
Project Finance Modelling
We build risk models for lending against a single asset and the cash flow it generates.
Equipment Leasing Modelling
We model risk on asset finance and leasing books, including the secondary market value of the equipment.
CRE Modelling
We build risk models for commercial real estate lending, covering bridging, development and buy-to-let, structured around sale, refinance and exit.
Trade Finance Modelling
We build risk and pricing models for lending against invoices, inventory and trade flows.
Motor Lending Modelling
We build credit risk and pricing models for motor finance, covering hire purchase, personal contract purchase and the residual value the contract is written against.
Banking Case Studies
Banking
Independent validation of a Pillar 2A concentration add-on
A European bank needed to know that the capital it holds against concentrated lending would stand up to scrutiny, so it asked us for an independent view of the model behind the number.
Banking
Redeveloping every IFRS 9 model across two portfolios
A UK lender needed every model in its IFRS 9 expected credit loss suite rebuilt, at the point when the team that had built them was no longer there.
Frequently Asked Questions
-
Generic scorecards break on specialist economics, usually because the exit rather than the monthly payment carries the risk. At Gini we build models tuned to that: SME and asset-based lending, bridging and development finance, and buy-to-let, where the security, the exit route and the borrower’s other exposures matter more than a conventional application score.
-
Yes, with the uncertainty stated rather than hidden. Where internal defaults are too few to support a segment, Gini pools across comparable exposures, uses external and market data to anchor the estimate, and quantifies the margin of conservatism that the thin data justifies. The alternative, a confident-looking model fitted to twenty defaults, is the version that fails review.
-
Separately from payment performance, because a loan that has never missed a payment can still fail at redemption. At Gini we model the exit itself: the refinance or sale that repays the facility, the time it takes, and the value achieved against the original valuation. Development exposures are modelled by stage, so drawn cost and remaining cost to complete are held apart.
-
Yes. Specialist portfolios tend to fail validation on segmentation and on the treatment of exceptions rather than on the statistics, so Gini tests whether the segments still hold given how the book has grown and whether the overrides have quietly become the strategy. Where an override rate is high, we quantify what it is worth before recommending anything.