LGD models.
From recovery cash flows to downturn LGD.

Workout-based LGD for secured and unsecured portfolios, including downturn adjustments and the treatment of defaulted exposures.

Cumulative recovery after defaultIllustrative
80%60%40%20%0%012243660 m Ultimate recovery 78% → realised LGD 22% before costs
Regulatory basis
CRR · EBA GL on PD & LGD estimation · EBA GL on downturn LGD
Portfolios
Mortgages, SME, corporate, unsecured retail
Outputs
Long-run LGD, downturn LGD, ELBE and LGD in-default
Used for
IRB capital, IFRS 9, stress testing
Approach

Every recovery, discounted and costed. Every downturn assumption on the record.

01

Realised LGD

Discounted recoveries and workout costs per defaulted facility, incomplete workouts treated explicitly.

02

Risk drivers

Segment by collateral, loan-to-value, seniority and cure likelihood.

03

Downturn adjustment

Identify downturn periods and quantify their impact on loss rates.

04

Defaulted assets

Estimate ELBE and LGD in-default by time in default.

Data

What you need. And what happens if you don't have it.

DataNeededIf missing or incomplete
Recovery cash flowsRequired—
Workout costsRequiredCost proxy from recent cohorts, with added conservatism
Collateral valuationsRequired (secured)Indexed valuations
Cure eventsHelpfulCure rate estimated at segment level
House price and macro seriesRequired for downturnPublic series
In Model Studio 0.1.0

Every choice is a decision for a person. Every test has its threshold on the page.

Decisions the agents propose

Each comes with a recommended option, the reason and any data gap. The model owner approves the set at the design gate.

  1. Discount rate
    Which rate discounts recoveries and costs to the default date?
  2. Maximum workout period
    After how many years is an open workout treated as closed?
  3. Risk drivers
    How are facilities segmented?
  4. Downturn period
    Which default years form the downturn period?
  5. Margin of conservatism
    How is general estimation error added?
Tests the engine runs

Shown next to each result. A test outside its threshold stays visible and becomes a limitation in the documentation.

TestGroupThreshold
Realised versus estimated LGDBack-testingRecent realised ≤ estimate + 2%
Sensitivity to the discount rateSensitivityReported
Treatment of incomplete workoutsData|difference| ≤ 2 points
Downturn impact analysisDownturnDownturn LGD ≥ long-run LGD
Back-testing by segmentBack-testingCoefficient of variation across years reported
Related models
All of Model Studio →
PD modelsBuilt from the defaults you can observeExplore →EAD & CCFWhat gets drawn before defaultExplore →Satellite modelsMacro scenarios into risk parametersExplore →

Rethinking your LGD framework?
See it done on your data.

Talk to sales

We'll walk through your use case and show the product on data structured like yours.

Talk to sales Book a 30-minute call
[sales@yourdomain.eu][+31 20 000 0000]
Talk to sales